A phylogenetic analysis of the arachnid orders based on morphological characters

45
Zoological Journal of the Linnean Society, 2007, 150, 221–265. With 6 figures © 2007 The Linnean Society of London, Zoological Journal of the Linnean Society, 2007, 150, 221–265 221 Blackwell Publishing LtdOxford, UKZOJZoological Journal of the Linnean Society0024-4082© 2007 The Linnean Society of London? 2007 1502 221265 Original Article PHYLOGENY OF THE ARACHNID ORDERSJ. W. SHULTZ *E-mail: [email protected] A phylogenetic analysis of the arachnid orders based on morphological characters JEFFREY W. SHULTZ* Department of Entomology, University of Maryland, College Park, MD 20742, USA Received October 2005; accepted for publication June 2006 Morphological evidence for resolving relationships among arachnid orders was surveyed and assembled in a matrix comprising 59 euchelicerate genera (41 extant, 18 fossil) and 202 binary and unordered multistate characters. Par- simony analysis of extant genera recovered a monophyletic Arachnida with the topology (Palpigradi (Acaromorpha (Tetrapulmonata (Haplocnemata, Stomothecata nom. nov.)))), with Acaromorpha containing Ricinulei and Acari, Tetrapulmonata containing Araneae and Pedipalpi (Amblypygi, Uropygi), Haplocnemata (Pseudoscorpiones, Solif- ugae) and Stomothecata (Scorpiones, Opiliones). However, nodal support and results from exploratory implied weights analysis indicated that relationships among the five clades were effectively unresolved. Analysis of extant and fossil genera recovered a clade, Pantetrapulmonata nom nov., with the topology (Trigonotarbida (Araneae (Hap- topoda (Pedipalpi)))). Arachnida was recovered as monophyletic with the internal relationships (Stomothecata (Pal- pigradi, Acaromorpha (Haplocnemata, Pantetrapulmonata))). Nodal support and exploratory implied weights indicated that relationships among these five clades were effectively unresolved. Thus, some interordinal relation- ships were strongly and/or consistently supported by morphology, but arachnid phylogeny is unresolved at its deepest levels. Alternative hypotheses proposed in the recent literature were evaluated by constraining analyses to recover hypothesized clades, an exercise that often resulted in the collapse of otherwise well-supported clades. These results suggest that attempts to resolve specific nodes based on individual characters, lists of similarities, evolutionary sce- narios, etc., are problematic, as they ignore broader impacts on homoplasy and analytical effects on non-target nodes. © 2007 The Linnean Society of London, Zoological Journal of the Linnean Society, 2007, 150, 221–265. ADDITIONAL KEYWORDS: Arachnida – fossil – morphology – phylogeny – systematics INTRODUCTION Despite an ever-increasing reliance on molecular sequence data for phylogeny reconstruction and evo- lutionary inference, morphological characters remain an important source of phylogenetic signal (both alone and in combination with molecular data) and are essential for reconstructing and exploring patterns in organismal evolution. Construction and maintenance of digital databases of structural information are essential if morphology and the results of morpholog- ical analyses are to remain useful. In the present work, I define, homologize and code morphological and other non-molecular characters that vary among orders and major intraordinal groups of arachnids. The results are summarized in a taxon-by-character matrix, and the phylogenetic signal within the matrix is explored using parsimony-based analyses. This research clarifies the strengths and weaknesses of cur- rent understanding of arachnid phylogeny and high- lights several aspects of phylogenetic practice that may impede progress in the evolutionary morphology and phylogeny of Arachnida. Several interordinal relationships within Arachnida can be considered well established. The monophyly of Arachnida itself is well supported by morphological characters, although some palaeontologists continue a long tradition of placing scorpions outside arachnids with Eurypterida (e.g. Dunlop & Braddy, 2001). The- lyphonida and Schizomida include the tailed whip- scorpions and form an undisputed monophyletic group, Uropygi s.l. (= Camarostomata). Recent work indicates that Amblypygi, or whipspiders, is the sister group to Uropygi s.l. (Shear et al., 1987; Shultz, 1990, 1999; Giribet et al., 2002; but see Alberti, 2005) and together form the clade Pedipalpi. Earlier studies

Transcript of A phylogenetic analysis of the arachnid orders based on morphological characters

Zoological Journal of the Linnean Society

2007

150

221ndash265 With 6 figures

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

221

Blackwell Publishing Ltd

Oxford UK

ZOJZoological Journal of the Linnean Society

0024-4082copy 2007

The Linnean Society of London 2007

150

2

221265Original Article

PHYLOGENY OF THE ARACHNID ORDERSJ W SHULTZ

E-mail jshultzumdedu

A phylogenetic analysis of the arachnid orders based on morphological characters

JEFFREY W SHULTZ

Department of Entomology University of Maryland College Park MD 20742 USA

Received October 2005 accepted for publication June 2006

Morphological evidence for resolving relationships among arachnid orders was surveyed and assembled in a matrixcomprising 59 euchelicerate genera (41 extant 18 fossil) and 202 binary and unordered multistate characters Par-simony analysis of extant genera recovered a monophyletic Arachnida with the topology (Palpigradi (Acaromorpha(Tetrapulmonata (Haplocnemata Stomothecata

nom nov

)))) with Acaromorpha containing Ricinulei and AcariTetrapulmonata containing Araneae and Pedipalpi (Amblypygi Uropygi) Haplocnemata (Pseudoscorpiones Solif-ugae) and Stomothecata (Scorpiones Opiliones) However nodal support and results from exploratory impliedweights analysis indicated that relationships among the five clades were effectively unresolved Analysis of extantand fossil genera recovered a clade Pantetrapulmonata

nom nov

with the topology (Trigonotarbida (Araneae (Hap-topoda (Pedipalpi)))) Arachnida was recovered as monophyletic with the internal relationships (Stomothecata (Pal-pigradi Acaromorpha (Haplocnemata Pantetrapulmonata))) Nodal support and exploratory implied weightsindicated that relationships among these five clades were effectively unresolved Thus some interordinal relation-ships were strongly andor consistently supported by morphology but arachnid phylogeny is unresolved at its deepestlevels Alternative hypotheses proposed in the recent literature were evaluated by constraining analyses to recoverhypothesized clades an exercise that often resulted in the collapse of otherwise well-supported clades These resultssuggest that attempts to resolve specific nodes based on individual characters lists of similarities evolutionary sce-narios etc are problematic as they ignore broader impacts on homoplasy and analytical effects on non-targetnodes copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

ADDITIONAL KEYWORDS Arachnida ndash fossil ndash morphology ndash phylogeny ndash systematics

INTRODUCTION

Despite an ever-increasing reliance on molecularsequence data for phylogeny reconstruction and evo-lutionary inference morphological characters remainan important source of phylogenetic signal (both aloneand in combination with molecular data) and areessential for reconstructing and exploring patterns inorganismal evolution Construction and maintenanceof digital databases of structural information areessential if morphology and the results of morpholog-ical analyses are to remain useful In the presentwork I define homologize and code morphological andother non-molecular characters that vary amongorders and major intraordinal groups of arachnidsThe results are summarized in a taxon-by-charactermatrix and the phylogenetic signal within the matrix

is explored using parsimony-based analyses Thisresearch clarifies the strengths and weaknesses of cur-rent understanding of arachnid phylogeny and high-lights several aspects of phylogenetic practice thatmay impede progress in the evolutionary morphologyand phylogeny of Arachnida

Several interordinal relationships within Arachnidacan be considered well established The monophyly ofArachnida itself is well supported by morphologicalcharacters although some palaeontologists continue along tradition of placing scorpions outside arachnidswith Eurypterida (eg Dunlop amp Braddy 2001) The-lyphonida and Schizomida include the tailed whip-scorpions and form an undisputed monophyleticgroup Uropygi

sl

(

=

Camarostomata) Recent workindicates that Amblypygi or whipspiders is the sistergroup to Uropygi

sl

(Shear

et al

1987 Shultz 19901999 Giribet

et al

2002 but see Alberti 2005) andtogether form the clade Pedipalpi Earlier studies

222

J W SHULTZ

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

tended to favour a sister-group relationship betweenAmblypygi and Araneae the spiders (eg Platnick ampGertsch 1976 Weygoldt amp Paulus 1979) but this wasbased on a few similarities that do not appear to besynapomorphic (Shultz 1990 1999 present study)Araneae and Pedipalpi form a well-supported cladewith the fossil order Trigonotarbida being its likelysister group Pseudoscorpiones and Solifugae or sun-spiders are often united within a clade Haplocne-mata (

=

Apatellata) (Weygoldt amp Paulus 1979 Vander Hammen 1989 Shultz 1990 Wheeler amp Hayashi1998 Giribet

et al

2002 but see Alberti amp Peretti2002)

Some interordinal relationships are often recoveredin phylogenetic analyses but are supported by rela-tively few characters These include Dromopoda (Opil-iones

+

Scorpiones

+

Haplocnemata) Megoperculata(

=

Palpigradi

+

Tetrapulmonata) and Acaromorpha(Ricinulei

+

Acari) (Shultz 1990 Wheeler amp Hayashi1998 Giribet

et al

2002) The status of Acaromorphais further complicated by substantial morphologicaldivergence within and between the two principal lin-eages of Acari (ie Anactinotrichida and Acariformes)(Lindquist 1984) which has led some workers to sug-gest that Acari is diphyletic (Van der Hammen 1989)Relationships among Palpigradi Tetrapulmonata

Acaromorpha Haplocnemata Opiliones and Scorpi-ones are effectively unresolved as is the placement ofseveral fossil taxa (eg Haptopoda)

Results from the present analysis (Fig 1) essen-tially affirm this characterization of our currentunderstanding of arachnid phylogeny based on mor-phology and other non-molecular characters but alsooffer new proposals and evaluate alternative interpre-tations that have emerged in the last 17 years iesince my previous attempt to resolve arachnid phylog-eny (Shultz 1990) Specifically parsimony-based anal-yses corroborate the monophyly of Arachnida as wellas Uropygi Pedipalpi Haplocnemata and Acaromor-pha The data also support more recent proposalsincluding the monophyly of Opiliones and Scorpiones(

=

Stomothecata

nom nov

) (Shultz 2000) and Pan-tetrapulmonata

nom nov

with an internal structureanticipated by Dunlop (1999 2002c) ie (Trigonotar-bida (Araneae (Haptopoda Pedipalpi))) Howeverrelationships among Palpigradi Acaromorpha Hap-locnemata Stomothecata and Pantetrapulmonata areeffectively unresolved These results indicate thatmorphology offers important phylogenetic informa-tion but it is not yet sufficient to resolve relationshipsat the deepest levels within Arachnida

METHODS

T

ERMINAL

TAXA

The study was based on 59 euchelicerate genera (41extant 18 fossil) with most represented by one spe-cies (detailed below) coded for 202 binary and unor-dered multistate characters (Table 1 Appendix)

Xiphosura (horseshoe crabs)

The xiphosurans are anancient (SilurianndashRecent) aquatic lineage with itsgreatest diversity occurring in the fossil record Itcomprises two main groups Synziphosurida (Sil-urianndashDevonian ~ ten genera) and Xiphosurida (Car-boniferousndashRecent ~14 genera) Synziphosurids areprobably paraphyletic and retain plesiomorphic fea-tures such as a ten-segmented opisthosoma withthree segmented metasoma (Anderson amp Selden1997) They were represented in the matrix by

Wein-bergina opitzi

one of the few fossil xiphosurans withpreserved appendages (Moore Briggs amp Bartels2005) and a more typically preserved synziphosurid

Limuloides limuloides

Extant xiphosurans (threegenera four spp) were represented by the intensivelystudied Atlantic horseshoe crab

Limulus polyphemus

and an Asian horseshoe crab

Carcinoscorpiusrotundicauda

with supplemental information drawnfrom another Asian species

Tachypleus tridentatus

Eurypterida (sea scorpions)

A diverse (

gt

60 genera)aquatic group of fossil euchelicerates that ranged from

Figure 1

Cladogram summarizing results from analysisof major euchelicerate lineages generated by parsimonyanalysis of 59 genera and 202 non-molecular charactersOnly those relationships that were well supported by boot-strap analysis or consistently recovered in sensitivity anal-ysis are depicted deepest relationships within Arachnidaare effectively unresolved

Xiphosura

Eurypterida sstr

Chasmataspidida

Scorpiones

Opiliones

Pseudoscorpiones

Solifugae

Palpigradi

Trigonotarbida

Araneae

Haptopoda

Amblypygi

Thelyphonida

Schizomida

Ricinulei

Anactinotrichida

Acariformes

Acaromorpha

Uropygi

Pedipalpi

Schizotarsata nom nov

Tetrapulmonata

Pantetrapulmonata nom nov

Haplocnemata

Arachnida

Stomothecata nom nov

Eurypterida slat

Acari

PHYLOGENY OF THE ARACHNID ORDERS

223

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Table 1

Data matrix comprising 59 euchelicerata genera and 202 binary and unordered multistate characters Fossil taxaare indicated by an asterisk Ambiguity codes A

=

[345] B

=

[12] C

=

[02] D

=

[34] E

=

[01]

Weinbergina

00010-011 00 - - 000 10-0001 0300000 0- -1210-0 -

Limuloides

00010-011 01 - - 300000 0- -1210-0 -

Euproops

00010-011 01 - - 200010 0- -1010-0 - 1

Limulus

10010-0011 0000-01000 0000-00000 -0010-0000 0010000000 011110001 0100010021 0001011011 000--01101 1000200010 000110-000 00-0010-0 01110-0110 0106112110 100--00000 0000000000 0010200000 0001000000 10-0010000 0-00000111 01

Tachypleus

10010-0011 0000-01000 0000-00000 -0010-0000 0010000000 011110001 0100010021 0001011011 000--01101 1000200010 000110-000 00-0010-0 01110-0110 0106112110 100--00000 0000000000 0010000000 0001000000 10-0010000 0-00000111 01

Baltoeurypterus

00000-011 00000000 0-0 -0000-0 0010000010 0110000 0011-0020 00000 100-00001 005000-0 00-00 0-0310-0 01- 1010 --00000 0 1 100 0-

Stylonurus

00000-011 00000000 - -000-0 001000000 010000 0011-000 000 10-0001 005000-0 00-0 0-0310-0 01- 1010 000 0 1

Chasmataspis

00000-011 01 - - 1 0 1601-0 0- -1410-0 00- 11 0 0

Diploaspis

0000-01 01 - - 1 0001 0601-0 0- -0410-0 00- 11 0

Octoberaspis

0010-01 01 - - 1 00 601-0 0- -0410-0 00- 11 0 1

Prokoenenia

0000110000 0010-00000 000000000 -0000-00 001000000 000000001 1000000020 00000110 0101000001 0014000-0 000001-010 10-021201 00000010 10310-00- -010000100 1001000 000--000 001 0001100000 1011000111 00

Eukoenenia

0000110000 0010-00000 0000000000 -0000-0000 001000000 0000000001 1000000020 000001011 0101000001 00104000-0 000001-010 00--021201 000000100 110310-00- -010000100 1001000 000--000 001 0001100000 101100010- --

Plesiosiro

00110-000 00 -0 -0- 0000 0000 0000 0 11000 0A00-0 00100 0--00--1 1 00 0 0

224

J W SHULTZ

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Palaeocharinus

00000-0000 00100110 0000 -000-00 00100000 0000000 0000002 02--0 100-0001 04100-1 001000 10--110--1 110000 101B1000 0 0 10 11

Gilboarachne

00000-0000 0000110 000 -000-00 000000 0000000 0002 02--0 100-0001 04100-1 001000 0--110--1 110000 101B 100 0 0

Liphistius

00000-0000 0011000110 0000001000 -0000-0000 0110000000 0000000110 00000100220010011011 0100-00001 00115010-0 0000010010 1101000--1 1100001000 00030110120110000011 1001000000 0111310000 1000000101 1001101000 1110100111 10

Aphonopelma

00000-0000 0011000110 0000001000 -0000-0000 0110000000 0000000110 00000100220010011011 0100-00001 00115010-0 0000011010 1110000--1 1100001000 00031110120110000011 1001000000 0111310000 1000000101 1001101000 1011100111 10

Hypochilus

00000-0000 0011000110 0000001000 -0000-0000 0110000000 0000000110 0000010022 0010011011 0100-00001 00115010-0 0000011010 1110000--1 1100001000 0003011012 0110000011 1001000000 0111310000 1000000101 001101000 1110100111 10

Charinus

00100-0000 0011000110 0000100000 -00010001 0010010000 000001011 0000010023 1000101011 0101100000 01105000-0 0000010000 10--020--1 1100001000 1015010012 01100000 1011001100 0111310000 1110000101 1001100000 1110001011 10

Phrynus

00100-0000 0011000110 0000100000 -000100101 0000-10000 0001001011 0000010023 1000101011 0101100000 00105000-0 0000010000 00--020--1 1100001000 1015010012 01100000 1011001100 0111310000 1110000101 00-1100000 1110001011 10

Stenochrus

0010110000 0211000110 000000001 0000110100 0010010000 000000111 00000101230000001011 0101100000 00115000-0 0100010000 00--021211 1000001001 01410-014111000000 1111001110 0111310000 110000101 00-1100000 111000100 00

Protoschizomus

0010110000 0211000110 000000001 00001100 0010010000 0000011 000000123 00000 0101100000 0015000-0 0100010000 00--021211 100000100 050-00- -111000000 111001110 0111310000 110000101 00-1100000 1000100 00

Mastigoproctus

00110-0000 0211000110 0000100001 0000100100 000--10000 0001001111 0000010123 0000001011 0101100000 00115000-0 0100010000 00--021201 1100001001 101511001B 0111000000 1111001110 0111310000 1110000101 00-1100000 1110001010 00

Proschizomus

00100-000 02001 1 0010 0010000 0000 100 05000-0 001000 0--02121 0 00 0 0

Terpsicroton

00000-000 10 - 0- 00000 0000 0110 0 0510-0 100- --120--1 0 50-13 00 0

Poliochera

00000-000 10 -1 0- 00000 0000 0110 0 0510-0 100- --120--1 0 50-13 00 0

Table 1

Continued

PHYLOGENY OF THE ARACHNID ORDERS

225

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Cryptocellus

00000-0000 1211000100 00-00001 0000-000 0010001000 000000001 0001100020 02--000-11 0101000001 00105120-0 100000-000 00--120--1 000003 1500-014 10-000011 10010000 011120000 0010101 00-1100000 0-0000000- --

Ricinoides

00000-0000 1211000100 00-00001 0000-000 0010001000 000000001 0001100020 02--000-11 0101000001 00105120-0 100000-000 00--120--1 000003 1500-014 10-000011 10010000 011120000 0010101 00-1100000 0-0000000- --

Neocarus

00001-0000 0011000020 0000-00001 10100-000 0010010000 00000001 000111120 02--00 0101000011 0114000-0 000000-000 00--000--1 000005 00-013 010--00010 1001110 000--01000 011101 00-0100000 0-0000000- --

Siamacarus

00001-0000 0011000020 0000-00001 1100-000 0010010000 0000000 0011112 02--00 10100011 014000-0 000000-000 00--000--1 000005 00-012 110000010 101110 00--10 01101 -0100000 0-0 --

Australothyrus

01000-0100 0001000020 0000-00001 10100-000 0011000000 0000000 00001020 02--00 010-00011 01000-0 000000-100 00--000--1 000003 00-014 10--00010 1001100 000--01000 0101 00-1100000 0-000 --

Allothyrus

01000-0100 0011000020 0000-00001 10100-000 0011000000 0000000 00001020 02--00 010-00011 01000-0 000000-100 00--000--1 000003 00-00- -10--00010 101100 00--10 0101 00-1100000 0-0 --

Glyptholaspis01000-0000 0011000020 000-00001 10100-000 0011000000 000000001 0000001020 02--011011 010-00011 0110000-0 000000-000 00--000--1 00000300 00-00- -10--00000 1001100000 000--01000 000010101 00-1100000 0-0000000- --

Amblyomma00000-0000 0000-00020 0000-00001 10000-000 000--00000 000000001 0000001020 02--011011 010-00011 01100000-0 000000-000 00--000--1 00000300 100-014 010--00010 1001100000 000--01000 0000010101 00-1100000 0-0000000- --

Argas00000-0000 0000-00020 0000-00001 10000-000 000--00000 000000001 0000001020 02--011011 010-00011 00100000-0 000000-000 00--000--1 00000300 100-014 010--00010 101100000 00--10 0000010101 00-1100000 0-00000- --

Alycus000001000 000110020 000-00001 10100-000 000--00000 000000001 000010020 02--0011 000--00011 00101000-0 000000-001 10--000--1 000000 00-013 110100000 10100 00 0111 0010100000 0-000- --

Allothrombium0000000 000110020 00-00001 10000-000 000--00000 000000001 00111-0020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000004 00-0130110100000 1001001000 000--00000 010011011 0011100000 0-0000000- --

Microcaeculus00000000 000110020 00-00001 10000-000 000--00000 000000001 00111-0020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000004 0120130110100000 10010010 00--0000 01001011 001100000 0-0000000- --

Palaeacarus000011000 040110020 000-00001 10100-000 000--00000 000000001 00111-0020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000000 010 110100000 101011000 00--00 010011001 0010100000 0-00000- --

Table 1 Continued

226 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Archegozetes000011000 040110020 000-00001 10100-000 000--00000 000000001 0000000020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000000 00-00- -110100000 1001011000 000--00000 0010011001 0010100000 0-0000000- --

Cyphophthalmus0000100100 0100-00000 100-00000 -0000-1000 0010000001 1000000001 0000000020 01--010-11 0100-00001 00112000-0 000000-000 00--0013-1 000001000 010100 10--00000 1001110001 001100010 0000000 10-100110 0-0001000- --

Chileogovea0000100100 0100-00000 1010-00000 -0000-1000 0010000001 1000000001 0000000020 01--010-11 0100-00001 00112000-0 000000-000 00--0013-1 000001000 010100 10--00000 1001110001 0000 0000000 10-100110 0-0001000- --

Caddo0000100100 0000-00000 100-00000 -0000-000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 01010100--10--00000 10-1110001 000000 0000000 10-0100110 10001000- --

Leiobunum0000100100 0000-00000 1010-00000 -0000-0000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 010B01C00- -10--00000 1001110001 000--00000 1000000000 10-0100110 101001000- --

Sclerobunus0000000100 0000-00000 100-00000 -0001-000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 010001C00- -10--00000 1001110001 00--00000 1000000000 10-0100110 10001000- --

Gonyleptes0000000100 0000-00000 1010-00000 -0001-000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 010001C00- -10--00000 1001110001 010--00000 100000000 10-0100110 10001000- --

Centruroides0000100000 0100-10000 0111-00000 -0000-1000 100--00001 1000000001 0000010010 01--012110 0100-10001 00116000-0 001000-000 00--0311-1 0011001000 0005110011 1110001011 1101101000 0010110100 0000100100 10-1100111 0-10010011 00

Hadrurus0000100000 0100-10000 0111-00000 -0000-1000 100--00001 1000000001 0000010010 01--012110 0100-10001 00116000-0 001000-000 00--0311-1 0011001000 0005110011 1110001011 1101101000 0010110111 0000100100 10-1100111 0-10010011 00

Heterometrus0000100000 0100-10000 0111-00000 -0000-1000 100--00001 1000000001 0000010010 01--012110 0100-10001 00116000-0 001000-000 00--0311-1 0011001000 00051100111110001011 1101101000 001020111 0000100100 10-1100111 0-10010011 00

Prearcturus000010000 010 -0 -000- 100--001 000 000-0 00- --0311- 0 0

Palaeoscorpius00000000 0D0 -0 -000- 100--0000 00 00000 0 10-0001 06000-0 000-00 --0311- 1 01 0 0

Stoermeroscorpio00000000 00000 -0 -000- 100--0000 000 00000 0 10-0001 06000-0 000-00 --0311- 110 01 1 0

Table 1 Continued

PHYLOGENY OF THE ARACHNID ORDERS 227

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

the Ordovician to the Permian (Tollerton 1989) Phy-logenetic relationships among the eurypterids havenot been rigorously addressed (but see Beall amp Laban-deira 1990) However members of Stylonuroideaappear to be plesiomorphic in retaining relativelyunspecialized legs and were represented by Stylonu-rus (eg Clarke amp Ruedemann 1912 Waterston1979) In the remaining eurypterids the last leg ispaddle-like (Tollerton 1989) The non-stylonuroideurypterids were represented in the matrix by thewell-preserved and well-studied Baltoeurypterus tet-ragonophthalmus especially as described by Selden(1981)

Chasmataspidida Members of this fossil aquaticgroup (OrdovicianndashDevonian six genera) have a four-segmented mesosoma and a nine-segmented meta-soma (Dunlop Poschmann amp Anderson 2001 DunlopAnderson amp Braddy 2004 Dunlop 2002a) Three gen-era were included Chasmataspis from the monotypicfamily Chasmataspididae as well as Diploaspis andOctoberaspis from Diploaspididae

Haptopoda This is a fossil terrestrial group (Carbon-iferous) containing one known species Plesiosiromadeleyi The known specimens have been re-examined by Dunlop (1999)

Palpigradi The extant palpigrades (two families sixgenera ~50 spp) (Harvey 2002) are a poorly studiedgroup of small-bodied terrestrial (some semi-aquatic)arachnids The two most well-studied generaEukoenenia (Eukoeneniidae) especially E mirabilisand Prokoenenia (Prokoeneniidae) especiallyP wheeleri were included in the matrix The oneknown fossil species Paleokoenenia mordax is rela-tively recent (Pliocene) and to the extent known isfairly similar to extant palpigrades (Rowland amp Sis-som 1980) It was not included in the analysis

Trigonotarbida This is a fossil terrestrial order(Upper Silurian to Lower Permian ~ten families ~50spp) The genera coded here were drawn from twoLower Devonian Lagerstatten Gilboarachne from theGilboa Formation (New York) (Shear et al 1987) and

Proscorpius00000000 00000 -0 -000- 100--0000 010 00000 0 10-0001 06000-0 000-00 --0311- 110 01 1 0

Chthonius0000100000 0310-00120 0000-10100 -1000-0000 100--00100 0000000001 0000010010 01--012100 0E00-00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Neobisium0000100000 0310-00121 0000-10100 -1000-0010 100--00100 0000000001 0000010010 01--012100 0100-00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Feaella0000100000 0310-00120 0000-10100 -1000-0000 100--00100 0000000001 0000010010 01--012100 000--00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Chelifer0000100000 0310-00121 0000-10100 -1000-0010 100--00100 0000000001 0000010010 01--012100 000--00001 01115000-0 000000-000 00--000--1 0000021000 01700-014 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Eremocosta0000110000 0310-00121 0000-00010 -1000-0000 001010100 0000000001 00011-0000 02--010-10 1101010001 01114000-0 000000-000 00--000--1 00000200 13111013 010--10000 1001000000 000--00100 1010000100 01-1100001 0-00000110 00

Galeodes0000110000 0310-00121 0000-00010 -1000-0000 001010100 0000000001 00011-0000 02--010-10 1101010001 01114000-0 000000-000 00--000--1 00000200 13111013 010--10000 1001000000 000--00100 1010000100 011100001 0-00000110 00

Table 1 Continued

228 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Palaeocharinus from the Rhynie Chert (Scotland)(Dunlop 1994 Fayers Dunlop amp Trewin 2004)

Araneae (spiders) This is a very large order (110 fam-ilies ~3600 genera ~39 000 extant spp) (Platnick2005) represented here by one genus from each of thethree principal extant lineages (Platnick amp Gertsch1976 Coddington amp Levi 1991 Coddington 2005)Liphistius (Mesothelae) Aphonopelma (Mygalomor-phae) and Hypochilus (Araneomorphae) The sampleis small with respect to known diversity but the basalphylogeny and relevant groundplan states of the orderare well established and states derived from the rep-resented taxa are consistent with them

Amblypygi (whipspiders) This is a small extant ter-restrial order (five families 17 genera 136 spp)(Harvey 2002) and was represented in the matrix bytwo fairly well-studied genera Charinus and Phry-nus Charinus retains several features that appear tobe plesiomorphic for the order (eg adult pedal pul-villi coxal glands associated with leg 3 eversiblevesicles) Weygoldt (1996) has examined the phylog-eny of the order and has provided an important sum-mary of morphology and general biology (Weygoldt2000)

Thelyphonida or Uropygi ss (whipscorpions vinega-roons) Extant whipscorpions (16 genera 106 spp)(Harvey 2002) were represented here by one well-studied species Mastigoproctus giganteus althoughadditional information was drawn from Typopeltis andThelyphonus The morphology of the group is highlyconserved there have been no modern studies ofintraordinal phylogeny A controversial fossil speciesfrom the late Carboniferous Proschizomus (Dunlop ampHorrocks 19951996) was also included but less well-preserved Carboniferous forms were not

Schizomida This small order (two extant families~34 genera ~200 spp) (Harvey 2002) is widelyregarded as the sister group of Thelyphonida Thetaxon sample included one representative from eachextant family Protoschizomus (Protoschizomidae) andStenochrus (Hubbardiidae) especially S portoricen-sis There are three recent (Pliocene) fossil speciesall from the same locality these were not includedhere

Ricinulei The extant ricinuleids (three genera 55spp) (Harvey 2002) were represented by two generaCryptocellus and Ricinoides Detailed studies of grosscuticular anatomy and post-embryonic developmentare available for a representative of each genus (Cryp-tocellus Pittard amp Mitchell 1972 Ricinoides Legg1976) and some information on internal anatomy isavailable for Ricinoides (Millot 1945) and this hasbeen extrapolated to Cryptocellus in the matrix There

are two basic fossil types (Selden 1992) one resem-bling modern taxa and another with a unique opistho-soma that superficially resembles the closed elytra of abeetle (ie curculioids) Two fossils from the formergroup were included Terpsicroton and Poliochera asthey are reasonably well preserved and show impor-tant characters (eg two pairs of eyes or evidence ofopisthosomal diplosegmentation) not expressed inextant forms No curculioid ricinuleids were includedas they appear to offer no additional information rel-evant to resolving ordinal relationships

Opilioacariformes (= Opilioacarida Notostigmata)The opilioacariform mites (nine genera 20 spp) (Har-vey 2002) are generally regarded as plesiomorphicAcari and are fairly conserved in their morphologyThe group was represented by two species Neocarustexanus and Siamacarus withi N texanus has a typi-cal opilioacariform morphology and its external anat-omy has been particularly well studied (eg Van derHammen 1989 Klompen 2000) S withi differs frommost other opilioacariforms in having trichobothriaand three rather than two pairs of lateral eyes (the lat-ter also in Paracarus) features that are potentiallysignificant for assessing ordinal relationships Someinformation on internal morphology of Opilioacaruswas taken from With (1904) and extrapolated to Neo-carus and Siamacarus

Parasitiformes This large group of mites comprisesthree clades Holothyrida Ixodida (ticks) and Mesos-tigmata (= Gamasida) Holothyrida (three familiesfive genera lt 30 spp) (Walter amp Proctor 1999) wasrepresented here by two species Australothyrus ocel-latus and Allothyrus constrictus Ixodids (three fami-lies 22 genera ~860 spp) (Kierans amp Robbins 1999)were represented by Amblyomma and Argas andmesostigmatids (~70 families ~10 000 spp) were rep-resented by Glyptholaspis confusa

Acariformes (= Actinotrichida) This is a very largediverse group of mites that includes the Endeostig-mata Sarcoptiformes and Prostigmata Endeostig-mata is an apparently paraphyletic assemblage of~ten families A morphologically generalized speciesAlycus roseus (Bimichaeliidae) was included herewith character states extracted largely from Van derHammen (1989) Sarcoptiformes (gt 200 families) iseffectively synonymous with Oribatida with Astig-mata being a large derived lsquooribatidrsquo clade (Norton1998 Maraun et al 2004) The group was representedby an early divergent plesiomorphic genus Palaeac-arus (Palaeosomata) and a more derived and well-studied genus Archegozetes (especially A longisetusand A magnus) The Prostigmata (~120 families~7000 spp) (Walter amp Proctor 1999) was representedby two genera Allothrombium and Microcaeculus

PHYLOGENY OF THE ARACHNID ORDERS 229

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Pseudoscorpiones The pseudoscorpions (24 families425 genera ~3200 spp) (Harvey 2002) were repre-sented in the matrix by four terminal genera Chtho-nius (Chthoniidae) was drawn from the basallydivergent superfamily Chthonioidea (Harvey 1992)Feaella (Feaellidae) represented the enigmatic super-family Feaelloidea which has an unusual combinationof characters and may be an early divergent lineageclose to Chthonioidea (Harvey 1992) Neobisium (Neo-bisiidae) is morphologically intermediate between theearly divergent families and the higher families rep-resented by Chelifer (Cheliferidae)

Solifugae (= Solpugida) (sun spiders) The order issmall (~1000 spp) but the absence of modern phylo-genetic treatments precludes meaningful estimates ofgenera and families (Harvey 2002) The taxon sampleincluded two genera one from the New World Eremo-costa and one from the Old World Galeodes

Opiliones (harvestmen) Extant harvestmen (~25 fam-ilies ~500 genera ~6000 spp) are divided into twosuborders Cyphophthalmi and Phalangida Cyphoph-thalmids were represented here by Chileogovea oedi-pus (Petallidae) which is currently under study by theauthor and Cyphophthamus duricorius (Sironidae)which is probably the most well-studied member of thesuborder (eg Janczyk 1956) Phalangida includesthree major groups Laniatores Eupnoi and Dyspnoibut there is disagreement as to whether Dyspnoi is thesister group to Laniatores or to Eupnoi (Shultz ampRegier 2001 Giribet et al 2002) Laniatores was rep-resented by Sclerobunus and Gonyleptes and Eupnoiby Leiobunum and Caddo

Scorpiones (scorpions) The extant scorpions (16 fam-ilies 155 genera 1279 spp) (Fet et al 2000) wererepresented by three taxa Centruroides vittatus(Buthidae) Hadrurus arizonensis (Iuridae) and Heter-ometrus spinifer (Scorpionidae) Buthidae is widelyregarded as the sister group to other extant lineagesand similarities among the represented terminals arelikely to be ground plan features of extant scorpionsgenerally The morphology of fossil scorpions is sub-stantially more diverse (Kjellesvig-Waering 1986)but their phylogenetic relationships are unclear (butsee Jeram 1998) The fossil taxa used in this study(Proscorpius Stoermeroscorpio Palaeoscorpius Pre-arcturus) were chosen for quality of preservation andor presence of a phylogenetically significant constella-tion of characters

CHARACTER CODING

Most character states were determined from directobservation the primary literature and authoritativereviews In some cases states were assigned to termi-

nals based on observations from related species asnoted in the Appendix In the matrix state lsquo-rsquo indicatesthat the character is inapplicable because the taxonlacks a more general character For example if a taxonlacks eyes then special features of the eyes (numberposition retinal configuration etc) are inapplicable tothat taxon A lsquorsquo indicates that the state is unknown oruncertain An entry with multiple states (eg 01 345) should be treated as an ambiguity code not a poly-morphism it indicates that two or more interpreta-tions of homology are applicable and that assignmentwas established analytically by character concor-dance Although lsquo-rsquo lsquorsquo and lsquo01rsquo are analyticallyidentical when applied to binary characters theyprovide information about the empirical status of thecharacter state in specific taxa Characters are citedthroughout the text as italicized numbers in parenthe-ses and are discussed in the Appendix

PHYLOGENETIC ANALYSIS

Phylogenetic analysis of the full data matrix wasperformed using the program Tree Analysis usingNew Technologies (TNT) ver 1 (Goloboff Farris ampNixon 2000) using lsquotraditionalrsquo search based on1000 replicates using TBR branch swapping Resultswere compared to those obtaining using the ratchetalgorithm (Nixon 1999) to determine any differencedue to analytical approach Nodal support for theminimal-length topology was evaluated by boot-strap (Felsenstein 1985) and Bremer support(Bremer 1994) Bootstrap analysis was conducted inTNT and based on 1000 pseudoreplicates each anal-ysed by ten random-addition replicates using TBRbranch swapping A nexus file containing the result-ing 1000 trees was imported into PAUP ver 10(Swofford 2002) to obtain bootstrap frequenciesBremer support was determined in TNT by con-straining specific nodes in the minimal-length topol-ogy and then determining the shortest tree that didnot recover the specified clade The difference inlength between the unconstrained and constrainedminimal-length trees is the Bremer support Theeffect of homoplasy on results was explored by con-ducting implied weights analysis (Goloboff 1993) inTNT Six analyses were conducted each with con-stant of concavity (k) set to a different integer valueof 1ndash6 where 1 is weighted most severely againsthomoplasious characters Each implied weightsanalysis was conducted using lsquotraditionalrsquo searchbased on 1000 replicates using TBR branch swap-ping The same procedures were used in analysing amatrix that included only extant taxa Howevercharacters rendered uninformative by removal offossil taxa (ie 49 98 99 100 113 114 125 161)were excluded prior to analysis

230 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

COMPARISONS OF ALTERNATIVE HYPOTHESES

Phylogenetic hypotheses proposed in the recent liter-ature were also evaluated including those thatattempted to resolve arachnid phylogeny completely(Fig 2) their hypothesized subclades (Fig 3) andhypotheses that proposed only specific nodes (Figs 34) The fully resolved topologies of Weygoldt amp Paulus(1979) Van der Hammen (1989) Shultz (1990)Wheeler amp Hayashi (1998) and Giribet et al (2002)were compared with the optimal topology using theTempleton test (Templeton 1983) as implementedin PAUP Internal relationships of multisampledorders were constrained to match those of the optimaltopology unless the original authors explicitlyfavoured an alternative Node-specific hypotheseswere evaluated by determining the frequency withwhich the node was recovered in bootstrap analysisThey were also evaluated by constraining parsimonyanalysis in TNT to recover the shortest tree contain-ing each specific node and then assessing the effect onrelative tree length and overall resolution takingnote of effects on otherwise stable or well-supportedclades The entire matrix was used to assess the twocases where node-specific hypotheses involved fossiltaxa (Fig 4)

RESULTS

EXTANT TAXA

Unweighted analysis of extant taxa produced two min-imal-length trees (length 383 CI 057) (Fig 5A) withconflicts limited to relationships within Pseudoscorpi-ones Multiply sampled orders were recovered asmonophyletic with high nodal support as measured bybootstrap percentage (BP gt 80) except Acari (BP 30)There were few well-supported interordinal groupsexcept Uropygi (BP 100) Pedipalpi (BP 100) Tetra-pulmonata (BP 93) and ArachnidaXiphosura (BP100) All other interordinal relationships were recov-ered with BP less than 60

Results from implied weights analysis (IWA) indi-cated a significant effect of homoplasy on resolution ofthe deepest interordinal relationships although theinterordinal clades Stomotheca Haplocnemata Acar-omorpha Tetrapulmonata Pedipalpi and Uropygiwere recovered under all values of k IWA with k = 1 (1tree best score = 5124) recovered Megoperculata(= Palpigradi + Tetrapulmonata) and Micrura(= Megoperculata + Acaromorpha) with Micrura beingthe sister group to Haplocnemata (= Pseudoscorpiones+ Solifugae) and with Stomothecata (= Opiliones +Scorpiones) being the sister group to all other arach-nids (Fig 5B) With k = 2 (1 tree best score = 3807)and k = 3 (1 tree best score = 3054) Megoperculatawas recovered and Stomotheca remained the sister

group to other arachnids but the relative positions ofHaplocnemata and Acaromorpha changed (Fig 5C)IWA using k = 4 (1 tree best score = 2559) k = 5 (1tree best score = 2205) and k = 6 (1 tree bestscore = 1938) recovered topologies consistent with theminimal-length topology derived from unweightedcharacters (Fig 5A)

EXTANT AND FOSSIL TAXA

Unweighted analysis of extant and fossil taxa pro-duced 18 minimal-length trees (length 426 CI 056)(Fig 6A) with conflicts limited to relationships withinPseudoscorpiones Scorpiones and Araneae Multiplysampled orders were again recovered as monophyleticwith high nodal support (BP gt 80) except Acari whichwas reconstructed as diphyletic with Anactinotrichidabeing the sister group to Ricinulei (BP 33) Therewere no well-supported interordinal groupings (ieBP gt 80) except Uropygi (BP 89) Pedipalpi (BP 92)and Arachnida (BP 90)

Results from IWA indicated a significant effect ofhomoplasy on resolution of the deepest interordinalrelationships in Arachnida although StomothecataHaplocnemata Acaromorpha Cryptognomae Pantet-rapulmonata Tetrapulmonata Schizotarsata Pedi-palpi and Uropygi were recovered under all values ofk IWA with k = 1 (7 trees best score = 5675) recov-ered Megoperculata and Micrura with Micrura beingthe sister group to Haplocnemata Stomothecata wasnot recovered in the strict consensus (Fig 6B) Withk = 2 (3 trees best score = 4243) IWA recovered thetopology shown in Fig 6C with Palpigradi being thesister group to all other arachnids The strict consen-sus topologies reconstructed using k = 3ndash6 were iden-tical to those obtained with k = 2 (k = 3 3 trees bestscore = 3419 k = 4 3 trees best score = 2874 k = 53 trees best score = 2483 k = 6 3 trees bestscore = 2188)

COMPARISONS OF ALTERNATIVE HYPOTHESES

The hypothesis of Weygoldt amp Paulus (1979) (Fig 2)was 398 steps (15 steps longer than the unconstrainedminimal-length tree) Van der Hammen (1989) was400 steps (17 steps longer) Shultz (1990) was 399steps (six steps longer) Wheeler amp Hayashi (1998)was 391 steps (eight steps longer) and Giribet et al(2002) was 394 steps (11 steps longer) Templetontests conducted at the 005 significance level rejectedthe hypothesis of no difference between the optimaltopology and those of Weygoldt amp Paulus (P = 00002)Van der Hammen (P = 00011) Wheeler amp Hayashi(P = 00018) and Giribet et al (P = 00343) The Shultztopology was not significantly different (P = 01444)from the optimal topology

PHYLOGENY OF THE ARACHNID ORDERS 231

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 2 Fully resolved phylogenetic hypotheses of extant euchelicerate groups proposed in the recent literature Note thesimilarity in the topologies of the parsimony-based analyses by Shultz (1990) Wheeler amp Hayashi (1998) and Giribet et al(2002) The Giribet et al topology is based on neontological data (morphology and molecules) and the original lsquoROOTrsquo maybe an artefact from use of the highly divergent pycnogonids as an outgroup

Xip

hosu

raS

corp

ione

sO

pilio

nes

Ara

neae

Am

blyp

ygi

Uro

pygi

Ric

inul

eiA

nact

inot

richi

daS

olifu

gae

Pse

udos

corp

ione

sP

alpi

gra

diA

carif

orm

es

Xip

hosu

raS

corp

ione

sA

rane

aeA

mbl

ypyg

iU

ropy

giP

alpi

gra

diS

olifu

gae

Pse

udos

corp

ione

sO

pilio

nes

Ric

inul

eiA

cari

Lipoctena

Arachnida

Apulmonata

Holotracheata

Cryptoperculata

Acaromorpha(ldquoAcarinomorphardquo)

Tetrapulmonata(rdquoMegoperculatardquo)

LabellataHaplo-cnemata

Weygoldt amp Paulus (1979)

Labellata

Tetrapulmonata(rdquoArachnidardquo)

Haplocnemata(rdquoApatellatardquo)

Van der Hammen (1989)

Xip

hosu

raP

alpi

gra

diA

rane

aeA

mbl

ypyg

iU

ropy

giR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Pedipalpi Haplocnemata

Novogenuata

Dromopoda

Shultz (1990)

Acaro-morpha

Micrura

Xip

hosu

raP

alpi

gra

diU

ropy

giA

mbl

ypyg

iA

rane

aeR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Labellata Haplocnemata

Novogenuata

Dromopoda

Wheeler amp Hayashi (1998)

Acaro-morpha

Micrura

Xip

hosu

raA

cari

Pal

pig

radi

Ric

inul

eiA

rane

aeA

mbl

ypyg

iU

ropy

giO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Tetrapulmonata

PedipalpiHaplocnemata

Novogenuata

Dromopoda

Giribet et al (2002)

ROOT

Micrura

Crypto-gnomae

Sternocoxata

Epimerata

RostrosomataMyliosomata

Neosternata

232 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 3 Consensus trees produced by parsimony analysis of neontological data constrained to produce relationships pro-posed in the recent literature The constrained (target) node is indicated by a black dot and the taxa encompassed by theconstraint are enclosed in a box Numbers below each tree represent the number of minimal-length constrained treeslength of minimal-length trees difference in the length of the unconstrained minimal-length tree and the constrained min-imal-length tree and percentage unconstrained bootstrap trees in which the target node was recovered respectively Thesetrees indicate the effect on branch length imposed by specific hypotheses and impact of constraining target nodes on non-target nodes

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Lipoctena4 385 2 lt5

Xi

Sc

Op

Ps

So

Pa

Ur

Am

Ar

Ri

An

Ac

Labellata6 397 14 lt5

Xi

Sc

Ar

Am

Ur

Op

Pa

Ps

So

Ri

An

Ac

Apulmonata2 390 7 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

So

Ri

An

Ac

Holotracheata5 390 7 lt5

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptoperculata2 388 5 lt5

Xi

Sc

Op

Pa

So

Ps

Ar

Am

Ur

Ri

An

Ac

Rostrosomata6 384 1 25

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Neosternata3 388 5 lt5

Xi

Sc

Op

Ar

Am

Ur

Ri

An

Ps

So

Pa

Ac

Coxisternata2 388 5 lt5

Epimerata7 387 4 11

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptognomae6 384 1 24

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Micrura3 385 2 7

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Megoperculata3 384 1 28

Xi

Sc

Ps

So

Op

Pa

Ar

Am

Ur

Ri

An

Ac

Novogenuata10 388 5 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

Ri

So

An

Ac

Solifugae + Acari4 388 5 lt5

Shultz (1990) Alberti amp Peretti (2002)

Van der Hammen (1989)

Weygoldt amp Paulus (1979)

Xi

Ar

Am

Ur

So

Ps

Sc

Op

Ri

Pa

An

Ac

PHYLOGENY OF THE ARACHNID ORDERS 233

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Results from analyses of node-specific hypothesesare summarized in Figures 3 4 In most cases pro-posed relationships were recovered in fewer than 5of bootstrap pseudoreplicates The notable exceptionswere Rostrosomata (25) Cryptognomae (24) andMegoperculata (28) and the strict-consensus con-strained minimal-length tree for each was only onestep longer than the unconstrained optimal tree Inmost cases constrained topologies did not add manysteps most were less than five steps longer and nonewas greater than 14 steps longer than the un-constrained minimal-length tree It is noteworthyhowever that even constrained nodes that imposerelatively few extra steps sometimes supportedimprobable relationships or eliminated clades thatwere well supported or stable in the unconstrainedanalysis Specific examples and their implications arediscussed below

DISCUSSION

RECENT ISSUES IN ARACHNID PHYLOGENY

Available morphological evidence consistently re-solves some interordinal relationships and fails to

resolve others (Fig 1) Continued progress depends onthe ability of arachnologists to discover new charac-ters and to assess the evidence critically Whilereviewing the recent arachnological literature severalaspects of phylogenetic practice emerged that seemedcounterproductive to both the perception and theactual rate of progress toward resolving arachnid phy-logeny Some of these are summarized here withspecific examples given in the remainder of theDiscussion and in the Appendix

There is a tendency to portray arachnid ordinal phy-logeny as more poorly resolved and contentious than isactually the case (Coddington et al 2004) Phyloge-netic hypotheses generated during different historicalperiods and using differing standards of evidence areoften cited as examples of current disagreement (egSelden 1993 Dunlop 1996 Selden amp Dunlop 1998Wheeler amp Hayashi 1998) In fact recent parsimony-based analyses of morphology have tended to convergeon topologies with internal structures congruent withthose found here (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) (Fig 2) Arachnidphylogeny is not fully resolved especially at its deep-est levels (Fig 1) but this does not mean that allaspects of arachnid phylogeny are controversial orpoorly supported by the available evidence

Matrices are sometimes constructed by uncriticallsquorecyclingrsquo of erroneous or problematic charactersbased on diverse secondary sources (Jenner 2001)Conclusions derived from mixtures of valid invalidspeculative and redundant characters are sometimesportrayed as the phylogenetic signal provided by mor-phology Data recycling can perpetuate errors (see 1320 39 52 136 140 144 152 169 171ndash174) legiti-mize speculations (see 13 32 63 95) or create dupli-cate or non-independent characters (see 13 30 70 77172) The lsquolateral organrsquo is a notable exampleYoshikura (1975) equated the embryonicearly post-embryonic lsquolateral organsrsquo of Amblypygi Thelyphonidaand Solifugae with the dissimilar lsquolateral organrsquo ofXiphosura but failed to note the very similar Cla-paregravede organ of Acariformes (see 173 174) This cod-ing was recycled by Wheeler amp Hayashi (1998) andthen by Giribet et al (2002) Error is probably inevi-table when assembling a large matrix from morphol-ogy including the one presented here but this can beminimized by making original observations and byconsulting primary sources

Some workers advocate weighting characters a pri-ori on the basis of structural or functional complexity(eg Kraus 1998 Dunlop amp Braddy 2001) and dismissphylogenetic conclusions derived from equal-weightsparsimony In short these workers criticize parsimonyfor emphasizing data quantity over data quality Thiscriticism ignores the intense debate in systematic biol-ogy that eventually led to widespread adoption of

Figure 4 Consensus trees produced by parsimony analy-sis of the full data matrix constrained to produce relation-ships proposed in the recent literature See legend toFigure 3 for details

Xi

Eu

Sc

Op

Pa

Ac

Ri

An

So

Ps

Tr

Ar

Ha

Am

UrEurypterida slat + Scorpiones12 435 9 lt5

Xi

Eu

Sc

Op

Pa

Ac

An

So

Ps

Ri

Tr

Ar

Ha

Am

UrTrigonotarbida+ Ricinulei28 434 8 lt5

Dunlop amp Braddy (2002)et al

Dunlop (1996)

234 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 5 Results from analysis of neontological data A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology AFor B and C relationships within terminal clades are the same as those shown in A

LimulusTachypleusProkoeneniaEukoeneniaCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAllothyrusAustralothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesLiphistiusAphonopelmaHypochilusCharinusPhrynusStenochrusProtoschizomusMastigoproctusCyphophthalmusChileogoveaCaddoLeiobunumSclerobunusGonyleptesCentruroidesHadrurusHeterometrusChthoniusFeaellaNeobisiumCheliferEremocostaGaleodes

10031

10018451

501561

733

723

944

633

10010

832

901

943

903

522

341

1007

1008

Acaromorpha

Haplocnemata

10012

Stomothecata

Opiliones

Scorpiones

Pseudo-scorpiones

Solifugae

Acariformes

Parasitiformes

Opilioacariformes

Ricinulei

Xiphosura

Palpigradi

Thelyphonida

Schizomida

Araneae

Amblypygi

10011

936853

100141005

1009

Tetrapulmonata Pedipalpi

Uropygi

692

522572

968

331

603

81

511

161

101

Acari

Dromopoda

Arachnida

A

Xiphosura

Stomothecata

Haplocnemata

Acaromorpha

Palpigradi

TetrapulmonataMicrura

Megoperculata

BXiphosura

Stomothecata

Acaromorpha

Haplocnemata

Palpigradi

TetrapulmonataMegoperculata

C

PHYLOGENY OF THE ARACHNID ORDERS 235

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 6 Results from analysis of the full data matrix A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology CFor B and C relationships within terminal clades are the same as those shown in A

WeinberginaLimuloidesEuproopsTachypleusLimulusStylonurusBaltoeurypterusChasmataspisDiploaspisOctoberaspisCyphophthamusChileogoveaCaddoLeiobunumSclerobunusGonyleptesProscorpiusPalaeoscorpiusStoermeroscorpioPrearcturusCentruroidesHadrurusHeterometrusProkoeneniaEukoeneniaTerpsicrotonPoliocheraCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAustralothyrusAllothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesEremocostaGaleodesChthoniusFeaellaNeobisiumCheliferGilboarachnePalaeocharinusLiphistiusAphonopelmaHypochilusPlesiosiroCharinusPhrynusStenochrusProtoschizomusProschizomusMastigoproctus

631

893

923

984

984

671

341

671

986

10011

423

974

Xiphosura

Eurypterida sstr

Chasmataspidida

Opiliones

Scorpiones

Palpigradi

Ricinulei

Opilioacariformes

Parasitiformes

Acariformes

Solifugae

Pseudoscorpiones

Trigonotarbida

Araneae

Haptopoda

Amblypygi

Schizomida

Thelyphonida

842

251331

361311

552

832

882

925

913

573

561291

443

976

541

954

743644

732521

571

331

833

663591

966

573

71

221

905

832

351

Uropygi

Pedipalpi

Haplocnemata

Tetrapulmonata

Anactinotrichida

Cryptognomae

Acaromorpha

Stomothecata

Pantetrapulmonata

Arachnida

Eurypterida slat

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatProscorpiusStoermeroscorpioPalaeoscorpiusPrearcturusCentruroidesHadrurusHeterometrusOpilionesHaplocnemataRicinuleiAnactinotrichidaAcariformesPalpigradiPantetrapulmonata

Megoperculata

Micrura

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatPalpigradiStomothecataHaplocnemataPantetrapulmonataAcaromorpha

A

B

C

Schizotarsata

236 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

parsimony and instead advocates a return to thespeculative and subjective approaches of the late 19thand early 20th century that once threatened thescientific legitimacy of the discipline (Bowler 1996)Furthermore it misrepresents the properties of parsi-mony-based analyses Specifically characters withfunctional significance are readily encompassed byparsimony analyses and in fact characters derivedfrom locomotor systems (eg 46ndash94) have played animportant role in developing current ideas aboutarachnid phylogeny (eg Shultz 1989 1991) In addi-tion lsquocomplex charactersrsquo can be viewed as compositesof several characters such that morphological com-plexity is effectively weighted by the number of inde-pendent lsquosubcharactersrsquo it contains For example thelsquosucking stomachrsquo once considered a synapomorphy ofLabellata (= Araneae + Amblypygi Fig 3) is codedhere as a composite of three characters (199ndash201) TheLabellata hypothesis was not corroborated in thepresent analysis (Figs 1 3ndash5) and in fact was highlydisfavoured (Fig 3) but this result cannot be dis-missed as a failure to acknowledge the complexity ofthe character

Some workers support specific (target) clades withone or more similarities without exploring the impacton overall homoplasy or relationships among non-target clades Each character offers its own phyloge-netic hypothesis which may or may not be consistentwith relationships implied by other characters It isexceedingly rare for all characters to be perfectly com-patible in the phylogenetic hypotheses they supportand criteria such as parsimony have been developed todiscover those hypotheses that minimize the conflict-ing phylogenetic signals of different characters Eventhough character conflict (homoplasy) is a virtuallyinescapable phenomenon in comparative biology it isnot uncommon for workers to discover one or morecharacters and to promote their phylogeneticimplications over alternative hypotheses even thosethat otherwise appear to be well supported Thisapproach may have value in highlighting new dataand perspectives but accomplishes this by promotingthe erroneous impression that all aspects of arachnidphylogeny are so tenuous that a single character canfalsify even well-supported hypotheses Several exam-ples of this approach have appeared in the recentarachnological literature

For example Alberti amp Peretti (2002) argued thataflagellate spermatozoa (163) are a compelling syna-pomorphy for a Solifugae + Acari clade and dismissedsome of the characters that support Solifugae + Pseu-doscorpiones as having lsquodebatable valuersquo Yet theirproposal rejects two hypotheses that are consistentlyrecovered in recent phylogenetic analyses that isHaplocnemata (= Solifugae + Pseudoscorpiones) andAcaromorpha (= Acari + Ricinulei) (Figs 1 2) Rejec-

tion of Haplocnemata would require its presumed syn-apomorphies to be reinterpreted as homoplasiesincluding features of the chelicerae (18 19 20) pre-oral chamber (13 32) legs (12 48) and respiratorysystem (126) The same reason applies to Acaromor-pha and its synapomorphies It is noteworthy that aSolifugae + Acari clade was recovered in the presentanalysis in fewer than 5 of bootstrap pseudorepli-cates and that analyses constrained to recover thisclade were five steps longer than the minimal-lengthtree and favoured a problematic clade unitingRicinulei and Pseudoscorpiones (Fig 3)

In another example Dunlop (1996) proposed a closerelationship between Trigonotarbida and Ricinuleibased on two-segmented chelicerae (18) prosomandashopisthosoma coupling mechanism (96) diplotergites(100 101) and longitudinally divided opisthosomaltergites (115) However phylogenetic analyses con-strained to recover this relationship required eightadditional steps eliminated support for Acari and nec-essarily rejected Acaromorpha and forced its synapo-morphies to be reinterpreted as homoplasies Alberti(2005) proposed an interesting hypothesis for the evo-lution of male gonads in tetrapulmonates (153) butchose to accept Labellata (= Araneae + Amblypygi) indeveloping his argument over the much more well-supported Pedipalpi (= Amblypygi + Uropygi) (Shultz1999) a phylogenetic reconfiguration that was recov-ered here in fewer than 5 of bootstrap pseudorepli-cates increased tree length by a minimum of 14 stepsand resulted in the collapse of Stomothecata and Hap-locnemata (Fig 3) Many other examples can be cited

Promoting or defending a specific phylogenetichypothesis via lists of compatible synapomorphies is acommon but problematic approach By restrictingattention to the states of specific characters at one ortwo target nodes one can easily overlook the unin-tended impact of the hypothesis on phylogenetic sig-nal elsewhere and its effect on non-target clades Anode supported by a long list of synapomorphies mayseem convincing taken in isolation but may becomeless acceptable when its full phylogenetic implicationsare explored

SUMMARY OF INTERORDINAL ARACHNID CLADES RECOVERED IN THIS ANALYSIS

Arachnida Lamarck 1801Analyses consistently recovered Arachnida as a mono-phyletic group with high bootstrap support (Figs 1 45) Possible synapomorphies include the loss of thecarapacal pleural doublure (9) cardiac lobe (10) pedalgnathobases (52) and moveable endites (53) and thegain of aerial respiration (120) and an anteriorly oranteroventrally directed mouth (185) Some tradi-tional synapomorphies such as slit sensilla (142) and

PHYLOGENY OF THE ARACHNID ORDERS 237

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

fluid feeding (184) may have appeared later in arach-nid evolution but this can only be decided once theinternal phylogeny of Arachnida has been determined

Some workers regard many proposed arachnid syn-apomorphies as adaptations to terrestrial life andthereby link the hypothesis of arachnid monophyly tothe hypothesis of a single ancestral aquatic-to-terres-trial transition and arachnid polyphyly to multipletransitions (eg Selden amp Jeram 1989 Dunlop 1997Dunlop amp Webster 1999 Dunlop amp Braddy 2001)Given this line of reasoning the existence of appar-ently aquatic scorpions in the fossil record (Kjellesvig-Waering 1986 Jeram 1998) and the inference thatterrestrialization occurred late in scorpion evolutionthese workers conclude that character states support-ing arachnid monophyly are actually convergencesand do not necessarily support arachnid monophyly

However this approach to assessing phylogenetichypotheses is founded on overly simplistic assump-tions such as the ability of the investigator to discrim-inate unerringly between characters that existexclusively in aquatic organisms (including fossils)from those that occur exclusively in terrestrial organ-isms There also appears to be an assumption thathomoplasy can be generated through parallelism (iemultiple aquatic-to-terrestrial events) but not throughterrestrial-to-aquatic reversals Furthermore anassumed dichotomy between exclusively aquatic andterrestrial life histories in ancestral arachnids is sim-plistic as illustrated by the amphibious life cycles ofbasal vertebrates and pterygote hexapods In factthese examples demonstrate that there is no neces-sary inconsistency in basing a hypothesis of arachnidmonophyly on the derived terrestrial features of anamphibious ancestor whose descendants thencompleted terrestrialization once or several timesindependently or even returned to a fully aquaticexistence Workers who link the frequency and direc-tion of aquaticndashterrestrial transitions to the assess-ment of arachnid phylogeny do so by endowingthemselves with substantially greater insight thanseems prudent by ignoring the huge gaps in ourunderstanding of early arachnid evolution and bydenying to arachnids the evolutionary complexityknown to exist in other groups

Several palaeontologists have been particularlyactive during the past decade in proposing new char-acters with the stated goal of removing Scorpionesfrom Arachnida and erecting a Scorpiones + Eurypter-ida clade (Braddy amp Dunlop 1997 Dunlop amp Braddy1997 Dunlop 1998 Braddy et al 1999 Dunlop ampWebster 1999) Dunlop amp Braddy (2001) recentlysummarized this evidence and conducted aparsimony-based analysis of Xiphosura EurypteridaScorpiones Opiliones and Tetrapulmonata (but notHaplocnemata Acaromorpha or Palpigradi) using 33

morphological characters Their analysis produced atopology congruent with those generated here (Figs 15 6) including recovery of Stomothecata (= Scorpiones+ Opiliones) However they rejected this result as aproduct of lsquoempirical cladisticsrsquo because it gives thesame weight to prosomal characters that supportarachnid monophyly and to selected opisthosomalcharacters that support their favoured Eurypterida +Scorpiones clade namely a five-segmented metasoma(116) suppression of opisthosomal tergite 1 (95) lossof lamellate respiratory organs on the postgenitalsomite (122) Kiemenplatten (125) loss of respiratorylamellae on the genital segment (121) and a lsquonon-stainingrsquo exocuticle (but see Grainge amp Pearson 1966for evidence of this in Opiliones see Appendix for com-ments on the other characters)

A Eurypterida + Scorpiones clade was not favouredin the present analysis (Fig 6) the strict consensus ofminimal-length trees constrained to recover this clade(Fig 4) is nine steps longer than the tree recovered byanalysis without this constraint It is also noteworthythat Opiliones was consistently reconstructed as thesister group to Eurypterida + Scorpiones in the mini-mal-length constrained trees a provocative result thatwas probably unintended and unanticipated by sup-porters of the Eurypterida + Scorpiones Given thatDunlop amp Braddy (2001) reject equal weights parsi-mony as an arbiter of phylogenetic hypotheses theywould presumably dismiss these results as irrelevantto their argument just as they dismissed their ownparsimony-based results However if the Eurypterida+ Scorpiones hypothesis is to be credible it must beopen to evaluation and potential falsification usingobjective criteria and the subjective or intuitive a pri-ori weighting of characters advocated by Dunlop ampBraddy clearly does not qualify At present it isunclear how one would objectively evaluate Dunlop ampBraddyrsquos proposal with criteria compatible with thoseused in its original formulation For now the Euryp-terida + Scorpiones concept advocated by Dunlop ampBraddy may persist outside the mainstream of mod-ern systematic practice but it is increasingly problem-atic within it

Stomothecata nomnovOpiliones and Scorpiones were consistently recoveredas a monophyletic group The proposed name acknowl-edges a unique preoral chamber the stomothecaformed by coxapophyses of the palp and leg 1 (50)often with an auxiliary role played by the coxapophy-sis of leg 2 (51) In addition the epistome appears tohave been modified for adduction of the palpal coxaeThe lateral walls of the epistome are fused to themedial surfaces of the palpal coxae and the epistomallumen is spanned by a transverse muscle (188) which

238 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

apparently adducts the palpal coxae thereby constrict-ing the stomothecal chamber Scorpions and opilionsare also unique in having a pair of large epistomalarms projecting rearward into the prosoma andattaching to the endosternite (189) The epistomalarms provide attachment sites for pharyngeal dilatormuscles (196) and extrinsic muscles of anterior proso-mal appendages (eg 37) The chelicera is equippedwith a muscle that arises on the carapace and insertson the ventral margin of the second cheliceral article(23) There is an anteriorly placed genital opening(155)

Several notable similarities were found whilereviewing the literature but information from otherarachnid groups was considered too incomplete toallow them to be included in the matrix For exampleboth orders have apparent haemocytopoeitic organsassociated with major nerves of the anterior opistho-soma These are termed supraneural organs in scorpi-ons (Farley 1999) and perineural organs in opilions(Kaumlstner 1935) Haemocytes develop in the cardiacwall in spiders (Seitz 1972) and perhaps amblypygids(Weygoldt 2000) but haemocytopoeitic organs areunknown in most other arachnid groups Germ cellsdifferentiate very early during embryogenesis in bothscorpions and opilions relative to spiders (Moritz1957 Anderson 1973) Additional research is requiredto determine the phylogenetic utility of thesecharacters

Haplocnemata Boumlrner 1904Several workers recognize a close phylogenetic rela-tionship between Pseudoscorpiones and Solifugae(Weygoldt amp Paulus 1979 Van der Hammen 1989Shultz 1990 Wheeler amp Hayashi 1998 Giribetet al 2002) and this was consistently recovered inthe present analysis Haplocnemata is supported byseveral synapomorphies including (i) a feeding com-plex formed by two-segmented chelicerae (18) with aventrolateral intrinsic articulation (19) and perhapsa dorsolateral articulation with the carapace (20)and (ii) a rostrosoma (32) a preoral apparatus formedby an anteriorly projecting epistome affixed dorsallyto enlarged palpal coxae and bordered by lateral pal-pal projections The epistome forms the dorsal compo-nent of a lsquobeakrsquo with the ventral part formed by amidventral sternapophysis (13) fused between thepalpal coxae The locomotor apparatus consists ofcoxae that meet along the ventral midline (12) elon-gate femur-like patellae (48) and an apotele in theadult modified as an eversible empodium or pulvillus(92) The respiratory system is formed entirely (Pseu-doscorpiones) or in part (Solifugae) by paired trachealstigmata opening on opisthosomal somites 3 and 4(126)

Acaromorpha Dubinin 1957A clade comprising Ricinulei and Acari is consistentlyrecovered in the present analysis It is united here bytwo unique and seven homoplasious synapomorphiesThese include a gnathosoma (31) defined in part bymedial fusion of the palpal coxae (30) although pres-ence of a gnathosoma in Ricinulei is debatable as dis-cussed in the Appendix Acaromorphs also have aunique post-embryonic development consisting of ahexapodal larva and up to three octopodal nymphalinstars (176) The pedal patellandashtibia joints are formedby a bicondylar dorsal hinge rather than a single mid-dorsal condyle (69) all postcheliceral segmental gan-glia are unified in the subesophageal ganglion (130)and a postcerebral pharynx is absent (192) The groupis also tentatively united by presence of differentiatedpedal basi- and telofemora (63 64) and the absence ofa ventral (sternal) pharyngeal dilator muscle (199)but these may be symplesiomorphies erroneouslyreconstructed as a synapomorphies

The internal phylogenetic structure of Acaromorphais controversial with many recent workers favouringa monophyletic Acari (Weygoldt amp Paulus 1979Shultz 1990 argued most thoroughly by Lindquist1984) and others advocating a diphyletic Acari withAcariformes being the sister group to Cryptognomae(= Ricinulei + Anactinotrichida) (especially Van derHammen 1979 1989) (Fig 3) Acari was recovered asmonophyletic when fossils were excluded (Fig 2) butnot when they were included (Fig 3) A lsquomite-centredrsquosurvey of arachnid characters may be needed if mor-phology is to offer a compelling solution to the internalphylogeny of Acari and its placement within Arach-nida These issues are not resolved by the presentanalysis

Pantetrapulmonata nom novPantetrapulmonata includes the extinct orders Trigo-notarbida Haptopoda and the extant orders AraneaeAmblypygi Schizomida and Thelyphonida The cladeis united by cheliceral structure (18 19) a megoper-culum (106) booklungs on the genital and first post-genital somites (121 122) and enlargement of theepipharyngeal sclerite (192) It is important to notehowever that most of these characters were coded asuncertain in Plesiosiro (Haptopoda) Aside from theplacement of Haptopoda the monophyly of Trigono-tarbida and the extant orders was anticipated byShear et al (1987) and is generally regarded as amonophyletic group

Tetrapulmonata Shultz 1990Tetrapulmonata was originally proposed on the basisof neontological analyses and encompassed Araneae

PHYLOGENY OF THE ARACHNID ORDERS 239

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Amblypygi Schizomida and Thelyphonida (Shultz1990) and was also recovered here in analysis of neon-tological data (Fig 5) However results from analysisof all taxa required that the Tetrapulmonata conceptbe expanded to include the fossil order HaptopodaFeatures uniting this group are problematic howeveras many states in Haptopoda were coded as uncertain

Schizotarsata nomnovHaptopoda and Pedipalpi (= Amblypygi + Schizomida+ Thelyphonida) are united here in a group named forpossession of divided pedal telotarsi (84) Synapomor-phies include a pointed anterior carapacal margin (3)and elongation of leg 1 (46) As already noted theplacement of Haptopoda should probably be regardedas tentative because the state of many characters inthis group was coded as unknown The placement ofHaptopoda as the sister group to Pedipalpi was antic-ipated by Dunlop (1999 2002c)

Pedipalpi Boumlrner 1904Pedipalpi encompasses Amblypygi and Uropygi(= Thelyphonida + Schizomida) Although widely rec-ognized in the past Pedipalpi was set aside in the mid-20th century in favour of Labellata (= Amblypygi +Araneae) based largely on the presumed synapomor-phies of a narrow prosomandashopisthosoma juncture (97)and sucking stomach (199ndash201) The situation beganto reverse when Shear et al (1987) again highlightedsimilarities of Amblypygi and Uropygi such as rapto-rial palps (35) and antenniform leg 1 (46) Shultz(1989 1990 1999) subsequently described numerousderived similarities including asymmetrical flexormuscles at the pedal femurndashpatella joint (70) threetelotarsomeres on legs 2ndash4 (85) and modification ofthe palpal coxae to provide support for extrinsic pha-ryngeal muscles (197) Pedipalpi has emerged as oneof the most well-supported interordinal relationshipsin Arachnida

Uropygi Thorell 1882Uropygi has long been accepted as a monophyleticunion of Thelyphonida and Schizomida Synapomor-phies include a unique mating behaviour (159) fusedpalpal coxae (30) 2-1-1-1 arrangement of tibial tricho-bothria (144) posterior defensive glands (102) elon-gated patella of leg 1 (68) and many others

ACKNOWLEDGEMENTS

I thank Nikolaj Scharff and three anonymous review-ers for useful comments This work was supported bythe Maryland Agricultural Experiment Station the

General Research Board of the University of Mary-land and National Science Foundation Grant 0416628to Susan Masta

REFERENCES

Abd el-Wahab A 1952 Notes on the morphology of the scor-pion Buthus quinquestriatus (HE) Publications de lrsquoInsti-tut Fouad 1er du Desert 3 1ndash129

Alberti G 1979 Licht- und elektronenmikroskopische Unter-suchungen an Coxaldruumlsen von Walzenspinnen (ArachnidaSolifugae) Zoologischer Anzeiger 203 48ndash64

Alberti G 1995 Comparative spermatology of Cheliceratareview and perspective Memoires du Museum NationaldrsquoHistoire Naturelle 166 203ndash230

Alberti G 2005 Double spermatogenesis in ChelicerataJournal of Morphology 266 281ndash297

Alberti G Coons LB 1999 Acari mites In Harrison FWFoelix RF eds Microscopic anatomy of invertebratesVol 8C Chelicerate Arthropoda New York Wiley Liss515ndash1215

Alberti G Peretti AV 2002 Fine structure of male genitalsystem and sperm in Solifugae does not support a sister-group relationship with Pseudoscorpiones (Arachnida)Journal of Arachnology 30 268ndash274

Alexander AJ 1967 Problems of limb extension in the scor-pion Opisthophthalmus latimanus Koch Transactions of theRoyal Society of South Africa 37 165ndash181

Anderson DT 1973 Embryology and phylogeny in annelidsand arthropods Oxford Pergamon Press

Anderson LI Selden PA 1997 Opisthosomal fusion andphylogeny of Palaeozoic Xiphosura Lethaia 30 19ndash31

Beall BS Labandeira CC 1990 Macroevolutionary patternsof the Chelicerata and Tracheata In Culver SJ ed Arthro-pod paleobiology Short Courses in Paleontology No 3 Lon-don The Paleontological Society 257ndash284

Bernard HM 1896 The comparative morphology of the Gale-odidae Transactions of the Linnean Society of London Sec-ond Series 6 305ndash417

Boumlrner C 1902a Arachnologische Studien (II und III) Zool-ogischer Anzeiger 25 433ndash466

Boumlrner C 1902b Arachnologische Studien IV Die Genitalor-gane der Pedipalpen Zoologischer Anzeiger 26 81ndash92

Boumlrner C 1904 Beitraumlge zur Morphologie der Arthropoden IEin Beitrag zur Kenntnis der Pedipalpen Zoologica (Stut-tgart) 42 1ndash174

Bowler PJ 1996 Lifersquos splendid drama Chicago Universityof Chicago Press

Braddy SJ Alridge RJ Gabbott SE Theron JN 1999Lamellate book-gills in a late Ordovician eurypterid from theSoom Shale South Africa support for a eurypterid-scorpionclade Lethaia 32 72ndash74

Braddy SJ Dunlop JA 1997 The functional morphology ofmating in the Silurian eurypterid Baltoeurypterus tet-ragonophthalmus (Fischer 1839) Zoological Journal of theLinnean Society 121 435ndash461

Brauer A 1895 Beitrage zur Kenntnis der Entwicklungsge-schichte des Skorpiones II Zeitschrift fuumlr WissenschaftlicheZoologie 59 351ndash433

240 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bremer K 1994 Branch support and tree stability Cladistics10 295ndash304

Bristowe WS 1932 The liphistiid spiders Proceedings of theZoological Society of London Part 4 1016ndash1057

Butt AG Taylor HH 1991 The function of spider coxalorgans effects of feeding and salt-loading on Porrhotheleantipodiana (Mygalomorpha Dipluridae) Journal of Exper-imental Biology 158 439ndash461

Buxton BH 1913 Coxal glands of arachnids ZoologischeJahrbucher (Suppl XIV) 16ndash50

Buxton BH 1917 Notes on the anatomy of arachnids Jour-nal of Morphology 29 1ndash31

Camin JH Clark GM Bourdeau FG 1956 The palpallsquotined setarsquo in the Mesostigmata a homologue of the palpalclaw in the Onychopalpida (Acarina) In Becker EC ed Pro-ceedings of the Tenth International Congress of EntomologyOttawa Mortimer 903ndash908

Chamberlin JC 1931 The arachnid order ChelonethidaStanford University Stanford University Press

Chapman RF 1998 The insects structure and function 4thedn Cambridge Cambridge University Press

Clarke KU 1979 Visceral anatomy and arthropod phylogenyIn Gupta AP ed Arthropod phylogeny New York Van Nos-trand Reinhold 467ndash549

Clarke J 1984 On the relationship between structure andfunction in the leg joints of Heteropoda venatoria (L) (Ara-neae Eusparassidae) Bulletin of the British ArachnologicalSociety 6 181ndash192

Clarke J 1986 The comparative functional morphology of theleg joints and muscles of five spiders Bulletin of the BritishArachnological Society 7 37ndash47

Clarke JM Ruedemann R 1912 The Eurypterida of NewYork New York State Museum Memoir 14 (1ndash2) 1ndash628

Coddington JA 2005 Phylogeny and classification of spi-ders In Ubick D Paquin P Cushing PE Roth V eds Spi-ders of North America an identification manual AmericanArachnological Society 18ndash24

Coddington JA Giribet G Harvey MS Prendini LWalter DE 2004 Arachnida In Cracraft J Donoghue MJeds Assembling the tree of life Oxford Oxford UniversityPress 269ndash318

Coddington JA Levi HW 1991 Systematics and evolutionof spiders (Araneae) Annual Review of Ecology and System-atics 22 565ndash592

Coineau Y Van der Hammen L 1979 The postembryonicdevelopment of Opilioacarida with notes on new taxa and ona general model for the evolution In Piffle E ed Proceed-ings of the 4th International Congress of Acarology Budap-est Akadeacutemiai Kiadoacute 437ndash441

Cokendolpher JC Reddell JR 1992 Revision of the Pro-toschizomidae (Arachnida Schizomida) with notes on thephylogeny of the order Texas Memorial Museum Speleolog-ical Monographs 3 31ndash74

Condeacute B 1991a Le dimorphisme sexuel des Palpigrades Bul-letin de la Socieacuteteacute Neuchacircteloise des Sciences Naturelles 11667ndash73

Condeacute B 1991b Lrsquoaire genitale male des Palpigrades Bulle-tin de la Socieacuteteacute Europeenne drsquoArachnologie 1990 64ndash69

Dearden PK Donly C Grbicccc M 2002 Expression of pair-rule gene homologues in a chelicerate early patterning ofthe two-spotted spider mite Tetranychus urticae Develop-ment 129 5461ndash5472

Delle Cave L 1975 The ldquocleaning brushrdquo in Damon diademaSimon 1876 and Phrynichus jayakari Pocock 1894 (Ambly-pygi Arachnida) Monitore Zoologico Italiano (NS) 9 175ndash184

Dunlop JA 1994 Filtration mechanisms in the mouthparts oftetrapulmonate arachnids (Trigonotarbida Araneae Ambly-pygi Uropygi Schizomida) Bulletin of the British Arachno-logical Society 9 267ndash273

Dunlop JA 1996 Evidence for a sister group relationshipbetween Ricinulei and Trigonotarbida Bulletin of the BritishArachnological Society 10 193ndash204

Dunlop JA 1997 Palaeozoic arachnids and their significancefor arachnid phylogeny In Zabka M ed Proceedings of the16th European colloquium of arachnology Siedlce PolandWyzsza Szkola Rolnicko-Pedagogiczna 65ndash82

Dunlop JA 1998 The origins of tetrapulmonate book lungsand their significance for chelicerate phylogeny In SeldenPA ed Proceedings of the 17th European Colloquium ofArachnology Edinburgh 1997 Burnham Beeches BritishArachnological Society 9ndash16

Dunlop JA 1999 A redescription of the Carboniferous arach-nid Plesiosiro madeleyi Pocock 1911 (Arachnida Hapto-poda) Transactions of the Royal Society of Edinburgh EarthSciences 90 29ndash47

Dunlop JA 2000 The epistomo-labral plate and lateral lips insolifuges pseudoscorpions and mites Ekoloacutegia (Bratislava)19 67ndash78

Dunlop JA 2002a Arthropods from the Lower Devonian Sev-ernaya Zemlya Formation of October Revolution Island(Russia) Geodiversitas 24 349ndash379

Dunlop JA 2002b Character states and evolution of the che-licerate claws In Toft S Scharff N eds European Arach-nology 2000 Aarhus Aarhus University Press 345ndash354

Dunlop JA 2002c Phylogeny of Chelicerata In LlorenteBousquets J Morrone JJ eds Biodiversidad Taxonomiacutea yBiogeografiacutea de Arthroacutepodos de Meacutexico Hacia Una Siacutentesisde Su Conocimiento Vol III Prensas de Ciencias Univer-sidad Nacional Autonoma de Meacutexico 117ndash141

Dunlop JA Anderson LI Braddy SJ 2004 A redescriptionof Chasmataspis laurencii Caster amp Brooks 1956 (Chelicer-ata Chasmataspidida) from the Middle Ordovician of Ten-nessee USA with remarks on chasmataspid phylogenyTransactions of the Royal Society of Edinburgh Earth Sci-ences 94 207ndash225

Dunlop JA Braddy SJ 1997 Slit-like structures on theprosomal appendages of the eurypterid BaltoeurypterusNeues Jahrbuch fuumlr Geologie und Palaumlontologie-Monatshefte1997 31ndash38

Dunlop JA Braddy SJ 2001 Scorpions and their sister-group relationships In Fet V Selden PA eds Scorpions2001 in Memoriam Gary A Polis London British Arachno-logical Society 1ndash24

Dunlop JA Horrocks CA 19951996 A new Upper Carbon-iferous whip scorpion (Arachnida Uropygi Thelyphonida)

PHYLOGENY OF THE ARACHNID ORDERS 241

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

with a revision of the British Carboniferous Uropygi Zoolo-gischer Anzeiger 234 293ndash306

Dunlop JA Poschmann M Anderson L 2001 On theEmsian (early Devonian) arthropods of the Rhenish SlateMountains 3 The chasmataspidid Diploaspis Palaumlon-tolgische Zeitschrift 75 253ndash269

Dunlop JA Webster M 1999 Fossil evidence terrestrializa-tion and arachnid phylogeny Journal of Arachnology 27 86ndash93

Edgecombe GD Wilson GDF Colgan DJ Gray MR Cas-sis G 2000 Arthropod cladistics combined analysis of his-tone H3 and U2 snRNA sequences and morphologyCladistics 16 155ndash203

Evans GO 1992 Principles of acarology Wallingford UKCABI

Fahrenbach WH 1999 Merostomata In Harrison FW Foe-lix RF eds Microscopic anatomy of invertebrates Vol 8AChelicerate arthropods New York Wiley-Liss 21ndash115

Farley RD 1999 Scorpiones In Harrison FW Foelix RF edsMicroscopic anatomy of invertebrates Vol 8A Cheliceratearthropods New York Wiley-Liss 117ndash222

Farley RD 2005 Developmental changes in the embryo pro-nymph and first molt of the scorpion Centruroides vittatus(Scorpiones Buthidae) Journal of Morphology 265 1ndash27

Fayers SR Dunlop JA Trewin NH 2004 A new earlyDevonian trigonotarbid arachnid from the Windyfield ChertRhynie Scotland Journal of Systematic Palaeontology 2269ndash284

Felsenstein J 1985 Confidence limits on phylogenies anapproach using the bootstrap Evolution 39 783ndash791

Fet V Sissom WD Lowe G Braunwalder ME 2000 Cat-alog of the scorpions of the world (1758ndash1998) New YorkNew York Entomological Society

Firstman B 1954 The central nervous system musculatureand segmentation of the cephalothorax of a tarantula (Eu-rypelma californicum Ausserer) Microentomology 87 1ndash40

Firstman B 1973 The relationship of the chelicerate arterialsystem to the evolution of the endosternite Journal ofArachnology 1 1ndash54

Foelix RF 1985 Mechano- and chemoreceptive sensilla InBarth FG ed Neurobiology of arachnids Berlin Spinger-Verlag 118ndash137

Foelix RF 1996 Biology of spiders 2nd edn Oxford OxfordUniversity Press

Giribet G Edgecombe GD Wheeler WC Babbit C 2002Phylogeny and systematic position of Opiliones a combinedanalysis of chelicerate relationships using morphologicaland molecular data Cladistics 18 5ndash70

Goloboff P 1993 Estimating character weights during treesearch Cladistics 9 83ndash91

Goloboff PA Farris JS Nixon K 2000 TNT Tree AnalysisUsing New Technologies Version 1 Program and documen-tation httpwwwzmucdkpublicphylogenytnt

Grainge CA Pearson RG 1966 Cuticular structure in thePhalangida Nature 211 866

Grandjean F 1938 Observations sur les Bdelles (Acariens)Annales de la Socieacuteteacute Entomologique de France Paris 107 1ndash24

Hansen HJ Soslashrensen W 1904 On two orders of ArachnidaCambridge Cambridge University Press

Hansen HJ Soslashrensen W 1905 The Tartarides a tribe of theorder Pedipalpi Arkiv fuumlr Zoologi 2 1ndash78

Hara MR Gnaspini P 2003 Comparative study of the defen-sive behavior and morphology of the gland opening areaamong harvestmen (Arachnida Opiliones Gonyleptidae)under a phylogenetic perspective Arthropod Structure andDevelopment 32 257ndash275

Harvey MS 1992 The phylogeny and classification of thePseudoscorpionida (Chelicerata Arachnida) InvertebrateTaxonomy 6 1373ndash1435

Harvey MS 2002 The neglected cousins what do we knowabout the smaller arachnid orders Journal of Arachnology30 357ndash372

Haupt J Song D 1996 Revision of East Asian whip scorpions(Arachnida Uropygi Theyphonida) I China and JapanArthropoda Selecta 5 43ndash52

Hjelle JT 1990 Anatomy and morphology In Polis GA edThe biology of scorpions Stanford University Stanford Uni-versity Press 9ndash63

Janczyk FSW 1956 Anatomie von Siro duricorius Joseph imVergleich mit anderen Opilioniden Sitzungsberichte Oster-reichischen Akademie der Wissenschaften Mathematisch-Natruwissenschaftliche Klasse 165 474ndash522

Jenner RA 2001 Bilaterian phylogeny and uncritical recy-cling of morphological data sets Systematic Biology 50 730ndash742

Jeram AJ 1998 Phylogeny classification and evolution of Sil-urian and Devonian scorpions In Selden PA ed Proceed-ings of the 17th European Colloquium of ArachnologyEdinburgh 1997 Burnham Beeches UK British Arachno-logical Society 15ndash31

Karaman IM 2005 Evidence of spermatophores in Cyphoph-thalmi (Arachnida Opiliones) Revue Suisse de Zoologie 1123ndash11

Kaumlstner A 1935 Die Funktion der sogenannten sympathis-chen Ganglien und die Exkretion bei den PhalangiidenZoologischer Anzeiger 109 273ndash288

Kierans JE Robbins RG 1999 A world checklist of generasubgenera and species of ticks (Acari Ixodida) publishedfrom 1973 to 1997 Journal of Vector Ecology 24 115ndash129

Kjellesvig-Waering EN 1986 A restudy of the fossil Scorpi-onida of the world Palaeontographica Americana 55 1ndash287

Klompen JSH 2000 Prelarva and larva of Opilioacarus (Neo-carus) texanus (Chamberlin and Mulaik) (Acari Opilioacar-ida) with notes on the patterns of setation and lyrifissuresJournal of Natural History 34 1977ndash1992

Kraus O 1998 Elucidating the historical process of phylog-eny phylogenetic systematics versus cladistic techniquesIn Selden PA ed Proceedings of the 17th European Collo-quium of Arachnology Edinburgh 1997 Burnham BeechesUK British Arachnological Society 1ndash7

Kraus O Kraus M 1993 Divergent transformation of cheli-cerae and original arrangement of eyes in spiders (Arach-nida Araneae) Memoirs of the Queensland Museum 33579ndash584

Lankester ER Benham WBS Beck EJ 1885 On the mus-

242 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

cular and endoskeletal systems of Limulus and Scorpio withsome notes on the anatomy and generic characters of scor-pions Transactions of the Zoological Society of London 11311ndash384

Leclerc P 1989 Considerations paleobiogeographiques a pro-pos la decouverte en Thailande drsquoopilioacariens nouveaux(Acari ndash Notostigmata) Compte Rendu des Seances de laSociete de Biogeographie 65 162ndash174

Legg G 1976 The external morphology of a new species ofricinuleid (Arachnida) from Sierra Leone Zoological Journalof the Linnean Society 59 1ndash58

Lindquist EE 1984 Current theories on the evolution ofmajor groups of Acari and on their relationships with othergroups of Arachnida with consequent implications for theirclassificaton In Griffiths DA Bowman CE eds AcarologyVI Vol 1 New York John Wiley amp Sons 28ndash62

Manning PL Dunlop JA 1995 The respiratory organs ofeurypterids Palaeontology 38 287ndash297

Manton SM 1958 Hydrostatic pressure and leg exten-sion in arthropods with special reference to arachnidsAnnals and Magazine of Natural History Series 13 1161ndash182

Manton SM 1964 Mandibular mechanisms and the evolutionof arthropods Philosophical Transactions of the Royal Soci-ety of London B 7 1ndash183

Maraun M Heethoff M Schneider K Scheu S WeigmannG Cianciolo J Thomas RH Norton RA 2004 Molecularphylogeny of oribatid mites (Oribatida Acari) evidence formultiple radiations of parthenogenetic lineages Experimen-tal and Applied Acarology 33 183ndash201

Marples BJ 1967 The spinnerets and epiandrous glands ofspiders Journal of the Linnean Society (Zoology) 46 209ndash222

Marples BJ 1983 Observations on the structure of the fore-gut of spiders Bulletin of the British Arachnological Society8 46ndash52

Millot J 1933 Notes compleacutementaires sur lrsquoanatomie desliphistiides et des hypchilides a propos drsquoun travail reacutecent deA Petrunkevitch Bulletin de la Socieacuteteacute Zoologique de France68 217ndash235

Millot J 1942 Sur lrsquoanatomie et lrsquohistophysiologie deKoenenia mirabilis Grassi (Arachnida Palpigradi) RevueFranccedilaise drsquoEntomologie 9 33ndash51

Millot J 1943 Notes complementaires sur lrsquoanatomielrsquohistologie et la repartition geographique en France deKoenenia mirabilis Grassi Revue Franccedilaise drsquoEntomologie 9127ndash135

Millot J 1945 Lrsquoanatomie interne des Ricinulei (Arachnides)Annales des Sciences Naturelles (Zoologie) Series 11 7 1ndash29

Millot J 1949a Morphologie generale et anatomie interne InGrasseacute P-P ed Traite de zoologie VI Paris Masson et Cie263ndash319

Millot J 1949b Ordre des Amblypyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 563ndash588

Millot J 1949c Ordre des Araneacuteides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 589ndash743

Millot J 1949d Ordre des Palpigrades In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 520ndash532

Millot J 1949e Ordre des Uropyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 533ndash562

Millot J 1949f Ordre des Ricinuleides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 744ndash760

Millot J Vachon M 1949 Ordre Des Solifuges In Grasseacute P-P ed Traite de zoologie VI Paris Masson et Cie 482ndash519

Moore RA Briggs DEG Bartels C 2005 A new specimen ofWeinbergina opitzi (Chelicerata Xiphosura) from the LowerDevonian Huumlnsruck Slate Germany PalaumlontologischeZeitschrift 79 399ndash408

Moritz M 1957 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) unter besonderer Beruumlcksichtigungder ausseren Morphologie der Bildung des Mitteldarmesund der Genitalanlage Zoologische Jahrbuumlcher Abteilungfuumlr Anatomie und Ontogenie der Tiere 76 331ndash370

Moritz M 1959 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) II Die Anlage und Entwicklung derCoxaldruumlse bei Phalangium opilio L Zoologische Jahr-buumlcher Abteilung fuumlr Anatomie und Ontogenie der Tiere 77229ndash240

Nixon K 1999 The parsimony ratchet a new method forrapid parsimony analysis Cladistics 15 407ndash414

Norton RA 1998 Morphological evidence for the evolutionaryorigin of Astigmata (Acari Acariformes) Experimental andApplied Acarology 22 559ndash594

Palmgren P 1978 On the muscular anatomy of spiders ActaZoologica Fennica 155 1ndash41

Patten W 1912 The evolution of the vertebrates and their kinPhiladelphia P Blakistonrsquos Son

Patten W Hazen AP 1900 The development of the coxalgland branchial cartilages and genital ducts of Limuluspolyphemus Journal of Morphology 16 459ndash490

Paulus HF 1979 Eye structure and the monophyly of theArthropoda In Gupta AP ed Arthropod phylogeny NewYork Van Nostrand Reinhold 299ndash383

Petrunkevitch A 1949 A study of Palaeozoic ArachnidaTransactions of the Connecticut Academy of Arts and Sci-ences 37 69ndash315

Pittard K Mitchell RW 1972 Comparative morphology ofthe life stages of Cryptocellus pelaezi (Arachnida Ricinulei)Graduate Studies Texas Tech University 1 1ndash77

Platnick NI 2005 The world spider catalog Version55 American Museum of Natural History httpresearchamnhorgentomologyspiderscatalogindexhtm

Platnick NI Gertsch WJ 1976 The suborders of spiders acladistic analysis (Arachnida Araneae) American MuseumNovitates 2607 1ndash15

Pocock RI 1893 On some points in the morphology of theArachnida (ss) with notes on the classification of the groupAnnals and Magazine of Natural History Series 6 11 1ndash19

Pocock RI 1902 Some points in the morphology and classi-fication of the Opiliones Annals and Magazine of NaturalHistory Series 7 10 504ndash516

Popadicccc A Panganiban G Rusch D Shear WA KaufmanTC 1998 Molecular evidence for the gnathobasic derivationof arthropod mandibles and for the appendicular origin ofthe labrum and other structures Development Genes andEvolution 208 142ndash150

PHYLOGENY OF THE ARACHNID ORDERS 243

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Raw F 1957 Origin of chelicerates Journal of Paleontology31 139ndash192

Reddell JR Cokendolpher JC 1995 Catalogue bibliogra-phy and generic revision of the order Schizomida (Arach-nida) Texas Memorial Museum Speleological Monographs4 1ndash170

Reissland A Goumlrner P 1985 Trichobothria In Barth FGed Neurobiology of arachnids Berlin Springer-Verlag 138ndash161

Roewer C-F 1934 Solifugae Palpigradi Bronns Klassen undOrdnungen ses Tierreich 5 IV 4 1ndash713

Rowland JM Cooke JAL 1973 Systematics of the arachnidorder Uropygida (=Thelyphonida) Journal of Arachnology 155ndash71

Rowland JM Sissom WD 1980 Report on a fossil palpi-grade from the Tertiary of Arizona and a review of themorphology and systematics of the order (ArachnidaPalpigradida) Journal of Arachnology 8 69ndash86

Rucker A 1901 The Texan Koenenia American Naturalist35 615ndash630

Ruhland M Rathmayer W 1978 Die Beinmuskulatur undihre Innervation bei der Vogelspinne Dugesiella henzti (Ch)(Araneae Aviculariidae) Zoomorphologie 89 33ndash46

Saboori A Kamali K 2000 Description of Allothrombiumtriticium adult (Acari Trombidiidae) from Iran Systematicand Applied Acarology 5 207ndash208

Schliwa M 1979 The retina of the phalangid Opilio raven-nae with particular reference to arhabdomeric cells Celland Tissue Research 204 473ndash495

Scholl G 1977 Beitrage zur Embryonalentwicklung von Lim-ulus polyphemus L (Chelicerata Xiphosura) Zoomorpholo-gie 86 99ndash154

Scholtz G Kamenz C 2006 The book lungs of Scorpionesand Tetrapulmonata (Chelicerata Arachnida) evidence forhomology and a single terrestrialization event of a commonarachnid ancestor Zoology 109 2ndash13

Seitz KA 1972 Zur Histologie und Feinstruktur des Herzensund der Haumlmocyten von Cupiennius salei Keys (AraneaeCtenidae) I Herzwandung Bildung und Differenzierungder Haumlmocyten Zoologische Jahrbuumlcher Abteilung fuumlr Anat-omie und Ontogenie der Tiere 89 351ndash384

Sekiguchi K ed 1988 Biology of horseshoe crabs Tokyo Sci-ence House

Selden PA 1981 Functional morphology of the prosoma ofBaltoeurypterus tetragonophthalmus (Fischer) (ChelicerataEurypterida) Transactions of the Royal Society of Edin-burgh Earth Sciences 72 9ndash48

Selden P 1992 Revision of the fossil ricinuleids Transactionsof the Royal Society of Edinburgh Earth Sciences 83 595ndash634

Selden PA 1993 Fossil arachnids ndash recent advances andfuture prospects Memoirs of the Queensland Museum 33389ndash400

Selden PA Dunlop JA 1998 Fossil taxa and relationshipsof chelicerates In Edgecombe GD ed Arthropod fossilsand phylogeny New York Columbia University Press 303ndash331

Selden PA Jeram AJ 1989 Palaeophysiology of terrestrial-

isation in the Chelicerata Transactions of the Royal Societyof Edinburgh Earth Sciences 80 303ndash310

Selden PA Shear WA Bonamo PM 1991 A spider and otherarachnids from the Devonian of New York and reinterpre-tations of Devonian Araneae Palaeontology 34 241ndash281

Sensenig A Shultz JW 2003 Mechanics of cuticular elasticenergy storage in leg joints lacking extensor muscles inarachnids Journal of Experimental Biology 206 771ndash784

Shear WA 1993 New species in the opilionid genus Stylocel-lus from Malaysia Indonesia and the Philippines (OpilionesCyphophthalmi Stylocellidae) Bulletin of the British Arach-nological Society 9 174ndash188

Shear WA Selden PA Rolfe WDI Bonamo PM GriersonJD 1987 New terrestrial arachnids from the Devonian ofGilboa New York (Arachnida Trigonotarbida) AmericanMuseum Novitates 2901 1ndash74

Shultz JW 1987 The origin of the spinning apparatus in spi-ders Biological Reviews of the Cambridge PhilosophicalSociety 62 89ndash113

Shultz JW 1989 Morphology of locomotor appendages inArachnida evolutionary trends and phylogenetic implica-tions Zoological Journal of the Linnean Society 97 1ndash56

Shultz JW 1990 Evolutionary morphology and phylogeny ofArachnida Cladistics 6 1ndash31

Shultz JW 1991 Evolution of locomotion in Arachnida thehydraulic pressure pump of the giant whipscorpion Masti-goproctus giganteus (Uropygi) Journal of Morphology 21013ndash31

Shultz JW 1993 Muscular anatomy of the giant whipscor-pion Mastigoproctus giganteus (Arachnida Uropygi) and itsevolutionary significance Zoological Journal of the LinneanSociety 108 335ndash365

Shultz JW 1998 Phylogeny of Opiliones (Arachnida) anassessment of the lsquoCyphopalpatoresrsquo concept Journal ofArachnology 26 257ndash272

Shultz JW 1999 Muscular anatomy of a whipspider Phrynuslongipes (Amblypygi) and its evolutionary significance Zoo-logical Journal of the Linnean Society 126 81ndash116

Shultz JW 2000 Skeletomuscular anatomy of the harvest-man Leiobunum aldrichi (Weed 1893) (Arachnida Opil-iones Palpatores) and its evolutionary significanceZoological Journal of the Linnean Society 128 401ndash438

Shultz JW 2001 Gross muscular anatomy of Limuluspolyphemus (Chelicerata Xiphosura) and its bearing on evo-lution in the Arachnida Journal of Arachnology 29 283ndash303

Shultz JW 2007 Morphology of the prosomal exoskeleton ofScorpiones (Arachnida) and a new hypothesis for the evolu-tion of cuticular cephalic endoskeletons in arthropodsArthropod Structure amp Development 36 77ndash102

Shultz JW Pinto da Rocha R 2007 Morphology and func-tional anatomy In Pinto da Rocha R Machado G Giribet Geds The harvestmen the biology of Opiliones CambridgeMA Harvard University Press 14ndash61

Shultz JW Regier JC 2001 Phylogenetic analysis of Pha-langida (Arachnida Opiliones) using two nuclear protein-encoding genes supports monophyly of Palpatores Journalof Arachnology 29 189ndash200

Sitnikova LG 1978 The main evolutionary trends of the

244 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Acari and the problem of their monophyletism Entomologis-cheskoe Obozrenie 57 431ndash457

Snodgrass RE 1948 The feeding organs of Arachnidaincluding mites and ticks Smithsonian Miscellaneous Col-lections 110 (10) 1ndash94

Snodgrass RE 1952 A textbook of arthropod anatomy Ith-aca NY Comstock Publishing

Soslashrensen W 1879 Om Bygningen af Gonyleptiderne en Typeaf Arachnidernes Classe Naturhistorisk Tidsskrifter Series3 8 97ndash222

Stockwell SA 1989 Revision of the phylogeny and higherclassification of scorpions (Chelicerata) Unpublished doc-toral dissertation University of California at Berkeley

Stoslashrmer L 1944 On the relationships and phylogeny of fossiland Recent Arachnomorpha Skrifter utgitt av Det NorskeVidenskaps-Akademi i Oslo I 5 1ndash158

Stoslashrmer L 1955 Merostomata In Moore RC ed Treatise oninvertebrate paleontology Part P Arthropoda 2 LawrenceKS University of Kansas Press and Geological Society ofAmerica 4ndash41

Swofford DL 2002 PAUP Phylogenetic analysis using par-simony (and other methods) Version Beta 10 SunderlandMA Sinauer Associates

Talarico G Palacios-Vargas JG Fuentes Silva M AlbertiG 2005 First ultrastructural observations of the tarsal poreorgan of Pseudocellus pearsei and P boneti (ArachnidaRicinulei) Journal of Arachnology 33 604ndash612

Templeton AR 1983 Phylogenetic inference from restrictionendonuclease cleavage site maps with articular reference tothe evolution of humans and the apes Evolution 37 221ndash244

Tetlie OE Braddy SJ 2004 The first Silurian chasmataspidLoganamaraspis dunlopi gen et sp nov (Chelicerata Chas-mataspidida) from Lesmahagow Scotland and its implica-tions for eurypterid phylogeny Transactions of the RoyalSociety of Edinburgh Earth Sciences 94 227ndash234

Thomas RH Telford MJ 1999 Appendage development inembryos of the oribatid mite Archegozetes longisetosus(Acari Oribatei Trhypochthoniidae) Acta Zoologica (Stock-holm) 80 193ndash200

Tollerton VP Jr 1989 Morphology taxonomy and classifica-tion of the order Eurypterida Burmeister 1843 Journal ofPaleontology 63 642ndash657

Vachon M 1949 Ordre des Pseudoscorpions In Grasseacute P-Ped Traite de zoologie VI Paris Masson et Cie 436ndash481

Van der Hammen L 1979 Comparative studies in Chelicer-ata I The Cryptognomae (Ricinulei Architarbi and Anacti-notrichida) Zoologische Verhandelingen 174 1ndash62

Van der Hammen L 1982 Comparative studies in Chelicer-ata II Epimerata (Palpigradi and Actinotrichida) Zoolo-gische Verhandelingen 196 1ndash70

Van der Hammen L 1985 Functional morphology and affin-ities of extant Chelicerata in evolutionary perspectiveTransactions of the Royal Society of Edinburgh 76 137ndash146

Van der Hammen L 1989 An introduction to comparativearachnology Leiden SPB Academic Publishing

Vyas AB 1970 Studies in myology of the scorpion Heter-ometrus fulvipes Koch Unpublished PhD DissertationDepartment of Zoology Gujarat University India

Vyas AB 1974 The cheliceral muscles of the scorpion Heter-ometrus fulvipes Bulletin Southern California Academy ofSciences 73 9ndash14

Walter DE 1988 Predation and mycophagy by endeostigma-tid mites (Acariformes Prostigmata) Experimental andApplied Acarology 4 159ndash166

Walter DE Proctor HC 1998 Feeding behaviour and phy-logeny observations on early derivative Acari Experimentaland Applied Acarology 22 39ndash50

Walter DE Proctor HC 1999 Mites ecology evolution andbehaviour Sydney University of New South Wales

Waterston CD 1979 Problems of functional morphology andclassification in stylonuroid eurypterids (Chelierata Meros-tomata) with observations on the Scottish Silurian Sty-lonuroidea Transactions of the Royal Society of EdinburghEarth Science 70 251ndash322

Weygoldt P 1969 Biology of pseudoscorpions CambridgeMA Harvard University Press

Weygoldt P 1984 Lrsquoautotomie chez les Amblypyges RevueArachnologique 5 321ndash327

Weygoldt P 1996 Evolutionary morphology of whip spiderstowards a phylogenetic system (Chelicerata ArachnidaAmblypygi) Journal of Zoological Systematics and Olution-ary Research 34 185ndash202

Weygoldt P 1998 Evolution and systematics of the Chelicer-ata Experimental and Applied Acarology 22 63ndash79

Weygoldt P 2000 Whip spiders (Chelicerata Amblypygi)their biology morphology and systematics StenstrupDenmark Apollo Books

Weygoldt P Paulus HF 1979 Untersuchungen zur Morphol-ogie Taxonomie und Phylogenie der Chelicerata Zeitschriftfuumlr Zoologische Systematik und Evolutionsforschung 17 85ndash116 177ndash200

Wheeler WC Hayashi CY 1998 The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192

Winkler D 1957 Die Entwicklung der aumlusseren Koumlrperge-stalt bei den Phalangiidae (Opiliones) Mitteilungen aus demZoologischen Museum in Berlin 33 355ndash389

With CJ 1904 The Notostigmata a new suborder of AcariVidenskabelige Meddelelser fra Dansk Naturhistorisk Foren-ingi Kjobenhaus 1904 137ndash192

Yamasaki T Makioka T Saito J 1988 Morphology InSekiguchi K ed Biology of horseshoe crabs Tokyo ScienceHouse 69ndash132

Yoshikura M 1975 Comparative embryology and phylogeny ofArachnida Kumamoto Journal of Science Biology 12 71ndash142

APPENDIX

Each character used in this analysis is numbereddefined and cross-referenced with characters used infour previous studies that are abbreviated in bracketsspecifically WP = Weygoldt amp Paulus (1979)S = Shultz (1990) WH = Wheeler amp Hayashi (1998)GEWB = Giribet et al (2002) Descriptions includecorrections of errors discussions of controversies orambiguities and justifications for state assignmentsto lsquoproblemrsquo taxa

PHYLOGENY OF THE ARACHNID ORDERS 245

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

PROSOMA

1 Dorsal sclerite formed by fusion of the prosomal car-apace the dorsal portion of the first opisthosomalsomite and the dorsomedial (axial) portion of the sec-ond opisthosomal somite 0 absent 1 presentState 1 is present in extant xiphosurids and close fos-sil relatives (Scholl 1977 Anderson amp Selden 1997Shultz 2001) The fossil record suggests that thisstructure originated with the disappearance ofthe already reduced first opisthosomal tergite(= microtergite) and axial portion of the secondopisthosomal tergite (Anderson amp Selden 1997)2 Single dorsal sclerite covering entire dorsal surfaceof body no lines indicating original segmentation 0absent 1 presentState 1 occurs throughout Holothyrida (Acari) (Vander Hammen 1989) and in many Mesostigmata(Alberti amp Coons 1999) including Glyptholaspis (Vander Hammen 1989)3 Anterior end of dorsal prosoma with medianmarginal or submarginal pointed process 0 absent1 presentState 1 occurs throughout Schizomida (Protoschizo-mus Stenochrus Reddell amp Cokendolpher 1995) The-lyphonida (Rowland amp Cooke 1973 Shultz unpublobs Proschizomus Dunlop amp Horrocks 19951996)Amblypygi (Charinus Millot 1949b Phrynus Shultz1999) and Plesiosiro (Dunlop 1999) Citing a figure inRowland amp Cooke (1973) Dunlop amp Horrocks assumedthat the anterior process was limited to lsquohypoctonidrsquothelyphonids and united them with Proschizomus andSchizomida If valid this would make Thelyphonidaparaphyletic However the doubtful monophyly oflsquohypoctonidsrsquo (Haupt amp Song 1996) and widespreadpresence of a submarginal anterior process in Pedi-palpi weaken this proposal (Harvey 2002)4 Ophthalmic ridges pair of longitudinal crests inter-secting or passing near the region of the lateral eyes orcomparable region where lateral eyes are absent 0absent 1 presentState 1 is present throughout Xiphosura (Anderson ampSelden 1997) Similar structures are present in Plesi-osiro (Dunlop 1999) and lsquonon-hypoctonidrsquo Thely-phonida (Mastigoproctus Rowland amp Cooke 1973)Dunlop (2002a) described apparent ophthalmic ridgesin the chasmataspidid Octoberaspis but these werenot observed in Diploaspis (Dunlop et al 2001) eitherdue to absence or inadequate preservation of speci-mens nor in Chasmataspis (Dunlop et al 2004)5 Carapace with demarcations (eg grooves scleritesphragmata) between pro- meso- or metapeltidia 0absent 1 present [S 1 WH 48 GEWB 29]State 1 is present in extant Scorpiones Cyphoph-thalmi (Hansen amp Soslashrensen 1904) and many Palpa-tores (Opiliones) including Caddo and Leiobunum

and most Pseudoscorpiones (Chamberlin 1931) Giri-bet et al (2002) coded this character as absent inCyphophthalmi (Opiliones) but a transverseprocurved groove is present throughout the group(Hansen amp Soslashrensen 1904) and represents the meso-metapeltidial border as indicated by attachment ofthe dorsal endosternal suspensor of postoral somite VIin Chileogovea oedipus and Siro exilis (Shultz unpublobserv) Kjellesvig-Waering (1986) did not explicitlydescribe such grooves in fossil scorpions although heillustrated corresponding structures in PrearcturusArchaeophonus Archaeoctonus Palaeophonus andother Palaeozoic scorpions (Kjellesvig-Waering 1986figs 14A 21E 27D 59) (coded as uncertain here) butnot in Proscorpius (contra Giribet et al 2002) Dis-tinct carapacal sclerites are present throughout Palpi-gradi Schizomida and Solifugae (6)

The condition in Acari is controversial but is codedhere as State 1 Van der Hammen (1989) consideredthe tergal region of the leg-bearing somites (postoralsomites IIIndashVI) or podosoma to be reduced andreplaced by posterior migration of the aspidosoma (ietergal region assumed to be associated with append-ages of the gnathosoma) and anterior migration of theopisthosomal tergal region to form the hysterosomaThe dorsal proterosomal and hysterosomal elementsmeet at a transverse sejugal furrow that continues lat-erally and passes ventrally between the coxae of legs 2and 3 According to this scheme the prosomandashopisthosoma border is expressed as the disjugal fur-row which passes from the ventral posterior margin ofthe podosoma anterodorsally to join the dorsal part ofthe sejugal furrow Based on the arrangement of setaeand slit sensilla (= lyrifissures) Van der Hammen(1989) interpreted the region above coxae 3 and 4 (ieregion C) as a fusion of the dorsal parts of the first twoopisthosomal somites (= postoral somites VII and VIII)in early divergent Anactinotrichida (ie Opilioacari-formes) and Acariformes (eg Alycus) These interpre-tations have been followed by many acarologistsalthough its speculative aspects are widely acknowl-edged (Evans 1992 Alberti amp Coons 1999)

From the standpoint of a general arachnologist theVan der Hammen system seems unnecessarily compli-cated it appears to have been formulated to explainbroad morphological themes in Oribatida but was thenextrapolated to other mites A more conservativelsquoarachnologicalrsquo scheme adopted here equates the pro-podosoma with the pars cephalica (= propeltidium) ofthe arachnid prosoma and region C with the pars tho-racica (= mesopeltidium + metapetidium) In fact theearliest and most recent treatments of Opilioacari-formes (With 1904 Klompen 2000) reached similarconclusions This interpretation is also consistent withevidence from Acariformes For example the regiondorsal to the coxa of legs 3 and 4 in Alycus is as readily

246 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

explained by the persistence of primitive prosomal ter-gal elements as by the supposition that these disap-peared and were replaced by two opisthosomalelements Furthermore several diverse lineages ofbasally divergent Acariformes display features that donot fit readily within the GrandjeanndashVan der Ham-men system but seem to correspond to the cephalicathoracica division of the prosoma They have a sejugalfurrow that may have demarcated the cephalicatho-racica border and a postpodosomal furrow thatappears to represent the prosomandashopisthosoma junc-tion there is no disjugal furrow Examples includepalaeosomate (especially aphelacarid) and pediculoch-elid Oribatida alicorhagiid and mircopsammidlsquoEndeostigmatarsquo and paratydeid Prostigmata (Albertiamp Coons 1999 fig 42) This organization may be sym-plesiomorphic for Acariformes but could be a conver-gence brought about by selection for enhancedflexibility6 Carapace with distinct pro- meso- or metapeltidialsclerites 0 absent 1 present - inapplicable due toabsence of pro- meso- or metapeltidial demarcations(5)State 1 occurs throughout Palpigradi Schizomida andSolifugae It also appears sporadically within Opil-iones (eg Leiobunum flavum Shultz unpubl observ)and Pseudoscorpiones (eg Pseudochiridium) (Cham-berlin 1931)7 Sejugal furrow circumferential zone of body flexi-bility that passes between the coxae of legs 2 and 3 0absent 1 presentState 1 is probably synapomorphic for Acariformes asit occurs in most lsquoendeostigmatidsrsquo including Alycus(Van der Hammen 1989) and is variously developedthroughout Sarcoptiformes (Alberti amp Coons 1999) Itis probably the primitive condition for Prostigmatabut is either absent or weakly expressed in represen-tatives included here See 5 for alternative morpholog-ical interpretations of this body region in Acari8 Prosomal ozopores 0 absent 1 present [WH 46GEWB 12]State 1 occurs throughout Opiliones and Holothyrida(Van der Hammen 1989)9 Carapacal pleural doublure 0 absent 1 present[WP 12 S 2 WH 49 GEWB 27]A carapacal pleural doublure occurs throughoutXiphosura (Stoslashrmer 1944) Eurypterida (Clarke ampRuedemann 1912 Selden 1981) and Chasmataspid-ida (Dunlop et al 2001 2004 Dunlop 2002a) and isabsent in all known Arachnida10 Cardiac lobe a longitudinal axial elevation of thecarapace 0 absent 1 presentState 1 occurs throughout Xiphosura (Anderson ampSelden 1997) and Eurypterida (Clarke amp Ruedemann1912 Selden 1981) Carapaces are not preserved suf-ficiently to determine this condition in the chasmatas-

pidids Diploaspis (Dunlop et al 2001) andOctoberaspis (Dunlop 2002a) but an apparent cardiaclobe occurs in the Chasmataspis (Dunlop et al 2004)A cardiac lobe is absent in Arachnida11 Moveable cucullus 0 absent 1 present [WP 54GEWB 22]State 1 is a synapomorphy of Ricinulei (Selden 1992)12 Medial intercoxal lsquosternalrsquo region 0 all pedalcoxae separated medially 1 anterior pedal coxae abut-ting medially posterior coxae separated 2 anteriorpedal coxae separated medially posterior coxae abut-ting 3 all pedal coxae abutting medially 4 epimeracoxae undifferentiated medially from ventral body wall[S sim3 WH sim50 GEWB sim17]Prosomal sternites appear to be those portions of theventral body wall not occupied by coxae Despite earlyattempts to assess metamerism of the ventralprosoma (protosternum deutosternum tritosternumetc) (Boumlrner 1902a Millot 1949a) no apparent mor-phological features such as borders of sclerites ormuscle attachments reliably demarcate the ventralbody wall of one somite from that of an adjacentsomite Thus coding schemes that focus on describingmetameric components of the sternal region are prob-lematic Shultz (1990) attempted to code the shape ofthe entire intercoxal region and subsequent workershave adopted this approach All pedal coxae are sepa-rated medially in Plesiosiro (Dunlop 1999) Palpigradi(Roewer 1934) Araneae (Millot 1949c) Amblypygi(Weygoldt 2000) and many Acari including Opilioac-ariformes Holothryrida Mesostigmata Ixodida Aly-cus and many Prostigmata (Van der Hammen 1989Evans 1992 Alberti amp Coons 1999) The anteriorcoxae are separated in extant Thelyphonida Schizo-mida (Hansen amp Soslashrensen 1905 Millot 1949e) andRicinulei (Pittard amp Mitchell 1972 Legg 1976) Thecoxae are fused to the ventral body wall in Sarcopti-formes (Van der Hammen 1989 Evans 1992)

Interpreting this character is complicated for theanterior coxae of Scorpiones and Opiliones and allcoxae in Xiphosura and Eurypterida due to the pres-ence of coxapophyses endites andor gnathobaseswhich form components of the preoral chambers inthese groups Thus coxae may abut medially whenthese structures are considered part of the coxae butare separate medially if these structures are ignoredAs coxapophyses etc are here coded as separate char-acters (50ndash53) I have chosen to code this character asif these structures did not exist Consequently pha-langid Opiliones are coded as having all coxae sepa-rated (Pocock 1902 Hansen amp Soslashrensen 1904) Fossilscorpions also show substantial variation in this char-acter All pairs of pedal coxae in Palaeoscorpius abutmedially but also have gnathobases (Jeram 1998)and this condition is coded here as uncertain A ster-num separates all pairs of pedal coxae medially in

PHYLOGENY OF THE ARACHNID ORDERS

247

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Proscorpius

and

Stoermeroscorpio

(State 0) and

Pre-arcturus

approximates modern scorpions (State 1)(Kjellesvig-Waering 1986 Jeram 1998) The coxalbases of Eurypterida and Xiphosura are separated bya small sternite the endostoma (Selden 1981 Shultz2001)

13 Postoral sternapophysis (

=

tritosternum labium)a cuticular evagination of the ventral body wall poste-riorly adjacent to the palpal coxae forming the poste-rior border of the preoral chamber in some taxa ordisplaced posteriorly by fusion of the palpal coxae inothers 0 absent 1 present

[S

sim

10 WH

sim

56 GEWB19

=

33]State 1 is present throughout Palpigradi AraneaeAmblypygi (Snodgrass 1948) Schizomida (Hansen ampSoslashrensen 1905 Van der Hammen 1989) Thely-phonida (Van der Hammen 1989) extant Ricinulei(Pittard amp Mitchell 1972 Legg 1976) Opilioacari-formes and Mesostigmata (Van der Hammen 1989Alberti amp Coons 1999)

Allothyrus constrictus

is theonly holothyrid mite known to have sternapophysesAn apparent sternapophysis has been described in apalaeocharinid trigonotarbid (Dunlop 1994) Giribet

et al

(2002) miscoded Ixodida and Acariformes as hav-ing a lsquolabium (

=

tritosternum)rsquo The lophognath ofpseudoscorpions (Chamberlin 1931) and lsquolabiumrsquo ofSolifugae (Roewer 1934) also appear to be sternapo-physes that have been incorporated into the ros-trosoma (

32

) The labium of phalangid Opiliones(Pocock 1902 Hansen amp Soslashrensen 1904) is a ster-napophysis but is associated with leg 1 rather thepedipalp (Winkler 1957) A small anterior sclerite inthe fossil scorpions

Proscorpius

Waeringoscorpio

and

Labriscorpio

has been interpreted as a lsquolabiumrsquo homol-ogous with the sternapophyses of other arachnids(Weygoldt 1998) despite absence of any indication ofthe relationship of this sclerite to the mouth and pos-sible alternative interpretations (eg

16

) Giribet

et al

coded this sclerite as a sternapophysis but the char-acter is coded here as uncertain in fossil scorpions

14 Channels on the body surface linking openings ofcoxal organs to preoral chamber 0 absent 1 present

Van der Hammen (1989) described bilaterally pairedcuticular tracts (lsquotaenidiarsquo) connecting orifices of coxalorgans to the preoral chamber in some arachnids andor a ventromedian groove (

=

intercoxal or subcapitulargutter) that leads in turn to the preoral chamber inother taxa Taenidia and intercoxal gutters arepresent in Opilioacariformes (Van der Hammen1989) Holothyrida (Van der Hammen 1989) Mesos-tigmata (

Glyptholapis

Van der Hammen 1989)Ricinulei (Pittard amp Mitchell 1972 Van der Hammen1989) Araneae (

Heptathela

Aphonopelma

Hypochi-lus

Shultz unpubl observ also

Porrhothele

(Mygalo-morphae Dipluridae) Butt amp Taylor 1991

Segestria

(Araneomorphae Segestriidae) Van der Hammen

1989) Amblypygi (Van der Hammen 1989) Schizo-mida (

Stenochrus

Shultz unpubl observ unidenti-fied Hubbardiinae Van der Hammen 1989) andThelyphonida (

Mastigoproctus

Shultz unpublobserv also

Tetrabalius

Van der Hammen 1989) Theconduction of large volumes of fluid from orifices of thecoxal organs to the preoral chamber has been observedin the mygalomorph spider

Porrhothele

(Dipluridae)(Butt amp Taylor 1991) and

Mastigoproctus

(Thely-phonida) (Shultz unpubl observ) The podocephaliccanal (

15

) is present in Acariformes (Acari) (Lindquist1984) and appears to have evolved as an invaginatedsupracoxal channel (Grandjean 1938 G Alberti inEvans 1992) Van der Hammen (1982) could not dis-cern with certainty whether a channel exists in

Eukoenenia

(Palpigradi) but none has been reportedby previous authors and they have not been observedby the present author Palpigradi is coded as havingState 0

15 Podocephalic canal cuticular channel andorduct draining multiple glands and opening nearmouthparts 0 absent 1 present - inapplicable due toabsence of channel (14)

State 1 occurs throughout Acariformes (Acari)(Lindquist 1984)

16 Heavily sclerotized suboral sclerite serving inpart as basal pivot point for coxae of appendages ofpostoral somites IIndashIV (

=

arachnid palp and legs 1 and2) 0 absent 1 present

The sclerite forms the posterior wall of the true mouthand is heavily sclerotized in extant scorpions It is notvisible externally due to the tightly fitting coxae of theanterior prosomal appendages (Shultz 2007) It maycorrespond to the so-called labium (Weygoldt 1998) orlabrum (Kjellesvig-Waering 1986) of certain fossilscorpions (see

13

)

17 Genal angles 0 rounded 1 pointed

Anderson amp Selden (1997) originally coded a pointedgenal angle and genal spine as separate charactersbut they are combined here State 1 is present in manyXiphosura (Anderson amp Selden 1997) and Chas-mataspidida (Dunlop

et al

2001 2004 Dunlop2002a)

A

PPENDAGES

OF

POSTORAL

SOMITE

I

CHELICERAE

18 Cheliceral segmentation 0 three articles 1 twoarticles

[WP

sim

25

+

42 WH

sim

15 GEWB

sim

44]State 0 is present in extant Xiphosura Eurypterida(Clarke amp Ruedemann 1912 Selden 1981) Palpi-gradi (Roewer 1934) Opiliones (Hansen amp Soslashrensen1904) Scorpiones (Hjelle 1990) and throughout non-acariform Acari (Evans 1992) State 1 is presentthroughout known Trigonotarbida Araneae Ambly-pygi Thelyphonida Schizomida Pseudoscorpionesand Solifugae Acariform mites have two unambigu-

248 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ous cheliceral articles but may have a reduced thirdarticle (Van der Hammen 1989 Evans 1992) acari-form mites are thus coded as uncertain Chelicerae arenot known or their condition is unclear in several fos-sil taxa including Xiphosura Chasmataspidida (Dun-lop et al 2001 2004 Dunlop 2002a) Riciniulei(Selden 1992) Plesiosiro (Dunlop 1999) Proschizo-mus (Dunlop amp Horrocks 19951996) and the fossilscorpions Praearcturus and Palaeoscorpius (Kjelles-vig-Waering 1986) Kjellesvig-Waering (1986) consis-tently interpreted chelicerae of fossil scorpions ashaving four articles but this is probably a misinter-pretation of the basal article (see also Stockwell1989) The basal article in extant scorpions consists ofa distal collar of cuticle and a large proximal processand this arrangement would probably appear as twoarticles in compressed specimens In any event thereare at least three cheliceral articles in Stoermeroscor-pio and Proscorpius (Kjellesvig-Waering 1986)19 Terminal cheliceral joint 0 laterally placedbicondylar hinge 1 dorsally placed bicondylar hinge2 ventrally placed bicondylar hinge [WP 25 + 42 Ssim11 + 12 WH sim15 GEWB sim44]State 0 is present in Xiphosura (Shultz 2001)extant Scorpiones (Millot amp Vachon 1949) Opil-iones (Hansen amp Soslashrensen 1904) Palpigradi(Boumlrner 1904 Millot 1949d) Ricinulei (Pittard ampMitchell 1972 Legg 1976) and Eurypterida includ-ing Baltoeurypterus (Selden 1981) and Stylonurus(Clarke amp Ruedemann 1912) State 0 is present inthe fossil scorpions Proscorpius and Stoermeroscor-pio but the state of the chelicerae is unclear in theother fossil scorpions represented here (Kjellesvig-Waering 1986) State 1 is present in Trigonotarbida(Shear et al 1987) mygalomorph Araneae Ambly-pygi Schizomida and Theylphonida (Weygoldt ampPaulus 1979) State 2 is present in Pseudoscorpi-ones (Chamberlin 1931) and Solifugae (Roewer1934) and appears to be a groundplan feature for allmajor groups of mites (eg Van der Hammen 1989Alberti amp Coons 1999) The difference between theorthognathy of mygalomorph Araneae and lsquoplagiog-nathyrsquo mesothele (Liphistius Heptathela) and palae-ocribellate Araneae (Hypochilus) (Kraus amp Kraus1993) is small compared with the variation thatoccurs throughout arachnids and plagiognathy iscoded as State 1 here20 Chelicera articulating with carapace at anterolat-eral pivot 0 absent 1 present [S 13 WH 57 GEWB45]State 1 occurs in Solifugae (Roewer 1934) and Pseu-doscorpiones except Chthonioidea and Feaelloidea(Chamberlin 1931) Giribet et al (2002) coded thecharacter as present in Pseudoscorpiones21 Chelicera pivoting on dorsal protuberance of epis-tome 0 absent 1 present

State 1 appears to be a unique synapomorphy of Opil-iones (Shultz 2000)22 Extrinsic cheliceral muscle arising on carapaceand inserting on dorsal margin of nonbasal cheliceralarticle 0 absent 1 present (Note Characters 22 and23 refer to two different muscles)State 1 is known only in extant Scorpiones (LankesterBenham amp Beck 1885 Vyas 1970 1974 Shultz2007) but may occur in acariform mites if their cheli-cerae are composed of three rather than two articles(Evans 1992) (see 18) Acariformes are coded asuncertain23 Extrinsic cheliceral muscle arising on carapaceand inserting on ventral margin of nonbasal cheliceralarticle 0 absent 1 present [GEWB 252] (Note Char-acters 22 and 23 refer to two different muscles)State 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) somemesostigmatid Acari (Evans 1992) and representa-tive Opiliones particularly Leiobunum (Shultz 2000)Chileogovea and Gonyleptes (Shultz unpubl observ)The status of the character in other opilions is not yetknown The character was originally defined as anextrinsic muscle inserting on the second segment ofthree-segmented chelicerae (Shultz 2000 Giribetet al 2002) but it has been redefined here to make itapplicable to arachnids with two-segmented cheli-cerae24 Extrinsic cheliceral muscles attaching to epistome0 absent 1 presentState 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) but mayoccur in Prostigmata if the sigmoid piece to which thecheliceral protractors attach (Evans 1992) is a modi-fication of the epistome25 Lateral tergocheliceral muscle with three heads 0absent 1 present - coded only for extant tetrapulmo-nates and the palpigrade Eukoenenia homology isunclear in other taxa [GEWB 230]State 1 is present in Pedipalpi (Shultz 1999)26 Cheliceral silk glands 0 absent 1 present [GEWB47]State 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931) The silk gland of spider mites(eg Tetranychus) are associated with the pedipalps(Evans 1992)27 Cheliceral venom glands 0 absent 1 present [WP34 WH 40 GEWB 46]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)28 Cheliceral serrula interior and exterior 0 absent1 presentState 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931 Harvey 1992)29 Cheliceral lsquoflagellumrsquo in male 0 absent 1 present[GEWB 48]

PHYLOGENY OF THE ARACHNID ORDERS 249

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITE II LEG 1 IN NONARACHNIDS AND PALP IN ARACHNIDS

30 Palpal coxae fused ventromedially and formingposterior wall of preoral chamber 0 absent 1 present[WP 27 S 18 WH 17 = 62 GEWB 33 + 63]State 1 occurs in Schizomida Thelyphonida Acari andRicinulei31 Gnathosoma 0 absent 1 present - inapplicablepalpal coxae not fused (30) [WP 56 S 19 WH sim63GEWB sim36 + 37]The gnathosoma is a functional complex comprisingan epistome attached to the dorsal surface of mediallyfused palpal coxae (= infra- or subcapitulum) (see 30)the chelicerae and in some taxa a supracheliceral tec-tum These components are tightly integrated andmove in unison relative to the body The gnathosomahas traditionally been considered a feature of AcariHowever Van der Hammen (1989) argued that a gna-thosoma is present in Ricinulei and this interpreta-tion has received wide acceptance (eg Lindquist1984) However it is not clear that the relationship ofthe chelicerae and subcapitulum is as intimate inRicinulei as in Acari and the ricinuleid subcapitulumdoes not appear to be as moveable In fact the lsquosubca-pitulumrsquo in Ricinulei is similar in many respects to thecondition in Schizomida and Thelyphonida (Hansen ampSoslashrensen 1904) which has never been regarded as agnathosoma Consequently I have coded the conditionof Ricinulei as uncertain32 Rostrosoma long narrow subcylindrical epistomeprojecting anteriorly with base fixed to dorsal surfaceof palpal coxae bordered laterally by lobes projectingfrom palpal coxae ventral wall of preoral chamberformed by anterior element of prosoma (sternapophy-sis) 0 absent 1 present [S 15 WH 59 GEWB 20 + 36]State 1 is present in Pseudoscorpiones and Solifugae(Chamberlin 1931 Roewer 1934 Van der Hammen1989) Dunlop (2000) attempted to homologize com-ponents of the rostrosoma and gnathosoma (31) anargument accepted by Giribet et al (2002) Howeverfusion of the palpal coxae (30) apparently evolvedindependently in the two structures Coxal fusion inthe gnathosoma is complete ventrally and excludesthe suboral sternapophysis (13) while fusion isincomplete ventrally in the rostrosoma and incorpo-rates the sternapophysis Unlike the gnathosomathe rostrosoma is largely immobile and cheliceralmovement is not coupled with that of the palpalcoxae33 Rutellacorniculi hypertrophied setae modified asmouthparts located on the anterior processes of the pal-pal coxae 0 absent 1 present [GEWB 39]

State 1 occurs in the anactinotrichid groups Opilioac-ariformes Holothyrida and Mesostigmata but not Ixo-dida (Van der Hammen 1989) It also occurs inSarcoptiformes (Evans 1992) and certain endeostig-matids (eg Alycus Van der Hammen 1989) but not inProstigmata Comparable structures are unknown inother chelicerate groups34 Terminal segments specially modified in adultmale as a clasper used to engage the female 0 absent1 presentState 1 is known only in extant Xiphosura (YamasakiMakioka amp Saito 1988)35 Robust raptorial 0 absent 1 present [S 17 WH61 GEWB sim59 97]State 1 is present in Amblypygi Thelyphonida andSchizomida (Shear et al 1987) and in Laniatores(Opiliones)36 Orientation of robust raptorial appendage 0operating in subtransverse plane 1 operating in sub-vertical plane - inapplicable coded only for PedipalpiState 0 is present in extant Thelyphonida and mostAmblypygi and State 1 is present in Schizomida (Cok-endolpher amp Reddell 1992) and in the basally diver-gent amblypygid Paracharon caecus (Weygoldt 2000)The orientation of the pedipalps in the fossil Proschi-zomus is uncertain (Dunlop amp Horrocks 19951996)37 Extrinsic muscle attaching to epistome 0 absent1 presentState 1 occurs in Chileogovea and Siro (OpilionesCyphophthalmi) and extant Scorpiones (Vyas 1970Shultz 2007)38 Muscle originating and inserting within coxa 0absent 1 present [GEWB sim232]State 1 occurs is in Amblypygi (Phrynus Shultz 1999Phrynichus Boumlrner 1904) Thelyphonida (Mastigo-proctus Shultz 1993 Thelyphonus Boumlrner 1904) andSchizomida (Stenochrus Shultz unpubl observ)There are two intracoxal muscles in amblypygids andthelyphonids the one coded here and another that isserially homologous with a muscle in the pedal coxae(54)39 Tarsus andor tibia with venom glands 0 absent1 present [GEWB 64]State 1 is known only from extant Pseudoscorpi-ones except the superfamilies Chthonioidea andFeaelloidea (Chamberlin 1931 Harvey 1992) Giri-bet et al (2002) coded this character as present forPseudoscorpiones40 Tarsal grooming organ 0 absent 1 present[GEWB 98]State 1 is a unique synapomorphy of Amblypygi (DelleCave 1975 Weygoldt 2000)41 Scorpionoid chela a large well-developed chelaformed by tibia (manus + fixed finger) and tarsus(moveable finger) 0 absent 1 present [S 16 WH 60GEWB 62]

250 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs throughout Scorpiones andPseudoscorpiones42 Modified in male as copulatory organ 0 absent 1present [WP 35 WH 39 GEWB 67]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)43 Apotele 0 apparently absent not differentiatedexternally from penultimate article 1 present [GEWB69]The apotele exists as a structure (ie dactyl clawempodium) differentiated from the tibiotarsus tarsusor telotarsus and is typically associated with a pair ofantagonistic muscles The apotele occurs as a distinctstructure in Xiphosura (Snodgrass 1948) Eurypter-ida (Clarke amp Ruedemann 1912) Schizomida (Coken-dolpher amp Reddell 1992) Araneae (Foelix 1996)early divergent Amblypygi (Charinus Weygoldt 2000not Phrynus Shultz 1999) Anactinotrichida (CaminClark amp Bourdeau 1956 Lindquist 1984) Ricinulei(Pittard amp Mitchell 1972 Legg 1976) and Solifugae(Roewer 1934) Giribet et al (2002) coded Solifugae asnot having a palpal apotele but it is present as anadhesive organ (45) Dunlop (1999) reconstructedHaptopoda as having a palpal claw but notes in hisdescription that the distal ends were not preserved inthe known specimens44 Apotele (claw) position 0 terminal 1 subtermi-nal - inapplicable coded only for taxa with a distinctapotele (43)State 1 occurs in Parasitiformes (Acari) (Evans 1992)45 Terminal adhesive organ 0 absent 1 present -inapplicable due to lack of an apotele (43) [GEWB 66]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITES IIIndashVI LEGS 2ndash5 IN NON-ARACHNIDS AND LEGS 1ndash4 IN ARACHNIDS

46 Appendage III (= arachnid leg 1) extremely elon-gate antenniform 0 absent 1 present [WP sim24 + 32S 20 WH 14 GEWB sim84]State 1 occurs in Amblypygi Thelyphonida Schizo-mida (Shear et al 1987) and Opilioacariformes (Vander Hammen 1989) State 1 is approximated in Solif-ugae Palpigradi (Roewer 1934) and Haptopoda(Dunlop 1999) and it is coded as uncertain in thesetaxa47 Appendage V (= arachnid leg 3) of male specializedfor sperm transfer 0 absent 1 present [WP 55 GEWB94]State 1 is known only from extant Ricinulei48 Appendages V and VI (= arachnid legs 3 and 4)with femur shorter than patella and with principal siteof flexionextension at patella-tibia joint (lsquoapatellatersquocondition sensu Van der Hammen 1989) 0 absent 1present [S 25 GEWB 86]

State 1 occurs in Pseudoscorpiones and Solifugae(Shultz 1989)49 Appendage VI (= arachnid leg 4) with terminusmodified as flattened blade 0 absent 1 presentState 1 occurs in some Chasmataspidida (DiploapisDunlop et al 2001 Octoberaspis Dunlop 2002a)uncertain in Chasmataspis Dunlop et al 2004 State0 in Loganamaraspis Tetlie amp Braddy 2004) andEurypterida except stylonuroids (Clarke amp Ruede-mann 1912 Waterston 1979 Tollerton 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI COXA AND BODYndashCOXA JOINT

50 Appendage III (= arachnid leg 1) with coxapophy-sis forming floor or wall of preoral chamber 0 absent1 present [S sim14 WH sim58]State 1 occurs in Opiliones (Hansen amp Soslashrensen 1904)and higher scorpions (Jeram 1998) A lobe is presenton the coxa of leg 1 in the fossil scorpion Prearcturusand perhaps Stoermeroscorpio and is coded here as acoxapophysis Coxapophysis of leg 1 together with theepistome and coxapophysis of the palp form a uniquepreoral chamber the stomotheca Weygoldt (1998) andDunlop amp Braddy (2001) reject the homology of cox-apophyses and stomothecae in Opiliones and Scorpi-ones because these structures are not apparent infossils of those scorpions thought to have been aquaticHowever the coxapophyses in Opiliones are formedlargely from soft cuticular lsquolipsrsquo and similar structuresin fossil scorpions would probably not have beenpreserved51 Appendage IV (= arachnid leg 2) with coxapophy-sis 0 absent 1 present [S sim14 WH sim58 GEWB 75]State 1 occurs in extant Scorpiones and most Opil-iones except Dyspnoi (Shultz 1998)52 At least one pair of coxal gnathobases on append-ages III-VI (= arachnid legs 1ndash4) 0 absent 1 present[S sim23 WH sim66 GEWB 78]State 1 occurs in all appropriately preserved Euryp-terida including Baltoeurypterus (Selden 1981) andStylonurus (Clarke amp Ruedemann 1912) extantXiphosura on appendages IIndashVI and apparentlyProscorpius on appendage III (= arachnid leg 1)(Kjellesvig-Waering 1986 Jeram 1998 contra Giribetet al 2002) Fossilized appendages with gnathobaseshave been found in association with Chasmataspis(Chasmataspidida) suggesting that they may havehad State 1 (Dunlop et al 2004)53 Coxae of appendages IIIndashV (= arachnid legs 1ndash3)with jointed moveable endites 0 absent 1 present [Ssim23]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) and in adequately preserved eurypterids(Clarke amp Ruedemann 1912) (BaltoeurypterusSelden 1981)

PHYLOGENY OF THE ARACHNID ORDERS 251

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

54 Intracoxal muscle a muscle arising on anteriorwall of coxa and inserting on posterior wall 0 absent1 present [GEWB 237]State 1 occurs in Thelyphonida (MastigoproctusShultz 1993) and Amblypygi (Phryus Shultz 1999)It is unclear whether the muscle operating the move-able endite (53) in extant Xiphosura (Manton 1964Shultz 2001) is homologous with this muscle and iscoded here as uncertain It is not present in Steno-chrus (Schizomida) (Shultz unpubl observ)55 Coxae of appendages IIndashVI (= arachnid palp andlegs) with dorsal articulation with carapace 0 absent1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)56 Flabellum (exite) on coxa of appendage VI(= arachnid leg 4) 0 absent 1 present [GEWB sim110]State 1 occurs in extant Xiphosura (Yamasaki et al1988) An apparent exite occurs on appendages asso-ciated with Chasmataspis (Chasmataspidida) (Dunlopet al 2004)57 Insertion process of anteromedial tergo-coxal mus-cle 0 weakly developed 1 well to extremely well devel-oped [GEWB 238]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Boumlrner 1904 Shultz 1999)58 Musculi laterales enlarged lateral tergocoxalmuscle with attachment shifted from coxa to adjacentpleural membrane 0 absent 1 present [S 22 WH 65GEWB 242]State 1 occurs in Araneae and Thelyphonida (Shultz1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI TROCHANTER AND COXAndashTROCHANTER JOINT

59 Coxandashtrochanter joint with complex posteriorarticulation composed of two articulating sclerites 0absent 1 present [S 24 WH 67 GEWB 85]State 1 occurs in Araneae Amblypygi Thelyphonidaand Schizomida (Shultz 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI FEMUR AND TROCHANTERndashFEMUR JOINT

60 Depressor muscle (or homologue) of trochanter-femur joint 0 absent 1 present [WH 42 GEWB 220]State 0 appears to be a unique synapomorphy of Ara-neae (Liphistius Shultz 1989 Aphonopelma Ruh-land amp Rathmayer 1978 Hypochilus present studyother examples Clarke 1984 1986)61 Trochanterndashfemur joint with dorsal hinge or pivotoperated by flexor muscles only 0 absent 1 present[WH sim44 GEWB sim91]State 1 occurs in Palpigradi Other chelicerates have abicondylar articulation and are typically equipped

with antagonistic muscles (Van der Hammen 1985Shultz 1989) but see 60 for an exception62 Superior trochanterndashfemur muscle (or homologue)originating broadly in femur inserting on distal mar-gin of trochanter 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 19892001)63 Basifemurndashtelofemur joint of appendages III andIV (= arachnid legs 1 and 2) in adult 0 absent 1present [S 25 WH 68 GEWB 86]The basifemur and telofemur apparently develop aslsquosisterrsquo articles through division of a parental articlethe femur (eg opilioacariform endeostigmatidpalaeosomatid mites) (Coineau amp Van der Hammen1979 Evans 1992) A single animal can have all legswith divided or undivided femora or have some com-bination of the two typically with the more posteriorlegs having the divided femora Thus anterior legsoften tend to be paedomorphic with respect toposterior legs Most chelicerates have undivided fem-ora on some or all legs but muscles homologous withthose of the basifemurndashtelofemur joint may stilldevelop (66) (Shultz 1989) suggesting that the devel-opment of cuticular and muscular components of ajoint have a degree of developmental and evolutionaryindependence

Undivided femora occur in most extant euchelicer-ates including Xiphosura Palpigradi AraneaeAmblypygi Thelyphonida Schizomida ScorpionesOpiliones Pseudoscorpiones higher Sarcoptiformes(Archegozetes Van der Hammen 1989) many Prostig-mata (Tetranychus Evans 1992) and throughout Par-asitiformes Divided femora occur in all legs in manyearly divergent Acariformes including many Endeo-stigmata basally divergent Sarcoptiformes (egPalaeacarus) and Prostigmata (eg AllothrombiumMicrocaeculus) (Van der Hammen 1989 Evans1992) Only leg 4 has basi- and telofemora in Alycus(Van der Hammen 1989) Basi- and telofemora occuron legs 3 and 4 in Solifugae Ricinulei Opilioacari-formes and Eurypterida (Clarke amp Ruedemann 1912Selden 1981 Shultz 1989)

Giribet et al (2002) coded the lsquoannulusrsquo of certainTrigonotarbida (Shear et al 1987) as a basifemur buta similar non-muscularized annulus is present in theposterior legs of extant Thelyphonida In both casesthe condyles of the trochanterndashannulus and annulusndashfemur joints are aligned in such away that they per-form essentially identical movements and appear notto enhance either the range or degree of leg move-ment [This contrasts with the condition in the lasttwo pairs of legs in Baltoeurypterus where condyles ofthe apparent trochanter-basifemur and basifemurndashtelofemur joints have different arrangements (Selden1981)] The annulus in Thelyphonida appears to func-tion in maintaining constant volume of the tro-

252 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

chanterndashfemur joint by preventing ballooning of thearthrodial membrane thereby minimizing the effect ofinternal fluid pressure on movement64 Basifemurndashtelofemur joint of appendage V(= arachnid leg 3) in adult 0 absent 1 present See 6365 Basifemurndashtelofemur joint of appendage VI(= arachnid leg 4) in adult 0 absent 1 present See6366 Cuticular differentiation of basifemur-telofemurjoint absent but muscles present 0 absent 1 present- inapplicable due to presence of joint (63ndash65)State 1 occurs in appendages of postoral somites IIIndashIV (= arachnid legs 1ndash4) in Xiphosura AraneaeAmblypygi Thelyphonida Schizomida Scorpionesand Pseudoscorpiones (Shultz 1989)67 Circumfemoral ring 0 absent 1 presentThe femora of one or more pairs of legs in anactinot-richid Acari have a basal groove or ring associatedwith slit sensilla Acarologists term that portionproximal to the ring the lsquobasifemurrsquo and that portiondistal to the ring the lsquotelofemurrsquo and they refer tothe basifemur as the lsquosecond trochanterrsquo and thetelofemur as the lsquofemurrsquo Despite the confusingterminology the basifemur discussed in 63ndash65 is nothomologous with the lsquobasifemurrsquo of opilioacariformand parasitiform Acari as both a true basifemurndashtelofemur joint and a circumfemoral ring are presentin legs 3 and 4 of Opilioacariformes (Van der Ham-men 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI PATELLA AND FEMURndashPATELLA JOINT

68 Patella of appendage of postoral somite III(= arachnid leg 1) proportionally much longer thanthose of more posterior appendages 0 absent 1presentState 1 occurs in Thelyphonida Schizomida (Hansenamp Soslashrensen 1905 Shultz 1989) and Opilioacari-formes (Van der Hammen 1989)69 Femurndashpatella joint 0 monocondylar several axesof movement and multifunctional muscles 1 bicondy-lar hinge one axis of movement and antagonistic mus-cles 2 hinge one axis of movement flexor muscleswithout muscular antagonists [S sim26 WH sim69 GEWBsim87]State 0 occurs in Solifugae State 1 occurs in Phalang-ida (Opiliones) Scorpiones and PseudoscorpionesState 2 occurs in the remaining taxa (Shultz 1989)70 Patellar plagula 0 absent or with simple medianattachment 1 symmetrical Y-shaped with long prox-imal stem 2 symmetrical U-shaped (= arcuate scler-ite) 3 asymmetrical attaching to patella only atanterior margin [S 27 WH sim70 GEWB 234 = 243]State 1 occurs in extant Xiphosura (Shultz 1989)State 2 occurs in Araneae (Manton 1958 Ruhland amp

Rathmayer 1978 Clarke 1986 Shultz 1989) State 3occurs in Amblypygi Thelyphonida and Schizomida(Shultz 1989) Giribet et al (2002) included this char-acter twice

APPENDAGES OF POSTORAL SOMITES IIIndashVI TIBIA AND PATELLAndashTIBIA JOINT

71 Tibiae divided by one or more joints 0 absent 1presentState 1 occurs in Amblypygi (Weygoldt 2000)72 Patellandashtibia joint 0 monocondylar with or with-out CZY (73) 1 bicondylar hinge one axis of move-ment antagonistic muscles 2 hinge one axis ofmovement muscles without muscular antagonists [Ssim31 WH 74 GEWB sim92] (Shultz 1989 Selden Shearamp Bonamo 1991)73 Patellandashtibia joint with posterior compression zone(lsquoCZYrsquo) 0 absent 1 present - inapplicable due toabsence of monocondylar articulation [GEWB 99]State 1 is a unique synapomorphy of Araneae (Seldenet al 1991)74 Patellandashtibia joint of appendages IIIndashVI(= arachnid legs 1ndash4) with deep-set monocondylar pivotbordered by a pair of tibial processes to which extensormuscles attach 0 absent 1 present - inapplicabledue to absence of monocondylar joint (72)State 1 occurs in extant Xiphosura (Snodgrass 1952Shultz 1989) although a similar arrangement ispresent in the appendage III (= arachnid leg 1) in The-lyphonida (Shultz 1993)75 Patellandashtibia joint largely immobile specializedfor autotomy 0 absent 1 presentState 1 occurs throughout Amblypygi (Weygoldt 19842000)76 Anterior femurndashtibia or femoropatellandashtibia(transpatellar) muscle 0 absent 1 present [S sim30GEWB 222] (Shultz 1989)77 Proximal attachment of posterior femurndashtibia orfemoropatellarndashtibia (transpatellar) muscle 0 muscleabsent 1 dorsoposterior surface of femur andor pos-terior surface of patella 2 distal process of femur mus-cle acting as extensor of femur-patella joint [S sim28 WHsim71 + 72 GEWB 235 = 244 245] (Shultz 1989)78 Distal attachment of posterior femurndashtibia (trans-patellar) muscle 0 posterior 1 dorsal acting as exten-sor of patellandashtibia joint - inapplicable due to absenceof muscle (77) [S 29 GEWB 221] (Shultz 1989)79 Anterior patellandashtibia muscle 0 absent 1 present[S sim32 WH sim73 GEWB 246] (Shultz 1989)80 Posterior patellandashtibia muscle 0 absent 1 present[S 33 WH 76 GEWB 248] (Shultz 1989)81 Patellandashtibia joint spanned by elastic (lsquospringlikersquo)sclerite 0 absent 1 present [GEWB 107]State 1 is a unique synapomorphy of Solifugae (Sens-enig amp Shultz 2003)

PHYLOGENY OF THE ARACHNID ORDERS 253

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

APPENDAGES OF POSTORAL SOMITES IIIndashVI TARSUS AND TIBIAndashTARSUS JOINT

82 Tarsus divided into proximal basitarsus(= metatarsus) and distal telotarsus (= distitarsus orlsquotarsusrsquo) 0 absent 1 presentState 0 occurs in extant Xiphosura (Shultz 1989) andthroughout Acariformes (Lindquist 1984 Evans1992) in the anterior two pairs of legs in chthonioidpseudoscorpions and all legs in lsquomonosphyronidrsquo pseu-doscorpions including Feaelloidea and Cheliferoidea(Chamberlin 1931) A circumtarsal ring (83) associ-ated with slit sensilla in Parasitiformes (Acari) mayrepresent a joint between a telotarsus and basitarsus(Evans 1992)83 Circumtarsal ring 0 absent 1 presentState 1 is an apparent synapomorphy of Parasiti-formes (Evans 1992)84 Telotarsus in adult with two or more tarsomeres 0absent 1 present - inapplicable due to absence of telo-tarsus (82) [GEWB 71]State 1 occurs in Schizomida Thelyphonida Ambly-pygi Palpigradi Ricinulei Solifugae (Shultz 1989)Opilioacariformes (Van der Hammen 1989) and Ple-siosiro (Dunlop 1999)85 Three telotarsomeres on appendages of postoralsomites IV-VI (= arachnid legs 2ndash4) 0 absent 1present - inapplicable due to absence of telotarsus (82)or absence of telotarsomeres (84) [S 34 WH 77 GEWB102]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Shultz 1989)86 Tibiandashtarsus joint spanned by well-developed elas-tic (lsquospringlikersquo) sclerite 0 absent 1 present [GEWB107]State 1 occurs in Scorpiones phalangid Opiliones andSolifugae (Alexander 1967 Shultz 2000 Sensenig ampShultz 2003) Extant Xiphosura is coded for the lastprosomal appendage only (see 88)87 Appendage of postoral somite VI (= arachnid leg 4)with ring of large basally articulated spatulate pro-cesses at tibia-tarsus joint 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)88 Appendages postoral somites IIIndashV (= arachnidlegs 1ndash3) with tibiotarsus (ie tibia and tarsus not dif-ferentiated) 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 1989) butnot apparently in Weinbergina (Moore et al 2005) theonly synxiphosurid in which appendages are suffi-ciently preserved to determine the number ofpodomeres89 Ambulacrum peduncle-like extension of the tarsuswith internal condylophores and terminating distallywith apotele (eg claws) andor pulvillus 0 absent 1present

State 1 is present throughout Acari (Alberti amp Coons1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI APOTELE AND TARSUSndashAPOTELE JOINT

90 Apotele of appendage III (= arachnid leg 1) 0absent or not apparent 1 present [GEWB 101]State 0 occurs in Amblypygi Thelyphonida and Schi-zomida (Shultz 1999) However the apotele is proba-bly present but undifferentiated because muscles thatwould normally insert on the apotele terminate on theend of the lsquotarsusrsquo Dunlop (1999) reconstructed Hap-topoda as lacking an apotelic claw on leg 1 but noted inthe text that this was speculative Dunlop (2002b) hasreviewed apotelic diversity in Chelicerata91 Appendages of postoral somites IIIndashVI (= arachnidlegs) chelate with chela formed from tibiotarsus andapotele or tarsus and apotele 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988) The condition appears to occur in Chasmataspisbased on associated appendages (Dunlop et al 2004)but not in Diploaspis and Octoberaspis (Dunlop et al2001 Dunlop 2002a)92 Apotele with eversible or padlike empodium(= pulvillus) in adult 0 absent although empodialclaw may be present 1 present [S sim38 WH 81 GEWB95]State 1 occurs throughout Pseudoscorpiones (Cham-berlin 1931) Solifugae (Roewer 1934) Opilioacari-formes Holothyrida Ixodida and free-livingMesostigmata (Evans 1992) An eversible lsquopulvillusrsquo isprobably a primitive feature of Amblypygi (eg Char-inus Weygoldt 2000) but is absent in higher groups(eg Phrynus Weygoldt 2000)93 Inferior apotele muscle (= claw depressor) with tib-ial attachment 0 absent 1 present [S 35 WH 78GEWB 249]The character is coded for extant Xiphosura usingappendage of postoral somite VI (= arachnid leg 4) asother legs lack a differentiated tibia (88) State 1appears to occur throughout Arachnida (Shultz 1989)94 Inferior apotele muscle (= claw depressor) withpatellar attachment 0 absent 1 present [S 36 WH 79GEWB 250]State 0 occurs in extant Xiphosura Acari AmblypygiPalpigradi and Ricinulei State 1 occurs in all otherarachnids examined here (Shultz 1989)

OPISTHOSOMA

95 Number of opisthosomal somites in adult 0 five1 eight 2 nine 3 10 4 11 5 12 6 13 [WH sim33GEWB sim190]Chasmataspidids appear to have 13 opisthosomalsomites (Dunlop et al 2004) The synziphosurans

254 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Weinbergina (Moore et al 2005) and Limuloides(Anderson amp Selden 1997) appear to have ten opistho-somal somites Moore et al (2005) found no evidence ofan anterior microtergite in Weinbergina as proposedby Anderson amp Selden (1997) Xiphosurids appear tohave nine opisthosomal somites (Scholl 1977 Ander-son amp Selden 1997 Shultz 2001) Members of the fol-lowing orders have 12 opisthosomal somites Araneae(Millot 1949ac) Amblypygi (Weygoldt 2000) Thely-phonida and Schizomida (Hansen amp Soslashrensen 1905)and Pseudoscorpiones (Chamberlin 1931) Elevensomites are present in Palpigradi Solifugae (Roewer1934) and Trigonotarbida (Gilboarachne Shear et al1987 Palaeocharinus Fayers et al 2004) Plesiosiromay have 11 or 12 somites (Dunlop 1999) Justifica-tion for coding of more controversial taxa is providedbelow

Scorpiones Two principal hypotheses regarding thenumber of opisthosomal somites in scorpions havebeen advocated a 13-somite hypothesis derived fromembryological studies (Brauer 1895 Patten 1912Farley 1999 2005) and a 12-somite hypothesis basedon comparative anatomy of adults (Weygoldt amp Pau-lus 1979) The embryological interpretation is basedon the observation of pregenital genital and pectinalsomites (each with segmental ganglia and paired limbbuds) in early scorpion embryos followed by extremereduction or loss of the pregenital somite in laterembryos According to this view a missing pregenitalsomite should be added to the 12 apparent opisthoso-mal somites of post-embryonic scorpions to achieve afinal number of 13 The anatomy-based hypothesiswas introduced by Weygoldt amp Paulus (1979) whoadvocated a literal interpretation of post-embryonicsegmentation based on opisthosomal tergites Specifi-cally these authors argued that the last pair of dorsalendosternal suspensor muscles of non-scorpionarachnids especially Pedipalpi attach to the first(= pregenital) somite that this condition also occurs inscorpions and that there is no reason to invoke a miss-ing pregenital tergite They proposed that the pectinesbelong to the genital somite and that functional spe-cializations of the nervous system for pectinal functiongive the apparence of an extra neuromere duringembryonic development

I recently dissected the prosoma and anterioropisthosoma of the scorpions Centruroides Hadrurusand Heterometrus and focused on the composition ofthe muscular diaphragm (103) that separates thehaemocoelic compartments of the prosoma andopisthosoma (Shultz 2007) The diaphragm is com-posed of a metameric series of axial muscles fromthree somites the anterior somite corresponds to thelast prosomal somite and the posterior somite corre-sponds to the genital somite The middle elements

insert dorsally along a tranverse tendon attached tothe anterior margin of the first tergite These obser-vations are consistent with the embryological inter-pretation that a pregenital somite is present but itstergite is not expressed It appears likely that the thepregenital somite was compressed longitudinally dur-ing the evolution of the diaphragm I code scorpions ashaving 13 opisthosomal somites

Eurypterida The eurypterid opisthosoma is widelyassumed to have 12 somites (Clarke amp Ruedemann1912 Stoslashrmer 1944) However in a speculative paperon the evolutionary morphology of trilobites and che-licerates Raw (1957) proposed that both scorpions andeurypterids have 15 opisthosomal somites Rawassumed that scorpions have 13 apparent opisthoso-mal somites based on the transient pregenital somiteof scorpion embryology that scorpions and eurypteridsare close relatives and should have the same numberof somites and that olenellid trilobites and chelicer-ates always have somites in multiples of three Rawachieved 15 somites in scorpions by assuming the lastopisthosomal somite to be a diplosomite and that thetelson is a postanal somite even though there is noevidence for either of these proposals He attributedthese features to eurypterids as well He also notedthat the connection between the prosomal carapaceand first opisthosomal tergite in Eurypterida differedstructurally from the connection between adjacentopisthosomal tergites and regarded this as evidencefor a reduced pregenital tergite in the prosomandashopisthosoma junction Rawrsquos speculations were largelyforgotten until Dunlop amp Webster (1999) resurrectedthe proposal that eurypterids have a reduced opistho-somal tergite and therefore share a unique similaritywith scorpions Unfortunately Dunlop amp Bullocktreated Rawrsquos conjecture as if it were based on empir-ical evidence rather than an attempt to force euryp-terid morphology into a peculiar numerical system Inthe absence of convincing evidence to the contrary Ihave coded eurypterids as having 12 opisthosomalsomites

Ricinulei The opisthosoma consists of a membranouspedicel (97) bearing the gonopore ventrally Thefemale gonopore is bordered by an anterior plate and aposterior plate The remainder of the opisthosoma iscomposed of thick sclerites separated by less heavilysclerotized cuticle The first dorsal sclerite is short andfunctions as part of a prosomandashopisthosoma couplingmechanism (96) that may or may not be a specializedcomponent of the following tergite This is followed byfour tergites and sternites and a three-segmentedmetasoma (= lsquopygidiumrsquo) Millot (1945 1949f) rea-soned that the opisthosoma contains ten somites withsomites VIIndashIX incorporated into the pedicel XndashXIIIexpressed as tergites and sternites and the metasoma

PHYLOGENY OF THE ARACHNID ORDERS 255

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

comprising four somites He did not regard the dorsalcoupling sclerite as a separate tergite and his inter-pretation of four rather than three metasomal somiteshas been rejected Pittard amp Mitchell (1972) also pro-posed ten opisthosomal somites but achieved thisnumber by regarding the dorsal coupling sclerite as atergite of somite IX and by recognizing three metaso-mal somites Legg (1976) adopted the system proposedby Pittard amp Mitchell but did not regard the dorsalcoupling sclerite as separate from the following tergite(X) Van der Hammen (1979 1989) reconstructed 13somites The pedicel was regarded as having twosomites (VII + VIII) the following four sets of tergitesand sternites were considered diplosomites (IX + XXI + XII XIII + XIV XV + XVI) and the metasoma hadthree somites (XVIIndashXIX) This interpretation mayhave been biased by Van der Hammenrsquos attempt tounite Ricinulei with anactinotrichid Acari which healso regarded as having a primitive number of 13somites Selden (1992) described a fossil ricinuleidTerpsicroton that shows two pairs of depressions onthe three large premetasomal tergites which con-trasts with the single pair seen in extant species Thisobservation appears to corroborate the diplosomitehypothesis but the evidence does not indicate that thetergite anterior to these is a diplosomite

Dunlop (1996) attempted to homologize theprosomandashopisthosoma coupling mechanisms ofRicinulei and Trigonotarbida a goal that required anovel and rather forced interpretation of the dorsalsclerites He regarded the coupling sclerite as homol-ogous with the first opisthosomal tergite (VII) of trig-onotarbids and then followed Van der Hammenrsquosdiplosomite hypothesis to achieve 12 somites in totalDunloprsquos scheme differs from previous systems in sug-gesting that the pedicel does not contain the dorsalelements of the first and second somites and is in-consistent with Millotrsquos (1945) observation that thepre- and postgenital plates each have dorsoventralmuscles Here I code Ricinulei as having 12 somitesThere are three metasomal somites (XVIndashXVIII)three diplosomites (= six somites) (XndashXV) one cou-pling somite (IX) and two somites in the pedicel (iethe genital and pregenital somites) (VII VIII)

Opiliones Harvestmen have nine opisthosomalsomites and an anal operculum that is traditionallyregarded as the tergite of a tenth somite a tenth ster-nite is lacking (Hansen amp Soslashrensen 1904 Winkler1957) However the anal operculum appears to repre-sent a persistent embryonic telson (Moritz 1957) andis therefore likely to be a postsegmental structurecomparable with the stinger of scorpions flagella ofthelyphonids etc (Shultz 2000)

Acari The number of opisthosomal somites is probe-matic for most mite taxa due to uncertainty about the

location of the prosomandashopisthosoma boundry (5)extensive simplification or loss of metamericallyarranged sclerites and muscle attachments a paucityof developmental studies of engrailed expression andin Acariformes opisthosomal anamorphosis and het-erochronic modification of somite number (Evans1992) Ixodids appear to be the exception developmen-tal studies indicate five somites in the opisthosoma ofticks (Evans 1992)

Some mites retain external evidence of segmenta-tion ndash metameric patterns of furrows muscle attach-ments and slit sensilla ndash and with certainassumptions the number of opisthosomal somites canbe estimated Here it is assumed that the dorsal sur-faces of the last two prosomal somites are present inmites and retain their primitive association with legs3 and 4 (contra Van der Hammen 1989) (see 5 for jus-tification) Given this there appear to be 11 somites inthe opisthosoma of Opilioacariformes a conclusionalso reached by other workers (eg With 1904 Sitni-kova 1978 Klompen 2000) Similar reasoning sug-gests that Alycus has seven opisthosomal somites notnine as advocated by Van der Hammen (1989) Unfor-tunately external evidence is ambiguous for determin-ing the number of somites in other mite lineages andthese are coded as uncertain96 Prosoma-opisthosoma coupling mechanism 0absent 1 present [GEWB sim24]The posterior margin of the carapace and the anteriormargin of the first apparent opisthosomal tergite arespecialized as a coupling mechanism in Ricinulei andTrigonotarbida and Dunlop amp Horrocks (1996) pro-posed that these are synapomorphic for the twogroups However there is uncertainty about thehomology of the anterior opisthosomal somites in Trig-onotarbida and Ricinulei (95)97 Pedicel 0 absent 1 aranean type 2 ricinuleidtype [WP 30 S sim40 WH sim20 GEWB sim126]The body narrows at or near the prosomandashopisthosoma juncture in several arachnid lineages (ieSolifugae Palpigradi Amblypygi Araneae Ricinulei)and this lsquowaistrsquo has often been used as character at theinterordinal level (eg Pocock 1893) However thisinterpretation is rejected here for being subjective andfor uniting non-homologous conditions For exampleAraneae and Amblypygi are often grouped on the basisof a lsquopedicelrsquo yet it is a highly specialized structure inAraneae and its parts are not readily homologizedwith those of Amblypygi In contrast the condition inAmblypygi is a slightly narrower version of the highlymoveable prosomandashopisthosoma juncture in Uropygiwhich is not generally considered a lsquopedicelrsquo Thepedicel in Ricinulei is also unique a weakly sclero-tized stalk containing the genital opening (Cryptocel-lus Pittard amp Mitchell 1972 Ricinoides Legg 1976)See 95

256 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

98 Opisthosoma with three- segmented lsquobucklerrsquo 0absent 1 presentState 1 occurs throughout Chasmatapsidida (Dunlop2002a)99 Thoracetron consolidation of tergites of postgeni-tal somites 0 absent 1 present - inapplicable codedonly for Xiphosura (Anderson amp Selden 1997)100 Fusion of tergites of postoral somites VIII and IX(= opisthosomal somites 2 and 3) only 0 absent 1present [GEWB 146]State 1 appears to be a synapomorphy Trigonotarbida(Shear et al 1987) Ricinuleids also have diplotergites(101) but these appear to encompass a different com-bination of somites (see 95)101 Three diplotergites 0 absent presentState 1 is a unique synapomorphy of Ricinulei (Selden1992) See 95102 Paired opisthosomal defensive glands openingvia ducts on either side of the anus 0 absent 1present [S 46 WH 86 GEWB 122]State 1 occurs in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905)103 Muscular diaphragm separating prosomal andopisthosomal compartments formed by dorsoventralmuscles of postoral somites VIndashVIII and extrinsic mus-cles of leg 4 0 absent 1 presentState 1 is known only in extant scorpions (Lankesteret al 1885 Firstman 1973) The diaphragm is oftentreated as a single structure but recent anatomicalwork (Shultz 2007) has shown that it is a composite ofdorsal endosternal suspensors and extrinsic leg mus-cles (see 95) A diaphragm is also present in Solifugaebut it is located more posteriorly and does not appearto be homologous with that of scorpions (Roewer1934)104 Opisthosomal appendicular chondrites 0absent 1 presentThese cartilage-like columns of mesodermally derivedtissue are associated with each opisthosomal append-age including the chilaria in extant Xiphosura(Patten amp Hazen 1900 Yamasaki et al 1988 Fahr-enbach 1999 Shultz 2001)105 Paired appendages on ventral surface of postoralsomite VII (= opisthosomal somite 1) in adult 0absent 1 present [WP sim13 S 39 WH 82 GEWB 143]State 1 is known in Weinbergina (Moore et al 2005)and extant Xiphosura The bilobed structure of themetastoma in certain eurypterids as well as the cor-responding placement of chilaria in Xiphosura haveled some workers to regard the metastoma as fusedappendages of postoral somite VII (eg Stoslashrmer1955) This hypothesis is problematic given that (i) nospecial explanation is needed to account for bilaterallysymmetrical structures in bilaterally symmetricalorganisms (ii) incorporation of the ventral part of thefirst opisthosomal sternite as a functional element of

the prosoma is typical of arachnids and perhapsEuchelicerata generally and (iii) some workers regardthe metastoma as a sternite (eg Jeram 1998) Thusthe metastoma is coded here as uncertain Dunlop hasproposed that the sternum of scorpions is derivedfrom appendages citing the presence of transient limbbuds in scorpion development (Brauer 1895 Patten1912) but this is not compelling evidence for a persis-tent appendicular contribution to the sternite of theadult106 Megoperculum 0 absent 1 present [S sim41 + 42WH sim83 GEWB 162 + 163]State 1 occurs in Thelyphonida Schizomida Ambly-pygi Araneae (Shultz 1993 1999) Trigonotarbida(Shear et al 1987) and probably Haptopoda (Dunlop1999) and Palpigradi The megoperculum consists ofappendages of opisthosomal somite 2 (= genital somiteor postoral somite VIII) (Shultz 1993 1999 Popadicet al 1998) that have fused medially and displacedthe ventral body wall of the somite anteriorly (oftenrepresented by a small sternite to which dorsoventralmuscles of postoral somite VIII attach) The megoper-culum projects posteriorly to form the ventral surfaceof a pregenital chamber The dorsal surface of thechamber is formed by the ventral body wall of opistho-somal somite 3 (= postoral somite IX) which serves asthe ventral attachment of the paired dorsoventralmuscles of that somite The megoperculum bears book-lungs (where present) and sometimes paired gonopodsthat are probably derived from telopodites A similarbut less developed operculum is present on theopisthosomal somite 3 which may also bear booklungsand eversible vesicles (eg Amblypygi) (see 111) corre-sponding to the booklungs and gonopods of the mego-perculum (Shultz 1999)

I contend here that a megoperculum is present inPalpigradi and that this is particularly evident infemales Specifically a large unpaired lobe projectsposteriorly from the genital somite over the ventroan-terior surface of the first postgenital somite therebyforming a pregenital chamber The dorsoventral mus-cles of the genital somite attach ventrally near theanterior border of postoral somite VIII and those ofpostoral somite IX attach to the upper surface of thepregenital chamber (Rucker 1901 Boumlrner 1902bRoewer 1934 Millot 1949d Van der Hammen 1989)Weygoldt (1998) has questioned this interpretation ofPalpigradi107 Postgenital operculum or lsquosternitersquo 0 sclerotized1 not sclerotized - inapplicable coded only for Pan-tetrapulmonata (Platnick amp Gertsch 1976)State 1 is present in Trigonotarbida Amblypygi The-lyphonida Schizomida and Mesothelae (Araneae) butnot Opisthothelae (Araneae)108 Genital opening of female guarded by four plates(one pregenital one postgenital two laterogenitals)

PHYLOGENY OF THE ARACHNID ORDERS 257

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

genital opening of male guarded by two plates 0absent 1 presentThis an apparent synapomorphy of Holothyrida(Acari) (Van der Hammen 1989)109 Anterior margin of genital opening in male withglands secreting via fusules 0 absent 1 presentState 1 occurs in Araneae as epigastric or lsquoepiandrousrsquoglands (Marples 1967) and in Palpigradi (Condeacute1991b)110 Paired valve-like plates apparently formed fromcomponents of three somites covering triradiate genitalopening 0 absent 1 presentThis is an apparent synapomorphy of Acariformes(Evans 1992 Alberti amp Coons 1999)111 Eversible lsquoappendagesrsquo on the ventral surface ofpostgenital somites 0 absent 1 present [S sim43 WHsim84 GEWB sim127]Paired ventral lsquoappendagesrsquo operated in part byhaemolymph pressure occur in the form of ventralsacs in certain Amblypygi (postoral somite IX) (Wey-goldt 2000) (Charinus Millot 1949b not PhrynusShultz 1999) prokoenenian Palpigradi (postoralsomites XndashXII) (Rucker 1901 Condeacute 1991a) as spin-nerets in Araneae (XndashXI) (Shultz 1987) and as geni-tal papillae in many Acariformes (Alberti amp Coons1999) including Oribatida many Endeostigmata (Aly-cus Van der Hammen 1989) and Prostigmata (Allo-thrombium Saboori amp Kamali 2000 MicrocaeculusEvans 1992) Eversible vesicles similar to those ofamblypygids occur in fossils of the trigonotarbidPalaeocharinus (postoral somite IX) (Fayers et al2004) Two cuticular structures (lsquogenital verrucaersquo)occur anterior to the genital opening in Opilioacari-formes and each covers the opening to a thin-walledinvaginated sac (lsquogenital papillarsquo) of unknown function(Van der Hammen 1989) Van der Hammen proposedthat it is homologous to a genital papilla of Acari-formes but Alberti amp Coons (1999) have questionedthis interpretation on several grounds and refer tothese structures as pregenital capsules112 Opisthosomal silk glands and spinnerets derivedfrom appendages on postoral somites X and XI(= opisthosomal somites 4 and 5) 0 absent 1 present[WP 33 WH 41 GEWB 123 + 142]State 1 is a unique synapomorphy of Araneae (Plat-nick amp Gertsch 1976) although potentially homolo-gous glands and spigot-like setae occur on the ventralopisthosomal surface in Palpigradi (Millot 1943Condeacute 1991a)113 Opisthsosomal spinnerets location 0 near mid-dle of opisthosoma 1 near posterior end of opisthso-som ndash inapplicable coded only for Araneae (Platnickamp Gertsch 1976)114 Anterior medial lsquospinneretsrsquo 0 absent 1 present- inapplicable coded only for Araneae (Platnick ampGertsch 1976)

115 Opisthosomal tergites divided longitudinally intoone median and two lateral plates 0 absent 1 present[GEWB 145]State 1 occurs in non-curculioid Ricinulei (Selden1992) and many Trigonotarbida (Dunlop 1996) Thisfeature or something very like it (eg distinct opistho-somal trilobation) occurs sporadically in severaleuchelicerate groups including Chasmataspis (Chas-mataspidida) (Dunlop et al 2004) most fossil xipho-surans (Anderson amp Selden 1997) and a few derivedEurypterida (especially Mixopteroidea) (Tollerton1989)116 Number of metasomal somites 0 zero 1 two 2three 3 five 4 nine [S sim44 WH sim85 GEWB 128 144]The metasoma is a preanal region comprising multiplesomites that are substantially narrower than the pre-ceding somites or mesosoma Metasomal somitesoften lack pleural membranes and take the form ofsclerotized rings State 1 occurs in TrigonotarbidaState 2 is present in synziphosurid Xiphosura (Ander-son amp Selden 1997) Amblypyi Thelyphonida Schizo-mida and Ricinulei State 3 occurs in Scorpiones andEurypterida and State 4 occurs in Chasmatispidida

Some workers appear to consider a five-segmentedmetasoma as conclusive evidence for the monophylyof a Eurypterida + Scorpiones clade (eg Dunlop ampBraddy 2001) presumably because the character rep-resents a kind of tagmosis and should therefore begiven substantial phylogenetic weight However Wey-goldt (1998) has pointed out that a metasoma contain-ing three somites is probably part of the ground planof Euchelicerata The known diversity of the eucheli-cerate metasoma indicates that this feature can inincrease or decrease its segmental composition in evo-lution or be eliminated entirely without necessarilychanging the total number of opisthosomal somitesConsequently it is problematic to assume that thefive-segmented metasoma has any special immunityto homoplasy117 Postanal structure (telson) 0 absent or not obvi-ously developed 1 presentPostanal structures are considered non-segmentalbecause they are modifications of the embryonicregion posterior to the site of somite addition (embry-onic growth zone) A well-developed postanal struc-ture persists in adults in Xiphosura EurypteridaChasmataspidida Scorpiones Schizomida Thely-phonida and Palpigradi The anal operculum of Opil-iones is generally considered by morphologists to bethe tergite of the tenth opisthosomal or anal somite(eg Hansen amp Soslashrensen 1904) but embryological evi-dence (Moritz 1957) suggests that it is derived fromthe postproliferative zone and thus corresponds to thetelson of other chelicerates (see 95)118 Postanal structure shape 0 caudal spine 1aculeus 2 flagellum 3 anal operculum - inapplica-

258 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ble due to absence of postanal structure (117) [WP sim41S sim45 WH sim24 GEWB sim121 129 + 147]State 0 is present in Xiphosura Eurypterida andChasmataspidida State 1 occurs in Scorpiones State2 occurs in Palpigradi Thelyphonida and SchizomidaState 3 occurs in Opiliones (see 95 117)119 Specialized postanal flagellum in male (see 157ndash159) 0 absent 1 present - inapplicable coded onlyfor taxa with postanal flagellum (118) [GEWB 131]State 1 is a unique synapomorphy of Schizomida(Hansen amp Soslashrensen 1905 Cokendolpher amp Reddell1992 Reddell amp Cokendolpher 1995)

RESPIRATORY SYSTEM

120 Respiratory medium 0 water 1 airAmong the terminal taxa included only fossil scorpi-ons are problematic for this character They are codedhere as uncertain Based on a study of book lungmicrosculpture Scholtz amp Kamenz (2006) have arguedthat arachnids are primitively terrestrial and pulmo-nate (see also Firstman 1973) and have questionedwhether any fossil scorpions were aquatic No positionon this proposal is taken here121 Respiratory lamellae on opisthosomal somite 2(= genital somite postoral somite VIII) 0 absent 1present [WP sim37 S sim51 WH sim22 GWEB sim133ndash137]This character encompasses book gills (= lamellae thatfunction in water) and book lungs (= lamellae thatfunction in air) The distinction between book gills andbook lungs is accommodated here by 120 in combina-tion with 121ndash124 Respiratory lamellae are presenton the genital somite in Trigonotarbida AraneaeAmblypygi Thelyphonida and Schizomida Petrunk-evitch (1949) reconstructed Plesiosiro as having booklungs but Dunlop (1999) could not corroborate thisThis character was used by Dunlop amp Webster (1999)to propose that Xiphosura and Scorpiones are closelyrelated because they both lack respiratory lamellae onthe genital somite Dunlop amp Braddy (2001) alsoargued for the placement of Eurypterida with Xipho-sura and Scorpiones based in part on this characterAt least some eurypterids may have had respiratorylamellae (eg Manning amp Dunlop 1995) but the onlyevidence of their segmental distribution is derivedfrom one specimen (Braddy et al 1999)122 Respiratory lamellae on opisthosomal somite 3(= postoral somite IX) 0 absent 1 present [GEWB136]State 1 occurs in Amblypygi Thelyphonida and Xipho-sura It is a groundplan feature of Araneae (Platnick ampGertsch 1976) and is present in all representativetaxa included here Braddy et al (1999) proposed theexistence of respiratory lamellae on postoral somitesIXndashXII in Eurypterida based on evidence from onespecimen This character is coded as uncertain for the

Eurypterida included here Extant scorpions are codedhere as lacking respiratory lamellae on this somite(= pectinal somite) based on the interpretation of theopisthosomal segmentation discussed in 95123 Respiratory lamellae on opisthosomal somites 4ndash6 (= postoral somites XndashXII) 0 absent 1 presentState 1 is definitely present in extant Xiphosura andScorpiones and may have occurred in Eurypterida124 Respiratory lamellae on opisthosomal somite 7(= postoral somite XIII)State 1 occurs in extant Xiphosura and Scorpionesonly125 Kiemenplatten 0 absent 1 presentThese structures are located on the roof of opercularchambers in Eurypterida (Clarke amp Ruedemann1912) where they take the form of ventrally projectingcones with a distinct cuticular microsculputure (Man-ning amp Dunlop 1995) Dunlop amp Braddy (2001)inferred the existence of Kiemenplatten in all Palaeo-zoic scorpions based on a description and photos of onespecimen of Paraisobuthus duobicarinatus by Kjelles-vig-Waering (1986 pls 16ndash18) The plates depict darkcone-like rearward-pointing denticles distributedwithin a white amorphous material The denticlesappear to lack microsculpture even though the mag-nifications at which the photos were taken (times90ndashtimes330)are comparable with those illustrating the cones ofKiemenplatten (times170ndashtimes350 in Manning amp Dunlop1995 figs 1 2) Given the diversity of cuticular struc-tures present in the atria of booklungs book gills andtracheae the denticles appear to bear no special sim-ilarity to the cones of Kiemenplatten126 Tracheal system 0 absent 1 paired ventral stig-mata on postoral somite VIII (= opisthosomal somite2) 2 paired ventral stigmata on postoral somites IXand X 3 one pair of stigmata opening near legs 3 or 44 paired stigmata associated with chelicerae 5 fourpairs of stigmata on dorsal surface of opisthosoma -inapplicable aquatic (120) [WP 40 sim43 S sim52 + 5354 WH sim23 45 88 GEWB sim138 139]Firstman (1973) and Weygoldt amp Paulus (1979)hypothesized that tracheae are homologous in all tra-cheate arachnids except spiders and subsequentworkers have entertained this hypothesis by includinga character for the presenceabsence of tracheal sys-tems that ignores the diverse arrangement of stig-mata in arachnids (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) However thisapproach assumes that internal tracheal systems areconserved but that tracheal openings (stigmata)appear and disappear on different parts of the bodywith higher evolutionary frequency Here I assumethat stigmata are conserved in evolution and that dif-ferences in their anatomical placement reflect the evo-lution of new tracheae State 1 occurs throughoutOpiliones State 2 occurs throughout Solifugae

PHYLOGENY OF THE ARACHNID ORDERS 259

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

(Roewer 1934) and Pseudoscorpiones (Chamberlin1931) State 3 recognizes the possible homology of stig-mata in Ricinulei Opilioacariformes and Parasiti-formes (Acari) (Lindquist 1984 Van der Hammen1989) Tracheal systems are apparently absent in thegroundplan of Acariformes but Prostigmata typicallyhave tracheae associated with the chelicerae Tra-cheae derived from the posterior pair of book lungs arewidespread in araneomorph Araneae but the booklungs are retained and tracheae are absent in themost basally divergent groups (eg Hypochilus)

BOXndashTRUSS AXIAL MUSCLE SYSTEM (BTAMS)

127 Posterior oblique muscles of BTAMS of postoralsomites IndashVI 0 absent 1 present in one or moresomitesState 0 occurs in Xiphosura and state 1 occurs inPalpigradi Araneae Amblypygi Thelyphonida andSchizomida (Shultz 2001) The condition in AcariRicinulei Opiliones and Solifugae is not known128 Anterior oblique muscles of BTAMS posterior topostoral somite VI 0 absent 1 presentState 1 occurs in Xiphosura which is probably theprimitive condition based on comparison with otherarthropods (Shultz 2001) State 0 occurs in all arach-nids examined thus far129 Ventral attachments of posterior oblique musclesof opisthosomal BTAMS located in prosoma 0 absent1 presentState 1 occurs in extant Xiphosura (Shultz 2001)130 Endosternite fenestrate 0 absent 1 present [S 8WH 54 GEWB 35]State 1 occurs in Thelyphonida and hubbardiid Schi-zomida (Firstman 1973) The condition in protoschi-zomid Schizomida is not known131 Suboral suspensor a tendon that arises from theBTAMS and inserts on the ventral surface of the oralcavity via muscle 0 absent 1 presentState 1 occurs in Palpigradi (Eukoenenia Millot1943) Amblypygi (Shultz 1999) and Thelyphonida(Shultz 1993)132 Perineural vascular membrane in adult 0absent 1 present [WH 28 GEWB 155]Wheeler amp Hayashi (1998) coded extant Xiphosura asunknown although the primary source for this char-acter (Firstman 1973) stated that adult Limulus havea perineural vascular membrane Within Arachnidapresence of a perineural vascular membrane is appar-ently correlated with the presence of tracheae (126)133 Ventral endosternal suspensor attaching on coxaof anteriorly adjacent somite 0 absent 1 present[GEWB 239 253]State 1 occurs in Amblypygi (Shultz 1999) Thely-phonida (Shultz 1993) and Schizomida (Shultzunpubl observ)

134 Posteriormost postoral somite with a pair of dor-soventral muscles 0 VI 1 VII 2 VIII 3 XII 4 XIII5 XIV 6 XV 7 XVIDorsoventral muscles tend to run in a continuousmetameric series beginning in the prosoma Deter-mining the last tergite on which the muscle seriesends is substantially easier than counting especiallygiven frequent anatomical complexities near theprosomandashopisthosoma juncture The dorsoventralmuscle series ends on postoral somite VII inCyphophthalmi (eg Chileogovea) and Laniatores(Opiliones) (Shultz unpubl observ) on postoralsomite VII or VIII in Eupnoi (Leiobunum Shultz2000 unpubl observ) on postoral somite XII in Pal-pigradi (Roewer 1934) Solifugae (Bernard 1896Roewer 1934 Millot amp Vachon 1949) and Araneae(Liphistius Hypochilus Millot 1933 but segmenta-tion reinterpreted here following Shultz 1993 1999)on postoral somite XIII in hubbardiid Schizomida(Cokendolpher amp Reddell 1992) and Scorpiones(Lankester et al 1885) and on postoral somite XIVin extant Xiphosura (Shultz 2001) Thelyphonida(Shultz 1993) protoschizomid Schizomida (Cokendol-pher amp Reddell 1992) and Amblypygi (Shultz 1999)Dunlop (1999) reconstructed Plesiosiro as havingpaired tergal apodemes ending on postoral somite XVbut his figures indicate that the series ends on posto-ral somite XIV The interpretation of opisthosomalsegmentation in Ricinulei coded here (95) indicatesthat the series ends on postoral somite XV in Terpsi-croton and this is treated as the groundplan for theorder The posteriormost dorsoventral muscles inPseudoscorpiones occur on postoral somite XVI(Vachon 1949)

NERVOUS SYSTEM

135 Segmental ganglia 0 consolidated in prosoma1 one or more present in opisthosoma [WH 30 GEWB210 + 216]Adult neuromeres are exclusively prosomal in Ambly-pygi liphistiomorph and araneomorph Araneae Opil-iones Pseudoscorpiones Ricinulei and Acari (Millot1949a)136 Dorsal median eyes 0 absent 1 present [WP14 + 47 S 50 WH sim7 GEWB 1]Median eyes are a groundplan feature of Euchelicer-ata and are retained in extant Xiphosura Eurypteridaand occur in all arachnid orders except PalpigradiSchizomida Ricinulei and Pseudoscorpiones (Paulus1979) Median eyes are also absent in opilioacariformand parasitiform Acari (Lindquist 1984) and perhapsthe fossil whipscorpion Proschizomus (Dunlop amp Hor-rocks 19951996) They are unknown in synziphosu-ran Xiphosura Median eyes are present in a variety ofbasal acariform mites including Endeostigmata (but

260 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

not Alycus) Prostigmata (Microcaeculus) and certainoribatids (Palaeacarus) (Evans 1992 Alberti amp Coons1999) Giribet et al (2002) coded all representativeAcari as lacking median eyes It is not known whetherthe eyes of some cyphophthalmid opilions are medianor lateral evidence from tracheal branching inCyphophthalmus (Janczyk 1956) suggests that theyare median eyes (Shultz amp Pinto da Rocha 2007) andpresence of a tapetum in Stylocellus is consistent withlateral eyes (Shear 1993)137 Retinula cells of dorsal median eyes 0 organizedinto closed rhabdoms 1 organized into network ofrhabdomeres 2 disorganized - inapplicable due toabsence of median eyes (136) [GEWB sim3]State 0 is present in Scorpiones Thelyphonida andAmblypygi State 1 is present in Solifugae and Ara-neae state 2 is present in Xiphosura (Paulus 1979)The retinula cells of median eyes have been studied inseveral prostigmatid Acariformes and are organized ina network in some taxa and in an irregular pattern inothers (Alberti amp Coons 1999) Retinulae in phalangidOpiliones have state 0 proximally and state 2 distally(Schliwa 1979)138 Ventral median eyes 0 absent 1 presentState 1 occurs in early instars of extant Xiphosura(Paulus 1979)139 Lateral eyes 0 absent 1 presentLateral eyes are primitively present in Chelicerataand are absent in Palpigradi and Opiliones (Paulus1979) It is unclear whether the eyes of cyphoph-thalmid opilions are median or lateral (see 136)140 Arrangement and number of lateral eyes 0 com-pound many 1 five or more pairs (includes micro-lenses) 2 three primary pairs (excludes microlenses)3 two pairs 4 one pair - inapplicable due to absenceof lateral eyes (139) [WP sim13 + 18 + 38 + 44 + 52 Ssim49 WH sim10 GEWB sim4 + 5]True compound eyes are present in XiphosuraEurypterida Chasmataspidida and many PalaeozoicScorpiones Among the anactinotrichid Acari Opilioa-cariformes have two (Neocarus) or three (Siamacarus)pairs of lateral eyes (although at least one species ofSiamacarus lacks eyes) the allothryrid Holothyridaand many Ixodida have a single pair of lateral lensesand the remainder apparently lack eyes (Evans1992) Thelyphonida was coded as lsquo12rsquo to reflect fivepairs of lenses comprising three pairs of primarylenses and two pairs of small accessory lenses (notsimply three pairs as coded by Giribet et al 2002)Trigonotarbida is also coded as lsquo12rsquo to reflect threepairs of primary lenses and multiple small accessorylenses Dunlop (1999) illustrated Haptopoda as havingpaired triads of lateral eyes but noted that there wasactually no evidence of this in the fossils The lateraleyes of fossil ricinuleids have two pairs of lenses(Selden 1992) One pair of eyes or eyespots are

present in extant Ricinulei and many Schizomidaalthough five genera of hubbardiid Schizomida have apair of lenses (Reddell amp Cokendolpher 1995) State 1is the groundplan for extant Scorpiones and State 3occurs in the pseudoscorpions Chthonius FeaellaNeobisium (but State 4 in Chelifer) and Solifugae(Paulus 1979)141 Lateral eyes with closed rhabdoms 0 absent 1present - inapplicable due to absence of lateral eyes(139) [WP 21 WH 12 GEWB 6]Presence of closed rhabdoms is probably primitive forChelicerata and is retained in extant Xiphosura andScorpiones Retinula cells form a network of rhab-domeres in other extant chelicerates (Paulus 1979)but these networks can differ substantially in detailWeygoldt (1998) and other workers have given sub-stantial weight to the network character in unitingnon-scorpion arachnids but these authors seem not togrant comparable phylogenetic significance to theanalogous character of the median eyes shared bySolifugae and Araneae (137)142 Slit sensilla 0 absent 1 present [WP 19 S 47WH 11 GEWB 209]State 1 occurs in all arachnid orders except Palpigradi(Shultz 1990) The proposal that lsquoprimitiversquo slit sen-silla were present in Eurypterida (Dunlop amp Braddy1997) appears to be based on the over-interpretation ofa comparatively large notch that occurs in Baltoeu-rypterus at the terminus of the tibia (= podomere 7)(see also Edgecombe et al 2000 Giribet et al 2002)The neural construction of slit sensilla like function-ally similar campaniform sensilla of hexapods is sim-ilar to that of trichoid sensilla (Chapman 1998Klompen 2000) Thus these cuticular stress receptorsmay represent modified bases of sensory setae a viewsupported by the replacement of setae by slit sensilladuring post-embryonic development in some Opilioac-ariformes (Klompen 2000)143 Trichobothria 0 absent 1 present [GEWB 213]Trichobothria are present in extant Scorpiones(Jeram 1998) Pseudoscorpiones (Chamberlin 1931)some endeostigmatid (Alycus Van der Hammen 1989)and prostigmatid Acariformes and most Oribatida(Lindquist 1984) They occur on the ventral surface ofthe pedal femora in the opilioacariform Siamacarus(Leclerc 1989) They are also present in Araneae (Foe-lix 1996) Amblypygi (Weygoldt 2000) Schizomidaand Thelyphonida (Hansen amp Soslashrensen 1905) Theyare apparently absent in all non-arachnid chelicer-ates Solifugae Ricinulei Opiliones (Reissland ampGoumlrner 1985) and Parasitiformes (Acari) (Lindquist1984)144 Tibial trichobothria with 2-1-1-1 pattern onappendages IIIndashVI (= arachnid legs 1ndash4) 0 absent 1present - inapplicable due to absence of trichobothria(143) [S 48 WH 87 GEWB 88]

PHYLOGENY OF THE ARACHNID ORDERS 261

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs only in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905) Note that Shultz (1990)erroneously described this character as a 2-2-1-1 pat-tern and that this error was repeated by Giribet et al(2002) and described as 2-1-1 by Wheeler amp Hayashi(1998)145 Paired trichobothria on dorsal surface ofprosoma 0 absent 1 present - inapplicable due toabsence of trichobothria (143)State 1 is an apparent synapomorphy of Acariformesand is present in all acariforms included in this study(Alberti amp Coons 1999)146 Malleoli 0 absent 1 present [WP 45 GEWB 96]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)147 Pectines 0 absent 1 present [WP 20 WH 38GEWB 120]State 1 is a unique synapomorphy of Scorpiones148 Intercheliceral median organ 0 absent 1present [GEWB 212]State 1 is a unique synapomorphy of Palpigradi(Roewer 1934)149 Tarsal organ on appendage of postoral somite III(= arachnid leg 1) (= Hallerrsquos organ) sensilla containedwithin a cuticular depression on the superior surface ofthe tarsus of appendage III (= arachnid leg 1) 0absent 1 present [GEWB sim100]Klompen (2000) has noted that State 1 occurs in Opil-ioacariformes Parasitiformes (except Mesostigmata)and Ricinulei where it also occurs on leg 2 (Talaricoet al 2005) State 1 occurs on all legs in Araneae andperhaps Scorpiones (Foelix 1985)150 Tarsal organ on appendage of postoral somite IV(= arachnid leg 2) 0 absent 1 present (See 149)

REPRODUCTION

151 Gonads 0 primarily prosomal 1 primarilyopisthosomalState 0 is limited to Xiphosura (Sekiguchi 1988)State 1 occurs throughout Arachnida (Millot1949a)152 Ladder-like opisthosomal gonadsaccessoryglands (see 153) 0 absent 1 present [WH sim37 GEWBsim158]Giribet et al (2002) followed Wheeler amp Hayashi(1998) who followed Clarke (1979) in coding gonads asreticulate (Xiphosura) ladder-like (Scorpiones Thely-phonida Schizomida) or lsquosaccularrsquo (all remainingArachnida) However comparisons between Xipho-sura and Arachnida are problematic given that thexiphosuran gonads are primarily prosomal and thoseof arachnids are primarily opisthosomal (151) Fur-ther the reticulate pattern in Limulus (Xiphosura)but not other extant xiphosurans contains a distinctlyladder-like component The lsquosaccularrsquo state is probably

artificial as it encompasses a wide variety of pairedand unpaired structures153 Male gonads in two distinct parts one producingsperm and another (tubular gland) producing a holo-crine secretion similar to degenerate sperm 0 absent1 presentState 1 occurs in Thelyphonida Schizomida andAmblypygi although the holocrine material is pro-duced by ventral organs in Amblypygi and dorsalorgans in Thelyphonida and Schizomida (Alberti2005)154 Number of gonopores 0 two 1 oneExtant xiphosurans have two small genital openingson the base of the genital telopodite and all extantarachnids have a single opening The condition inEurypterida is not known Clarke amp Ruedemann(1912) located a pair of openings near the base of themedian organ (161) that are the outlets of the lsquohornorgansrsquo but it is unclear whether these are genitalducts or accessory structures Braddy amp Dunlop (1997)have developed numerous speculations about thesestructures and extended their arguments far beyondthe available evidence155 Genital opening (ie gonopore or gonostome)appearing to open in prosomal region (ie between legcoxae or anterior to posterior carapacal margin) 0absent 1 present [WP sim50 WH 26 GEWB 166]The genital opening in Euchelicerata is located onpostoral somite VIII (= opisthosomal somite 2) but ithas shifted anterior to the posterior margin of thecarapace or between the coxae of the last pair of legsin most Scorpiones (but not in PalaeoscorpiusKjellesvig-Waering 1986) and Opiliones The geni-tal opening occurs near or anterior to the last coxaein the opilioacariform and parasitiform Acari repre-sented here It is variable in Acariformes but islocated posterior to the coxae in all representativetaxa156 Ovipositor 0 absent 1 present [WP sim51 S sim60WH sim91 GEWB sim172]An ovipositor with a trilobed terminus is an apparentgroundplan character of Oribatida (Lindquist 1984Alberti amp Coons 1999) An ovipositor is also present inOpilioacariformes and Opiliones (Van der Hammen1989)157 Stalked spermatophore attached to substratum0 absent 1 present [S 57]State 1 occurs in Scorpiones PseudoscorpionesAmblypygi Thelyphonida and Schizomida The mech-anism of sperm transfer is unknown in PalpigradiOpilioacariformes and Holothryrida158 Male turns posterior end toward female duringmating behaviour 0 absent 1 presentState 1 occurs in Amblypygi (Weygoldt 2000) The-lyphonida and Schizomida (Weygoldt amp Paulus1979)

262 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

159 Female grasps male opisthosoma during matingbehaviour 0 absent 1 present [WP 29 S 58 WH 19GEWB 188]State 1 occurs in Thelyphonida and Schizomida (Wey-goldt amp Paulus 1979) Evolution of this behaviour wasprobably facilitated by 158160 Penis 0 absent 1 present [WP sim51 GEWB 167]State 1 is a unique synapomorphy of Opiliones A truepenis occurs in Phalangida A clearly homologousstructure is present in Cyphophthalmi (Opiliones) andapparently functions in depositing a spermatophore inthe femalersquos genital chamber (Karaman 2005) A truepenis may occur in some mites but it does not appearto be a groundplan feature of any major group (Evans1992) The lsquopenisrsquo in Oribatida is really a spermatop-ositor it functions in construction of a spermatophore(Alberti amp Coons 1999)161 Median organ 0 absent 1 presentState 1 occurs throughout Eurypterida (Clarke amp Rue-demann 1912) and has been observed in diploaspididChasmataspidida (Dunlop 2002a also Loganamaras-pis Tetlie amp Braddy 2004)

SPERM MORPHOLOGY

162 Nucleus with manchette of microtubules 0absent 1 present [S 54 WH 89 GEWB 193]State 1 occurs in Araneae Amblypygi Thely-phonida Schizomida Ricinulei and gonyleptid Lani-atores (Opiliones) (Alberti 1995 Giribet et al2002)163 Axoneme 0 absent 1 present [WP sim49 S sim55WH sim25 GEWB 195]State 1 occurs in Xiphosura Scorpiones Cyphoph-thalmi (Opiliones) Pseudoscorpiones RicinuleiAraneae Amblypygi Thelyphonida and Schizomida(Alberti 1995)164 Coiled axoneme 0 absent 1 present - inappli-cable due to absence of axoneme (163) [WP 22 S sim55WH 13 GEWB 196]State 1 occurs in Pseudoscorpiones Ricinulei Ara-neae Amblypygi Thelyphonida and Schizomida(Alberti 1995)165 Microtubule arrangement in axoneme 0 9 + 0 19 + 1 2 9 + 2 3 9 + 3 - inapplicable due to absence ofaxoneme (163) [WP sim26 S sim56 WH sim16 GEWB 198](Alberti 1995)166 Helical or corkscrew shaped nucleus 0 absent 1present [GEWB sim204]State 1 occurs in Araneae Amblypygi ThelyphonidaSchizomida and certain Scorpiones (eg Hadrurus)(Alberti 1995)167 Vacuolated-type sperm 0 absent 1 present[GEWB 205]State 1 is unique to Opilioacariformes and Parasiti-formes (Acari) (Alberti 1995)

168 Sperm aggregates 0 absent 1 present [GEWB206] (Alberti 1995)

DEVELOPMENT

169 Yolk in early embryo 0 concentrated (centroleci-thal or telolecithal) 1 evenly distributed (isolecithal)[WH sim34 GEWB sim191]Wheeler amp Hayashi (1998) and Giribet et al (2002)coded this character as lsquo0 isolecithal or telolecithal 1centrolecithalrsquo based on information presented byYoshikura (1975) Their coding appears to highlightseparate conditions found in Scorpiones (ie isoleci-thal and telolecithal) rather than a property intrinsicto the character itself170 Embryonic nutrition other than yolk 0 absent 1present [WP sim49]State 1 occurs in extant Scorpiones (except buthidsand chaerilids) and Pseudoscorpiones (Weygoldt1969)171 Embryological growth zone 0 initiating segmentaddition within prosoma 1 initiating segment addi-tion posterior to prosoma [GEWB sim192]Giribet et al (2002) followed Dunlop amp Webster (1999)in stating that only Xiphosura and Scorpiones havestate 0 among Chelicerata However the last prosomalsomite develops from the growth zone in Ixodida(Anderson 1973 Evans 1992) A similar processapparently occurs in the acariform mites Tyroglyphus(Sarcoptiformes Astigmata) (Yoshikura 1975) andArchegozetes (Sarcoptiformes Oribatida) (Thomas ampTelford 1999) but not apparently in Tetranychus(Prostigmata) (Dearden Donly amp Grbic 2002)172 Eggsembryos maintained in external attachedbrood sac secreted by genital glands 0 absent 1present [WP sim46 S 59 WH 90 GEWB 208 = 219]State 1 occurs throughout Amblypygi SchizomidaThelyphonida and Pseudoscorpions Giribet et al(2002) followed Shear et al (1987) in homologizingspinneret-derived silken egg sacs of spiders with gen-ital-gland-derived brood sacs of Pedipalpi However asthese structures are not homologous in secretory ori-gin construction or composition (Shultz 1987) theyare not regarded here as homologous Giribet et al(2002) miscoded the character as absent in Pseudo-scorpiones and without justification in PalpigradiFemale palpigrades have well-developed glands asso-ciated with postgenital somites and an array of setaefusules and other structures (112) (Boumlrner 1902b Mil-lot 1942 Condeacute 1991a) which are consistent withsome sort of brood care and Palpigradi should there-fore be coded as unknown for this character173 Embryonic and early postembryonic lsquolateralrsquo orClaparegravede organs associated with coxa of postoralsomite IV (= arachnid leg 2) 0 absent 1 present [WHsim31 GEWB sim211]

PHYLOGENY OF THE ARACHNID ORDERS 263

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bilaterally paired rounded protuberances locatedbetween coxae of appendages of postoral somites IIIand IV (arachnid legs 1 and 2) in prelarval and larvalinstars appear to be a primitive feature of Acariformes(Lindquist 1984 Evans 1992 Alberti amp Coons 1999)It is known to be derived embryologically from thecoxa of leg 2 in the oribatid Archegozetes (Thomas ampTelford 1999) Structures apparently homologouswith the acariform Claparegravede organ are present inembryonic Amblypygi (Weygoldt 2000) and Thely-phonida (Yoshikura 1975) and in embryonic and earlypost-embryonic stages of Solifugae (Roewer 1934)where they are termed lateral organs The so-calledlsquolateral organrsquo of embryonic Xiphosura (174) occurs onthe lateral surface of the carapace and does not appearto be homologous with the coxa-associated structuressharing their name In fact the xiphosuran lateralorgan develops even when the coxa of postoral somiteIV has been excised (Sekiguchi 1988) In contrastWheeler amp Hayashi (1998) and Giribet et al (2002)followed Yoshikura (1975) in regarding all lsquolateralorgansrsquo as homologous and in miscoding this characteras absent in Acariformes Van der Hammen (1989)suggested that the sternal verrucae of Opilioacari-formes are homologous with the Claparegravede organ butit differs in gross structure and ontogenetic timing174 Embryonic lateral organ associated with cara-pace 0 absent 1 present [WH sim31 GEWB sim211]State 1 is known only from embryonic stages of extantXiphosura (Yoshikura 1975) In contrast to previousinterpretations this feature does not appear to behomologous with the lsquolateralrsquo or Clarapegravede organ ofcertain arachnids (173)175 Live birth 0 absent 1 presentState 1 occurs in extant Scorpiones and has evolvedseveral times in a few mites (Evans 1992)176 Hexapodal larva and 1ndash3 nymphal stages 0absent 1 present [WP 53 S sim61 WH 27 GEWB 186]State 1 occurs in Acari and Ricinulei (Lindquist1984)177 Hexapodal prelarva 0 absent 1 present [S sim61WH 47]State 1 occurs in Opilioacariformes (NeocarusKlompen 2000) and many Acariformes (Evans 1992)

EXCRETORYOSMOREGULATORY SYSTEM

178 Malpighian tubules 0 absent 1 present [WP 17S 62 WH 9 GEWB 153]State 1 occurs in all major arachnid groups except Pal-pigradi Pseudoscorpiones Opiliones Oribatida (Sar-coptiformes) and Prostigmata Within Acari State 1occurs in all major anactinotrichid lineages (Opilioac-ariformes Holothyrida Mesostigmata Ixodida) andsimilar structures are present in certain Astigmata(Evans 1992 Alberti amp Coons 1999)

179 Dorsomedian excretory organ 0 absent 1presentThis is a specialized excretoryosmoregulatory organformed by the postventriculus and proctodeum inProstigmata (Alberti amp Coons 1999)180 Adult coxal organ opening on or near coxa ofappendage III (= arachnid leg 1) 0 absent 1 present[S 64 WH 93 GEWB 90]State 1 occurs throughout Acari (Evans 1992 Albertiamp Coons 1999) Ricinulei (Pittard amp Mitchell 1972Legg 1976) Palpigradi (Millot 1942) AraneaeAmblypygi Thelyphonida and Schizomida (Buxton1913 1917) There is evidence that the second lsquoozo-porersquo of gonyleptids (Opiliones Laniatores) (Hara ampGnaspini 2003) is a persistent opening to the coxalorgan associated with leg 1 (Soslashrensen 1879) whichalso occurs in embryonic Phalangium (Opiliones Eup-noi) (Moritz 1959)181 Adult coxal organ opening on or near coxa ofappendage V (= arachnid leg 3) 0 absent 1 present [S63 WH 92 GEWB 89]State 1 occurs in extant Xiphosura (Yamasaki et al1988) Scorpiones mygalomorph (Buxton 1913 1917)and mesothele Araneae (Liphistius J Millot in Bris-towe 1932) State 1 occurs in basal Amblypygi(Charon Charinus Buxton 1913 1917) but degener-ates prior to the adult stage in higher groups(Phrynus Weygoldt 2000) Adequately preservedeurypterids show a small submarginal opening on thecoxa of appendage V (Baltoeurypterus Selden 1981also Eurypterus Hughmilleria Clarke amp Ruedemann1912) and this is interpreted here as the opening to acoxal organ Giribet et al (2002) coded this characteras uncertain for Eurypterida182 Adult coxal organ opening on coxa of appendageII (= arachnid palp) 0 absent 1 present [GEWB 65]Buxton (1913 1917) noted that the coxal organ ofSolifugae opens on the prolateral surface of the palpalcoxa The proximal end of the duct leading to the ori-fice is associated with glands and apparently theglandular secretion andor coxal fluid serve as saliva(Alberti 1979) (see also 14 15)183 Genital papillae 0 absent 1 present - inappli-cable coded only for those taxa with eversible lsquoappend-agesrsquo (111)The genital papillae are associated with the genitalopening in postlarval Acariformes (Evans 1992Alberti amp Coons 1999) These structures are probablyserially homologous with the prosomal Claparegravedeorgan (173) and both structures probably function inwater and ion regulation

DIGESTIVE SYSTEM

184 Ingestion 0 solid food 1 primarily liquid foodwith or without preoral digestion [WP sim16 WH sim8GEWB sim218]

264 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Extant xiphosurans ingest solid food and members ofmost extant arachnid orders ingest fluids Opilionesand Acari are the principal exceptions The phalangidOpiliones ingest solids but Shultz (2000 unpublobserv) has shown that the precerebral pharyngealapparatus of Cyphophthalmi is very similar to that ofthe fluid-feeding Scorpiones and differs substantiallyfrom phalangid Opiliones Examination of gut con-tents of the cyphophthalmids Siro acaroides S exilisand Chileogovea oedipus have failed to reveal solidparticles typically found in the guts of phalangids(Shultz unpubl observ) the character is coded asunknown for Cyphophthalmi Within Acari consump-tion of particles has been documented in Opilioacari-formes (Neocarus Van der Hammen 1989unidentified Walter amp Proctor 1998) Holothyridanon-parasitic Sarcoptiformes (Evans 1992 Walter ampProctor 1998 Alberti amp Coons 1999) and many endeo-stigmatids including the nematophagous Alycusroseus (Walter 1988)185 Mouth 0 directed posteroventrally 1 directedanteroventrally [S 9 WH 55 GEWB 28]State 1 occurs throughout Arachnida (Shultz 1990)186 Oral cavity dilated by muscles arising from coxaeand constricted by large circular sphincter 0 absent1 presentState 1 occurs in extant Xiphosura (Manton 1964Scholl 1977 Shultz 2001)187 Palate plate 0 absent 1 present [GEWB 159]State 1 is a unique synapomorphy of Araneae (Dunlop1994)188 Lateral walls of epistome broadly fused to medialwalls of palpal coxae opposite sides connected by well-developed transverse epistomal muscle 0 absent 1presentState 1 occurs throughout Opiliones and Scorpiones(Shultz 2000 unpubl observ)189 Epistome with a pair of lateral arms that projectsposteriorly into the prosoma on either side of the phar-ynx 0 absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)190 Epistome with four pairs of suspensor musclesattaching to the carapace 0 absent 1 presentState 1 occurs in extant Scorpiones (Lankester et al1885 Vyas 1970 Shultz unpubl observ) and appar-ently in Solifugae (Roewer 1934)191 Intercheliceral epipharyngeal sclerite 0 absent1 present [GEWB 31]State 1 occurs in Palpigradi (Boumlrner 1904) Trigo-notarbida (Palaeocharinus Dunlop 1994) Araneae(Firstman 1954 Marples 1983) Amblypygi (Mil-lot 1949b Shultz 1999) Schizomida (Millot1949e) Thelyphonida (Millot 1949e Shultz 1993)and Phalangida (Opiliones) (Shultz 2000 unpublobserv)

192 Epipharyngeal sclerite large projecting posteri-orly 0 absent 1 present - inapplicable due to absenceof sclerite (191) [GEWB 32]State 1 occurs in Amblypygi (Millot 1949b Shultz1999) Schizomida (Millot 1949e) Thelyphonida (Mil-lot 1949e Shultz 1993) some Araneae (Heptathela ndashcoded as Liphistius ndash and Hypochilus Marples 1983not Aphonopelma Firstman 1954) and perhaps Trig-onotarbida (Palaeocharinus Dunlop 1994)193 Dorsal dilator muscle of precerebral pharynxattaching to intercheliceral septum or associatedepipharyngeal sclerite (191) 0 absent 1 presentState 1 occurs in Palpigradi (Millot 1943) Araneae(Marples 1983) Amblypygi Schizomida and Thely-phonida (Millot 1949e Shultz 1993 1999)Extremely fine muscles have been documented insome Scorpiones (Shultz 2007) and in Leiobunum(Opiliones) (Shultz 2000) Lankester et al (1885)reported a muscle in Limulus that would appear tocorrespond to this muscle but the muscle does notexist (Manton 1964 Shultz 2001)194 Dorsal dilator muscle of precerebral pharynxattaching to dorsal surface of prosoma 0 absent 1present [GEWB 228]State 1 occurs in Palpigradi (Roewer 1934) and mostAraneae (Palmgren 1978 Aphonopelma Firstman1973) but not in Heptathela (coded as Liphistius) orHypochilus (Marples 1983)195 Lateral dilator muscle of precerebral pharynxattaching to endosternite 0 absent 1 present [GEWBsim226]State 1 is a unique feature of Araneae (Marples 1983)196 Lateral dilator muscle of precerebral pharynxattaching to lateral surface of epistomal processes 0absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)197 Lateral dilator muscle of precerebral pharynxattaching to medial process of coxa of appendage II(= arachnid palp) 0 absent 1 present [GEWB sim226]State 1 occurs in Amblypygi Thelyphonida and Schi-zomida (Boumlrner 1904 Shultz 1993 1999)198 Dilator muscle of precerebral pharynx andorpreoral cavity attaching to ventral surface of prosoma0 absent 1 present [GEWB 227]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) Araneae (Marples 1983) Palpigradi(Millot 1943) and Solifugae (Roewer 1934)199 Postcerebral pharynx 0 absent 1 present [WPsim31 S sim5 WH sim21 GEWB sim154]The cuticle-lined foregut passes through the centralnervous system in many arthropods That portion pos-terior to the CNS is here termed the postcerebralpharynx regardless of the details of its morphology orfunctional specialization A substantial postcerebralpharynx is known in extant Xiphosura (Lankester

PHYLOGENY OF THE ARACHNID ORDERS 265

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

et al 1885 Manton 1964 Yamasaki et al 1988)Scorpiones (Centruroides Hadrurus HeterometrusShultz 2007 also Androctonus Abd el-Wahab 1952)Solifugae (Millot amp Vachon 1949) Araneae Ambly-pygi Thelyphonida and Schizomida (Millot 1949andasheShultz 1993) Weygoldt amp Paulus (1979) applied theterm lsquopostcerebral pharynxrsquo only to the lsquosucking stom-achrsquo of Araneae and Amblypygi but the sucking stom-ach is here regarded as a complex of three characters199ndash201

200 Dilator muscle of postcerebral pharynx attach-ing to endosternite 0 absent 1 present - inapplicabledue to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Lankester et al1885 Manton 1964 Shultz 2001) Scorpiones (Cen-truroides Hadrurus Heterometrus Shultz 2007 alsoAndroctonus Abd el-Wahab 1952) Araneae (First-

man 1954 Palmgren 1978) Amblypygi (Millot1949b Shultz 1999) and apparently the palpigradeProkoenenia wheeleri (Rucker 1901) but not inEukoenenia miriabilis (Millot 1943 contra Boumlrner1904)201 Dilator muscle of postcerebral pharynx attachingto dorsal surface of prosoma 0 absent 1 present -inapplicable due to absence of postcerebral pharynx(199)State 1 occurs throughout Araneae (Millot 1949cFirstman 1954 Palmgren 1978) and Amblypygi(Charinus Millot 1949b Phrynus Shultz 1999)202 Crop and gizzard 0 absent 1 present - inap-plicable due to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Sekiguchi 1988)but not in extant Arachnida

222

J W SHULTZ

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

tended to favour a sister-group relationship betweenAmblypygi and Araneae the spiders (eg Platnick ampGertsch 1976 Weygoldt amp Paulus 1979) but this wasbased on a few similarities that do not appear to besynapomorphic (Shultz 1990 1999 present study)Araneae and Pedipalpi form a well-supported cladewith the fossil order Trigonotarbida being its likelysister group Pseudoscorpiones and Solifugae or sun-spiders are often united within a clade Haplocne-mata (

=

Apatellata) (Weygoldt amp Paulus 1979 Vander Hammen 1989 Shultz 1990 Wheeler amp Hayashi1998 Giribet

et al

2002 but see Alberti amp Peretti2002)

Some interordinal relationships are often recoveredin phylogenetic analyses but are supported by rela-tively few characters These include Dromopoda (Opil-iones

+

Scorpiones

+

Haplocnemata) Megoperculata(

=

Palpigradi

+

Tetrapulmonata) and Acaromorpha(Ricinulei

+

Acari) (Shultz 1990 Wheeler amp Hayashi1998 Giribet

et al

2002) The status of Acaromorphais further complicated by substantial morphologicaldivergence within and between the two principal lin-eages of Acari (ie Anactinotrichida and Acariformes)(Lindquist 1984) which has led some workers to sug-gest that Acari is diphyletic (Van der Hammen 1989)Relationships among Palpigradi Tetrapulmonata

Acaromorpha Haplocnemata Opiliones and Scorpi-ones are effectively unresolved as is the placement ofseveral fossil taxa (eg Haptopoda)

Results from the present analysis (Fig 1) essen-tially affirm this characterization of our currentunderstanding of arachnid phylogeny based on mor-phology and other non-molecular characters but alsooffer new proposals and evaluate alternative interpre-tations that have emerged in the last 17 years iesince my previous attempt to resolve arachnid phylog-eny (Shultz 1990) Specifically parsimony-based anal-yses corroborate the monophyly of Arachnida as wellas Uropygi Pedipalpi Haplocnemata and Acaromor-pha The data also support more recent proposalsincluding the monophyly of Opiliones and Scorpiones(

=

Stomothecata

nom nov

) (Shultz 2000) and Pan-tetrapulmonata

nom nov

with an internal structureanticipated by Dunlop (1999 2002c) ie (Trigonotar-bida (Araneae (Haptopoda Pedipalpi))) Howeverrelationships among Palpigradi Acaromorpha Hap-locnemata Stomothecata and Pantetrapulmonata areeffectively unresolved These results indicate thatmorphology offers important phylogenetic informa-tion but it is not yet sufficient to resolve relationshipsat the deepest levels within Arachnida

METHODS

T

ERMINAL

TAXA

The study was based on 59 euchelicerate genera (41extant 18 fossil) with most represented by one spe-cies (detailed below) coded for 202 binary and unor-dered multistate characters (Table 1 Appendix)

Xiphosura (horseshoe crabs)

The xiphosurans are anancient (SilurianndashRecent) aquatic lineage with itsgreatest diversity occurring in the fossil record Itcomprises two main groups Synziphosurida (Sil-urianndashDevonian ~ ten genera) and Xiphosurida (Car-boniferousndashRecent ~14 genera) Synziphosurids areprobably paraphyletic and retain plesiomorphic fea-tures such as a ten-segmented opisthosoma withthree segmented metasoma (Anderson amp Selden1997) They were represented in the matrix by

Wein-bergina opitzi

one of the few fossil xiphosurans withpreserved appendages (Moore Briggs amp Bartels2005) and a more typically preserved synziphosurid

Limuloides limuloides

Extant xiphosurans (threegenera four spp) were represented by the intensivelystudied Atlantic horseshoe crab

Limulus polyphemus

and an Asian horseshoe crab

Carcinoscorpiusrotundicauda

with supplemental information drawnfrom another Asian species

Tachypleus tridentatus

Eurypterida (sea scorpions)

A diverse (

gt

60 genera)aquatic group of fossil euchelicerates that ranged from

Figure 1

Cladogram summarizing results from analysisof major euchelicerate lineages generated by parsimonyanalysis of 59 genera and 202 non-molecular charactersOnly those relationships that were well supported by boot-strap analysis or consistently recovered in sensitivity anal-ysis are depicted deepest relationships within Arachnidaare effectively unresolved

Xiphosura

Eurypterida sstr

Chasmataspidida

Scorpiones

Opiliones

Pseudoscorpiones

Solifugae

Palpigradi

Trigonotarbida

Araneae

Haptopoda

Amblypygi

Thelyphonida

Schizomida

Ricinulei

Anactinotrichida

Acariformes

Acaromorpha

Uropygi

Pedipalpi

Schizotarsata nom nov

Tetrapulmonata

Pantetrapulmonata nom nov

Haplocnemata

Arachnida

Stomothecata nom nov

Eurypterida slat

Acari

PHYLOGENY OF THE ARACHNID ORDERS

223

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Table 1

Data matrix comprising 59 euchelicerata genera and 202 binary and unordered multistate characters Fossil taxaare indicated by an asterisk Ambiguity codes A

=

[345] B

=

[12] C

=

[02] D

=

[34] E

=

[01]

Weinbergina

00010-011 00 - - 000 10-0001 0300000 0- -1210-0 -

Limuloides

00010-011 01 - - 300000 0- -1210-0 -

Euproops

00010-011 01 - - 200010 0- -1010-0 - 1

Limulus

10010-0011 0000-01000 0000-00000 -0010-0000 0010000000 011110001 0100010021 0001011011 000--01101 1000200010 000110-000 00-0010-0 01110-0110 0106112110 100--00000 0000000000 0010200000 0001000000 10-0010000 0-00000111 01

Tachypleus

10010-0011 0000-01000 0000-00000 -0010-0000 0010000000 011110001 0100010021 0001011011 000--01101 1000200010 000110-000 00-0010-0 01110-0110 0106112110 100--00000 0000000000 0010000000 0001000000 10-0010000 0-00000111 01

Baltoeurypterus

00000-011 00000000 0-0 -0000-0 0010000010 0110000 0011-0020 00000 100-00001 005000-0 00-00 0-0310-0 01- 1010 --00000 0 1 100 0-

Stylonurus

00000-011 00000000 - -000-0 001000000 010000 0011-000 000 10-0001 005000-0 00-0 0-0310-0 01- 1010 000 0 1

Chasmataspis

00000-011 01 - - 1 0 1601-0 0- -1410-0 00- 11 0 0

Diploaspis

0000-01 01 - - 1 0001 0601-0 0- -0410-0 00- 11 0

Octoberaspis

0010-01 01 - - 1 00 601-0 0- -0410-0 00- 11 0 1

Prokoenenia

0000110000 0010-00000 000000000 -0000-00 001000000 000000001 1000000020 00000110 0101000001 0014000-0 000001-010 10-021201 00000010 10310-00- -010000100 1001000 000--000 001 0001100000 1011000111 00

Eukoenenia

0000110000 0010-00000 0000000000 -0000-0000 001000000 0000000001 1000000020 000001011 0101000001 00104000-0 000001-010 00--021201 000000100 110310-00- -010000100 1001000 000--000 001 0001100000 101100010- --

Plesiosiro

00110-000 00 -0 -0- 0000 0000 0000 0 11000 0A00-0 00100 0--00--1 1 00 0 0

224

J W SHULTZ

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Palaeocharinus

00000-0000 00100110 0000 -000-00 00100000 0000000 0000002 02--0 100-0001 04100-1 001000 10--110--1 110000 101B1000 0 0 10 11

Gilboarachne

00000-0000 0000110 000 -000-00 000000 0000000 0002 02--0 100-0001 04100-1 001000 0--110--1 110000 101B 100 0 0

Liphistius

00000-0000 0011000110 0000001000 -0000-0000 0110000000 0000000110 00000100220010011011 0100-00001 00115010-0 0000010010 1101000--1 1100001000 00030110120110000011 1001000000 0111310000 1000000101 1001101000 1110100111 10

Aphonopelma

00000-0000 0011000110 0000001000 -0000-0000 0110000000 0000000110 00000100220010011011 0100-00001 00115010-0 0000011010 1110000--1 1100001000 00031110120110000011 1001000000 0111310000 1000000101 1001101000 1011100111 10

Hypochilus

00000-0000 0011000110 0000001000 -0000-0000 0110000000 0000000110 0000010022 0010011011 0100-00001 00115010-0 0000011010 1110000--1 1100001000 0003011012 0110000011 1001000000 0111310000 1000000101 001101000 1110100111 10

Charinus

00100-0000 0011000110 0000100000 -00010001 0010010000 000001011 0000010023 1000101011 0101100000 01105000-0 0000010000 10--020--1 1100001000 1015010012 01100000 1011001100 0111310000 1110000101 1001100000 1110001011 10

Phrynus

00100-0000 0011000110 0000100000 -000100101 0000-10000 0001001011 0000010023 1000101011 0101100000 00105000-0 0000010000 00--020--1 1100001000 1015010012 01100000 1011001100 0111310000 1110000101 00-1100000 1110001011 10

Stenochrus

0010110000 0211000110 000000001 0000110100 0010010000 000000111 00000101230000001011 0101100000 00115000-0 0100010000 00--021211 1000001001 01410-014111000000 1111001110 0111310000 110000101 00-1100000 111000100 00

Protoschizomus

0010110000 0211000110 000000001 00001100 0010010000 0000011 000000123 00000 0101100000 0015000-0 0100010000 00--021211 100000100 050-00- -111000000 111001110 0111310000 110000101 00-1100000 1000100 00

Mastigoproctus

00110-0000 0211000110 0000100001 0000100100 000--10000 0001001111 0000010123 0000001011 0101100000 00115000-0 0100010000 00--021201 1100001001 101511001B 0111000000 1111001110 0111310000 1110000101 00-1100000 1110001010 00

Proschizomus

00100-000 02001 1 0010 0010000 0000 100 05000-0 001000 0--02121 0 00 0 0

Terpsicroton

00000-000 10 - 0- 00000 0000 0110 0 0510-0 100- --120--1 0 50-13 00 0

Poliochera

00000-000 10 -1 0- 00000 0000 0110 0 0510-0 100- --120--1 0 50-13 00 0

Table 1

Continued

PHYLOGENY OF THE ARACHNID ORDERS

225

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Cryptocellus

00000-0000 1211000100 00-00001 0000-000 0010001000 000000001 0001100020 02--000-11 0101000001 00105120-0 100000-000 00--120--1 000003 1500-014 10-000011 10010000 011120000 0010101 00-1100000 0-0000000- --

Ricinoides

00000-0000 1211000100 00-00001 0000-000 0010001000 000000001 0001100020 02--000-11 0101000001 00105120-0 100000-000 00--120--1 000003 1500-014 10-000011 10010000 011120000 0010101 00-1100000 0-0000000- --

Neocarus

00001-0000 0011000020 0000-00001 10100-000 0010010000 00000001 000111120 02--00 0101000011 0114000-0 000000-000 00--000--1 000005 00-013 010--00010 1001110 000--01000 011101 00-0100000 0-0000000- --

Siamacarus

00001-0000 0011000020 0000-00001 1100-000 0010010000 0000000 0011112 02--00 10100011 014000-0 000000-000 00--000--1 000005 00-012 110000010 101110 00--10 01101 -0100000 0-0 --

Australothyrus

01000-0100 0001000020 0000-00001 10100-000 0011000000 0000000 00001020 02--00 010-00011 01000-0 000000-100 00--000--1 000003 00-014 10--00010 1001100 000--01000 0101 00-1100000 0-000 --

Allothyrus

01000-0100 0011000020 0000-00001 10100-000 0011000000 0000000 00001020 02--00 010-00011 01000-0 000000-100 00--000--1 000003 00-00- -10--00010 101100 00--10 0101 00-1100000 0-0 --

Glyptholaspis01000-0000 0011000020 000-00001 10100-000 0011000000 000000001 0000001020 02--011011 010-00011 0110000-0 000000-000 00--000--1 00000300 00-00- -10--00000 1001100000 000--01000 000010101 00-1100000 0-0000000- --

Amblyomma00000-0000 0000-00020 0000-00001 10000-000 000--00000 000000001 0000001020 02--011011 010-00011 01100000-0 000000-000 00--000--1 00000300 100-014 010--00010 1001100000 000--01000 0000010101 00-1100000 0-0000000- --

Argas00000-0000 0000-00020 0000-00001 10000-000 000--00000 000000001 0000001020 02--011011 010-00011 00100000-0 000000-000 00--000--1 00000300 100-014 010--00010 101100000 00--10 0000010101 00-1100000 0-00000- --

Alycus000001000 000110020 000-00001 10100-000 000--00000 000000001 000010020 02--0011 000--00011 00101000-0 000000-001 10--000--1 000000 00-013 110100000 10100 00 0111 0010100000 0-000- --

Allothrombium0000000 000110020 00-00001 10000-000 000--00000 000000001 00111-0020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000004 00-0130110100000 1001001000 000--00000 010011011 0011100000 0-0000000- --

Microcaeculus00000000 000110020 00-00001 10000-000 000--00000 000000001 00111-0020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000004 0120130110100000 10010010 00--0000 01001011 001100000 0-0000000- --

Palaeacarus000011000 040110020 000-00001 10100-000 000--00000 000000001 00111-0020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000000 010 110100000 101011000 00--00 010011001 0010100000 0-00000- --

Table 1 Continued

226 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Archegozetes000011000 040110020 000-00001 10100-000 000--00000 000000001 0000000020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000000 00-00- -110100000 1001011000 000--00000 0010011001 0010100000 0-0000000- --

Cyphophthalmus0000100100 0100-00000 100-00000 -0000-1000 0010000001 1000000001 0000000020 01--010-11 0100-00001 00112000-0 000000-000 00--0013-1 000001000 010100 10--00000 1001110001 001100010 0000000 10-100110 0-0001000- --

Chileogovea0000100100 0100-00000 1010-00000 -0000-1000 0010000001 1000000001 0000000020 01--010-11 0100-00001 00112000-0 000000-000 00--0013-1 000001000 010100 10--00000 1001110001 0000 0000000 10-100110 0-0001000- --

Caddo0000100100 0000-00000 100-00000 -0000-000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 01010100--10--00000 10-1110001 000000 0000000 10-0100110 10001000- --

Leiobunum0000100100 0000-00000 1010-00000 -0000-0000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 010B01C00- -10--00000 1001110001 000--00000 1000000000 10-0100110 101001000- --

Sclerobunus0000000100 0000-00000 100-00000 -0001-000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 010001C00- -10--00000 1001110001 00--00000 1000000000 10-0100110 10001000- --

Gonyleptes0000000100 0000-00000 1010-00000 -0001-000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 010001C00- -10--00000 1001110001 010--00000 100000000 10-0100110 10001000- --

Centruroides0000100000 0100-10000 0111-00000 -0000-1000 100--00001 1000000001 0000010010 01--012110 0100-10001 00116000-0 001000-000 00--0311-1 0011001000 0005110011 1110001011 1101101000 0010110100 0000100100 10-1100111 0-10010011 00

Hadrurus0000100000 0100-10000 0111-00000 -0000-1000 100--00001 1000000001 0000010010 01--012110 0100-10001 00116000-0 001000-000 00--0311-1 0011001000 0005110011 1110001011 1101101000 0010110111 0000100100 10-1100111 0-10010011 00

Heterometrus0000100000 0100-10000 0111-00000 -0000-1000 100--00001 1000000001 0000010010 01--012110 0100-10001 00116000-0 001000-000 00--0311-1 0011001000 00051100111110001011 1101101000 001020111 0000100100 10-1100111 0-10010011 00

Prearcturus000010000 010 -0 -000- 100--001 000 000-0 00- --0311- 0 0

Palaeoscorpius00000000 0D0 -0 -000- 100--0000 00 00000 0 10-0001 06000-0 000-00 --0311- 1 01 0 0

Stoermeroscorpio00000000 00000 -0 -000- 100--0000 000 00000 0 10-0001 06000-0 000-00 --0311- 110 01 1 0

Table 1 Continued

PHYLOGENY OF THE ARACHNID ORDERS 227

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

the Ordovician to the Permian (Tollerton 1989) Phy-logenetic relationships among the eurypterids havenot been rigorously addressed (but see Beall amp Laban-deira 1990) However members of Stylonuroideaappear to be plesiomorphic in retaining relativelyunspecialized legs and were represented by Stylonu-rus (eg Clarke amp Ruedemann 1912 Waterston1979) In the remaining eurypterids the last leg ispaddle-like (Tollerton 1989) The non-stylonuroideurypterids were represented in the matrix by thewell-preserved and well-studied Baltoeurypterus tet-ragonophthalmus especially as described by Selden(1981)

Chasmataspidida Members of this fossil aquaticgroup (OrdovicianndashDevonian six genera) have a four-segmented mesosoma and a nine-segmented meta-soma (Dunlop Poschmann amp Anderson 2001 DunlopAnderson amp Braddy 2004 Dunlop 2002a) Three gen-era were included Chasmataspis from the monotypicfamily Chasmataspididae as well as Diploaspis andOctoberaspis from Diploaspididae

Haptopoda This is a fossil terrestrial group (Carbon-iferous) containing one known species Plesiosiromadeleyi The known specimens have been re-examined by Dunlop (1999)

Palpigradi The extant palpigrades (two families sixgenera ~50 spp) (Harvey 2002) are a poorly studiedgroup of small-bodied terrestrial (some semi-aquatic)arachnids The two most well-studied generaEukoenenia (Eukoeneniidae) especially E mirabilisand Prokoenenia (Prokoeneniidae) especiallyP wheeleri were included in the matrix The oneknown fossil species Paleokoenenia mordax is rela-tively recent (Pliocene) and to the extent known isfairly similar to extant palpigrades (Rowland amp Sis-som 1980) It was not included in the analysis

Trigonotarbida This is a fossil terrestrial order(Upper Silurian to Lower Permian ~ten families ~50spp) The genera coded here were drawn from twoLower Devonian Lagerstatten Gilboarachne from theGilboa Formation (New York) (Shear et al 1987) and

Proscorpius00000000 00000 -0 -000- 100--0000 010 00000 0 10-0001 06000-0 000-00 --0311- 110 01 1 0

Chthonius0000100000 0310-00120 0000-10100 -1000-0000 100--00100 0000000001 0000010010 01--012100 0E00-00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Neobisium0000100000 0310-00121 0000-10100 -1000-0010 100--00100 0000000001 0000010010 01--012100 0100-00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Feaella0000100000 0310-00120 0000-10100 -1000-0000 100--00100 0000000001 0000010010 01--012100 000--00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Chelifer0000100000 0310-00121 0000-10100 -1000-0010 100--00100 0000000001 0000010010 01--012100 000--00001 01115000-0 000000-000 00--000--1 0000021000 01700-014 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Eremocosta0000110000 0310-00121 0000-00010 -1000-0000 001010100 0000000001 00011-0000 02--010-10 1101010001 01114000-0 000000-000 00--000--1 00000200 13111013 010--10000 1001000000 000--00100 1010000100 01-1100001 0-00000110 00

Galeodes0000110000 0310-00121 0000-00010 -1000-0000 001010100 0000000001 00011-0000 02--010-10 1101010001 01114000-0 000000-000 00--000--1 00000200 13111013 010--10000 1001000000 000--00100 1010000100 011100001 0-00000110 00

Table 1 Continued

228 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Palaeocharinus from the Rhynie Chert (Scotland)(Dunlop 1994 Fayers Dunlop amp Trewin 2004)

Araneae (spiders) This is a very large order (110 fam-ilies ~3600 genera ~39 000 extant spp) (Platnick2005) represented here by one genus from each of thethree principal extant lineages (Platnick amp Gertsch1976 Coddington amp Levi 1991 Coddington 2005)Liphistius (Mesothelae) Aphonopelma (Mygalomor-phae) and Hypochilus (Araneomorphae) The sampleis small with respect to known diversity but the basalphylogeny and relevant groundplan states of the orderare well established and states derived from the rep-resented taxa are consistent with them

Amblypygi (whipspiders) This is a small extant ter-restrial order (five families 17 genera 136 spp)(Harvey 2002) and was represented in the matrix bytwo fairly well-studied genera Charinus and Phry-nus Charinus retains several features that appear tobe plesiomorphic for the order (eg adult pedal pul-villi coxal glands associated with leg 3 eversiblevesicles) Weygoldt (1996) has examined the phylog-eny of the order and has provided an important sum-mary of morphology and general biology (Weygoldt2000)

Thelyphonida or Uropygi ss (whipscorpions vinega-roons) Extant whipscorpions (16 genera 106 spp)(Harvey 2002) were represented here by one well-studied species Mastigoproctus giganteus althoughadditional information was drawn from Typopeltis andThelyphonus The morphology of the group is highlyconserved there have been no modern studies ofintraordinal phylogeny A controversial fossil speciesfrom the late Carboniferous Proschizomus (Dunlop ampHorrocks 19951996) was also included but less well-preserved Carboniferous forms were not

Schizomida This small order (two extant families~34 genera ~200 spp) (Harvey 2002) is widelyregarded as the sister group of Thelyphonida Thetaxon sample included one representative from eachextant family Protoschizomus (Protoschizomidae) andStenochrus (Hubbardiidae) especially S portoricen-sis There are three recent (Pliocene) fossil speciesall from the same locality these were not includedhere

Ricinulei The extant ricinuleids (three genera 55spp) (Harvey 2002) were represented by two generaCryptocellus and Ricinoides Detailed studies of grosscuticular anatomy and post-embryonic developmentare available for a representative of each genus (Cryp-tocellus Pittard amp Mitchell 1972 Ricinoides Legg1976) and some information on internal anatomy isavailable for Ricinoides (Millot 1945) and this hasbeen extrapolated to Cryptocellus in the matrix There

are two basic fossil types (Selden 1992) one resem-bling modern taxa and another with a unique opistho-soma that superficially resembles the closed elytra of abeetle (ie curculioids) Two fossils from the formergroup were included Terpsicroton and Poliochera asthey are reasonably well preserved and show impor-tant characters (eg two pairs of eyes or evidence ofopisthosomal diplosegmentation) not expressed inextant forms No curculioid ricinuleids were includedas they appear to offer no additional information rel-evant to resolving ordinal relationships

Opilioacariformes (= Opilioacarida Notostigmata)The opilioacariform mites (nine genera 20 spp) (Har-vey 2002) are generally regarded as plesiomorphicAcari and are fairly conserved in their morphologyThe group was represented by two species Neocarustexanus and Siamacarus withi N texanus has a typi-cal opilioacariform morphology and its external anat-omy has been particularly well studied (eg Van derHammen 1989 Klompen 2000) S withi differs frommost other opilioacariforms in having trichobothriaand three rather than two pairs of lateral eyes (the lat-ter also in Paracarus) features that are potentiallysignificant for assessing ordinal relationships Someinformation on internal morphology of Opilioacaruswas taken from With (1904) and extrapolated to Neo-carus and Siamacarus

Parasitiformes This large group of mites comprisesthree clades Holothyrida Ixodida (ticks) and Mesos-tigmata (= Gamasida) Holothyrida (three familiesfive genera lt 30 spp) (Walter amp Proctor 1999) wasrepresented here by two species Australothyrus ocel-latus and Allothyrus constrictus Ixodids (three fami-lies 22 genera ~860 spp) (Kierans amp Robbins 1999)were represented by Amblyomma and Argas andmesostigmatids (~70 families ~10 000 spp) were rep-resented by Glyptholaspis confusa

Acariformes (= Actinotrichida) This is a very largediverse group of mites that includes the Endeostig-mata Sarcoptiformes and Prostigmata Endeostig-mata is an apparently paraphyletic assemblage of~ten families A morphologically generalized speciesAlycus roseus (Bimichaeliidae) was included herewith character states extracted largely from Van derHammen (1989) Sarcoptiformes (gt 200 families) iseffectively synonymous with Oribatida with Astig-mata being a large derived lsquooribatidrsquo clade (Norton1998 Maraun et al 2004) The group was representedby an early divergent plesiomorphic genus Palaeac-arus (Palaeosomata) and a more derived and well-studied genus Archegozetes (especially A longisetusand A magnus) The Prostigmata (~120 families~7000 spp) (Walter amp Proctor 1999) was representedby two genera Allothrombium and Microcaeculus

PHYLOGENY OF THE ARACHNID ORDERS 229

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Pseudoscorpiones The pseudoscorpions (24 families425 genera ~3200 spp) (Harvey 2002) were repre-sented in the matrix by four terminal genera Chtho-nius (Chthoniidae) was drawn from the basallydivergent superfamily Chthonioidea (Harvey 1992)Feaella (Feaellidae) represented the enigmatic super-family Feaelloidea which has an unusual combinationof characters and may be an early divergent lineageclose to Chthonioidea (Harvey 1992) Neobisium (Neo-bisiidae) is morphologically intermediate between theearly divergent families and the higher families rep-resented by Chelifer (Cheliferidae)

Solifugae (= Solpugida) (sun spiders) The order issmall (~1000 spp) but the absence of modern phylo-genetic treatments precludes meaningful estimates ofgenera and families (Harvey 2002) The taxon sampleincluded two genera one from the New World Eremo-costa and one from the Old World Galeodes

Opiliones (harvestmen) Extant harvestmen (~25 fam-ilies ~500 genera ~6000 spp) are divided into twosuborders Cyphophthalmi and Phalangida Cyphoph-thalmids were represented here by Chileogovea oedi-pus (Petallidae) which is currently under study by theauthor and Cyphophthamus duricorius (Sironidae)which is probably the most well-studied member of thesuborder (eg Janczyk 1956) Phalangida includesthree major groups Laniatores Eupnoi and Dyspnoibut there is disagreement as to whether Dyspnoi is thesister group to Laniatores or to Eupnoi (Shultz ampRegier 2001 Giribet et al 2002) Laniatores was rep-resented by Sclerobunus and Gonyleptes and Eupnoiby Leiobunum and Caddo

Scorpiones (scorpions) The extant scorpions (16 fam-ilies 155 genera 1279 spp) (Fet et al 2000) wererepresented by three taxa Centruroides vittatus(Buthidae) Hadrurus arizonensis (Iuridae) and Heter-ometrus spinifer (Scorpionidae) Buthidae is widelyregarded as the sister group to other extant lineagesand similarities among the represented terminals arelikely to be ground plan features of extant scorpionsgenerally The morphology of fossil scorpions is sub-stantially more diverse (Kjellesvig-Waering 1986)but their phylogenetic relationships are unclear (butsee Jeram 1998) The fossil taxa used in this study(Proscorpius Stoermeroscorpio Palaeoscorpius Pre-arcturus) were chosen for quality of preservation andor presence of a phylogenetically significant constella-tion of characters

CHARACTER CODING

Most character states were determined from directobservation the primary literature and authoritativereviews In some cases states were assigned to termi-

nals based on observations from related species asnoted in the Appendix In the matrix state lsquo-rsquo indicatesthat the character is inapplicable because the taxonlacks a more general character For example if a taxonlacks eyes then special features of the eyes (numberposition retinal configuration etc) are inapplicable tothat taxon A lsquorsquo indicates that the state is unknown oruncertain An entry with multiple states (eg 01 345) should be treated as an ambiguity code not a poly-morphism it indicates that two or more interpreta-tions of homology are applicable and that assignmentwas established analytically by character concor-dance Although lsquo-rsquo lsquorsquo and lsquo01rsquo are analyticallyidentical when applied to binary characters theyprovide information about the empirical status of thecharacter state in specific taxa Characters are citedthroughout the text as italicized numbers in parenthe-ses and are discussed in the Appendix

PHYLOGENETIC ANALYSIS

Phylogenetic analysis of the full data matrix wasperformed using the program Tree Analysis usingNew Technologies (TNT) ver 1 (Goloboff Farris ampNixon 2000) using lsquotraditionalrsquo search based on1000 replicates using TBR branch swapping Resultswere compared to those obtaining using the ratchetalgorithm (Nixon 1999) to determine any differencedue to analytical approach Nodal support for theminimal-length topology was evaluated by boot-strap (Felsenstein 1985) and Bremer support(Bremer 1994) Bootstrap analysis was conducted inTNT and based on 1000 pseudoreplicates each anal-ysed by ten random-addition replicates using TBRbranch swapping A nexus file containing the result-ing 1000 trees was imported into PAUP ver 10(Swofford 2002) to obtain bootstrap frequenciesBremer support was determined in TNT by con-straining specific nodes in the minimal-length topol-ogy and then determining the shortest tree that didnot recover the specified clade The difference inlength between the unconstrained and constrainedminimal-length trees is the Bremer support Theeffect of homoplasy on results was explored by con-ducting implied weights analysis (Goloboff 1993) inTNT Six analyses were conducted each with con-stant of concavity (k) set to a different integer valueof 1ndash6 where 1 is weighted most severely againsthomoplasious characters Each implied weightsanalysis was conducted using lsquotraditionalrsquo searchbased on 1000 replicates using TBR branch swap-ping The same procedures were used in analysing amatrix that included only extant taxa Howevercharacters rendered uninformative by removal offossil taxa (ie 49 98 99 100 113 114 125 161)were excluded prior to analysis

230 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

COMPARISONS OF ALTERNATIVE HYPOTHESES

Phylogenetic hypotheses proposed in the recent liter-ature were also evaluated including those thatattempted to resolve arachnid phylogeny completely(Fig 2) their hypothesized subclades (Fig 3) andhypotheses that proposed only specific nodes (Figs 34) The fully resolved topologies of Weygoldt amp Paulus(1979) Van der Hammen (1989) Shultz (1990)Wheeler amp Hayashi (1998) and Giribet et al (2002)were compared with the optimal topology using theTempleton test (Templeton 1983) as implementedin PAUP Internal relationships of multisampledorders were constrained to match those of the optimaltopology unless the original authors explicitlyfavoured an alternative Node-specific hypotheseswere evaluated by determining the frequency withwhich the node was recovered in bootstrap analysisThey were also evaluated by constraining parsimonyanalysis in TNT to recover the shortest tree contain-ing each specific node and then assessing the effect onrelative tree length and overall resolution takingnote of effects on otherwise stable or well-supportedclades The entire matrix was used to assess the twocases where node-specific hypotheses involved fossiltaxa (Fig 4)

RESULTS

EXTANT TAXA

Unweighted analysis of extant taxa produced two min-imal-length trees (length 383 CI 057) (Fig 5A) withconflicts limited to relationships within Pseudoscorpi-ones Multiply sampled orders were recovered asmonophyletic with high nodal support as measured bybootstrap percentage (BP gt 80) except Acari (BP 30)There were few well-supported interordinal groupsexcept Uropygi (BP 100) Pedipalpi (BP 100) Tetra-pulmonata (BP 93) and ArachnidaXiphosura (BP100) All other interordinal relationships were recov-ered with BP less than 60

Results from implied weights analysis (IWA) indi-cated a significant effect of homoplasy on resolution ofthe deepest interordinal relationships although theinterordinal clades Stomotheca Haplocnemata Acar-omorpha Tetrapulmonata Pedipalpi and Uropygiwere recovered under all values of k IWA with k = 1 (1tree best score = 5124) recovered Megoperculata(= Palpigradi + Tetrapulmonata) and Micrura(= Megoperculata + Acaromorpha) with Micrura beingthe sister group to Haplocnemata (= Pseudoscorpiones+ Solifugae) and with Stomothecata (= Opiliones +Scorpiones) being the sister group to all other arach-nids (Fig 5B) With k = 2 (1 tree best score = 3807)and k = 3 (1 tree best score = 3054) Megoperculatawas recovered and Stomotheca remained the sister

group to other arachnids but the relative positions ofHaplocnemata and Acaromorpha changed (Fig 5C)IWA using k = 4 (1 tree best score = 2559) k = 5 (1tree best score = 2205) and k = 6 (1 tree bestscore = 1938) recovered topologies consistent with theminimal-length topology derived from unweightedcharacters (Fig 5A)

EXTANT AND FOSSIL TAXA

Unweighted analysis of extant and fossil taxa pro-duced 18 minimal-length trees (length 426 CI 056)(Fig 6A) with conflicts limited to relationships withinPseudoscorpiones Scorpiones and Araneae Multiplysampled orders were again recovered as monophyleticwith high nodal support (BP gt 80) except Acari whichwas reconstructed as diphyletic with Anactinotrichidabeing the sister group to Ricinulei (BP 33) Therewere no well-supported interordinal groupings (ieBP gt 80) except Uropygi (BP 89) Pedipalpi (BP 92)and Arachnida (BP 90)

Results from IWA indicated a significant effect ofhomoplasy on resolution of the deepest interordinalrelationships in Arachnida although StomothecataHaplocnemata Acaromorpha Cryptognomae Pantet-rapulmonata Tetrapulmonata Schizotarsata Pedi-palpi and Uropygi were recovered under all values ofk IWA with k = 1 (7 trees best score = 5675) recov-ered Megoperculata and Micrura with Micrura beingthe sister group to Haplocnemata Stomothecata wasnot recovered in the strict consensus (Fig 6B) Withk = 2 (3 trees best score = 4243) IWA recovered thetopology shown in Fig 6C with Palpigradi being thesister group to all other arachnids The strict consen-sus topologies reconstructed using k = 3ndash6 were iden-tical to those obtained with k = 2 (k = 3 3 trees bestscore = 3419 k = 4 3 trees best score = 2874 k = 53 trees best score = 2483 k = 6 3 trees bestscore = 2188)

COMPARISONS OF ALTERNATIVE HYPOTHESES

The hypothesis of Weygoldt amp Paulus (1979) (Fig 2)was 398 steps (15 steps longer than the unconstrainedminimal-length tree) Van der Hammen (1989) was400 steps (17 steps longer) Shultz (1990) was 399steps (six steps longer) Wheeler amp Hayashi (1998)was 391 steps (eight steps longer) and Giribet et al(2002) was 394 steps (11 steps longer) Templetontests conducted at the 005 significance level rejectedthe hypothesis of no difference between the optimaltopology and those of Weygoldt amp Paulus (P = 00002)Van der Hammen (P = 00011) Wheeler amp Hayashi(P = 00018) and Giribet et al (P = 00343) The Shultztopology was not significantly different (P = 01444)from the optimal topology

PHYLOGENY OF THE ARACHNID ORDERS 231

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 2 Fully resolved phylogenetic hypotheses of extant euchelicerate groups proposed in the recent literature Note thesimilarity in the topologies of the parsimony-based analyses by Shultz (1990) Wheeler amp Hayashi (1998) and Giribet et al(2002) The Giribet et al topology is based on neontological data (morphology and molecules) and the original lsquoROOTrsquo maybe an artefact from use of the highly divergent pycnogonids as an outgroup

Xip

hosu

raS

corp

ione

sO

pilio

nes

Ara

neae

Am

blyp

ygi

Uro

pygi

Ric

inul

eiA

nact

inot

richi

daS

olifu

gae

Pse

udos

corp

ione

sP

alpi

gra

diA

carif

orm

es

Xip

hosu

raS

corp

ione

sA

rane

aeA

mbl

ypyg

iU

ropy

giP

alpi

gra

diS

olifu

gae

Pse

udos

corp

ione

sO

pilio

nes

Ric

inul

eiA

cari

Lipoctena

Arachnida

Apulmonata

Holotracheata

Cryptoperculata

Acaromorpha(ldquoAcarinomorphardquo)

Tetrapulmonata(rdquoMegoperculatardquo)

LabellataHaplo-cnemata

Weygoldt amp Paulus (1979)

Labellata

Tetrapulmonata(rdquoArachnidardquo)

Haplocnemata(rdquoApatellatardquo)

Van der Hammen (1989)

Xip

hosu

raP

alpi

gra

diA

rane

aeA

mbl

ypyg

iU

ropy

giR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Pedipalpi Haplocnemata

Novogenuata

Dromopoda

Shultz (1990)

Acaro-morpha

Micrura

Xip

hosu

raP

alpi

gra

diU

ropy

giA

mbl

ypyg

iA

rane

aeR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Labellata Haplocnemata

Novogenuata

Dromopoda

Wheeler amp Hayashi (1998)

Acaro-morpha

Micrura

Xip

hosu

raA

cari

Pal

pig

radi

Ric

inul

eiA

rane

aeA

mbl

ypyg

iU

ropy

giO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Tetrapulmonata

PedipalpiHaplocnemata

Novogenuata

Dromopoda

Giribet et al (2002)

ROOT

Micrura

Crypto-gnomae

Sternocoxata

Epimerata

RostrosomataMyliosomata

Neosternata

232 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 3 Consensus trees produced by parsimony analysis of neontological data constrained to produce relationships pro-posed in the recent literature The constrained (target) node is indicated by a black dot and the taxa encompassed by theconstraint are enclosed in a box Numbers below each tree represent the number of minimal-length constrained treeslength of minimal-length trees difference in the length of the unconstrained minimal-length tree and the constrained min-imal-length tree and percentage unconstrained bootstrap trees in which the target node was recovered respectively Thesetrees indicate the effect on branch length imposed by specific hypotheses and impact of constraining target nodes on non-target nodes

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Lipoctena4 385 2 lt5

Xi

Sc

Op

Ps

So

Pa

Ur

Am

Ar

Ri

An

Ac

Labellata6 397 14 lt5

Xi

Sc

Ar

Am

Ur

Op

Pa

Ps

So

Ri

An

Ac

Apulmonata2 390 7 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

So

Ri

An

Ac

Holotracheata5 390 7 lt5

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptoperculata2 388 5 lt5

Xi

Sc

Op

Pa

So

Ps

Ar

Am

Ur

Ri

An

Ac

Rostrosomata6 384 1 25

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Neosternata3 388 5 lt5

Xi

Sc

Op

Ar

Am

Ur

Ri

An

Ps

So

Pa

Ac

Coxisternata2 388 5 lt5

Epimerata7 387 4 11

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptognomae6 384 1 24

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Micrura3 385 2 7

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Megoperculata3 384 1 28

Xi

Sc

Ps

So

Op

Pa

Ar

Am

Ur

Ri

An

Ac

Novogenuata10 388 5 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

Ri

So

An

Ac

Solifugae + Acari4 388 5 lt5

Shultz (1990) Alberti amp Peretti (2002)

Van der Hammen (1989)

Weygoldt amp Paulus (1979)

Xi

Ar

Am

Ur

So

Ps

Sc

Op

Ri

Pa

An

Ac

PHYLOGENY OF THE ARACHNID ORDERS 233

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Results from analyses of node-specific hypothesesare summarized in Figures 3 4 In most cases pro-posed relationships were recovered in fewer than 5of bootstrap pseudoreplicates The notable exceptionswere Rostrosomata (25) Cryptognomae (24) andMegoperculata (28) and the strict-consensus con-strained minimal-length tree for each was only onestep longer than the unconstrained optimal tree Inmost cases constrained topologies did not add manysteps most were less than five steps longer and nonewas greater than 14 steps longer than the un-constrained minimal-length tree It is noteworthyhowever that even constrained nodes that imposerelatively few extra steps sometimes supportedimprobable relationships or eliminated clades thatwere well supported or stable in the unconstrainedanalysis Specific examples and their implications arediscussed below

DISCUSSION

RECENT ISSUES IN ARACHNID PHYLOGENY

Available morphological evidence consistently re-solves some interordinal relationships and fails to

resolve others (Fig 1) Continued progress depends onthe ability of arachnologists to discover new charac-ters and to assess the evidence critically Whilereviewing the recent arachnological literature severalaspects of phylogenetic practice emerged that seemedcounterproductive to both the perception and theactual rate of progress toward resolving arachnid phy-logeny Some of these are summarized here withspecific examples given in the remainder of theDiscussion and in the Appendix

There is a tendency to portray arachnid ordinal phy-logeny as more poorly resolved and contentious than isactually the case (Coddington et al 2004) Phyloge-netic hypotheses generated during different historicalperiods and using differing standards of evidence areoften cited as examples of current disagreement (egSelden 1993 Dunlop 1996 Selden amp Dunlop 1998Wheeler amp Hayashi 1998) In fact recent parsimony-based analyses of morphology have tended to convergeon topologies with internal structures congruent withthose found here (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) (Fig 2) Arachnidphylogeny is not fully resolved especially at its deep-est levels (Fig 1) but this does not mean that allaspects of arachnid phylogeny are controversial orpoorly supported by the available evidence

Matrices are sometimes constructed by uncriticallsquorecyclingrsquo of erroneous or problematic charactersbased on diverse secondary sources (Jenner 2001)Conclusions derived from mixtures of valid invalidspeculative and redundant characters are sometimesportrayed as the phylogenetic signal provided by mor-phology Data recycling can perpetuate errors (see 1320 39 52 136 140 144 152 169 171ndash174) legiti-mize speculations (see 13 32 63 95) or create dupli-cate or non-independent characters (see 13 30 70 77172) The lsquolateral organrsquo is a notable exampleYoshikura (1975) equated the embryonicearly post-embryonic lsquolateral organsrsquo of Amblypygi Thelyphonidaand Solifugae with the dissimilar lsquolateral organrsquo ofXiphosura but failed to note the very similar Cla-paregravede organ of Acariformes (see 173 174) This cod-ing was recycled by Wheeler amp Hayashi (1998) andthen by Giribet et al (2002) Error is probably inevi-table when assembling a large matrix from morphol-ogy including the one presented here but this can beminimized by making original observations and byconsulting primary sources

Some workers advocate weighting characters a pri-ori on the basis of structural or functional complexity(eg Kraus 1998 Dunlop amp Braddy 2001) and dismissphylogenetic conclusions derived from equal-weightsparsimony In short these workers criticize parsimonyfor emphasizing data quantity over data quality Thiscriticism ignores the intense debate in systematic biol-ogy that eventually led to widespread adoption of

Figure 4 Consensus trees produced by parsimony analy-sis of the full data matrix constrained to produce relation-ships proposed in the recent literature See legend toFigure 3 for details

Xi

Eu

Sc

Op

Pa

Ac

Ri

An

So

Ps

Tr

Ar

Ha

Am

UrEurypterida slat + Scorpiones12 435 9 lt5

Xi

Eu

Sc

Op

Pa

Ac

An

So

Ps

Ri

Tr

Ar

Ha

Am

UrTrigonotarbida+ Ricinulei28 434 8 lt5

Dunlop amp Braddy (2002)et al

Dunlop (1996)

234 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 5 Results from analysis of neontological data A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology AFor B and C relationships within terminal clades are the same as those shown in A

LimulusTachypleusProkoeneniaEukoeneniaCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAllothyrusAustralothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesLiphistiusAphonopelmaHypochilusCharinusPhrynusStenochrusProtoschizomusMastigoproctusCyphophthalmusChileogoveaCaddoLeiobunumSclerobunusGonyleptesCentruroidesHadrurusHeterometrusChthoniusFeaellaNeobisiumCheliferEremocostaGaleodes

10031

10018451

501561

733

723

944

633

10010

832

901

943

903

522

341

1007

1008

Acaromorpha

Haplocnemata

10012

Stomothecata

Opiliones

Scorpiones

Pseudo-scorpiones

Solifugae

Acariformes

Parasitiformes

Opilioacariformes

Ricinulei

Xiphosura

Palpigradi

Thelyphonida

Schizomida

Araneae

Amblypygi

10011

936853

100141005

1009

Tetrapulmonata Pedipalpi

Uropygi

692

522572

968

331

603

81

511

161

101

Acari

Dromopoda

Arachnida

A

Xiphosura

Stomothecata

Haplocnemata

Acaromorpha

Palpigradi

TetrapulmonataMicrura

Megoperculata

BXiphosura

Stomothecata

Acaromorpha

Haplocnemata

Palpigradi

TetrapulmonataMegoperculata

C

PHYLOGENY OF THE ARACHNID ORDERS 235

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 6 Results from analysis of the full data matrix A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology CFor B and C relationships within terminal clades are the same as those shown in A

WeinberginaLimuloidesEuproopsTachypleusLimulusStylonurusBaltoeurypterusChasmataspisDiploaspisOctoberaspisCyphophthamusChileogoveaCaddoLeiobunumSclerobunusGonyleptesProscorpiusPalaeoscorpiusStoermeroscorpioPrearcturusCentruroidesHadrurusHeterometrusProkoeneniaEukoeneniaTerpsicrotonPoliocheraCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAustralothyrusAllothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesEremocostaGaleodesChthoniusFeaellaNeobisiumCheliferGilboarachnePalaeocharinusLiphistiusAphonopelmaHypochilusPlesiosiroCharinusPhrynusStenochrusProtoschizomusProschizomusMastigoproctus

631

893

923

984

984

671

341

671

986

10011

423

974

Xiphosura

Eurypterida sstr

Chasmataspidida

Opiliones

Scorpiones

Palpigradi

Ricinulei

Opilioacariformes

Parasitiformes

Acariformes

Solifugae

Pseudoscorpiones

Trigonotarbida

Araneae

Haptopoda

Amblypygi

Schizomida

Thelyphonida

842

251331

361311

552

832

882

925

913

573

561291

443

976

541

954

743644

732521

571

331

833

663591

966

573

71

221

905

832

351

Uropygi

Pedipalpi

Haplocnemata

Tetrapulmonata

Anactinotrichida

Cryptognomae

Acaromorpha

Stomothecata

Pantetrapulmonata

Arachnida

Eurypterida slat

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatProscorpiusStoermeroscorpioPalaeoscorpiusPrearcturusCentruroidesHadrurusHeterometrusOpilionesHaplocnemataRicinuleiAnactinotrichidaAcariformesPalpigradiPantetrapulmonata

Megoperculata

Micrura

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatPalpigradiStomothecataHaplocnemataPantetrapulmonataAcaromorpha

A

B

C

Schizotarsata

236 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

parsimony and instead advocates a return to thespeculative and subjective approaches of the late 19thand early 20th century that once threatened thescientific legitimacy of the discipline (Bowler 1996)Furthermore it misrepresents the properties of parsi-mony-based analyses Specifically characters withfunctional significance are readily encompassed byparsimony analyses and in fact characters derivedfrom locomotor systems (eg 46ndash94) have played animportant role in developing current ideas aboutarachnid phylogeny (eg Shultz 1989 1991) In addi-tion lsquocomplex charactersrsquo can be viewed as compositesof several characters such that morphological com-plexity is effectively weighted by the number of inde-pendent lsquosubcharactersrsquo it contains For example thelsquosucking stomachrsquo once considered a synapomorphy ofLabellata (= Araneae + Amblypygi Fig 3) is codedhere as a composite of three characters (199ndash201) TheLabellata hypothesis was not corroborated in thepresent analysis (Figs 1 3ndash5) and in fact was highlydisfavoured (Fig 3) but this result cannot be dis-missed as a failure to acknowledge the complexity ofthe character

Some workers support specific (target) clades withone or more similarities without exploring the impacton overall homoplasy or relationships among non-target clades Each character offers its own phyloge-netic hypothesis which may or may not be consistentwith relationships implied by other characters It isexceedingly rare for all characters to be perfectly com-patible in the phylogenetic hypotheses they supportand criteria such as parsimony have been developed todiscover those hypotheses that minimize the conflict-ing phylogenetic signals of different characters Eventhough character conflict (homoplasy) is a virtuallyinescapable phenomenon in comparative biology it isnot uncommon for workers to discover one or morecharacters and to promote their phylogeneticimplications over alternative hypotheses even thosethat otherwise appear to be well supported Thisapproach may have value in highlighting new dataand perspectives but accomplishes this by promotingthe erroneous impression that all aspects of arachnidphylogeny are so tenuous that a single character canfalsify even well-supported hypotheses Several exam-ples of this approach have appeared in the recentarachnological literature

For example Alberti amp Peretti (2002) argued thataflagellate spermatozoa (163) are a compelling syna-pomorphy for a Solifugae + Acari clade and dismissedsome of the characters that support Solifugae + Pseu-doscorpiones as having lsquodebatable valuersquo Yet theirproposal rejects two hypotheses that are consistentlyrecovered in recent phylogenetic analyses that isHaplocnemata (= Solifugae + Pseudoscorpiones) andAcaromorpha (= Acari + Ricinulei) (Figs 1 2) Rejec-

tion of Haplocnemata would require its presumed syn-apomorphies to be reinterpreted as homoplasiesincluding features of the chelicerae (18 19 20) pre-oral chamber (13 32) legs (12 48) and respiratorysystem (126) The same reason applies to Acaromor-pha and its synapomorphies It is noteworthy that aSolifugae + Acari clade was recovered in the presentanalysis in fewer than 5 of bootstrap pseudorepli-cates and that analyses constrained to recover thisclade were five steps longer than the minimal-lengthtree and favoured a problematic clade unitingRicinulei and Pseudoscorpiones (Fig 3)

In another example Dunlop (1996) proposed a closerelationship between Trigonotarbida and Ricinuleibased on two-segmented chelicerae (18) prosomandashopisthosoma coupling mechanism (96) diplotergites(100 101) and longitudinally divided opisthosomaltergites (115) However phylogenetic analyses con-strained to recover this relationship required eightadditional steps eliminated support for Acari and nec-essarily rejected Acaromorpha and forced its synapo-morphies to be reinterpreted as homoplasies Alberti(2005) proposed an interesting hypothesis for the evo-lution of male gonads in tetrapulmonates (153) butchose to accept Labellata (= Araneae + Amblypygi) indeveloping his argument over the much more well-supported Pedipalpi (= Amblypygi + Uropygi) (Shultz1999) a phylogenetic reconfiguration that was recov-ered here in fewer than 5 of bootstrap pseudorepli-cates increased tree length by a minimum of 14 stepsand resulted in the collapse of Stomothecata and Hap-locnemata (Fig 3) Many other examples can be cited

Promoting or defending a specific phylogenetichypothesis via lists of compatible synapomorphies is acommon but problematic approach By restrictingattention to the states of specific characters at one ortwo target nodes one can easily overlook the unin-tended impact of the hypothesis on phylogenetic sig-nal elsewhere and its effect on non-target clades Anode supported by a long list of synapomorphies mayseem convincing taken in isolation but may becomeless acceptable when its full phylogenetic implicationsare explored

SUMMARY OF INTERORDINAL ARACHNID CLADES RECOVERED IN THIS ANALYSIS

Arachnida Lamarck 1801Analyses consistently recovered Arachnida as a mono-phyletic group with high bootstrap support (Figs 1 45) Possible synapomorphies include the loss of thecarapacal pleural doublure (9) cardiac lobe (10) pedalgnathobases (52) and moveable endites (53) and thegain of aerial respiration (120) and an anteriorly oranteroventrally directed mouth (185) Some tradi-tional synapomorphies such as slit sensilla (142) and

PHYLOGENY OF THE ARACHNID ORDERS 237

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

fluid feeding (184) may have appeared later in arach-nid evolution but this can only be decided once theinternal phylogeny of Arachnida has been determined

Some workers regard many proposed arachnid syn-apomorphies as adaptations to terrestrial life andthereby link the hypothesis of arachnid monophyly tothe hypothesis of a single ancestral aquatic-to-terres-trial transition and arachnid polyphyly to multipletransitions (eg Selden amp Jeram 1989 Dunlop 1997Dunlop amp Webster 1999 Dunlop amp Braddy 2001)Given this line of reasoning the existence of appar-ently aquatic scorpions in the fossil record (Kjellesvig-Waering 1986 Jeram 1998) and the inference thatterrestrialization occurred late in scorpion evolutionthese workers conclude that character states support-ing arachnid monophyly are actually convergencesand do not necessarily support arachnid monophyly

However this approach to assessing phylogenetichypotheses is founded on overly simplistic assump-tions such as the ability of the investigator to discrim-inate unerringly between characters that existexclusively in aquatic organisms (including fossils)from those that occur exclusively in terrestrial organ-isms There also appears to be an assumption thathomoplasy can be generated through parallelism (iemultiple aquatic-to-terrestrial events) but not throughterrestrial-to-aquatic reversals Furthermore anassumed dichotomy between exclusively aquatic andterrestrial life histories in ancestral arachnids is sim-plistic as illustrated by the amphibious life cycles ofbasal vertebrates and pterygote hexapods In factthese examples demonstrate that there is no neces-sary inconsistency in basing a hypothesis of arachnidmonophyly on the derived terrestrial features of anamphibious ancestor whose descendants thencompleted terrestrialization once or several timesindependently or even returned to a fully aquaticexistence Workers who link the frequency and direc-tion of aquaticndashterrestrial transitions to the assess-ment of arachnid phylogeny do so by endowingthemselves with substantially greater insight thanseems prudent by ignoring the huge gaps in ourunderstanding of early arachnid evolution and bydenying to arachnids the evolutionary complexityknown to exist in other groups

Several palaeontologists have been particularlyactive during the past decade in proposing new char-acters with the stated goal of removing Scorpionesfrom Arachnida and erecting a Scorpiones + Eurypter-ida clade (Braddy amp Dunlop 1997 Dunlop amp Braddy1997 Dunlop 1998 Braddy et al 1999 Dunlop ampWebster 1999) Dunlop amp Braddy (2001) recentlysummarized this evidence and conducted aparsimony-based analysis of Xiphosura EurypteridaScorpiones Opiliones and Tetrapulmonata (but notHaplocnemata Acaromorpha or Palpigradi) using 33

morphological characters Their analysis produced atopology congruent with those generated here (Figs 15 6) including recovery of Stomothecata (= Scorpiones+ Opiliones) However they rejected this result as aproduct of lsquoempirical cladisticsrsquo because it gives thesame weight to prosomal characters that supportarachnid monophyly and to selected opisthosomalcharacters that support their favoured Eurypterida +Scorpiones clade namely a five-segmented metasoma(116) suppression of opisthosomal tergite 1 (95) lossof lamellate respiratory organs on the postgenitalsomite (122) Kiemenplatten (125) loss of respiratorylamellae on the genital segment (121) and a lsquonon-stainingrsquo exocuticle (but see Grainge amp Pearson 1966for evidence of this in Opiliones see Appendix for com-ments on the other characters)

A Eurypterida + Scorpiones clade was not favouredin the present analysis (Fig 6) the strict consensus ofminimal-length trees constrained to recover this clade(Fig 4) is nine steps longer than the tree recovered byanalysis without this constraint It is also noteworthythat Opiliones was consistently reconstructed as thesister group to Eurypterida + Scorpiones in the mini-mal-length constrained trees a provocative result thatwas probably unintended and unanticipated by sup-porters of the Eurypterida + Scorpiones Given thatDunlop amp Braddy (2001) reject equal weights parsi-mony as an arbiter of phylogenetic hypotheses theywould presumably dismiss these results as irrelevantto their argument just as they dismissed their ownparsimony-based results However if the Eurypterida+ Scorpiones hypothesis is to be credible it must beopen to evaluation and potential falsification usingobjective criteria and the subjective or intuitive a pri-ori weighting of characters advocated by Dunlop ampBraddy clearly does not qualify At present it isunclear how one would objectively evaluate Dunlop ampBraddyrsquos proposal with criteria compatible with thoseused in its original formulation For now the Euryp-terida + Scorpiones concept advocated by Dunlop ampBraddy may persist outside the mainstream of mod-ern systematic practice but it is increasingly problem-atic within it

Stomothecata nomnovOpiliones and Scorpiones were consistently recoveredas a monophyletic group The proposed name acknowl-edges a unique preoral chamber the stomothecaformed by coxapophyses of the palp and leg 1 (50)often with an auxiliary role played by the coxapophy-sis of leg 2 (51) In addition the epistome appears tohave been modified for adduction of the palpal coxaeThe lateral walls of the epistome are fused to themedial surfaces of the palpal coxae and the epistomallumen is spanned by a transverse muscle (188) which

238 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

apparently adducts the palpal coxae thereby constrict-ing the stomothecal chamber Scorpions and opilionsare also unique in having a pair of large epistomalarms projecting rearward into the prosoma andattaching to the endosternite (189) The epistomalarms provide attachment sites for pharyngeal dilatormuscles (196) and extrinsic muscles of anterior proso-mal appendages (eg 37) The chelicera is equippedwith a muscle that arises on the carapace and insertson the ventral margin of the second cheliceral article(23) There is an anteriorly placed genital opening(155)

Several notable similarities were found whilereviewing the literature but information from otherarachnid groups was considered too incomplete toallow them to be included in the matrix For exampleboth orders have apparent haemocytopoeitic organsassociated with major nerves of the anterior opistho-soma These are termed supraneural organs in scorpi-ons (Farley 1999) and perineural organs in opilions(Kaumlstner 1935) Haemocytes develop in the cardiacwall in spiders (Seitz 1972) and perhaps amblypygids(Weygoldt 2000) but haemocytopoeitic organs areunknown in most other arachnid groups Germ cellsdifferentiate very early during embryogenesis in bothscorpions and opilions relative to spiders (Moritz1957 Anderson 1973) Additional research is requiredto determine the phylogenetic utility of thesecharacters

Haplocnemata Boumlrner 1904Several workers recognize a close phylogenetic rela-tionship between Pseudoscorpiones and Solifugae(Weygoldt amp Paulus 1979 Van der Hammen 1989Shultz 1990 Wheeler amp Hayashi 1998 Giribetet al 2002) and this was consistently recovered inthe present analysis Haplocnemata is supported byseveral synapomorphies including (i) a feeding com-plex formed by two-segmented chelicerae (18) with aventrolateral intrinsic articulation (19) and perhapsa dorsolateral articulation with the carapace (20)and (ii) a rostrosoma (32) a preoral apparatus formedby an anteriorly projecting epistome affixed dorsallyto enlarged palpal coxae and bordered by lateral pal-pal projections The epistome forms the dorsal compo-nent of a lsquobeakrsquo with the ventral part formed by amidventral sternapophysis (13) fused between thepalpal coxae The locomotor apparatus consists ofcoxae that meet along the ventral midline (12) elon-gate femur-like patellae (48) and an apotele in theadult modified as an eversible empodium or pulvillus(92) The respiratory system is formed entirely (Pseu-doscorpiones) or in part (Solifugae) by paired trachealstigmata opening on opisthosomal somites 3 and 4(126)

Acaromorpha Dubinin 1957A clade comprising Ricinulei and Acari is consistentlyrecovered in the present analysis It is united here bytwo unique and seven homoplasious synapomorphiesThese include a gnathosoma (31) defined in part bymedial fusion of the palpal coxae (30) although pres-ence of a gnathosoma in Ricinulei is debatable as dis-cussed in the Appendix Acaromorphs also have aunique post-embryonic development consisting of ahexapodal larva and up to three octopodal nymphalinstars (176) The pedal patellandashtibia joints are formedby a bicondylar dorsal hinge rather than a single mid-dorsal condyle (69) all postcheliceral segmental gan-glia are unified in the subesophageal ganglion (130)and a postcerebral pharynx is absent (192) The groupis also tentatively united by presence of differentiatedpedal basi- and telofemora (63 64) and the absence ofa ventral (sternal) pharyngeal dilator muscle (199)but these may be symplesiomorphies erroneouslyreconstructed as a synapomorphies

The internal phylogenetic structure of Acaromorphais controversial with many recent workers favouringa monophyletic Acari (Weygoldt amp Paulus 1979Shultz 1990 argued most thoroughly by Lindquist1984) and others advocating a diphyletic Acari withAcariformes being the sister group to Cryptognomae(= Ricinulei + Anactinotrichida) (especially Van derHammen 1979 1989) (Fig 3) Acari was recovered asmonophyletic when fossils were excluded (Fig 2) butnot when they were included (Fig 3) A lsquomite-centredrsquosurvey of arachnid characters may be needed if mor-phology is to offer a compelling solution to the internalphylogeny of Acari and its placement within Arach-nida These issues are not resolved by the presentanalysis

Pantetrapulmonata nom novPantetrapulmonata includes the extinct orders Trigo-notarbida Haptopoda and the extant orders AraneaeAmblypygi Schizomida and Thelyphonida The cladeis united by cheliceral structure (18 19) a megoper-culum (106) booklungs on the genital and first post-genital somites (121 122) and enlargement of theepipharyngeal sclerite (192) It is important to notehowever that most of these characters were coded asuncertain in Plesiosiro (Haptopoda) Aside from theplacement of Haptopoda the monophyly of Trigono-tarbida and the extant orders was anticipated byShear et al (1987) and is generally regarded as amonophyletic group

Tetrapulmonata Shultz 1990Tetrapulmonata was originally proposed on the basisof neontological analyses and encompassed Araneae

PHYLOGENY OF THE ARACHNID ORDERS 239

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Amblypygi Schizomida and Thelyphonida (Shultz1990) and was also recovered here in analysis of neon-tological data (Fig 5) However results from analysisof all taxa required that the Tetrapulmonata conceptbe expanded to include the fossil order HaptopodaFeatures uniting this group are problematic howeveras many states in Haptopoda were coded as uncertain

Schizotarsata nomnovHaptopoda and Pedipalpi (= Amblypygi + Schizomida+ Thelyphonida) are united here in a group named forpossession of divided pedal telotarsi (84) Synapomor-phies include a pointed anterior carapacal margin (3)and elongation of leg 1 (46) As already noted theplacement of Haptopoda should probably be regardedas tentative because the state of many characters inthis group was coded as unknown The placement ofHaptopoda as the sister group to Pedipalpi was antic-ipated by Dunlop (1999 2002c)

Pedipalpi Boumlrner 1904Pedipalpi encompasses Amblypygi and Uropygi(= Thelyphonida + Schizomida) Although widely rec-ognized in the past Pedipalpi was set aside in the mid-20th century in favour of Labellata (= Amblypygi +Araneae) based largely on the presumed synapomor-phies of a narrow prosomandashopisthosoma juncture (97)and sucking stomach (199ndash201) The situation beganto reverse when Shear et al (1987) again highlightedsimilarities of Amblypygi and Uropygi such as rapto-rial palps (35) and antenniform leg 1 (46) Shultz(1989 1990 1999) subsequently described numerousderived similarities including asymmetrical flexormuscles at the pedal femurndashpatella joint (70) threetelotarsomeres on legs 2ndash4 (85) and modification ofthe palpal coxae to provide support for extrinsic pha-ryngeal muscles (197) Pedipalpi has emerged as oneof the most well-supported interordinal relationshipsin Arachnida

Uropygi Thorell 1882Uropygi has long been accepted as a monophyleticunion of Thelyphonida and Schizomida Synapomor-phies include a unique mating behaviour (159) fusedpalpal coxae (30) 2-1-1-1 arrangement of tibial tricho-bothria (144) posterior defensive glands (102) elon-gated patella of leg 1 (68) and many others

ACKNOWLEDGEMENTS

I thank Nikolaj Scharff and three anonymous review-ers for useful comments This work was supported bythe Maryland Agricultural Experiment Station the

General Research Board of the University of Mary-land and National Science Foundation Grant 0416628to Susan Masta

REFERENCES

Abd el-Wahab A 1952 Notes on the morphology of the scor-pion Buthus quinquestriatus (HE) Publications de lrsquoInsti-tut Fouad 1er du Desert 3 1ndash129

Alberti G 1979 Licht- und elektronenmikroskopische Unter-suchungen an Coxaldruumlsen von Walzenspinnen (ArachnidaSolifugae) Zoologischer Anzeiger 203 48ndash64

Alberti G 1995 Comparative spermatology of Cheliceratareview and perspective Memoires du Museum NationaldrsquoHistoire Naturelle 166 203ndash230

Alberti G 2005 Double spermatogenesis in ChelicerataJournal of Morphology 266 281ndash297

Alberti G Coons LB 1999 Acari mites In Harrison FWFoelix RF eds Microscopic anatomy of invertebratesVol 8C Chelicerate Arthropoda New York Wiley Liss515ndash1215

Alberti G Peretti AV 2002 Fine structure of male genitalsystem and sperm in Solifugae does not support a sister-group relationship with Pseudoscorpiones (Arachnida)Journal of Arachnology 30 268ndash274

Alexander AJ 1967 Problems of limb extension in the scor-pion Opisthophthalmus latimanus Koch Transactions of theRoyal Society of South Africa 37 165ndash181

Anderson DT 1973 Embryology and phylogeny in annelidsand arthropods Oxford Pergamon Press

Anderson LI Selden PA 1997 Opisthosomal fusion andphylogeny of Palaeozoic Xiphosura Lethaia 30 19ndash31

Beall BS Labandeira CC 1990 Macroevolutionary patternsof the Chelicerata and Tracheata In Culver SJ ed Arthro-pod paleobiology Short Courses in Paleontology No 3 Lon-don The Paleontological Society 257ndash284

Bernard HM 1896 The comparative morphology of the Gale-odidae Transactions of the Linnean Society of London Sec-ond Series 6 305ndash417

Boumlrner C 1902a Arachnologische Studien (II und III) Zool-ogischer Anzeiger 25 433ndash466

Boumlrner C 1902b Arachnologische Studien IV Die Genitalor-gane der Pedipalpen Zoologischer Anzeiger 26 81ndash92

Boumlrner C 1904 Beitraumlge zur Morphologie der Arthropoden IEin Beitrag zur Kenntnis der Pedipalpen Zoologica (Stut-tgart) 42 1ndash174

Bowler PJ 1996 Lifersquos splendid drama Chicago Universityof Chicago Press

Braddy SJ Alridge RJ Gabbott SE Theron JN 1999Lamellate book-gills in a late Ordovician eurypterid from theSoom Shale South Africa support for a eurypterid-scorpionclade Lethaia 32 72ndash74

Braddy SJ Dunlop JA 1997 The functional morphology ofmating in the Silurian eurypterid Baltoeurypterus tet-ragonophthalmus (Fischer 1839) Zoological Journal of theLinnean Society 121 435ndash461

Brauer A 1895 Beitrage zur Kenntnis der Entwicklungsge-schichte des Skorpiones II Zeitschrift fuumlr WissenschaftlicheZoologie 59 351ndash433

240 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bremer K 1994 Branch support and tree stability Cladistics10 295ndash304

Bristowe WS 1932 The liphistiid spiders Proceedings of theZoological Society of London Part 4 1016ndash1057

Butt AG Taylor HH 1991 The function of spider coxalorgans effects of feeding and salt-loading on Porrhotheleantipodiana (Mygalomorpha Dipluridae) Journal of Exper-imental Biology 158 439ndash461

Buxton BH 1913 Coxal glands of arachnids ZoologischeJahrbucher (Suppl XIV) 16ndash50

Buxton BH 1917 Notes on the anatomy of arachnids Jour-nal of Morphology 29 1ndash31

Camin JH Clark GM Bourdeau FG 1956 The palpallsquotined setarsquo in the Mesostigmata a homologue of the palpalclaw in the Onychopalpida (Acarina) In Becker EC ed Pro-ceedings of the Tenth International Congress of EntomologyOttawa Mortimer 903ndash908

Chamberlin JC 1931 The arachnid order ChelonethidaStanford University Stanford University Press

Chapman RF 1998 The insects structure and function 4thedn Cambridge Cambridge University Press

Clarke KU 1979 Visceral anatomy and arthropod phylogenyIn Gupta AP ed Arthropod phylogeny New York Van Nos-trand Reinhold 467ndash549

Clarke J 1984 On the relationship between structure andfunction in the leg joints of Heteropoda venatoria (L) (Ara-neae Eusparassidae) Bulletin of the British ArachnologicalSociety 6 181ndash192

Clarke J 1986 The comparative functional morphology of theleg joints and muscles of five spiders Bulletin of the BritishArachnological Society 7 37ndash47

Clarke JM Ruedemann R 1912 The Eurypterida of NewYork New York State Museum Memoir 14 (1ndash2) 1ndash628

Coddington JA 2005 Phylogeny and classification of spi-ders In Ubick D Paquin P Cushing PE Roth V eds Spi-ders of North America an identification manual AmericanArachnological Society 18ndash24

Coddington JA Giribet G Harvey MS Prendini LWalter DE 2004 Arachnida In Cracraft J Donoghue MJeds Assembling the tree of life Oxford Oxford UniversityPress 269ndash318

Coddington JA Levi HW 1991 Systematics and evolutionof spiders (Araneae) Annual Review of Ecology and System-atics 22 565ndash592

Coineau Y Van der Hammen L 1979 The postembryonicdevelopment of Opilioacarida with notes on new taxa and ona general model for the evolution In Piffle E ed Proceed-ings of the 4th International Congress of Acarology Budap-est Akadeacutemiai Kiadoacute 437ndash441

Cokendolpher JC Reddell JR 1992 Revision of the Pro-toschizomidae (Arachnida Schizomida) with notes on thephylogeny of the order Texas Memorial Museum Speleolog-ical Monographs 3 31ndash74

Condeacute B 1991a Le dimorphisme sexuel des Palpigrades Bul-letin de la Socieacuteteacute Neuchacircteloise des Sciences Naturelles 11667ndash73

Condeacute B 1991b Lrsquoaire genitale male des Palpigrades Bulle-tin de la Socieacuteteacute Europeenne drsquoArachnologie 1990 64ndash69

Dearden PK Donly C Grbicccc M 2002 Expression of pair-rule gene homologues in a chelicerate early patterning ofthe two-spotted spider mite Tetranychus urticae Develop-ment 129 5461ndash5472

Delle Cave L 1975 The ldquocleaning brushrdquo in Damon diademaSimon 1876 and Phrynichus jayakari Pocock 1894 (Ambly-pygi Arachnida) Monitore Zoologico Italiano (NS) 9 175ndash184

Dunlop JA 1994 Filtration mechanisms in the mouthparts oftetrapulmonate arachnids (Trigonotarbida Araneae Ambly-pygi Uropygi Schizomida) Bulletin of the British Arachno-logical Society 9 267ndash273

Dunlop JA 1996 Evidence for a sister group relationshipbetween Ricinulei and Trigonotarbida Bulletin of the BritishArachnological Society 10 193ndash204

Dunlop JA 1997 Palaeozoic arachnids and their significancefor arachnid phylogeny In Zabka M ed Proceedings of the16th European colloquium of arachnology Siedlce PolandWyzsza Szkola Rolnicko-Pedagogiczna 65ndash82

Dunlop JA 1998 The origins of tetrapulmonate book lungsand their significance for chelicerate phylogeny In SeldenPA ed Proceedings of the 17th European Colloquium ofArachnology Edinburgh 1997 Burnham Beeches BritishArachnological Society 9ndash16

Dunlop JA 1999 A redescription of the Carboniferous arach-nid Plesiosiro madeleyi Pocock 1911 (Arachnida Hapto-poda) Transactions of the Royal Society of Edinburgh EarthSciences 90 29ndash47

Dunlop JA 2000 The epistomo-labral plate and lateral lips insolifuges pseudoscorpions and mites Ekoloacutegia (Bratislava)19 67ndash78

Dunlop JA 2002a Arthropods from the Lower Devonian Sev-ernaya Zemlya Formation of October Revolution Island(Russia) Geodiversitas 24 349ndash379

Dunlop JA 2002b Character states and evolution of the che-licerate claws In Toft S Scharff N eds European Arach-nology 2000 Aarhus Aarhus University Press 345ndash354

Dunlop JA 2002c Phylogeny of Chelicerata In LlorenteBousquets J Morrone JJ eds Biodiversidad Taxonomiacutea yBiogeografiacutea de Arthroacutepodos de Meacutexico Hacia Una Siacutentesisde Su Conocimiento Vol III Prensas de Ciencias Univer-sidad Nacional Autonoma de Meacutexico 117ndash141

Dunlop JA Anderson LI Braddy SJ 2004 A redescriptionof Chasmataspis laurencii Caster amp Brooks 1956 (Chelicer-ata Chasmataspidida) from the Middle Ordovician of Ten-nessee USA with remarks on chasmataspid phylogenyTransactions of the Royal Society of Edinburgh Earth Sci-ences 94 207ndash225

Dunlop JA Braddy SJ 1997 Slit-like structures on theprosomal appendages of the eurypterid BaltoeurypterusNeues Jahrbuch fuumlr Geologie und Palaumlontologie-Monatshefte1997 31ndash38

Dunlop JA Braddy SJ 2001 Scorpions and their sister-group relationships In Fet V Selden PA eds Scorpions2001 in Memoriam Gary A Polis London British Arachno-logical Society 1ndash24

Dunlop JA Horrocks CA 19951996 A new Upper Carbon-iferous whip scorpion (Arachnida Uropygi Thelyphonida)

PHYLOGENY OF THE ARACHNID ORDERS 241

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

with a revision of the British Carboniferous Uropygi Zoolo-gischer Anzeiger 234 293ndash306

Dunlop JA Poschmann M Anderson L 2001 On theEmsian (early Devonian) arthropods of the Rhenish SlateMountains 3 The chasmataspidid Diploaspis Palaumlon-tolgische Zeitschrift 75 253ndash269

Dunlop JA Webster M 1999 Fossil evidence terrestrializa-tion and arachnid phylogeny Journal of Arachnology 27 86ndash93

Edgecombe GD Wilson GDF Colgan DJ Gray MR Cas-sis G 2000 Arthropod cladistics combined analysis of his-tone H3 and U2 snRNA sequences and morphologyCladistics 16 155ndash203

Evans GO 1992 Principles of acarology Wallingford UKCABI

Fahrenbach WH 1999 Merostomata In Harrison FW Foe-lix RF eds Microscopic anatomy of invertebrates Vol 8AChelicerate arthropods New York Wiley-Liss 21ndash115

Farley RD 1999 Scorpiones In Harrison FW Foelix RF edsMicroscopic anatomy of invertebrates Vol 8A Cheliceratearthropods New York Wiley-Liss 117ndash222

Farley RD 2005 Developmental changes in the embryo pro-nymph and first molt of the scorpion Centruroides vittatus(Scorpiones Buthidae) Journal of Morphology 265 1ndash27

Fayers SR Dunlop JA Trewin NH 2004 A new earlyDevonian trigonotarbid arachnid from the Windyfield ChertRhynie Scotland Journal of Systematic Palaeontology 2269ndash284

Felsenstein J 1985 Confidence limits on phylogenies anapproach using the bootstrap Evolution 39 783ndash791

Fet V Sissom WD Lowe G Braunwalder ME 2000 Cat-alog of the scorpions of the world (1758ndash1998) New YorkNew York Entomological Society

Firstman B 1954 The central nervous system musculatureand segmentation of the cephalothorax of a tarantula (Eu-rypelma californicum Ausserer) Microentomology 87 1ndash40

Firstman B 1973 The relationship of the chelicerate arterialsystem to the evolution of the endosternite Journal ofArachnology 1 1ndash54

Foelix RF 1985 Mechano- and chemoreceptive sensilla InBarth FG ed Neurobiology of arachnids Berlin Spinger-Verlag 118ndash137

Foelix RF 1996 Biology of spiders 2nd edn Oxford OxfordUniversity Press

Giribet G Edgecombe GD Wheeler WC Babbit C 2002Phylogeny and systematic position of Opiliones a combinedanalysis of chelicerate relationships using morphologicaland molecular data Cladistics 18 5ndash70

Goloboff P 1993 Estimating character weights during treesearch Cladistics 9 83ndash91

Goloboff PA Farris JS Nixon K 2000 TNT Tree AnalysisUsing New Technologies Version 1 Program and documen-tation httpwwwzmucdkpublicphylogenytnt

Grainge CA Pearson RG 1966 Cuticular structure in thePhalangida Nature 211 866

Grandjean F 1938 Observations sur les Bdelles (Acariens)Annales de la Socieacuteteacute Entomologique de France Paris 107 1ndash24

Hansen HJ Soslashrensen W 1904 On two orders of ArachnidaCambridge Cambridge University Press

Hansen HJ Soslashrensen W 1905 The Tartarides a tribe of theorder Pedipalpi Arkiv fuumlr Zoologi 2 1ndash78

Hara MR Gnaspini P 2003 Comparative study of the defen-sive behavior and morphology of the gland opening areaamong harvestmen (Arachnida Opiliones Gonyleptidae)under a phylogenetic perspective Arthropod Structure andDevelopment 32 257ndash275

Harvey MS 1992 The phylogeny and classification of thePseudoscorpionida (Chelicerata Arachnida) InvertebrateTaxonomy 6 1373ndash1435

Harvey MS 2002 The neglected cousins what do we knowabout the smaller arachnid orders Journal of Arachnology30 357ndash372

Haupt J Song D 1996 Revision of East Asian whip scorpions(Arachnida Uropygi Theyphonida) I China and JapanArthropoda Selecta 5 43ndash52

Hjelle JT 1990 Anatomy and morphology In Polis GA edThe biology of scorpions Stanford University Stanford Uni-versity Press 9ndash63

Janczyk FSW 1956 Anatomie von Siro duricorius Joseph imVergleich mit anderen Opilioniden Sitzungsberichte Oster-reichischen Akademie der Wissenschaften Mathematisch-Natruwissenschaftliche Klasse 165 474ndash522

Jenner RA 2001 Bilaterian phylogeny and uncritical recy-cling of morphological data sets Systematic Biology 50 730ndash742

Jeram AJ 1998 Phylogeny classification and evolution of Sil-urian and Devonian scorpions In Selden PA ed Proceed-ings of the 17th European Colloquium of ArachnologyEdinburgh 1997 Burnham Beeches UK British Arachno-logical Society 15ndash31

Karaman IM 2005 Evidence of spermatophores in Cyphoph-thalmi (Arachnida Opiliones) Revue Suisse de Zoologie 1123ndash11

Kaumlstner A 1935 Die Funktion der sogenannten sympathis-chen Ganglien und die Exkretion bei den PhalangiidenZoologischer Anzeiger 109 273ndash288

Kierans JE Robbins RG 1999 A world checklist of generasubgenera and species of ticks (Acari Ixodida) publishedfrom 1973 to 1997 Journal of Vector Ecology 24 115ndash129

Kjellesvig-Waering EN 1986 A restudy of the fossil Scorpi-onida of the world Palaeontographica Americana 55 1ndash287

Klompen JSH 2000 Prelarva and larva of Opilioacarus (Neo-carus) texanus (Chamberlin and Mulaik) (Acari Opilioacar-ida) with notes on the patterns of setation and lyrifissuresJournal of Natural History 34 1977ndash1992

Kraus O 1998 Elucidating the historical process of phylog-eny phylogenetic systematics versus cladistic techniquesIn Selden PA ed Proceedings of the 17th European Collo-quium of Arachnology Edinburgh 1997 Burnham BeechesUK British Arachnological Society 1ndash7

Kraus O Kraus M 1993 Divergent transformation of cheli-cerae and original arrangement of eyes in spiders (Arach-nida Araneae) Memoirs of the Queensland Museum 33579ndash584

Lankester ER Benham WBS Beck EJ 1885 On the mus-

242 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

cular and endoskeletal systems of Limulus and Scorpio withsome notes on the anatomy and generic characters of scor-pions Transactions of the Zoological Society of London 11311ndash384

Leclerc P 1989 Considerations paleobiogeographiques a pro-pos la decouverte en Thailande drsquoopilioacariens nouveaux(Acari ndash Notostigmata) Compte Rendu des Seances de laSociete de Biogeographie 65 162ndash174

Legg G 1976 The external morphology of a new species ofricinuleid (Arachnida) from Sierra Leone Zoological Journalof the Linnean Society 59 1ndash58

Lindquist EE 1984 Current theories on the evolution ofmajor groups of Acari and on their relationships with othergroups of Arachnida with consequent implications for theirclassificaton In Griffiths DA Bowman CE eds AcarologyVI Vol 1 New York John Wiley amp Sons 28ndash62

Manning PL Dunlop JA 1995 The respiratory organs ofeurypterids Palaeontology 38 287ndash297

Manton SM 1958 Hydrostatic pressure and leg exten-sion in arthropods with special reference to arachnidsAnnals and Magazine of Natural History Series 13 1161ndash182

Manton SM 1964 Mandibular mechanisms and the evolutionof arthropods Philosophical Transactions of the Royal Soci-ety of London B 7 1ndash183

Maraun M Heethoff M Schneider K Scheu S WeigmannG Cianciolo J Thomas RH Norton RA 2004 Molecularphylogeny of oribatid mites (Oribatida Acari) evidence formultiple radiations of parthenogenetic lineages Experimen-tal and Applied Acarology 33 183ndash201

Marples BJ 1967 The spinnerets and epiandrous glands ofspiders Journal of the Linnean Society (Zoology) 46 209ndash222

Marples BJ 1983 Observations on the structure of the fore-gut of spiders Bulletin of the British Arachnological Society8 46ndash52

Millot J 1933 Notes compleacutementaires sur lrsquoanatomie desliphistiides et des hypchilides a propos drsquoun travail reacutecent deA Petrunkevitch Bulletin de la Socieacuteteacute Zoologique de France68 217ndash235

Millot J 1942 Sur lrsquoanatomie et lrsquohistophysiologie deKoenenia mirabilis Grassi (Arachnida Palpigradi) RevueFranccedilaise drsquoEntomologie 9 33ndash51

Millot J 1943 Notes complementaires sur lrsquoanatomielrsquohistologie et la repartition geographique en France deKoenenia mirabilis Grassi Revue Franccedilaise drsquoEntomologie 9127ndash135

Millot J 1945 Lrsquoanatomie interne des Ricinulei (Arachnides)Annales des Sciences Naturelles (Zoologie) Series 11 7 1ndash29

Millot J 1949a Morphologie generale et anatomie interne InGrasseacute P-P ed Traite de zoologie VI Paris Masson et Cie263ndash319

Millot J 1949b Ordre des Amblypyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 563ndash588

Millot J 1949c Ordre des Araneacuteides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 589ndash743

Millot J 1949d Ordre des Palpigrades In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 520ndash532

Millot J 1949e Ordre des Uropyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 533ndash562

Millot J 1949f Ordre des Ricinuleides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 744ndash760

Millot J Vachon M 1949 Ordre Des Solifuges In Grasseacute P-P ed Traite de zoologie VI Paris Masson et Cie 482ndash519

Moore RA Briggs DEG Bartels C 2005 A new specimen ofWeinbergina opitzi (Chelicerata Xiphosura) from the LowerDevonian Huumlnsruck Slate Germany PalaumlontologischeZeitschrift 79 399ndash408

Moritz M 1957 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) unter besonderer Beruumlcksichtigungder ausseren Morphologie der Bildung des Mitteldarmesund der Genitalanlage Zoologische Jahrbuumlcher Abteilungfuumlr Anatomie und Ontogenie der Tiere 76 331ndash370

Moritz M 1959 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) II Die Anlage und Entwicklung derCoxaldruumlse bei Phalangium opilio L Zoologische Jahr-buumlcher Abteilung fuumlr Anatomie und Ontogenie der Tiere 77229ndash240

Nixon K 1999 The parsimony ratchet a new method forrapid parsimony analysis Cladistics 15 407ndash414

Norton RA 1998 Morphological evidence for the evolutionaryorigin of Astigmata (Acari Acariformes) Experimental andApplied Acarology 22 559ndash594

Palmgren P 1978 On the muscular anatomy of spiders ActaZoologica Fennica 155 1ndash41

Patten W 1912 The evolution of the vertebrates and their kinPhiladelphia P Blakistonrsquos Son

Patten W Hazen AP 1900 The development of the coxalgland branchial cartilages and genital ducts of Limuluspolyphemus Journal of Morphology 16 459ndash490

Paulus HF 1979 Eye structure and the monophyly of theArthropoda In Gupta AP ed Arthropod phylogeny NewYork Van Nostrand Reinhold 299ndash383

Petrunkevitch A 1949 A study of Palaeozoic ArachnidaTransactions of the Connecticut Academy of Arts and Sci-ences 37 69ndash315

Pittard K Mitchell RW 1972 Comparative morphology ofthe life stages of Cryptocellus pelaezi (Arachnida Ricinulei)Graduate Studies Texas Tech University 1 1ndash77

Platnick NI 2005 The world spider catalog Version55 American Museum of Natural History httpresearchamnhorgentomologyspiderscatalogindexhtm

Platnick NI Gertsch WJ 1976 The suborders of spiders acladistic analysis (Arachnida Araneae) American MuseumNovitates 2607 1ndash15

Pocock RI 1893 On some points in the morphology of theArachnida (ss) with notes on the classification of the groupAnnals and Magazine of Natural History Series 6 11 1ndash19

Pocock RI 1902 Some points in the morphology and classi-fication of the Opiliones Annals and Magazine of NaturalHistory Series 7 10 504ndash516

Popadicccc A Panganiban G Rusch D Shear WA KaufmanTC 1998 Molecular evidence for the gnathobasic derivationof arthropod mandibles and for the appendicular origin ofthe labrum and other structures Development Genes andEvolution 208 142ndash150

PHYLOGENY OF THE ARACHNID ORDERS 243

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Raw F 1957 Origin of chelicerates Journal of Paleontology31 139ndash192

Reddell JR Cokendolpher JC 1995 Catalogue bibliogra-phy and generic revision of the order Schizomida (Arach-nida) Texas Memorial Museum Speleological Monographs4 1ndash170

Reissland A Goumlrner P 1985 Trichobothria In Barth FGed Neurobiology of arachnids Berlin Springer-Verlag 138ndash161

Roewer C-F 1934 Solifugae Palpigradi Bronns Klassen undOrdnungen ses Tierreich 5 IV 4 1ndash713

Rowland JM Cooke JAL 1973 Systematics of the arachnidorder Uropygida (=Thelyphonida) Journal of Arachnology 155ndash71

Rowland JM Sissom WD 1980 Report on a fossil palpi-grade from the Tertiary of Arizona and a review of themorphology and systematics of the order (ArachnidaPalpigradida) Journal of Arachnology 8 69ndash86

Rucker A 1901 The Texan Koenenia American Naturalist35 615ndash630

Ruhland M Rathmayer W 1978 Die Beinmuskulatur undihre Innervation bei der Vogelspinne Dugesiella henzti (Ch)(Araneae Aviculariidae) Zoomorphologie 89 33ndash46

Saboori A Kamali K 2000 Description of Allothrombiumtriticium adult (Acari Trombidiidae) from Iran Systematicand Applied Acarology 5 207ndash208

Schliwa M 1979 The retina of the phalangid Opilio raven-nae with particular reference to arhabdomeric cells Celland Tissue Research 204 473ndash495

Scholl G 1977 Beitrage zur Embryonalentwicklung von Lim-ulus polyphemus L (Chelicerata Xiphosura) Zoomorpholo-gie 86 99ndash154

Scholtz G Kamenz C 2006 The book lungs of Scorpionesand Tetrapulmonata (Chelicerata Arachnida) evidence forhomology and a single terrestrialization event of a commonarachnid ancestor Zoology 109 2ndash13

Seitz KA 1972 Zur Histologie und Feinstruktur des Herzensund der Haumlmocyten von Cupiennius salei Keys (AraneaeCtenidae) I Herzwandung Bildung und Differenzierungder Haumlmocyten Zoologische Jahrbuumlcher Abteilung fuumlr Anat-omie und Ontogenie der Tiere 89 351ndash384

Sekiguchi K ed 1988 Biology of horseshoe crabs Tokyo Sci-ence House

Selden PA 1981 Functional morphology of the prosoma ofBaltoeurypterus tetragonophthalmus (Fischer) (ChelicerataEurypterida) Transactions of the Royal Society of Edin-burgh Earth Sciences 72 9ndash48

Selden P 1992 Revision of the fossil ricinuleids Transactionsof the Royal Society of Edinburgh Earth Sciences 83 595ndash634

Selden PA 1993 Fossil arachnids ndash recent advances andfuture prospects Memoirs of the Queensland Museum 33389ndash400

Selden PA Dunlop JA 1998 Fossil taxa and relationshipsof chelicerates In Edgecombe GD ed Arthropod fossilsand phylogeny New York Columbia University Press 303ndash331

Selden PA Jeram AJ 1989 Palaeophysiology of terrestrial-

isation in the Chelicerata Transactions of the Royal Societyof Edinburgh Earth Sciences 80 303ndash310

Selden PA Shear WA Bonamo PM 1991 A spider and otherarachnids from the Devonian of New York and reinterpre-tations of Devonian Araneae Palaeontology 34 241ndash281

Sensenig A Shultz JW 2003 Mechanics of cuticular elasticenergy storage in leg joints lacking extensor muscles inarachnids Journal of Experimental Biology 206 771ndash784

Shear WA 1993 New species in the opilionid genus Stylocel-lus from Malaysia Indonesia and the Philippines (OpilionesCyphophthalmi Stylocellidae) Bulletin of the British Arach-nological Society 9 174ndash188

Shear WA Selden PA Rolfe WDI Bonamo PM GriersonJD 1987 New terrestrial arachnids from the Devonian ofGilboa New York (Arachnida Trigonotarbida) AmericanMuseum Novitates 2901 1ndash74

Shultz JW 1987 The origin of the spinning apparatus in spi-ders Biological Reviews of the Cambridge PhilosophicalSociety 62 89ndash113

Shultz JW 1989 Morphology of locomotor appendages inArachnida evolutionary trends and phylogenetic implica-tions Zoological Journal of the Linnean Society 97 1ndash56

Shultz JW 1990 Evolutionary morphology and phylogeny ofArachnida Cladistics 6 1ndash31

Shultz JW 1991 Evolution of locomotion in Arachnida thehydraulic pressure pump of the giant whipscorpion Masti-goproctus giganteus (Uropygi) Journal of Morphology 21013ndash31

Shultz JW 1993 Muscular anatomy of the giant whipscor-pion Mastigoproctus giganteus (Arachnida Uropygi) and itsevolutionary significance Zoological Journal of the LinneanSociety 108 335ndash365

Shultz JW 1998 Phylogeny of Opiliones (Arachnida) anassessment of the lsquoCyphopalpatoresrsquo concept Journal ofArachnology 26 257ndash272

Shultz JW 1999 Muscular anatomy of a whipspider Phrynuslongipes (Amblypygi) and its evolutionary significance Zoo-logical Journal of the Linnean Society 126 81ndash116

Shultz JW 2000 Skeletomuscular anatomy of the harvest-man Leiobunum aldrichi (Weed 1893) (Arachnida Opil-iones Palpatores) and its evolutionary significanceZoological Journal of the Linnean Society 128 401ndash438

Shultz JW 2001 Gross muscular anatomy of Limuluspolyphemus (Chelicerata Xiphosura) and its bearing on evo-lution in the Arachnida Journal of Arachnology 29 283ndash303

Shultz JW 2007 Morphology of the prosomal exoskeleton ofScorpiones (Arachnida) and a new hypothesis for the evolu-tion of cuticular cephalic endoskeletons in arthropodsArthropod Structure amp Development 36 77ndash102

Shultz JW Pinto da Rocha R 2007 Morphology and func-tional anatomy In Pinto da Rocha R Machado G Giribet Geds The harvestmen the biology of Opiliones CambridgeMA Harvard University Press 14ndash61

Shultz JW Regier JC 2001 Phylogenetic analysis of Pha-langida (Arachnida Opiliones) using two nuclear protein-encoding genes supports monophyly of Palpatores Journalof Arachnology 29 189ndash200

Sitnikova LG 1978 The main evolutionary trends of the

244 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Acari and the problem of their monophyletism Entomologis-cheskoe Obozrenie 57 431ndash457

Snodgrass RE 1948 The feeding organs of Arachnidaincluding mites and ticks Smithsonian Miscellaneous Col-lections 110 (10) 1ndash94

Snodgrass RE 1952 A textbook of arthropod anatomy Ith-aca NY Comstock Publishing

Soslashrensen W 1879 Om Bygningen af Gonyleptiderne en Typeaf Arachnidernes Classe Naturhistorisk Tidsskrifter Series3 8 97ndash222

Stockwell SA 1989 Revision of the phylogeny and higherclassification of scorpions (Chelicerata) Unpublished doc-toral dissertation University of California at Berkeley

Stoslashrmer L 1944 On the relationships and phylogeny of fossiland Recent Arachnomorpha Skrifter utgitt av Det NorskeVidenskaps-Akademi i Oslo I 5 1ndash158

Stoslashrmer L 1955 Merostomata In Moore RC ed Treatise oninvertebrate paleontology Part P Arthropoda 2 LawrenceKS University of Kansas Press and Geological Society ofAmerica 4ndash41

Swofford DL 2002 PAUP Phylogenetic analysis using par-simony (and other methods) Version Beta 10 SunderlandMA Sinauer Associates

Talarico G Palacios-Vargas JG Fuentes Silva M AlbertiG 2005 First ultrastructural observations of the tarsal poreorgan of Pseudocellus pearsei and P boneti (ArachnidaRicinulei) Journal of Arachnology 33 604ndash612

Templeton AR 1983 Phylogenetic inference from restrictionendonuclease cleavage site maps with articular reference tothe evolution of humans and the apes Evolution 37 221ndash244

Tetlie OE Braddy SJ 2004 The first Silurian chasmataspidLoganamaraspis dunlopi gen et sp nov (Chelicerata Chas-mataspidida) from Lesmahagow Scotland and its implica-tions for eurypterid phylogeny Transactions of the RoyalSociety of Edinburgh Earth Sciences 94 227ndash234

Thomas RH Telford MJ 1999 Appendage development inembryos of the oribatid mite Archegozetes longisetosus(Acari Oribatei Trhypochthoniidae) Acta Zoologica (Stock-holm) 80 193ndash200

Tollerton VP Jr 1989 Morphology taxonomy and classifica-tion of the order Eurypterida Burmeister 1843 Journal ofPaleontology 63 642ndash657

Vachon M 1949 Ordre des Pseudoscorpions In Grasseacute P-Ped Traite de zoologie VI Paris Masson et Cie 436ndash481

Van der Hammen L 1979 Comparative studies in Chelicer-ata I The Cryptognomae (Ricinulei Architarbi and Anacti-notrichida) Zoologische Verhandelingen 174 1ndash62

Van der Hammen L 1982 Comparative studies in Chelicer-ata II Epimerata (Palpigradi and Actinotrichida) Zoolo-gische Verhandelingen 196 1ndash70

Van der Hammen L 1985 Functional morphology and affin-ities of extant Chelicerata in evolutionary perspectiveTransactions of the Royal Society of Edinburgh 76 137ndash146

Van der Hammen L 1989 An introduction to comparativearachnology Leiden SPB Academic Publishing

Vyas AB 1970 Studies in myology of the scorpion Heter-ometrus fulvipes Koch Unpublished PhD DissertationDepartment of Zoology Gujarat University India

Vyas AB 1974 The cheliceral muscles of the scorpion Heter-ometrus fulvipes Bulletin Southern California Academy ofSciences 73 9ndash14

Walter DE 1988 Predation and mycophagy by endeostigma-tid mites (Acariformes Prostigmata) Experimental andApplied Acarology 4 159ndash166

Walter DE Proctor HC 1998 Feeding behaviour and phy-logeny observations on early derivative Acari Experimentaland Applied Acarology 22 39ndash50

Walter DE Proctor HC 1999 Mites ecology evolution andbehaviour Sydney University of New South Wales

Waterston CD 1979 Problems of functional morphology andclassification in stylonuroid eurypterids (Chelierata Meros-tomata) with observations on the Scottish Silurian Sty-lonuroidea Transactions of the Royal Society of EdinburghEarth Science 70 251ndash322

Weygoldt P 1969 Biology of pseudoscorpions CambridgeMA Harvard University Press

Weygoldt P 1984 Lrsquoautotomie chez les Amblypyges RevueArachnologique 5 321ndash327

Weygoldt P 1996 Evolutionary morphology of whip spiderstowards a phylogenetic system (Chelicerata ArachnidaAmblypygi) Journal of Zoological Systematics and Olution-ary Research 34 185ndash202

Weygoldt P 1998 Evolution and systematics of the Chelicer-ata Experimental and Applied Acarology 22 63ndash79

Weygoldt P 2000 Whip spiders (Chelicerata Amblypygi)their biology morphology and systematics StenstrupDenmark Apollo Books

Weygoldt P Paulus HF 1979 Untersuchungen zur Morphol-ogie Taxonomie und Phylogenie der Chelicerata Zeitschriftfuumlr Zoologische Systematik und Evolutionsforschung 17 85ndash116 177ndash200

Wheeler WC Hayashi CY 1998 The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192

Winkler D 1957 Die Entwicklung der aumlusseren Koumlrperge-stalt bei den Phalangiidae (Opiliones) Mitteilungen aus demZoologischen Museum in Berlin 33 355ndash389

With CJ 1904 The Notostigmata a new suborder of AcariVidenskabelige Meddelelser fra Dansk Naturhistorisk Foren-ingi Kjobenhaus 1904 137ndash192

Yamasaki T Makioka T Saito J 1988 Morphology InSekiguchi K ed Biology of horseshoe crabs Tokyo ScienceHouse 69ndash132

Yoshikura M 1975 Comparative embryology and phylogeny ofArachnida Kumamoto Journal of Science Biology 12 71ndash142

APPENDIX

Each character used in this analysis is numbereddefined and cross-referenced with characters used infour previous studies that are abbreviated in bracketsspecifically WP = Weygoldt amp Paulus (1979)S = Shultz (1990) WH = Wheeler amp Hayashi (1998)GEWB = Giribet et al (2002) Descriptions includecorrections of errors discussions of controversies orambiguities and justifications for state assignmentsto lsquoproblemrsquo taxa

PHYLOGENY OF THE ARACHNID ORDERS 245

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

PROSOMA

1 Dorsal sclerite formed by fusion of the prosomal car-apace the dorsal portion of the first opisthosomalsomite and the dorsomedial (axial) portion of the sec-ond opisthosomal somite 0 absent 1 presentState 1 is present in extant xiphosurids and close fos-sil relatives (Scholl 1977 Anderson amp Selden 1997Shultz 2001) The fossil record suggests that thisstructure originated with the disappearance ofthe already reduced first opisthosomal tergite(= microtergite) and axial portion of the secondopisthosomal tergite (Anderson amp Selden 1997)2 Single dorsal sclerite covering entire dorsal surfaceof body no lines indicating original segmentation 0absent 1 presentState 1 occurs throughout Holothyrida (Acari) (Vander Hammen 1989) and in many Mesostigmata(Alberti amp Coons 1999) including Glyptholaspis (Vander Hammen 1989)3 Anterior end of dorsal prosoma with medianmarginal or submarginal pointed process 0 absent1 presentState 1 occurs throughout Schizomida (Protoschizo-mus Stenochrus Reddell amp Cokendolpher 1995) The-lyphonida (Rowland amp Cooke 1973 Shultz unpublobs Proschizomus Dunlop amp Horrocks 19951996)Amblypygi (Charinus Millot 1949b Phrynus Shultz1999) and Plesiosiro (Dunlop 1999) Citing a figure inRowland amp Cooke (1973) Dunlop amp Horrocks assumedthat the anterior process was limited to lsquohypoctonidrsquothelyphonids and united them with Proschizomus andSchizomida If valid this would make Thelyphonidaparaphyletic However the doubtful monophyly oflsquohypoctonidsrsquo (Haupt amp Song 1996) and widespreadpresence of a submarginal anterior process in Pedi-palpi weaken this proposal (Harvey 2002)4 Ophthalmic ridges pair of longitudinal crests inter-secting or passing near the region of the lateral eyes orcomparable region where lateral eyes are absent 0absent 1 presentState 1 is present throughout Xiphosura (Anderson ampSelden 1997) Similar structures are present in Plesi-osiro (Dunlop 1999) and lsquonon-hypoctonidrsquo Thely-phonida (Mastigoproctus Rowland amp Cooke 1973)Dunlop (2002a) described apparent ophthalmic ridgesin the chasmataspidid Octoberaspis but these werenot observed in Diploaspis (Dunlop et al 2001) eitherdue to absence or inadequate preservation of speci-mens nor in Chasmataspis (Dunlop et al 2004)5 Carapace with demarcations (eg grooves scleritesphragmata) between pro- meso- or metapeltidia 0absent 1 present [S 1 WH 48 GEWB 29]State 1 is present in extant Scorpiones Cyphoph-thalmi (Hansen amp Soslashrensen 1904) and many Palpa-tores (Opiliones) including Caddo and Leiobunum

and most Pseudoscorpiones (Chamberlin 1931) Giri-bet et al (2002) coded this character as absent inCyphophthalmi (Opiliones) but a transverseprocurved groove is present throughout the group(Hansen amp Soslashrensen 1904) and represents the meso-metapeltidial border as indicated by attachment ofthe dorsal endosternal suspensor of postoral somite VIin Chileogovea oedipus and Siro exilis (Shultz unpublobserv) Kjellesvig-Waering (1986) did not explicitlydescribe such grooves in fossil scorpions although heillustrated corresponding structures in PrearcturusArchaeophonus Archaeoctonus Palaeophonus andother Palaeozoic scorpions (Kjellesvig-Waering 1986figs 14A 21E 27D 59) (coded as uncertain here) butnot in Proscorpius (contra Giribet et al 2002) Dis-tinct carapacal sclerites are present throughout Palpi-gradi Schizomida and Solifugae (6)

The condition in Acari is controversial but is codedhere as State 1 Van der Hammen (1989) consideredthe tergal region of the leg-bearing somites (postoralsomites IIIndashVI) or podosoma to be reduced andreplaced by posterior migration of the aspidosoma (ietergal region assumed to be associated with append-ages of the gnathosoma) and anterior migration of theopisthosomal tergal region to form the hysterosomaThe dorsal proterosomal and hysterosomal elementsmeet at a transverse sejugal furrow that continues lat-erally and passes ventrally between the coxae of legs 2and 3 According to this scheme the prosomandashopisthosoma border is expressed as the disjugal fur-row which passes from the ventral posterior margin ofthe podosoma anterodorsally to join the dorsal part ofthe sejugal furrow Based on the arrangement of setaeand slit sensilla (= lyrifissures) Van der Hammen(1989) interpreted the region above coxae 3 and 4 (ieregion C) as a fusion of the dorsal parts of the first twoopisthosomal somites (= postoral somites VII and VIII)in early divergent Anactinotrichida (ie Opilioacari-formes) and Acariformes (eg Alycus) These interpre-tations have been followed by many acarologistsalthough its speculative aspects are widely acknowl-edged (Evans 1992 Alberti amp Coons 1999)

From the standpoint of a general arachnologist theVan der Hammen system seems unnecessarily compli-cated it appears to have been formulated to explainbroad morphological themes in Oribatida but was thenextrapolated to other mites A more conservativelsquoarachnologicalrsquo scheme adopted here equates the pro-podosoma with the pars cephalica (= propeltidium) ofthe arachnid prosoma and region C with the pars tho-racica (= mesopeltidium + metapetidium) In fact theearliest and most recent treatments of Opilioacari-formes (With 1904 Klompen 2000) reached similarconclusions This interpretation is also consistent withevidence from Acariformes For example the regiondorsal to the coxa of legs 3 and 4 in Alycus is as readily

246 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

explained by the persistence of primitive prosomal ter-gal elements as by the supposition that these disap-peared and were replaced by two opisthosomalelements Furthermore several diverse lineages ofbasally divergent Acariformes display features that donot fit readily within the GrandjeanndashVan der Ham-men system but seem to correspond to the cephalicathoracica division of the prosoma They have a sejugalfurrow that may have demarcated the cephalicatho-racica border and a postpodosomal furrow thatappears to represent the prosomandashopisthosoma junc-tion there is no disjugal furrow Examples includepalaeosomate (especially aphelacarid) and pediculoch-elid Oribatida alicorhagiid and mircopsammidlsquoEndeostigmatarsquo and paratydeid Prostigmata (Albertiamp Coons 1999 fig 42) This organization may be sym-plesiomorphic for Acariformes but could be a conver-gence brought about by selection for enhancedflexibility6 Carapace with distinct pro- meso- or metapeltidialsclerites 0 absent 1 present - inapplicable due toabsence of pro- meso- or metapeltidial demarcations(5)State 1 occurs throughout Palpigradi Schizomida andSolifugae It also appears sporadically within Opil-iones (eg Leiobunum flavum Shultz unpubl observ)and Pseudoscorpiones (eg Pseudochiridium) (Cham-berlin 1931)7 Sejugal furrow circumferential zone of body flexi-bility that passes between the coxae of legs 2 and 3 0absent 1 presentState 1 is probably synapomorphic for Acariformes asit occurs in most lsquoendeostigmatidsrsquo including Alycus(Van der Hammen 1989) and is variously developedthroughout Sarcoptiformes (Alberti amp Coons 1999) Itis probably the primitive condition for Prostigmatabut is either absent or weakly expressed in represen-tatives included here See 5 for alternative morpholog-ical interpretations of this body region in Acari8 Prosomal ozopores 0 absent 1 present [WH 46GEWB 12]State 1 occurs throughout Opiliones and Holothyrida(Van der Hammen 1989)9 Carapacal pleural doublure 0 absent 1 present[WP 12 S 2 WH 49 GEWB 27]A carapacal pleural doublure occurs throughoutXiphosura (Stoslashrmer 1944) Eurypterida (Clarke ampRuedemann 1912 Selden 1981) and Chasmataspid-ida (Dunlop et al 2001 2004 Dunlop 2002a) and isabsent in all known Arachnida10 Cardiac lobe a longitudinal axial elevation of thecarapace 0 absent 1 presentState 1 occurs throughout Xiphosura (Anderson ampSelden 1997) and Eurypterida (Clarke amp Ruedemann1912 Selden 1981) Carapaces are not preserved suf-ficiently to determine this condition in the chasmatas-

pidids Diploaspis (Dunlop et al 2001) andOctoberaspis (Dunlop 2002a) but an apparent cardiaclobe occurs in the Chasmataspis (Dunlop et al 2004)A cardiac lobe is absent in Arachnida11 Moveable cucullus 0 absent 1 present [WP 54GEWB 22]State 1 is a synapomorphy of Ricinulei (Selden 1992)12 Medial intercoxal lsquosternalrsquo region 0 all pedalcoxae separated medially 1 anterior pedal coxae abut-ting medially posterior coxae separated 2 anteriorpedal coxae separated medially posterior coxae abut-ting 3 all pedal coxae abutting medially 4 epimeracoxae undifferentiated medially from ventral body wall[S sim3 WH sim50 GEWB sim17]Prosomal sternites appear to be those portions of theventral body wall not occupied by coxae Despite earlyattempts to assess metamerism of the ventralprosoma (protosternum deutosternum tritosternumetc) (Boumlrner 1902a Millot 1949a) no apparent mor-phological features such as borders of sclerites ormuscle attachments reliably demarcate the ventralbody wall of one somite from that of an adjacentsomite Thus coding schemes that focus on describingmetameric components of the sternal region are prob-lematic Shultz (1990) attempted to code the shape ofthe entire intercoxal region and subsequent workershave adopted this approach All pedal coxae are sepa-rated medially in Plesiosiro (Dunlop 1999) Palpigradi(Roewer 1934) Araneae (Millot 1949c) Amblypygi(Weygoldt 2000) and many Acari including Opilioac-ariformes Holothryrida Mesostigmata Ixodida Aly-cus and many Prostigmata (Van der Hammen 1989Evans 1992 Alberti amp Coons 1999) The anteriorcoxae are separated in extant Thelyphonida Schizo-mida (Hansen amp Soslashrensen 1905 Millot 1949e) andRicinulei (Pittard amp Mitchell 1972 Legg 1976) Thecoxae are fused to the ventral body wall in Sarcopti-formes (Van der Hammen 1989 Evans 1992)

Interpreting this character is complicated for theanterior coxae of Scorpiones and Opiliones and allcoxae in Xiphosura and Eurypterida due to the pres-ence of coxapophyses endites andor gnathobaseswhich form components of the preoral chambers inthese groups Thus coxae may abut medially whenthese structures are considered part of the coxae butare separate medially if these structures are ignoredAs coxapophyses etc are here coded as separate char-acters (50ndash53) I have chosen to code this character asif these structures did not exist Consequently pha-langid Opiliones are coded as having all coxae sepa-rated (Pocock 1902 Hansen amp Soslashrensen 1904) Fossilscorpions also show substantial variation in this char-acter All pairs of pedal coxae in Palaeoscorpius abutmedially but also have gnathobases (Jeram 1998)and this condition is coded here as uncertain A ster-num separates all pairs of pedal coxae medially in

PHYLOGENY OF THE ARACHNID ORDERS

247

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Proscorpius

and

Stoermeroscorpio

(State 0) and

Pre-arcturus

approximates modern scorpions (State 1)(Kjellesvig-Waering 1986 Jeram 1998) The coxalbases of Eurypterida and Xiphosura are separated bya small sternite the endostoma (Selden 1981 Shultz2001)

13 Postoral sternapophysis (

=

tritosternum labium)a cuticular evagination of the ventral body wall poste-riorly adjacent to the palpal coxae forming the poste-rior border of the preoral chamber in some taxa ordisplaced posteriorly by fusion of the palpal coxae inothers 0 absent 1 present

[S

sim

10 WH

sim

56 GEWB19

=

33]State 1 is present throughout Palpigradi AraneaeAmblypygi (Snodgrass 1948) Schizomida (Hansen ampSoslashrensen 1905 Van der Hammen 1989) Thely-phonida (Van der Hammen 1989) extant Ricinulei(Pittard amp Mitchell 1972 Legg 1976) Opilioacari-formes and Mesostigmata (Van der Hammen 1989Alberti amp Coons 1999)

Allothyrus constrictus

is theonly holothyrid mite known to have sternapophysesAn apparent sternapophysis has been described in apalaeocharinid trigonotarbid (Dunlop 1994) Giribet

et al

(2002) miscoded Ixodida and Acariformes as hav-ing a lsquolabium (

=

tritosternum)rsquo The lophognath ofpseudoscorpions (Chamberlin 1931) and lsquolabiumrsquo ofSolifugae (Roewer 1934) also appear to be sternapo-physes that have been incorporated into the ros-trosoma (

32

) The labium of phalangid Opiliones(Pocock 1902 Hansen amp Soslashrensen 1904) is a ster-napophysis but is associated with leg 1 rather thepedipalp (Winkler 1957) A small anterior sclerite inthe fossil scorpions

Proscorpius

Waeringoscorpio

and

Labriscorpio

has been interpreted as a lsquolabiumrsquo homol-ogous with the sternapophyses of other arachnids(Weygoldt 1998) despite absence of any indication ofthe relationship of this sclerite to the mouth and pos-sible alternative interpretations (eg

16

) Giribet

et al

coded this sclerite as a sternapophysis but the char-acter is coded here as uncertain in fossil scorpions

14 Channels on the body surface linking openings ofcoxal organs to preoral chamber 0 absent 1 present

Van der Hammen (1989) described bilaterally pairedcuticular tracts (lsquotaenidiarsquo) connecting orifices of coxalorgans to the preoral chamber in some arachnids andor a ventromedian groove (

=

intercoxal or subcapitulargutter) that leads in turn to the preoral chamber inother taxa Taenidia and intercoxal gutters arepresent in Opilioacariformes (Van der Hammen1989) Holothyrida (Van der Hammen 1989) Mesos-tigmata (

Glyptholapis

Van der Hammen 1989)Ricinulei (Pittard amp Mitchell 1972 Van der Hammen1989) Araneae (

Heptathela

Aphonopelma

Hypochi-lus

Shultz unpubl observ also

Porrhothele

(Mygalo-morphae Dipluridae) Butt amp Taylor 1991

Segestria

(Araneomorphae Segestriidae) Van der Hammen

1989) Amblypygi (Van der Hammen 1989) Schizo-mida (

Stenochrus

Shultz unpubl observ unidenti-fied Hubbardiinae Van der Hammen 1989) andThelyphonida (

Mastigoproctus

Shultz unpublobserv also

Tetrabalius

Van der Hammen 1989) Theconduction of large volumes of fluid from orifices of thecoxal organs to the preoral chamber has been observedin the mygalomorph spider

Porrhothele

(Dipluridae)(Butt amp Taylor 1991) and

Mastigoproctus

(Thely-phonida) (Shultz unpubl observ) The podocephaliccanal (

15

) is present in Acariformes (Acari) (Lindquist1984) and appears to have evolved as an invaginatedsupracoxal channel (Grandjean 1938 G Alberti inEvans 1992) Van der Hammen (1982) could not dis-cern with certainty whether a channel exists in

Eukoenenia

(Palpigradi) but none has been reportedby previous authors and they have not been observedby the present author Palpigradi is coded as havingState 0

15 Podocephalic canal cuticular channel andorduct draining multiple glands and opening nearmouthparts 0 absent 1 present - inapplicable due toabsence of channel (14)

State 1 occurs throughout Acariformes (Acari)(Lindquist 1984)

16 Heavily sclerotized suboral sclerite serving inpart as basal pivot point for coxae of appendages ofpostoral somites IIndashIV (

=

arachnid palp and legs 1 and2) 0 absent 1 present

The sclerite forms the posterior wall of the true mouthand is heavily sclerotized in extant scorpions It is notvisible externally due to the tightly fitting coxae of theanterior prosomal appendages (Shultz 2007) It maycorrespond to the so-called labium (Weygoldt 1998) orlabrum (Kjellesvig-Waering 1986) of certain fossilscorpions (see

13

)

17 Genal angles 0 rounded 1 pointed

Anderson amp Selden (1997) originally coded a pointedgenal angle and genal spine as separate charactersbut they are combined here State 1 is present in manyXiphosura (Anderson amp Selden 1997) and Chas-mataspidida (Dunlop

et al

2001 2004 Dunlop2002a)

A

PPENDAGES

OF

POSTORAL

SOMITE

I

CHELICERAE

18 Cheliceral segmentation 0 three articles 1 twoarticles

[WP

sim

25

+

42 WH

sim

15 GEWB

sim

44]State 0 is present in extant Xiphosura Eurypterida(Clarke amp Ruedemann 1912 Selden 1981) Palpi-gradi (Roewer 1934) Opiliones (Hansen amp Soslashrensen1904) Scorpiones (Hjelle 1990) and throughout non-acariform Acari (Evans 1992) State 1 is presentthroughout known Trigonotarbida Araneae Ambly-pygi Thelyphonida Schizomida Pseudoscorpionesand Solifugae Acariform mites have two unambigu-

248 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ous cheliceral articles but may have a reduced thirdarticle (Van der Hammen 1989 Evans 1992) acari-form mites are thus coded as uncertain Chelicerae arenot known or their condition is unclear in several fos-sil taxa including Xiphosura Chasmataspidida (Dun-lop et al 2001 2004 Dunlop 2002a) Riciniulei(Selden 1992) Plesiosiro (Dunlop 1999) Proschizo-mus (Dunlop amp Horrocks 19951996) and the fossilscorpions Praearcturus and Palaeoscorpius (Kjelles-vig-Waering 1986) Kjellesvig-Waering (1986) consis-tently interpreted chelicerae of fossil scorpions ashaving four articles but this is probably a misinter-pretation of the basal article (see also Stockwell1989) The basal article in extant scorpions consists ofa distal collar of cuticle and a large proximal processand this arrangement would probably appear as twoarticles in compressed specimens In any event thereare at least three cheliceral articles in Stoermeroscor-pio and Proscorpius (Kjellesvig-Waering 1986)19 Terminal cheliceral joint 0 laterally placedbicondylar hinge 1 dorsally placed bicondylar hinge2 ventrally placed bicondylar hinge [WP 25 + 42 Ssim11 + 12 WH sim15 GEWB sim44]State 0 is present in Xiphosura (Shultz 2001)extant Scorpiones (Millot amp Vachon 1949) Opil-iones (Hansen amp Soslashrensen 1904) Palpigradi(Boumlrner 1904 Millot 1949d) Ricinulei (Pittard ampMitchell 1972 Legg 1976) and Eurypterida includ-ing Baltoeurypterus (Selden 1981) and Stylonurus(Clarke amp Ruedemann 1912) State 0 is present inthe fossil scorpions Proscorpius and Stoermeroscor-pio but the state of the chelicerae is unclear in theother fossil scorpions represented here (Kjellesvig-Waering 1986) State 1 is present in Trigonotarbida(Shear et al 1987) mygalomorph Araneae Ambly-pygi Schizomida and Theylphonida (Weygoldt ampPaulus 1979) State 2 is present in Pseudoscorpi-ones (Chamberlin 1931) and Solifugae (Roewer1934) and appears to be a groundplan feature for allmajor groups of mites (eg Van der Hammen 1989Alberti amp Coons 1999) The difference between theorthognathy of mygalomorph Araneae and lsquoplagiog-nathyrsquo mesothele (Liphistius Heptathela) and palae-ocribellate Araneae (Hypochilus) (Kraus amp Kraus1993) is small compared with the variation thatoccurs throughout arachnids and plagiognathy iscoded as State 1 here20 Chelicera articulating with carapace at anterolat-eral pivot 0 absent 1 present [S 13 WH 57 GEWB45]State 1 occurs in Solifugae (Roewer 1934) and Pseu-doscorpiones except Chthonioidea and Feaelloidea(Chamberlin 1931) Giribet et al (2002) coded thecharacter as present in Pseudoscorpiones21 Chelicera pivoting on dorsal protuberance of epis-tome 0 absent 1 present

State 1 appears to be a unique synapomorphy of Opil-iones (Shultz 2000)22 Extrinsic cheliceral muscle arising on carapaceand inserting on dorsal margin of nonbasal cheliceralarticle 0 absent 1 present (Note Characters 22 and23 refer to two different muscles)State 1 is known only in extant Scorpiones (LankesterBenham amp Beck 1885 Vyas 1970 1974 Shultz2007) but may occur in acariform mites if their cheli-cerae are composed of three rather than two articles(Evans 1992) (see 18) Acariformes are coded asuncertain23 Extrinsic cheliceral muscle arising on carapaceand inserting on ventral margin of nonbasal cheliceralarticle 0 absent 1 present [GEWB 252] (Note Char-acters 22 and 23 refer to two different muscles)State 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) somemesostigmatid Acari (Evans 1992) and representa-tive Opiliones particularly Leiobunum (Shultz 2000)Chileogovea and Gonyleptes (Shultz unpubl observ)The status of the character in other opilions is not yetknown The character was originally defined as anextrinsic muscle inserting on the second segment ofthree-segmented chelicerae (Shultz 2000 Giribetet al 2002) but it has been redefined here to make itapplicable to arachnids with two-segmented cheli-cerae24 Extrinsic cheliceral muscles attaching to epistome0 absent 1 presentState 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) but mayoccur in Prostigmata if the sigmoid piece to which thecheliceral protractors attach (Evans 1992) is a modi-fication of the epistome25 Lateral tergocheliceral muscle with three heads 0absent 1 present - coded only for extant tetrapulmo-nates and the palpigrade Eukoenenia homology isunclear in other taxa [GEWB 230]State 1 is present in Pedipalpi (Shultz 1999)26 Cheliceral silk glands 0 absent 1 present [GEWB47]State 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931) The silk gland of spider mites(eg Tetranychus) are associated with the pedipalps(Evans 1992)27 Cheliceral venom glands 0 absent 1 present [WP34 WH 40 GEWB 46]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)28 Cheliceral serrula interior and exterior 0 absent1 presentState 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931 Harvey 1992)29 Cheliceral lsquoflagellumrsquo in male 0 absent 1 present[GEWB 48]

PHYLOGENY OF THE ARACHNID ORDERS 249

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITE II LEG 1 IN NONARACHNIDS AND PALP IN ARACHNIDS

30 Palpal coxae fused ventromedially and formingposterior wall of preoral chamber 0 absent 1 present[WP 27 S 18 WH 17 = 62 GEWB 33 + 63]State 1 occurs in Schizomida Thelyphonida Acari andRicinulei31 Gnathosoma 0 absent 1 present - inapplicablepalpal coxae not fused (30) [WP 56 S 19 WH sim63GEWB sim36 + 37]The gnathosoma is a functional complex comprisingan epistome attached to the dorsal surface of mediallyfused palpal coxae (= infra- or subcapitulum) (see 30)the chelicerae and in some taxa a supracheliceral tec-tum These components are tightly integrated andmove in unison relative to the body The gnathosomahas traditionally been considered a feature of AcariHowever Van der Hammen (1989) argued that a gna-thosoma is present in Ricinulei and this interpreta-tion has received wide acceptance (eg Lindquist1984) However it is not clear that the relationship ofthe chelicerae and subcapitulum is as intimate inRicinulei as in Acari and the ricinuleid subcapitulumdoes not appear to be as moveable In fact the lsquosubca-pitulumrsquo in Ricinulei is similar in many respects to thecondition in Schizomida and Thelyphonida (Hansen ampSoslashrensen 1904) which has never been regarded as agnathosoma Consequently I have coded the conditionof Ricinulei as uncertain32 Rostrosoma long narrow subcylindrical epistomeprojecting anteriorly with base fixed to dorsal surfaceof palpal coxae bordered laterally by lobes projectingfrom palpal coxae ventral wall of preoral chamberformed by anterior element of prosoma (sternapophy-sis) 0 absent 1 present [S 15 WH 59 GEWB 20 + 36]State 1 is present in Pseudoscorpiones and Solifugae(Chamberlin 1931 Roewer 1934 Van der Hammen1989) Dunlop (2000) attempted to homologize com-ponents of the rostrosoma and gnathosoma (31) anargument accepted by Giribet et al (2002) Howeverfusion of the palpal coxae (30) apparently evolvedindependently in the two structures Coxal fusion inthe gnathosoma is complete ventrally and excludesthe suboral sternapophysis (13) while fusion isincomplete ventrally in the rostrosoma and incorpo-rates the sternapophysis Unlike the gnathosomathe rostrosoma is largely immobile and cheliceralmovement is not coupled with that of the palpalcoxae33 Rutellacorniculi hypertrophied setae modified asmouthparts located on the anterior processes of the pal-pal coxae 0 absent 1 present [GEWB 39]

State 1 occurs in the anactinotrichid groups Opilioac-ariformes Holothyrida and Mesostigmata but not Ixo-dida (Van der Hammen 1989) It also occurs inSarcoptiformes (Evans 1992) and certain endeostig-matids (eg Alycus Van der Hammen 1989) but not inProstigmata Comparable structures are unknown inother chelicerate groups34 Terminal segments specially modified in adultmale as a clasper used to engage the female 0 absent1 presentState 1 is known only in extant Xiphosura (YamasakiMakioka amp Saito 1988)35 Robust raptorial 0 absent 1 present [S 17 WH61 GEWB sim59 97]State 1 is present in Amblypygi Thelyphonida andSchizomida (Shear et al 1987) and in Laniatores(Opiliones)36 Orientation of robust raptorial appendage 0operating in subtransverse plane 1 operating in sub-vertical plane - inapplicable coded only for PedipalpiState 0 is present in extant Thelyphonida and mostAmblypygi and State 1 is present in Schizomida (Cok-endolpher amp Reddell 1992) and in the basally diver-gent amblypygid Paracharon caecus (Weygoldt 2000)The orientation of the pedipalps in the fossil Proschi-zomus is uncertain (Dunlop amp Horrocks 19951996)37 Extrinsic muscle attaching to epistome 0 absent1 presentState 1 occurs in Chileogovea and Siro (OpilionesCyphophthalmi) and extant Scorpiones (Vyas 1970Shultz 2007)38 Muscle originating and inserting within coxa 0absent 1 present [GEWB sim232]State 1 occurs is in Amblypygi (Phrynus Shultz 1999Phrynichus Boumlrner 1904) Thelyphonida (Mastigo-proctus Shultz 1993 Thelyphonus Boumlrner 1904) andSchizomida (Stenochrus Shultz unpubl observ)There are two intracoxal muscles in amblypygids andthelyphonids the one coded here and another that isserially homologous with a muscle in the pedal coxae(54)39 Tarsus andor tibia with venom glands 0 absent1 present [GEWB 64]State 1 is known only from extant Pseudoscorpi-ones except the superfamilies Chthonioidea andFeaelloidea (Chamberlin 1931 Harvey 1992) Giri-bet et al (2002) coded this character as present forPseudoscorpiones40 Tarsal grooming organ 0 absent 1 present[GEWB 98]State 1 is a unique synapomorphy of Amblypygi (DelleCave 1975 Weygoldt 2000)41 Scorpionoid chela a large well-developed chelaformed by tibia (manus + fixed finger) and tarsus(moveable finger) 0 absent 1 present [S 16 WH 60GEWB 62]

250 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs throughout Scorpiones andPseudoscorpiones42 Modified in male as copulatory organ 0 absent 1present [WP 35 WH 39 GEWB 67]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)43 Apotele 0 apparently absent not differentiatedexternally from penultimate article 1 present [GEWB69]The apotele exists as a structure (ie dactyl clawempodium) differentiated from the tibiotarsus tarsusor telotarsus and is typically associated with a pair ofantagonistic muscles The apotele occurs as a distinctstructure in Xiphosura (Snodgrass 1948) Eurypter-ida (Clarke amp Ruedemann 1912) Schizomida (Coken-dolpher amp Reddell 1992) Araneae (Foelix 1996)early divergent Amblypygi (Charinus Weygoldt 2000not Phrynus Shultz 1999) Anactinotrichida (CaminClark amp Bourdeau 1956 Lindquist 1984) Ricinulei(Pittard amp Mitchell 1972 Legg 1976) and Solifugae(Roewer 1934) Giribet et al (2002) coded Solifugae asnot having a palpal apotele but it is present as anadhesive organ (45) Dunlop (1999) reconstructedHaptopoda as having a palpal claw but notes in hisdescription that the distal ends were not preserved inthe known specimens44 Apotele (claw) position 0 terminal 1 subtermi-nal - inapplicable coded only for taxa with a distinctapotele (43)State 1 occurs in Parasitiformes (Acari) (Evans 1992)45 Terminal adhesive organ 0 absent 1 present -inapplicable due to lack of an apotele (43) [GEWB 66]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITES IIIndashVI LEGS 2ndash5 IN NON-ARACHNIDS AND LEGS 1ndash4 IN ARACHNIDS

46 Appendage III (= arachnid leg 1) extremely elon-gate antenniform 0 absent 1 present [WP sim24 + 32S 20 WH 14 GEWB sim84]State 1 occurs in Amblypygi Thelyphonida Schizo-mida (Shear et al 1987) and Opilioacariformes (Vander Hammen 1989) State 1 is approximated in Solif-ugae Palpigradi (Roewer 1934) and Haptopoda(Dunlop 1999) and it is coded as uncertain in thesetaxa47 Appendage V (= arachnid leg 3) of male specializedfor sperm transfer 0 absent 1 present [WP 55 GEWB94]State 1 is known only from extant Ricinulei48 Appendages V and VI (= arachnid legs 3 and 4)with femur shorter than patella and with principal siteof flexionextension at patella-tibia joint (lsquoapatellatersquocondition sensu Van der Hammen 1989) 0 absent 1present [S 25 GEWB 86]

State 1 occurs in Pseudoscorpiones and Solifugae(Shultz 1989)49 Appendage VI (= arachnid leg 4) with terminusmodified as flattened blade 0 absent 1 presentState 1 occurs in some Chasmataspidida (DiploapisDunlop et al 2001 Octoberaspis Dunlop 2002a)uncertain in Chasmataspis Dunlop et al 2004 State0 in Loganamaraspis Tetlie amp Braddy 2004) andEurypterida except stylonuroids (Clarke amp Ruede-mann 1912 Waterston 1979 Tollerton 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI COXA AND BODYndashCOXA JOINT

50 Appendage III (= arachnid leg 1) with coxapophy-sis forming floor or wall of preoral chamber 0 absent1 present [S sim14 WH sim58]State 1 occurs in Opiliones (Hansen amp Soslashrensen 1904)and higher scorpions (Jeram 1998) A lobe is presenton the coxa of leg 1 in the fossil scorpion Prearcturusand perhaps Stoermeroscorpio and is coded here as acoxapophysis Coxapophysis of leg 1 together with theepistome and coxapophysis of the palp form a uniquepreoral chamber the stomotheca Weygoldt (1998) andDunlop amp Braddy (2001) reject the homology of cox-apophyses and stomothecae in Opiliones and Scorpi-ones because these structures are not apparent infossils of those scorpions thought to have been aquaticHowever the coxapophyses in Opiliones are formedlargely from soft cuticular lsquolipsrsquo and similar structuresin fossil scorpions would probably not have beenpreserved51 Appendage IV (= arachnid leg 2) with coxapophy-sis 0 absent 1 present [S sim14 WH sim58 GEWB 75]State 1 occurs in extant Scorpiones and most Opil-iones except Dyspnoi (Shultz 1998)52 At least one pair of coxal gnathobases on append-ages III-VI (= arachnid legs 1ndash4) 0 absent 1 present[S sim23 WH sim66 GEWB 78]State 1 occurs in all appropriately preserved Euryp-terida including Baltoeurypterus (Selden 1981) andStylonurus (Clarke amp Ruedemann 1912) extantXiphosura on appendages IIndashVI and apparentlyProscorpius on appendage III (= arachnid leg 1)(Kjellesvig-Waering 1986 Jeram 1998 contra Giribetet al 2002) Fossilized appendages with gnathobaseshave been found in association with Chasmataspis(Chasmataspidida) suggesting that they may havehad State 1 (Dunlop et al 2004)53 Coxae of appendages IIIndashV (= arachnid legs 1ndash3)with jointed moveable endites 0 absent 1 present [Ssim23]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) and in adequately preserved eurypterids(Clarke amp Ruedemann 1912) (BaltoeurypterusSelden 1981)

PHYLOGENY OF THE ARACHNID ORDERS 251

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

54 Intracoxal muscle a muscle arising on anteriorwall of coxa and inserting on posterior wall 0 absent1 present [GEWB 237]State 1 occurs in Thelyphonida (MastigoproctusShultz 1993) and Amblypygi (Phryus Shultz 1999)It is unclear whether the muscle operating the move-able endite (53) in extant Xiphosura (Manton 1964Shultz 2001) is homologous with this muscle and iscoded here as uncertain It is not present in Steno-chrus (Schizomida) (Shultz unpubl observ)55 Coxae of appendages IIndashVI (= arachnid palp andlegs) with dorsal articulation with carapace 0 absent1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)56 Flabellum (exite) on coxa of appendage VI(= arachnid leg 4) 0 absent 1 present [GEWB sim110]State 1 occurs in extant Xiphosura (Yamasaki et al1988) An apparent exite occurs on appendages asso-ciated with Chasmataspis (Chasmataspidida) (Dunlopet al 2004)57 Insertion process of anteromedial tergo-coxal mus-cle 0 weakly developed 1 well to extremely well devel-oped [GEWB 238]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Boumlrner 1904 Shultz 1999)58 Musculi laterales enlarged lateral tergocoxalmuscle with attachment shifted from coxa to adjacentpleural membrane 0 absent 1 present [S 22 WH 65GEWB 242]State 1 occurs in Araneae and Thelyphonida (Shultz1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI TROCHANTER AND COXAndashTROCHANTER JOINT

59 Coxandashtrochanter joint with complex posteriorarticulation composed of two articulating sclerites 0absent 1 present [S 24 WH 67 GEWB 85]State 1 occurs in Araneae Amblypygi Thelyphonidaand Schizomida (Shultz 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI FEMUR AND TROCHANTERndashFEMUR JOINT

60 Depressor muscle (or homologue) of trochanter-femur joint 0 absent 1 present [WH 42 GEWB 220]State 0 appears to be a unique synapomorphy of Ara-neae (Liphistius Shultz 1989 Aphonopelma Ruh-land amp Rathmayer 1978 Hypochilus present studyother examples Clarke 1984 1986)61 Trochanterndashfemur joint with dorsal hinge or pivotoperated by flexor muscles only 0 absent 1 present[WH sim44 GEWB sim91]State 1 occurs in Palpigradi Other chelicerates have abicondylar articulation and are typically equipped

with antagonistic muscles (Van der Hammen 1985Shultz 1989) but see 60 for an exception62 Superior trochanterndashfemur muscle (or homologue)originating broadly in femur inserting on distal mar-gin of trochanter 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 19892001)63 Basifemurndashtelofemur joint of appendages III andIV (= arachnid legs 1 and 2) in adult 0 absent 1present [S 25 WH 68 GEWB 86]The basifemur and telofemur apparently develop aslsquosisterrsquo articles through division of a parental articlethe femur (eg opilioacariform endeostigmatidpalaeosomatid mites) (Coineau amp Van der Hammen1979 Evans 1992) A single animal can have all legswith divided or undivided femora or have some com-bination of the two typically with the more posteriorlegs having the divided femora Thus anterior legsoften tend to be paedomorphic with respect toposterior legs Most chelicerates have undivided fem-ora on some or all legs but muscles homologous withthose of the basifemurndashtelofemur joint may stilldevelop (66) (Shultz 1989) suggesting that the devel-opment of cuticular and muscular components of ajoint have a degree of developmental and evolutionaryindependence

Undivided femora occur in most extant euchelicer-ates including Xiphosura Palpigradi AraneaeAmblypygi Thelyphonida Schizomida ScorpionesOpiliones Pseudoscorpiones higher Sarcoptiformes(Archegozetes Van der Hammen 1989) many Prostig-mata (Tetranychus Evans 1992) and throughout Par-asitiformes Divided femora occur in all legs in manyearly divergent Acariformes including many Endeo-stigmata basally divergent Sarcoptiformes (egPalaeacarus) and Prostigmata (eg AllothrombiumMicrocaeculus) (Van der Hammen 1989 Evans1992) Only leg 4 has basi- and telofemora in Alycus(Van der Hammen 1989) Basi- and telofemora occuron legs 3 and 4 in Solifugae Ricinulei Opilioacari-formes and Eurypterida (Clarke amp Ruedemann 1912Selden 1981 Shultz 1989)

Giribet et al (2002) coded the lsquoannulusrsquo of certainTrigonotarbida (Shear et al 1987) as a basifemur buta similar non-muscularized annulus is present in theposterior legs of extant Thelyphonida In both casesthe condyles of the trochanterndashannulus and annulusndashfemur joints are aligned in such away that they per-form essentially identical movements and appear notto enhance either the range or degree of leg move-ment [This contrasts with the condition in the lasttwo pairs of legs in Baltoeurypterus where condyles ofthe apparent trochanter-basifemur and basifemurndashtelofemur joints have different arrangements (Selden1981)] The annulus in Thelyphonida appears to func-tion in maintaining constant volume of the tro-

252 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

chanterndashfemur joint by preventing ballooning of thearthrodial membrane thereby minimizing the effect ofinternal fluid pressure on movement64 Basifemurndashtelofemur joint of appendage V(= arachnid leg 3) in adult 0 absent 1 present See 6365 Basifemurndashtelofemur joint of appendage VI(= arachnid leg 4) in adult 0 absent 1 present See6366 Cuticular differentiation of basifemur-telofemurjoint absent but muscles present 0 absent 1 present- inapplicable due to presence of joint (63ndash65)State 1 occurs in appendages of postoral somites IIIndashIV (= arachnid legs 1ndash4) in Xiphosura AraneaeAmblypygi Thelyphonida Schizomida Scorpionesand Pseudoscorpiones (Shultz 1989)67 Circumfemoral ring 0 absent 1 presentThe femora of one or more pairs of legs in anactinot-richid Acari have a basal groove or ring associatedwith slit sensilla Acarologists term that portionproximal to the ring the lsquobasifemurrsquo and that portiondistal to the ring the lsquotelofemurrsquo and they refer tothe basifemur as the lsquosecond trochanterrsquo and thetelofemur as the lsquofemurrsquo Despite the confusingterminology the basifemur discussed in 63ndash65 is nothomologous with the lsquobasifemurrsquo of opilioacariformand parasitiform Acari as both a true basifemurndashtelofemur joint and a circumfemoral ring are presentin legs 3 and 4 of Opilioacariformes (Van der Ham-men 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI PATELLA AND FEMURndashPATELLA JOINT

68 Patella of appendage of postoral somite III(= arachnid leg 1) proportionally much longer thanthose of more posterior appendages 0 absent 1presentState 1 occurs in Thelyphonida Schizomida (Hansenamp Soslashrensen 1905 Shultz 1989) and Opilioacari-formes (Van der Hammen 1989)69 Femurndashpatella joint 0 monocondylar several axesof movement and multifunctional muscles 1 bicondy-lar hinge one axis of movement and antagonistic mus-cles 2 hinge one axis of movement flexor muscleswithout muscular antagonists [S sim26 WH sim69 GEWBsim87]State 0 occurs in Solifugae State 1 occurs in Phalang-ida (Opiliones) Scorpiones and PseudoscorpionesState 2 occurs in the remaining taxa (Shultz 1989)70 Patellar plagula 0 absent or with simple medianattachment 1 symmetrical Y-shaped with long prox-imal stem 2 symmetrical U-shaped (= arcuate scler-ite) 3 asymmetrical attaching to patella only atanterior margin [S 27 WH sim70 GEWB 234 = 243]State 1 occurs in extant Xiphosura (Shultz 1989)State 2 occurs in Araneae (Manton 1958 Ruhland amp

Rathmayer 1978 Clarke 1986 Shultz 1989) State 3occurs in Amblypygi Thelyphonida and Schizomida(Shultz 1989) Giribet et al (2002) included this char-acter twice

APPENDAGES OF POSTORAL SOMITES IIIndashVI TIBIA AND PATELLAndashTIBIA JOINT

71 Tibiae divided by one or more joints 0 absent 1presentState 1 occurs in Amblypygi (Weygoldt 2000)72 Patellandashtibia joint 0 monocondylar with or with-out CZY (73) 1 bicondylar hinge one axis of move-ment antagonistic muscles 2 hinge one axis ofmovement muscles without muscular antagonists [Ssim31 WH 74 GEWB sim92] (Shultz 1989 Selden Shearamp Bonamo 1991)73 Patellandashtibia joint with posterior compression zone(lsquoCZYrsquo) 0 absent 1 present - inapplicable due toabsence of monocondylar articulation [GEWB 99]State 1 is a unique synapomorphy of Araneae (Seldenet al 1991)74 Patellandashtibia joint of appendages IIIndashVI(= arachnid legs 1ndash4) with deep-set monocondylar pivotbordered by a pair of tibial processes to which extensormuscles attach 0 absent 1 present - inapplicabledue to absence of monocondylar joint (72)State 1 occurs in extant Xiphosura (Snodgrass 1952Shultz 1989) although a similar arrangement ispresent in the appendage III (= arachnid leg 1) in The-lyphonida (Shultz 1993)75 Patellandashtibia joint largely immobile specializedfor autotomy 0 absent 1 presentState 1 occurs throughout Amblypygi (Weygoldt 19842000)76 Anterior femurndashtibia or femoropatellandashtibia(transpatellar) muscle 0 absent 1 present [S sim30GEWB 222] (Shultz 1989)77 Proximal attachment of posterior femurndashtibia orfemoropatellarndashtibia (transpatellar) muscle 0 muscleabsent 1 dorsoposterior surface of femur andor pos-terior surface of patella 2 distal process of femur mus-cle acting as extensor of femur-patella joint [S sim28 WHsim71 + 72 GEWB 235 = 244 245] (Shultz 1989)78 Distal attachment of posterior femurndashtibia (trans-patellar) muscle 0 posterior 1 dorsal acting as exten-sor of patellandashtibia joint - inapplicable due to absenceof muscle (77) [S 29 GEWB 221] (Shultz 1989)79 Anterior patellandashtibia muscle 0 absent 1 present[S sim32 WH sim73 GEWB 246] (Shultz 1989)80 Posterior patellandashtibia muscle 0 absent 1 present[S 33 WH 76 GEWB 248] (Shultz 1989)81 Patellandashtibia joint spanned by elastic (lsquospringlikersquo)sclerite 0 absent 1 present [GEWB 107]State 1 is a unique synapomorphy of Solifugae (Sens-enig amp Shultz 2003)

PHYLOGENY OF THE ARACHNID ORDERS 253

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

APPENDAGES OF POSTORAL SOMITES IIIndashVI TARSUS AND TIBIAndashTARSUS JOINT

82 Tarsus divided into proximal basitarsus(= metatarsus) and distal telotarsus (= distitarsus orlsquotarsusrsquo) 0 absent 1 presentState 0 occurs in extant Xiphosura (Shultz 1989) andthroughout Acariformes (Lindquist 1984 Evans1992) in the anterior two pairs of legs in chthonioidpseudoscorpions and all legs in lsquomonosphyronidrsquo pseu-doscorpions including Feaelloidea and Cheliferoidea(Chamberlin 1931) A circumtarsal ring (83) associ-ated with slit sensilla in Parasitiformes (Acari) mayrepresent a joint between a telotarsus and basitarsus(Evans 1992)83 Circumtarsal ring 0 absent 1 presentState 1 is an apparent synapomorphy of Parasiti-formes (Evans 1992)84 Telotarsus in adult with two or more tarsomeres 0absent 1 present - inapplicable due to absence of telo-tarsus (82) [GEWB 71]State 1 occurs in Schizomida Thelyphonida Ambly-pygi Palpigradi Ricinulei Solifugae (Shultz 1989)Opilioacariformes (Van der Hammen 1989) and Ple-siosiro (Dunlop 1999)85 Three telotarsomeres on appendages of postoralsomites IV-VI (= arachnid legs 2ndash4) 0 absent 1present - inapplicable due to absence of telotarsus (82)or absence of telotarsomeres (84) [S 34 WH 77 GEWB102]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Shultz 1989)86 Tibiandashtarsus joint spanned by well-developed elas-tic (lsquospringlikersquo) sclerite 0 absent 1 present [GEWB107]State 1 occurs in Scorpiones phalangid Opiliones andSolifugae (Alexander 1967 Shultz 2000 Sensenig ampShultz 2003) Extant Xiphosura is coded for the lastprosomal appendage only (see 88)87 Appendage of postoral somite VI (= arachnid leg 4)with ring of large basally articulated spatulate pro-cesses at tibia-tarsus joint 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)88 Appendages postoral somites IIIndashV (= arachnidlegs 1ndash3) with tibiotarsus (ie tibia and tarsus not dif-ferentiated) 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 1989) butnot apparently in Weinbergina (Moore et al 2005) theonly synxiphosurid in which appendages are suffi-ciently preserved to determine the number ofpodomeres89 Ambulacrum peduncle-like extension of the tarsuswith internal condylophores and terminating distallywith apotele (eg claws) andor pulvillus 0 absent 1present

State 1 is present throughout Acari (Alberti amp Coons1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI APOTELE AND TARSUSndashAPOTELE JOINT

90 Apotele of appendage III (= arachnid leg 1) 0absent or not apparent 1 present [GEWB 101]State 0 occurs in Amblypygi Thelyphonida and Schi-zomida (Shultz 1999) However the apotele is proba-bly present but undifferentiated because muscles thatwould normally insert on the apotele terminate on theend of the lsquotarsusrsquo Dunlop (1999) reconstructed Hap-topoda as lacking an apotelic claw on leg 1 but noted inthe text that this was speculative Dunlop (2002b) hasreviewed apotelic diversity in Chelicerata91 Appendages of postoral somites IIIndashVI (= arachnidlegs) chelate with chela formed from tibiotarsus andapotele or tarsus and apotele 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988) The condition appears to occur in Chasmataspisbased on associated appendages (Dunlop et al 2004)but not in Diploaspis and Octoberaspis (Dunlop et al2001 Dunlop 2002a)92 Apotele with eversible or padlike empodium(= pulvillus) in adult 0 absent although empodialclaw may be present 1 present [S sim38 WH 81 GEWB95]State 1 occurs throughout Pseudoscorpiones (Cham-berlin 1931) Solifugae (Roewer 1934) Opilioacari-formes Holothyrida Ixodida and free-livingMesostigmata (Evans 1992) An eversible lsquopulvillusrsquo isprobably a primitive feature of Amblypygi (eg Char-inus Weygoldt 2000) but is absent in higher groups(eg Phrynus Weygoldt 2000)93 Inferior apotele muscle (= claw depressor) with tib-ial attachment 0 absent 1 present [S 35 WH 78GEWB 249]The character is coded for extant Xiphosura usingappendage of postoral somite VI (= arachnid leg 4) asother legs lack a differentiated tibia (88) State 1appears to occur throughout Arachnida (Shultz 1989)94 Inferior apotele muscle (= claw depressor) withpatellar attachment 0 absent 1 present [S 36 WH 79GEWB 250]State 0 occurs in extant Xiphosura Acari AmblypygiPalpigradi and Ricinulei State 1 occurs in all otherarachnids examined here (Shultz 1989)

OPISTHOSOMA

95 Number of opisthosomal somites in adult 0 five1 eight 2 nine 3 10 4 11 5 12 6 13 [WH sim33GEWB sim190]Chasmataspidids appear to have 13 opisthosomalsomites (Dunlop et al 2004) The synziphosurans

254 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Weinbergina (Moore et al 2005) and Limuloides(Anderson amp Selden 1997) appear to have ten opistho-somal somites Moore et al (2005) found no evidence ofan anterior microtergite in Weinbergina as proposedby Anderson amp Selden (1997) Xiphosurids appear tohave nine opisthosomal somites (Scholl 1977 Ander-son amp Selden 1997 Shultz 2001) Members of the fol-lowing orders have 12 opisthosomal somites Araneae(Millot 1949ac) Amblypygi (Weygoldt 2000) Thely-phonida and Schizomida (Hansen amp Soslashrensen 1905)and Pseudoscorpiones (Chamberlin 1931) Elevensomites are present in Palpigradi Solifugae (Roewer1934) and Trigonotarbida (Gilboarachne Shear et al1987 Palaeocharinus Fayers et al 2004) Plesiosiromay have 11 or 12 somites (Dunlop 1999) Justifica-tion for coding of more controversial taxa is providedbelow

Scorpiones Two principal hypotheses regarding thenumber of opisthosomal somites in scorpions havebeen advocated a 13-somite hypothesis derived fromembryological studies (Brauer 1895 Patten 1912Farley 1999 2005) and a 12-somite hypothesis basedon comparative anatomy of adults (Weygoldt amp Pau-lus 1979) The embryological interpretation is basedon the observation of pregenital genital and pectinalsomites (each with segmental ganglia and paired limbbuds) in early scorpion embryos followed by extremereduction or loss of the pregenital somite in laterembryos According to this view a missing pregenitalsomite should be added to the 12 apparent opisthoso-mal somites of post-embryonic scorpions to achieve afinal number of 13 The anatomy-based hypothesiswas introduced by Weygoldt amp Paulus (1979) whoadvocated a literal interpretation of post-embryonicsegmentation based on opisthosomal tergites Specifi-cally these authors argued that the last pair of dorsalendosternal suspensor muscles of non-scorpionarachnids especially Pedipalpi attach to the first(= pregenital) somite that this condition also occurs inscorpions and that there is no reason to invoke a miss-ing pregenital tergite They proposed that the pectinesbelong to the genital somite and that functional spe-cializations of the nervous system for pectinal functiongive the apparence of an extra neuromere duringembryonic development

I recently dissected the prosoma and anterioropisthosoma of the scorpions Centruroides Hadrurusand Heterometrus and focused on the composition ofthe muscular diaphragm (103) that separates thehaemocoelic compartments of the prosoma andopisthosoma (Shultz 2007) The diaphragm is com-posed of a metameric series of axial muscles fromthree somites the anterior somite corresponds to thelast prosomal somite and the posterior somite corre-sponds to the genital somite The middle elements

insert dorsally along a tranverse tendon attached tothe anterior margin of the first tergite These obser-vations are consistent with the embryological inter-pretation that a pregenital somite is present but itstergite is not expressed It appears likely that the thepregenital somite was compressed longitudinally dur-ing the evolution of the diaphragm I code scorpions ashaving 13 opisthosomal somites

Eurypterida The eurypterid opisthosoma is widelyassumed to have 12 somites (Clarke amp Ruedemann1912 Stoslashrmer 1944) However in a speculative paperon the evolutionary morphology of trilobites and che-licerates Raw (1957) proposed that both scorpions andeurypterids have 15 opisthosomal somites Rawassumed that scorpions have 13 apparent opisthoso-mal somites based on the transient pregenital somiteof scorpion embryology that scorpions and eurypteridsare close relatives and should have the same numberof somites and that olenellid trilobites and chelicer-ates always have somites in multiples of three Rawachieved 15 somites in scorpions by assuming the lastopisthosomal somite to be a diplosomite and that thetelson is a postanal somite even though there is noevidence for either of these proposals He attributedthese features to eurypterids as well He also notedthat the connection between the prosomal carapaceand first opisthosomal tergite in Eurypterida differedstructurally from the connection between adjacentopisthosomal tergites and regarded this as evidencefor a reduced pregenital tergite in the prosomandashopisthosoma junction Rawrsquos speculations were largelyforgotten until Dunlop amp Webster (1999) resurrectedthe proposal that eurypterids have a reduced opistho-somal tergite and therefore share a unique similaritywith scorpions Unfortunately Dunlop amp Bullocktreated Rawrsquos conjecture as if it were based on empir-ical evidence rather than an attempt to force euryp-terid morphology into a peculiar numerical system Inthe absence of convincing evidence to the contrary Ihave coded eurypterids as having 12 opisthosomalsomites

Ricinulei The opisthosoma consists of a membranouspedicel (97) bearing the gonopore ventrally Thefemale gonopore is bordered by an anterior plate and aposterior plate The remainder of the opisthosoma iscomposed of thick sclerites separated by less heavilysclerotized cuticle The first dorsal sclerite is short andfunctions as part of a prosomandashopisthosoma couplingmechanism (96) that may or may not be a specializedcomponent of the following tergite This is followed byfour tergites and sternites and a three-segmentedmetasoma (= lsquopygidiumrsquo) Millot (1945 1949f) rea-soned that the opisthosoma contains ten somites withsomites VIIndashIX incorporated into the pedicel XndashXIIIexpressed as tergites and sternites and the metasoma

PHYLOGENY OF THE ARACHNID ORDERS 255

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

comprising four somites He did not regard the dorsalcoupling sclerite as a separate tergite and his inter-pretation of four rather than three metasomal somiteshas been rejected Pittard amp Mitchell (1972) also pro-posed ten opisthosomal somites but achieved thisnumber by regarding the dorsal coupling sclerite as atergite of somite IX and by recognizing three metaso-mal somites Legg (1976) adopted the system proposedby Pittard amp Mitchell but did not regard the dorsalcoupling sclerite as separate from the following tergite(X) Van der Hammen (1979 1989) reconstructed 13somites The pedicel was regarded as having twosomites (VII + VIII) the following four sets of tergitesand sternites were considered diplosomites (IX + XXI + XII XIII + XIV XV + XVI) and the metasoma hadthree somites (XVIIndashXIX) This interpretation mayhave been biased by Van der Hammenrsquos attempt tounite Ricinulei with anactinotrichid Acari which healso regarded as having a primitive number of 13somites Selden (1992) described a fossil ricinuleidTerpsicroton that shows two pairs of depressions onthe three large premetasomal tergites which con-trasts with the single pair seen in extant species Thisobservation appears to corroborate the diplosomitehypothesis but the evidence does not indicate that thetergite anterior to these is a diplosomite

Dunlop (1996) attempted to homologize theprosomandashopisthosoma coupling mechanisms ofRicinulei and Trigonotarbida a goal that required anovel and rather forced interpretation of the dorsalsclerites He regarded the coupling sclerite as homol-ogous with the first opisthosomal tergite (VII) of trig-onotarbids and then followed Van der Hammenrsquosdiplosomite hypothesis to achieve 12 somites in totalDunloprsquos scheme differs from previous systems in sug-gesting that the pedicel does not contain the dorsalelements of the first and second somites and is in-consistent with Millotrsquos (1945) observation that thepre- and postgenital plates each have dorsoventralmuscles Here I code Ricinulei as having 12 somitesThere are three metasomal somites (XVIndashXVIII)three diplosomites (= six somites) (XndashXV) one cou-pling somite (IX) and two somites in the pedicel (iethe genital and pregenital somites) (VII VIII)

Opiliones Harvestmen have nine opisthosomalsomites and an anal operculum that is traditionallyregarded as the tergite of a tenth somite a tenth ster-nite is lacking (Hansen amp Soslashrensen 1904 Winkler1957) However the anal operculum appears to repre-sent a persistent embryonic telson (Moritz 1957) andis therefore likely to be a postsegmental structurecomparable with the stinger of scorpions flagella ofthelyphonids etc (Shultz 2000)

Acari The number of opisthosomal somites is probe-matic for most mite taxa due to uncertainty about the

location of the prosomandashopisthosoma boundry (5)extensive simplification or loss of metamericallyarranged sclerites and muscle attachments a paucityof developmental studies of engrailed expression andin Acariformes opisthosomal anamorphosis and het-erochronic modification of somite number (Evans1992) Ixodids appear to be the exception developmen-tal studies indicate five somites in the opisthosoma ofticks (Evans 1992)

Some mites retain external evidence of segmenta-tion ndash metameric patterns of furrows muscle attach-ments and slit sensilla ndash and with certainassumptions the number of opisthosomal somites canbe estimated Here it is assumed that the dorsal sur-faces of the last two prosomal somites are present inmites and retain their primitive association with legs3 and 4 (contra Van der Hammen 1989) (see 5 for jus-tification) Given this there appear to be 11 somites inthe opisthosoma of Opilioacariformes a conclusionalso reached by other workers (eg With 1904 Sitni-kova 1978 Klompen 2000) Similar reasoning sug-gests that Alycus has seven opisthosomal somites notnine as advocated by Van der Hammen (1989) Unfor-tunately external evidence is ambiguous for determin-ing the number of somites in other mite lineages andthese are coded as uncertain96 Prosoma-opisthosoma coupling mechanism 0absent 1 present [GEWB sim24]The posterior margin of the carapace and the anteriormargin of the first apparent opisthosomal tergite arespecialized as a coupling mechanism in Ricinulei andTrigonotarbida and Dunlop amp Horrocks (1996) pro-posed that these are synapomorphic for the twogroups However there is uncertainty about thehomology of the anterior opisthosomal somites in Trig-onotarbida and Ricinulei (95)97 Pedicel 0 absent 1 aranean type 2 ricinuleidtype [WP 30 S sim40 WH sim20 GEWB sim126]The body narrows at or near the prosomandashopisthosoma juncture in several arachnid lineages (ieSolifugae Palpigradi Amblypygi Araneae Ricinulei)and this lsquowaistrsquo has often been used as character at theinterordinal level (eg Pocock 1893) However thisinterpretation is rejected here for being subjective andfor uniting non-homologous conditions For exampleAraneae and Amblypygi are often grouped on the basisof a lsquopedicelrsquo yet it is a highly specialized structure inAraneae and its parts are not readily homologizedwith those of Amblypygi In contrast the condition inAmblypygi is a slightly narrower version of the highlymoveable prosomandashopisthosoma juncture in Uropygiwhich is not generally considered a lsquopedicelrsquo Thepedicel in Ricinulei is also unique a weakly sclero-tized stalk containing the genital opening (Cryptocel-lus Pittard amp Mitchell 1972 Ricinoides Legg 1976)See 95

256 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

98 Opisthosoma with three- segmented lsquobucklerrsquo 0absent 1 presentState 1 occurs throughout Chasmatapsidida (Dunlop2002a)99 Thoracetron consolidation of tergites of postgeni-tal somites 0 absent 1 present - inapplicable codedonly for Xiphosura (Anderson amp Selden 1997)100 Fusion of tergites of postoral somites VIII and IX(= opisthosomal somites 2 and 3) only 0 absent 1present [GEWB 146]State 1 appears to be a synapomorphy Trigonotarbida(Shear et al 1987) Ricinuleids also have diplotergites(101) but these appear to encompass a different com-bination of somites (see 95)101 Three diplotergites 0 absent presentState 1 is a unique synapomorphy of Ricinulei (Selden1992) See 95102 Paired opisthosomal defensive glands openingvia ducts on either side of the anus 0 absent 1present [S 46 WH 86 GEWB 122]State 1 occurs in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905)103 Muscular diaphragm separating prosomal andopisthosomal compartments formed by dorsoventralmuscles of postoral somites VIndashVIII and extrinsic mus-cles of leg 4 0 absent 1 presentState 1 is known only in extant scorpions (Lankesteret al 1885 Firstman 1973) The diaphragm is oftentreated as a single structure but recent anatomicalwork (Shultz 2007) has shown that it is a composite ofdorsal endosternal suspensors and extrinsic leg mus-cles (see 95) A diaphragm is also present in Solifugaebut it is located more posteriorly and does not appearto be homologous with that of scorpions (Roewer1934)104 Opisthosomal appendicular chondrites 0absent 1 presentThese cartilage-like columns of mesodermally derivedtissue are associated with each opisthosomal append-age including the chilaria in extant Xiphosura(Patten amp Hazen 1900 Yamasaki et al 1988 Fahr-enbach 1999 Shultz 2001)105 Paired appendages on ventral surface of postoralsomite VII (= opisthosomal somite 1) in adult 0absent 1 present [WP sim13 S 39 WH 82 GEWB 143]State 1 is known in Weinbergina (Moore et al 2005)and extant Xiphosura The bilobed structure of themetastoma in certain eurypterids as well as the cor-responding placement of chilaria in Xiphosura haveled some workers to regard the metastoma as fusedappendages of postoral somite VII (eg Stoslashrmer1955) This hypothesis is problematic given that (i) nospecial explanation is needed to account for bilaterallysymmetrical structures in bilaterally symmetricalorganisms (ii) incorporation of the ventral part of thefirst opisthosomal sternite as a functional element of

the prosoma is typical of arachnids and perhapsEuchelicerata generally and (iii) some workers regardthe metastoma as a sternite (eg Jeram 1998) Thusthe metastoma is coded here as uncertain Dunlop hasproposed that the sternum of scorpions is derivedfrom appendages citing the presence of transient limbbuds in scorpion development (Brauer 1895 Patten1912) but this is not compelling evidence for a persis-tent appendicular contribution to the sternite of theadult106 Megoperculum 0 absent 1 present [S sim41 + 42WH sim83 GEWB 162 + 163]State 1 occurs in Thelyphonida Schizomida Ambly-pygi Araneae (Shultz 1993 1999) Trigonotarbida(Shear et al 1987) and probably Haptopoda (Dunlop1999) and Palpigradi The megoperculum consists ofappendages of opisthosomal somite 2 (= genital somiteor postoral somite VIII) (Shultz 1993 1999 Popadicet al 1998) that have fused medially and displacedthe ventral body wall of the somite anteriorly (oftenrepresented by a small sternite to which dorsoventralmuscles of postoral somite VIII attach) The megoper-culum projects posteriorly to form the ventral surfaceof a pregenital chamber The dorsal surface of thechamber is formed by the ventral body wall of opistho-somal somite 3 (= postoral somite IX) which serves asthe ventral attachment of the paired dorsoventralmuscles of that somite The megoperculum bears book-lungs (where present) and sometimes paired gonopodsthat are probably derived from telopodites A similarbut less developed operculum is present on theopisthosomal somite 3 which may also bear booklungsand eversible vesicles (eg Amblypygi) (see 111) corre-sponding to the booklungs and gonopods of the mego-perculum (Shultz 1999)

I contend here that a megoperculum is present inPalpigradi and that this is particularly evident infemales Specifically a large unpaired lobe projectsposteriorly from the genital somite over the ventroan-terior surface of the first postgenital somite therebyforming a pregenital chamber The dorsoventral mus-cles of the genital somite attach ventrally near theanterior border of postoral somite VIII and those ofpostoral somite IX attach to the upper surface of thepregenital chamber (Rucker 1901 Boumlrner 1902bRoewer 1934 Millot 1949d Van der Hammen 1989)Weygoldt (1998) has questioned this interpretation ofPalpigradi107 Postgenital operculum or lsquosternitersquo 0 sclerotized1 not sclerotized - inapplicable coded only for Pan-tetrapulmonata (Platnick amp Gertsch 1976)State 1 is present in Trigonotarbida Amblypygi The-lyphonida Schizomida and Mesothelae (Araneae) butnot Opisthothelae (Araneae)108 Genital opening of female guarded by four plates(one pregenital one postgenital two laterogenitals)

PHYLOGENY OF THE ARACHNID ORDERS 257

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

genital opening of male guarded by two plates 0absent 1 presentThis an apparent synapomorphy of Holothyrida(Acari) (Van der Hammen 1989)109 Anterior margin of genital opening in male withglands secreting via fusules 0 absent 1 presentState 1 occurs in Araneae as epigastric or lsquoepiandrousrsquoglands (Marples 1967) and in Palpigradi (Condeacute1991b)110 Paired valve-like plates apparently formed fromcomponents of three somites covering triradiate genitalopening 0 absent 1 presentThis is an apparent synapomorphy of Acariformes(Evans 1992 Alberti amp Coons 1999)111 Eversible lsquoappendagesrsquo on the ventral surface ofpostgenital somites 0 absent 1 present [S sim43 WHsim84 GEWB sim127]Paired ventral lsquoappendagesrsquo operated in part byhaemolymph pressure occur in the form of ventralsacs in certain Amblypygi (postoral somite IX) (Wey-goldt 2000) (Charinus Millot 1949b not PhrynusShultz 1999) prokoenenian Palpigradi (postoralsomites XndashXII) (Rucker 1901 Condeacute 1991a) as spin-nerets in Araneae (XndashXI) (Shultz 1987) and as geni-tal papillae in many Acariformes (Alberti amp Coons1999) including Oribatida many Endeostigmata (Aly-cus Van der Hammen 1989) and Prostigmata (Allo-thrombium Saboori amp Kamali 2000 MicrocaeculusEvans 1992) Eversible vesicles similar to those ofamblypygids occur in fossils of the trigonotarbidPalaeocharinus (postoral somite IX) (Fayers et al2004) Two cuticular structures (lsquogenital verrucaersquo)occur anterior to the genital opening in Opilioacari-formes and each covers the opening to a thin-walledinvaginated sac (lsquogenital papillarsquo) of unknown function(Van der Hammen 1989) Van der Hammen proposedthat it is homologous to a genital papilla of Acari-formes but Alberti amp Coons (1999) have questionedthis interpretation on several grounds and refer tothese structures as pregenital capsules112 Opisthosomal silk glands and spinnerets derivedfrom appendages on postoral somites X and XI(= opisthosomal somites 4 and 5) 0 absent 1 present[WP 33 WH 41 GEWB 123 + 142]State 1 is a unique synapomorphy of Araneae (Plat-nick amp Gertsch 1976) although potentially homolo-gous glands and spigot-like setae occur on the ventralopisthosomal surface in Palpigradi (Millot 1943Condeacute 1991a)113 Opisthsosomal spinnerets location 0 near mid-dle of opisthosoma 1 near posterior end of opisthso-som ndash inapplicable coded only for Araneae (Platnickamp Gertsch 1976)114 Anterior medial lsquospinneretsrsquo 0 absent 1 present- inapplicable coded only for Araneae (Platnick ampGertsch 1976)

115 Opisthosomal tergites divided longitudinally intoone median and two lateral plates 0 absent 1 present[GEWB 145]State 1 occurs in non-curculioid Ricinulei (Selden1992) and many Trigonotarbida (Dunlop 1996) Thisfeature or something very like it (eg distinct opistho-somal trilobation) occurs sporadically in severaleuchelicerate groups including Chasmataspis (Chas-mataspidida) (Dunlop et al 2004) most fossil xipho-surans (Anderson amp Selden 1997) and a few derivedEurypterida (especially Mixopteroidea) (Tollerton1989)116 Number of metasomal somites 0 zero 1 two 2three 3 five 4 nine [S sim44 WH sim85 GEWB 128 144]The metasoma is a preanal region comprising multiplesomites that are substantially narrower than the pre-ceding somites or mesosoma Metasomal somitesoften lack pleural membranes and take the form ofsclerotized rings State 1 occurs in TrigonotarbidaState 2 is present in synziphosurid Xiphosura (Ander-son amp Selden 1997) Amblypyi Thelyphonida Schizo-mida and Ricinulei State 3 occurs in Scorpiones andEurypterida and State 4 occurs in Chasmatispidida

Some workers appear to consider a five-segmentedmetasoma as conclusive evidence for the monophylyof a Eurypterida + Scorpiones clade (eg Dunlop ampBraddy 2001) presumably because the character rep-resents a kind of tagmosis and should therefore begiven substantial phylogenetic weight However Wey-goldt (1998) has pointed out that a metasoma contain-ing three somites is probably part of the ground planof Euchelicerata The known diversity of the eucheli-cerate metasoma indicates that this feature can inincrease or decrease its segmental composition in evo-lution or be eliminated entirely without necessarilychanging the total number of opisthosomal somitesConsequently it is problematic to assume that thefive-segmented metasoma has any special immunityto homoplasy117 Postanal structure (telson) 0 absent or not obvi-ously developed 1 presentPostanal structures are considered non-segmentalbecause they are modifications of the embryonicregion posterior to the site of somite addition (embry-onic growth zone) A well-developed postanal struc-ture persists in adults in Xiphosura EurypteridaChasmataspidida Scorpiones Schizomida Thely-phonida and Palpigradi The anal operculum of Opil-iones is generally considered by morphologists to bethe tergite of the tenth opisthosomal or anal somite(eg Hansen amp Soslashrensen 1904) but embryological evi-dence (Moritz 1957) suggests that it is derived fromthe postproliferative zone and thus corresponds to thetelson of other chelicerates (see 95)118 Postanal structure shape 0 caudal spine 1aculeus 2 flagellum 3 anal operculum - inapplica-

258 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ble due to absence of postanal structure (117) [WP sim41S sim45 WH sim24 GEWB sim121 129 + 147]State 0 is present in Xiphosura Eurypterida andChasmataspidida State 1 occurs in Scorpiones State2 occurs in Palpigradi Thelyphonida and SchizomidaState 3 occurs in Opiliones (see 95 117)119 Specialized postanal flagellum in male (see 157ndash159) 0 absent 1 present - inapplicable coded onlyfor taxa with postanal flagellum (118) [GEWB 131]State 1 is a unique synapomorphy of Schizomida(Hansen amp Soslashrensen 1905 Cokendolpher amp Reddell1992 Reddell amp Cokendolpher 1995)

RESPIRATORY SYSTEM

120 Respiratory medium 0 water 1 airAmong the terminal taxa included only fossil scorpi-ons are problematic for this character They are codedhere as uncertain Based on a study of book lungmicrosculpture Scholtz amp Kamenz (2006) have arguedthat arachnids are primitively terrestrial and pulmo-nate (see also Firstman 1973) and have questionedwhether any fossil scorpions were aquatic No positionon this proposal is taken here121 Respiratory lamellae on opisthosomal somite 2(= genital somite postoral somite VIII) 0 absent 1present [WP sim37 S sim51 WH sim22 GWEB sim133ndash137]This character encompasses book gills (= lamellae thatfunction in water) and book lungs (= lamellae thatfunction in air) The distinction between book gills andbook lungs is accommodated here by 120 in combina-tion with 121ndash124 Respiratory lamellae are presenton the genital somite in Trigonotarbida AraneaeAmblypygi Thelyphonida and Schizomida Petrunk-evitch (1949) reconstructed Plesiosiro as having booklungs but Dunlop (1999) could not corroborate thisThis character was used by Dunlop amp Webster (1999)to propose that Xiphosura and Scorpiones are closelyrelated because they both lack respiratory lamellae onthe genital somite Dunlop amp Braddy (2001) alsoargued for the placement of Eurypterida with Xipho-sura and Scorpiones based in part on this characterAt least some eurypterids may have had respiratorylamellae (eg Manning amp Dunlop 1995) but the onlyevidence of their segmental distribution is derivedfrom one specimen (Braddy et al 1999)122 Respiratory lamellae on opisthosomal somite 3(= postoral somite IX) 0 absent 1 present [GEWB136]State 1 occurs in Amblypygi Thelyphonida and Xipho-sura It is a groundplan feature of Araneae (Platnick ampGertsch 1976) and is present in all representativetaxa included here Braddy et al (1999) proposed theexistence of respiratory lamellae on postoral somitesIXndashXII in Eurypterida based on evidence from onespecimen This character is coded as uncertain for the

Eurypterida included here Extant scorpions are codedhere as lacking respiratory lamellae on this somite(= pectinal somite) based on the interpretation of theopisthosomal segmentation discussed in 95123 Respiratory lamellae on opisthosomal somites 4ndash6 (= postoral somites XndashXII) 0 absent 1 presentState 1 is definitely present in extant Xiphosura andScorpiones and may have occurred in Eurypterida124 Respiratory lamellae on opisthosomal somite 7(= postoral somite XIII)State 1 occurs in extant Xiphosura and Scorpionesonly125 Kiemenplatten 0 absent 1 presentThese structures are located on the roof of opercularchambers in Eurypterida (Clarke amp Ruedemann1912) where they take the form of ventrally projectingcones with a distinct cuticular microsculputure (Man-ning amp Dunlop 1995) Dunlop amp Braddy (2001)inferred the existence of Kiemenplatten in all Palaeo-zoic scorpions based on a description and photos of onespecimen of Paraisobuthus duobicarinatus by Kjelles-vig-Waering (1986 pls 16ndash18) The plates depict darkcone-like rearward-pointing denticles distributedwithin a white amorphous material The denticlesappear to lack microsculpture even though the mag-nifications at which the photos were taken (times90ndashtimes330)are comparable with those illustrating the cones ofKiemenplatten (times170ndashtimes350 in Manning amp Dunlop1995 figs 1 2) Given the diversity of cuticular struc-tures present in the atria of booklungs book gills andtracheae the denticles appear to bear no special sim-ilarity to the cones of Kiemenplatten126 Tracheal system 0 absent 1 paired ventral stig-mata on postoral somite VIII (= opisthosomal somite2) 2 paired ventral stigmata on postoral somites IXand X 3 one pair of stigmata opening near legs 3 or 44 paired stigmata associated with chelicerae 5 fourpairs of stigmata on dorsal surface of opisthosoma -inapplicable aquatic (120) [WP 40 sim43 S sim52 + 5354 WH sim23 45 88 GEWB sim138 139]Firstman (1973) and Weygoldt amp Paulus (1979)hypothesized that tracheae are homologous in all tra-cheate arachnids except spiders and subsequentworkers have entertained this hypothesis by includinga character for the presenceabsence of tracheal sys-tems that ignores the diverse arrangement of stig-mata in arachnids (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) However thisapproach assumes that internal tracheal systems areconserved but that tracheal openings (stigmata)appear and disappear on different parts of the bodywith higher evolutionary frequency Here I assumethat stigmata are conserved in evolution and that dif-ferences in their anatomical placement reflect the evo-lution of new tracheae State 1 occurs throughoutOpiliones State 2 occurs throughout Solifugae

PHYLOGENY OF THE ARACHNID ORDERS 259

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

(Roewer 1934) and Pseudoscorpiones (Chamberlin1931) State 3 recognizes the possible homology of stig-mata in Ricinulei Opilioacariformes and Parasiti-formes (Acari) (Lindquist 1984 Van der Hammen1989) Tracheal systems are apparently absent in thegroundplan of Acariformes but Prostigmata typicallyhave tracheae associated with the chelicerae Tra-cheae derived from the posterior pair of book lungs arewidespread in araneomorph Araneae but the booklungs are retained and tracheae are absent in themost basally divergent groups (eg Hypochilus)

BOXndashTRUSS AXIAL MUSCLE SYSTEM (BTAMS)

127 Posterior oblique muscles of BTAMS of postoralsomites IndashVI 0 absent 1 present in one or moresomitesState 0 occurs in Xiphosura and state 1 occurs inPalpigradi Araneae Amblypygi Thelyphonida andSchizomida (Shultz 2001) The condition in AcariRicinulei Opiliones and Solifugae is not known128 Anterior oblique muscles of BTAMS posterior topostoral somite VI 0 absent 1 presentState 1 occurs in Xiphosura which is probably theprimitive condition based on comparison with otherarthropods (Shultz 2001) State 0 occurs in all arach-nids examined thus far129 Ventral attachments of posterior oblique musclesof opisthosomal BTAMS located in prosoma 0 absent1 presentState 1 occurs in extant Xiphosura (Shultz 2001)130 Endosternite fenestrate 0 absent 1 present [S 8WH 54 GEWB 35]State 1 occurs in Thelyphonida and hubbardiid Schi-zomida (Firstman 1973) The condition in protoschi-zomid Schizomida is not known131 Suboral suspensor a tendon that arises from theBTAMS and inserts on the ventral surface of the oralcavity via muscle 0 absent 1 presentState 1 occurs in Palpigradi (Eukoenenia Millot1943) Amblypygi (Shultz 1999) and Thelyphonida(Shultz 1993)132 Perineural vascular membrane in adult 0absent 1 present [WH 28 GEWB 155]Wheeler amp Hayashi (1998) coded extant Xiphosura asunknown although the primary source for this char-acter (Firstman 1973) stated that adult Limulus havea perineural vascular membrane Within Arachnidapresence of a perineural vascular membrane is appar-ently correlated with the presence of tracheae (126)133 Ventral endosternal suspensor attaching on coxaof anteriorly adjacent somite 0 absent 1 present[GEWB 239 253]State 1 occurs in Amblypygi (Shultz 1999) Thely-phonida (Shultz 1993) and Schizomida (Shultzunpubl observ)

134 Posteriormost postoral somite with a pair of dor-soventral muscles 0 VI 1 VII 2 VIII 3 XII 4 XIII5 XIV 6 XV 7 XVIDorsoventral muscles tend to run in a continuousmetameric series beginning in the prosoma Deter-mining the last tergite on which the muscle seriesends is substantially easier than counting especiallygiven frequent anatomical complexities near theprosomandashopisthosoma juncture The dorsoventralmuscle series ends on postoral somite VII inCyphophthalmi (eg Chileogovea) and Laniatores(Opiliones) (Shultz unpubl observ) on postoralsomite VII or VIII in Eupnoi (Leiobunum Shultz2000 unpubl observ) on postoral somite XII in Pal-pigradi (Roewer 1934) Solifugae (Bernard 1896Roewer 1934 Millot amp Vachon 1949) and Araneae(Liphistius Hypochilus Millot 1933 but segmenta-tion reinterpreted here following Shultz 1993 1999)on postoral somite XIII in hubbardiid Schizomida(Cokendolpher amp Reddell 1992) and Scorpiones(Lankester et al 1885) and on postoral somite XIVin extant Xiphosura (Shultz 2001) Thelyphonida(Shultz 1993) protoschizomid Schizomida (Cokendol-pher amp Reddell 1992) and Amblypygi (Shultz 1999)Dunlop (1999) reconstructed Plesiosiro as havingpaired tergal apodemes ending on postoral somite XVbut his figures indicate that the series ends on posto-ral somite XIV The interpretation of opisthosomalsegmentation in Ricinulei coded here (95) indicatesthat the series ends on postoral somite XV in Terpsi-croton and this is treated as the groundplan for theorder The posteriormost dorsoventral muscles inPseudoscorpiones occur on postoral somite XVI(Vachon 1949)

NERVOUS SYSTEM

135 Segmental ganglia 0 consolidated in prosoma1 one or more present in opisthosoma [WH 30 GEWB210 + 216]Adult neuromeres are exclusively prosomal in Ambly-pygi liphistiomorph and araneomorph Araneae Opil-iones Pseudoscorpiones Ricinulei and Acari (Millot1949a)136 Dorsal median eyes 0 absent 1 present [WP14 + 47 S 50 WH sim7 GEWB 1]Median eyes are a groundplan feature of Euchelicer-ata and are retained in extant Xiphosura Eurypteridaand occur in all arachnid orders except PalpigradiSchizomida Ricinulei and Pseudoscorpiones (Paulus1979) Median eyes are also absent in opilioacariformand parasitiform Acari (Lindquist 1984) and perhapsthe fossil whipscorpion Proschizomus (Dunlop amp Hor-rocks 19951996) They are unknown in synziphosu-ran Xiphosura Median eyes are present in a variety ofbasal acariform mites including Endeostigmata (but

260 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

not Alycus) Prostigmata (Microcaeculus) and certainoribatids (Palaeacarus) (Evans 1992 Alberti amp Coons1999) Giribet et al (2002) coded all representativeAcari as lacking median eyes It is not known whetherthe eyes of some cyphophthalmid opilions are medianor lateral evidence from tracheal branching inCyphophthalmus (Janczyk 1956) suggests that theyare median eyes (Shultz amp Pinto da Rocha 2007) andpresence of a tapetum in Stylocellus is consistent withlateral eyes (Shear 1993)137 Retinula cells of dorsal median eyes 0 organizedinto closed rhabdoms 1 organized into network ofrhabdomeres 2 disorganized - inapplicable due toabsence of median eyes (136) [GEWB sim3]State 0 is present in Scorpiones Thelyphonida andAmblypygi State 1 is present in Solifugae and Ara-neae state 2 is present in Xiphosura (Paulus 1979)The retinula cells of median eyes have been studied inseveral prostigmatid Acariformes and are organized ina network in some taxa and in an irregular pattern inothers (Alberti amp Coons 1999) Retinulae in phalangidOpiliones have state 0 proximally and state 2 distally(Schliwa 1979)138 Ventral median eyes 0 absent 1 presentState 1 occurs in early instars of extant Xiphosura(Paulus 1979)139 Lateral eyes 0 absent 1 presentLateral eyes are primitively present in Chelicerataand are absent in Palpigradi and Opiliones (Paulus1979) It is unclear whether the eyes of cyphoph-thalmid opilions are median or lateral (see 136)140 Arrangement and number of lateral eyes 0 com-pound many 1 five or more pairs (includes micro-lenses) 2 three primary pairs (excludes microlenses)3 two pairs 4 one pair - inapplicable due to absenceof lateral eyes (139) [WP sim13 + 18 + 38 + 44 + 52 Ssim49 WH sim10 GEWB sim4 + 5]True compound eyes are present in XiphosuraEurypterida Chasmataspidida and many PalaeozoicScorpiones Among the anactinotrichid Acari Opilioa-cariformes have two (Neocarus) or three (Siamacarus)pairs of lateral eyes (although at least one species ofSiamacarus lacks eyes) the allothryrid Holothyridaand many Ixodida have a single pair of lateral lensesand the remainder apparently lack eyes (Evans1992) Thelyphonida was coded as lsquo12rsquo to reflect fivepairs of lenses comprising three pairs of primarylenses and two pairs of small accessory lenses (notsimply three pairs as coded by Giribet et al 2002)Trigonotarbida is also coded as lsquo12rsquo to reflect threepairs of primary lenses and multiple small accessorylenses Dunlop (1999) illustrated Haptopoda as havingpaired triads of lateral eyes but noted that there wasactually no evidence of this in the fossils The lateraleyes of fossil ricinuleids have two pairs of lenses(Selden 1992) One pair of eyes or eyespots are

present in extant Ricinulei and many Schizomidaalthough five genera of hubbardiid Schizomida have apair of lenses (Reddell amp Cokendolpher 1995) State 1is the groundplan for extant Scorpiones and State 3occurs in the pseudoscorpions Chthonius FeaellaNeobisium (but State 4 in Chelifer) and Solifugae(Paulus 1979)141 Lateral eyes with closed rhabdoms 0 absent 1present - inapplicable due to absence of lateral eyes(139) [WP 21 WH 12 GEWB 6]Presence of closed rhabdoms is probably primitive forChelicerata and is retained in extant Xiphosura andScorpiones Retinula cells form a network of rhab-domeres in other extant chelicerates (Paulus 1979)but these networks can differ substantially in detailWeygoldt (1998) and other workers have given sub-stantial weight to the network character in unitingnon-scorpion arachnids but these authors seem not togrant comparable phylogenetic significance to theanalogous character of the median eyes shared bySolifugae and Araneae (137)142 Slit sensilla 0 absent 1 present [WP 19 S 47WH 11 GEWB 209]State 1 occurs in all arachnid orders except Palpigradi(Shultz 1990) The proposal that lsquoprimitiversquo slit sen-silla were present in Eurypterida (Dunlop amp Braddy1997) appears to be based on the over-interpretation ofa comparatively large notch that occurs in Baltoeu-rypterus at the terminus of the tibia (= podomere 7)(see also Edgecombe et al 2000 Giribet et al 2002)The neural construction of slit sensilla like function-ally similar campaniform sensilla of hexapods is sim-ilar to that of trichoid sensilla (Chapman 1998Klompen 2000) Thus these cuticular stress receptorsmay represent modified bases of sensory setae a viewsupported by the replacement of setae by slit sensilladuring post-embryonic development in some Opilioac-ariformes (Klompen 2000)143 Trichobothria 0 absent 1 present [GEWB 213]Trichobothria are present in extant Scorpiones(Jeram 1998) Pseudoscorpiones (Chamberlin 1931)some endeostigmatid (Alycus Van der Hammen 1989)and prostigmatid Acariformes and most Oribatida(Lindquist 1984) They occur on the ventral surface ofthe pedal femora in the opilioacariform Siamacarus(Leclerc 1989) They are also present in Araneae (Foe-lix 1996) Amblypygi (Weygoldt 2000) Schizomidaand Thelyphonida (Hansen amp Soslashrensen 1905) Theyare apparently absent in all non-arachnid chelicer-ates Solifugae Ricinulei Opiliones (Reissland ampGoumlrner 1985) and Parasitiformes (Acari) (Lindquist1984)144 Tibial trichobothria with 2-1-1-1 pattern onappendages IIIndashVI (= arachnid legs 1ndash4) 0 absent 1present - inapplicable due to absence of trichobothria(143) [S 48 WH 87 GEWB 88]

PHYLOGENY OF THE ARACHNID ORDERS 261

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs only in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905) Note that Shultz (1990)erroneously described this character as a 2-2-1-1 pat-tern and that this error was repeated by Giribet et al(2002) and described as 2-1-1 by Wheeler amp Hayashi(1998)145 Paired trichobothria on dorsal surface ofprosoma 0 absent 1 present - inapplicable due toabsence of trichobothria (143)State 1 is an apparent synapomorphy of Acariformesand is present in all acariforms included in this study(Alberti amp Coons 1999)146 Malleoli 0 absent 1 present [WP 45 GEWB 96]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)147 Pectines 0 absent 1 present [WP 20 WH 38GEWB 120]State 1 is a unique synapomorphy of Scorpiones148 Intercheliceral median organ 0 absent 1present [GEWB 212]State 1 is a unique synapomorphy of Palpigradi(Roewer 1934)149 Tarsal organ on appendage of postoral somite III(= arachnid leg 1) (= Hallerrsquos organ) sensilla containedwithin a cuticular depression on the superior surface ofthe tarsus of appendage III (= arachnid leg 1) 0absent 1 present [GEWB sim100]Klompen (2000) has noted that State 1 occurs in Opil-ioacariformes Parasitiformes (except Mesostigmata)and Ricinulei where it also occurs on leg 2 (Talaricoet al 2005) State 1 occurs on all legs in Araneae andperhaps Scorpiones (Foelix 1985)150 Tarsal organ on appendage of postoral somite IV(= arachnid leg 2) 0 absent 1 present (See 149)

REPRODUCTION

151 Gonads 0 primarily prosomal 1 primarilyopisthosomalState 0 is limited to Xiphosura (Sekiguchi 1988)State 1 occurs throughout Arachnida (Millot1949a)152 Ladder-like opisthosomal gonadsaccessoryglands (see 153) 0 absent 1 present [WH sim37 GEWBsim158]Giribet et al (2002) followed Wheeler amp Hayashi(1998) who followed Clarke (1979) in coding gonads asreticulate (Xiphosura) ladder-like (Scorpiones Thely-phonida Schizomida) or lsquosaccularrsquo (all remainingArachnida) However comparisons between Xipho-sura and Arachnida are problematic given that thexiphosuran gonads are primarily prosomal and thoseof arachnids are primarily opisthosomal (151) Fur-ther the reticulate pattern in Limulus (Xiphosura)but not other extant xiphosurans contains a distinctlyladder-like component The lsquosaccularrsquo state is probably

artificial as it encompasses a wide variety of pairedand unpaired structures153 Male gonads in two distinct parts one producingsperm and another (tubular gland) producing a holo-crine secretion similar to degenerate sperm 0 absent1 presentState 1 occurs in Thelyphonida Schizomida andAmblypygi although the holocrine material is pro-duced by ventral organs in Amblypygi and dorsalorgans in Thelyphonida and Schizomida (Alberti2005)154 Number of gonopores 0 two 1 oneExtant xiphosurans have two small genital openingson the base of the genital telopodite and all extantarachnids have a single opening The condition inEurypterida is not known Clarke amp Ruedemann(1912) located a pair of openings near the base of themedian organ (161) that are the outlets of the lsquohornorgansrsquo but it is unclear whether these are genitalducts or accessory structures Braddy amp Dunlop (1997)have developed numerous speculations about thesestructures and extended their arguments far beyondthe available evidence155 Genital opening (ie gonopore or gonostome)appearing to open in prosomal region (ie between legcoxae or anterior to posterior carapacal margin) 0absent 1 present [WP sim50 WH 26 GEWB 166]The genital opening in Euchelicerata is located onpostoral somite VIII (= opisthosomal somite 2) but ithas shifted anterior to the posterior margin of thecarapace or between the coxae of the last pair of legsin most Scorpiones (but not in PalaeoscorpiusKjellesvig-Waering 1986) and Opiliones The geni-tal opening occurs near or anterior to the last coxaein the opilioacariform and parasitiform Acari repre-sented here It is variable in Acariformes but islocated posterior to the coxae in all representativetaxa156 Ovipositor 0 absent 1 present [WP sim51 S sim60WH sim91 GEWB sim172]An ovipositor with a trilobed terminus is an apparentgroundplan character of Oribatida (Lindquist 1984Alberti amp Coons 1999) An ovipositor is also present inOpilioacariformes and Opiliones (Van der Hammen1989)157 Stalked spermatophore attached to substratum0 absent 1 present [S 57]State 1 occurs in Scorpiones PseudoscorpionesAmblypygi Thelyphonida and Schizomida The mech-anism of sperm transfer is unknown in PalpigradiOpilioacariformes and Holothryrida158 Male turns posterior end toward female duringmating behaviour 0 absent 1 presentState 1 occurs in Amblypygi (Weygoldt 2000) The-lyphonida and Schizomida (Weygoldt amp Paulus1979)

262 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

159 Female grasps male opisthosoma during matingbehaviour 0 absent 1 present [WP 29 S 58 WH 19GEWB 188]State 1 occurs in Thelyphonida and Schizomida (Wey-goldt amp Paulus 1979) Evolution of this behaviour wasprobably facilitated by 158160 Penis 0 absent 1 present [WP sim51 GEWB 167]State 1 is a unique synapomorphy of Opiliones A truepenis occurs in Phalangida A clearly homologousstructure is present in Cyphophthalmi (Opiliones) andapparently functions in depositing a spermatophore inthe femalersquos genital chamber (Karaman 2005) A truepenis may occur in some mites but it does not appearto be a groundplan feature of any major group (Evans1992) The lsquopenisrsquo in Oribatida is really a spermatop-ositor it functions in construction of a spermatophore(Alberti amp Coons 1999)161 Median organ 0 absent 1 presentState 1 occurs throughout Eurypterida (Clarke amp Rue-demann 1912) and has been observed in diploaspididChasmataspidida (Dunlop 2002a also Loganamaras-pis Tetlie amp Braddy 2004)

SPERM MORPHOLOGY

162 Nucleus with manchette of microtubules 0absent 1 present [S 54 WH 89 GEWB 193]State 1 occurs in Araneae Amblypygi Thely-phonida Schizomida Ricinulei and gonyleptid Lani-atores (Opiliones) (Alberti 1995 Giribet et al2002)163 Axoneme 0 absent 1 present [WP sim49 S sim55WH sim25 GEWB 195]State 1 occurs in Xiphosura Scorpiones Cyphoph-thalmi (Opiliones) Pseudoscorpiones RicinuleiAraneae Amblypygi Thelyphonida and Schizomida(Alberti 1995)164 Coiled axoneme 0 absent 1 present - inappli-cable due to absence of axoneme (163) [WP 22 S sim55WH 13 GEWB 196]State 1 occurs in Pseudoscorpiones Ricinulei Ara-neae Amblypygi Thelyphonida and Schizomida(Alberti 1995)165 Microtubule arrangement in axoneme 0 9 + 0 19 + 1 2 9 + 2 3 9 + 3 - inapplicable due to absence ofaxoneme (163) [WP sim26 S sim56 WH sim16 GEWB 198](Alberti 1995)166 Helical or corkscrew shaped nucleus 0 absent 1present [GEWB sim204]State 1 occurs in Araneae Amblypygi ThelyphonidaSchizomida and certain Scorpiones (eg Hadrurus)(Alberti 1995)167 Vacuolated-type sperm 0 absent 1 present[GEWB 205]State 1 is unique to Opilioacariformes and Parasiti-formes (Acari) (Alberti 1995)

168 Sperm aggregates 0 absent 1 present [GEWB206] (Alberti 1995)

DEVELOPMENT

169 Yolk in early embryo 0 concentrated (centroleci-thal or telolecithal) 1 evenly distributed (isolecithal)[WH sim34 GEWB sim191]Wheeler amp Hayashi (1998) and Giribet et al (2002)coded this character as lsquo0 isolecithal or telolecithal 1centrolecithalrsquo based on information presented byYoshikura (1975) Their coding appears to highlightseparate conditions found in Scorpiones (ie isoleci-thal and telolecithal) rather than a property intrinsicto the character itself170 Embryonic nutrition other than yolk 0 absent 1present [WP sim49]State 1 occurs in extant Scorpiones (except buthidsand chaerilids) and Pseudoscorpiones (Weygoldt1969)171 Embryological growth zone 0 initiating segmentaddition within prosoma 1 initiating segment addi-tion posterior to prosoma [GEWB sim192]Giribet et al (2002) followed Dunlop amp Webster (1999)in stating that only Xiphosura and Scorpiones havestate 0 among Chelicerata However the last prosomalsomite develops from the growth zone in Ixodida(Anderson 1973 Evans 1992) A similar processapparently occurs in the acariform mites Tyroglyphus(Sarcoptiformes Astigmata) (Yoshikura 1975) andArchegozetes (Sarcoptiformes Oribatida) (Thomas ampTelford 1999) but not apparently in Tetranychus(Prostigmata) (Dearden Donly amp Grbic 2002)172 Eggsembryos maintained in external attachedbrood sac secreted by genital glands 0 absent 1present [WP sim46 S 59 WH 90 GEWB 208 = 219]State 1 occurs throughout Amblypygi SchizomidaThelyphonida and Pseudoscorpions Giribet et al(2002) followed Shear et al (1987) in homologizingspinneret-derived silken egg sacs of spiders with gen-ital-gland-derived brood sacs of Pedipalpi However asthese structures are not homologous in secretory ori-gin construction or composition (Shultz 1987) theyare not regarded here as homologous Giribet et al(2002) miscoded the character as absent in Pseudo-scorpiones and without justification in PalpigradiFemale palpigrades have well-developed glands asso-ciated with postgenital somites and an array of setaefusules and other structures (112) (Boumlrner 1902b Mil-lot 1942 Condeacute 1991a) which are consistent withsome sort of brood care and Palpigradi should there-fore be coded as unknown for this character173 Embryonic and early postembryonic lsquolateralrsquo orClaparegravede organs associated with coxa of postoralsomite IV (= arachnid leg 2) 0 absent 1 present [WHsim31 GEWB sim211]

PHYLOGENY OF THE ARACHNID ORDERS 263

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bilaterally paired rounded protuberances locatedbetween coxae of appendages of postoral somites IIIand IV (arachnid legs 1 and 2) in prelarval and larvalinstars appear to be a primitive feature of Acariformes(Lindquist 1984 Evans 1992 Alberti amp Coons 1999)It is known to be derived embryologically from thecoxa of leg 2 in the oribatid Archegozetes (Thomas ampTelford 1999) Structures apparently homologouswith the acariform Claparegravede organ are present inembryonic Amblypygi (Weygoldt 2000) and Thely-phonida (Yoshikura 1975) and in embryonic and earlypost-embryonic stages of Solifugae (Roewer 1934)where they are termed lateral organs The so-calledlsquolateral organrsquo of embryonic Xiphosura (174) occurs onthe lateral surface of the carapace and does not appearto be homologous with the coxa-associated structuressharing their name In fact the xiphosuran lateralorgan develops even when the coxa of postoral somiteIV has been excised (Sekiguchi 1988) In contrastWheeler amp Hayashi (1998) and Giribet et al (2002)followed Yoshikura (1975) in regarding all lsquolateralorgansrsquo as homologous and in miscoding this characteras absent in Acariformes Van der Hammen (1989)suggested that the sternal verrucae of Opilioacari-formes are homologous with the Claparegravede organ butit differs in gross structure and ontogenetic timing174 Embryonic lateral organ associated with cara-pace 0 absent 1 present [WH sim31 GEWB sim211]State 1 is known only from embryonic stages of extantXiphosura (Yoshikura 1975) In contrast to previousinterpretations this feature does not appear to behomologous with the lsquolateralrsquo or Clarapegravede organ ofcertain arachnids (173)175 Live birth 0 absent 1 presentState 1 occurs in extant Scorpiones and has evolvedseveral times in a few mites (Evans 1992)176 Hexapodal larva and 1ndash3 nymphal stages 0absent 1 present [WP 53 S sim61 WH 27 GEWB 186]State 1 occurs in Acari and Ricinulei (Lindquist1984)177 Hexapodal prelarva 0 absent 1 present [S sim61WH 47]State 1 occurs in Opilioacariformes (NeocarusKlompen 2000) and many Acariformes (Evans 1992)

EXCRETORYOSMOREGULATORY SYSTEM

178 Malpighian tubules 0 absent 1 present [WP 17S 62 WH 9 GEWB 153]State 1 occurs in all major arachnid groups except Pal-pigradi Pseudoscorpiones Opiliones Oribatida (Sar-coptiformes) and Prostigmata Within Acari State 1occurs in all major anactinotrichid lineages (Opilioac-ariformes Holothyrida Mesostigmata Ixodida) andsimilar structures are present in certain Astigmata(Evans 1992 Alberti amp Coons 1999)

179 Dorsomedian excretory organ 0 absent 1presentThis is a specialized excretoryosmoregulatory organformed by the postventriculus and proctodeum inProstigmata (Alberti amp Coons 1999)180 Adult coxal organ opening on or near coxa ofappendage III (= arachnid leg 1) 0 absent 1 present[S 64 WH 93 GEWB 90]State 1 occurs throughout Acari (Evans 1992 Albertiamp Coons 1999) Ricinulei (Pittard amp Mitchell 1972Legg 1976) Palpigradi (Millot 1942) AraneaeAmblypygi Thelyphonida and Schizomida (Buxton1913 1917) There is evidence that the second lsquoozo-porersquo of gonyleptids (Opiliones Laniatores) (Hara ampGnaspini 2003) is a persistent opening to the coxalorgan associated with leg 1 (Soslashrensen 1879) whichalso occurs in embryonic Phalangium (Opiliones Eup-noi) (Moritz 1959)181 Adult coxal organ opening on or near coxa ofappendage V (= arachnid leg 3) 0 absent 1 present [S63 WH 92 GEWB 89]State 1 occurs in extant Xiphosura (Yamasaki et al1988) Scorpiones mygalomorph (Buxton 1913 1917)and mesothele Araneae (Liphistius J Millot in Bris-towe 1932) State 1 occurs in basal Amblypygi(Charon Charinus Buxton 1913 1917) but degener-ates prior to the adult stage in higher groups(Phrynus Weygoldt 2000) Adequately preservedeurypterids show a small submarginal opening on thecoxa of appendage V (Baltoeurypterus Selden 1981also Eurypterus Hughmilleria Clarke amp Ruedemann1912) and this is interpreted here as the opening to acoxal organ Giribet et al (2002) coded this characteras uncertain for Eurypterida182 Adult coxal organ opening on coxa of appendageII (= arachnid palp) 0 absent 1 present [GEWB 65]Buxton (1913 1917) noted that the coxal organ ofSolifugae opens on the prolateral surface of the palpalcoxa The proximal end of the duct leading to the ori-fice is associated with glands and apparently theglandular secretion andor coxal fluid serve as saliva(Alberti 1979) (see also 14 15)183 Genital papillae 0 absent 1 present - inappli-cable coded only for those taxa with eversible lsquoappend-agesrsquo (111)The genital papillae are associated with the genitalopening in postlarval Acariformes (Evans 1992Alberti amp Coons 1999) These structures are probablyserially homologous with the prosomal Claparegravedeorgan (173) and both structures probably function inwater and ion regulation

DIGESTIVE SYSTEM

184 Ingestion 0 solid food 1 primarily liquid foodwith or without preoral digestion [WP sim16 WH sim8GEWB sim218]

264 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Extant xiphosurans ingest solid food and members ofmost extant arachnid orders ingest fluids Opilionesand Acari are the principal exceptions The phalangidOpiliones ingest solids but Shultz (2000 unpublobserv) has shown that the precerebral pharyngealapparatus of Cyphophthalmi is very similar to that ofthe fluid-feeding Scorpiones and differs substantiallyfrom phalangid Opiliones Examination of gut con-tents of the cyphophthalmids Siro acaroides S exilisand Chileogovea oedipus have failed to reveal solidparticles typically found in the guts of phalangids(Shultz unpubl observ) the character is coded asunknown for Cyphophthalmi Within Acari consump-tion of particles has been documented in Opilioacari-formes (Neocarus Van der Hammen 1989unidentified Walter amp Proctor 1998) Holothyridanon-parasitic Sarcoptiformes (Evans 1992 Walter ampProctor 1998 Alberti amp Coons 1999) and many endeo-stigmatids including the nematophagous Alycusroseus (Walter 1988)185 Mouth 0 directed posteroventrally 1 directedanteroventrally [S 9 WH 55 GEWB 28]State 1 occurs throughout Arachnida (Shultz 1990)186 Oral cavity dilated by muscles arising from coxaeand constricted by large circular sphincter 0 absent1 presentState 1 occurs in extant Xiphosura (Manton 1964Scholl 1977 Shultz 2001)187 Palate plate 0 absent 1 present [GEWB 159]State 1 is a unique synapomorphy of Araneae (Dunlop1994)188 Lateral walls of epistome broadly fused to medialwalls of palpal coxae opposite sides connected by well-developed transverse epistomal muscle 0 absent 1presentState 1 occurs throughout Opiliones and Scorpiones(Shultz 2000 unpubl observ)189 Epistome with a pair of lateral arms that projectsposteriorly into the prosoma on either side of the phar-ynx 0 absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)190 Epistome with four pairs of suspensor musclesattaching to the carapace 0 absent 1 presentState 1 occurs in extant Scorpiones (Lankester et al1885 Vyas 1970 Shultz unpubl observ) and appar-ently in Solifugae (Roewer 1934)191 Intercheliceral epipharyngeal sclerite 0 absent1 present [GEWB 31]State 1 occurs in Palpigradi (Boumlrner 1904) Trigo-notarbida (Palaeocharinus Dunlop 1994) Araneae(Firstman 1954 Marples 1983) Amblypygi (Mil-lot 1949b Shultz 1999) Schizomida (Millot1949e) Thelyphonida (Millot 1949e Shultz 1993)and Phalangida (Opiliones) (Shultz 2000 unpublobserv)

192 Epipharyngeal sclerite large projecting posteri-orly 0 absent 1 present - inapplicable due to absenceof sclerite (191) [GEWB 32]State 1 occurs in Amblypygi (Millot 1949b Shultz1999) Schizomida (Millot 1949e) Thelyphonida (Mil-lot 1949e Shultz 1993) some Araneae (Heptathela ndashcoded as Liphistius ndash and Hypochilus Marples 1983not Aphonopelma Firstman 1954) and perhaps Trig-onotarbida (Palaeocharinus Dunlop 1994)193 Dorsal dilator muscle of precerebral pharynxattaching to intercheliceral septum or associatedepipharyngeal sclerite (191) 0 absent 1 presentState 1 occurs in Palpigradi (Millot 1943) Araneae(Marples 1983) Amblypygi Schizomida and Thely-phonida (Millot 1949e Shultz 1993 1999)Extremely fine muscles have been documented insome Scorpiones (Shultz 2007) and in Leiobunum(Opiliones) (Shultz 2000) Lankester et al (1885)reported a muscle in Limulus that would appear tocorrespond to this muscle but the muscle does notexist (Manton 1964 Shultz 2001)194 Dorsal dilator muscle of precerebral pharynxattaching to dorsal surface of prosoma 0 absent 1present [GEWB 228]State 1 occurs in Palpigradi (Roewer 1934) and mostAraneae (Palmgren 1978 Aphonopelma Firstman1973) but not in Heptathela (coded as Liphistius) orHypochilus (Marples 1983)195 Lateral dilator muscle of precerebral pharynxattaching to endosternite 0 absent 1 present [GEWBsim226]State 1 is a unique feature of Araneae (Marples 1983)196 Lateral dilator muscle of precerebral pharynxattaching to lateral surface of epistomal processes 0absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)197 Lateral dilator muscle of precerebral pharynxattaching to medial process of coxa of appendage II(= arachnid palp) 0 absent 1 present [GEWB sim226]State 1 occurs in Amblypygi Thelyphonida and Schi-zomida (Boumlrner 1904 Shultz 1993 1999)198 Dilator muscle of precerebral pharynx andorpreoral cavity attaching to ventral surface of prosoma0 absent 1 present [GEWB 227]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) Araneae (Marples 1983) Palpigradi(Millot 1943) and Solifugae (Roewer 1934)199 Postcerebral pharynx 0 absent 1 present [WPsim31 S sim5 WH sim21 GEWB sim154]The cuticle-lined foregut passes through the centralnervous system in many arthropods That portion pos-terior to the CNS is here termed the postcerebralpharynx regardless of the details of its morphology orfunctional specialization A substantial postcerebralpharynx is known in extant Xiphosura (Lankester

PHYLOGENY OF THE ARACHNID ORDERS 265

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

et al 1885 Manton 1964 Yamasaki et al 1988)Scorpiones (Centruroides Hadrurus HeterometrusShultz 2007 also Androctonus Abd el-Wahab 1952)Solifugae (Millot amp Vachon 1949) Araneae Ambly-pygi Thelyphonida and Schizomida (Millot 1949andasheShultz 1993) Weygoldt amp Paulus (1979) applied theterm lsquopostcerebral pharynxrsquo only to the lsquosucking stom-achrsquo of Araneae and Amblypygi but the sucking stom-ach is here regarded as a complex of three characters199ndash201

200 Dilator muscle of postcerebral pharynx attach-ing to endosternite 0 absent 1 present - inapplicabledue to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Lankester et al1885 Manton 1964 Shultz 2001) Scorpiones (Cen-truroides Hadrurus Heterometrus Shultz 2007 alsoAndroctonus Abd el-Wahab 1952) Araneae (First-

man 1954 Palmgren 1978) Amblypygi (Millot1949b Shultz 1999) and apparently the palpigradeProkoenenia wheeleri (Rucker 1901) but not inEukoenenia miriabilis (Millot 1943 contra Boumlrner1904)201 Dilator muscle of postcerebral pharynx attachingto dorsal surface of prosoma 0 absent 1 present -inapplicable due to absence of postcerebral pharynx(199)State 1 occurs throughout Araneae (Millot 1949cFirstman 1954 Palmgren 1978) and Amblypygi(Charinus Millot 1949b Phrynus Shultz 1999)202 Crop and gizzard 0 absent 1 present - inap-plicable due to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Sekiguchi 1988)but not in extant Arachnida

PHYLOGENY OF THE ARACHNID ORDERS

223

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Table 1

Data matrix comprising 59 euchelicerata genera and 202 binary and unordered multistate characters Fossil taxaare indicated by an asterisk Ambiguity codes A

=

[345] B

=

[12] C

=

[02] D

=

[34] E

=

[01]

Weinbergina

00010-011 00 - - 000 10-0001 0300000 0- -1210-0 -

Limuloides

00010-011 01 - - 300000 0- -1210-0 -

Euproops

00010-011 01 - - 200010 0- -1010-0 - 1

Limulus

10010-0011 0000-01000 0000-00000 -0010-0000 0010000000 011110001 0100010021 0001011011 000--01101 1000200010 000110-000 00-0010-0 01110-0110 0106112110 100--00000 0000000000 0010200000 0001000000 10-0010000 0-00000111 01

Tachypleus

10010-0011 0000-01000 0000-00000 -0010-0000 0010000000 011110001 0100010021 0001011011 000--01101 1000200010 000110-000 00-0010-0 01110-0110 0106112110 100--00000 0000000000 0010000000 0001000000 10-0010000 0-00000111 01

Baltoeurypterus

00000-011 00000000 0-0 -0000-0 0010000010 0110000 0011-0020 00000 100-00001 005000-0 00-00 0-0310-0 01- 1010 --00000 0 1 100 0-

Stylonurus

00000-011 00000000 - -000-0 001000000 010000 0011-000 000 10-0001 005000-0 00-0 0-0310-0 01- 1010 000 0 1

Chasmataspis

00000-011 01 - - 1 0 1601-0 0- -1410-0 00- 11 0 0

Diploaspis

0000-01 01 - - 1 0001 0601-0 0- -0410-0 00- 11 0

Octoberaspis

0010-01 01 - - 1 00 601-0 0- -0410-0 00- 11 0 1

Prokoenenia

0000110000 0010-00000 000000000 -0000-00 001000000 000000001 1000000020 00000110 0101000001 0014000-0 000001-010 10-021201 00000010 10310-00- -010000100 1001000 000--000 001 0001100000 1011000111 00

Eukoenenia

0000110000 0010-00000 0000000000 -0000-0000 001000000 0000000001 1000000020 000001011 0101000001 00104000-0 000001-010 00--021201 000000100 110310-00- -010000100 1001000 000--000 001 0001100000 101100010- --

Plesiosiro

00110-000 00 -0 -0- 0000 0000 0000 0 11000 0A00-0 00100 0--00--1 1 00 0 0

224

J W SHULTZ

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Palaeocharinus

00000-0000 00100110 0000 -000-00 00100000 0000000 0000002 02--0 100-0001 04100-1 001000 10--110--1 110000 101B1000 0 0 10 11

Gilboarachne

00000-0000 0000110 000 -000-00 000000 0000000 0002 02--0 100-0001 04100-1 001000 0--110--1 110000 101B 100 0 0

Liphistius

00000-0000 0011000110 0000001000 -0000-0000 0110000000 0000000110 00000100220010011011 0100-00001 00115010-0 0000010010 1101000--1 1100001000 00030110120110000011 1001000000 0111310000 1000000101 1001101000 1110100111 10

Aphonopelma

00000-0000 0011000110 0000001000 -0000-0000 0110000000 0000000110 00000100220010011011 0100-00001 00115010-0 0000011010 1110000--1 1100001000 00031110120110000011 1001000000 0111310000 1000000101 1001101000 1011100111 10

Hypochilus

00000-0000 0011000110 0000001000 -0000-0000 0110000000 0000000110 0000010022 0010011011 0100-00001 00115010-0 0000011010 1110000--1 1100001000 0003011012 0110000011 1001000000 0111310000 1000000101 001101000 1110100111 10

Charinus

00100-0000 0011000110 0000100000 -00010001 0010010000 000001011 0000010023 1000101011 0101100000 01105000-0 0000010000 10--020--1 1100001000 1015010012 01100000 1011001100 0111310000 1110000101 1001100000 1110001011 10

Phrynus

00100-0000 0011000110 0000100000 -000100101 0000-10000 0001001011 0000010023 1000101011 0101100000 00105000-0 0000010000 00--020--1 1100001000 1015010012 01100000 1011001100 0111310000 1110000101 00-1100000 1110001011 10

Stenochrus

0010110000 0211000110 000000001 0000110100 0010010000 000000111 00000101230000001011 0101100000 00115000-0 0100010000 00--021211 1000001001 01410-014111000000 1111001110 0111310000 110000101 00-1100000 111000100 00

Protoschizomus

0010110000 0211000110 000000001 00001100 0010010000 0000011 000000123 00000 0101100000 0015000-0 0100010000 00--021211 100000100 050-00- -111000000 111001110 0111310000 110000101 00-1100000 1000100 00

Mastigoproctus

00110-0000 0211000110 0000100001 0000100100 000--10000 0001001111 0000010123 0000001011 0101100000 00115000-0 0100010000 00--021201 1100001001 101511001B 0111000000 1111001110 0111310000 1110000101 00-1100000 1110001010 00

Proschizomus

00100-000 02001 1 0010 0010000 0000 100 05000-0 001000 0--02121 0 00 0 0

Terpsicroton

00000-000 10 - 0- 00000 0000 0110 0 0510-0 100- --120--1 0 50-13 00 0

Poliochera

00000-000 10 -1 0- 00000 0000 0110 0 0510-0 100- --120--1 0 50-13 00 0

Table 1

Continued

PHYLOGENY OF THE ARACHNID ORDERS

225

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Cryptocellus

00000-0000 1211000100 00-00001 0000-000 0010001000 000000001 0001100020 02--000-11 0101000001 00105120-0 100000-000 00--120--1 000003 1500-014 10-000011 10010000 011120000 0010101 00-1100000 0-0000000- --

Ricinoides

00000-0000 1211000100 00-00001 0000-000 0010001000 000000001 0001100020 02--000-11 0101000001 00105120-0 100000-000 00--120--1 000003 1500-014 10-000011 10010000 011120000 0010101 00-1100000 0-0000000- --

Neocarus

00001-0000 0011000020 0000-00001 10100-000 0010010000 00000001 000111120 02--00 0101000011 0114000-0 000000-000 00--000--1 000005 00-013 010--00010 1001110 000--01000 011101 00-0100000 0-0000000- --

Siamacarus

00001-0000 0011000020 0000-00001 1100-000 0010010000 0000000 0011112 02--00 10100011 014000-0 000000-000 00--000--1 000005 00-012 110000010 101110 00--10 01101 -0100000 0-0 --

Australothyrus

01000-0100 0001000020 0000-00001 10100-000 0011000000 0000000 00001020 02--00 010-00011 01000-0 000000-100 00--000--1 000003 00-014 10--00010 1001100 000--01000 0101 00-1100000 0-000 --

Allothyrus

01000-0100 0011000020 0000-00001 10100-000 0011000000 0000000 00001020 02--00 010-00011 01000-0 000000-100 00--000--1 000003 00-00- -10--00010 101100 00--10 0101 00-1100000 0-0 --

Glyptholaspis01000-0000 0011000020 000-00001 10100-000 0011000000 000000001 0000001020 02--011011 010-00011 0110000-0 000000-000 00--000--1 00000300 00-00- -10--00000 1001100000 000--01000 000010101 00-1100000 0-0000000- --

Amblyomma00000-0000 0000-00020 0000-00001 10000-000 000--00000 000000001 0000001020 02--011011 010-00011 01100000-0 000000-000 00--000--1 00000300 100-014 010--00010 1001100000 000--01000 0000010101 00-1100000 0-0000000- --

Argas00000-0000 0000-00020 0000-00001 10000-000 000--00000 000000001 0000001020 02--011011 010-00011 00100000-0 000000-000 00--000--1 00000300 100-014 010--00010 101100000 00--10 0000010101 00-1100000 0-00000- --

Alycus000001000 000110020 000-00001 10100-000 000--00000 000000001 000010020 02--0011 000--00011 00101000-0 000000-001 10--000--1 000000 00-013 110100000 10100 00 0111 0010100000 0-000- --

Allothrombium0000000 000110020 00-00001 10000-000 000--00000 000000001 00111-0020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000004 00-0130110100000 1001001000 000--00000 010011011 0011100000 0-0000000- --

Microcaeculus00000000 000110020 00-00001 10000-000 000--00000 000000001 00111-0020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000004 0120130110100000 10010010 00--0000 01001011 001100000 0-0000000- --

Palaeacarus000011000 040110020 000-00001 10100-000 000--00000 000000001 00111-0020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000000 010 110100000 101011000 00--00 010011001 0010100000 0-00000- --

Table 1 Continued

226 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Archegozetes000011000 040110020 000-00001 10100-000 000--00000 000000001 0000000020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000000 00-00- -110100000 1001011000 000--00000 0010011001 0010100000 0-0000000- --

Cyphophthalmus0000100100 0100-00000 100-00000 -0000-1000 0010000001 1000000001 0000000020 01--010-11 0100-00001 00112000-0 000000-000 00--0013-1 000001000 010100 10--00000 1001110001 001100010 0000000 10-100110 0-0001000- --

Chileogovea0000100100 0100-00000 1010-00000 -0000-1000 0010000001 1000000001 0000000020 01--010-11 0100-00001 00112000-0 000000-000 00--0013-1 000001000 010100 10--00000 1001110001 0000 0000000 10-100110 0-0001000- --

Caddo0000100100 0000-00000 100-00000 -0000-000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 01010100--10--00000 10-1110001 000000 0000000 10-0100110 10001000- --

Leiobunum0000100100 0000-00000 1010-00000 -0000-0000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 010B01C00- -10--00000 1001110001 000--00000 1000000000 10-0100110 101001000- --

Sclerobunus0000000100 0000-00000 100-00000 -0001-000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 010001C00- -10--00000 1001110001 00--00000 1000000000 10-0100110 10001000- --

Gonyleptes0000000100 0000-00000 1010-00000 -0001-000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 010001C00- -10--00000 1001110001 010--00000 100000000 10-0100110 10001000- --

Centruroides0000100000 0100-10000 0111-00000 -0000-1000 100--00001 1000000001 0000010010 01--012110 0100-10001 00116000-0 001000-000 00--0311-1 0011001000 0005110011 1110001011 1101101000 0010110100 0000100100 10-1100111 0-10010011 00

Hadrurus0000100000 0100-10000 0111-00000 -0000-1000 100--00001 1000000001 0000010010 01--012110 0100-10001 00116000-0 001000-000 00--0311-1 0011001000 0005110011 1110001011 1101101000 0010110111 0000100100 10-1100111 0-10010011 00

Heterometrus0000100000 0100-10000 0111-00000 -0000-1000 100--00001 1000000001 0000010010 01--012110 0100-10001 00116000-0 001000-000 00--0311-1 0011001000 00051100111110001011 1101101000 001020111 0000100100 10-1100111 0-10010011 00

Prearcturus000010000 010 -0 -000- 100--001 000 000-0 00- --0311- 0 0

Palaeoscorpius00000000 0D0 -0 -000- 100--0000 00 00000 0 10-0001 06000-0 000-00 --0311- 1 01 0 0

Stoermeroscorpio00000000 00000 -0 -000- 100--0000 000 00000 0 10-0001 06000-0 000-00 --0311- 110 01 1 0

Table 1 Continued

PHYLOGENY OF THE ARACHNID ORDERS 227

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

the Ordovician to the Permian (Tollerton 1989) Phy-logenetic relationships among the eurypterids havenot been rigorously addressed (but see Beall amp Laban-deira 1990) However members of Stylonuroideaappear to be plesiomorphic in retaining relativelyunspecialized legs and were represented by Stylonu-rus (eg Clarke amp Ruedemann 1912 Waterston1979) In the remaining eurypterids the last leg ispaddle-like (Tollerton 1989) The non-stylonuroideurypterids were represented in the matrix by thewell-preserved and well-studied Baltoeurypterus tet-ragonophthalmus especially as described by Selden(1981)

Chasmataspidida Members of this fossil aquaticgroup (OrdovicianndashDevonian six genera) have a four-segmented mesosoma and a nine-segmented meta-soma (Dunlop Poschmann amp Anderson 2001 DunlopAnderson amp Braddy 2004 Dunlop 2002a) Three gen-era were included Chasmataspis from the monotypicfamily Chasmataspididae as well as Diploaspis andOctoberaspis from Diploaspididae

Haptopoda This is a fossil terrestrial group (Carbon-iferous) containing one known species Plesiosiromadeleyi The known specimens have been re-examined by Dunlop (1999)

Palpigradi The extant palpigrades (two families sixgenera ~50 spp) (Harvey 2002) are a poorly studiedgroup of small-bodied terrestrial (some semi-aquatic)arachnids The two most well-studied generaEukoenenia (Eukoeneniidae) especially E mirabilisand Prokoenenia (Prokoeneniidae) especiallyP wheeleri were included in the matrix The oneknown fossil species Paleokoenenia mordax is rela-tively recent (Pliocene) and to the extent known isfairly similar to extant palpigrades (Rowland amp Sis-som 1980) It was not included in the analysis

Trigonotarbida This is a fossil terrestrial order(Upper Silurian to Lower Permian ~ten families ~50spp) The genera coded here were drawn from twoLower Devonian Lagerstatten Gilboarachne from theGilboa Formation (New York) (Shear et al 1987) and

Proscorpius00000000 00000 -0 -000- 100--0000 010 00000 0 10-0001 06000-0 000-00 --0311- 110 01 1 0

Chthonius0000100000 0310-00120 0000-10100 -1000-0000 100--00100 0000000001 0000010010 01--012100 0E00-00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Neobisium0000100000 0310-00121 0000-10100 -1000-0010 100--00100 0000000001 0000010010 01--012100 0100-00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Feaella0000100000 0310-00120 0000-10100 -1000-0000 100--00100 0000000001 0000010010 01--012100 000--00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Chelifer0000100000 0310-00121 0000-10100 -1000-0010 100--00100 0000000001 0000010010 01--012100 000--00001 01115000-0 000000-000 00--000--1 0000021000 01700-014 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Eremocosta0000110000 0310-00121 0000-00010 -1000-0000 001010100 0000000001 00011-0000 02--010-10 1101010001 01114000-0 000000-000 00--000--1 00000200 13111013 010--10000 1001000000 000--00100 1010000100 01-1100001 0-00000110 00

Galeodes0000110000 0310-00121 0000-00010 -1000-0000 001010100 0000000001 00011-0000 02--010-10 1101010001 01114000-0 000000-000 00--000--1 00000200 13111013 010--10000 1001000000 000--00100 1010000100 011100001 0-00000110 00

Table 1 Continued

228 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Palaeocharinus from the Rhynie Chert (Scotland)(Dunlop 1994 Fayers Dunlop amp Trewin 2004)

Araneae (spiders) This is a very large order (110 fam-ilies ~3600 genera ~39 000 extant spp) (Platnick2005) represented here by one genus from each of thethree principal extant lineages (Platnick amp Gertsch1976 Coddington amp Levi 1991 Coddington 2005)Liphistius (Mesothelae) Aphonopelma (Mygalomor-phae) and Hypochilus (Araneomorphae) The sampleis small with respect to known diversity but the basalphylogeny and relevant groundplan states of the orderare well established and states derived from the rep-resented taxa are consistent with them

Amblypygi (whipspiders) This is a small extant ter-restrial order (five families 17 genera 136 spp)(Harvey 2002) and was represented in the matrix bytwo fairly well-studied genera Charinus and Phry-nus Charinus retains several features that appear tobe plesiomorphic for the order (eg adult pedal pul-villi coxal glands associated with leg 3 eversiblevesicles) Weygoldt (1996) has examined the phylog-eny of the order and has provided an important sum-mary of morphology and general biology (Weygoldt2000)

Thelyphonida or Uropygi ss (whipscorpions vinega-roons) Extant whipscorpions (16 genera 106 spp)(Harvey 2002) were represented here by one well-studied species Mastigoproctus giganteus althoughadditional information was drawn from Typopeltis andThelyphonus The morphology of the group is highlyconserved there have been no modern studies ofintraordinal phylogeny A controversial fossil speciesfrom the late Carboniferous Proschizomus (Dunlop ampHorrocks 19951996) was also included but less well-preserved Carboniferous forms were not

Schizomida This small order (two extant families~34 genera ~200 spp) (Harvey 2002) is widelyregarded as the sister group of Thelyphonida Thetaxon sample included one representative from eachextant family Protoschizomus (Protoschizomidae) andStenochrus (Hubbardiidae) especially S portoricen-sis There are three recent (Pliocene) fossil speciesall from the same locality these were not includedhere

Ricinulei The extant ricinuleids (three genera 55spp) (Harvey 2002) were represented by two generaCryptocellus and Ricinoides Detailed studies of grosscuticular anatomy and post-embryonic developmentare available for a representative of each genus (Cryp-tocellus Pittard amp Mitchell 1972 Ricinoides Legg1976) and some information on internal anatomy isavailable for Ricinoides (Millot 1945) and this hasbeen extrapolated to Cryptocellus in the matrix There

are two basic fossil types (Selden 1992) one resem-bling modern taxa and another with a unique opistho-soma that superficially resembles the closed elytra of abeetle (ie curculioids) Two fossils from the formergroup were included Terpsicroton and Poliochera asthey are reasonably well preserved and show impor-tant characters (eg two pairs of eyes or evidence ofopisthosomal diplosegmentation) not expressed inextant forms No curculioid ricinuleids were includedas they appear to offer no additional information rel-evant to resolving ordinal relationships

Opilioacariformes (= Opilioacarida Notostigmata)The opilioacariform mites (nine genera 20 spp) (Har-vey 2002) are generally regarded as plesiomorphicAcari and are fairly conserved in their morphologyThe group was represented by two species Neocarustexanus and Siamacarus withi N texanus has a typi-cal opilioacariform morphology and its external anat-omy has been particularly well studied (eg Van derHammen 1989 Klompen 2000) S withi differs frommost other opilioacariforms in having trichobothriaand three rather than two pairs of lateral eyes (the lat-ter also in Paracarus) features that are potentiallysignificant for assessing ordinal relationships Someinformation on internal morphology of Opilioacaruswas taken from With (1904) and extrapolated to Neo-carus and Siamacarus

Parasitiformes This large group of mites comprisesthree clades Holothyrida Ixodida (ticks) and Mesos-tigmata (= Gamasida) Holothyrida (three familiesfive genera lt 30 spp) (Walter amp Proctor 1999) wasrepresented here by two species Australothyrus ocel-latus and Allothyrus constrictus Ixodids (three fami-lies 22 genera ~860 spp) (Kierans amp Robbins 1999)were represented by Amblyomma and Argas andmesostigmatids (~70 families ~10 000 spp) were rep-resented by Glyptholaspis confusa

Acariformes (= Actinotrichida) This is a very largediverse group of mites that includes the Endeostig-mata Sarcoptiformes and Prostigmata Endeostig-mata is an apparently paraphyletic assemblage of~ten families A morphologically generalized speciesAlycus roseus (Bimichaeliidae) was included herewith character states extracted largely from Van derHammen (1989) Sarcoptiformes (gt 200 families) iseffectively synonymous with Oribatida with Astig-mata being a large derived lsquooribatidrsquo clade (Norton1998 Maraun et al 2004) The group was representedby an early divergent plesiomorphic genus Palaeac-arus (Palaeosomata) and a more derived and well-studied genus Archegozetes (especially A longisetusand A magnus) The Prostigmata (~120 families~7000 spp) (Walter amp Proctor 1999) was representedby two genera Allothrombium and Microcaeculus

PHYLOGENY OF THE ARACHNID ORDERS 229

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Pseudoscorpiones The pseudoscorpions (24 families425 genera ~3200 spp) (Harvey 2002) were repre-sented in the matrix by four terminal genera Chtho-nius (Chthoniidae) was drawn from the basallydivergent superfamily Chthonioidea (Harvey 1992)Feaella (Feaellidae) represented the enigmatic super-family Feaelloidea which has an unusual combinationof characters and may be an early divergent lineageclose to Chthonioidea (Harvey 1992) Neobisium (Neo-bisiidae) is morphologically intermediate between theearly divergent families and the higher families rep-resented by Chelifer (Cheliferidae)

Solifugae (= Solpugida) (sun spiders) The order issmall (~1000 spp) but the absence of modern phylo-genetic treatments precludes meaningful estimates ofgenera and families (Harvey 2002) The taxon sampleincluded two genera one from the New World Eremo-costa and one from the Old World Galeodes

Opiliones (harvestmen) Extant harvestmen (~25 fam-ilies ~500 genera ~6000 spp) are divided into twosuborders Cyphophthalmi and Phalangida Cyphoph-thalmids were represented here by Chileogovea oedi-pus (Petallidae) which is currently under study by theauthor and Cyphophthamus duricorius (Sironidae)which is probably the most well-studied member of thesuborder (eg Janczyk 1956) Phalangida includesthree major groups Laniatores Eupnoi and Dyspnoibut there is disagreement as to whether Dyspnoi is thesister group to Laniatores or to Eupnoi (Shultz ampRegier 2001 Giribet et al 2002) Laniatores was rep-resented by Sclerobunus and Gonyleptes and Eupnoiby Leiobunum and Caddo

Scorpiones (scorpions) The extant scorpions (16 fam-ilies 155 genera 1279 spp) (Fet et al 2000) wererepresented by three taxa Centruroides vittatus(Buthidae) Hadrurus arizonensis (Iuridae) and Heter-ometrus spinifer (Scorpionidae) Buthidae is widelyregarded as the sister group to other extant lineagesand similarities among the represented terminals arelikely to be ground plan features of extant scorpionsgenerally The morphology of fossil scorpions is sub-stantially more diverse (Kjellesvig-Waering 1986)but their phylogenetic relationships are unclear (butsee Jeram 1998) The fossil taxa used in this study(Proscorpius Stoermeroscorpio Palaeoscorpius Pre-arcturus) were chosen for quality of preservation andor presence of a phylogenetically significant constella-tion of characters

CHARACTER CODING

Most character states were determined from directobservation the primary literature and authoritativereviews In some cases states were assigned to termi-

nals based on observations from related species asnoted in the Appendix In the matrix state lsquo-rsquo indicatesthat the character is inapplicable because the taxonlacks a more general character For example if a taxonlacks eyes then special features of the eyes (numberposition retinal configuration etc) are inapplicable tothat taxon A lsquorsquo indicates that the state is unknown oruncertain An entry with multiple states (eg 01 345) should be treated as an ambiguity code not a poly-morphism it indicates that two or more interpreta-tions of homology are applicable and that assignmentwas established analytically by character concor-dance Although lsquo-rsquo lsquorsquo and lsquo01rsquo are analyticallyidentical when applied to binary characters theyprovide information about the empirical status of thecharacter state in specific taxa Characters are citedthroughout the text as italicized numbers in parenthe-ses and are discussed in the Appendix

PHYLOGENETIC ANALYSIS

Phylogenetic analysis of the full data matrix wasperformed using the program Tree Analysis usingNew Technologies (TNT) ver 1 (Goloboff Farris ampNixon 2000) using lsquotraditionalrsquo search based on1000 replicates using TBR branch swapping Resultswere compared to those obtaining using the ratchetalgorithm (Nixon 1999) to determine any differencedue to analytical approach Nodal support for theminimal-length topology was evaluated by boot-strap (Felsenstein 1985) and Bremer support(Bremer 1994) Bootstrap analysis was conducted inTNT and based on 1000 pseudoreplicates each anal-ysed by ten random-addition replicates using TBRbranch swapping A nexus file containing the result-ing 1000 trees was imported into PAUP ver 10(Swofford 2002) to obtain bootstrap frequenciesBremer support was determined in TNT by con-straining specific nodes in the minimal-length topol-ogy and then determining the shortest tree that didnot recover the specified clade The difference inlength between the unconstrained and constrainedminimal-length trees is the Bremer support Theeffect of homoplasy on results was explored by con-ducting implied weights analysis (Goloboff 1993) inTNT Six analyses were conducted each with con-stant of concavity (k) set to a different integer valueof 1ndash6 where 1 is weighted most severely againsthomoplasious characters Each implied weightsanalysis was conducted using lsquotraditionalrsquo searchbased on 1000 replicates using TBR branch swap-ping The same procedures were used in analysing amatrix that included only extant taxa Howevercharacters rendered uninformative by removal offossil taxa (ie 49 98 99 100 113 114 125 161)were excluded prior to analysis

230 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

COMPARISONS OF ALTERNATIVE HYPOTHESES

Phylogenetic hypotheses proposed in the recent liter-ature were also evaluated including those thatattempted to resolve arachnid phylogeny completely(Fig 2) their hypothesized subclades (Fig 3) andhypotheses that proposed only specific nodes (Figs 34) The fully resolved topologies of Weygoldt amp Paulus(1979) Van der Hammen (1989) Shultz (1990)Wheeler amp Hayashi (1998) and Giribet et al (2002)were compared with the optimal topology using theTempleton test (Templeton 1983) as implementedin PAUP Internal relationships of multisampledorders were constrained to match those of the optimaltopology unless the original authors explicitlyfavoured an alternative Node-specific hypotheseswere evaluated by determining the frequency withwhich the node was recovered in bootstrap analysisThey were also evaluated by constraining parsimonyanalysis in TNT to recover the shortest tree contain-ing each specific node and then assessing the effect onrelative tree length and overall resolution takingnote of effects on otherwise stable or well-supportedclades The entire matrix was used to assess the twocases where node-specific hypotheses involved fossiltaxa (Fig 4)

RESULTS

EXTANT TAXA

Unweighted analysis of extant taxa produced two min-imal-length trees (length 383 CI 057) (Fig 5A) withconflicts limited to relationships within Pseudoscorpi-ones Multiply sampled orders were recovered asmonophyletic with high nodal support as measured bybootstrap percentage (BP gt 80) except Acari (BP 30)There were few well-supported interordinal groupsexcept Uropygi (BP 100) Pedipalpi (BP 100) Tetra-pulmonata (BP 93) and ArachnidaXiphosura (BP100) All other interordinal relationships were recov-ered with BP less than 60

Results from implied weights analysis (IWA) indi-cated a significant effect of homoplasy on resolution ofthe deepest interordinal relationships although theinterordinal clades Stomotheca Haplocnemata Acar-omorpha Tetrapulmonata Pedipalpi and Uropygiwere recovered under all values of k IWA with k = 1 (1tree best score = 5124) recovered Megoperculata(= Palpigradi + Tetrapulmonata) and Micrura(= Megoperculata + Acaromorpha) with Micrura beingthe sister group to Haplocnemata (= Pseudoscorpiones+ Solifugae) and with Stomothecata (= Opiliones +Scorpiones) being the sister group to all other arach-nids (Fig 5B) With k = 2 (1 tree best score = 3807)and k = 3 (1 tree best score = 3054) Megoperculatawas recovered and Stomotheca remained the sister

group to other arachnids but the relative positions ofHaplocnemata and Acaromorpha changed (Fig 5C)IWA using k = 4 (1 tree best score = 2559) k = 5 (1tree best score = 2205) and k = 6 (1 tree bestscore = 1938) recovered topologies consistent with theminimal-length topology derived from unweightedcharacters (Fig 5A)

EXTANT AND FOSSIL TAXA

Unweighted analysis of extant and fossil taxa pro-duced 18 minimal-length trees (length 426 CI 056)(Fig 6A) with conflicts limited to relationships withinPseudoscorpiones Scorpiones and Araneae Multiplysampled orders were again recovered as monophyleticwith high nodal support (BP gt 80) except Acari whichwas reconstructed as diphyletic with Anactinotrichidabeing the sister group to Ricinulei (BP 33) Therewere no well-supported interordinal groupings (ieBP gt 80) except Uropygi (BP 89) Pedipalpi (BP 92)and Arachnida (BP 90)

Results from IWA indicated a significant effect ofhomoplasy on resolution of the deepest interordinalrelationships in Arachnida although StomothecataHaplocnemata Acaromorpha Cryptognomae Pantet-rapulmonata Tetrapulmonata Schizotarsata Pedi-palpi and Uropygi were recovered under all values ofk IWA with k = 1 (7 trees best score = 5675) recov-ered Megoperculata and Micrura with Micrura beingthe sister group to Haplocnemata Stomothecata wasnot recovered in the strict consensus (Fig 6B) Withk = 2 (3 trees best score = 4243) IWA recovered thetopology shown in Fig 6C with Palpigradi being thesister group to all other arachnids The strict consen-sus topologies reconstructed using k = 3ndash6 were iden-tical to those obtained with k = 2 (k = 3 3 trees bestscore = 3419 k = 4 3 trees best score = 2874 k = 53 trees best score = 2483 k = 6 3 trees bestscore = 2188)

COMPARISONS OF ALTERNATIVE HYPOTHESES

The hypothesis of Weygoldt amp Paulus (1979) (Fig 2)was 398 steps (15 steps longer than the unconstrainedminimal-length tree) Van der Hammen (1989) was400 steps (17 steps longer) Shultz (1990) was 399steps (six steps longer) Wheeler amp Hayashi (1998)was 391 steps (eight steps longer) and Giribet et al(2002) was 394 steps (11 steps longer) Templetontests conducted at the 005 significance level rejectedthe hypothesis of no difference between the optimaltopology and those of Weygoldt amp Paulus (P = 00002)Van der Hammen (P = 00011) Wheeler amp Hayashi(P = 00018) and Giribet et al (P = 00343) The Shultztopology was not significantly different (P = 01444)from the optimal topology

PHYLOGENY OF THE ARACHNID ORDERS 231

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 2 Fully resolved phylogenetic hypotheses of extant euchelicerate groups proposed in the recent literature Note thesimilarity in the topologies of the parsimony-based analyses by Shultz (1990) Wheeler amp Hayashi (1998) and Giribet et al(2002) The Giribet et al topology is based on neontological data (morphology and molecules) and the original lsquoROOTrsquo maybe an artefact from use of the highly divergent pycnogonids as an outgroup

Xip

hosu

raS

corp

ione

sO

pilio

nes

Ara

neae

Am

blyp

ygi

Uro

pygi

Ric

inul

eiA

nact

inot

richi

daS

olifu

gae

Pse

udos

corp

ione

sP

alpi

gra

diA

carif

orm

es

Xip

hosu

raS

corp

ione

sA

rane

aeA

mbl

ypyg

iU

ropy

giP

alpi

gra

diS

olifu

gae

Pse

udos

corp

ione

sO

pilio

nes

Ric

inul

eiA

cari

Lipoctena

Arachnida

Apulmonata

Holotracheata

Cryptoperculata

Acaromorpha(ldquoAcarinomorphardquo)

Tetrapulmonata(rdquoMegoperculatardquo)

LabellataHaplo-cnemata

Weygoldt amp Paulus (1979)

Labellata

Tetrapulmonata(rdquoArachnidardquo)

Haplocnemata(rdquoApatellatardquo)

Van der Hammen (1989)

Xip

hosu

raP

alpi

gra

diA

rane

aeA

mbl

ypyg

iU

ropy

giR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Pedipalpi Haplocnemata

Novogenuata

Dromopoda

Shultz (1990)

Acaro-morpha

Micrura

Xip

hosu

raP

alpi

gra

diU

ropy

giA

mbl

ypyg

iA

rane

aeR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Labellata Haplocnemata

Novogenuata

Dromopoda

Wheeler amp Hayashi (1998)

Acaro-morpha

Micrura

Xip

hosu

raA

cari

Pal

pig

radi

Ric

inul

eiA

rane

aeA

mbl

ypyg

iU

ropy

giO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Tetrapulmonata

PedipalpiHaplocnemata

Novogenuata

Dromopoda

Giribet et al (2002)

ROOT

Micrura

Crypto-gnomae

Sternocoxata

Epimerata

RostrosomataMyliosomata

Neosternata

232 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 3 Consensus trees produced by parsimony analysis of neontological data constrained to produce relationships pro-posed in the recent literature The constrained (target) node is indicated by a black dot and the taxa encompassed by theconstraint are enclosed in a box Numbers below each tree represent the number of minimal-length constrained treeslength of minimal-length trees difference in the length of the unconstrained minimal-length tree and the constrained min-imal-length tree and percentage unconstrained bootstrap trees in which the target node was recovered respectively Thesetrees indicate the effect on branch length imposed by specific hypotheses and impact of constraining target nodes on non-target nodes

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Lipoctena4 385 2 lt5

Xi

Sc

Op

Ps

So

Pa

Ur

Am

Ar

Ri

An

Ac

Labellata6 397 14 lt5

Xi

Sc

Ar

Am

Ur

Op

Pa

Ps

So

Ri

An

Ac

Apulmonata2 390 7 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

So

Ri

An

Ac

Holotracheata5 390 7 lt5

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptoperculata2 388 5 lt5

Xi

Sc

Op

Pa

So

Ps

Ar

Am

Ur

Ri

An

Ac

Rostrosomata6 384 1 25

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Neosternata3 388 5 lt5

Xi

Sc

Op

Ar

Am

Ur

Ri

An

Ps

So

Pa

Ac

Coxisternata2 388 5 lt5

Epimerata7 387 4 11

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptognomae6 384 1 24

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Micrura3 385 2 7

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Megoperculata3 384 1 28

Xi

Sc

Ps

So

Op

Pa

Ar

Am

Ur

Ri

An

Ac

Novogenuata10 388 5 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

Ri

So

An

Ac

Solifugae + Acari4 388 5 lt5

Shultz (1990) Alberti amp Peretti (2002)

Van der Hammen (1989)

Weygoldt amp Paulus (1979)

Xi

Ar

Am

Ur

So

Ps

Sc

Op

Ri

Pa

An

Ac

PHYLOGENY OF THE ARACHNID ORDERS 233

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Results from analyses of node-specific hypothesesare summarized in Figures 3 4 In most cases pro-posed relationships were recovered in fewer than 5of bootstrap pseudoreplicates The notable exceptionswere Rostrosomata (25) Cryptognomae (24) andMegoperculata (28) and the strict-consensus con-strained minimal-length tree for each was only onestep longer than the unconstrained optimal tree Inmost cases constrained topologies did not add manysteps most were less than five steps longer and nonewas greater than 14 steps longer than the un-constrained minimal-length tree It is noteworthyhowever that even constrained nodes that imposerelatively few extra steps sometimes supportedimprobable relationships or eliminated clades thatwere well supported or stable in the unconstrainedanalysis Specific examples and their implications arediscussed below

DISCUSSION

RECENT ISSUES IN ARACHNID PHYLOGENY

Available morphological evidence consistently re-solves some interordinal relationships and fails to

resolve others (Fig 1) Continued progress depends onthe ability of arachnologists to discover new charac-ters and to assess the evidence critically Whilereviewing the recent arachnological literature severalaspects of phylogenetic practice emerged that seemedcounterproductive to both the perception and theactual rate of progress toward resolving arachnid phy-logeny Some of these are summarized here withspecific examples given in the remainder of theDiscussion and in the Appendix

There is a tendency to portray arachnid ordinal phy-logeny as more poorly resolved and contentious than isactually the case (Coddington et al 2004) Phyloge-netic hypotheses generated during different historicalperiods and using differing standards of evidence areoften cited as examples of current disagreement (egSelden 1993 Dunlop 1996 Selden amp Dunlop 1998Wheeler amp Hayashi 1998) In fact recent parsimony-based analyses of morphology have tended to convergeon topologies with internal structures congruent withthose found here (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) (Fig 2) Arachnidphylogeny is not fully resolved especially at its deep-est levels (Fig 1) but this does not mean that allaspects of arachnid phylogeny are controversial orpoorly supported by the available evidence

Matrices are sometimes constructed by uncriticallsquorecyclingrsquo of erroneous or problematic charactersbased on diverse secondary sources (Jenner 2001)Conclusions derived from mixtures of valid invalidspeculative and redundant characters are sometimesportrayed as the phylogenetic signal provided by mor-phology Data recycling can perpetuate errors (see 1320 39 52 136 140 144 152 169 171ndash174) legiti-mize speculations (see 13 32 63 95) or create dupli-cate or non-independent characters (see 13 30 70 77172) The lsquolateral organrsquo is a notable exampleYoshikura (1975) equated the embryonicearly post-embryonic lsquolateral organsrsquo of Amblypygi Thelyphonidaand Solifugae with the dissimilar lsquolateral organrsquo ofXiphosura but failed to note the very similar Cla-paregravede organ of Acariformes (see 173 174) This cod-ing was recycled by Wheeler amp Hayashi (1998) andthen by Giribet et al (2002) Error is probably inevi-table when assembling a large matrix from morphol-ogy including the one presented here but this can beminimized by making original observations and byconsulting primary sources

Some workers advocate weighting characters a pri-ori on the basis of structural or functional complexity(eg Kraus 1998 Dunlop amp Braddy 2001) and dismissphylogenetic conclusions derived from equal-weightsparsimony In short these workers criticize parsimonyfor emphasizing data quantity over data quality Thiscriticism ignores the intense debate in systematic biol-ogy that eventually led to widespread adoption of

Figure 4 Consensus trees produced by parsimony analy-sis of the full data matrix constrained to produce relation-ships proposed in the recent literature See legend toFigure 3 for details

Xi

Eu

Sc

Op

Pa

Ac

Ri

An

So

Ps

Tr

Ar

Ha

Am

UrEurypterida slat + Scorpiones12 435 9 lt5

Xi

Eu

Sc

Op

Pa

Ac

An

So

Ps

Ri

Tr

Ar

Ha

Am

UrTrigonotarbida+ Ricinulei28 434 8 lt5

Dunlop amp Braddy (2002)et al

Dunlop (1996)

234 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 5 Results from analysis of neontological data A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology AFor B and C relationships within terminal clades are the same as those shown in A

LimulusTachypleusProkoeneniaEukoeneniaCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAllothyrusAustralothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesLiphistiusAphonopelmaHypochilusCharinusPhrynusStenochrusProtoschizomusMastigoproctusCyphophthalmusChileogoveaCaddoLeiobunumSclerobunusGonyleptesCentruroidesHadrurusHeterometrusChthoniusFeaellaNeobisiumCheliferEremocostaGaleodes

10031

10018451

501561

733

723

944

633

10010

832

901

943

903

522

341

1007

1008

Acaromorpha

Haplocnemata

10012

Stomothecata

Opiliones

Scorpiones

Pseudo-scorpiones

Solifugae

Acariformes

Parasitiformes

Opilioacariformes

Ricinulei

Xiphosura

Palpigradi

Thelyphonida

Schizomida

Araneae

Amblypygi

10011

936853

100141005

1009

Tetrapulmonata Pedipalpi

Uropygi

692

522572

968

331

603

81

511

161

101

Acari

Dromopoda

Arachnida

A

Xiphosura

Stomothecata

Haplocnemata

Acaromorpha

Palpigradi

TetrapulmonataMicrura

Megoperculata

BXiphosura

Stomothecata

Acaromorpha

Haplocnemata

Palpigradi

TetrapulmonataMegoperculata

C

PHYLOGENY OF THE ARACHNID ORDERS 235

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 6 Results from analysis of the full data matrix A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology CFor B and C relationships within terminal clades are the same as those shown in A

WeinberginaLimuloidesEuproopsTachypleusLimulusStylonurusBaltoeurypterusChasmataspisDiploaspisOctoberaspisCyphophthamusChileogoveaCaddoLeiobunumSclerobunusGonyleptesProscorpiusPalaeoscorpiusStoermeroscorpioPrearcturusCentruroidesHadrurusHeterometrusProkoeneniaEukoeneniaTerpsicrotonPoliocheraCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAustralothyrusAllothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesEremocostaGaleodesChthoniusFeaellaNeobisiumCheliferGilboarachnePalaeocharinusLiphistiusAphonopelmaHypochilusPlesiosiroCharinusPhrynusStenochrusProtoschizomusProschizomusMastigoproctus

631

893

923

984

984

671

341

671

986

10011

423

974

Xiphosura

Eurypterida sstr

Chasmataspidida

Opiliones

Scorpiones

Palpigradi

Ricinulei

Opilioacariformes

Parasitiformes

Acariformes

Solifugae

Pseudoscorpiones

Trigonotarbida

Araneae

Haptopoda

Amblypygi

Schizomida

Thelyphonida

842

251331

361311

552

832

882

925

913

573

561291

443

976

541

954

743644

732521

571

331

833

663591

966

573

71

221

905

832

351

Uropygi

Pedipalpi

Haplocnemata

Tetrapulmonata

Anactinotrichida

Cryptognomae

Acaromorpha

Stomothecata

Pantetrapulmonata

Arachnida

Eurypterida slat

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatProscorpiusStoermeroscorpioPalaeoscorpiusPrearcturusCentruroidesHadrurusHeterometrusOpilionesHaplocnemataRicinuleiAnactinotrichidaAcariformesPalpigradiPantetrapulmonata

Megoperculata

Micrura

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatPalpigradiStomothecataHaplocnemataPantetrapulmonataAcaromorpha

A

B

C

Schizotarsata

236 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

parsimony and instead advocates a return to thespeculative and subjective approaches of the late 19thand early 20th century that once threatened thescientific legitimacy of the discipline (Bowler 1996)Furthermore it misrepresents the properties of parsi-mony-based analyses Specifically characters withfunctional significance are readily encompassed byparsimony analyses and in fact characters derivedfrom locomotor systems (eg 46ndash94) have played animportant role in developing current ideas aboutarachnid phylogeny (eg Shultz 1989 1991) In addi-tion lsquocomplex charactersrsquo can be viewed as compositesof several characters such that morphological com-plexity is effectively weighted by the number of inde-pendent lsquosubcharactersrsquo it contains For example thelsquosucking stomachrsquo once considered a synapomorphy ofLabellata (= Araneae + Amblypygi Fig 3) is codedhere as a composite of three characters (199ndash201) TheLabellata hypothesis was not corroborated in thepresent analysis (Figs 1 3ndash5) and in fact was highlydisfavoured (Fig 3) but this result cannot be dis-missed as a failure to acknowledge the complexity ofthe character

Some workers support specific (target) clades withone or more similarities without exploring the impacton overall homoplasy or relationships among non-target clades Each character offers its own phyloge-netic hypothesis which may or may not be consistentwith relationships implied by other characters It isexceedingly rare for all characters to be perfectly com-patible in the phylogenetic hypotheses they supportand criteria such as parsimony have been developed todiscover those hypotheses that minimize the conflict-ing phylogenetic signals of different characters Eventhough character conflict (homoplasy) is a virtuallyinescapable phenomenon in comparative biology it isnot uncommon for workers to discover one or morecharacters and to promote their phylogeneticimplications over alternative hypotheses even thosethat otherwise appear to be well supported Thisapproach may have value in highlighting new dataand perspectives but accomplishes this by promotingthe erroneous impression that all aspects of arachnidphylogeny are so tenuous that a single character canfalsify even well-supported hypotheses Several exam-ples of this approach have appeared in the recentarachnological literature

For example Alberti amp Peretti (2002) argued thataflagellate spermatozoa (163) are a compelling syna-pomorphy for a Solifugae + Acari clade and dismissedsome of the characters that support Solifugae + Pseu-doscorpiones as having lsquodebatable valuersquo Yet theirproposal rejects two hypotheses that are consistentlyrecovered in recent phylogenetic analyses that isHaplocnemata (= Solifugae + Pseudoscorpiones) andAcaromorpha (= Acari + Ricinulei) (Figs 1 2) Rejec-

tion of Haplocnemata would require its presumed syn-apomorphies to be reinterpreted as homoplasiesincluding features of the chelicerae (18 19 20) pre-oral chamber (13 32) legs (12 48) and respiratorysystem (126) The same reason applies to Acaromor-pha and its synapomorphies It is noteworthy that aSolifugae + Acari clade was recovered in the presentanalysis in fewer than 5 of bootstrap pseudorepli-cates and that analyses constrained to recover thisclade were five steps longer than the minimal-lengthtree and favoured a problematic clade unitingRicinulei and Pseudoscorpiones (Fig 3)

In another example Dunlop (1996) proposed a closerelationship between Trigonotarbida and Ricinuleibased on two-segmented chelicerae (18) prosomandashopisthosoma coupling mechanism (96) diplotergites(100 101) and longitudinally divided opisthosomaltergites (115) However phylogenetic analyses con-strained to recover this relationship required eightadditional steps eliminated support for Acari and nec-essarily rejected Acaromorpha and forced its synapo-morphies to be reinterpreted as homoplasies Alberti(2005) proposed an interesting hypothesis for the evo-lution of male gonads in tetrapulmonates (153) butchose to accept Labellata (= Araneae + Amblypygi) indeveloping his argument over the much more well-supported Pedipalpi (= Amblypygi + Uropygi) (Shultz1999) a phylogenetic reconfiguration that was recov-ered here in fewer than 5 of bootstrap pseudorepli-cates increased tree length by a minimum of 14 stepsand resulted in the collapse of Stomothecata and Hap-locnemata (Fig 3) Many other examples can be cited

Promoting or defending a specific phylogenetichypothesis via lists of compatible synapomorphies is acommon but problematic approach By restrictingattention to the states of specific characters at one ortwo target nodes one can easily overlook the unin-tended impact of the hypothesis on phylogenetic sig-nal elsewhere and its effect on non-target clades Anode supported by a long list of synapomorphies mayseem convincing taken in isolation but may becomeless acceptable when its full phylogenetic implicationsare explored

SUMMARY OF INTERORDINAL ARACHNID CLADES RECOVERED IN THIS ANALYSIS

Arachnida Lamarck 1801Analyses consistently recovered Arachnida as a mono-phyletic group with high bootstrap support (Figs 1 45) Possible synapomorphies include the loss of thecarapacal pleural doublure (9) cardiac lobe (10) pedalgnathobases (52) and moveable endites (53) and thegain of aerial respiration (120) and an anteriorly oranteroventrally directed mouth (185) Some tradi-tional synapomorphies such as slit sensilla (142) and

PHYLOGENY OF THE ARACHNID ORDERS 237

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

fluid feeding (184) may have appeared later in arach-nid evolution but this can only be decided once theinternal phylogeny of Arachnida has been determined

Some workers regard many proposed arachnid syn-apomorphies as adaptations to terrestrial life andthereby link the hypothesis of arachnid monophyly tothe hypothesis of a single ancestral aquatic-to-terres-trial transition and arachnid polyphyly to multipletransitions (eg Selden amp Jeram 1989 Dunlop 1997Dunlop amp Webster 1999 Dunlop amp Braddy 2001)Given this line of reasoning the existence of appar-ently aquatic scorpions in the fossil record (Kjellesvig-Waering 1986 Jeram 1998) and the inference thatterrestrialization occurred late in scorpion evolutionthese workers conclude that character states support-ing arachnid monophyly are actually convergencesand do not necessarily support arachnid monophyly

However this approach to assessing phylogenetichypotheses is founded on overly simplistic assump-tions such as the ability of the investigator to discrim-inate unerringly between characters that existexclusively in aquatic organisms (including fossils)from those that occur exclusively in terrestrial organ-isms There also appears to be an assumption thathomoplasy can be generated through parallelism (iemultiple aquatic-to-terrestrial events) but not throughterrestrial-to-aquatic reversals Furthermore anassumed dichotomy between exclusively aquatic andterrestrial life histories in ancestral arachnids is sim-plistic as illustrated by the amphibious life cycles ofbasal vertebrates and pterygote hexapods In factthese examples demonstrate that there is no neces-sary inconsistency in basing a hypothesis of arachnidmonophyly on the derived terrestrial features of anamphibious ancestor whose descendants thencompleted terrestrialization once or several timesindependently or even returned to a fully aquaticexistence Workers who link the frequency and direc-tion of aquaticndashterrestrial transitions to the assess-ment of arachnid phylogeny do so by endowingthemselves with substantially greater insight thanseems prudent by ignoring the huge gaps in ourunderstanding of early arachnid evolution and bydenying to arachnids the evolutionary complexityknown to exist in other groups

Several palaeontologists have been particularlyactive during the past decade in proposing new char-acters with the stated goal of removing Scorpionesfrom Arachnida and erecting a Scorpiones + Eurypter-ida clade (Braddy amp Dunlop 1997 Dunlop amp Braddy1997 Dunlop 1998 Braddy et al 1999 Dunlop ampWebster 1999) Dunlop amp Braddy (2001) recentlysummarized this evidence and conducted aparsimony-based analysis of Xiphosura EurypteridaScorpiones Opiliones and Tetrapulmonata (but notHaplocnemata Acaromorpha or Palpigradi) using 33

morphological characters Their analysis produced atopology congruent with those generated here (Figs 15 6) including recovery of Stomothecata (= Scorpiones+ Opiliones) However they rejected this result as aproduct of lsquoempirical cladisticsrsquo because it gives thesame weight to prosomal characters that supportarachnid monophyly and to selected opisthosomalcharacters that support their favoured Eurypterida +Scorpiones clade namely a five-segmented metasoma(116) suppression of opisthosomal tergite 1 (95) lossof lamellate respiratory organs on the postgenitalsomite (122) Kiemenplatten (125) loss of respiratorylamellae on the genital segment (121) and a lsquonon-stainingrsquo exocuticle (but see Grainge amp Pearson 1966for evidence of this in Opiliones see Appendix for com-ments on the other characters)

A Eurypterida + Scorpiones clade was not favouredin the present analysis (Fig 6) the strict consensus ofminimal-length trees constrained to recover this clade(Fig 4) is nine steps longer than the tree recovered byanalysis without this constraint It is also noteworthythat Opiliones was consistently reconstructed as thesister group to Eurypterida + Scorpiones in the mini-mal-length constrained trees a provocative result thatwas probably unintended and unanticipated by sup-porters of the Eurypterida + Scorpiones Given thatDunlop amp Braddy (2001) reject equal weights parsi-mony as an arbiter of phylogenetic hypotheses theywould presumably dismiss these results as irrelevantto their argument just as they dismissed their ownparsimony-based results However if the Eurypterida+ Scorpiones hypothesis is to be credible it must beopen to evaluation and potential falsification usingobjective criteria and the subjective or intuitive a pri-ori weighting of characters advocated by Dunlop ampBraddy clearly does not qualify At present it isunclear how one would objectively evaluate Dunlop ampBraddyrsquos proposal with criteria compatible with thoseused in its original formulation For now the Euryp-terida + Scorpiones concept advocated by Dunlop ampBraddy may persist outside the mainstream of mod-ern systematic practice but it is increasingly problem-atic within it

Stomothecata nomnovOpiliones and Scorpiones were consistently recoveredas a monophyletic group The proposed name acknowl-edges a unique preoral chamber the stomothecaformed by coxapophyses of the palp and leg 1 (50)often with an auxiliary role played by the coxapophy-sis of leg 2 (51) In addition the epistome appears tohave been modified for adduction of the palpal coxaeThe lateral walls of the epistome are fused to themedial surfaces of the palpal coxae and the epistomallumen is spanned by a transverse muscle (188) which

238 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

apparently adducts the palpal coxae thereby constrict-ing the stomothecal chamber Scorpions and opilionsare also unique in having a pair of large epistomalarms projecting rearward into the prosoma andattaching to the endosternite (189) The epistomalarms provide attachment sites for pharyngeal dilatormuscles (196) and extrinsic muscles of anterior proso-mal appendages (eg 37) The chelicera is equippedwith a muscle that arises on the carapace and insertson the ventral margin of the second cheliceral article(23) There is an anteriorly placed genital opening(155)

Several notable similarities were found whilereviewing the literature but information from otherarachnid groups was considered too incomplete toallow them to be included in the matrix For exampleboth orders have apparent haemocytopoeitic organsassociated with major nerves of the anterior opistho-soma These are termed supraneural organs in scorpi-ons (Farley 1999) and perineural organs in opilions(Kaumlstner 1935) Haemocytes develop in the cardiacwall in spiders (Seitz 1972) and perhaps amblypygids(Weygoldt 2000) but haemocytopoeitic organs areunknown in most other arachnid groups Germ cellsdifferentiate very early during embryogenesis in bothscorpions and opilions relative to spiders (Moritz1957 Anderson 1973) Additional research is requiredto determine the phylogenetic utility of thesecharacters

Haplocnemata Boumlrner 1904Several workers recognize a close phylogenetic rela-tionship between Pseudoscorpiones and Solifugae(Weygoldt amp Paulus 1979 Van der Hammen 1989Shultz 1990 Wheeler amp Hayashi 1998 Giribetet al 2002) and this was consistently recovered inthe present analysis Haplocnemata is supported byseveral synapomorphies including (i) a feeding com-plex formed by two-segmented chelicerae (18) with aventrolateral intrinsic articulation (19) and perhapsa dorsolateral articulation with the carapace (20)and (ii) a rostrosoma (32) a preoral apparatus formedby an anteriorly projecting epistome affixed dorsallyto enlarged palpal coxae and bordered by lateral pal-pal projections The epistome forms the dorsal compo-nent of a lsquobeakrsquo with the ventral part formed by amidventral sternapophysis (13) fused between thepalpal coxae The locomotor apparatus consists ofcoxae that meet along the ventral midline (12) elon-gate femur-like patellae (48) and an apotele in theadult modified as an eversible empodium or pulvillus(92) The respiratory system is formed entirely (Pseu-doscorpiones) or in part (Solifugae) by paired trachealstigmata opening on opisthosomal somites 3 and 4(126)

Acaromorpha Dubinin 1957A clade comprising Ricinulei and Acari is consistentlyrecovered in the present analysis It is united here bytwo unique and seven homoplasious synapomorphiesThese include a gnathosoma (31) defined in part bymedial fusion of the palpal coxae (30) although pres-ence of a gnathosoma in Ricinulei is debatable as dis-cussed in the Appendix Acaromorphs also have aunique post-embryonic development consisting of ahexapodal larva and up to three octopodal nymphalinstars (176) The pedal patellandashtibia joints are formedby a bicondylar dorsal hinge rather than a single mid-dorsal condyle (69) all postcheliceral segmental gan-glia are unified in the subesophageal ganglion (130)and a postcerebral pharynx is absent (192) The groupis also tentatively united by presence of differentiatedpedal basi- and telofemora (63 64) and the absence ofa ventral (sternal) pharyngeal dilator muscle (199)but these may be symplesiomorphies erroneouslyreconstructed as a synapomorphies

The internal phylogenetic structure of Acaromorphais controversial with many recent workers favouringa monophyletic Acari (Weygoldt amp Paulus 1979Shultz 1990 argued most thoroughly by Lindquist1984) and others advocating a diphyletic Acari withAcariformes being the sister group to Cryptognomae(= Ricinulei + Anactinotrichida) (especially Van derHammen 1979 1989) (Fig 3) Acari was recovered asmonophyletic when fossils were excluded (Fig 2) butnot when they were included (Fig 3) A lsquomite-centredrsquosurvey of arachnid characters may be needed if mor-phology is to offer a compelling solution to the internalphylogeny of Acari and its placement within Arach-nida These issues are not resolved by the presentanalysis

Pantetrapulmonata nom novPantetrapulmonata includes the extinct orders Trigo-notarbida Haptopoda and the extant orders AraneaeAmblypygi Schizomida and Thelyphonida The cladeis united by cheliceral structure (18 19) a megoper-culum (106) booklungs on the genital and first post-genital somites (121 122) and enlargement of theepipharyngeal sclerite (192) It is important to notehowever that most of these characters were coded asuncertain in Plesiosiro (Haptopoda) Aside from theplacement of Haptopoda the monophyly of Trigono-tarbida and the extant orders was anticipated byShear et al (1987) and is generally regarded as amonophyletic group

Tetrapulmonata Shultz 1990Tetrapulmonata was originally proposed on the basisof neontological analyses and encompassed Araneae

PHYLOGENY OF THE ARACHNID ORDERS 239

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Amblypygi Schizomida and Thelyphonida (Shultz1990) and was also recovered here in analysis of neon-tological data (Fig 5) However results from analysisof all taxa required that the Tetrapulmonata conceptbe expanded to include the fossil order HaptopodaFeatures uniting this group are problematic howeveras many states in Haptopoda were coded as uncertain

Schizotarsata nomnovHaptopoda and Pedipalpi (= Amblypygi + Schizomida+ Thelyphonida) are united here in a group named forpossession of divided pedal telotarsi (84) Synapomor-phies include a pointed anterior carapacal margin (3)and elongation of leg 1 (46) As already noted theplacement of Haptopoda should probably be regardedas tentative because the state of many characters inthis group was coded as unknown The placement ofHaptopoda as the sister group to Pedipalpi was antic-ipated by Dunlop (1999 2002c)

Pedipalpi Boumlrner 1904Pedipalpi encompasses Amblypygi and Uropygi(= Thelyphonida + Schizomida) Although widely rec-ognized in the past Pedipalpi was set aside in the mid-20th century in favour of Labellata (= Amblypygi +Araneae) based largely on the presumed synapomor-phies of a narrow prosomandashopisthosoma juncture (97)and sucking stomach (199ndash201) The situation beganto reverse when Shear et al (1987) again highlightedsimilarities of Amblypygi and Uropygi such as rapto-rial palps (35) and antenniform leg 1 (46) Shultz(1989 1990 1999) subsequently described numerousderived similarities including asymmetrical flexormuscles at the pedal femurndashpatella joint (70) threetelotarsomeres on legs 2ndash4 (85) and modification ofthe palpal coxae to provide support for extrinsic pha-ryngeal muscles (197) Pedipalpi has emerged as oneof the most well-supported interordinal relationshipsin Arachnida

Uropygi Thorell 1882Uropygi has long been accepted as a monophyleticunion of Thelyphonida and Schizomida Synapomor-phies include a unique mating behaviour (159) fusedpalpal coxae (30) 2-1-1-1 arrangement of tibial tricho-bothria (144) posterior defensive glands (102) elon-gated patella of leg 1 (68) and many others

ACKNOWLEDGEMENTS

I thank Nikolaj Scharff and three anonymous review-ers for useful comments This work was supported bythe Maryland Agricultural Experiment Station the

General Research Board of the University of Mary-land and National Science Foundation Grant 0416628to Susan Masta

REFERENCES

Abd el-Wahab A 1952 Notes on the morphology of the scor-pion Buthus quinquestriatus (HE) Publications de lrsquoInsti-tut Fouad 1er du Desert 3 1ndash129

Alberti G 1979 Licht- und elektronenmikroskopische Unter-suchungen an Coxaldruumlsen von Walzenspinnen (ArachnidaSolifugae) Zoologischer Anzeiger 203 48ndash64

Alberti G 1995 Comparative spermatology of Cheliceratareview and perspective Memoires du Museum NationaldrsquoHistoire Naturelle 166 203ndash230

Alberti G 2005 Double spermatogenesis in ChelicerataJournal of Morphology 266 281ndash297

Alberti G Coons LB 1999 Acari mites In Harrison FWFoelix RF eds Microscopic anatomy of invertebratesVol 8C Chelicerate Arthropoda New York Wiley Liss515ndash1215

Alberti G Peretti AV 2002 Fine structure of male genitalsystem and sperm in Solifugae does not support a sister-group relationship with Pseudoscorpiones (Arachnida)Journal of Arachnology 30 268ndash274

Alexander AJ 1967 Problems of limb extension in the scor-pion Opisthophthalmus latimanus Koch Transactions of theRoyal Society of South Africa 37 165ndash181

Anderson DT 1973 Embryology and phylogeny in annelidsand arthropods Oxford Pergamon Press

Anderson LI Selden PA 1997 Opisthosomal fusion andphylogeny of Palaeozoic Xiphosura Lethaia 30 19ndash31

Beall BS Labandeira CC 1990 Macroevolutionary patternsof the Chelicerata and Tracheata In Culver SJ ed Arthro-pod paleobiology Short Courses in Paleontology No 3 Lon-don The Paleontological Society 257ndash284

Bernard HM 1896 The comparative morphology of the Gale-odidae Transactions of the Linnean Society of London Sec-ond Series 6 305ndash417

Boumlrner C 1902a Arachnologische Studien (II und III) Zool-ogischer Anzeiger 25 433ndash466

Boumlrner C 1902b Arachnologische Studien IV Die Genitalor-gane der Pedipalpen Zoologischer Anzeiger 26 81ndash92

Boumlrner C 1904 Beitraumlge zur Morphologie der Arthropoden IEin Beitrag zur Kenntnis der Pedipalpen Zoologica (Stut-tgart) 42 1ndash174

Bowler PJ 1996 Lifersquos splendid drama Chicago Universityof Chicago Press

Braddy SJ Alridge RJ Gabbott SE Theron JN 1999Lamellate book-gills in a late Ordovician eurypterid from theSoom Shale South Africa support for a eurypterid-scorpionclade Lethaia 32 72ndash74

Braddy SJ Dunlop JA 1997 The functional morphology ofmating in the Silurian eurypterid Baltoeurypterus tet-ragonophthalmus (Fischer 1839) Zoological Journal of theLinnean Society 121 435ndash461

Brauer A 1895 Beitrage zur Kenntnis der Entwicklungsge-schichte des Skorpiones II Zeitschrift fuumlr WissenschaftlicheZoologie 59 351ndash433

240 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bremer K 1994 Branch support and tree stability Cladistics10 295ndash304

Bristowe WS 1932 The liphistiid spiders Proceedings of theZoological Society of London Part 4 1016ndash1057

Butt AG Taylor HH 1991 The function of spider coxalorgans effects of feeding and salt-loading on Porrhotheleantipodiana (Mygalomorpha Dipluridae) Journal of Exper-imental Biology 158 439ndash461

Buxton BH 1913 Coxal glands of arachnids ZoologischeJahrbucher (Suppl XIV) 16ndash50

Buxton BH 1917 Notes on the anatomy of arachnids Jour-nal of Morphology 29 1ndash31

Camin JH Clark GM Bourdeau FG 1956 The palpallsquotined setarsquo in the Mesostigmata a homologue of the palpalclaw in the Onychopalpida (Acarina) In Becker EC ed Pro-ceedings of the Tenth International Congress of EntomologyOttawa Mortimer 903ndash908

Chamberlin JC 1931 The arachnid order ChelonethidaStanford University Stanford University Press

Chapman RF 1998 The insects structure and function 4thedn Cambridge Cambridge University Press

Clarke KU 1979 Visceral anatomy and arthropod phylogenyIn Gupta AP ed Arthropod phylogeny New York Van Nos-trand Reinhold 467ndash549

Clarke J 1984 On the relationship between structure andfunction in the leg joints of Heteropoda venatoria (L) (Ara-neae Eusparassidae) Bulletin of the British ArachnologicalSociety 6 181ndash192

Clarke J 1986 The comparative functional morphology of theleg joints and muscles of five spiders Bulletin of the BritishArachnological Society 7 37ndash47

Clarke JM Ruedemann R 1912 The Eurypterida of NewYork New York State Museum Memoir 14 (1ndash2) 1ndash628

Coddington JA 2005 Phylogeny and classification of spi-ders In Ubick D Paquin P Cushing PE Roth V eds Spi-ders of North America an identification manual AmericanArachnological Society 18ndash24

Coddington JA Giribet G Harvey MS Prendini LWalter DE 2004 Arachnida In Cracraft J Donoghue MJeds Assembling the tree of life Oxford Oxford UniversityPress 269ndash318

Coddington JA Levi HW 1991 Systematics and evolutionof spiders (Araneae) Annual Review of Ecology and System-atics 22 565ndash592

Coineau Y Van der Hammen L 1979 The postembryonicdevelopment of Opilioacarida with notes on new taxa and ona general model for the evolution In Piffle E ed Proceed-ings of the 4th International Congress of Acarology Budap-est Akadeacutemiai Kiadoacute 437ndash441

Cokendolpher JC Reddell JR 1992 Revision of the Pro-toschizomidae (Arachnida Schizomida) with notes on thephylogeny of the order Texas Memorial Museum Speleolog-ical Monographs 3 31ndash74

Condeacute B 1991a Le dimorphisme sexuel des Palpigrades Bul-letin de la Socieacuteteacute Neuchacircteloise des Sciences Naturelles 11667ndash73

Condeacute B 1991b Lrsquoaire genitale male des Palpigrades Bulle-tin de la Socieacuteteacute Europeenne drsquoArachnologie 1990 64ndash69

Dearden PK Donly C Grbicccc M 2002 Expression of pair-rule gene homologues in a chelicerate early patterning ofthe two-spotted spider mite Tetranychus urticae Develop-ment 129 5461ndash5472

Delle Cave L 1975 The ldquocleaning brushrdquo in Damon diademaSimon 1876 and Phrynichus jayakari Pocock 1894 (Ambly-pygi Arachnida) Monitore Zoologico Italiano (NS) 9 175ndash184

Dunlop JA 1994 Filtration mechanisms in the mouthparts oftetrapulmonate arachnids (Trigonotarbida Araneae Ambly-pygi Uropygi Schizomida) Bulletin of the British Arachno-logical Society 9 267ndash273

Dunlop JA 1996 Evidence for a sister group relationshipbetween Ricinulei and Trigonotarbida Bulletin of the BritishArachnological Society 10 193ndash204

Dunlop JA 1997 Palaeozoic arachnids and their significancefor arachnid phylogeny In Zabka M ed Proceedings of the16th European colloquium of arachnology Siedlce PolandWyzsza Szkola Rolnicko-Pedagogiczna 65ndash82

Dunlop JA 1998 The origins of tetrapulmonate book lungsand their significance for chelicerate phylogeny In SeldenPA ed Proceedings of the 17th European Colloquium ofArachnology Edinburgh 1997 Burnham Beeches BritishArachnological Society 9ndash16

Dunlop JA 1999 A redescription of the Carboniferous arach-nid Plesiosiro madeleyi Pocock 1911 (Arachnida Hapto-poda) Transactions of the Royal Society of Edinburgh EarthSciences 90 29ndash47

Dunlop JA 2000 The epistomo-labral plate and lateral lips insolifuges pseudoscorpions and mites Ekoloacutegia (Bratislava)19 67ndash78

Dunlop JA 2002a Arthropods from the Lower Devonian Sev-ernaya Zemlya Formation of October Revolution Island(Russia) Geodiversitas 24 349ndash379

Dunlop JA 2002b Character states and evolution of the che-licerate claws In Toft S Scharff N eds European Arach-nology 2000 Aarhus Aarhus University Press 345ndash354

Dunlop JA 2002c Phylogeny of Chelicerata In LlorenteBousquets J Morrone JJ eds Biodiversidad Taxonomiacutea yBiogeografiacutea de Arthroacutepodos de Meacutexico Hacia Una Siacutentesisde Su Conocimiento Vol III Prensas de Ciencias Univer-sidad Nacional Autonoma de Meacutexico 117ndash141

Dunlop JA Anderson LI Braddy SJ 2004 A redescriptionof Chasmataspis laurencii Caster amp Brooks 1956 (Chelicer-ata Chasmataspidida) from the Middle Ordovician of Ten-nessee USA with remarks on chasmataspid phylogenyTransactions of the Royal Society of Edinburgh Earth Sci-ences 94 207ndash225

Dunlop JA Braddy SJ 1997 Slit-like structures on theprosomal appendages of the eurypterid BaltoeurypterusNeues Jahrbuch fuumlr Geologie und Palaumlontologie-Monatshefte1997 31ndash38

Dunlop JA Braddy SJ 2001 Scorpions and their sister-group relationships In Fet V Selden PA eds Scorpions2001 in Memoriam Gary A Polis London British Arachno-logical Society 1ndash24

Dunlop JA Horrocks CA 19951996 A new Upper Carbon-iferous whip scorpion (Arachnida Uropygi Thelyphonida)

PHYLOGENY OF THE ARACHNID ORDERS 241

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

with a revision of the British Carboniferous Uropygi Zoolo-gischer Anzeiger 234 293ndash306

Dunlop JA Poschmann M Anderson L 2001 On theEmsian (early Devonian) arthropods of the Rhenish SlateMountains 3 The chasmataspidid Diploaspis Palaumlon-tolgische Zeitschrift 75 253ndash269

Dunlop JA Webster M 1999 Fossil evidence terrestrializa-tion and arachnid phylogeny Journal of Arachnology 27 86ndash93

Edgecombe GD Wilson GDF Colgan DJ Gray MR Cas-sis G 2000 Arthropod cladistics combined analysis of his-tone H3 and U2 snRNA sequences and morphologyCladistics 16 155ndash203

Evans GO 1992 Principles of acarology Wallingford UKCABI

Fahrenbach WH 1999 Merostomata In Harrison FW Foe-lix RF eds Microscopic anatomy of invertebrates Vol 8AChelicerate arthropods New York Wiley-Liss 21ndash115

Farley RD 1999 Scorpiones In Harrison FW Foelix RF edsMicroscopic anatomy of invertebrates Vol 8A Cheliceratearthropods New York Wiley-Liss 117ndash222

Farley RD 2005 Developmental changes in the embryo pro-nymph and first molt of the scorpion Centruroides vittatus(Scorpiones Buthidae) Journal of Morphology 265 1ndash27

Fayers SR Dunlop JA Trewin NH 2004 A new earlyDevonian trigonotarbid arachnid from the Windyfield ChertRhynie Scotland Journal of Systematic Palaeontology 2269ndash284

Felsenstein J 1985 Confidence limits on phylogenies anapproach using the bootstrap Evolution 39 783ndash791

Fet V Sissom WD Lowe G Braunwalder ME 2000 Cat-alog of the scorpions of the world (1758ndash1998) New YorkNew York Entomological Society

Firstman B 1954 The central nervous system musculatureand segmentation of the cephalothorax of a tarantula (Eu-rypelma californicum Ausserer) Microentomology 87 1ndash40

Firstman B 1973 The relationship of the chelicerate arterialsystem to the evolution of the endosternite Journal ofArachnology 1 1ndash54

Foelix RF 1985 Mechano- and chemoreceptive sensilla InBarth FG ed Neurobiology of arachnids Berlin Spinger-Verlag 118ndash137

Foelix RF 1996 Biology of spiders 2nd edn Oxford OxfordUniversity Press

Giribet G Edgecombe GD Wheeler WC Babbit C 2002Phylogeny and systematic position of Opiliones a combinedanalysis of chelicerate relationships using morphologicaland molecular data Cladistics 18 5ndash70

Goloboff P 1993 Estimating character weights during treesearch Cladistics 9 83ndash91

Goloboff PA Farris JS Nixon K 2000 TNT Tree AnalysisUsing New Technologies Version 1 Program and documen-tation httpwwwzmucdkpublicphylogenytnt

Grainge CA Pearson RG 1966 Cuticular structure in thePhalangida Nature 211 866

Grandjean F 1938 Observations sur les Bdelles (Acariens)Annales de la Socieacuteteacute Entomologique de France Paris 107 1ndash24

Hansen HJ Soslashrensen W 1904 On two orders of ArachnidaCambridge Cambridge University Press

Hansen HJ Soslashrensen W 1905 The Tartarides a tribe of theorder Pedipalpi Arkiv fuumlr Zoologi 2 1ndash78

Hara MR Gnaspini P 2003 Comparative study of the defen-sive behavior and morphology of the gland opening areaamong harvestmen (Arachnida Opiliones Gonyleptidae)under a phylogenetic perspective Arthropod Structure andDevelopment 32 257ndash275

Harvey MS 1992 The phylogeny and classification of thePseudoscorpionida (Chelicerata Arachnida) InvertebrateTaxonomy 6 1373ndash1435

Harvey MS 2002 The neglected cousins what do we knowabout the smaller arachnid orders Journal of Arachnology30 357ndash372

Haupt J Song D 1996 Revision of East Asian whip scorpions(Arachnida Uropygi Theyphonida) I China and JapanArthropoda Selecta 5 43ndash52

Hjelle JT 1990 Anatomy and morphology In Polis GA edThe biology of scorpions Stanford University Stanford Uni-versity Press 9ndash63

Janczyk FSW 1956 Anatomie von Siro duricorius Joseph imVergleich mit anderen Opilioniden Sitzungsberichte Oster-reichischen Akademie der Wissenschaften Mathematisch-Natruwissenschaftliche Klasse 165 474ndash522

Jenner RA 2001 Bilaterian phylogeny and uncritical recy-cling of morphological data sets Systematic Biology 50 730ndash742

Jeram AJ 1998 Phylogeny classification and evolution of Sil-urian and Devonian scorpions In Selden PA ed Proceed-ings of the 17th European Colloquium of ArachnologyEdinburgh 1997 Burnham Beeches UK British Arachno-logical Society 15ndash31

Karaman IM 2005 Evidence of spermatophores in Cyphoph-thalmi (Arachnida Opiliones) Revue Suisse de Zoologie 1123ndash11

Kaumlstner A 1935 Die Funktion der sogenannten sympathis-chen Ganglien und die Exkretion bei den PhalangiidenZoologischer Anzeiger 109 273ndash288

Kierans JE Robbins RG 1999 A world checklist of generasubgenera and species of ticks (Acari Ixodida) publishedfrom 1973 to 1997 Journal of Vector Ecology 24 115ndash129

Kjellesvig-Waering EN 1986 A restudy of the fossil Scorpi-onida of the world Palaeontographica Americana 55 1ndash287

Klompen JSH 2000 Prelarva and larva of Opilioacarus (Neo-carus) texanus (Chamberlin and Mulaik) (Acari Opilioacar-ida) with notes on the patterns of setation and lyrifissuresJournal of Natural History 34 1977ndash1992

Kraus O 1998 Elucidating the historical process of phylog-eny phylogenetic systematics versus cladistic techniquesIn Selden PA ed Proceedings of the 17th European Collo-quium of Arachnology Edinburgh 1997 Burnham BeechesUK British Arachnological Society 1ndash7

Kraus O Kraus M 1993 Divergent transformation of cheli-cerae and original arrangement of eyes in spiders (Arach-nida Araneae) Memoirs of the Queensland Museum 33579ndash584

Lankester ER Benham WBS Beck EJ 1885 On the mus-

242 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

cular and endoskeletal systems of Limulus and Scorpio withsome notes on the anatomy and generic characters of scor-pions Transactions of the Zoological Society of London 11311ndash384

Leclerc P 1989 Considerations paleobiogeographiques a pro-pos la decouverte en Thailande drsquoopilioacariens nouveaux(Acari ndash Notostigmata) Compte Rendu des Seances de laSociete de Biogeographie 65 162ndash174

Legg G 1976 The external morphology of a new species ofricinuleid (Arachnida) from Sierra Leone Zoological Journalof the Linnean Society 59 1ndash58

Lindquist EE 1984 Current theories on the evolution ofmajor groups of Acari and on their relationships with othergroups of Arachnida with consequent implications for theirclassificaton In Griffiths DA Bowman CE eds AcarologyVI Vol 1 New York John Wiley amp Sons 28ndash62

Manning PL Dunlop JA 1995 The respiratory organs ofeurypterids Palaeontology 38 287ndash297

Manton SM 1958 Hydrostatic pressure and leg exten-sion in arthropods with special reference to arachnidsAnnals and Magazine of Natural History Series 13 1161ndash182

Manton SM 1964 Mandibular mechanisms and the evolutionof arthropods Philosophical Transactions of the Royal Soci-ety of London B 7 1ndash183

Maraun M Heethoff M Schneider K Scheu S WeigmannG Cianciolo J Thomas RH Norton RA 2004 Molecularphylogeny of oribatid mites (Oribatida Acari) evidence formultiple radiations of parthenogenetic lineages Experimen-tal and Applied Acarology 33 183ndash201

Marples BJ 1967 The spinnerets and epiandrous glands ofspiders Journal of the Linnean Society (Zoology) 46 209ndash222

Marples BJ 1983 Observations on the structure of the fore-gut of spiders Bulletin of the British Arachnological Society8 46ndash52

Millot J 1933 Notes compleacutementaires sur lrsquoanatomie desliphistiides et des hypchilides a propos drsquoun travail reacutecent deA Petrunkevitch Bulletin de la Socieacuteteacute Zoologique de France68 217ndash235

Millot J 1942 Sur lrsquoanatomie et lrsquohistophysiologie deKoenenia mirabilis Grassi (Arachnida Palpigradi) RevueFranccedilaise drsquoEntomologie 9 33ndash51

Millot J 1943 Notes complementaires sur lrsquoanatomielrsquohistologie et la repartition geographique en France deKoenenia mirabilis Grassi Revue Franccedilaise drsquoEntomologie 9127ndash135

Millot J 1945 Lrsquoanatomie interne des Ricinulei (Arachnides)Annales des Sciences Naturelles (Zoologie) Series 11 7 1ndash29

Millot J 1949a Morphologie generale et anatomie interne InGrasseacute P-P ed Traite de zoologie VI Paris Masson et Cie263ndash319

Millot J 1949b Ordre des Amblypyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 563ndash588

Millot J 1949c Ordre des Araneacuteides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 589ndash743

Millot J 1949d Ordre des Palpigrades In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 520ndash532

Millot J 1949e Ordre des Uropyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 533ndash562

Millot J 1949f Ordre des Ricinuleides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 744ndash760

Millot J Vachon M 1949 Ordre Des Solifuges In Grasseacute P-P ed Traite de zoologie VI Paris Masson et Cie 482ndash519

Moore RA Briggs DEG Bartels C 2005 A new specimen ofWeinbergina opitzi (Chelicerata Xiphosura) from the LowerDevonian Huumlnsruck Slate Germany PalaumlontologischeZeitschrift 79 399ndash408

Moritz M 1957 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) unter besonderer Beruumlcksichtigungder ausseren Morphologie der Bildung des Mitteldarmesund der Genitalanlage Zoologische Jahrbuumlcher Abteilungfuumlr Anatomie und Ontogenie der Tiere 76 331ndash370

Moritz M 1959 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) II Die Anlage und Entwicklung derCoxaldruumlse bei Phalangium opilio L Zoologische Jahr-buumlcher Abteilung fuumlr Anatomie und Ontogenie der Tiere 77229ndash240

Nixon K 1999 The parsimony ratchet a new method forrapid parsimony analysis Cladistics 15 407ndash414

Norton RA 1998 Morphological evidence for the evolutionaryorigin of Astigmata (Acari Acariformes) Experimental andApplied Acarology 22 559ndash594

Palmgren P 1978 On the muscular anatomy of spiders ActaZoologica Fennica 155 1ndash41

Patten W 1912 The evolution of the vertebrates and their kinPhiladelphia P Blakistonrsquos Son

Patten W Hazen AP 1900 The development of the coxalgland branchial cartilages and genital ducts of Limuluspolyphemus Journal of Morphology 16 459ndash490

Paulus HF 1979 Eye structure and the monophyly of theArthropoda In Gupta AP ed Arthropod phylogeny NewYork Van Nostrand Reinhold 299ndash383

Petrunkevitch A 1949 A study of Palaeozoic ArachnidaTransactions of the Connecticut Academy of Arts and Sci-ences 37 69ndash315

Pittard K Mitchell RW 1972 Comparative morphology ofthe life stages of Cryptocellus pelaezi (Arachnida Ricinulei)Graduate Studies Texas Tech University 1 1ndash77

Platnick NI 2005 The world spider catalog Version55 American Museum of Natural History httpresearchamnhorgentomologyspiderscatalogindexhtm

Platnick NI Gertsch WJ 1976 The suborders of spiders acladistic analysis (Arachnida Araneae) American MuseumNovitates 2607 1ndash15

Pocock RI 1893 On some points in the morphology of theArachnida (ss) with notes on the classification of the groupAnnals and Magazine of Natural History Series 6 11 1ndash19

Pocock RI 1902 Some points in the morphology and classi-fication of the Opiliones Annals and Magazine of NaturalHistory Series 7 10 504ndash516

Popadicccc A Panganiban G Rusch D Shear WA KaufmanTC 1998 Molecular evidence for the gnathobasic derivationof arthropod mandibles and for the appendicular origin ofthe labrum and other structures Development Genes andEvolution 208 142ndash150

PHYLOGENY OF THE ARACHNID ORDERS 243

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Raw F 1957 Origin of chelicerates Journal of Paleontology31 139ndash192

Reddell JR Cokendolpher JC 1995 Catalogue bibliogra-phy and generic revision of the order Schizomida (Arach-nida) Texas Memorial Museum Speleological Monographs4 1ndash170

Reissland A Goumlrner P 1985 Trichobothria In Barth FGed Neurobiology of arachnids Berlin Springer-Verlag 138ndash161

Roewer C-F 1934 Solifugae Palpigradi Bronns Klassen undOrdnungen ses Tierreich 5 IV 4 1ndash713

Rowland JM Cooke JAL 1973 Systematics of the arachnidorder Uropygida (=Thelyphonida) Journal of Arachnology 155ndash71

Rowland JM Sissom WD 1980 Report on a fossil palpi-grade from the Tertiary of Arizona and a review of themorphology and systematics of the order (ArachnidaPalpigradida) Journal of Arachnology 8 69ndash86

Rucker A 1901 The Texan Koenenia American Naturalist35 615ndash630

Ruhland M Rathmayer W 1978 Die Beinmuskulatur undihre Innervation bei der Vogelspinne Dugesiella henzti (Ch)(Araneae Aviculariidae) Zoomorphologie 89 33ndash46

Saboori A Kamali K 2000 Description of Allothrombiumtriticium adult (Acari Trombidiidae) from Iran Systematicand Applied Acarology 5 207ndash208

Schliwa M 1979 The retina of the phalangid Opilio raven-nae with particular reference to arhabdomeric cells Celland Tissue Research 204 473ndash495

Scholl G 1977 Beitrage zur Embryonalentwicklung von Lim-ulus polyphemus L (Chelicerata Xiphosura) Zoomorpholo-gie 86 99ndash154

Scholtz G Kamenz C 2006 The book lungs of Scorpionesand Tetrapulmonata (Chelicerata Arachnida) evidence forhomology and a single terrestrialization event of a commonarachnid ancestor Zoology 109 2ndash13

Seitz KA 1972 Zur Histologie und Feinstruktur des Herzensund der Haumlmocyten von Cupiennius salei Keys (AraneaeCtenidae) I Herzwandung Bildung und Differenzierungder Haumlmocyten Zoologische Jahrbuumlcher Abteilung fuumlr Anat-omie und Ontogenie der Tiere 89 351ndash384

Sekiguchi K ed 1988 Biology of horseshoe crabs Tokyo Sci-ence House

Selden PA 1981 Functional morphology of the prosoma ofBaltoeurypterus tetragonophthalmus (Fischer) (ChelicerataEurypterida) Transactions of the Royal Society of Edin-burgh Earth Sciences 72 9ndash48

Selden P 1992 Revision of the fossil ricinuleids Transactionsof the Royal Society of Edinburgh Earth Sciences 83 595ndash634

Selden PA 1993 Fossil arachnids ndash recent advances andfuture prospects Memoirs of the Queensland Museum 33389ndash400

Selden PA Dunlop JA 1998 Fossil taxa and relationshipsof chelicerates In Edgecombe GD ed Arthropod fossilsand phylogeny New York Columbia University Press 303ndash331

Selden PA Jeram AJ 1989 Palaeophysiology of terrestrial-

isation in the Chelicerata Transactions of the Royal Societyof Edinburgh Earth Sciences 80 303ndash310

Selden PA Shear WA Bonamo PM 1991 A spider and otherarachnids from the Devonian of New York and reinterpre-tations of Devonian Araneae Palaeontology 34 241ndash281

Sensenig A Shultz JW 2003 Mechanics of cuticular elasticenergy storage in leg joints lacking extensor muscles inarachnids Journal of Experimental Biology 206 771ndash784

Shear WA 1993 New species in the opilionid genus Stylocel-lus from Malaysia Indonesia and the Philippines (OpilionesCyphophthalmi Stylocellidae) Bulletin of the British Arach-nological Society 9 174ndash188

Shear WA Selden PA Rolfe WDI Bonamo PM GriersonJD 1987 New terrestrial arachnids from the Devonian ofGilboa New York (Arachnida Trigonotarbida) AmericanMuseum Novitates 2901 1ndash74

Shultz JW 1987 The origin of the spinning apparatus in spi-ders Biological Reviews of the Cambridge PhilosophicalSociety 62 89ndash113

Shultz JW 1989 Morphology of locomotor appendages inArachnida evolutionary trends and phylogenetic implica-tions Zoological Journal of the Linnean Society 97 1ndash56

Shultz JW 1990 Evolutionary morphology and phylogeny ofArachnida Cladistics 6 1ndash31

Shultz JW 1991 Evolution of locomotion in Arachnida thehydraulic pressure pump of the giant whipscorpion Masti-goproctus giganteus (Uropygi) Journal of Morphology 21013ndash31

Shultz JW 1993 Muscular anatomy of the giant whipscor-pion Mastigoproctus giganteus (Arachnida Uropygi) and itsevolutionary significance Zoological Journal of the LinneanSociety 108 335ndash365

Shultz JW 1998 Phylogeny of Opiliones (Arachnida) anassessment of the lsquoCyphopalpatoresrsquo concept Journal ofArachnology 26 257ndash272

Shultz JW 1999 Muscular anatomy of a whipspider Phrynuslongipes (Amblypygi) and its evolutionary significance Zoo-logical Journal of the Linnean Society 126 81ndash116

Shultz JW 2000 Skeletomuscular anatomy of the harvest-man Leiobunum aldrichi (Weed 1893) (Arachnida Opil-iones Palpatores) and its evolutionary significanceZoological Journal of the Linnean Society 128 401ndash438

Shultz JW 2001 Gross muscular anatomy of Limuluspolyphemus (Chelicerata Xiphosura) and its bearing on evo-lution in the Arachnida Journal of Arachnology 29 283ndash303

Shultz JW 2007 Morphology of the prosomal exoskeleton ofScorpiones (Arachnida) and a new hypothesis for the evolu-tion of cuticular cephalic endoskeletons in arthropodsArthropod Structure amp Development 36 77ndash102

Shultz JW Pinto da Rocha R 2007 Morphology and func-tional anatomy In Pinto da Rocha R Machado G Giribet Geds The harvestmen the biology of Opiliones CambridgeMA Harvard University Press 14ndash61

Shultz JW Regier JC 2001 Phylogenetic analysis of Pha-langida (Arachnida Opiliones) using two nuclear protein-encoding genes supports monophyly of Palpatores Journalof Arachnology 29 189ndash200

Sitnikova LG 1978 The main evolutionary trends of the

244 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Acari and the problem of their monophyletism Entomologis-cheskoe Obozrenie 57 431ndash457

Snodgrass RE 1948 The feeding organs of Arachnidaincluding mites and ticks Smithsonian Miscellaneous Col-lections 110 (10) 1ndash94

Snodgrass RE 1952 A textbook of arthropod anatomy Ith-aca NY Comstock Publishing

Soslashrensen W 1879 Om Bygningen af Gonyleptiderne en Typeaf Arachnidernes Classe Naturhistorisk Tidsskrifter Series3 8 97ndash222

Stockwell SA 1989 Revision of the phylogeny and higherclassification of scorpions (Chelicerata) Unpublished doc-toral dissertation University of California at Berkeley

Stoslashrmer L 1944 On the relationships and phylogeny of fossiland Recent Arachnomorpha Skrifter utgitt av Det NorskeVidenskaps-Akademi i Oslo I 5 1ndash158

Stoslashrmer L 1955 Merostomata In Moore RC ed Treatise oninvertebrate paleontology Part P Arthropoda 2 LawrenceKS University of Kansas Press and Geological Society ofAmerica 4ndash41

Swofford DL 2002 PAUP Phylogenetic analysis using par-simony (and other methods) Version Beta 10 SunderlandMA Sinauer Associates

Talarico G Palacios-Vargas JG Fuentes Silva M AlbertiG 2005 First ultrastructural observations of the tarsal poreorgan of Pseudocellus pearsei and P boneti (ArachnidaRicinulei) Journal of Arachnology 33 604ndash612

Templeton AR 1983 Phylogenetic inference from restrictionendonuclease cleavage site maps with articular reference tothe evolution of humans and the apes Evolution 37 221ndash244

Tetlie OE Braddy SJ 2004 The first Silurian chasmataspidLoganamaraspis dunlopi gen et sp nov (Chelicerata Chas-mataspidida) from Lesmahagow Scotland and its implica-tions for eurypterid phylogeny Transactions of the RoyalSociety of Edinburgh Earth Sciences 94 227ndash234

Thomas RH Telford MJ 1999 Appendage development inembryos of the oribatid mite Archegozetes longisetosus(Acari Oribatei Trhypochthoniidae) Acta Zoologica (Stock-holm) 80 193ndash200

Tollerton VP Jr 1989 Morphology taxonomy and classifica-tion of the order Eurypterida Burmeister 1843 Journal ofPaleontology 63 642ndash657

Vachon M 1949 Ordre des Pseudoscorpions In Grasseacute P-Ped Traite de zoologie VI Paris Masson et Cie 436ndash481

Van der Hammen L 1979 Comparative studies in Chelicer-ata I The Cryptognomae (Ricinulei Architarbi and Anacti-notrichida) Zoologische Verhandelingen 174 1ndash62

Van der Hammen L 1982 Comparative studies in Chelicer-ata II Epimerata (Palpigradi and Actinotrichida) Zoolo-gische Verhandelingen 196 1ndash70

Van der Hammen L 1985 Functional morphology and affin-ities of extant Chelicerata in evolutionary perspectiveTransactions of the Royal Society of Edinburgh 76 137ndash146

Van der Hammen L 1989 An introduction to comparativearachnology Leiden SPB Academic Publishing

Vyas AB 1970 Studies in myology of the scorpion Heter-ometrus fulvipes Koch Unpublished PhD DissertationDepartment of Zoology Gujarat University India

Vyas AB 1974 The cheliceral muscles of the scorpion Heter-ometrus fulvipes Bulletin Southern California Academy ofSciences 73 9ndash14

Walter DE 1988 Predation and mycophagy by endeostigma-tid mites (Acariformes Prostigmata) Experimental andApplied Acarology 4 159ndash166

Walter DE Proctor HC 1998 Feeding behaviour and phy-logeny observations on early derivative Acari Experimentaland Applied Acarology 22 39ndash50

Walter DE Proctor HC 1999 Mites ecology evolution andbehaviour Sydney University of New South Wales

Waterston CD 1979 Problems of functional morphology andclassification in stylonuroid eurypterids (Chelierata Meros-tomata) with observations on the Scottish Silurian Sty-lonuroidea Transactions of the Royal Society of EdinburghEarth Science 70 251ndash322

Weygoldt P 1969 Biology of pseudoscorpions CambridgeMA Harvard University Press

Weygoldt P 1984 Lrsquoautotomie chez les Amblypyges RevueArachnologique 5 321ndash327

Weygoldt P 1996 Evolutionary morphology of whip spiderstowards a phylogenetic system (Chelicerata ArachnidaAmblypygi) Journal of Zoological Systematics and Olution-ary Research 34 185ndash202

Weygoldt P 1998 Evolution and systematics of the Chelicer-ata Experimental and Applied Acarology 22 63ndash79

Weygoldt P 2000 Whip spiders (Chelicerata Amblypygi)their biology morphology and systematics StenstrupDenmark Apollo Books

Weygoldt P Paulus HF 1979 Untersuchungen zur Morphol-ogie Taxonomie und Phylogenie der Chelicerata Zeitschriftfuumlr Zoologische Systematik und Evolutionsforschung 17 85ndash116 177ndash200

Wheeler WC Hayashi CY 1998 The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192

Winkler D 1957 Die Entwicklung der aumlusseren Koumlrperge-stalt bei den Phalangiidae (Opiliones) Mitteilungen aus demZoologischen Museum in Berlin 33 355ndash389

With CJ 1904 The Notostigmata a new suborder of AcariVidenskabelige Meddelelser fra Dansk Naturhistorisk Foren-ingi Kjobenhaus 1904 137ndash192

Yamasaki T Makioka T Saito J 1988 Morphology InSekiguchi K ed Biology of horseshoe crabs Tokyo ScienceHouse 69ndash132

Yoshikura M 1975 Comparative embryology and phylogeny ofArachnida Kumamoto Journal of Science Biology 12 71ndash142

APPENDIX

Each character used in this analysis is numbereddefined and cross-referenced with characters used infour previous studies that are abbreviated in bracketsspecifically WP = Weygoldt amp Paulus (1979)S = Shultz (1990) WH = Wheeler amp Hayashi (1998)GEWB = Giribet et al (2002) Descriptions includecorrections of errors discussions of controversies orambiguities and justifications for state assignmentsto lsquoproblemrsquo taxa

PHYLOGENY OF THE ARACHNID ORDERS 245

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

PROSOMA

1 Dorsal sclerite formed by fusion of the prosomal car-apace the dorsal portion of the first opisthosomalsomite and the dorsomedial (axial) portion of the sec-ond opisthosomal somite 0 absent 1 presentState 1 is present in extant xiphosurids and close fos-sil relatives (Scholl 1977 Anderson amp Selden 1997Shultz 2001) The fossil record suggests that thisstructure originated with the disappearance ofthe already reduced first opisthosomal tergite(= microtergite) and axial portion of the secondopisthosomal tergite (Anderson amp Selden 1997)2 Single dorsal sclerite covering entire dorsal surfaceof body no lines indicating original segmentation 0absent 1 presentState 1 occurs throughout Holothyrida (Acari) (Vander Hammen 1989) and in many Mesostigmata(Alberti amp Coons 1999) including Glyptholaspis (Vander Hammen 1989)3 Anterior end of dorsal prosoma with medianmarginal or submarginal pointed process 0 absent1 presentState 1 occurs throughout Schizomida (Protoschizo-mus Stenochrus Reddell amp Cokendolpher 1995) The-lyphonida (Rowland amp Cooke 1973 Shultz unpublobs Proschizomus Dunlop amp Horrocks 19951996)Amblypygi (Charinus Millot 1949b Phrynus Shultz1999) and Plesiosiro (Dunlop 1999) Citing a figure inRowland amp Cooke (1973) Dunlop amp Horrocks assumedthat the anterior process was limited to lsquohypoctonidrsquothelyphonids and united them with Proschizomus andSchizomida If valid this would make Thelyphonidaparaphyletic However the doubtful monophyly oflsquohypoctonidsrsquo (Haupt amp Song 1996) and widespreadpresence of a submarginal anterior process in Pedi-palpi weaken this proposal (Harvey 2002)4 Ophthalmic ridges pair of longitudinal crests inter-secting or passing near the region of the lateral eyes orcomparable region where lateral eyes are absent 0absent 1 presentState 1 is present throughout Xiphosura (Anderson ampSelden 1997) Similar structures are present in Plesi-osiro (Dunlop 1999) and lsquonon-hypoctonidrsquo Thely-phonida (Mastigoproctus Rowland amp Cooke 1973)Dunlop (2002a) described apparent ophthalmic ridgesin the chasmataspidid Octoberaspis but these werenot observed in Diploaspis (Dunlop et al 2001) eitherdue to absence or inadequate preservation of speci-mens nor in Chasmataspis (Dunlop et al 2004)5 Carapace with demarcations (eg grooves scleritesphragmata) between pro- meso- or metapeltidia 0absent 1 present [S 1 WH 48 GEWB 29]State 1 is present in extant Scorpiones Cyphoph-thalmi (Hansen amp Soslashrensen 1904) and many Palpa-tores (Opiliones) including Caddo and Leiobunum

and most Pseudoscorpiones (Chamberlin 1931) Giri-bet et al (2002) coded this character as absent inCyphophthalmi (Opiliones) but a transverseprocurved groove is present throughout the group(Hansen amp Soslashrensen 1904) and represents the meso-metapeltidial border as indicated by attachment ofthe dorsal endosternal suspensor of postoral somite VIin Chileogovea oedipus and Siro exilis (Shultz unpublobserv) Kjellesvig-Waering (1986) did not explicitlydescribe such grooves in fossil scorpions although heillustrated corresponding structures in PrearcturusArchaeophonus Archaeoctonus Palaeophonus andother Palaeozoic scorpions (Kjellesvig-Waering 1986figs 14A 21E 27D 59) (coded as uncertain here) butnot in Proscorpius (contra Giribet et al 2002) Dis-tinct carapacal sclerites are present throughout Palpi-gradi Schizomida and Solifugae (6)

The condition in Acari is controversial but is codedhere as State 1 Van der Hammen (1989) consideredthe tergal region of the leg-bearing somites (postoralsomites IIIndashVI) or podosoma to be reduced andreplaced by posterior migration of the aspidosoma (ietergal region assumed to be associated with append-ages of the gnathosoma) and anterior migration of theopisthosomal tergal region to form the hysterosomaThe dorsal proterosomal and hysterosomal elementsmeet at a transverse sejugal furrow that continues lat-erally and passes ventrally between the coxae of legs 2and 3 According to this scheme the prosomandashopisthosoma border is expressed as the disjugal fur-row which passes from the ventral posterior margin ofthe podosoma anterodorsally to join the dorsal part ofthe sejugal furrow Based on the arrangement of setaeand slit sensilla (= lyrifissures) Van der Hammen(1989) interpreted the region above coxae 3 and 4 (ieregion C) as a fusion of the dorsal parts of the first twoopisthosomal somites (= postoral somites VII and VIII)in early divergent Anactinotrichida (ie Opilioacari-formes) and Acariformes (eg Alycus) These interpre-tations have been followed by many acarologistsalthough its speculative aspects are widely acknowl-edged (Evans 1992 Alberti amp Coons 1999)

From the standpoint of a general arachnologist theVan der Hammen system seems unnecessarily compli-cated it appears to have been formulated to explainbroad morphological themes in Oribatida but was thenextrapolated to other mites A more conservativelsquoarachnologicalrsquo scheme adopted here equates the pro-podosoma with the pars cephalica (= propeltidium) ofthe arachnid prosoma and region C with the pars tho-racica (= mesopeltidium + metapetidium) In fact theearliest and most recent treatments of Opilioacari-formes (With 1904 Klompen 2000) reached similarconclusions This interpretation is also consistent withevidence from Acariformes For example the regiondorsal to the coxa of legs 3 and 4 in Alycus is as readily

246 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

explained by the persistence of primitive prosomal ter-gal elements as by the supposition that these disap-peared and were replaced by two opisthosomalelements Furthermore several diverse lineages ofbasally divergent Acariformes display features that donot fit readily within the GrandjeanndashVan der Ham-men system but seem to correspond to the cephalicathoracica division of the prosoma They have a sejugalfurrow that may have demarcated the cephalicatho-racica border and a postpodosomal furrow thatappears to represent the prosomandashopisthosoma junc-tion there is no disjugal furrow Examples includepalaeosomate (especially aphelacarid) and pediculoch-elid Oribatida alicorhagiid and mircopsammidlsquoEndeostigmatarsquo and paratydeid Prostigmata (Albertiamp Coons 1999 fig 42) This organization may be sym-plesiomorphic for Acariformes but could be a conver-gence brought about by selection for enhancedflexibility6 Carapace with distinct pro- meso- or metapeltidialsclerites 0 absent 1 present - inapplicable due toabsence of pro- meso- or metapeltidial demarcations(5)State 1 occurs throughout Palpigradi Schizomida andSolifugae It also appears sporadically within Opil-iones (eg Leiobunum flavum Shultz unpubl observ)and Pseudoscorpiones (eg Pseudochiridium) (Cham-berlin 1931)7 Sejugal furrow circumferential zone of body flexi-bility that passes between the coxae of legs 2 and 3 0absent 1 presentState 1 is probably synapomorphic for Acariformes asit occurs in most lsquoendeostigmatidsrsquo including Alycus(Van der Hammen 1989) and is variously developedthroughout Sarcoptiformes (Alberti amp Coons 1999) Itis probably the primitive condition for Prostigmatabut is either absent or weakly expressed in represen-tatives included here See 5 for alternative morpholog-ical interpretations of this body region in Acari8 Prosomal ozopores 0 absent 1 present [WH 46GEWB 12]State 1 occurs throughout Opiliones and Holothyrida(Van der Hammen 1989)9 Carapacal pleural doublure 0 absent 1 present[WP 12 S 2 WH 49 GEWB 27]A carapacal pleural doublure occurs throughoutXiphosura (Stoslashrmer 1944) Eurypterida (Clarke ampRuedemann 1912 Selden 1981) and Chasmataspid-ida (Dunlop et al 2001 2004 Dunlop 2002a) and isabsent in all known Arachnida10 Cardiac lobe a longitudinal axial elevation of thecarapace 0 absent 1 presentState 1 occurs throughout Xiphosura (Anderson ampSelden 1997) and Eurypterida (Clarke amp Ruedemann1912 Selden 1981) Carapaces are not preserved suf-ficiently to determine this condition in the chasmatas-

pidids Diploaspis (Dunlop et al 2001) andOctoberaspis (Dunlop 2002a) but an apparent cardiaclobe occurs in the Chasmataspis (Dunlop et al 2004)A cardiac lobe is absent in Arachnida11 Moveable cucullus 0 absent 1 present [WP 54GEWB 22]State 1 is a synapomorphy of Ricinulei (Selden 1992)12 Medial intercoxal lsquosternalrsquo region 0 all pedalcoxae separated medially 1 anterior pedal coxae abut-ting medially posterior coxae separated 2 anteriorpedal coxae separated medially posterior coxae abut-ting 3 all pedal coxae abutting medially 4 epimeracoxae undifferentiated medially from ventral body wall[S sim3 WH sim50 GEWB sim17]Prosomal sternites appear to be those portions of theventral body wall not occupied by coxae Despite earlyattempts to assess metamerism of the ventralprosoma (protosternum deutosternum tritosternumetc) (Boumlrner 1902a Millot 1949a) no apparent mor-phological features such as borders of sclerites ormuscle attachments reliably demarcate the ventralbody wall of one somite from that of an adjacentsomite Thus coding schemes that focus on describingmetameric components of the sternal region are prob-lematic Shultz (1990) attempted to code the shape ofthe entire intercoxal region and subsequent workershave adopted this approach All pedal coxae are sepa-rated medially in Plesiosiro (Dunlop 1999) Palpigradi(Roewer 1934) Araneae (Millot 1949c) Amblypygi(Weygoldt 2000) and many Acari including Opilioac-ariformes Holothryrida Mesostigmata Ixodida Aly-cus and many Prostigmata (Van der Hammen 1989Evans 1992 Alberti amp Coons 1999) The anteriorcoxae are separated in extant Thelyphonida Schizo-mida (Hansen amp Soslashrensen 1905 Millot 1949e) andRicinulei (Pittard amp Mitchell 1972 Legg 1976) Thecoxae are fused to the ventral body wall in Sarcopti-formes (Van der Hammen 1989 Evans 1992)

Interpreting this character is complicated for theanterior coxae of Scorpiones and Opiliones and allcoxae in Xiphosura and Eurypterida due to the pres-ence of coxapophyses endites andor gnathobaseswhich form components of the preoral chambers inthese groups Thus coxae may abut medially whenthese structures are considered part of the coxae butare separate medially if these structures are ignoredAs coxapophyses etc are here coded as separate char-acters (50ndash53) I have chosen to code this character asif these structures did not exist Consequently pha-langid Opiliones are coded as having all coxae sepa-rated (Pocock 1902 Hansen amp Soslashrensen 1904) Fossilscorpions also show substantial variation in this char-acter All pairs of pedal coxae in Palaeoscorpius abutmedially but also have gnathobases (Jeram 1998)and this condition is coded here as uncertain A ster-num separates all pairs of pedal coxae medially in

PHYLOGENY OF THE ARACHNID ORDERS

247

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Proscorpius

and

Stoermeroscorpio

(State 0) and

Pre-arcturus

approximates modern scorpions (State 1)(Kjellesvig-Waering 1986 Jeram 1998) The coxalbases of Eurypterida and Xiphosura are separated bya small sternite the endostoma (Selden 1981 Shultz2001)

13 Postoral sternapophysis (

=

tritosternum labium)a cuticular evagination of the ventral body wall poste-riorly adjacent to the palpal coxae forming the poste-rior border of the preoral chamber in some taxa ordisplaced posteriorly by fusion of the palpal coxae inothers 0 absent 1 present

[S

sim

10 WH

sim

56 GEWB19

=

33]State 1 is present throughout Palpigradi AraneaeAmblypygi (Snodgrass 1948) Schizomida (Hansen ampSoslashrensen 1905 Van der Hammen 1989) Thely-phonida (Van der Hammen 1989) extant Ricinulei(Pittard amp Mitchell 1972 Legg 1976) Opilioacari-formes and Mesostigmata (Van der Hammen 1989Alberti amp Coons 1999)

Allothyrus constrictus

is theonly holothyrid mite known to have sternapophysesAn apparent sternapophysis has been described in apalaeocharinid trigonotarbid (Dunlop 1994) Giribet

et al

(2002) miscoded Ixodida and Acariformes as hav-ing a lsquolabium (

=

tritosternum)rsquo The lophognath ofpseudoscorpions (Chamberlin 1931) and lsquolabiumrsquo ofSolifugae (Roewer 1934) also appear to be sternapo-physes that have been incorporated into the ros-trosoma (

32

) The labium of phalangid Opiliones(Pocock 1902 Hansen amp Soslashrensen 1904) is a ster-napophysis but is associated with leg 1 rather thepedipalp (Winkler 1957) A small anterior sclerite inthe fossil scorpions

Proscorpius

Waeringoscorpio

and

Labriscorpio

has been interpreted as a lsquolabiumrsquo homol-ogous with the sternapophyses of other arachnids(Weygoldt 1998) despite absence of any indication ofthe relationship of this sclerite to the mouth and pos-sible alternative interpretations (eg

16

) Giribet

et al

coded this sclerite as a sternapophysis but the char-acter is coded here as uncertain in fossil scorpions

14 Channels on the body surface linking openings ofcoxal organs to preoral chamber 0 absent 1 present

Van der Hammen (1989) described bilaterally pairedcuticular tracts (lsquotaenidiarsquo) connecting orifices of coxalorgans to the preoral chamber in some arachnids andor a ventromedian groove (

=

intercoxal or subcapitulargutter) that leads in turn to the preoral chamber inother taxa Taenidia and intercoxal gutters arepresent in Opilioacariformes (Van der Hammen1989) Holothyrida (Van der Hammen 1989) Mesos-tigmata (

Glyptholapis

Van der Hammen 1989)Ricinulei (Pittard amp Mitchell 1972 Van der Hammen1989) Araneae (

Heptathela

Aphonopelma

Hypochi-lus

Shultz unpubl observ also

Porrhothele

(Mygalo-morphae Dipluridae) Butt amp Taylor 1991

Segestria

(Araneomorphae Segestriidae) Van der Hammen

1989) Amblypygi (Van der Hammen 1989) Schizo-mida (

Stenochrus

Shultz unpubl observ unidenti-fied Hubbardiinae Van der Hammen 1989) andThelyphonida (

Mastigoproctus

Shultz unpublobserv also

Tetrabalius

Van der Hammen 1989) Theconduction of large volumes of fluid from orifices of thecoxal organs to the preoral chamber has been observedin the mygalomorph spider

Porrhothele

(Dipluridae)(Butt amp Taylor 1991) and

Mastigoproctus

(Thely-phonida) (Shultz unpubl observ) The podocephaliccanal (

15

) is present in Acariformes (Acari) (Lindquist1984) and appears to have evolved as an invaginatedsupracoxal channel (Grandjean 1938 G Alberti inEvans 1992) Van der Hammen (1982) could not dis-cern with certainty whether a channel exists in

Eukoenenia

(Palpigradi) but none has been reportedby previous authors and they have not been observedby the present author Palpigradi is coded as havingState 0

15 Podocephalic canal cuticular channel andorduct draining multiple glands and opening nearmouthparts 0 absent 1 present - inapplicable due toabsence of channel (14)

State 1 occurs throughout Acariformes (Acari)(Lindquist 1984)

16 Heavily sclerotized suboral sclerite serving inpart as basal pivot point for coxae of appendages ofpostoral somites IIndashIV (

=

arachnid palp and legs 1 and2) 0 absent 1 present

The sclerite forms the posterior wall of the true mouthand is heavily sclerotized in extant scorpions It is notvisible externally due to the tightly fitting coxae of theanterior prosomal appendages (Shultz 2007) It maycorrespond to the so-called labium (Weygoldt 1998) orlabrum (Kjellesvig-Waering 1986) of certain fossilscorpions (see

13

)

17 Genal angles 0 rounded 1 pointed

Anderson amp Selden (1997) originally coded a pointedgenal angle and genal spine as separate charactersbut they are combined here State 1 is present in manyXiphosura (Anderson amp Selden 1997) and Chas-mataspidida (Dunlop

et al

2001 2004 Dunlop2002a)

A

PPENDAGES

OF

POSTORAL

SOMITE

I

CHELICERAE

18 Cheliceral segmentation 0 three articles 1 twoarticles

[WP

sim

25

+

42 WH

sim

15 GEWB

sim

44]State 0 is present in extant Xiphosura Eurypterida(Clarke amp Ruedemann 1912 Selden 1981) Palpi-gradi (Roewer 1934) Opiliones (Hansen amp Soslashrensen1904) Scorpiones (Hjelle 1990) and throughout non-acariform Acari (Evans 1992) State 1 is presentthroughout known Trigonotarbida Araneae Ambly-pygi Thelyphonida Schizomida Pseudoscorpionesand Solifugae Acariform mites have two unambigu-

248 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ous cheliceral articles but may have a reduced thirdarticle (Van der Hammen 1989 Evans 1992) acari-form mites are thus coded as uncertain Chelicerae arenot known or their condition is unclear in several fos-sil taxa including Xiphosura Chasmataspidida (Dun-lop et al 2001 2004 Dunlop 2002a) Riciniulei(Selden 1992) Plesiosiro (Dunlop 1999) Proschizo-mus (Dunlop amp Horrocks 19951996) and the fossilscorpions Praearcturus and Palaeoscorpius (Kjelles-vig-Waering 1986) Kjellesvig-Waering (1986) consis-tently interpreted chelicerae of fossil scorpions ashaving four articles but this is probably a misinter-pretation of the basal article (see also Stockwell1989) The basal article in extant scorpions consists ofa distal collar of cuticle and a large proximal processand this arrangement would probably appear as twoarticles in compressed specimens In any event thereare at least three cheliceral articles in Stoermeroscor-pio and Proscorpius (Kjellesvig-Waering 1986)19 Terminal cheliceral joint 0 laterally placedbicondylar hinge 1 dorsally placed bicondylar hinge2 ventrally placed bicondylar hinge [WP 25 + 42 Ssim11 + 12 WH sim15 GEWB sim44]State 0 is present in Xiphosura (Shultz 2001)extant Scorpiones (Millot amp Vachon 1949) Opil-iones (Hansen amp Soslashrensen 1904) Palpigradi(Boumlrner 1904 Millot 1949d) Ricinulei (Pittard ampMitchell 1972 Legg 1976) and Eurypterida includ-ing Baltoeurypterus (Selden 1981) and Stylonurus(Clarke amp Ruedemann 1912) State 0 is present inthe fossil scorpions Proscorpius and Stoermeroscor-pio but the state of the chelicerae is unclear in theother fossil scorpions represented here (Kjellesvig-Waering 1986) State 1 is present in Trigonotarbida(Shear et al 1987) mygalomorph Araneae Ambly-pygi Schizomida and Theylphonida (Weygoldt ampPaulus 1979) State 2 is present in Pseudoscorpi-ones (Chamberlin 1931) and Solifugae (Roewer1934) and appears to be a groundplan feature for allmajor groups of mites (eg Van der Hammen 1989Alberti amp Coons 1999) The difference between theorthognathy of mygalomorph Araneae and lsquoplagiog-nathyrsquo mesothele (Liphistius Heptathela) and palae-ocribellate Araneae (Hypochilus) (Kraus amp Kraus1993) is small compared with the variation thatoccurs throughout arachnids and plagiognathy iscoded as State 1 here20 Chelicera articulating with carapace at anterolat-eral pivot 0 absent 1 present [S 13 WH 57 GEWB45]State 1 occurs in Solifugae (Roewer 1934) and Pseu-doscorpiones except Chthonioidea and Feaelloidea(Chamberlin 1931) Giribet et al (2002) coded thecharacter as present in Pseudoscorpiones21 Chelicera pivoting on dorsal protuberance of epis-tome 0 absent 1 present

State 1 appears to be a unique synapomorphy of Opil-iones (Shultz 2000)22 Extrinsic cheliceral muscle arising on carapaceand inserting on dorsal margin of nonbasal cheliceralarticle 0 absent 1 present (Note Characters 22 and23 refer to two different muscles)State 1 is known only in extant Scorpiones (LankesterBenham amp Beck 1885 Vyas 1970 1974 Shultz2007) but may occur in acariform mites if their cheli-cerae are composed of three rather than two articles(Evans 1992) (see 18) Acariformes are coded asuncertain23 Extrinsic cheliceral muscle arising on carapaceand inserting on ventral margin of nonbasal cheliceralarticle 0 absent 1 present [GEWB 252] (Note Char-acters 22 and 23 refer to two different muscles)State 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) somemesostigmatid Acari (Evans 1992) and representa-tive Opiliones particularly Leiobunum (Shultz 2000)Chileogovea and Gonyleptes (Shultz unpubl observ)The status of the character in other opilions is not yetknown The character was originally defined as anextrinsic muscle inserting on the second segment ofthree-segmented chelicerae (Shultz 2000 Giribetet al 2002) but it has been redefined here to make itapplicable to arachnids with two-segmented cheli-cerae24 Extrinsic cheliceral muscles attaching to epistome0 absent 1 presentState 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) but mayoccur in Prostigmata if the sigmoid piece to which thecheliceral protractors attach (Evans 1992) is a modi-fication of the epistome25 Lateral tergocheliceral muscle with three heads 0absent 1 present - coded only for extant tetrapulmo-nates and the palpigrade Eukoenenia homology isunclear in other taxa [GEWB 230]State 1 is present in Pedipalpi (Shultz 1999)26 Cheliceral silk glands 0 absent 1 present [GEWB47]State 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931) The silk gland of spider mites(eg Tetranychus) are associated with the pedipalps(Evans 1992)27 Cheliceral venom glands 0 absent 1 present [WP34 WH 40 GEWB 46]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)28 Cheliceral serrula interior and exterior 0 absent1 presentState 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931 Harvey 1992)29 Cheliceral lsquoflagellumrsquo in male 0 absent 1 present[GEWB 48]

PHYLOGENY OF THE ARACHNID ORDERS 249

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITE II LEG 1 IN NONARACHNIDS AND PALP IN ARACHNIDS

30 Palpal coxae fused ventromedially and formingposterior wall of preoral chamber 0 absent 1 present[WP 27 S 18 WH 17 = 62 GEWB 33 + 63]State 1 occurs in Schizomida Thelyphonida Acari andRicinulei31 Gnathosoma 0 absent 1 present - inapplicablepalpal coxae not fused (30) [WP 56 S 19 WH sim63GEWB sim36 + 37]The gnathosoma is a functional complex comprisingan epistome attached to the dorsal surface of mediallyfused palpal coxae (= infra- or subcapitulum) (see 30)the chelicerae and in some taxa a supracheliceral tec-tum These components are tightly integrated andmove in unison relative to the body The gnathosomahas traditionally been considered a feature of AcariHowever Van der Hammen (1989) argued that a gna-thosoma is present in Ricinulei and this interpreta-tion has received wide acceptance (eg Lindquist1984) However it is not clear that the relationship ofthe chelicerae and subcapitulum is as intimate inRicinulei as in Acari and the ricinuleid subcapitulumdoes not appear to be as moveable In fact the lsquosubca-pitulumrsquo in Ricinulei is similar in many respects to thecondition in Schizomida and Thelyphonida (Hansen ampSoslashrensen 1904) which has never been regarded as agnathosoma Consequently I have coded the conditionof Ricinulei as uncertain32 Rostrosoma long narrow subcylindrical epistomeprojecting anteriorly with base fixed to dorsal surfaceof palpal coxae bordered laterally by lobes projectingfrom palpal coxae ventral wall of preoral chamberformed by anterior element of prosoma (sternapophy-sis) 0 absent 1 present [S 15 WH 59 GEWB 20 + 36]State 1 is present in Pseudoscorpiones and Solifugae(Chamberlin 1931 Roewer 1934 Van der Hammen1989) Dunlop (2000) attempted to homologize com-ponents of the rostrosoma and gnathosoma (31) anargument accepted by Giribet et al (2002) Howeverfusion of the palpal coxae (30) apparently evolvedindependently in the two structures Coxal fusion inthe gnathosoma is complete ventrally and excludesthe suboral sternapophysis (13) while fusion isincomplete ventrally in the rostrosoma and incorpo-rates the sternapophysis Unlike the gnathosomathe rostrosoma is largely immobile and cheliceralmovement is not coupled with that of the palpalcoxae33 Rutellacorniculi hypertrophied setae modified asmouthparts located on the anterior processes of the pal-pal coxae 0 absent 1 present [GEWB 39]

State 1 occurs in the anactinotrichid groups Opilioac-ariformes Holothyrida and Mesostigmata but not Ixo-dida (Van der Hammen 1989) It also occurs inSarcoptiformes (Evans 1992) and certain endeostig-matids (eg Alycus Van der Hammen 1989) but not inProstigmata Comparable structures are unknown inother chelicerate groups34 Terminal segments specially modified in adultmale as a clasper used to engage the female 0 absent1 presentState 1 is known only in extant Xiphosura (YamasakiMakioka amp Saito 1988)35 Robust raptorial 0 absent 1 present [S 17 WH61 GEWB sim59 97]State 1 is present in Amblypygi Thelyphonida andSchizomida (Shear et al 1987) and in Laniatores(Opiliones)36 Orientation of robust raptorial appendage 0operating in subtransverse plane 1 operating in sub-vertical plane - inapplicable coded only for PedipalpiState 0 is present in extant Thelyphonida and mostAmblypygi and State 1 is present in Schizomida (Cok-endolpher amp Reddell 1992) and in the basally diver-gent amblypygid Paracharon caecus (Weygoldt 2000)The orientation of the pedipalps in the fossil Proschi-zomus is uncertain (Dunlop amp Horrocks 19951996)37 Extrinsic muscle attaching to epistome 0 absent1 presentState 1 occurs in Chileogovea and Siro (OpilionesCyphophthalmi) and extant Scorpiones (Vyas 1970Shultz 2007)38 Muscle originating and inserting within coxa 0absent 1 present [GEWB sim232]State 1 occurs is in Amblypygi (Phrynus Shultz 1999Phrynichus Boumlrner 1904) Thelyphonida (Mastigo-proctus Shultz 1993 Thelyphonus Boumlrner 1904) andSchizomida (Stenochrus Shultz unpubl observ)There are two intracoxal muscles in amblypygids andthelyphonids the one coded here and another that isserially homologous with a muscle in the pedal coxae(54)39 Tarsus andor tibia with venom glands 0 absent1 present [GEWB 64]State 1 is known only from extant Pseudoscorpi-ones except the superfamilies Chthonioidea andFeaelloidea (Chamberlin 1931 Harvey 1992) Giri-bet et al (2002) coded this character as present forPseudoscorpiones40 Tarsal grooming organ 0 absent 1 present[GEWB 98]State 1 is a unique synapomorphy of Amblypygi (DelleCave 1975 Weygoldt 2000)41 Scorpionoid chela a large well-developed chelaformed by tibia (manus + fixed finger) and tarsus(moveable finger) 0 absent 1 present [S 16 WH 60GEWB 62]

250 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs throughout Scorpiones andPseudoscorpiones42 Modified in male as copulatory organ 0 absent 1present [WP 35 WH 39 GEWB 67]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)43 Apotele 0 apparently absent not differentiatedexternally from penultimate article 1 present [GEWB69]The apotele exists as a structure (ie dactyl clawempodium) differentiated from the tibiotarsus tarsusor telotarsus and is typically associated with a pair ofantagonistic muscles The apotele occurs as a distinctstructure in Xiphosura (Snodgrass 1948) Eurypter-ida (Clarke amp Ruedemann 1912) Schizomida (Coken-dolpher amp Reddell 1992) Araneae (Foelix 1996)early divergent Amblypygi (Charinus Weygoldt 2000not Phrynus Shultz 1999) Anactinotrichida (CaminClark amp Bourdeau 1956 Lindquist 1984) Ricinulei(Pittard amp Mitchell 1972 Legg 1976) and Solifugae(Roewer 1934) Giribet et al (2002) coded Solifugae asnot having a palpal apotele but it is present as anadhesive organ (45) Dunlop (1999) reconstructedHaptopoda as having a palpal claw but notes in hisdescription that the distal ends were not preserved inthe known specimens44 Apotele (claw) position 0 terminal 1 subtermi-nal - inapplicable coded only for taxa with a distinctapotele (43)State 1 occurs in Parasitiformes (Acari) (Evans 1992)45 Terminal adhesive organ 0 absent 1 present -inapplicable due to lack of an apotele (43) [GEWB 66]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITES IIIndashVI LEGS 2ndash5 IN NON-ARACHNIDS AND LEGS 1ndash4 IN ARACHNIDS

46 Appendage III (= arachnid leg 1) extremely elon-gate antenniform 0 absent 1 present [WP sim24 + 32S 20 WH 14 GEWB sim84]State 1 occurs in Amblypygi Thelyphonida Schizo-mida (Shear et al 1987) and Opilioacariformes (Vander Hammen 1989) State 1 is approximated in Solif-ugae Palpigradi (Roewer 1934) and Haptopoda(Dunlop 1999) and it is coded as uncertain in thesetaxa47 Appendage V (= arachnid leg 3) of male specializedfor sperm transfer 0 absent 1 present [WP 55 GEWB94]State 1 is known only from extant Ricinulei48 Appendages V and VI (= arachnid legs 3 and 4)with femur shorter than patella and with principal siteof flexionextension at patella-tibia joint (lsquoapatellatersquocondition sensu Van der Hammen 1989) 0 absent 1present [S 25 GEWB 86]

State 1 occurs in Pseudoscorpiones and Solifugae(Shultz 1989)49 Appendage VI (= arachnid leg 4) with terminusmodified as flattened blade 0 absent 1 presentState 1 occurs in some Chasmataspidida (DiploapisDunlop et al 2001 Octoberaspis Dunlop 2002a)uncertain in Chasmataspis Dunlop et al 2004 State0 in Loganamaraspis Tetlie amp Braddy 2004) andEurypterida except stylonuroids (Clarke amp Ruede-mann 1912 Waterston 1979 Tollerton 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI COXA AND BODYndashCOXA JOINT

50 Appendage III (= arachnid leg 1) with coxapophy-sis forming floor or wall of preoral chamber 0 absent1 present [S sim14 WH sim58]State 1 occurs in Opiliones (Hansen amp Soslashrensen 1904)and higher scorpions (Jeram 1998) A lobe is presenton the coxa of leg 1 in the fossil scorpion Prearcturusand perhaps Stoermeroscorpio and is coded here as acoxapophysis Coxapophysis of leg 1 together with theepistome and coxapophysis of the palp form a uniquepreoral chamber the stomotheca Weygoldt (1998) andDunlop amp Braddy (2001) reject the homology of cox-apophyses and stomothecae in Opiliones and Scorpi-ones because these structures are not apparent infossils of those scorpions thought to have been aquaticHowever the coxapophyses in Opiliones are formedlargely from soft cuticular lsquolipsrsquo and similar structuresin fossil scorpions would probably not have beenpreserved51 Appendage IV (= arachnid leg 2) with coxapophy-sis 0 absent 1 present [S sim14 WH sim58 GEWB 75]State 1 occurs in extant Scorpiones and most Opil-iones except Dyspnoi (Shultz 1998)52 At least one pair of coxal gnathobases on append-ages III-VI (= arachnid legs 1ndash4) 0 absent 1 present[S sim23 WH sim66 GEWB 78]State 1 occurs in all appropriately preserved Euryp-terida including Baltoeurypterus (Selden 1981) andStylonurus (Clarke amp Ruedemann 1912) extantXiphosura on appendages IIndashVI and apparentlyProscorpius on appendage III (= arachnid leg 1)(Kjellesvig-Waering 1986 Jeram 1998 contra Giribetet al 2002) Fossilized appendages with gnathobaseshave been found in association with Chasmataspis(Chasmataspidida) suggesting that they may havehad State 1 (Dunlop et al 2004)53 Coxae of appendages IIIndashV (= arachnid legs 1ndash3)with jointed moveable endites 0 absent 1 present [Ssim23]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) and in adequately preserved eurypterids(Clarke amp Ruedemann 1912) (BaltoeurypterusSelden 1981)

PHYLOGENY OF THE ARACHNID ORDERS 251

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

54 Intracoxal muscle a muscle arising on anteriorwall of coxa and inserting on posterior wall 0 absent1 present [GEWB 237]State 1 occurs in Thelyphonida (MastigoproctusShultz 1993) and Amblypygi (Phryus Shultz 1999)It is unclear whether the muscle operating the move-able endite (53) in extant Xiphosura (Manton 1964Shultz 2001) is homologous with this muscle and iscoded here as uncertain It is not present in Steno-chrus (Schizomida) (Shultz unpubl observ)55 Coxae of appendages IIndashVI (= arachnid palp andlegs) with dorsal articulation with carapace 0 absent1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)56 Flabellum (exite) on coxa of appendage VI(= arachnid leg 4) 0 absent 1 present [GEWB sim110]State 1 occurs in extant Xiphosura (Yamasaki et al1988) An apparent exite occurs on appendages asso-ciated with Chasmataspis (Chasmataspidida) (Dunlopet al 2004)57 Insertion process of anteromedial tergo-coxal mus-cle 0 weakly developed 1 well to extremely well devel-oped [GEWB 238]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Boumlrner 1904 Shultz 1999)58 Musculi laterales enlarged lateral tergocoxalmuscle with attachment shifted from coxa to adjacentpleural membrane 0 absent 1 present [S 22 WH 65GEWB 242]State 1 occurs in Araneae and Thelyphonida (Shultz1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI TROCHANTER AND COXAndashTROCHANTER JOINT

59 Coxandashtrochanter joint with complex posteriorarticulation composed of two articulating sclerites 0absent 1 present [S 24 WH 67 GEWB 85]State 1 occurs in Araneae Amblypygi Thelyphonidaand Schizomida (Shultz 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI FEMUR AND TROCHANTERndashFEMUR JOINT

60 Depressor muscle (or homologue) of trochanter-femur joint 0 absent 1 present [WH 42 GEWB 220]State 0 appears to be a unique synapomorphy of Ara-neae (Liphistius Shultz 1989 Aphonopelma Ruh-land amp Rathmayer 1978 Hypochilus present studyother examples Clarke 1984 1986)61 Trochanterndashfemur joint with dorsal hinge or pivotoperated by flexor muscles only 0 absent 1 present[WH sim44 GEWB sim91]State 1 occurs in Palpigradi Other chelicerates have abicondylar articulation and are typically equipped

with antagonistic muscles (Van der Hammen 1985Shultz 1989) but see 60 for an exception62 Superior trochanterndashfemur muscle (or homologue)originating broadly in femur inserting on distal mar-gin of trochanter 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 19892001)63 Basifemurndashtelofemur joint of appendages III andIV (= arachnid legs 1 and 2) in adult 0 absent 1present [S 25 WH 68 GEWB 86]The basifemur and telofemur apparently develop aslsquosisterrsquo articles through division of a parental articlethe femur (eg opilioacariform endeostigmatidpalaeosomatid mites) (Coineau amp Van der Hammen1979 Evans 1992) A single animal can have all legswith divided or undivided femora or have some com-bination of the two typically with the more posteriorlegs having the divided femora Thus anterior legsoften tend to be paedomorphic with respect toposterior legs Most chelicerates have undivided fem-ora on some or all legs but muscles homologous withthose of the basifemurndashtelofemur joint may stilldevelop (66) (Shultz 1989) suggesting that the devel-opment of cuticular and muscular components of ajoint have a degree of developmental and evolutionaryindependence

Undivided femora occur in most extant euchelicer-ates including Xiphosura Palpigradi AraneaeAmblypygi Thelyphonida Schizomida ScorpionesOpiliones Pseudoscorpiones higher Sarcoptiformes(Archegozetes Van der Hammen 1989) many Prostig-mata (Tetranychus Evans 1992) and throughout Par-asitiformes Divided femora occur in all legs in manyearly divergent Acariformes including many Endeo-stigmata basally divergent Sarcoptiformes (egPalaeacarus) and Prostigmata (eg AllothrombiumMicrocaeculus) (Van der Hammen 1989 Evans1992) Only leg 4 has basi- and telofemora in Alycus(Van der Hammen 1989) Basi- and telofemora occuron legs 3 and 4 in Solifugae Ricinulei Opilioacari-formes and Eurypterida (Clarke amp Ruedemann 1912Selden 1981 Shultz 1989)

Giribet et al (2002) coded the lsquoannulusrsquo of certainTrigonotarbida (Shear et al 1987) as a basifemur buta similar non-muscularized annulus is present in theposterior legs of extant Thelyphonida In both casesthe condyles of the trochanterndashannulus and annulusndashfemur joints are aligned in such away that they per-form essentially identical movements and appear notto enhance either the range or degree of leg move-ment [This contrasts with the condition in the lasttwo pairs of legs in Baltoeurypterus where condyles ofthe apparent trochanter-basifemur and basifemurndashtelofemur joints have different arrangements (Selden1981)] The annulus in Thelyphonida appears to func-tion in maintaining constant volume of the tro-

252 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

chanterndashfemur joint by preventing ballooning of thearthrodial membrane thereby minimizing the effect ofinternal fluid pressure on movement64 Basifemurndashtelofemur joint of appendage V(= arachnid leg 3) in adult 0 absent 1 present See 6365 Basifemurndashtelofemur joint of appendage VI(= arachnid leg 4) in adult 0 absent 1 present See6366 Cuticular differentiation of basifemur-telofemurjoint absent but muscles present 0 absent 1 present- inapplicable due to presence of joint (63ndash65)State 1 occurs in appendages of postoral somites IIIndashIV (= arachnid legs 1ndash4) in Xiphosura AraneaeAmblypygi Thelyphonida Schizomida Scorpionesand Pseudoscorpiones (Shultz 1989)67 Circumfemoral ring 0 absent 1 presentThe femora of one or more pairs of legs in anactinot-richid Acari have a basal groove or ring associatedwith slit sensilla Acarologists term that portionproximal to the ring the lsquobasifemurrsquo and that portiondistal to the ring the lsquotelofemurrsquo and they refer tothe basifemur as the lsquosecond trochanterrsquo and thetelofemur as the lsquofemurrsquo Despite the confusingterminology the basifemur discussed in 63ndash65 is nothomologous with the lsquobasifemurrsquo of opilioacariformand parasitiform Acari as both a true basifemurndashtelofemur joint and a circumfemoral ring are presentin legs 3 and 4 of Opilioacariformes (Van der Ham-men 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI PATELLA AND FEMURndashPATELLA JOINT

68 Patella of appendage of postoral somite III(= arachnid leg 1) proportionally much longer thanthose of more posterior appendages 0 absent 1presentState 1 occurs in Thelyphonida Schizomida (Hansenamp Soslashrensen 1905 Shultz 1989) and Opilioacari-formes (Van der Hammen 1989)69 Femurndashpatella joint 0 monocondylar several axesof movement and multifunctional muscles 1 bicondy-lar hinge one axis of movement and antagonistic mus-cles 2 hinge one axis of movement flexor muscleswithout muscular antagonists [S sim26 WH sim69 GEWBsim87]State 0 occurs in Solifugae State 1 occurs in Phalang-ida (Opiliones) Scorpiones and PseudoscorpionesState 2 occurs in the remaining taxa (Shultz 1989)70 Patellar plagula 0 absent or with simple medianattachment 1 symmetrical Y-shaped with long prox-imal stem 2 symmetrical U-shaped (= arcuate scler-ite) 3 asymmetrical attaching to patella only atanterior margin [S 27 WH sim70 GEWB 234 = 243]State 1 occurs in extant Xiphosura (Shultz 1989)State 2 occurs in Araneae (Manton 1958 Ruhland amp

Rathmayer 1978 Clarke 1986 Shultz 1989) State 3occurs in Amblypygi Thelyphonida and Schizomida(Shultz 1989) Giribet et al (2002) included this char-acter twice

APPENDAGES OF POSTORAL SOMITES IIIndashVI TIBIA AND PATELLAndashTIBIA JOINT

71 Tibiae divided by one or more joints 0 absent 1presentState 1 occurs in Amblypygi (Weygoldt 2000)72 Patellandashtibia joint 0 monocondylar with or with-out CZY (73) 1 bicondylar hinge one axis of move-ment antagonistic muscles 2 hinge one axis ofmovement muscles without muscular antagonists [Ssim31 WH 74 GEWB sim92] (Shultz 1989 Selden Shearamp Bonamo 1991)73 Patellandashtibia joint with posterior compression zone(lsquoCZYrsquo) 0 absent 1 present - inapplicable due toabsence of monocondylar articulation [GEWB 99]State 1 is a unique synapomorphy of Araneae (Seldenet al 1991)74 Patellandashtibia joint of appendages IIIndashVI(= arachnid legs 1ndash4) with deep-set monocondylar pivotbordered by a pair of tibial processes to which extensormuscles attach 0 absent 1 present - inapplicabledue to absence of monocondylar joint (72)State 1 occurs in extant Xiphosura (Snodgrass 1952Shultz 1989) although a similar arrangement ispresent in the appendage III (= arachnid leg 1) in The-lyphonida (Shultz 1993)75 Patellandashtibia joint largely immobile specializedfor autotomy 0 absent 1 presentState 1 occurs throughout Amblypygi (Weygoldt 19842000)76 Anterior femurndashtibia or femoropatellandashtibia(transpatellar) muscle 0 absent 1 present [S sim30GEWB 222] (Shultz 1989)77 Proximal attachment of posterior femurndashtibia orfemoropatellarndashtibia (transpatellar) muscle 0 muscleabsent 1 dorsoposterior surface of femur andor pos-terior surface of patella 2 distal process of femur mus-cle acting as extensor of femur-patella joint [S sim28 WHsim71 + 72 GEWB 235 = 244 245] (Shultz 1989)78 Distal attachment of posterior femurndashtibia (trans-patellar) muscle 0 posterior 1 dorsal acting as exten-sor of patellandashtibia joint - inapplicable due to absenceof muscle (77) [S 29 GEWB 221] (Shultz 1989)79 Anterior patellandashtibia muscle 0 absent 1 present[S sim32 WH sim73 GEWB 246] (Shultz 1989)80 Posterior patellandashtibia muscle 0 absent 1 present[S 33 WH 76 GEWB 248] (Shultz 1989)81 Patellandashtibia joint spanned by elastic (lsquospringlikersquo)sclerite 0 absent 1 present [GEWB 107]State 1 is a unique synapomorphy of Solifugae (Sens-enig amp Shultz 2003)

PHYLOGENY OF THE ARACHNID ORDERS 253

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

APPENDAGES OF POSTORAL SOMITES IIIndashVI TARSUS AND TIBIAndashTARSUS JOINT

82 Tarsus divided into proximal basitarsus(= metatarsus) and distal telotarsus (= distitarsus orlsquotarsusrsquo) 0 absent 1 presentState 0 occurs in extant Xiphosura (Shultz 1989) andthroughout Acariformes (Lindquist 1984 Evans1992) in the anterior two pairs of legs in chthonioidpseudoscorpions and all legs in lsquomonosphyronidrsquo pseu-doscorpions including Feaelloidea and Cheliferoidea(Chamberlin 1931) A circumtarsal ring (83) associ-ated with slit sensilla in Parasitiformes (Acari) mayrepresent a joint between a telotarsus and basitarsus(Evans 1992)83 Circumtarsal ring 0 absent 1 presentState 1 is an apparent synapomorphy of Parasiti-formes (Evans 1992)84 Telotarsus in adult with two or more tarsomeres 0absent 1 present - inapplicable due to absence of telo-tarsus (82) [GEWB 71]State 1 occurs in Schizomida Thelyphonida Ambly-pygi Palpigradi Ricinulei Solifugae (Shultz 1989)Opilioacariformes (Van der Hammen 1989) and Ple-siosiro (Dunlop 1999)85 Three telotarsomeres on appendages of postoralsomites IV-VI (= arachnid legs 2ndash4) 0 absent 1present - inapplicable due to absence of telotarsus (82)or absence of telotarsomeres (84) [S 34 WH 77 GEWB102]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Shultz 1989)86 Tibiandashtarsus joint spanned by well-developed elas-tic (lsquospringlikersquo) sclerite 0 absent 1 present [GEWB107]State 1 occurs in Scorpiones phalangid Opiliones andSolifugae (Alexander 1967 Shultz 2000 Sensenig ampShultz 2003) Extant Xiphosura is coded for the lastprosomal appendage only (see 88)87 Appendage of postoral somite VI (= arachnid leg 4)with ring of large basally articulated spatulate pro-cesses at tibia-tarsus joint 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)88 Appendages postoral somites IIIndashV (= arachnidlegs 1ndash3) with tibiotarsus (ie tibia and tarsus not dif-ferentiated) 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 1989) butnot apparently in Weinbergina (Moore et al 2005) theonly synxiphosurid in which appendages are suffi-ciently preserved to determine the number ofpodomeres89 Ambulacrum peduncle-like extension of the tarsuswith internal condylophores and terminating distallywith apotele (eg claws) andor pulvillus 0 absent 1present

State 1 is present throughout Acari (Alberti amp Coons1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI APOTELE AND TARSUSndashAPOTELE JOINT

90 Apotele of appendage III (= arachnid leg 1) 0absent or not apparent 1 present [GEWB 101]State 0 occurs in Amblypygi Thelyphonida and Schi-zomida (Shultz 1999) However the apotele is proba-bly present but undifferentiated because muscles thatwould normally insert on the apotele terminate on theend of the lsquotarsusrsquo Dunlop (1999) reconstructed Hap-topoda as lacking an apotelic claw on leg 1 but noted inthe text that this was speculative Dunlop (2002b) hasreviewed apotelic diversity in Chelicerata91 Appendages of postoral somites IIIndashVI (= arachnidlegs) chelate with chela formed from tibiotarsus andapotele or tarsus and apotele 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988) The condition appears to occur in Chasmataspisbased on associated appendages (Dunlop et al 2004)but not in Diploaspis and Octoberaspis (Dunlop et al2001 Dunlop 2002a)92 Apotele with eversible or padlike empodium(= pulvillus) in adult 0 absent although empodialclaw may be present 1 present [S sim38 WH 81 GEWB95]State 1 occurs throughout Pseudoscorpiones (Cham-berlin 1931) Solifugae (Roewer 1934) Opilioacari-formes Holothyrida Ixodida and free-livingMesostigmata (Evans 1992) An eversible lsquopulvillusrsquo isprobably a primitive feature of Amblypygi (eg Char-inus Weygoldt 2000) but is absent in higher groups(eg Phrynus Weygoldt 2000)93 Inferior apotele muscle (= claw depressor) with tib-ial attachment 0 absent 1 present [S 35 WH 78GEWB 249]The character is coded for extant Xiphosura usingappendage of postoral somite VI (= arachnid leg 4) asother legs lack a differentiated tibia (88) State 1appears to occur throughout Arachnida (Shultz 1989)94 Inferior apotele muscle (= claw depressor) withpatellar attachment 0 absent 1 present [S 36 WH 79GEWB 250]State 0 occurs in extant Xiphosura Acari AmblypygiPalpigradi and Ricinulei State 1 occurs in all otherarachnids examined here (Shultz 1989)

OPISTHOSOMA

95 Number of opisthosomal somites in adult 0 five1 eight 2 nine 3 10 4 11 5 12 6 13 [WH sim33GEWB sim190]Chasmataspidids appear to have 13 opisthosomalsomites (Dunlop et al 2004) The synziphosurans

254 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Weinbergina (Moore et al 2005) and Limuloides(Anderson amp Selden 1997) appear to have ten opistho-somal somites Moore et al (2005) found no evidence ofan anterior microtergite in Weinbergina as proposedby Anderson amp Selden (1997) Xiphosurids appear tohave nine opisthosomal somites (Scholl 1977 Ander-son amp Selden 1997 Shultz 2001) Members of the fol-lowing orders have 12 opisthosomal somites Araneae(Millot 1949ac) Amblypygi (Weygoldt 2000) Thely-phonida and Schizomida (Hansen amp Soslashrensen 1905)and Pseudoscorpiones (Chamberlin 1931) Elevensomites are present in Palpigradi Solifugae (Roewer1934) and Trigonotarbida (Gilboarachne Shear et al1987 Palaeocharinus Fayers et al 2004) Plesiosiromay have 11 or 12 somites (Dunlop 1999) Justifica-tion for coding of more controversial taxa is providedbelow

Scorpiones Two principal hypotheses regarding thenumber of opisthosomal somites in scorpions havebeen advocated a 13-somite hypothesis derived fromembryological studies (Brauer 1895 Patten 1912Farley 1999 2005) and a 12-somite hypothesis basedon comparative anatomy of adults (Weygoldt amp Pau-lus 1979) The embryological interpretation is basedon the observation of pregenital genital and pectinalsomites (each with segmental ganglia and paired limbbuds) in early scorpion embryos followed by extremereduction or loss of the pregenital somite in laterembryos According to this view a missing pregenitalsomite should be added to the 12 apparent opisthoso-mal somites of post-embryonic scorpions to achieve afinal number of 13 The anatomy-based hypothesiswas introduced by Weygoldt amp Paulus (1979) whoadvocated a literal interpretation of post-embryonicsegmentation based on opisthosomal tergites Specifi-cally these authors argued that the last pair of dorsalendosternal suspensor muscles of non-scorpionarachnids especially Pedipalpi attach to the first(= pregenital) somite that this condition also occurs inscorpions and that there is no reason to invoke a miss-ing pregenital tergite They proposed that the pectinesbelong to the genital somite and that functional spe-cializations of the nervous system for pectinal functiongive the apparence of an extra neuromere duringembryonic development

I recently dissected the prosoma and anterioropisthosoma of the scorpions Centruroides Hadrurusand Heterometrus and focused on the composition ofthe muscular diaphragm (103) that separates thehaemocoelic compartments of the prosoma andopisthosoma (Shultz 2007) The diaphragm is com-posed of a metameric series of axial muscles fromthree somites the anterior somite corresponds to thelast prosomal somite and the posterior somite corre-sponds to the genital somite The middle elements

insert dorsally along a tranverse tendon attached tothe anterior margin of the first tergite These obser-vations are consistent with the embryological inter-pretation that a pregenital somite is present but itstergite is not expressed It appears likely that the thepregenital somite was compressed longitudinally dur-ing the evolution of the diaphragm I code scorpions ashaving 13 opisthosomal somites

Eurypterida The eurypterid opisthosoma is widelyassumed to have 12 somites (Clarke amp Ruedemann1912 Stoslashrmer 1944) However in a speculative paperon the evolutionary morphology of trilobites and che-licerates Raw (1957) proposed that both scorpions andeurypterids have 15 opisthosomal somites Rawassumed that scorpions have 13 apparent opisthoso-mal somites based on the transient pregenital somiteof scorpion embryology that scorpions and eurypteridsare close relatives and should have the same numberof somites and that olenellid trilobites and chelicer-ates always have somites in multiples of three Rawachieved 15 somites in scorpions by assuming the lastopisthosomal somite to be a diplosomite and that thetelson is a postanal somite even though there is noevidence for either of these proposals He attributedthese features to eurypterids as well He also notedthat the connection between the prosomal carapaceand first opisthosomal tergite in Eurypterida differedstructurally from the connection between adjacentopisthosomal tergites and regarded this as evidencefor a reduced pregenital tergite in the prosomandashopisthosoma junction Rawrsquos speculations were largelyforgotten until Dunlop amp Webster (1999) resurrectedthe proposal that eurypterids have a reduced opistho-somal tergite and therefore share a unique similaritywith scorpions Unfortunately Dunlop amp Bullocktreated Rawrsquos conjecture as if it were based on empir-ical evidence rather than an attempt to force euryp-terid morphology into a peculiar numerical system Inthe absence of convincing evidence to the contrary Ihave coded eurypterids as having 12 opisthosomalsomites

Ricinulei The opisthosoma consists of a membranouspedicel (97) bearing the gonopore ventrally Thefemale gonopore is bordered by an anterior plate and aposterior plate The remainder of the opisthosoma iscomposed of thick sclerites separated by less heavilysclerotized cuticle The first dorsal sclerite is short andfunctions as part of a prosomandashopisthosoma couplingmechanism (96) that may or may not be a specializedcomponent of the following tergite This is followed byfour tergites and sternites and a three-segmentedmetasoma (= lsquopygidiumrsquo) Millot (1945 1949f) rea-soned that the opisthosoma contains ten somites withsomites VIIndashIX incorporated into the pedicel XndashXIIIexpressed as tergites and sternites and the metasoma

PHYLOGENY OF THE ARACHNID ORDERS 255

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

comprising four somites He did not regard the dorsalcoupling sclerite as a separate tergite and his inter-pretation of four rather than three metasomal somiteshas been rejected Pittard amp Mitchell (1972) also pro-posed ten opisthosomal somites but achieved thisnumber by regarding the dorsal coupling sclerite as atergite of somite IX and by recognizing three metaso-mal somites Legg (1976) adopted the system proposedby Pittard amp Mitchell but did not regard the dorsalcoupling sclerite as separate from the following tergite(X) Van der Hammen (1979 1989) reconstructed 13somites The pedicel was regarded as having twosomites (VII + VIII) the following four sets of tergitesand sternites were considered diplosomites (IX + XXI + XII XIII + XIV XV + XVI) and the metasoma hadthree somites (XVIIndashXIX) This interpretation mayhave been biased by Van der Hammenrsquos attempt tounite Ricinulei with anactinotrichid Acari which healso regarded as having a primitive number of 13somites Selden (1992) described a fossil ricinuleidTerpsicroton that shows two pairs of depressions onthe three large premetasomal tergites which con-trasts with the single pair seen in extant species Thisobservation appears to corroborate the diplosomitehypothesis but the evidence does not indicate that thetergite anterior to these is a diplosomite

Dunlop (1996) attempted to homologize theprosomandashopisthosoma coupling mechanisms ofRicinulei and Trigonotarbida a goal that required anovel and rather forced interpretation of the dorsalsclerites He regarded the coupling sclerite as homol-ogous with the first opisthosomal tergite (VII) of trig-onotarbids and then followed Van der Hammenrsquosdiplosomite hypothesis to achieve 12 somites in totalDunloprsquos scheme differs from previous systems in sug-gesting that the pedicel does not contain the dorsalelements of the first and second somites and is in-consistent with Millotrsquos (1945) observation that thepre- and postgenital plates each have dorsoventralmuscles Here I code Ricinulei as having 12 somitesThere are three metasomal somites (XVIndashXVIII)three diplosomites (= six somites) (XndashXV) one cou-pling somite (IX) and two somites in the pedicel (iethe genital and pregenital somites) (VII VIII)

Opiliones Harvestmen have nine opisthosomalsomites and an anal operculum that is traditionallyregarded as the tergite of a tenth somite a tenth ster-nite is lacking (Hansen amp Soslashrensen 1904 Winkler1957) However the anal operculum appears to repre-sent a persistent embryonic telson (Moritz 1957) andis therefore likely to be a postsegmental structurecomparable with the stinger of scorpions flagella ofthelyphonids etc (Shultz 2000)

Acari The number of opisthosomal somites is probe-matic for most mite taxa due to uncertainty about the

location of the prosomandashopisthosoma boundry (5)extensive simplification or loss of metamericallyarranged sclerites and muscle attachments a paucityof developmental studies of engrailed expression andin Acariformes opisthosomal anamorphosis and het-erochronic modification of somite number (Evans1992) Ixodids appear to be the exception developmen-tal studies indicate five somites in the opisthosoma ofticks (Evans 1992)

Some mites retain external evidence of segmenta-tion ndash metameric patterns of furrows muscle attach-ments and slit sensilla ndash and with certainassumptions the number of opisthosomal somites canbe estimated Here it is assumed that the dorsal sur-faces of the last two prosomal somites are present inmites and retain their primitive association with legs3 and 4 (contra Van der Hammen 1989) (see 5 for jus-tification) Given this there appear to be 11 somites inthe opisthosoma of Opilioacariformes a conclusionalso reached by other workers (eg With 1904 Sitni-kova 1978 Klompen 2000) Similar reasoning sug-gests that Alycus has seven opisthosomal somites notnine as advocated by Van der Hammen (1989) Unfor-tunately external evidence is ambiguous for determin-ing the number of somites in other mite lineages andthese are coded as uncertain96 Prosoma-opisthosoma coupling mechanism 0absent 1 present [GEWB sim24]The posterior margin of the carapace and the anteriormargin of the first apparent opisthosomal tergite arespecialized as a coupling mechanism in Ricinulei andTrigonotarbida and Dunlop amp Horrocks (1996) pro-posed that these are synapomorphic for the twogroups However there is uncertainty about thehomology of the anterior opisthosomal somites in Trig-onotarbida and Ricinulei (95)97 Pedicel 0 absent 1 aranean type 2 ricinuleidtype [WP 30 S sim40 WH sim20 GEWB sim126]The body narrows at or near the prosomandashopisthosoma juncture in several arachnid lineages (ieSolifugae Palpigradi Amblypygi Araneae Ricinulei)and this lsquowaistrsquo has often been used as character at theinterordinal level (eg Pocock 1893) However thisinterpretation is rejected here for being subjective andfor uniting non-homologous conditions For exampleAraneae and Amblypygi are often grouped on the basisof a lsquopedicelrsquo yet it is a highly specialized structure inAraneae and its parts are not readily homologizedwith those of Amblypygi In contrast the condition inAmblypygi is a slightly narrower version of the highlymoveable prosomandashopisthosoma juncture in Uropygiwhich is not generally considered a lsquopedicelrsquo Thepedicel in Ricinulei is also unique a weakly sclero-tized stalk containing the genital opening (Cryptocel-lus Pittard amp Mitchell 1972 Ricinoides Legg 1976)See 95

256 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

98 Opisthosoma with three- segmented lsquobucklerrsquo 0absent 1 presentState 1 occurs throughout Chasmatapsidida (Dunlop2002a)99 Thoracetron consolidation of tergites of postgeni-tal somites 0 absent 1 present - inapplicable codedonly for Xiphosura (Anderson amp Selden 1997)100 Fusion of tergites of postoral somites VIII and IX(= opisthosomal somites 2 and 3) only 0 absent 1present [GEWB 146]State 1 appears to be a synapomorphy Trigonotarbida(Shear et al 1987) Ricinuleids also have diplotergites(101) but these appear to encompass a different com-bination of somites (see 95)101 Three diplotergites 0 absent presentState 1 is a unique synapomorphy of Ricinulei (Selden1992) See 95102 Paired opisthosomal defensive glands openingvia ducts on either side of the anus 0 absent 1present [S 46 WH 86 GEWB 122]State 1 occurs in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905)103 Muscular diaphragm separating prosomal andopisthosomal compartments formed by dorsoventralmuscles of postoral somites VIndashVIII and extrinsic mus-cles of leg 4 0 absent 1 presentState 1 is known only in extant scorpions (Lankesteret al 1885 Firstman 1973) The diaphragm is oftentreated as a single structure but recent anatomicalwork (Shultz 2007) has shown that it is a composite ofdorsal endosternal suspensors and extrinsic leg mus-cles (see 95) A diaphragm is also present in Solifugaebut it is located more posteriorly and does not appearto be homologous with that of scorpions (Roewer1934)104 Opisthosomal appendicular chondrites 0absent 1 presentThese cartilage-like columns of mesodermally derivedtissue are associated with each opisthosomal append-age including the chilaria in extant Xiphosura(Patten amp Hazen 1900 Yamasaki et al 1988 Fahr-enbach 1999 Shultz 2001)105 Paired appendages on ventral surface of postoralsomite VII (= opisthosomal somite 1) in adult 0absent 1 present [WP sim13 S 39 WH 82 GEWB 143]State 1 is known in Weinbergina (Moore et al 2005)and extant Xiphosura The bilobed structure of themetastoma in certain eurypterids as well as the cor-responding placement of chilaria in Xiphosura haveled some workers to regard the metastoma as fusedappendages of postoral somite VII (eg Stoslashrmer1955) This hypothesis is problematic given that (i) nospecial explanation is needed to account for bilaterallysymmetrical structures in bilaterally symmetricalorganisms (ii) incorporation of the ventral part of thefirst opisthosomal sternite as a functional element of

the prosoma is typical of arachnids and perhapsEuchelicerata generally and (iii) some workers regardthe metastoma as a sternite (eg Jeram 1998) Thusthe metastoma is coded here as uncertain Dunlop hasproposed that the sternum of scorpions is derivedfrom appendages citing the presence of transient limbbuds in scorpion development (Brauer 1895 Patten1912) but this is not compelling evidence for a persis-tent appendicular contribution to the sternite of theadult106 Megoperculum 0 absent 1 present [S sim41 + 42WH sim83 GEWB 162 + 163]State 1 occurs in Thelyphonida Schizomida Ambly-pygi Araneae (Shultz 1993 1999) Trigonotarbida(Shear et al 1987) and probably Haptopoda (Dunlop1999) and Palpigradi The megoperculum consists ofappendages of opisthosomal somite 2 (= genital somiteor postoral somite VIII) (Shultz 1993 1999 Popadicet al 1998) that have fused medially and displacedthe ventral body wall of the somite anteriorly (oftenrepresented by a small sternite to which dorsoventralmuscles of postoral somite VIII attach) The megoper-culum projects posteriorly to form the ventral surfaceof a pregenital chamber The dorsal surface of thechamber is formed by the ventral body wall of opistho-somal somite 3 (= postoral somite IX) which serves asthe ventral attachment of the paired dorsoventralmuscles of that somite The megoperculum bears book-lungs (where present) and sometimes paired gonopodsthat are probably derived from telopodites A similarbut less developed operculum is present on theopisthosomal somite 3 which may also bear booklungsand eversible vesicles (eg Amblypygi) (see 111) corre-sponding to the booklungs and gonopods of the mego-perculum (Shultz 1999)

I contend here that a megoperculum is present inPalpigradi and that this is particularly evident infemales Specifically a large unpaired lobe projectsposteriorly from the genital somite over the ventroan-terior surface of the first postgenital somite therebyforming a pregenital chamber The dorsoventral mus-cles of the genital somite attach ventrally near theanterior border of postoral somite VIII and those ofpostoral somite IX attach to the upper surface of thepregenital chamber (Rucker 1901 Boumlrner 1902bRoewer 1934 Millot 1949d Van der Hammen 1989)Weygoldt (1998) has questioned this interpretation ofPalpigradi107 Postgenital operculum or lsquosternitersquo 0 sclerotized1 not sclerotized - inapplicable coded only for Pan-tetrapulmonata (Platnick amp Gertsch 1976)State 1 is present in Trigonotarbida Amblypygi The-lyphonida Schizomida and Mesothelae (Araneae) butnot Opisthothelae (Araneae)108 Genital opening of female guarded by four plates(one pregenital one postgenital two laterogenitals)

PHYLOGENY OF THE ARACHNID ORDERS 257

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

genital opening of male guarded by two plates 0absent 1 presentThis an apparent synapomorphy of Holothyrida(Acari) (Van der Hammen 1989)109 Anterior margin of genital opening in male withglands secreting via fusules 0 absent 1 presentState 1 occurs in Araneae as epigastric or lsquoepiandrousrsquoglands (Marples 1967) and in Palpigradi (Condeacute1991b)110 Paired valve-like plates apparently formed fromcomponents of three somites covering triradiate genitalopening 0 absent 1 presentThis is an apparent synapomorphy of Acariformes(Evans 1992 Alberti amp Coons 1999)111 Eversible lsquoappendagesrsquo on the ventral surface ofpostgenital somites 0 absent 1 present [S sim43 WHsim84 GEWB sim127]Paired ventral lsquoappendagesrsquo operated in part byhaemolymph pressure occur in the form of ventralsacs in certain Amblypygi (postoral somite IX) (Wey-goldt 2000) (Charinus Millot 1949b not PhrynusShultz 1999) prokoenenian Palpigradi (postoralsomites XndashXII) (Rucker 1901 Condeacute 1991a) as spin-nerets in Araneae (XndashXI) (Shultz 1987) and as geni-tal papillae in many Acariformes (Alberti amp Coons1999) including Oribatida many Endeostigmata (Aly-cus Van der Hammen 1989) and Prostigmata (Allo-thrombium Saboori amp Kamali 2000 MicrocaeculusEvans 1992) Eversible vesicles similar to those ofamblypygids occur in fossils of the trigonotarbidPalaeocharinus (postoral somite IX) (Fayers et al2004) Two cuticular structures (lsquogenital verrucaersquo)occur anterior to the genital opening in Opilioacari-formes and each covers the opening to a thin-walledinvaginated sac (lsquogenital papillarsquo) of unknown function(Van der Hammen 1989) Van der Hammen proposedthat it is homologous to a genital papilla of Acari-formes but Alberti amp Coons (1999) have questionedthis interpretation on several grounds and refer tothese structures as pregenital capsules112 Opisthosomal silk glands and spinnerets derivedfrom appendages on postoral somites X and XI(= opisthosomal somites 4 and 5) 0 absent 1 present[WP 33 WH 41 GEWB 123 + 142]State 1 is a unique synapomorphy of Araneae (Plat-nick amp Gertsch 1976) although potentially homolo-gous glands and spigot-like setae occur on the ventralopisthosomal surface in Palpigradi (Millot 1943Condeacute 1991a)113 Opisthsosomal spinnerets location 0 near mid-dle of opisthosoma 1 near posterior end of opisthso-som ndash inapplicable coded only for Araneae (Platnickamp Gertsch 1976)114 Anterior medial lsquospinneretsrsquo 0 absent 1 present- inapplicable coded only for Araneae (Platnick ampGertsch 1976)

115 Opisthosomal tergites divided longitudinally intoone median and two lateral plates 0 absent 1 present[GEWB 145]State 1 occurs in non-curculioid Ricinulei (Selden1992) and many Trigonotarbida (Dunlop 1996) Thisfeature or something very like it (eg distinct opistho-somal trilobation) occurs sporadically in severaleuchelicerate groups including Chasmataspis (Chas-mataspidida) (Dunlop et al 2004) most fossil xipho-surans (Anderson amp Selden 1997) and a few derivedEurypterida (especially Mixopteroidea) (Tollerton1989)116 Number of metasomal somites 0 zero 1 two 2three 3 five 4 nine [S sim44 WH sim85 GEWB 128 144]The metasoma is a preanal region comprising multiplesomites that are substantially narrower than the pre-ceding somites or mesosoma Metasomal somitesoften lack pleural membranes and take the form ofsclerotized rings State 1 occurs in TrigonotarbidaState 2 is present in synziphosurid Xiphosura (Ander-son amp Selden 1997) Amblypyi Thelyphonida Schizo-mida and Ricinulei State 3 occurs in Scorpiones andEurypterida and State 4 occurs in Chasmatispidida

Some workers appear to consider a five-segmentedmetasoma as conclusive evidence for the monophylyof a Eurypterida + Scorpiones clade (eg Dunlop ampBraddy 2001) presumably because the character rep-resents a kind of tagmosis and should therefore begiven substantial phylogenetic weight However Wey-goldt (1998) has pointed out that a metasoma contain-ing three somites is probably part of the ground planof Euchelicerata The known diversity of the eucheli-cerate metasoma indicates that this feature can inincrease or decrease its segmental composition in evo-lution or be eliminated entirely without necessarilychanging the total number of opisthosomal somitesConsequently it is problematic to assume that thefive-segmented metasoma has any special immunityto homoplasy117 Postanal structure (telson) 0 absent or not obvi-ously developed 1 presentPostanal structures are considered non-segmentalbecause they are modifications of the embryonicregion posterior to the site of somite addition (embry-onic growth zone) A well-developed postanal struc-ture persists in adults in Xiphosura EurypteridaChasmataspidida Scorpiones Schizomida Thely-phonida and Palpigradi The anal operculum of Opil-iones is generally considered by morphologists to bethe tergite of the tenth opisthosomal or anal somite(eg Hansen amp Soslashrensen 1904) but embryological evi-dence (Moritz 1957) suggests that it is derived fromthe postproliferative zone and thus corresponds to thetelson of other chelicerates (see 95)118 Postanal structure shape 0 caudal spine 1aculeus 2 flagellum 3 anal operculum - inapplica-

258 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ble due to absence of postanal structure (117) [WP sim41S sim45 WH sim24 GEWB sim121 129 + 147]State 0 is present in Xiphosura Eurypterida andChasmataspidida State 1 occurs in Scorpiones State2 occurs in Palpigradi Thelyphonida and SchizomidaState 3 occurs in Opiliones (see 95 117)119 Specialized postanal flagellum in male (see 157ndash159) 0 absent 1 present - inapplicable coded onlyfor taxa with postanal flagellum (118) [GEWB 131]State 1 is a unique synapomorphy of Schizomida(Hansen amp Soslashrensen 1905 Cokendolpher amp Reddell1992 Reddell amp Cokendolpher 1995)

RESPIRATORY SYSTEM

120 Respiratory medium 0 water 1 airAmong the terminal taxa included only fossil scorpi-ons are problematic for this character They are codedhere as uncertain Based on a study of book lungmicrosculpture Scholtz amp Kamenz (2006) have arguedthat arachnids are primitively terrestrial and pulmo-nate (see also Firstman 1973) and have questionedwhether any fossil scorpions were aquatic No positionon this proposal is taken here121 Respiratory lamellae on opisthosomal somite 2(= genital somite postoral somite VIII) 0 absent 1present [WP sim37 S sim51 WH sim22 GWEB sim133ndash137]This character encompasses book gills (= lamellae thatfunction in water) and book lungs (= lamellae thatfunction in air) The distinction between book gills andbook lungs is accommodated here by 120 in combina-tion with 121ndash124 Respiratory lamellae are presenton the genital somite in Trigonotarbida AraneaeAmblypygi Thelyphonida and Schizomida Petrunk-evitch (1949) reconstructed Plesiosiro as having booklungs but Dunlop (1999) could not corroborate thisThis character was used by Dunlop amp Webster (1999)to propose that Xiphosura and Scorpiones are closelyrelated because they both lack respiratory lamellae onthe genital somite Dunlop amp Braddy (2001) alsoargued for the placement of Eurypterida with Xipho-sura and Scorpiones based in part on this characterAt least some eurypterids may have had respiratorylamellae (eg Manning amp Dunlop 1995) but the onlyevidence of their segmental distribution is derivedfrom one specimen (Braddy et al 1999)122 Respiratory lamellae on opisthosomal somite 3(= postoral somite IX) 0 absent 1 present [GEWB136]State 1 occurs in Amblypygi Thelyphonida and Xipho-sura It is a groundplan feature of Araneae (Platnick ampGertsch 1976) and is present in all representativetaxa included here Braddy et al (1999) proposed theexistence of respiratory lamellae on postoral somitesIXndashXII in Eurypterida based on evidence from onespecimen This character is coded as uncertain for the

Eurypterida included here Extant scorpions are codedhere as lacking respiratory lamellae on this somite(= pectinal somite) based on the interpretation of theopisthosomal segmentation discussed in 95123 Respiratory lamellae on opisthosomal somites 4ndash6 (= postoral somites XndashXII) 0 absent 1 presentState 1 is definitely present in extant Xiphosura andScorpiones and may have occurred in Eurypterida124 Respiratory lamellae on opisthosomal somite 7(= postoral somite XIII)State 1 occurs in extant Xiphosura and Scorpionesonly125 Kiemenplatten 0 absent 1 presentThese structures are located on the roof of opercularchambers in Eurypterida (Clarke amp Ruedemann1912) where they take the form of ventrally projectingcones with a distinct cuticular microsculputure (Man-ning amp Dunlop 1995) Dunlop amp Braddy (2001)inferred the existence of Kiemenplatten in all Palaeo-zoic scorpions based on a description and photos of onespecimen of Paraisobuthus duobicarinatus by Kjelles-vig-Waering (1986 pls 16ndash18) The plates depict darkcone-like rearward-pointing denticles distributedwithin a white amorphous material The denticlesappear to lack microsculpture even though the mag-nifications at which the photos were taken (times90ndashtimes330)are comparable with those illustrating the cones ofKiemenplatten (times170ndashtimes350 in Manning amp Dunlop1995 figs 1 2) Given the diversity of cuticular struc-tures present in the atria of booklungs book gills andtracheae the denticles appear to bear no special sim-ilarity to the cones of Kiemenplatten126 Tracheal system 0 absent 1 paired ventral stig-mata on postoral somite VIII (= opisthosomal somite2) 2 paired ventral stigmata on postoral somites IXand X 3 one pair of stigmata opening near legs 3 or 44 paired stigmata associated with chelicerae 5 fourpairs of stigmata on dorsal surface of opisthosoma -inapplicable aquatic (120) [WP 40 sim43 S sim52 + 5354 WH sim23 45 88 GEWB sim138 139]Firstman (1973) and Weygoldt amp Paulus (1979)hypothesized that tracheae are homologous in all tra-cheate arachnids except spiders and subsequentworkers have entertained this hypothesis by includinga character for the presenceabsence of tracheal sys-tems that ignores the diverse arrangement of stig-mata in arachnids (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) However thisapproach assumes that internal tracheal systems areconserved but that tracheal openings (stigmata)appear and disappear on different parts of the bodywith higher evolutionary frequency Here I assumethat stigmata are conserved in evolution and that dif-ferences in their anatomical placement reflect the evo-lution of new tracheae State 1 occurs throughoutOpiliones State 2 occurs throughout Solifugae

PHYLOGENY OF THE ARACHNID ORDERS 259

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

(Roewer 1934) and Pseudoscorpiones (Chamberlin1931) State 3 recognizes the possible homology of stig-mata in Ricinulei Opilioacariformes and Parasiti-formes (Acari) (Lindquist 1984 Van der Hammen1989) Tracheal systems are apparently absent in thegroundplan of Acariformes but Prostigmata typicallyhave tracheae associated with the chelicerae Tra-cheae derived from the posterior pair of book lungs arewidespread in araneomorph Araneae but the booklungs are retained and tracheae are absent in themost basally divergent groups (eg Hypochilus)

BOXndashTRUSS AXIAL MUSCLE SYSTEM (BTAMS)

127 Posterior oblique muscles of BTAMS of postoralsomites IndashVI 0 absent 1 present in one or moresomitesState 0 occurs in Xiphosura and state 1 occurs inPalpigradi Araneae Amblypygi Thelyphonida andSchizomida (Shultz 2001) The condition in AcariRicinulei Opiliones and Solifugae is not known128 Anterior oblique muscles of BTAMS posterior topostoral somite VI 0 absent 1 presentState 1 occurs in Xiphosura which is probably theprimitive condition based on comparison with otherarthropods (Shultz 2001) State 0 occurs in all arach-nids examined thus far129 Ventral attachments of posterior oblique musclesof opisthosomal BTAMS located in prosoma 0 absent1 presentState 1 occurs in extant Xiphosura (Shultz 2001)130 Endosternite fenestrate 0 absent 1 present [S 8WH 54 GEWB 35]State 1 occurs in Thelyphonida and hubbardiid Schi-zomida (Firstman 1973) The condition in protoschi-zomid Schizomida is not known131 Suboral suspensor a tendon that arises from theBTAMS and inserts on the ventral surface of the oralcavity via muscle 0 absent 1 presentState 1 occurs in Palpigradi (Eukoenenia Millot1943) Amblypygi (Shultz 1999) and Thelyphonida(Shultz 1993)132 Perineural vascular membrane in adult 0absent 1 present [WH 28 GEWB 155]Wheeler amp Hayashi (1998) coded extant Xiphosura asunknown although the primary source for this char-acter (Firstman 1973) stated that adult Limulus havea perineural vascular membrane Within Arachnidapresence of a perineural vascular membrane is appar-ently correlated with the presence of tracheae (126)133 Ventral endosternal suspensor attaching on coxaof anteriorly adjacent somite 0 absent 1 present[GEWB 239 253]State 1 occurs in Amblypygi (Shultz 1999) Thely-phonida (Shultz 1993) and Schizomida (Shultzunpubl observ)

134 Posteriormost postoral somite with a pair of dor-soventral muscles 0 VI 1 VII 2 VIII 3 XII 4 XIII5 XIV 6 XV 7 XVIDorsoventral muscles tend to run in a continuousmetameric series beginning in the prosoma Deter-mining the last tergite on which the muscle seriesends is substantially easier than counting especiallygiven frequent anatomical complexities near theprosomandashopisthosoma juncture The dorsoventralmuscle series ends on postoral somite VII inCyphophthalmi (eg Chileogovea) and Laniatores(Opiliones) (Shultz unpubl observ) on postoralsomite VII or VIII in Eupnoi (Leiobunum Shultz2000 unpubl observ) on postoral somite XII in Pal-pigradi (Roewer 1934) Solifugae (Bernard 1896Roewer 1934 Millot amp Vachon 1949) and Araneae(Liphistius Hypochilus Millot 1933 but segmenta-tion reinterpreted here following Shultz 1993 1999)on postoral somite XIII in hubbardiid Schizomida(Cokendolpher amp Reddell 1992) and Scorpiones(Lankester et al 1885) and on postoral somite XIVin extant Xiphosura (Shultz 2001) Thelyphonida(Shultz 1993) protoschizomid Schizomida (Cokendol-pher amp Reddell 1992) and Amblypygi (Shultz 1999)Dunlop (1999) reconstructed Plesiosiro as havingpaired tergal apodemes ending on postoral somite XVbut his figures indicate that the series ends on posto-ral somite XIV The interpretation of opisthosomalsegmentation in Ricinulei coded here (95) indicatesthat the series ends on postoral somite XV in Terpsi-croton and this is treated as the groundplan for theorder The posteriormost dorsoventral muscles inPseudoscorpiones occur on postoral somite XVI(Vachon 1949)

NERVOUS SYSTEM

135 Segmental ganglia 0 consolidated in prosoma1 one or more present in opisthosoma [WH 30 GEWB210 + 216]Adult neuromeres are exclusively prosomal in Ambly-pygi liphistiomorph and araneomorph Araneae Opil-iones Pseudoscorpiones Ricinulei and Acari (Millot1949a)136 Dorsal median eyes 0 absent 1 present [WP14 + 47 S 50 WH sim7 GEWB 1]Median eyes are a groundplan feature of Euchelicer-ata and are retained in extant Xiphosura Eurypteridaand occur in all arachnid orders except PalpigradiSchizomida Ricinulei and Pseudoscorpiones (Paulus1979) Median eyes are also absent in opilioacariformand parasitiform Acari (Lindquist 1984) and perhapsthe fossil whipscorpion Proschizomus (Dunlop amp Hor-rocks 19951996) They are unknown in synziphosu-ran Xiphosura Median eyes are present in a variety ofbasal acariform mites including Endeostigmata (but

260 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

not Alycus) Prostigmata (Microcaeculus) and certainoribatids (Palaeacarus) (Evans 1992 Alberti amp Coons1999) Giribet et al (2002) coded all representativeAcari as lacking median eyes It is not known whetherthe eyes of some cyphophthalmid opilions are medianor lateral evidence from tracheal branching inCyphophthalmus (Janczyk 1956) suggests that theyare median eyes (Shultz amp Pinto da Rocha 2007) andpresence of a tapetum in Stylocellus is consistent withlateral eyes (Shear 1993)137 Retinula cells of dorsal median eyes 0 organizedinto closed rhabdoms 1 organized into network ofrhabdomeres 2 disorganized - inapplicable due toabsence of median eyes (136) [GEWB sim3]State 0 is present in Scorpiones Thelyphonida andAmblypygi State 1 is present in Solifugae and Ara-neae state 2 is present in Xiphosura (Paulus 1979)The retinula cells of median eyes have been studied inseveral prostigmatid Acariformes and are organized ina network in some taxa and in an irregular pattern inothers (Alberti amp Coons 1999) Retinulae in phalangidOpiliones have state 0 proximally and state 2 distally(Schliwa 1979)138 Ventral median eyes 0 absent 1 presentState 1 occurs in early instars of extant Xiphosura(Paulus 1979)139 Lateral eyes 0 absent 1 presentLateral eyes are primitively present in Chelicerataand are absent in Palpigradi and Opiliones (Paulus1979) It is unclear whether the eyes of cyphoph-thalmid opilions are median or lateral (see 136)140 Arrangement and number of lateral eyes 0 com-pound many 1 five or more pairs (includes micro-lenses) 2 three primary pairs (excludes microlenses)3 two pairs 4 one pair - inapplicable due to absenceof lateral eyes (139) [WP sim13 + 18 + 38 + 44 + 52 Ssim49 WH sim10 GEWB sim4 + 5]True compound eyes are present in XiphosuraEurypterida Chasmataspidida and many PalaeozoicScorpiones Among the anactinotrichid Acari Opilioa-cariformes have two (Neocarus) or three (Siamacarus)pairs of lateral eyes (although at least one species ofSiamacarus lacks eyes) the allothryrid Holothyridaand many Ixodida have a single pair of lateral lensesand the remainder apparently lack eyes (Evans1992) Thelyphonida was coded as lsquo12rsquo to reflect fivepairs of lenses comprising three pairs of primarylenses and two pairs of small accessory lenses (notsimply three pairs as coded by Giribet et al 2002)Trigonotarbida is also coded as lsquo12rsquo to reflect threepairs of primary lenses and multiple small accessorylenses Dunlop (1999) illustrated Haptopoda as havingpaired triads of lateral eyes but noted that there wasactually no evidence of this in the fossils The lateraleyes of fossil ricinuleids have two pairs of lenses(Selden 1992) One pair of eyes or eyespots are

present in extant Ricinulei and many Schizomidaalthough five genera of hubbardiid Schizomida have apair of lenses (Reddell amp Cokendolpher 1995) State 1is the groundplan for extant Scorpiones and State 3occurs in the pseudoscorpions Chthonius FeaellaNeobisium (but State 4 in Chelifer) and Solifugae(Paulus 1979)141 Lateral eyes with closed rhabdoms 0 absent 1present - inapplicable due to absence of lateral eyes(139) [WP 21 WH 12 GEWB 6]Presence of closed rhabdoms is probably primitive forChelicerata and is retained in extant Xiphosura andScorpiones Retinula cells form a network of rhab-domeres in other extant chelicerates (Paulus 1979)but these networks can differ substantially in detailWeygoldt (1998) and other workers have given sub-stantial weight to the network character in unitingnon-scorpion arachnids but these authors seem not togrant comparable phylogenetic significance to theanalogous character of the median eyes shared bySolifugae and Araneae (137)142 Slit sensilla 0 absent 1 present [WP 19 S 47WH 11 GEWB 209]State 1 occurs in all arachnid orders except Palpigradi(Shultz 1990) The proposal that lsquoprimitiversquo slit sen-silla were present in Eurypterida (Dunlop amp Braddy1997) appears to be based on the over-interpretation ofa comparatively large notch that occurs in Baltoeu-rypterus at the terminus of the tibia (= podomere 7)(see also Edgecombe et al 2000 Giribet et al 2002)The neural construction of slit sensilla like function-ally similar campaniform sensilla of hexapods is sim-ilar to that of trichoid sensilla (Chapman 1998Klompen 2000) Thus these cuticular stress receptorsmay represent modified bases of sensory setae a viewsupported by the replacement of setae by slit sensilladuring post-embryonic development in some Opilioac-ariformes (Klompen 2000)143 Trichobothria 0 absent 1 present [GEWB 213]Trichobothria are present in extant Scorpiones(Jeram 1998) Pseudoscorpiones (Chamberlin 1931)some endeostigmatid (Alycus Van der Hammen 1989)and prostigmatid Acariformes and most Oribatida(Lindquist 1984) They occur on the ventral surface ofthe pedal femora in the opilioacariform Siamacarus(Leclerc 1989) They are also present in Araneae (Foe-lix 1996) Amblypygi (Weygoldt 2000) Schizomidaand Thelyphonida (Hansen amp Soslashrensen 1905) Theyare apparently absent in all non-arachnid chelicer-ates Solifugae Ricinulei Opiliones (Reissland ampGoumlrner 1985) and Parasitiformes (Acari) (Lindquist1984)144 Tibial trichobothria with 2-1-1-1 pattern onappendages IIIndashVI (= arachnid legs 1ndash4) 0 absent 1present - inapplicable due to absence of trichobothria(143) [S 48 WH 87 GEWB 88]

PHYLOGENY OF THE ARACHNID ORDERS 261

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs only in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905) Note that Shultz (1990)erroneously described this character as a 2-2-1-1 pat-tern and that this error was repeated by Giribet et al(2002) and described as 2-1-1 by Wheeler amp Hayashi(1998)145 Paired trichobothria on dorsal surface ofprosoma 0 absent 1 present - inapplicable due toabsence of trichobothria (143)State 1 is an apparent synapomorphy of Acariformesand is present in all acariforms included in this study(Alberti amp Coons 1999)146 Malleoli 0 absent 1 present [WP 45 GEWB 96]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)147 Pectines 0 absent 1 present [WP 20 WH 38GEWB 120]State 1 is a unique synapomorphy of Scorpiones148 Intercheliceral median organ 0 absent 1present [GEWB 212]State 1 is a unique synapomorphy of Palpigradi(Roewer 1934)149 Tarsal organ on appendage of postoral somite III(= arachnid leg 1) (= Hallerrsquos organ) sensilla containedwithin a cuticular depression on the superior surface ofthe tarsus of appendage III (= arachnid leg 1) 0absent 1 present [GEWB sim100]Klompen (2000) has noted that State 1 occurs in Opil-ioacariformes Parasitiformes (except Mesostigmata)and Ricinulei where it also occurs on leg 2 (Talaricoet al 2005) State 1 occurs on all legs in Araneae andperhaps Scorpiones (Foelix 1985)150 Tarsal organ on appendage of postoral somite IV(= arachnid leg 2) 0 absent 1 present (See 149)

REPRODUCTION

151 Gonads 0 primarily prosomal 1 primarilyopisthosomalState 0 is limited to Xiphosura (Sekiguchi 1988)State 1 occurs throughout Arachnida (Millot1949a)152 Ladder-like opisthosomal gonadsaccessoryglands (see 153) 0 absent 1 present [WH sim37 GEWBsim158]Giribet et al (2002) followed Wheeler amp Hayashi(1998) who followed Clarke (1979) in coding gonads asreticulate (Xiphosura) ladder-like (Scorpiones Thely-phonida Schizomida) or lsquosaccularrsquo (all remainingArachnida) However comparisons between Xipho-sura and Arachnida are problematic given that thexiphosuran gonads are primarily prosomal and thoseof arachnids are primarily opisthosomal (151) Fur-ther the reticulate pattern in Limulus (Xiphosura)but not other extant xiphosurans contains a distinctlyladder-like component The lsquosaccularrsquo state is probably

artificial as it encompasses a wide variety of pairedand unpaired structures153 Male gonads in two distinct parts one producingsperm and another (tubular gland) producing a holo-crine secretion similar to degenerate sperm 0 absent1 presentState 1 occurs in Thelyphonida Schizomida andAmblypygi although the holocrine material is pro-duced by ventral organs in Amblypygi and dorsalorgans in Thelyphonida and Schizomida (Alberti2005)154 Number of gonopores 0 two 1 oneExtant xiphosurans have two small genital openingson the base of the genital telopodite and all extantarachnids have a single opening The condition inEurypterida is not known Clarke amp Ruedemann(1912) located a pair of openings near the base of themedian organ (161) that are the outlets of the lsquohornorgansrsquo but it is unclear whether these are genitalducts or accessory structures Braddy amp Dunlop (1997)have developed numerous speculations about thesestructures and extended their arguments far beyondthe available evidence155 Genital opening (ie gonopore or gonostome)appearing to open in prosomal region (ie between legcoxae or anterior to posterior carapacal margin) 0absent 1 present [WP sim50 WH 26 GEWB 166]The genital opening in Euchelicerata is located onpostoral somite VIII (= opisthosomal somite 2) but ithas shifted anterior to the posterior margin of thecarapace or between the coxae of the last pair of legsin most Scorpiones (but not in PalaeoscorpiusKjellesvig-Waering 1986) and Opiliones The geni-tal opening occurs near or anterior to the last coxaein the opilioacariform and parasitiform Acari repre-sented here It is variable in Acariformes but islocated posterior to the coxae in all representativetaxa156 Ovipositor 0 absent 1 present [WP sim51 S sim60WH sim91 GEWB sim172]An ovipositor with a trilobed terminus is an apparentgroundplan character of Oribatida (Lindquist 1984Alberti amp Coons 1999) An ovipositor is also present inOpilioacariformes and Opiliones (Van der Hammen1989)157 Stalked spermatophore attached to substratum0 absent 1 present [S 57]State 1 occurs in Scorpiones PseudoscorpionesAmblypygi Thelyphonida and Schizomida The mech-anism of sperm transfer is unknown in PalpigradiOpilioacariformes and Holothryrida158 Male turns posterior end toward female duringmating behaviour 0 absent 1 presentState 1 occurs in Amblypygi (Weygoldt 2000) The-lyphonida and Schizomida (Weygoldt amp Paulus1979)

262 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

159 Female grasps male opisthosoma during matingbehaviour 0 absent 1 present [WP 29 S 58 WH 19GEWB 188]State 1 occurs in Thelyphonida and Schizomida (Wey-goldt amp Paulus 1979) Evolution of this behaviour wasprobably facilitated by 158160 Penis 0 absent 1 present [WP sim51 GEWB 167]State 1 is a unique synapomorphy of Opiliones A truepenis occurs in Phalangida A clearly homologousstructure is present in Cyphophthalmi (Opiliones) andapparently functions in depositing a spermatophore inthe femalersquos genital chamber (Karaman 2005) A truepenis may occur in some mites but it does not appearto be a groundplan feature of any major group (Evans1992) The lsquopenisrsquo in Oribatida is really a spermatop-ositor it functions in construction of a spermatophore(Alberti amp Coons 1999)161 Median organ 0 absent 1 presentState 1 occurs throughout Eurypterida (Clarke amp Rue-demann 1912) and has been observed in diploaspididChasmataspidida (Dunlop 2002a also Loganamaras-pis Tetlie amp Braddy 2004)

SPERM MORPHOLOGY

162 Nucleus with manchette of microtubules 0absent 1 present [S 54 WH 89 GEWB 193]State 1 occurs in Araneae Amblypygi Thely-phonida Schizomida Ricinulei and gonyleptid Lani-atores (Opiliones) (Alberti 1995 Giribet et al2002)163 Axoneme 0 absent 1 present [WP sim49 S sim55WH sim25 GEWB 195]State 1 occurs in Xiphosura Scorpiones Cyphoph-thalmi (Opiliones) Pseudoscorpiones RicinuleiAraneae Amblypygi Thelyphonida and Schizomida(Alberti 1995)164 Coiled axoneme 0 absent 1 present - inappli-cable due to absence of axoneme (163) [WP 22 S sim55WH 13 GEWB 196]State 1 occurs in Pseudoscorpiones Ricinulei Ara-neae Amblypygi Thelyphonida and Schizomida(Alberti 1995)165 Microtubule arrangement in axoneme 0 9 + 0 19 + 1 2 9 + 2 3 9 + 3 - inapplicable due to absence ofaxoneme (163) [WP sim26 S sim56 WH sim16 GEWB 198](Alberti 1995)166 Helical or corkscrew shaped nucleus 0 absent 1present [GEWB sim204]State 1 occurs in Araneae Amblypygi ThelyphonidaSchizomida and certain Scorpiones (eg Hadrurus)(Alberti 1995)167 Vacuolated-type sperm 0 absent 1 present[GEWB 205]State 1 is unique to Opilioacariformes and Parasiti-formes (Acari) (Alberti 1995)

168 Sperm aggregates 0 absent 1 present [GEWB206] (Alberti 1995)

DEVELOPMENT

169 Yolk in early embryo 0 concentrated (centroleci-thal or telolecithal) 1 evenly distributed (isolecithal)[WH sim34 GEWB sim191]Wheeler amp Hayashi (1998) and Giribet et al (2002)coded this character as lsquo0 isolecithal or telolecithal 1centrolecithalrsquo based on information presented byYoshikura (1975) Their coding appears to highlightseparate conditions found in Scorpiones (ie isoleci-thal and telolecithal) rather than a property intrinsicto the character itself170 Embryonic nutrition other than yolk 0 absent 1present [WP sim49]State 1 occurs in extant Scorpiones (except buthidsand chaerilids) and Pseudoscorpiones (Weygoldt1969)171 Embryological growth zone 0 initiating segmentaddition within prosoma 1 initiating segment addi-tion posterior to prosoma [GEWB sim192]Giribet et al (2002) followed Dunlop amp Webster (1999)in stating that only Xiphosura and Scorpiones havestate 0 among Chelicerata However the last prosomalsomite develops from the growth zone in Ixodida(Anderson 1973 Evans 1992) A similar processapparently occurs in the acariform mites Tyroglyphus(Sarcoptiformes Astigmata) (Yoshikura 1975) andArchegozetes (Sarcoptiformes Oribatida) (Thomas ampTelford 1999) but not apparently in Tetranychus(Prostigmata) (Dearden Donly amp Grbic 2002)172 Eggsembryos maintained in external attachedbrood sac secreted by genital glands 0 absent 1present [WP sim46 S 59 WH 90 GEWB 208 = 219]State 1 occurs throughout Amblypygi SchizomidaThelyphonida and Pseudoscorpions Giribet et al(2002) followed Shear et al (1987) in homologizingspinneret-derived silken egg sacs of spiders with gen-ital-gland-derived brood sacs of Pedipalpi However asthese structures are not homologous in secretory ori-gin construction or composition (Shultz 1987) theyare not regarded here as homologous Giribet et al(2002) miscoded the character as absent in Pseudo-scorpiones and without justification in PalpigradiFemale palpigrades have well-developed glands asso-ciated with postgenital somites and an array of setaefusules and other structures (112) (Boumlrner 1902b Mil-lot 1942 Condeacute 1991a) which are consistent withsome sort of brood care and Palpigradi should there-fore be coded as unknown for this character173 Embryonic and early postembryonic lsquolateralrsquo orClaparegravede organs associated with coxa of postoralsomite IV (= arachnid leg 2) 0 absent 1 present [WHsim31 GEWB sim211]

PHYLOGENY OF THE ARACHNID ORDERS 263

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bilaterally paired rounded protuberances locatedbetween coxae of appendages of postoral somites IIIand IV (arachnid legs 1 and 2) in prelarval and larvalinstars appear to be a primitive feature of Acariformes(Lindquist 1984 Evans 1992 Alberti amp Coons 1999)It is known to be derived embryologically from thecoxa of leg 2 in the oribatid Archegozetes (Thomas ampTelford 1999) Structures apparently homologouswith the acariform Claparegravede organ are present inembryonic Amblypygi (Weygoldt 2000) and Thely-phonida (Yoshikura 1975) and in embryonic and earlypost-embryonic stages of Solifugae (Roewer 1934)where they are termed lateral organs The so-calledlsquolateral organrsquo of embryonic Xiphosura (174) occurs onthe lateral surface of the carapace and does not appearto be homologous with the coxa-associated structuressharing their name In fact the xiphosuran lateralorgan develops even when the coxa of postoral somiteIV has been excised (Sekiguchi 1988) In contrastWheeler amp Hayashi (1998) and Giribet et al (2002)followed Yoshikura (1975) in regarding all lsquolateralorgansrsquo as homologous and in miscoding this characteras absent in Acariformes Van der Hammen (1989)suggested that the sternal verrucae of Opilioacari-formes are homologous with the Claparegravede organ butit differs in gross structure and ontogenetic timing174 Embryonic lateral organ associated with cara-pace 0 absent 1 present [WH sim31 GEWB sim211]State 1 is known only from embryonic stages of extantXiphosura (Yoshikura 1975) In contrast to previousinterpretations this feature does not appear to behomologous with the lsquolateralrsquo or Clarapegravede organ ofcertain arachnids (173)175 Live birth 0 absent 1 presentState 1 occurs in extant Scorpiones and has evolvedseveral times in a few mites (Evans 1992)176 Hexapodal larva and 1ndash3 nymphal stages 0absent 1 present [WP 53 S sim61 WH 27 GEWB 186]State 1 occurs in Acari and Ricinulei (Lindquist1984)177 Hexapodal prelarva 0 absent 1 present [S sim61WH 47]State 1 occurs in Opilioacariformes (NeocarusKlompen 2000) and many Acariformes (Evans 1992)

EXCRETORYOSMOREGULATORY SYSTEM

178 Malpighian tubules 0 absent 1 present [WP 17S 62 WH 9 GEWB 153]State 1 occurs in all major arachnid groups except Pal-pigradi Pseudoscorpiones Opiliones Oribatida (Sar-coptiformes) and Prostigmata Within Acari State 1occurs in all major anactinotrichid lineages (Opilioac-ariformes Holothyrida Mesostigmata Ixodida) andsimilar structures are present in certain Astigmata(Evans 1992 Alberti amp Coons 1999)

179 Dorsomedian excretory organ 0 absent 1presentThis is a specialized excretoryosmoregulatory organformed by the postventriculus and proctodeum inProstigmata (Alberti amp Coons 1999)180 Adult coxal organ opening on or near coxa ofappendage III (= arachnid leg 1) 0 absent 1 present[S 64 WH 93 GEWB 90]State 1 occurs throughout Acari (Evans 1992 Albertiamp Coons 1999) Ricinulei (Pittard amp Mitchell 1972Legg 1976) Palpigradi (Millot 1942) AraneaeAmblypygi Thelyphonida and Schizomida (Buxton1913 1917) There is evidence that the second lsquoozo-porersquo of gonyleptids (Opiliones Laniatores) (Hara ampGnaspini 2003) is a persistent opening to the coxalorgan associated with leg 1 (Soslashrensen 1879) whichalso occurs in embryonic Phalangium (Opiliones Eup-noi) (Moritz 1959)181 Adult coxal organ opening on or near coxa ofappendage V (= arachnid leg 3) 0 absent 1 present [S63 WH 92 GEWB 89]State 1 occurs in extant Xiphosura (Yamasaki et al1988) Scorpiones mygalomorph (Buxton 1913 1917)and mesothele Araneae (Liphistius J Millot in Bris-towe 1932) State 1 occurs in basal Amblypygi(Charon Charinus Buxton 1913 1917) but degener-ates prior to the adult stage in higher groups(Phrynus Weygoldt 2000) Adequately preservedeurypterids show a small submarginal opening on thecoxa of appendage V (Baltoeurypterus Selden 1981also Eurypterus Hughmilleria Clarke amp Ruedemann1912) and this is interpreted here as the opening to acoxal organ Giribet et al (2002) coded this characteras uncertain for Eurypterida182 Adult coxal organ opening on coxa of appendageII (= arachnid palp) 0 absent 1 present [GEWB 65]Buxton (1913 1917) noted that the coxal organ ofSolifugae opens on the prolateral surface of the palpalcoxa The proximal end of the duct leading to the ori-fice is associated with glands and apparently theglandular secretion andor coxal fluid serve as saliva(Alberti 1979) (see also 14 15)183 Genital papillae 0 absent 1 present - inappli-cable coded only for those taxa with eversible lsquoappend-agesrsquo (111)The genital papillae are associated with the genitalopening in postlarval Acariformes (Evans 1992Alberti amp Coons 1999) These structures are probablyserially homologous with the prosomal Claparegravedeorgan (173) and both structures probably function inwater and ion regulation

DIGESTIVE SYSTEM

184 Ingestion 0 solid food 1 primarily liquid foodwith or without preoral digestion [WP sim16 WH sim8GEWB sim218]

264 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Extant xiphosurans ingest solid food and members ofmost extant arachnid orders ingest fluids Opilionesand Acari are the principal exceptions The phalangidOpiliones ingest solids but Shultz (2000 unpublobserv) has shown that the precerebral pharyngealapparatus of Cyphophthalmi is very similar to that ofthe fluid-feeding Scorpiones and differs substantiallyfrom phalangid Opiliones Examination of gut con-tents of the cyphophthalmids Siro acaroides S exilisand Chileogovea oedipus have failed to reveal solidparticles typically found in the guts of phalangids(Shultz unpubl observ) the character is coded asunknown for Cyphophthalmi Within Acari consump-tion of particles has been documented in Opilioacari-formes (Neocarus Van der Hammen 1989unidentified Walter amp Proctor 1998) Holothyridanon-parasitic Sarcoptiformes (Evans 1992 Walter ampProctor 1998 Alberti amp Coons 1999) and many endeo-stigmatids including the nematophagous Alycusroseus (Walter 1988)185 Mouth 0 directed posteroventrally 1 directedanteroventrally [S 9 WH 55 GEWB 28]State 1 occurs throughout Arachnida (Shultz 1990)186 Oral cavity dilated by muscles arising from coxaeand constricted by large circular sphincter 0 absent1 presentState 1 occurs in extant Xiphosura (Manton 1964Scholl 1977 Shultz 2001)187 Palate plate 0 absent 1 present [GEWB 159]State 1 is a unique synapomorphy of Araneae (Dunlop1994)188 Lateral walls of epistome broadly fused to medialwalls of palpal coxae opposite sides connected by well-developed transverse epistomal muscle 0 absent 1presentState 1 occurs throughout Opiliones and Scorpiones(Shultz 2000 unpubl observ)189 Epistome with a pair of lateral arms that projectsposteriorly into the prosoma on either side of the phar-ynx 0 absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)190 Epistome with four pairs of suspensor musclesattaching to the carapace 0 absent 1 presentState 1 occurs in extant Scorpiones (Lankester et al1885 Vyas 1970 Shultz unpubl observ) and appar-ently in Solifugae (Roewer 1934)191 Intercheliceral epipharyngeal sclerite 0 absent1 present [GEWB 31]State 1 occurs in Palpigradi (Boumlrner 1904) Trigo-notarbida (Palaeocharinus Dunlop 1994) Araneae(Firstman 1954 Marples 1983) Amblypygi (Mil-lot 1949b Shultz 1999) Schizomida (Millot1949e) Thelyphonida (Millot 1949e Shultz 1993)and Phalangida (Opiliones) (Shultz 2000 unpublobserv)

192 Epipharyngeal sclerite large projecting posteri-orly 0 absent 1 present - inapplicable due to absenceof sclerite (191) [GEWB 32]State 1 occurs in Amblypygi (Millot 1949b Shultz1999) Schizomida (Millot 1949e) Thelyphonida (Mil-lot 1949e Shultz 1993) some Araneae (Heptathela ndashcoded as Liphistius ndash and Hypochilus Marples 1983not Aphonopelma Firstman 1954) and perhaps Trig-onotarbida (Palaeocharinus Dunlop 1994)193 Dorsal dilator muscle of precerebral pharynxattaching to intercheliceral septum or associatedepipharyngeal sclerite (191) 0 absent 1 presentState 1 occurs in Palpigradi (Millot 1943) Araneae(Marples 1983) Amblypygi Schizomida and Thely-phonida (Millot 1949e Shultz 1993 1999)Extremely fine muscles have been documented insome Scorpiones (Shultz 2007) and in Leiobunum(Opiliones) (Shultz 2000) Lankester et al (1885)reported a muscle in Limulus that would appear tocorrespond to this muscle but the muscle does notexist (Manton 1964 Shultz 2001)194 Dorsal dilator muscle of precerebral pharynxattaching to dorsal surface of prosoma 0 absent 1present [GEWB 228]State 1 occurs in Palpigradi (Roewer 1934) and mostAraneae (Palmgren 1978 Aphonopelma Firstman1973) but not in Heptathela (coded as Liphistius) orHypochilus (Marples 1983)195 Lateral dilator muscle of precerebral pharynxattaching to endosternite 0 absent 1 present [GEWBsim226]State 1 is a unique feature of Araneae (Marples 1983)196 Lateral dilator muscle of precerebral pharynxattaching to lateral surface of epistomal processes 0absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)197 Lateral dilator muscle of precerebral pharynxattaching to medial process of coxa of appendage II(= arachnid palp) 0 absent 1 present [GEWB sim226]State 1 occurs in Amblypygi Thelyphonida and Schi-zomida (Boumlrner 1904 Shultz 1993 1999)198 Dilator muscle of precerebral pharynx andorpreoral cavity attaching to ventral surface of prosoma0 absent 1 present [GEWB 227]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) Araneae (Marples 1983) Palpigradi(Millot 1943) and Solifugae (Roewer 1934)199 Postcerebral pharynx 0 absent 1 present [WPsim31 S sim5 WH sim21 GEWB sim154]The cuticle-lined foregut passes through the centralnervous system in many arthropods That portion pos-terior to the CNS is here termed the postcerebralpharynx regardless of the details of its morphology orfunctional specialization A substantial postcerebralpharynx is known in extant Xiphosura (Lankester

PHYLOGENY OF THE ARACHNID ORDERS 265

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

et al 1885 Manton 1964 Yamasaki et al 1988)Scorpiones (Centruroides Hadrurus HeterometrusShultz 2007 also Androctonus Abd el-Wahab 1952)Solifugae (Millot amp Vachon 1949) Araneae Ambly-pygi Thelyphonida and Schizomida (Millot 1949andasheShultz 1993) Weygoldt amp Paulus (1979) applied theterm lsquopostcerebral pharynxrsquo only to the lsquosucking stom-achrsquo of Araneae and Amblypygi but the sucking stom-ach is here regarded as a complex of three characters199ndash201

200 Dilator muscle of postcerebral pharynx attach-ing to endosternite 0 absent 1 present - inapplicabledue to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Lankester et al1885 Manton 1964 Shultz 2001) Scorpiones (Cen-truroides Hadrurus Heterometrus Shultz 2007 alsoAndroctonus Abd el-Wahab 1952) Araneae (First-

man 1954 Palmgren 1978) Amblypygi (Millot1949b Shultz 1999) and apparently the palpigradeProkoenenia wheeleri (Rucker 1901) but not inEukoenenia miriabilis (Millot 1943 contra Boumlrner1904)201 Dilator muscle of postcerebral pharynx attachingto dorsal surface of prosoma 0 absent 1 present -inapplicable due to absence of postcerebral pharynx(199)State 1 occurs throughout Araneae (Millot 1949cFirstman 1954 Palmgren 1978) and Amblypygi(Charinus Millot 1949b Phrynus Shultz 1999)202 Crop and gizzard 0 absent 1 present - inap-plicable due to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Sekiguchi 1988)but not in extant Arachnida

224

J W SHULTZ

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Palaeocharinus

00000-0000 00100110 0000 -000-00 00100000 0000000 0000002 02--0 100-0001 04100-1 001000 10--110--1 110000 101B1000 0 0 10 11

Gilboarachne

00000-0000 0000110 000 -000-00 000000 0000000 0002 02--0 100-0001 04100-1 001000 0--110--1 110000 101B 100 0 0

Liphistius

00000-0000 0011000110 0000001000 -0000-0000 0110000000 0000000110 00000100220010011011 0100-00001 00115010-0 0000010010 1101000--1 1100001000 00030110120110000011 1001000000 0111310000 1000000101 1001101000 1110100111 10

Aphonopelma

00000-0000 0011000110 0000001000 -0000-0000 0110000000 0000000110 00000100220010011011 0100-00001 00115010-0 0000011010 1110000--1 1100001000 00031110120110000011 1001000000 0111310000 1000000101 1001101000 1011100111 10

Hypochilus

00000-0000 0011000110 0000001000 -0000-0000 0110000000 0000000110 0000010022 0010011011 0100-00001 00115010-0 0000011010 1110000--1 1100001000 0003011012 0110000011 1001000000 0111310000 1000000101 001101000 1110100111 10

Charinus

00100-0000 0011000110 0000100000 -00010001 0010010000 000001011 0000010023 1000101011 0101100000 01105000-0 0000010000 10--020--1 1100001000 1015010012 01100000 1011001100 0111310000 1110000101 1001100000 1110001011 10

Phrynus

00100-0000 0011000110 0000100000 -000100101 0000-10000 0001001011 0000010023 1000101011 0101100000 00105000-0 0000010000 00--020--1 1100001000 1015010012 01100000 1011001100 0111310000 1110000101 00-1100000 1110001011 10

Stenochrus

0010110000 0211000110 000000001 0000110100 0010010000 000000111 00000101230000001011 0101100000 00115000-0 0100010000 00--021211 1000001001 01410-014111000000 1111001110 0111310000 110000101 00-1100000 111000100 00

Protoschizomus

0010110000 0211000110 000000001 00001100 0010010000 0000011 000000123 00000 0101100000 0015000-0 0100010000 00--021211 100000100 050-00- -111000000 111001110 0111310000 110000101 00-1100000 1000100 00

Mastigoproctus

00110-0000 0211000110 0000100001 0000100100 000--10000 0001001111 0000010123 0000001011 0101100000 00115000-0 0100010000 00--021201 1100001001 101511001B 0111000000 1111001110 0111310000 1110000101 00-1100000 1110001010 00

Proschizomus

00100-000 02001 1 0010 0010000 0000 100 05000-0 001000 0--02121 0 00 0 0

Terpsicroton

00000-000 10 - 0- 00000 0000 0110 0 0510-0 100- --120--1 0 50-13 00 0

Poliochera

00000-000 10 -1 0- 00000 0000 0110 0 0510-0 100- --120--1 0 50-13 00 0

Table 1

Continued

PHYLOGENY OF THE ARACHNID ORDERS

225

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Cryptocellus

00000-0000 1211000100 00-00001 0000-000 0010001000 000000001 0001100020 02--000-11 0101000001 00105120-0 100000-000 00--120--1 000003 1500-014 10-000011 10010000 011120000 0010101 00-1100000 0-0000000- --

Ricinoides

00000-0000 1211000100 00-00001 0000-000 0010001000 000000001 0001100020 02--000-11 0101000001 00105120-0 100000-000 00--120--1 000003 1500-014 10-000011 10010000 011120000 0010101 00-1100000 0-0000000- --

Neocarus

00001-0000 0011000020 0000-00001 10100-000 0010010000 00000001 000111120 02--00 0101000011 0114000-0 000000-000 00--000--1 000005 00-013 010--00010 1001110 000--01000 011101 00-0100000 0-0000000- --

Siamacarus

00001-0000 0011000020 0000-00001 1100-000 0010010000 0000000 0011112 02--00 10100011 014000-0 000000-000 00--000--1 000005 00-012 110000010 101110 00--10 01101 -0100000 0-0 --

Australothyrus

01000-0100 0001000020 0000-00001 10100-000 0011000000 0000000 00001020 02--00 010-00011 01000-0 000000-100 00--000--1 000003 00-014 10--00010 1001100 000--01000 0101 00-1100000 0-000 --

Allothyrus

01000-0100 0011000020 0000-00001 10100-000 0011000000 0000000 00001020 02--00 010-00011 01000-0 000000-100 00--000--1 000003 00-00- -10--00010 101100 00--10 0101 00-1100000 0-0 --

Glyptholaspis01000-0000 0011000020 000-00001 10100-000 0011000000 000000001 0000001020 02--011011 010-00011 0110000-0 000000-000 00--000--1 00000300 00-00- -10--00000 1001100000 000--01000 000010101 00-1100000 0-0000000- --

Amblyomma00000-0000 0000-00020 0000-00001 10000-000 000--00000 000000001 0000001020 02--011011 010-00011 01100000-0 000000-000 00--000--1 00000300 100-014 010--00010 1001100000 000--01000 0000010101 00-1100000 0-0000000- --

Argas00000-0000 0000-00020 0000-00001 10000-000 000--00000 000000001 0000001020 02--011011 010-00011 00100000-0 000000-000 00--000--1 00000300 100-014 010--00010 101100000 00--10 0000010101 00-1100000 0-00000- --

Alycus000001000 000110020 000-00001 10100-000 000--00000 000000001 000010020 02--0011 000--00011 00101000-0 000000-001 10--000--1 000000 00-013 110100000 10100 00 0111 0010100000 0-000- --

Allothrombium0000000 000110020 00-00001 10000-000 000--00000 000000001 00111-0020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000004 00-0130110100000 1001001000 000--00000 010011011 0011100000 0-0000000- --

Microcaeculus00000000 000110020 00-00001 10000-000 000--00000 000000001 00111-0020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000004 0120130110100000 10010010 00--0000 01001011 001100000 0-0000000- --

Palaeacarus000011000 040110020 000-00001 10100-000 000--00000 000000001 00111-0020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000000 010 110100000 101011000 00--00 010011001 0010100000 0-00000- --

Table 1 Continued

226 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Archegozetes000011000 040110020 000-00001 10100-000 000--00000 000000001 0000000020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000000 00-00- -110100000 1001011000 000--00000 0010011001 0010100000 0-0000000- --

Cyphophthalmus0000100100 0100-00000 100-00000 -0000-1000 0010000001 1000000001 0000000020 01--010-11 0100-00001 00112000-0 000000-000 00--0013-1 000001000 010100 10--00000 1001110001 001100010 0000000 10-100110 0-0001000- --

Chileogovea0000100100 0100-00000 1010-00000 -0000-1000 0010000001 1000000001 0000000020 01--010-11 0100-00001 00112000-0 000000-000 00--0013-1 000001000 010100 10--00000 1001110001 0000 0000000 10-100110 0-0001000- --

Caddo0000100100 0000-00000 100-00000 -0000-000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 01010100--10--00000 10-1110001 000000 0000000 10-0100110 10001000- --

Leiobunum0000100100 0000-00000 1010-00000 -0000-0000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 010B01C00- -10--00000 1001110001 000--00000 1000000000 10-0100110 101001000- --

Sclerobunus0000000100 0000-00000 100-00000 -0001-000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 010001C00- -10--00000 1001110001 00--00000 1000000000 10-0100110 10001000- --

Gonyleptes0000000100 0000-00000 1010-00000 -0001-000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 010001C00- -10--00000 1001110001 010--00000 100000000 10-0100110 10001000- --

Centruroides0000100000 0100-10000 0111-00000 -0000-1000 100--00001 1000000001 0000010010 01--012110 0100-10001 00116000-0 001000-000 00--0311-1 0011001000 0005110011 1110001011 1101101000 0010110100 0000100100 10-1100111 0-10010011 00

Hadrurus0000100000 0100-10000 0111-00000 -0000-1000 100--00001 1000000001 0000010010 01--012110 0100-10001 00116000-0 001000-000 00--0311-1 0011001000 0005110011 1110001011 1101101000 0010110111 0000100100 10-1100111 0-10010011 00

Heterometrus0000100000 0100-10000 0111-00000 -0000-1000 100--00001 1000000001 0000010010 01--012110 0100-10001 00116000-0 001000-000 00--0311-1 0011001000 00051100111110001011 1101101000 001020111 0000100100 10-1100111 0-10010011 00

Prearcturus000010000 010 -0 -000- 100--001 000 000-0 00- --0311- 0 0

Palaeoscorpius00000000 0D0 -0 -000- 100--0000 00 00000 0 10-0001 06000-0 000-00 --0311- 1 01 0 0

Stoermeroscorpio00000000 00000 -0 -000- 100--0000 000 00000 0 10-0001 06000-0 000-00 --0311- 110 01 1 0

Table 1 Continued

PHYLOGENY OF THE ARACHNID ORDERS 227

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

the Ordovician to the Permian (Tollerton 1989) Phy-logenetic relationships among the eurypterids havenot been rigorously addressed (but see Beall amp Laban-deira 1990) However members of Stylonuroideaappear to be plesiomorphic in retaining relativelyunspecialized legs and were represented by Stylonu-rus (eg Clarke amp Ruedemann 1912 Waterston1979) In the remaining eurypterids the last leg ispaddle-like (Tollerton 1989) The non-stylonuroideurypterids were represented in the matrix by thewell-preserved and well-studied Baltoeurypterus tet-ragonophthalmus especially as described by Selden(1981)

Chasmataspidida Members of this fossil aquaticgroup (OrdovicianndashDevonian six genera) have a four-segmented mesosoma and a nine-segmented meta-soma (Dunlop Poschmann amp Anderson 2001 DunlopAnderson amp Braddy 2004 Dunlop 2002a) Three gen-era were included Chasmataspis from the monotypicfamily Chasmataspididae as well as Diploaspis andOctoberaspis from Diploaspididae

Haptopoda This is a fossil terrestrial group (Carbon-iferous) containing one known species Plesiosiromadeleyi The known specimens have been re-examined by Dunlop (1999)

Palpigradi The extant palpigrades (two families sixgenera ~50 spp) (Harvey 2002) are a poorly studiedgroup of small-bodied terrestrial (some semi-aquatic)arachnids The two most well-studied generaEukoenenia (Eukoeneniidae) especially E mirabilisand Prokoenenia (Prokoeneniidae) especiallyP wheeleri were included in the matrix The oneknown fossil species Paleokoenenia mordax is rela-tively recent (Pliocene) and to the extent known isfairly similar to extant palpigrades (Rowland amp Sis-som 1980) It was not included in the analysis

Trigonotarbida This is a fossil terrestrial order(Upper Silurian to Lower Permian ~ten families ~50spp) The genera coded here were drawn from twoLower Devonian Lagerstatten Gilboarachne from theGilboa Formation (New York) (Shear et al 1987) and

Proscorpius00000000 00000 -0 -000- 100--0000 010 00000 0 10-0001 06000-0 000-00 --0311- 110 01 1 0

Chthonius0000100000 0310-00120 0000-10100 -1000-0000 100--00100 0000000001 0000010010 01--012100 0E00-00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Neobisium0000100000 0310-00121 0000-10100 -1000-0010 100--00100 0000000001 0000010010 01--012100 0100-00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Feaella0000100000 0310-00120 0000-10100 -1000-0000 100--00100 0000000001 0000010010 01--012100 000--00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Chelifer0000100000 0310-00121 0000-10100 -1000-0010 100--00100 0000000001 0000010010 01--012100 000--00001 01115000-0 000000-000 00--000--1 0000021000 01700-014 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Eremocosta0000110000 0310-00121 0000-00010 -1000-0000 001010100 0000000001 00011-0000 02--010-10 1101010001 01114000-0 000000-000 00--000--1 00000200 13111013 010--10000 1001000000 000--00100 1010000100 01-1100001 0-00000110 00

Galeodes0000110000 0310-00121 0000-00010 -1000-0000 001010100 0000000001 00011-0000 02--010-10 1101010001 01114000-0 000000-000 00--000--1 00000200 13111013 010--10000 1001000000 000--00100 1010000100 011100001 0-00000110 00

Table 1 Continued

228 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Palaeocharinus from the Rhynie Chert (Scotland)(Dunlop 1994 Fayers Dunlop amp Trewin 2004)

Araneae (spiders) This is a very large order (110 fam-ilies ~3600 genera ~39 000 extant spp) (Platnick2005) represented here by one genus from each of thethree principal extant lineages (Platnick amp Gertsch1976 Coddington amp Levi 1991 Coddington 2005)Liphistius (Mesothelae) Aphonopelma (Mygalomor-phae) and Hypochilus (Araneomorphae) The sampleis small with respect to known diversity but the basalphylogeny and relevant groundplan states of the orderare well established and states derived from the rep-resented taxa are consistent with them

Amblypygi (whipspiders) This is a small extant ter-restrial order (five families 17 genera 136 spp)(Harvey 2002) and was represented in the matrix bytwo fairly well-studied genera Charinus and Phry-nus Charinus retains several features that appear tobe plesiomorphic for the order (eg adult pedal pul-villi coxal glands associated with leg 3 eversiblevesicles) Weygoldt (1996) has examined the phylog-eny of the order and has provided an important sum-mary of morphology and general biology (Weygoldt2000)

Thelyphonida or Uropygi ss (whipscorpions vinega-roons) Extant whipscorpions (16 genera 106 spp)(Harvey 2002) were represented here by one well-studied species Mastigoproctus giganteus althoughadditional information was drawn from Typopeltis andThelyphonus The morphology of the group is highlyconserved there have been no modern studies ofintraordinal phylogeny A controversial fossil speciesfrom the late Carboniferous Proschizomus (Dunlop ampHorrocks 19951996) was also included but less well-preserved Carboniferous forms were not

Schizomida This small order (two extant families~34 genera ~200 spp) (Harvey 2002) is widelyregarded as the sister group of Thelyphonida Thetaxon sample included one representative from eachextant family Protoschizomus (Protoschizomidae) andStenochrus (Hubbardiidae) especially S portoricen-sis There are three recent (Pliocene) fossil speciesall from the same locality these were not includedhere

Ricinulei The extant ricinuleids (three genera 55spp) (Harvey 2002) were represented by two generaCryptocellus and Ricinoides Detailed studies of grosscuticular anatomy and post-embryonic developmentare available for a representative of each genus (Cryp-tocellus Pittard amp Mitchell 1972 Ricinoides Legg1976) and some information on internal anatomy isavailable for Ricinoides (Millot 1945) and this hasbeen extrapolated to Cryptocellus in the matrix There

are two basic fossil types (Selden 1992) one resem-bling modern taxa and another with a unique opistho-soma that superficially resembles the closed elytra of abeetle (ie curculioids) Two fossils from the formergroup were included Terpsicroton and Poliochera asthey are reasonably well preserved and show impor-tant characters (eg two pairs of eyes or evidence ofopisthosomal diplosegmentation) not expressed inextant forms No curculioid ricinuleids were includedas they appear to offer no additional information rel-evant to resolving ordinal relationships

Opilioacariformes (= Opilioacarida Notostigmata)The opilioacariform mites (nine genera 20 spp) (Har-vey 2002) are generally regarded as plesiomorphicAcari and are fairly conserved in their morphologyThe group was represented by two species Neocarustexanus and Siamacarus withi N texanus has a typi-cal opilioacariform morphology and its external anat-omy has been particularly well studied (eg Van derHammen 1989 Klompen 2000) S withi differs frommost other opilioacariforms in having trichobothriaand three rather than two pairs of lateral eyes (the lat-ter also in Paracarus) features that are potentiallysignificant for assessing ordinal relationships Someinformation on internal morphology of Opilioacaruswas taken from With (1904) and extrapolated to Neo-carus and Siamacarus

Parasitiformes This large group of mites comprisesthree clades Holothyrida Ixodida (ticks) and Mesos-tigmata (= Gamasida) Holothyrida (three familiesfive genera lt 30 spp) (Walter amp Proctor 1999) wasrepresented here by two species Australothyrus ocel-latus and Allothyrus constrictus Ixodids (three fami-lies 22 genera ~860 spp) (Kierans amp Robbins 1999)were represented by Amblyomma and Argas andmesostigmatids (~70 families ~10 000 spp) were rep-resented by Glyptholaspis confusa

Acariformes (= Actinotrichida) This is a very largediverse group of mites that includes the Endeostig-mata Sarcoptiformes and Prostigmata Endeostig-mata is an apparently paraphyletic assemblage of~ten families A morphologically generalized speciesAlycus roseus (Bimichaeliidae) was included herewith character states extracted largely from Van derHammen (1989) Sarcoptiformes (gt 200 families) iseffectively synonymous with Oribatida with Astig-mata being a large derived lsquooribatidrsquo clade (Norton1998 Maraun et al 2004) The group was representedby an early divergent plesiomorphic genus Palaeac-arus (Palaeosomata) and a more derived and well-studied genus Archegozetes (especially A longisetusand A magnus) The Prostigmata (~120 families~7000 spp) (Walter amp Proctor 1999) was representedby two genera Allothrombium and Microcaeculus

PHYLOGENY OF THE ARACHNID ORDERS 229

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Pseudoscorpiones The pseudoscorpions (24 families425 genera ~3200 spp) (Harvey 2002) were repre-sented in the matrix by four terminal genera Chtho-nius (Chthoniidae) was drawn from the basallydivergent superfamily Chthonioidea (Harvey 1992)Feaella (Feaellidae) represented the enigmatic super-family Feaelloidea which has an unusual combinationof characters and may be an early divergent lineageclose to Chthonioidea (Harvey 1992) Neobisium (Neo-bisiidae) is morphologically intermediate between theearly divergent families and the higher families rep-resented by Chelifer (Cheliferidae)

Solifugae (= Solpugida) (sun spiders) The order issmall (~1000 spp) but the absence of modern phylo-genetic treatments precludes meaningful estimates ofgenera and families (Harvey 2002) The taxon sampleincluded two genera one from the New World Eremo-costa and one from the Old World Galeodes

Opiliones (harvestmen) Extant harvestmen (~25 fam-ilies ~500 genera ~6000 spp) are divided into twosuborders Cyphophthalmi and Phalangida Cyphoph-thalmids were represented here by Chileogovea oedi-pus (Petallidae) which is currently under study by theauthor and Cyphophthamus duricorius (Sironidae)which is probably the most well-studied member of thesuborder (eg Janczyk 1956) Phalangida includesthree major groups Laniatores Eupnoi and Dyspnoibut there is disagreement as to whether Dyspnoi is thesister group to Laniatores or to Eupnoi (Shultz ampRegier 2001 Giribet et al 2002) Laniatores was rep-resented by Sclerobunus and Gonyleptes and Eupnoiby Leiobunum and Caddo

Scorpiones (scorpions) The extant scorpions (16 fam-ilies 155 genera 1279 spp) (Fet et al 2000) wererepresented by three taxa Centruroides vittatus(Buthidae) Hadrurus arizonensis (Iuridae) and Heter-ometrus spinifer (Scorpionidae) Buthidae is widelyregarded as the sister group to other extant lineagesand similarities among the represented terminals arelikely to be ground plan features of extant scorpionsgenerally The morphology of fossil scorpions is sub-stantially more diverse (Kjellesvig-Waering 1986)but their phylogenetic relationships are unclear (butsee Jeram 1998) The fossil taxa used in this study(Proscorpius Stoermeroscorpio Palaeoscorpius Pre-arcturus) were chosen for quality of preservation andor presence of a phylogenetically significant constella-tion of characters

CHARACTER CODING

Most character states were determined from directobservation the primary literature and authoritativereviews In some cases states were assigned to termi-

nals based on observations from related species asnoted in the Appendix In the matrix state lsquo-rsquo indicatesthat the character is inapplicable because the taxonlacks a more general character For example if a taxonlacks eyes then special features of the eyes (numberposition retinal configuration etc) are inapplicable tothat taxon A lsquorsquo indicates that the state is unknown oruncertain An entry with multiple states (eg 01 345) should be treated as an ambiguity code not a poly-morphism it indicates that two or more interpreta-tions of homology are applicable and that assignmentwas established analytically by character concor-dance Although lsquo-rsquo lsquorsquo and lsquo01rsquo are analyticallyidentical when applied to binary characters theyprovide information about the empirical status of thecharacter state in specific taxa Characters are citedthroughout the text as italicized numbers in parenthe-ses and are discussed in the Appendix

PHYLOGENETIC ANALYSIS

Phylogenetic analysis of the full data matrix wasperformed using the program Tree Analysis usingNew Technologies (TNT) ver 1 (Goloboff Farris ampNixon 2000) using lsquotraditionalrsquo search based on1000 replicates using TBR branch swapping Resultswere compared to those obtaining using the ratchetalgorithm (Nixon 1999) to determine any differencedue to analytical approach Nodal support for theminimal-length topology was evaluated by boot-strap (Felsenstein 1985) and Bremer support(Bremer 1994) Bootstrap analysis was conducted inTNT and based on 1000 pseudoreplicates each anal-ysed by ten random-addition replicates using TBRbranch swapping A nexus file containing the result-ing 1000 trees was imported into PAUP ver 10(Swofford 2002) to obtain bootstrap frequenciesBremer support was determined in TNT by con-straining specific nodes in the minimal-length topol-ogy and then determining the shortest tree that didnot recover the specified clade The difference inlength between the unconstrained and constrainedminimal-length trees is the Bremer support Theeffect of homoplasy on results was explored by con-ducting implied weights analysis (Goloboff 1993) inTNT Six analyses were conducted each with con-stant of concavity (k) set to a different integer valueof 1ndash6 where 1 is weighted most severely againsthomoplasious characters Each implied weightsanalysis was conducted using lsquotraditionalrsquo searchbased on 1000 replicates using TBR branch swap-ping The same procedures were used in analysing amatrix that included only extant taxa Howevercharacters rendered uninformative by removal offossil taxa (ie 49 98 99 100 113 114 125 161)were excluded prior to analysis

230 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

COMPARISONS OF ALTERNATIVE HYPOTHESES

Phylogenetic hypotheses proposed in the recent liter-ature were also evaluated including those thatattempted to resolve arachnid phylogeny completely(Fig 2) their hypothesized subclades (Fig 3) andhypotheses that proposed only specific nodes (Figs 34) The fully resolved topologies of Weygoldt amp Paulus(1979) Van der Hammen (1989) Shultz (1990)Wheeler amp Hayashi (1998) and Giribet et al (2002)were compared with the optimal topology using theTempleton test (Templeton 1983) as implementedin PAUP Internal relationships of multisampledorders were constrained to match those of the optimaltopology unless the original authors explicitlyfavoured an alternative Node-specific hypotheseswere evaluated by determining the frequency withwhich the node was recovered in bootstrap analysisThey were also evaluated by constraining parsimonyanalysis in TNT to recover the shortest tree contain-ing each specific node and then assessing the effect onrelative tree length and overall resolution takingnote of effects on otherwise stable or well-supportedclades The entire matrix was used to assess the twocases where node-specific hypotheses involved fossiltaxa (Fig 4)

RESULTS

EXTANT TAXA

Unweighted analysis of extant taxa produced two min-imal-length trees (length 383 CI 057) (Fig 5A) withconflicts limited to relationships within Pseudoscorpi-ones Multiply sampled orders were recovered asmonophyletic with high nodal support as measured bybootstrap percentage (BP gt 80) except Acari (BP 30)There were few well-supported interordinal groupsexcept Uropygi (BP 100) Pedipalpi (BP 100) Tetra-pulmonata (BP 93) and ArachnidaXiphosura (BP100) All other interordinal relationships were recov-ered with BP less than 60

Results from implied weights analysis (IWA) indi-cated a significant effect of homoplasy on resolution ofthe deepest interordinal relationships although theinterordinal clades Stomotheca Haplocnemata Acar-omorpha Tetrapulmonata Pedipalpi and Uropygiwere recovered under all values of k IWA with k = 1 (1tree best score = 5124) recovered Megoperculata(= Palpigradi + Tetrapulmonata) and Micrura(= Megoperculata + Acaromorpha) with Micrura beingthe sister group to Haplocnemata (= Pseudoscorpiones+ Solifugae) and with Stomothecata (= Opiliones +Scorpiones) being the sister group to all other arach-nids (Fig 5B) With k = 2 (1 tree best score = 3807)and k = 3 (1 tree best score = 3054) Megoperculatawas recovered and Stomotheca remained the sister

group to other arachnids but the relative positions ofHaplocnemata and Acaromorpha changed (Fig 5C)IWA using k = 4 (1 tree best score = 2559) k = 5 (1tree best score = 2205) and k = 6 (1 tree bestscore = 1938) recovered topologies consistent with theminimal-length topology derived from unweightedcharacters (Fig 5A)

EXTANT AND FOSSIL TAXA

Unweighted analysis of extant and fossil taxa pro-duced 18 minimal-length trees (length 426 CI 056)(Fig 6A) with conflicts limited to relationships withinPseudoscorpiones Scorpiones and Araneae Multiplysampled orders were again recovered as monophyleticwith high nodal support (BP gt 80) except Acari whichwas reconstructed as diphyletic with Anactinotrichidabeing the sister group to Ricinulei (BP 33) Therewere no well-supported interordinal groupings (ieBP gt 80) except Uropygi (BP 89) Pedipalpi (BP 92)and Arachnida (BP 90)

Results from IWA indicated a significant effect ofhomoplasy on resolution of the deepest interordinalrelationships in Arachnida although StomothecataHaplocnemata Acaromorpha Cryptognomae Pantet-rapulmonata Tetrapulmonata Schizotarsata Pedi-palpi and Uropygi were recovered under all values ofk IWA with k = 1 (7 trees best score = 5675) recov-ered Megoperculata and Micrura with Micrura beingthe sister group to Haplocnemata Stomothecata wasnot recovered in the strict consensus (Fig 6B) Withk = 2 (3 trees best score = 4243) IWA recovered thetopology shown in Fig 6C with Palpigradi being thesister group to all other arachnids The strict consen-sus topologies reconstructed using k = 3ndash6 were iden-tical to those obtained with k = 2 (k = 3 3 trees bestscore = 3419 k = 4 3 trees best score = 2874 k = 53 trees best score = 2483 k = 6 3 trees bestscore = 2188)

COMPARISONS OF ALTERNATIVE HYPOTHESES

The hypothesis of Weygoldt amp Paulus (1979) (Fig 2)was 398 steps (15 steps longer than the unconstrainedminimal-length tree) Van der Hammen (1989) was400 steps (17 steps longer) Shultz (1990) was 399steps (six steps longer) Wheeler amp Hayashi (1998)was 391 steps (eight steps longer) and Giribet et al(2002) was 394 steps (11 steps longer) Templetontests conducted at the 005 significance level rejectedthe hypothesis of no difference between the optimaltopology and those of Weygoldt amp Paulus (P = 00002)Van der Hammen (P = 00011) Wheeler amp Hayashi(P = 00018) and Giribet et al (P = 00343) The Shultztopology was not significantly different (P = 01444)from the optimal topology

PHYLOGENY OF THE ARACHNID ORDERS 231

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 2 Fully resolved phylogenetic hypotheses of extant euchelicerate groups proposed in the recent literature Note thesimilarity in the topologies of the parsimony-based analyses by Shultz (1990) Wheeler amp Hayashi (1998) and Giribet et al(2002) The Giribet et al topology is based on neontological data (morphology and molecules) and the original lsquoROOTrsquo maybe an artefact from use of the highly divergent pycnogonids as an outgroup

Xip

hosu

raS

corp

ione

sO

pilio

nes

Ara

neae

Am

blyp

ygi

Uro

pygi

Ric

inul

eiA

nact

inot

richi

daS

olifu

gae

Pse

udos

corp

ione

sP

alpi

gra

diA

carif

orm

es

Xip

hosu

raS

corp

ione

sA

rane

aeA

mbl

ypyg

iU

ropy

giP

alpi

gra

diS

olifu

gae

Pse

udos

corp

ione

sO

pilio

nes

Ric

inul

eiA

cari

Lipoctena

Arachnida

Apulmonata

Holotracheata

Cryptoperculata

Acaromorpha(ldquoAcarinomorphardquo)

Tetrapulmonata(rdquoMegoperculatardquo)

LabellataHaplo-cnemata

Weygoldt amp Paulus (1979)

Labellata

Tetrapulmonata(rdquoArachnidardquo)

Haplocnemata(rdquoApatellatardquo)

Van der Hammen (1989)

Xip

hosu

raP

alpi

gra

diA

rane

aeA

mbl

ypyg

iU

ropy

giR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Pedipalpi Haplocnemata

Novogenuata

Dromopoda

Shultz (1990)

Acaro-morpha

Micrura

Xip

hosu

raP

alpi

gra

diU

ropy

giA

mbl

ypyg

iA

rane

aeR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Labellata Haplocnemata

Novogenuata

Dromopoda

Wheeler amp Hayashi (1998)

Acaro-morpha

Micrura

Xip

hosu

raA

cari

Pal

pig

radi

Ric

inul

eiA

rane

aeA

mbl

ypyg

iU

ropy

giO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Tetrapulmonata

PedipalpiHaplocnemata

Novogenuata

Dromopoda

Giribet et al (2002)

ROOT

Micrura

Crypto-gnomae

Sternocoxata

Epimerata

RostrosomataMyliosomata

Neosternata

232 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 3 Consensus trees produced by parsimony analysis of neontological data constrained to produce relationships pro-posed in the recent literature The constrained (target) node is indicated by a black dot and the taxa encompassed by theconstraint are enclosed in a box Numbers below each tree represent the number of minimal-length constrained treeslength of minimal-length trees difference in the length of the unconstrained minimal-length tree and the constrained min-imal-length tree and percentage unconstrained bootstrap trees in which the target node was recovered respectively Thesetrees indicate the effect on branch length imposed by specific hypotheses and impact of constraining target nodes on non-target nodes

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Lipoctena4 385 2 lt5

Xi

Sc

Op

Ps

So

Pa

Ur

Am

Ar

Ri

An

Ac

Labellata6 397 14 lt5

Xi

Sc

Ar

Am

Ur

Op

Pa

Ps

So

Ri

An

Ac

Apulmonata2 390 7 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

So

Ri

An

Ac

Holotracheata5 390 7 lt5

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptoperculata2 388 5 lt5

Xi

Sc

Op

Pa

So

Ps

Ar

Am

Ur

Ri

An

Ac

Rostrosomata6 384 1 25

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Neosternata3 388 5 lt5

Xi

Sc

Op

Ar

Am

Ur

Ri

An

Ps

So

Pa

Ac

Coxisternata2 388 5 lt5

Epimerata7 387 4 11

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptognomae6 384 1 24

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Micrura3 385 2 7

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Megoperculata3 384 1 28

Xi

Sc

Ps

So

Op

Pa

Ar

Am

Ur

Ri

An

Ac

Novogenuata10 388 5 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

Ri

So

An

Ac

Solifugae + Acari4 388 5 lt5

Shultz (1990) Alberti amp Peretti (2002)

Van der Hammen (1989)

Weygoldt amp Paulus (1979)

Xi

Ar

Am

Ur

So

Ps

Sc

Op

Ri

Pa

An

Ac

PHYLOGENY OF THE ARACHNID ORDERS 233

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Results from analyses of node-specific hypothesesare summarized in Figures 3 4 In most cases pro-posed relationships were recovered in fewer than 5of bootstrap pseudoreplicates The notable exceptionswere Rostrosomata (25) Cryptognomae (24) andMegoperculata (28) and the strict-consensus con-strained minimal-length tree for each was only onestep longer than the unconstrained optimal tree Inmost cases constrained topologies did not add manysteps most were less than five steps longer and nonewas greater than 14 steps longer than the un-constrained minimal-length tree It is noteworthyhowever that even constrained nodes that imposerelatively few extra steps sometimes supportedimprobable relationships or eliminated clades thatwere well supported or stable in the unconstrainedanalysis Specific examples and their implications arediscussed below

DISCUSSION

RECENT ISSUES IN ARACHNID PHYLOGENY

Available morphological evidence consistently re-solves some interordinal relationships and fails to

resolve others (Fig 1) Continued progress depends onthe ability of arachnologists to discover new charac-ters and to assess the evidence critically Whilereviewing the recent arachnological literature severalaspects of phylogenetic practice emerged that seemedcounterproductive to both the perception and theactual rate of progress toward resolving arachnid phy-logeny Some of these are summarized here withspecific examples given in the remainder of theDiscussion and in the Appendix

There is a tendency to portray arachnid ordinal phy-logeny as more poorly resolved and contentious than isactually the case (Coddington et al 2004) Phyloge-netic hypotheses generated during different historicalperiods and using differing standards of evidence areoften cited as examples of current disagreement (egSelden 1993 Dunlop 1996 Selden amp Dunlop 1998Wheeler amp Hayashi 1998) In fact recent parsimony-based analyses of morphology have tended to convergeon topologies with internal structures congruent withthose found here (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) (Fig 2) Arachnidphylogeny is not fully resolved especially at its deep-est levels (Fig 1) but this does not mean that allaspects of arachnid phylogeny are controversial orpoorly supported by the available evidence

Matrices are sometimes constructed by uncriticallsquorecyclingrsquo of erroneous or problematic charactersbased on diverse secondary sources (Jenner 2001)Conclusions derived from mixtures of valid invalidspeculative and redundant characters are sometimesportrayed as the phylogenetic signal provided by mor-phology Data recycling can perpetuate errors (see 1320 39 52 136 140 144 152 169 171ndash174) legiti-mize speculations (see 13 32 63 95) or create dupli-cate or non-independent characters (see 13 30 70 77172) The lsquolateral organrsquo is a notable exampleYoshikura (1975) equated the embryonicearly post-embryonic lsquolateral organsrsquo of Amblypygi Thelyphonidaand Solifugae with the dissimilar lsquolateral organrsquo ofXiphosura but failed to note the very similar Cla-paregravede organ of Acariformes (see 173 174) This cod-ing was recycled by Wheeler amp Hayashi (1998) andthen by Giribet et al (2002) Error is probably inevi-table when assembling a large matrix from morphol-ogy including the one presented here but this can beminimized by making original observations and byconsulting primary sources

Some workers advocate weighting characters a pri-ori on the basis of structural or functional complexity(eg Kraus 1998 Dunlop amp Braddy 2001) and dismissphylogenetic conclusions derived from equal-weightsparsimony In short these workers criticize parsimonyfor emphasizing data quantity over data quality Thiscriticism ignores the intense debate in systematic biol-ogy that eventually led to widespread adoption of

Figure 4 Consensus trees produced by parsimony analy-sis of the full data matrix constrained to produce relation-ships proposed in the recent literature See legend toFigure 3 for details

Xi

Eu

Sc

Op

Pa

Ac

Ri

An

So

Ps

Tr

Ar

Ha

Am

UrEurypterida slat + Scorpiones12 435 9 lt5

Xi

Eu

Sc

Op

Pa

Ac

An

So

Ps

Ri

Tr

Ar

Ha

Am

UrTrigonotarbida+ Ricinulei28 434 8 lt5

Dunlop amp Braddy (2002)et al

Dunlop (1996)

234 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 5 Results from analysis of neontological data A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology AFor B and C relationships within terminal clades are the same as those shown in A

LimulusTachypleusProkoeneniaEukoeneniaCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAllothyrusAustralothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesLiphistiusAphonopelmaHypochilusCharinusPhrynusStenochrusProtoschizomusMastigoproctusCyphophthalmusChileogoveaCaddoLeiobunumSclerobunusGonyleptesCentruroidesHadrurusHeterometrusChthoniusFeaellaNeobisiumCheliferEremocostaGaleodes

10031

10018451

501561

733

723

944

633

10010

832

901

943

903

522

341

1007

1008

Acaromorpha

Haplocnemata

10012

Stomothecata

Opiliones

Scorpiones

Pseudo-scorpiones

Solifugae

Acariformes

Parasitiformes

Opilioacariformes

Ricinulei

Xiphosura

Palpigradi

Thelyphonida

Schizomida

Araneae

Amblypygi

10011

936853

100141005

1009

Tetrapulmonata Pedipalpi

Uropygi

692

522572

968

331

603

81

511

161

101

Acari

Dromopoda

Arachnida

A

Xiphosura

Stomothecata

Haplocnemata

Acaromorpha

Palpigradi

TetrapulmonataMicrura

Megoperculata

BXiphosura

Stomothecata

Acaromorpha

Haplocnemata

Palpigradi

TetrapulmonataMegoperculata

C

PHYLOGENY OF THE ARACHNID ORDERS 235

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 6 Results from analysis of the full data matrix A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology CFor B and C relationships within terminal clades are the same as those shown in A

WeinberginaLimuloidesEuproopsTachypleusLimulusStylonurusBaltoeurypterusChasmataspisDiploaspisOctoberaspisCyphophthamusChileogoveaCaddoLeiobunumSclerobunusGonyleptesProscorpiusPalaeoscorpiusStoermeroscorpioPrearcturusCentruroidesHadrurusHeterometrusProkoeneniaEukoeneniaTerpsicrotonPoliocheraCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAustralothyrusAllothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesEremocostaGaleodesChthoniusFeaellaNeobisiumCheliferGilboarachnePalaeocharinusLiphistiusAphonopelmaHypochilusPlesiosiroCharinusPhrynusStenochrusProtoschizomusProschizomusMastigoproctus

631

893

923

984

984

671

341

671

986

10011

423

974

Xiphosura

Eurypterida sstr

Chasmataspidida

Opiliones

Scorpiones

Palpigradi

Ricinulei

Opilioacariformes

Parasitiformes

Acariformes

Solifugae

Pseudoscorpiones

Trigonotarbida

Araneae

Haptopoda

Amblypygi

Schizomida

Thelyphonida

842

251331

361311

552

832

882

925

913

573

561291

443

976

541

954

743644

732521

571

331

833

663591

966

573

71

221

905

832

351

Uropygi

Pedipalpi

Haplocnemata

Tetrapulmonata

Anactinotrichida

Cryptognomae

Acaromorpha

Stomothecata

Pantetrapulmonata

Arachnida

Eurypterida slat

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatProscorpiusStoermeroscorpioPalaeoscorpiusPrearcturusCentruroidesHadrurusHeterometrusOpilionesHaplocnemataRicinuleiAnactinotrichidaAcariformesPalpigradiPantetrapulmonata

Megoperculata

Micrura

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatPalpigradiStomothecataHaplocnemataPantetrapulmonataAcaromorpha

A

B

C

Schizotarsata

236 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

parsimony and instead advocates a return to thespeculative and subjective approaches of the late 19thand early 20th century that once threatened thescientific legitimacy of the discipline (Bowler 1996)Furthermore it misrepresents the properties of parsi-mony-based analyses Specifically characters withfunctional significance are readily encompassed byparsimony analyses and in fact characters derivedfrom locomotor systems (eg 46ndash94) have played animportant role in developing current ideas aboutarachnid phylogeny (eg Shultz 1989 1991) In addi-tion lsquocomplex charactersrsquo can be viewed as compositesof several characters such that morphological com-plexity is effectively weighted by the number of inde-pendent lsquosubcharactersrsquo it contains For example thelsquosucking stomachrsquo once considered a synapomorphy ofLabellata (= Araneae + Amblypygi Fig 3) is codedhere as a composite of three characters (199ndash201) TheLabellata hypothesis was not corroborated in thepresent analysis (Figs 1 3ndash5) and in fact was highlydisfavoured (Fig 3) but this result cannot be dis-missed as a failure to acknowledge the complexity ofthe character

Some workers support specific (target) clades withone or more similarities without exploring the impacton overall homoplasy or relationships among non-target clades Each character offers its own phyloge-netic hypothesis which may or may not be consistentwith relationships implied by other characters It isexceedingly rare for all characters to be perfectly com-patible in the phylogenetic hypotheses they supportand criteria such as parsimony have been developed todiscover those hypotheses that minimize the conflict-ing phylogenetic signals of different characters Eventhough character conflict (homoplasy) is a virtuallyinescapable phenomenon in comparative biology it isnot uncommon for workers to discover one or morecharacters and to promote their phylogeneticimplications over alternative hypotheses even thosethat otherwise appear to be well supported Thisapproach may have value in highlighting new dataand perspectives but accomplishes this by promotingthe erroneous impression that all aspects of arachnidphylogeny are so tenuous that a single character canfalsify even well-supported hypotheses Several exam-ples of this approach have appeared in the recentarachnological literature

For example Alberti amp Peretti (2002) argued thataflagellate spermatozoa (163) are a compelling syna-pomorphy for a Solifugae + Acari clade and dismissedsome of the characters that support Solifugae + Pseu-doscorpiones as having lsquodebatable valuersquo Yet theirproposal rejects two hypotheses that are consistentlyrecovered in recent phylogenetic analyses that isHaplocnemata (= Solifugae + Pseudoscorpiones) andAcaromorpha (= Acari + Ricinulei) (Figs 1 2) Rejec-

tion of Haplocnemata would require its presumed syn-apomorphies to be reinterpreted as homoplasiesincluding features of the chelicerae (18 19 20) pre-oral chamber (13 32) legs (12 48) and respiratorysystem (126) The same reason applies to Acaromor-pha and its synapomorphies It is noteworthy that aSolifugae + Acari clade was recovered in the presentanalysis in fewer than 5 of bootstrap pseudorepli-cates and that analyses constrained to recover thisclade were five steps longer than the minimal-lengthtree and favoured a problematic clade unitingRicinulei and Pseudoscorpiones (Fig 3)

In another example Dunlop (1996) proposed a closerelationship between Trigonotarbida and Ricinuleibased on two-segmented chelicerae (18) prosomandashopisthosoma coupling mechanism (96) diplotergites(100 101) and longitudinally divided opisthosomaltergites (115) However phylogenetic analyses con-strained to recover this relationship required eightadditional steps eliminated support for Acari and nec-essarily rejected Acaromorpha and forced its synapo-morphies to be reinterpreted as homoplasies Alberti(2005) proposed an interesting hypothesis for the evo-lution of male gonads in tetrapulmonates (153) butchose to accept Labellata (= Araneae + Amblypygi) indeveloping his argument over the much more well-supported Pedipalpi (= Amblypygi + Uropygi) (Shultz1999) a phylogenetic reconfiguration that was recov-ered here in fewer than 5 of bootstrap pseudorepli-cates increased tree length by a minimum of 14 stepsand resulted in the collapse of Stomothecata and Hap-locnemata (Fig 3) Many other examples can be cited

Promoting or defending a specific phylogenetichypothesis via lists of compatible synapomorphies is acommon but problematic approach By restrictingattention to the states of specific characters at one ortwo target nodes one can easily overlook the unin-tended impact of the hypothesis on phylogenetic sig-nal elsewhere and its effect on non-target clades Anode supported by a long list of synapomorphies mayseem convincing taken in isolation but may becomeless acceptable when its full phylogenetic implicationsare explored

SUMMARY OF INTERORDINAL ARACHNID CLADES RECOVERED IN THIS ANALYSIS

Arachnida Lamarck 1801Analyses consistently recovered Arachnida as a mono-phyletic group with high bootstrap support (Figs 1 45) Possible synapomorphies include the loss of thecarapacal pleural doublure (9) cardiac lobe (10) pedalgnathobases (52) and moveable endites (53) and thegain of aerial respiration (120) and an anteriorly oranteroventrally directed mouth (185) Some tradi-tional synapomorphies such as slit sensilla (142) and

PHYLOGENY OF THE ARACHNID ORDERS 237

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

fluid feeding (184) may have appeared later in arach-nid evolution but this can only be decided once theinternal phylogeny of Arachnida has been determined

Some workers regard many proposed arachnid syn-apomorphies as adaptations to terrestrial life andthereby link the hypothesis of arachnid monophyly tothe hypothesis of a single ancestral aquatic-to-terres-trial transition and arachnid polyphyly to multipletransitions (eg Selden amp Jeram 1989 Dunlop 1997Dunlop amp Webster 1999 Dunlop amp Braddy 2001)Given this line of reasoning the existence of appar-ently aquatic scorpions in the fossil record (Kjellesvig-Waering 1986 Jeram 1998) and the inference thatterrestrialization occurred late in scorpion evolutionthese workers conclude that character states support-ing arachnid monophyly are actually convergencesand do not necessarily support arachnid monophyly

However this approach to assessing phylogenetichypotheses is founded on overly simplistic assump-tions such as the ability of the investigator to discrim-inate unerringly between characters that existexclusively in aquatic organisms (including fossils)from those that occur exclusively in terrestrial organ-isms There also appears to be an assumption thathomoplasy can be generated through parallelism (iemultiple aquatic-to-terrestrial events) but not throughterrestrial-to-aquatic reversals Furthermore anassumed dichotomy between exclusively aquatic andterrestrial life histories in ancestral arachnids is sim-plistic as illustrated by the amphibious life cycles ofbasal vertebrates and pterygote hexapods In factthese examples demonstrate that there is no neces-sary inconsistency in basing a hypothesis of arachnidmonophyly on the derived terrestrial features of anamphibious ancestor whose descendants thencompleted terrestrialization once or several timesindependently or even returned to a fully aquaticexistence Workers who link the frequency and direc-tion of aquaticndashterrestrial transitions to the assess-ment of arachnid phylogeny do so by endowingthemselves with substantially greater insight thanseems prudent by ignoring the huge gaps in ourunderstanding of early arachnid evolution and bydenying to arachnids the evolutionary complexityknown to exist in other groups

Several palaeontologists have been particularlyactive during the past decade in proposing new char-acters with the stated goal of removing Scorpionesfrom Arachnida and erecting a Scorpiones + Eurypter-ida clade (Braddy amp Dunlop 1997 Dunlop amp Braddy1997 Dunlop 1998 Braddy et al 1999 Dunlop ampWebster 1999) Dunlop amp Braddy (2001) recentlysummarized this evidence and conducted aparsimony-based analysis of Xiphosura EurypteridaScorpiones Opiliones and Tetrapulmonata (but notHaplocnemata Acaromorpha or Palpigradi) using 33

morphological characters Their analysis produced atopology congruent with those generated here (Figs 15 6) including recovery of Stomothecata (= Scorpiones+ Opiliones) However they rejected this result as aproduct of lsquoempirical cladisticsrsquo because it gives thesame weight to prosomal characters that supportarachnid monophyly and to selected opisthosomalcharacters that support their favoured Eurypterida +Scorpiones clade namely a five-segmented metasoma(116) suppression of opisthosomal tergite 1 (95) lossof lamellate respiratory organs on the postgenitalsomite (122) Kiemenplatten (125) loss of respiratorylamellae on the genital segment (121) and a lsquonon-stainingrsquo exocuticle (but see Grainge amp Pearson 1966for evidence of this in Opiliones see Appendix for com-ments on the other characters)

A Eurypterida + Scorpiones clade was not favouredin the present analysis (Fig 6) the strict consensus ofminimal-length trees constrained to recover this clade(Fig 4) is nine steps longer than the tree recovered byanalysis without this constraint It is also noteworthythat Opiliones was consistently reconstructed as thesister group to Eurypterida + Scorpiones in the mini-mal-length constrained trees a provocative result thatwas probably unintended and unanticipated by sup-porters of the Eurypterida + Scorpiones Given thatDunlop amp Braddy (2001) reject equal weights parsi-mony as an arbiter of phylogenetic hypotheses theywould presumably dismiss these results as irrelevantto their argument just as they dismissed their ownparsimony-based results However if the Eurypterida+ Scorpiones hypothesis is to be credible it must beopen to evaluation and potential falsification usingobjective criteria and the subjective or intuitive a pri-ori weighting of characters advocated by Dunlop ampBraddy clearly does not qualify At present it isunclear how one would objectively evaluate Dunlop ampBraddyrsquos proposal with criteria compatible with thoseused in its original formulation For now the Euryp-terida + Scorpiones concept advocated by Dunlop ampBraddy may persist outside the mainstream of mod-ern systematic practice but it is increasingly problem-atic within it

Stomothecata nomnovOpiliones and Scorpiones were consistently recoveredas a monophyletic group The proposed name acknowl-edges a unique preoral chamber the stomothecaformed by coxapophyses of the palp and leg 1 (50)often with an auxiliary role played by the coxapophy-sis of leg 2 (51) In addition the epistome appears tohave been modified for adduction of the palpal coxaeThe lateral walls of the epistome are fused to themedial surfaces of the palpal coxae and the epistomallumen is spanned by a transverse muscle (188) which

238 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

apparently adducts the palpal coxae thereby constrict-ing the stomothecal chamber Scorpions and opilionsare also unique in having a pair of large epistomalarms projecting rearward into the prosoma andattaching to the endosternite (189) The epistomalarms provide attachment sites for pharyngeal dilatormuscles (196) and extrinsic muscles of anterior proso-mal appendages (eg 37) The chelicera is equippedwith a muscle that arises on the carapace and insertson the ventral margin of the second cheliceral article(23) There is an anteriorly placed genital opening(155)

Several notable similarities were found whilereviewing the literature but information from otherarachnid groups was considered too incomplete toallow them to be included in the matrix For exampleboth orders have apparent haemocytopoeitic organsassociated with major nerves of the anterior opistho-soma These are termed supraneural organs in scorpi-ons (Farley 1999) and perineural organs in opilions(Kaumlstner 1935) Haemocytes develop in the cardiacwall in spiders (Seitz 1972) and perhaps amblypygids(Weygoldt 2000) but haemocytopoeitic organs areunknown in most other arachnid groups Germ cellsdifferentiate very early during embryogenesis in bothscorpions and opilions relative to spiders (Moritz1957 Anderson 1973) Additional research is requiredto determine the phylogenetic utility of thesecharacters

Haplocnemata Boumlrner 1904Several workers recognize a close phylogenetic rela-tionship between Pseudoscorpiones and Solifugae(Weygoldt amp Paulus 1979 Van der Hammen 1989Shultz 1990 Wheeler amp Hayashi 1998 Giribetet al 2002) and this was consistently recovered inthe present analysis Haplocnemata is supported byseveral synapomorphies including (i) a feeding com-plex formed by two-segmented chelicerae (18) with aventrolateral intrinsic articulation (19) and perhapsa dorsolateral articulation with the carapace (20)and (ii) a rostrosoma (32) a preoral apparatus formedby an anteriorly projecting epistome affixed dorsallyto enlarged palpal coxae and bordered by lateral pal-pal projections The epistome forms the dorsal compo-nent of a lsquobeakrsquo with the ventral part formed by amidventral sternapophysis (13) fused between thepalpal coxae The locomotor apparatus consists ofcoxae that meet along the ventral midline (12) elon-gate femur-like patellae (48) and an apotele in theadult modified as an eversible empodium or pulvillus(92) The respiratory system is formed entirely (Pseu-doscorpiones) or in part (Solifugae) by paired trachealstigmata opening on opisthosomal somites 3 and 4(126)

Acaromorpha Dubinin 1957A clade comprising Ricinulei and Acari is consistentlyrecovered in the present analysis It is united here bytwo unique and seven homoplasious synapomorphiesThese include a gnathosoma (31) defined in part bymedial fusion of the palpal coxae (30) although pres-ence of a gnathosoma in Ricinulei is debatable as dis-cussed in the Appendix Acaromorphs also have aunique post-embryonic development consisting of ahexapodal larva and up to three octopodal nymphalinstars (176) The pedal patellandashtibia joints are formedby a bicondylar dorsal hinge rather than a single mid-dorsal condyle (69) all postcheliceral segmental gan-glia are unified in the subesophageal ganglion (130)and a postcerebral pharynx is absent (192) The groupis also tentatively united by presence of differentiatedpedal basi- and telofemora (63 64) and the absence ofa ventral (sternal) pharyngeal dilator muscle (199)but these may be symplesiomorphies erroneouslyreconstructed as a synapomorphies

The internal phylogenetic structure of Acaromorphais controversial with many recent workers favouringa monophyletic Acari (Weygoldt amp Paulus 1979Shultz 1990 argued most thoroughly by Lindquist1984) and others advocating a diphyletic Acari withAcariformes being the sister group to Cryptognomae(= Ricinulei + Anactinotrichida) (especially Van derHammen 1979 1989) (Fig 3) Acari was recovered asmonophyletic when fossils were excluded (Fig 2) butnot when they were included (Fig 3) A lsquomite-centredrsquosurvey of arachnid characters may be needed if mor-phology is to offer a compelling solution to the internalphylogeny of Acari and its placement within Arach-nida These issues are not resolved by the presentanalysis

Pantetrapulmonata nom novPantetrapulmonata includes the extinct orders Trigo-notarbida Haptopoda and the extant orders AraneaeAmblypygi Schizomida and Thelyphonida The cladeis united by cheliceral structure (18 19) a megoper-culum (106) booklungs on the genital and first post-genital somites (121 122) and enlargement of theepipharyngeal sclerite (192) It is important to notehowever that most of these characters were coded asuncertain in Plesiosiro (Haptopoda) Aside from theplacement of Haptopoda the monophyly of Trigono-tarbida and the extant orders was anticipated byShear et al (1987) and is generally regarded as amonophyletic group

Tetrapulmonata Shultz 1990Tetrapulmonata was originally proposed on the basisof neontological analyses and encompassed Araneae

PHYLOGENY OF THE ARACHNID ORDERS 239

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Amblypygi Schizomida and Thelyphonida (Shultz1990) and was also recovered here in analysis of neon-tological data (Fig 5) However results from analysisof all taxa required that the Tetrapulmonata conceptbe expanded to include the fossil order HaptopodaFeatures uniting this group are problematic howeveras many states in Haptopoda were coded as uncertain

Schizotarsata nomnovHaptopoda and Pedipalpi (= Amblypygi + Schizomida+ Thelyphonida) are united here in a group named forpossession of divided pedal telotarsi (84) Synapomor-phies include a pointed anterior carapacal margin (3)and elongation of leg 1 (46) As already noted theplacement of Haptopoda should probably be regardedas tentative because the state of many characters inthis group was coded as unknown The placement ofHaptopoda as the sister group to Pedipalpi was antic-ipated by Dunlop (1999 2002c)

Pedipalpi Boumlrner 1904Pedipalpi encompasses Amblypygi and Uropygi(= Thelyphonida + Schizomida) Although widely rec-ognized in the past Pedipalpi was set aside in the mid-20th century in favour of Labellata (= Amblypygi +Araneae) based largely on the presumed synapomor-phies of a narrow prosomandashopisthosoma juncture (97)and sucking stomach (199ndash201) The situation beganto reverse when Shear et al (1987) again highlightedsimilarities of Amblypygi and Uropygi such as rapto-rial palps (35) and antenniform leg 1 (46) Shultz(1989 1990 1999) subsequently described numerousderived similarities including asymmetrical flexormuscles at the pedal femurndashpatella joint (70) threetelotarsomeres on legs 2ndash4 (85) and modification ofthe palpal coxae to provide support for extrinsic pha-ryngeal muscles (197) Pedipalpi has emerged as oneof the most well-supported interordinal relationshipsin Arachnida

Uropygi Thorell 1882Uropygi has long been accepted as a monophyleticunion of Thelyphonida and Schizomida Synapomor-phies include a unique mating behaviour (159) fusedpalpal coxae (30) 2-1-1-1 arrangement of tibial tricho-bothria (144) posterior defensive glands (102) elon-gated patella of leg 1 (68) and many others

ACKNOWLEDGEMENTS

I thank Nikolaj Scharff and three anonymous review-ers for useful comments This work was supported bythe Maryland Agricultural Experiment Station the

General Research Board of the University of Mary-land and National Science Foundation Grant 0416628to Susan Masta

REFERENCES

Abd el-Wahab A 1952 Notes on the morphology of the scor-pion Buthus quinquestriatus (HE) Publications de lrsquoInsti-tut Fouad 1er du Desert 3 1ndash129

Alberti G 1979 Licht- und elektronenmikroskopische Unter-suchungen an Coxaldruumlsen von Walzenspinnen (ArachnidaSolifugae) Zoologischer Anzeiger 203 48ndash64

Alberti G 1995 Comparative spermatology of Cheliceratareview and perspective Memoires du Museum NationaldrsquoHistoire Naturelle 166 203ndash230

Alberti G 2005 Double spermatogenesis in ChelicerataJournal of Morphology 266 281ndash297

Alberti G Coons LB 1999 Acari mites In Harrison FWFoelix RF eds Microscopic anatomy of invertebratesVol 8C Chelicerate Arthropoda New York Wiley Liss515ndash1215

Alberti G Peretti AV 2002 Fine structure of male genitalsystem and sperm in Solifugae does not support a sister-group relationship with Pseudoscorpiones (Arachnida)Journal of Arachnology 30 268ndash274

Alexander AJ 1967 Problems of limb extension in the scor-pion Opisthophthalmus latimanus Koch Transactions of theRoyal Society of South Africa 37 165ndash181

Anderson DT 1973 Embryology and phylogeny in annelidsand arthropods Oxford Pergamon Press

Anderson LI Selden PA 1997 Opisthosomal fusion andphylogeny of Palaeozoic Xiphosura Lethaia 30 19ndash31

Beall BS Labandeira CC 1990 Macroevolutionary patternsof the Chelicerata and Tracheata In Culver SJ ed Arthro-pod paleobiology Short Courses in Paleontology No 3 Lon-don The Paleontological Society 257ndash284

Bernard HM 1896 The comparative morphology of the Gale-odidae Transactions of the Linnean Society of London Sec-ond Series 6 305ndash417

Boumlrner C 1902a Arachnologische Studien (II und III) Zool-ogischer Anzeiger 25 433ndash466

Boumlrner C 1902b Arachnologische Studien IV Die Genitalor-gane der Pedipalpen Zoologischer Anzeiger 26 81ndash92

Boumlrner C 1904 Beitraumlge zur Morphologie der Arthropoden IEin Beitrag zur Kenntnis der Pedipalpen Zoologica (Stut-tgart) 42 1ndash174

Bowler PJ 1996 Lifersquos splendid drama Chicago Universityof Chicago Press

Braddy SJ Alridge RJ Gabbott SE Theron JN 1999Lamellate book-gills in a late Ordovician eurypterid from theSoom Shale South Africa support for a eurypterid-scorpionclade Lethaia 32 72ndash74

Braddy SJ Dunlop JA 1997 The functional morphology ofmating in the Silurian eurypterid Baltoeurypterus tet-ragonophthalmus (Fischer 1839) Zoological Journal of theLinnean Society 121 435ndash461

Brauer A 1895 Beitrage zur Kenntnis der Entwicklungsge-schichte des Skorpiones II Zeitschrift fuumlr WissenschaftlicheZoologie 59 351ndash433

240 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bremer K 1994 Branch support and tree stability Cladistics10 295ndash304

Bristowe WS 1932 The liphistiid spiders Proceedings of theZoological Society of London Part 4 1016ndash1057

Butt AG Taylor HH 1991 The function of spider coxalorgans effects of feeding and salt-loading on Porrhotheleantipodiana (Mygalomorpha Dipluridae) Journal of Exper-imental Biology 158 439ndash461

Buxton BH 1913 Coxal glands of arachnids ZoologischeJahrbucher (Suppl XIV) 16ndash50

Buxton BH 1917 Notes on the anatomy of arachnids Jour-nal of Morphology 29 1ndash31

Camin JH Clark GM Bourdeau FG 1956 The palpallsquotined setarsquo in the Mesostigmata a homologue of the palpalclaw in the Onychopalpida (Acarina) In Becker EC ed Pro-ceedings of the Tenth International Congress of EntomologyOttawa Mortimer 903ndash908

Chamberlin JC 1931 The arachnid order ChelonethidaStanford University Stanford University Press

Chapman RF 1998 The insects structure and function 4thedn Cambridge Cambridge University Press

Clarke KU 1979 Visceral anatomy and arthropod phylogenyIn Gupta AP ed Arthropod phylogeny New York Van Nos-trand Reinhold 467ndash549

Clarke J 1984 On the relationship between structure andfunction in the leg joints of Heteropoda venatoria (L) (Ara-neae Eusparassidae) Bulletin of the British ArachnologicalSociety 6 181ndash192

Clarke J 1986 The comparative functional morphology of theleg joints and muscles of five spiders Bulletin of the BritishArachnological Society 7 37ndash47

Clarke JM Ruedemann R 1912 The Eurypterida of NewYork New York State Museum Memoir 14 (1ndash2) 1ndash628

Coddington JA 2005 Phylogeny and classification of spi-ders In Ubick D Paquin P Cushing PE Roth V eds Spi-ders of North America an identification manual AmericanArachnological Society 18ndash24

Coddington JA Giribet G Harvey MS Prendini LWalter DE 2004 Arachnida In Cracraft J Donoghue MJeds Assembling the tree of life Oxford Oxford UniversityPress 269ndash318

Coddington JA Levi HW 1991 Systematics and evolutionof spiders (Araneae) Annual Review of Ecology and System-atics 22 565ndash592

Coineau Y Van der Hammen L 1979 The postembryonicdevelopment of Opilioacarida with notes on new taxa and ona general model for the evolution In Piffle E ed Proceed-ings of the 4th International Congress of Acarology Budap-est Akadeacutemiai Kiadoacute 437ndash441

Cokendolpher JC Reddell JR 1992 Revision of the Pro-toschizomidae (Arachnida Schizomida) with notes on thephylogeny of the order Texas Memorial Museum Speleolog-ical Monographs 3 31ndash74

Condeacute B 1991a Le dimorphisme sexuel des Palpigrades Bul-letin de la Socieacuteteacute Neuchacircteloise des Sciences Naturelles 11667ndash73

Condeacute B 1991b Lrsquoaire genitale male des Palpigrades Bulle-tin de la Socieacuteteacute Europeenne drsquoArachnologie 1990 64ndash69

Dearden PK Donly C Grbicccc M 2002 Expression of pair-rule gene homologues in a chelicerate early patterning ofthe two-spotted spider mite Tetranychus urticae Develop-ment 129 5461ndash5472

Delle Cave L 1975 The ldquocleaning brushrdquo in Damon diademaSimon 1876 and Phrynichus jayakari Pocock 1894 (Ambly-pygi Arachnida) Monitore Zoologico Italiano (NS) 9 175ndash184

Dunlop JA 1994 Filtration mechanisms in the mouthparts oftetrapulmonate arachnids (Trigonotarbida Araneae Ambly-pygi Uropygi Schizomida) Bulletin of the British Arachno-logical Society 9 267ndash273

Dunlop JA 1996 Evidence for a sister group relationshipbetween Ricinulei and Trigonotarbida Bulletin of the BritishArachnological Society 10 193ndash204

Dunlop JA 1997 Palaeozoic arachnids and their significancefor arachnid phylogeny In Zabka M ed Proceedings of the16th European colloquium of arachnology Siedlce PolandWyzsza Szkola Rolnicko-Pedagogiczna 65ndash82

Dunlop JA 1998 The origins of tetrapulmonate book lungsand their significance for chelicerate phylogeny In SeldenPA ed Proceedings of the 17th European Colloquium ofArachnology Edinburgh 1997 Burnham Beeches BritishArachnological Society 9ndash16

Dunlop JA 1999 A redescription of the Carboniferous arach-nid Plesiosiro madeleyi Pocock 1911 (Arachnida Hapto-poda) Transactions of the Royal Society of Edinburgh EarthSciences 90 29ndash47

Dunlop JA 2000 The epistomo-labral plate and lateral lips insolifuges pseudoscorpions and mites Ekoloacutegia (Bratislava)19 67ndash78

Dunlop JA 2002a Arthropods from the Lower Devonian Sev-ernaya Zemlya Formation of October Revolution Island(Russia) Geodiversitas 24 349ndash379

Dunlop JA 2002b Character states and evolution of the che-licerate claws In Toft S Scharff N eds European Arach-nology 2000 Aarhus Aarhus University Press 345ndash354

Dunlop JA 2002c Phylogeny of Chelicerata In LlorenteBousquets J Morrone JJ eds Biodiversidad Taxonomiacutea yBiogeografiacutea de Arthroacutepodos de Meacutexico Hacia Una Siacutentesisde Su Conocimiento Vol III Prensas de Ciencias Univer-sidad Nacional Autonoma de Meacutexico 117ndash141

Dunlop JA Anderson LI Braddy SJ 2004 A redescriptionof Chasmataspis laurencii Caster amp Brooks 1956 (Chelicer-ata Chasmataspidida) from the Middle Ordovician of Ten-nessee USA with remarks on chasmataspid phylogenyTransactions of the Royal Society of Edinburgh Earth Sci-ences 94 207ndash225

Dunlop JA Braddy SJ 1997 Slit-like structures on theprosomal appendages of the eurypterid BaltoeurypterusNeues Jahrbuch fuumlr Geologie und Palaumlontologie-Monatshefte1997 31ndash38

Dunlop JA Braddy SJ 2001 Scorpions and their sister-group relationships In Fet V Selden PA eds Scorpions2001 in Memoriam Gary A Polis London British Arachno-logical Society 1ndash24

Dunlop JA Horrocks CA 19951996 A new Upper Carbon-iferous whip scorpion (Arachnida Uropygi Thelyphonida)

PHYLOGENY OF THE ARACHNID ORDERS 241

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

with a revision of the British Carboniferous Uropygi Zoolo-gischer Anzeiger 234 293ndash306

Dunlop JA Poschmann M Anderson L 2001 On theEmsian (early Devonian) arthropods of the Rhenish SlateMountains 3 The chasmataspidid Diploaspis Palaumlon-tolgische Zeitschrift 75 253ndash269

Dunlop JA Webster M 1999 Fossil evidence terrestrializa-tion and arachnid phylogeny Journal of Arachnology 27 86ndash93

Edgecombe GD Wilson GDF Colgan DJ Gray MR Cas-sis G 2000 Arthropod cladistics combined analysis of his-tone H3 and U2 snRNA sequences and morphologyCladistics 16 155ndash203

Evans GO 1992 Principles of acarology Wallingford UKCABI

Fahrenbach WH 1999 Merostomata In Harrison FW Foe-lix RF eds Microscopic anatomy of invertebrates Vol 8AChelicerate arthropods New York Wiley-Liss 21ndash115

Farley RD 1999 Scorpiones In Harrison FW Foelix RF edsMicroscopic anatomy of invertebrates Vol 8A Cheliceratearthropods New York Wiley-Liss 117ndash222

Farley RD 2005 Developmental changes in the embryo pro-nymph and first molt of the scorpion Centruroides vittatus(Scorpiones Buthidae) Journal of Morphology 265 1ndash27

Fayers SR Dunlop JA Trewin NH 2004 A new earlyDevonian trigonotarbid arachnid from the Windyfield ChertRhynie Scotland Journal of Systematic Palaeontology 2269ndash284

Felsenstein J 1985 Confidence limits on phylogenies anapproach using the bootstrap Evolution 39 783ndash791

Fet V Sissom WD Lowe G Braunwalder ME 2000 Cat-alog of the scorpions of the world (1758ndash1998) New YorkNew York Entomological Society

Firstman B 1954 The central nervous system musculatureand segmentation of the cephalothorax of a tarantula (Eu-rypelma californicum Ausserer) Microentomology 87 1ndash40

Firstman B 1973 The relationship of the chelicerate arterialsystem to the evolution of the endosternite Journal ofArachnology 1 1ndash54

Foelix RF 1985 Mechano- and chemoreceptive sensilla InBarth FG ed Neurobiology of arachnids Berlin Spinger-Verlag 118ndash137

Foelix RF 1996 Biology of spiders 2nd edn Oxford OxfordUniversity Press

Giribet G Edgecombe GD Wheeler WC Babbit C 2002Phylogeny and systematic position of Opiliones a combinedanalysis of chelicerate relationships using morphologicaland molecular data Cladistics 18 5ndash70

Goloboff P 1993 Estimating character weights during treesearch Cladistics 9 83ndash91

Goloboff PA Farris JS Nixon K 2000 TNT Tree AnalysisUsing New Technologies Version 1 Program and documen-tation httpwwwzmucdkpublicphylogenytnt

Grainge CA Pearson RG 1966 Cuticular structure in thePhalangida Nature 211 866

Grandjean F 1938 Observations sur les Bdelles (Acariens)Annales de la Socieacuteteacute Entomologique de France Paris 107 1ndash24

Hansen HJ Soslashrensen W 1904 On two orders of ArachnidaCambridge Cambridge University Press

Hansen HJ Soslashrensen W 1905 The Tartarides a tribe of theorder Pedipalpi Arkiv fuumlr Zoologi 2 1ndash78

Hara MR Gnaspini P 2003 Comparative study of the defen-sive behavior and morphology of the gland opening areaamong harvestmen (Arachnida Opiliones Gonyleptidae)under a phylogenetic perspective Arthropod Structure andDevelopment 32 257ndash275

Harvey MS 1992 The phylogeny and classification of thePseudoscorpionida (Chelicerata Arachnida) InvertebrateTaxonomy 6 1373ndash1435

Harvey MS 2002 The neglected cousins what do we knowabout the smaller arachnid orders Journal of Arachnology30 357ndash372

Haupt J Song D 1996 Revision of East Asian whip scorpions(Arachnida Uropygi Theyphonida) I China and JapanArthropoda Selecta 5 43ndash52

Hjelle JT 1990 Anatomy and morphology In Polis GA edThe biology of scorpions Stanford University Stanford Uni-versity Press 9ndash63

Janczyk FSW 1956 Anatomie von Siro duricorius Joseph imVergleich mit anderen Opilioniden Sitzungsberichte Oster-reichischen Akademie der Wissenschaften Mathematisch-Natruwissenschaftliche Klasse 165 474ndash522

Jenner RA 2001 Bilaterian phylogeny and uncritical recy-cling of morphological data sets Systematic Biology 50 730ndash742

Jeram AJ 1998 Phylogeny classification and evolution of Sil-urian and Devonian scorpions In Selden PA ed Proceed-ings of the 17th European Colloquium of ArachnologyEdinburgh 1997 Burnham Beeches UK British Arachno-logical Society 15ndash31

Karaman IM 2005 Evidence of spermatophores in Cyphoph-thalmi (Arachnida Opiliones) Revue Suisse de Zoologie 1123ndash11

Kaumlstner A 1935 Die Funktion der sogenannten sympathis-chen Ganglien und die Exkretion bei den PhalangiidenZoologischer Anzeiger 109 273ndash288

Kierans JE Robbins RG 1999 A world checklist of generasubgenera and species of ticks (Acari Ixodida) publishedfrom 1973 to 1997 Journal of Vector Ecology 24 115ndash129

Kjellesvig-Waering EN 1986 A restudy of the fossil Scorpi-onida of the world Palaeontographica Americana 55 1ndash287

Klompen JSH 2000 Prelarva and larva of Opilioacarus (Neo-carus) texanus (Chamberlin and Mulaik) (Acari Opilioacar-ida) with notes on the patterns of setation and lyrifissuresJournal of Natural History 34 1977ndash1992

Kraus O 1998 Elucidating the historical process of phylog-eny phylogenetic systematics versus cladistic techniquesIn Selden PA ed Proceedings of the 17th European Collo-quium of Arachnology Edinburgh 1997 Burnham BeechesUK British Arachnological Society 1ndash7

Kraus O Kraus M 1993 Divergent transformation of cheli-cerae and original arrangement of eyes in spiders (Arach-nida Araneae) Memoirs of the Queensland Museum 33579ndash584

Lankester ER Benham WBS Beck EJ 1885 On the mus-

242 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

cular and endoskeletal systems of Limulus and Scorpio withsome notes on the anatomy and generic characters of scor-pions Transactions of the Zoological Society of London 11311ndash384

Leclerc P 1989 Considerations paleobiogeographiques a pro-pos la decouverte en Thailande drsquoopilioacariens nouveaux(Acari ndash Notostigmata) Compte Rendu des Seances de laSociete de Biogeographie 65 162ndash174

Legg G 1976 The external morphology of a new species ofricinuleid (Arachnida) from Sierra Leone Zoological Journalof the Linnean Society 59 1ndash58

Lindquist EE 1984 Current theories on the evolution ofmajor groups of Acari and on their relationships with othergroups of Arachnida with consequent implications for theirclassificaton In Griffiths DA Bowman CE eds AcarologyVI Vol 1 New York John Wiley amp Sons 28ndash62

Manning PL Dunlop JA 1995 The respiratory organs ofeurypterids Palaeontology 38 287ndash297

Manton SM 1958 Hydrostatic pressure and leg exten-sion in arthropods with special reference to arachnidsAnnals and Magazine of Natural History Series 13 1161ndash182

Manton SM 1964 Mandibular mechanisms and the evolutionof arthropods Philosophical Transactions of the Royal Soci-ety of London B 7 1ndash183

Maraun M Heethoff M Schneider K Scheu S WeigmannG Cianciolo J Thomas RH Norton RA 2004 Molecularphylogeny of oribatid mites (Oribatida Acari) evidence formultiple radiations of parthenogenetic lineages Experimen-tal and Applied Acarology 33 183ndash201

Marples BJ 1967 The spinnerets and epiandrous glands ofspiders Journal of the Linnean Society (Zoology) 46 209ndash222

Marples BJ 1983 Observations on the structure of the fore-gut of spiders Bulletin of the British Arachnological Society8 46ndash52

Millot J 1933 Notes compleacutementaires sur lrsquoanatomie desliphistiides et des hypchilides a propos drsquoun travail reacutecent deA Petrunkevitch Bulletin de la Socieacuteteacute Zoologique de France68 217ndash235

Millot J 1942 Sur lrsquoanatomie et lrsquohistophysiologie deKoenenia mirabilis Grassi (Arachnida Palpigradi) RevueFranccedilaise drsquoEntomologie 9 33ndash51

Millot J 1943 Notes complementaires sur lrsquoanatomielrsquohistologie et la repartition geographique en France deKoenenia mirabilis Grassi Revue Franccedilaise drsquoEntomologie 9127ndash135

Millot J 1945 Lrsquoanatomie interne des Ricinulei (Arachnides)Annales des Sciences Naturelles (Zoologie) Series 11 7 1ndash29

Millot J 1949a Morphologie generale et anatomie interne InGrasseacute P-P ed Traite de zoologie VI Paris Masson et Cie263ndash319

Millot J 1949b Ordre des Amblypyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 563ndash588

Millot J 1949c Ordre des Araneacuteides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 589ndash743

Millot J 1949d Ordre des Palpigrades In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 520ndash532

Millot J 1949e Ordre des Uropyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 533ndash562

Millot J 1949f Ordre des Ricinuleides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 744ndash760

Millot J Vachon M 1949 Ordre Des Solifuges In Grasseacute P-P ed Traite de zoologie VI Paris Masson et Cie 482ndash519

Moore RA Briggs DEG Bartels C 2005 A new specimen ofWeinbergina opitzi (Chelicerata Xiphosura) from the LowerDevonian Huumlnsruck Slate Germany PalaumlontologischeZeitschrift 79 399ndash408

Moritz M 1957 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) unter besonderer Beruumlcksichtigungder ausseren Morphologie der Bildung des Mitteldarmesund der Genitalanlage Zoologische Jahrbuumlcher Abteilungfuumlr Anatomie und Ontogenie der Tiere 76 331ndash370

Moritz M 1959 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) II Die Anlage und Entwicklung derCoxaldruumlse bei Phalangium opilio L Zoologische Jahr-buumlcher Abteilung fuumlr Anatomie und Ontogenie der Tiere 77229ndash240

Nixon K 1999 The parsimony ratchet a new method forrapid parsimony analysis Cladistics 15 407ndash414

Norton RA 1998 Morphological evidence for the evolutionaryorigin of Astigmata (Acari Acariformes) Experimental andApplied Acarology 22 559ndash594

Palmgren P 1978 On the muscular anatomy of spiders ActaZoologica Fennica 155 1ndash41

Patten W 1912 The evolution of the vertebrates and their kinPhiladelphia P Blakistonrsquos Son

Patten W Hazen AP 1900 The development of the coxalgland branchial cartilages and genital ducts of Limuluspolyphemus Journal of Morphology 16 459ndash490

Paulus HF 1979 Eye structure and the monophyly of theArthropoda In Gupta AP ed Arthropod phylogeny NewYork Van Nostrand Reinhold 299ndash383

Petrunkevitch A 1949 A study of Palaeozoic ArachnidaTransactions of the Connecticut Academy of Arts and Sci-ences 37 69ndash315

Pittard K Mitchell RW 1972 Comparative morphology ofthe life stages of Cryptocellus pelaezi (Arachnida Ricinulei)Graduate Studies Texas Tech University 1 1ndash77

Platnick NI 2005 The world spider catalog Version55 American Museum of Natural History httpresearchamnhorgentomologyspiderscatalogindexhtm

Platnick NI Gertsch WJ 1976 The suborders of spiders acladistic analysis (Arachnida Araneae) American MuseumNovitates 2607 1ndash15

Pocock RI 1893 On some points in the morphology of theArachnida (ss) with notes on the classification of the groupAnnals and Magazine of Natural History Series 6 11 1ndash19

Pocock RI 1902 Some points in the morphology and classi-fication of the Opiliones Annals and Magazine of NaturalHistory Series 7 10 504ndash516

Popadicccc A Panganiban G Rusch D Shear WA KaufmanTC 1998 Molecular evidence for the gnathobasic derivationof arthropod mandibles and for the appendicular origin ofthe labrum and other structures Development Genes andEvolution 208 142ndash150

PHYLOGENY OF THE ARACHNID ORDERS 243

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Raw F 1957 Origin of chelicerates Journal of Paleontology31 139ndash192

Reddell JR Cokendolpher JC 1995 Catalogue bibliogra-phy and generic revision of the order Schizomida (Arach-nida) Texas Memorial Museum Speleological Monographs4 1ndash170

Reissland A Goumlrner P 1985 Trichobothria In Barth FGed Neurobiology of arachnids Berlin Springer-Verlag 138ndash161

Roewer C-F 1934 Solifugae Palpigradi Bronns Klassen undOrdnungen ses Tierreich 5 IV 4 1ndash713

Rowland JM Cooke JAL 1973 Systematics of the arachnidorder Uropygida (=Thelyphonida) Journal of Arachnology 155ndash71

Rowland JM Sissom WD 1980 Report on a fossil palpi-grade from the Tertiary of Arizona and a review of themorphology and systematics of the order (ArachnidaPalpigradida) Journal of Arachnology 8 69ndash86

Rucker A 1901 The Texan Koenenia American Naturalist35 615ndash630

Ruhland M Rathmayer W 1978 Die Beinmuskulatur undihre Innervation bei der Vogelspinne Dugesiella henzti (Ch)(Araneae Aviculariidae) Zoomorphologie 89 33ndash46

Saboori A Kamali K 2000 Description of Allothrombiumtriticium adult (Acari Trombidiidae) from Iran Systematicand Applied Acarology 5 207ndash208

Schliwa M 1979 The retina of the phalangid Opilio raven-nae with particular reference to arhabdomeric cells Celland Tissue Research 204 473ndash495

Scholl G 1977 Beitrage zur Embryonalentwicklung von Lim-ulus polyphemus L (Chelicerata Xiphosura) Zoomorpholo-gie 86 99ndash154

Scholtz G Kamenz C 2006 The book lungs of Scorpionesand Tetrapulmonata (Chelicerata Arachnida) evidence forhomology and a single terrestrialization event of a commonarachnid ancestor Zoology 109 2ndash13

Seitz KA 1972 Zur Histologie und Feinstruktur des Herzensund der Haumlmocyten von Cupiennius salei Keys (AraneaeCtenidae) I Herzwandung Bildung und Differenzierungder Haumlmocyten Zoologische Jahrbuumlcher Abteilung fuumlr Anat-omie und Ontogenie der Tiere 89 351ndash384

Sekiguchi K ed 1988 Biology of horseshoe crabs Tokyo Sci-ence House

Selden PA 1981 Functional morphology of the prosoma ofBaltoeurypterus tetragonophthalmus (Fischer) (ChelicerataEurypterida) Transactions of the Royal Society of Edin-burgh Earth Sciences 72 9ndash48

Selden P 1992 Revision of the fossil ricinuleids Transactionsof the Royal Society of Edinburgh Earth Sciences 83 595ndash634

Selden PA 1993 Fossil arachnids ndash recent advances andfuture prospects Memoirs of the Queensland Museum 33389ndash400

Selden PA Dunlop JA 1998 Fossil taxa and relationshipsof chelicerates In Edgecombe GD ed Arthropod fossilsand phylogeny New York Columbia University Press 303ndash331

Selden PA Jeram AJ 1989 Palaeophysiology of terrestrial-

isation in the Chelicerata Transactions of the Royal Societyof Edinburgh Earth Sciences 80 303ndash310

Selden PA Shear WA Bonamo PM 1991 A spider and otherarachnids from the Devonian of New York and reinterpre-tations of Devonian Araneae Palaeontology 34 241ndash281

Sensenig A Shultz JW 2003 Mechanics of cuticular elasticenergy storage in leg joints lacking extensor muscles inarachnids Journal of Experimental Biology 206 771ndash784

Shear WA 1993 New species in the opilionid genus Stylocel-lus from Malaysia Indonesia and the Philippines (OpilionesCyphophthalmi Stylocellidae) Bulletin of the British Arach-nological Society 9 174ndash188

Shear WA Selden PA Rolfe WDI Bonamo PM GriersonJD 1987 New terrestrial arachnids from the Devonian ofGilboa New York (Arachnida Trigonotarbida) AmericanMuseum Novitates 2901 1ndash74

Shultz JW 1987 The origin of the spinning apparatus in spi-ders Biological Reviews of the Cambridge PhilosophicalSociety 62 89ndash113

Shultz JW 1989 Morphology of locomotor appendages inArachnida evolutionary trends and phylogenetic implica-tions Zoological Journal of the Linnean Society 97 1ndash56

Shultz JW 1990 Evolutionary morphology and phylogeny ofArachnida Cladistics 6 1ndash31

Shultz JW 1991 Evolution of locomotion in Arachnida thehydraulic pressure pump of the giant whipscorpion Masti-goproctus giganteus (Uropygi) Journal of Morphology 21013ndash31

Shultz JW 1993 Muscular anatomy of the giant whipscor-pion Mastigoproctus giganteus (Arachnida Uropygi) and itsevolutionary significance Zoological Journal of the LinneanSociety 108 335ndash365

Shultz JW 1998 Phylogeny of Opiliones (Arachnida) anassessment of the lsquoCyphopalpatoresrsquo concept Journal ofArachnology 26 257ndash272

Shultz JW 1999 Muscular anatomy of a whipspider Phrynuslongipes (Amblypygi) and its evolutionary significance Zoo-logical Journal of the Linnean Society 126 81ndash116

Shultz JW 2000 Skeletomuscular anatomy of the harvest-man Leiobunum aldrichi (Weed 1893) (Arachnida Opil-iones Palpatores) and its evolutionary significanceZoological Journal of the Linnean Society 128 401ndash438

Shultz JW 2001 Gross muscular anatomy of Limuluspolyphemus (Chelicerata Xiphosura) and its bearing on evo-lution in the Arachnida Journal of Arachnology 29 283ndash303

Shultz JW 2007 Morphology of the prosomal exoskeleton ofScorpiones (Arachnida) and a new hypothesis for the evolu-tion of cuticular cephalic endoskeletons in arthropodsArthropod Structure amp Development 36 77ndash102

Shultz JW Pinto da Rocha R 2007 Morphology and func-tional anatomy In Pinto da Rocha R Machado G Giribet Geds The harvestmen the biology of Opiliones CambridgeMA Harvard University Press 14ndash61

Shultz JW Regier JC 2001 Phylogenetic analysis of Pha-langida (Arachnida Opiliones) using two nuclear protein-encoding genes supports monophyly of Palpatores Journalof Arachnology 29 189ndash200

Sitnikova LG 1978 The main evolutionary trends of the

244 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Acari and the problem of their monophyletism Entomologis-cheskoe Obozrenie 57 431ndash457

Snodgrass RE 1948 The feeding organs of Arachnidaincluding mites and ticks Smithsonian Miscellaneous Col-lections 110 (10) 1ndash94

Snodgrass RE 1952 A textbook of arthropod anatomy Ith-aca NY Comstock Publishing

Soslashrensen W 1879 Om Bygningen af Gonyleptiderne en Typeaf Arachnidernes Classe Naturhistorisk Tidsskrifter Series3 8 97ndash222

Stockwell SA 1989 Revision of the phylogeny and higherclassification of scorpions (Chelicerata) Unpublished doc-toral dissertation University of California at Berkeley

Stoslashrmer L 1944 On the relationships and phylogeny of fossiland Recent Arachnomorpha Skrifter utgitt av Det NorskeVidenskaps-Akademi i Oslo I 5 1ndash158

Stoslashrmer L 1955 Merostomata In Moore RC ed Treatise oninvertebrate paleontology Part P Arthropoda 2 LawrenceKS University of Kansas Press and Geological Society ofAmerica 4ndash41

Swofford DL 2002 PAUP Phylogenetic analysis using par-simony (and other methods) Version Beta 10 SunderlandMA Sinauer Associates

Talarico G Palacios-Vargas JG Fuentes Silva M AlbertiG 2005 First ultrastructural observations of the tarsal poreorgan of Pseudocellus pearsei and P boneti (ArachnidaRicinulei) Journal of Arachnology 33 604ndash612

Templeton AR 1983 Phylogenetic inference from restrictionendonuclease cleavage site maps with articular reference tothe evolution of humans and the apes Evolution 37 221ndash244

Tetlie OE Braddy SJ 2004 The first Silurian chasmataspidLoganamaraspis dunlopi gen et sp nov (Chelicerata Chas-mataspidida) from Lesmahagow Scotland and its implica-tions for eurypterid phylogeny Transactions of the RoyalSociety of Edinburgh Earth Sciences 94 227ndash234

Thomas RH Telford MJ 1999 Appendage development inembryos of the oribatid mite Archegozetes longisetosus(Acari Oribatei Trhypochthoniidae) Acta Zoologica (Stock-holm) 80 193ndash200

Tollerton VP Jr 1989 Morphology taxonomy and classifica-tion of the order Eurypterida Burmeister 1843 Journal ofPaleontology 63 642ndash657

Vachon M 1949 Ordre des Pseudoscorpions In Grasseacute P-Ped Traite de zoologie VI Paris Masson et Cie 436ndash481

Van der Hammen L 1979 Comparative studies in Chelicer-ata I The Cryptognomae (Ricinulei Architarbi and Anacti-notrichida) Zoologische Verhandelingen 174 1ndash62

Van der Hammen L 1982 Comparative studies in Chelicer-ata II Epimerata (Palpigradi and Actinotrichida) Zoolo-gische Verhandelingen 196 1ndash70

Van der Hammen L 1985 Functional morphology and affin-ities of extant Chelicerata in evolutionary perspectiveTransactions of the Royal Society of Edinburgh 76 137ndash146

Van der Hammen L 1989 An introduction to comparativearachnology Leiden SPB Academic Publishing

Vyas AB 1970 Studies in myology of the scorpion Heter-ometrus fulvipes Koch Unpublished PhD DissertationDepartment of Zoology Gujarat University India

Vyas AB 1974 The cheliceral muscles of the scorpion Heter-ometrus fulvipes Bulletin Southern California Academy ofSciences 73 9ndash14

Walter DE 1988 Predation and mycophagy by endeostigma-tid mites (Acariformes Prostigmata) Experimental andApplied Acarology 4 159ndash166

Walter DE Proctor HC 1998 Feeding behaviour and phy-logeny observations on early derivative Acari Experimentaland Applied Acarology 22 39ndash50

Walter DE Proctor HC 1999 Mites ecology evolution andbehaviour Sydney University of New South Wales

Waterston CD 1979 Problems of functional morphology andclassification in stylonuroid eurypterids (Chelierata Meros-tomata) with observations on the Scottish Silurian Sty-lonuroidea Transactions of the Royal Society of EdinburghEarth Science 70 251ndash322

Weygoldt P 1969 Biology of pseudoscorpions CambridgeMA Harvard University Press

Weygoldt P 1984 Lrsquoautotomie chez les Amblypyges RevueArachnologique 5 321ndash327

Weygoldt P 1996 Evolutionary morphology of whip spiderstowards a phylogenetic system (Chelicerata ArachnidaAmblypygi) Journal of Zoological Systematics and Olution-ary Research 34 185ndash202

Weygoldt P 1998 Evolution and systematics of the Chelicer-ata Experimental and Applied Acarology 22 63ndash79

Weygoldt P 2000 Whip spiders (Chelicerata Amblypygi)their biology morphology and systematics StenstrupDenmark Apollo Books

Weygoldt P Paulus HF 1979 Untersuchungen zur Morphol-ogie Taxonomie und Phylogenie der Chelicerata Zeitschriftfuumlr Zoologische Systematik und Evolutionsforschung 17 85ndash116 177ndash200

Wheeler WC Hayashi CY 1998 The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192

Winkler D 1957 Die Entwicklung der aumlusseren Koumlrperge-stalt bei den Phalangiidae (Opiliones) Mitteilungen aus demZoologischen Museum in Berlin 33 355ndash389

With CJ 1904 The Notostigmata a new suborder of AcariVidenskabelige Meddelelser fra Dansk Naturhistorisk Foren-ingi Kjobenhaus 1904 137ndash192

Yamasaki T Makioka T Saito J 1988 Morphology InSekiguchi K ed Biology of horseshoe crabs Tokyo ScienceHouse 69ndash132

Yoshikura M 1975 Comparative embryology and phylogeny ofArachnida Kumamoto Journal of Science Biology 12 71ndash142

APPENDIX

Each character used in this analysis is numbereddefined and cross-referenced with characters used infour previous studies that are abbreviated in bracketsspecifically WP = Weygoldt amp Paulus (1979)S = Shultz (1990) WH = Wheeler amp Hayashi (1998)GEWB = Giribet et al (2002) Descriptions includecorrections of errors discussions of controversies orambiguities and justifications for state assignmentsto lsquoproblemrsquo taxa

PHYLOGENY OF THE ARACHNID ORDERS 245

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

PROSOMA

1 Dorsal sclerite formed by fusion of the prosomal car-apace the dorsal portion of the first opisthosomalsomite and the dorsomedial (axial) portion of the sec-ond opisthosomal somite 0 absent 1 presentState 1 is present in extant xiphosurids and close fos-sil relatives (Scholl 1977 Anderson amp Selden 1997Shultz 2001) The fossil record suggests that thisstructure originated with the disappearance ofthe already reduced first opisthosomal tergite(= microtergite) and axial portion of the secondopisthosomal tergite (Anderson amp Selden 1997)2 Single dorsal sclerite covering entire dorsal surfaceof body no lines indicating original segmentation 0absent 1 presentState 1 occurs throughout Holothyrida (Acari) (Vander Hammen 1989) and in many Mesostigmata(Alberti amp Coons 1999) including Glyptholaspis (Vander Hammen 1989)3 Anterior end of dorsal prosoma with medianmarginal or submarginal pointed process 0 absent1 presentState 1 occurs throughout Schizomida (Protoschizo-mus Stenochrus Reddell amp Cokendolpher 1995) The-lyphonida (Rowland amp Cooke 1973 Shultz unpublobs Proschizomus Dunlop amp Horrocks 19951996)Amblypygi (Charinus Millot 1949b Phrynus Shultz1999) and Plesiosiro (Dunlop 1999) Citing a figure inRowland amp Cooke (1973) Dunlop amp Horrocks assumedthat the anterior process was limited to lsquohypoctonidrsquothelyphonids and united them with Proschizomus andSchizomida If valid this would make Thelyphonidaparaphyletic However the doubtful monophyly oflsquohypoctonidsrsquo (Haupt amp Song 1996) and widespreadpresence of a submarginal anterior process in Pedi-palpi weaken this proposal (Harvey 2002)4 Ophthalmic ridges pair of longitudinal crests inter-secting or passing near the region of the lateral eyes orcomparable region where lateral eyes are absent 0absent 1 presentState 1 is present throughout Xiphosura (Anderson ampSelden 1997) Similar structures are present in Plesi-osiro (Dunlop 1999) and lsquonon-hypoctonidrsquo Thely-phonida (Mastigoproctus Rowland amp Cooke 1973)Dunlop (2002a) described apparent ophthalmic ridgesin the chasmataspidid Octoberaspis but these werenot observed in Diploaspis (Dunlop et al 2001) eitherdue to absence or inadequate preservation of speci-mens nor in Chasmataspis (Dunlop et al 2004)5 Carapace with demarcations (eg grooves scleritesphragmata) between pro- meso- or metapeltidia 0absent 1 present [S 1 WH 48 GEWB 29]State 1 is present in extant Scorpiones Cyphoph-thalmi (Hansen amp Soslashrensen 1904) and many Palpa-tores (Opiliones) including Caddo and Leiobunum

and most Pseudoscorpiones (Chamberlin 1931) Giri-bet et al (2002) coded this character as absent inCyphophthalmi (Opiliones) but a transverseprocurved groove is present throughout the group(Hansen amp Soslashrensen 1904) and represents the meso-metapeltidial border as indicated by attachment ofthe dorsal endosternal suspensor of postoral somite VIin Chileogovea oedipus and Siro exilis (Shultz unpublobserv) Kjellesvig-Waering (1986) did not explicitlydescribe such grooves in fossil scorpions although heillustrated corresponding structures in PrearcturusArchaeophonus Archaeoctonus Palaeophonus andother Palaeozoic scorpions (Kjellesvig-Waering 1986figs 14A 21E 27D 59) (coded as uncertain here) butnot in Proscorpius (contra Giribet et al 2002) Dis-tinct carapacal sclerites are present throughout Palpi-gradi Schizomida and Solifugae (6)

The condition in Acari is controversial but is codedhere as State 1 Van der Hammen (1989) consideredthe tergal region of the leg-bearing somites (postoralsomites IIIndashVI) or podosoma to be reduced andreplaced by posterior migration of the aspidosoma (ietergal region assumed to be associated with append-ages of the gnathosoma) and anterior migration of theopisthosomal tergal region to form the hysterosomaThe dorsal proterosomal and hysterosomal elementsmeet at a transverse sejugal furrow that continues lat-erally and passes ventrally between the coxae of legs 2and 3 According to this scheme the prosomandashopisthosoma border is expressed as the disjugal fur-row which passes from the ventral posterior margin ofthe podosoma anterodorsally to join the dorsal part ofthe sejugal furrow Based on the arrangement of setaeand slit sensilla (= lyrifissures) Van der Hammen(1989) interpreted the region above coxae 3 and 4 (ieregion C) as a fusion of the dorsal parts of the first twoopisthosomal somites (= postoral somites VII and VIII)in early divergent Anactinotrichida (ie Opilioacari-formes) and Acariformes (eg Alycus) These interpre-tations have been followed by many acarologistsalthough its speculative aspects are widely acknowl-edged (Evans 1992 Alberti amp Coons 1999)

From the standpoint of a general arachnologist theVan der Hammen system seems unnecessarily compli-cated it appears to have been formulated to explainbroad morphological themes in Oribatida but was thenextrapolated to other mites A more conservativelsquoarachnologicalrsquo scheme adopted here equates the pro-podosoma with the pars cephalica (= propeltidium) ofthe arachnid prosoma and region C with the pars tho-racica (= mesopeltidium + metapetidium) In fact theearliest and most recent treatments of Opilioacari-formes (With 1904 Klompen 2000) reached similarconclusions This interpretation is also consistent withevidence from Acariformes For example the regiondorsal to the coxa of legs 3 and 4 in Alycus is as readily

246 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

explained by the persistence of primitive prosomal ter-gal elements as by the supposition that these disap-peared and were replaced by two opisthosomalelements Furthermore several diverse lineages ofbasally divergent Acariformes display features that donot fit readily within the GrandjeanndashVan der Ham-men system but seem to correspond to the cephalicathoracica division of the prosoma They have a sejugalfurrow that may have demarcated the cephalicatho-racica border and a postpodosomal furrow thatappears to represent the prosomandashopisthosoma junc-tion there is no disjugal furrow Examples includepalaeosomate (especially aphelacarid) and pediculoch-elid Oribatida alicorhagiid and mircopsammidlsquoEndeostigmatarsquo and paratydeid Prostigmata (Albertiamp Coons 1999 fig 42) This organization may be sym-plesiomorphic for Acariformes but could be a conver-gence brought about by selection for enhancedflexibility6 Carapace with distinct pro- meso- or metapeltidialsclerites 0 absent 1 present - inapplicable due toabsence of pro- meso- or metapeltidial demarcations(5)State 1 occurs throughout Palpigradi Schizomida andSolifugae It also appears sporadically within Opil-iones (eg Leiobunum flavum Shultz unpubl observ)and Pseudoscorpiones (eg Pseudochiridium) (Cham-berlin 1931)7 Sejugal furrow circumferential zone of body flexi-bility that passes between the coxae of legs 2 and 3 0absent 1 presentState 1 is probably synapomorphic for Acariformes asit occurs in most lsquoendeostigmatidsrsquo including Alycus(Van der Hammen 1989) and is variously developedthroughout Sarcoptiformes (Alberti amp Coons 1999) Itis probably the primitive condition for Prostigmatabut is either absent or weakly expressed in represen-tatives included here See 5 for alternative morpholog-ical interpretations of this body region in Acari8 Prosomal ozopores 0 absent 1 present [WH 46GEWB 12]State 1 occurs throughout Opiliones and Holothyrida(Van der Hammen 1989)9 Carapacal pleural doublure 0 absent 1 present[WP 12 S 2 WH 49 GEWB 27]A carapacal pleural doublure occurs throughoutXiphosura (Stoslashrmer 1944) Eurypterida (Clarke ampRuedemann 1912 Selden 1981) and Chasmataspid-ida (Dunlop et al 2001 2004 Dunlop 2002a) and isabsent in all known Arachnida10 Cardiac lobe a longitudinal axial elevation of thecarapace 0 absent 1 presentState 1 occurs throughout Xiphosura (Anderson ampSelden 1997) and Eurypterida (Clarke amp Ruedemann1912 Selden 1981) Carapaces are not preserved suf-ficiently to determine this condition in the chasmatas-

pidids Diploaspis (Dunlop et al 2001) andOctoberaspis (Dunlop 2002a) but an apparent cardiaclobe occurs in the Chasmataspis (Dunlop et al 2004)A cardiac lobe is absent in Arachnida11 Moveable cucullus 0 absent 1 present [WP 54GEWB 22]State 1 is a synapomorphy of Ricinulei (Selden 1992)12 Medial intercoxal lsquosternalrsquo region 0 all pedalcoxae separated medially 1 anterior pedal coxae abut-ting medially posterior coxae separated 2 anteriorpedal coxae separated medially posterior coxae abut-ting 3 all pedal coxae abutting medially 4 epimeracoxae undifferentiated medially from ventral body wall[S sim3 WH sim50 GEWB sim17]Prosomal sternites appear to be those portions of theventral body wall not occupied by coxae Despite earlyattempts to assess metamerism of the ventralprosoma (protosternum deutosternum tritosternumetc) (Boumlrner 1902a Millot 1949a) no apparent mor-phological features such as borders of sclerites ormuscle attachments reliably demarcate the ventralbody wall of one somite from that of an adjacentsomite Thus coding schemes that focus on describingmetameric components of the sternal region are prob-lematic Shultz (1990) attempted to code the shape ofthe entire intercoxal region and subsequent workershave adopted this approach All pedal coxae are sepa-rated medially in Plesiosiro (Dunlop 1999) Palpigradi(Roewer 1934) Araneae (Millot 1949c) Amblypygi(Weygoldt 2000) and many Acari including Opilioac-ariformes Holothryrida Mesostigmata Ixodida Aly-cus and many Prostigmata (Van der Hammen 1989Evans 1992 Alberti amp Coons 1999) The anteriorcoxae are separated in extant Thelyphonida Schizo-mida (Hansen amp Soslashrensen 1905 Millot 1949e) andRicinulei (Pittard amp Mitchell 1972 Legg 1976) Thecoxae are fused to the ventral body wall in Sarcopti-formes (Van der Hammen 1989 Evans 1992)

Interpreting this character is complicated for theanterior coxae of Scorpiones and Opiliones and allcoxae in Xiphosura and Eurypterida due to the pres-ence of coxapophyses endites andor gnathobaseswhich form components of the preoral chambers inthese groups Thus coxae may abut medially whenthese structures are considered part of the coxae butare separate medially if these structures are ignoredAs coxapophyses etc are here coded as separate char-acters (50ndash53) I have chosen to code this character asif these structures did not exist Consequently pha-langid Opiliones are coded as having all coxae sepa-rated (Pocock 1902 Hansen amp Soslashrensen 1904) Fossilscorpions also show substantial variation in this char-acter All pairs of pedal coxae in Palaeoscorpius abutmedially but also have gnathobases (Jeram 1998)and this condition is coded here as uncertain A ster-num separates all pairs of pedal coxae medially in

PHYLOGENY OF THE ARACHNID ORDERS

247

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Proscorpius

and

Stoermeroscorpio

(State 0) and

Pre-arcturus

approximates modern scorpions (State 1)(Kjellesvig-Waering 1986 Jeram 1998) The coxalbases of Eurypterida and Xiphosura are separated bya small sternite the endostoma (Selden 1981 Shultz2001)

13 Postoral sternapophysis (

=

tritosternum labium)a cuticular evagination of the ventral body wall poste-riorly adjacent to the palpal coxae forming the poste-rior border of the preoral chamber in some taxa ordisplaced posteriorly by fusion of the palpal coxae inothers 0 absent 1 present

[S

sim

10 WH

sim

56 GEWB19

=

33]State 1 is present throughout Palpigradi AraneaeAmblypygi (Snodgrass 1948) Schizomida (Hansen ampSoslashrensen 1905 Van der Hammen 1989) Thely-phonida (Van der Hammen 1989) extant Ricinulei(Pittard amp Mitchell 1972 Legg 1976) Opilioacari-formes and Mesostigmata (Van der Hammen 1989Alberti amp Coons 1999)

Allothyrus constrictus

is theonly holothyrid mite known to have sternapophysesAn apparent sternapophysis has been described in apalaeocharinid trigonotarbid (Dunlop 1994) Giribet

et al

(2002) miscoded Ixodida and Acariformes as hav-ing a lsquolabium (

=

tritosternum)rsquo The lophognath ofpseudoscorpions (Chamberlin 1931) and lsquolabiumrsquo ofSolifugae (Roewer 1934) also appear to be sternapo-physes that have been incorporated into the ros-trosoma (

32

) The labium of phalangid Opiliones(Pocock 1902 Hansen amp Soslashrensen 1904) is a ster-napophysis but is associated with leg 1 rather thepedipalp (Winkler 1957) A small anterior sclerite inthe fossil scorpions

Proscorpius

Waeringoscorpio

and

Labriscorpio

has been interpreted as a lsquolabiumrsquo homol-ogous with the sternapophyses of other arachnids(Weygoldt 1998) despite absence of any indication ofthe relationship of this sclerite to the mouth and pos-sible alternative interpretations (eg

16

) Giribet

et al

coded this sclerite as a sternapophysis but the char-acter is coded here as uncertain in fossil scorpions

14 Channels on the body surface linking openings ofcoxal organs to preoral chamber 0 absent 1 present

Van der Hammen (1989) described bilaterally pairedcuticular tracts (lsquotaenidiarsquo) connecting orifices of coxalorgans to the preoral chamber in some arachnids andor a ventromedian groove (

=

intercoxal or subcapitulargutter) that leads in turn to the preoral chamber inother taxa Taenidia and intercoxal gutters arepresent in Opilioacariformes (Van der Hammen1989) Holothyrida (Van der Hammen 1989) Mesos-tigmata (

Glyptholapis

Van der Hammen 1989)Ricinulei (Pittard amp Mitchell 1972 Van der Hammen1989) Araneae (

Heptathela

Aphonopelma

Hypochi-lus

Shultz unpubl observ also

Porrhothele

(Mygalo-morphae Dipluridae) Butt amp Taylor 1991

Segestria

(Araneomorphae Segestriidae) Van der Hammen

1989) Amblypygi (Van der Hammen 1989) Schizo-mida (

Stenochrus

Shultz unpubl observ unidenti-fied Hubbardiinae Van der Hammen 1989) andThelyphonida (

Mastigoproctus

Shultz unpublobserv also

Tetrabalius

Van der Hammen 1989) Theconduction of large volumes of fluid from orifices of thecoxal organs to the preoral chamber has been observedin the mygalomorph spider

Porrhothele

(Dipluridae)(Butt amp Taylor 1991) and

Mastigoproctus

(Thely-phonida) (Shultz unpubl observ) The podocephaliccanal (

15

) is present in Acariformes (Acari) (Lindquist1984) and appears to have evolved as an invaginatedsupracoxal channel (Grandjean 1938 G Alberti inEvans 1992) Van der Hammen (1982) could not dis-cern with certainty whether a channel exists in

Eukoenenia

(Palpigradi) but none has been reportedby previous authors and they have not been observedby the present author Palpigradi is coded as havingState 0

15 Podocephalic canal cuticular channel andorduct draining multiple glands and opening nearmouthparts 0 absent 1 present - inapplicable due toabsence of channel (14)

State 1 occurs throughout Acariformes (Acari)(Lindquist 1984)

16 Heavily sclerotized suboral sclerite serving inpart as basal pivot point for coxae of appendages ofpostoral somites IIndashIV (

=

arachnid palp and legs 1 and2) 0 absent 1 present

The sclerite forms the posterior wall of the true mouthand is heavily sclerotized in extant scorpions It is notvisible externally due to the tightly fitting coxae of theanterior prosomal appendages (Shultz 2007) It maycorrespond to the so-called labium (Weygoldt 1998) orlabrum (Kjellesvig-Waering 1986) of certain fossilscorpions (see

13

)

17 Genal angles 0 rounded 1 pointed

Anderson amp Selden (1997) originally coded a pointedgenal angle and genal spine as separate charactersbut they are combined here State 1 is present in manyXiphosura (Anderson amp Selden 1997) and Chas-mataspidida (Dunlop

et al

2001 2004 Dunlop2002a)

A

PPENDAGES

OF

POSTORAL

SOMITE

I

CHELICERAE

18 Cheliceral segmentation 0 three articles 1 twoarticles

[WP

sim

25

+

42 WH

sim

15 GEWB

sim

44]State 0 is present in extant Xiphosura Eurypterida(Clarke amp Ruedemann 1912 Selden 1981) Palpi-gradi (Roewer 1934) Opiliones (Hansen amp Soslashrensen1904) Scorpiones (Hjelle 1990) and throughout non-acariform Acari (Evans 1992) State 1 is presentthroughout known Trigonotarbida Araneae Ambly-pygi Thelyphonida Schizomida Pseudoscorpionesand Solifugae Acariform mites have two unambigu-

248 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ous cheliceral articles but may have a reduced thirdarticle (Van der Hammen 1989 Evans 1992) acari-form mites are thus coded as uncertain Chelicerae arenot known or their condition is unclear in several fos-sil taxa including Xiphosura Chasmataspidida (Dun-lop et al 2001 2004 Dunlop 2002a) Riciniulei(Selden 1992) Plesiosiro (Dunlop 1999) Proschizo-mus (Dunlop amp Horrocks 19951996) and the fossilscorpions Praearcturus and Palaeoscorpius (Kjelles-vig-Waering 1986) Kjellesvig-Waering (1986) consis-tently interpreted chelicerae of fossil scorpions ashaving four articles but this is probably a misinter-pretation of the basal article (see also Stockwell1989) The basal article in extant scorpions consists ofa distal collar of cuticle and a large proximal processand this arrangement would probably appear as twoarticles in compressed specimens In any event thereare at least three cheliceral articles in Stoermeroscor-pio and Proscorpius (Kjellesvig-Waering 1986)19 Terminal cheliceral joint 0 laterally placedbicondylar hinge 1 dorsally placed bicondylar hinge2 ventrally placed bicondylar hinge [WP 25 + 42 Ssim11 + 12 WH sim15 GEWB sim44]State 0 is present in Xiphosura (Shultz 2001)extant Scorpiones (Millot amp Vachon 1949) Opil-iones (Hansen amp Soslashrensen 1904) Palpigradi(Boumlrner 1904 Millot 1949d) Ricinulei (Pittard ampMitchell 1972 Legg 1976) and Eurypterida includ-ing Baltoeurypterus (Selden 1981) and Stylonurus(Clarke amp Ruedemann 1912) State 0 is present inthe fossil scorpions Proscorpius and Stoermeroscor-pio but the state of the chelicerae is unclear in theother fossil scorpions represented here (Kjellesvig-Waering 1986) State 1 is present in Trigonotarbida(Shear et al 1987) mygalomorph Araneae Ambly-pygi Schizomida and Theylphonida (Weygoldt ampPaulus 1979) State 2 is present in Pseudoscorpi-ones (Chamberlin 1931) and Solifugae (Roewer1934) and appears to be a groundplan feature for allmajor groups of mites (eg Van der Hammen 1989Alberti amp Coons 1999) The difference between theorthognathy of mygalomorph Araneae and lsquoplagiog-nathyrsquo mesothele (Liphistius Heptathela) and palae-ocribellate Araneae (Hypochilus) (Kraus amp Kraus1993) is small compared with the variation thatoccurs throughout arachnids and plagiognathy iscoded as State 1 here20 Chelicera articulating with carapace at anterolat-eral pivot 0 absent 1 present [S 13 WH 57 GEWB45]State 1 occurs in Solifugae (Roewer 1934) and Pseu-doscorpiones except Chthonioidea and Feaelloidea(Chamberlin 1931) Giribet et al (2002) coded thecharacter as present in Pseudoscorpiones21 Chelicera pivoting on dorsal protuberance of epis-tome 0 absent 1 present

State 1 appears to be a unique synapomorphy of Opil-iones (Shultz 2000)22 Extrinsic cheliceral muscle arising on carapaceand inserting on dorsal margin of nonbasal cheliceralarticle 0 absent 1 present (Note Characters 22 and23 refer to two different muscles)State 1 is known only in extant Scorpiones (LankesterBenham amp Beck 1885 Vyas 1970 1974 Shultz2007) but may occur in acariform mites if their cheli-cerae are composed of three rather than two articles(Evans 1992) (see 18) Acariformes are coded asuncertain23 Extrinsic cheliceral muscle arising on carapaceand inserting on ventral margin of nonbasal cheliceralarticle 0 absent 1 present [GEWB 252] (Note Char-acters 22 and 23 refer to two different muscles)State 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) somemesostigmatid Acari (Evans 1992) and representa-tive Opiliones particularly Leiobunum (Shultz 2000)Chileogovea and Gonyleptes (Shultz unpubl observ)The status of the character in other opilions is not yetknown The character was originally defined as anextrinsic muscle inserting on the second segment ofthree-segmented chelicerae (Shultz 2000 Giribetet al 2002) but it has been redefined here to make itapplicable to arachnids with two-segmented cheli-cerae24 Extrinsic cheliceral muscles attaching to epistome0 absent 1 presentState 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) but mayoccur in Prostigmata if the sigmoid piece to which thecheliceral protractors attach (Evans 1992) is a modi-fication of the epistome25 Lateral tergocheliceral muscle with three heads 0absent 1 present - coded only for extant tetrapulmo-nates and the palpigrade Eukoenenia homology isunclear in other taxa [GEWB 230]State 1 is present in Pedipalpi (Shultz 1999)26 Cheliceral silk glands 0 absent 1 present [GEWB47]State 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931) The silk gland of spider mites(eg Tetranychus) are associated with the pedipalps(Evans 1992)27 Cheliceral venom glands 0 absent 1 present [WP34 WH 40 GEWB 46]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)28 Cheliceral serrula interior and exterior 0 absent1 presentState 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931 Harvey 1992)29 Cheliceral lsquoflagellumrsquo in male 0 absent 1 present[GEWB 48]

PHYLOGENY OF THE ARACHNID ORDERS 249

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITE II LEG 1 IN NONARACHNIDS AND PALP IN ARACHNIDS

30 Palpal coxae fused ventromedially and formingposterior wall of preoral chamber 0 absent 1 present[WP 27 S 18 WH 17 = 62 GEWB 33 + 63]State 1 occurs in Schizomida Thelyphonida Acari andRicinulei31 Gnathosoma 0 absent 1 present - inapplicablepalpal coxae not fused (30) [WP 56 S 19 WH sim63GEWB sim36 + 37]The gnathosoma is a functional complex comprisingan epistome attached to the dorsal surface of mediallyfused palpal coxae (= infra- or subcapitulum) (see 30)the chelicerae and in some taxa a supracheliceral tec-tum These components are tightly integrated andmove in unison relative to the body The gnathosomahas traditionally been considered a feature of AcariHowever Van der Hammen (1989) argued that a gna-thosoma is present in Ricinulei and this interpreta-tion has received wide acceptance (eg Lindquist1984) However it is not clear that the relationship ofthe chelicerae and subcapitulum is as intimate inRicinulei as in Acari and the ricinuleid subcapitulumdoes not appear to be as moveable In fact the lsquosubca-pitulumrsquo in Ricinulei is similar in many respects to thecondition in Schizomida and Thelyphonida (Hansen ampSoslashrensen 1904) which has never been regarded as agnathosoma Consequently I have coded the conditionof Ricinulei as uncertain32 Rostrosoma long narrow subcylindrical epistomeprojecting anteriorly with base fixed to dorsal surfaceof palpal coxae bordered laterally by lobes projectingfrom palpal coxae ventral wall of preoral chamberformed by anterior element of prosoma (sternapophy-sis) 0 absent 1 present [S 15 WH 59 GEWB 20 + 36]State 1 is present in Pseudoscorpiones and Solifugae(Chamberlin 1931 Roewer 1934 Van der Hammen1989) Dunlop (2000) attempted to homologize com-ponents of the rostrosoma and gnathosoma (31) anargument accepted by Giribet et al (2002) Howeverfusion of the palpal coxae (30) apparently evolvedindependently in the two structures Coxal fusion inthe gnathosoma is complete ventrally and excludesthe suboral sternapophysis (13) while fusion isincomplete ventrally in the rostrosoma and incorpo-rates the sternapophysis Unlike the gnathosomathe rostrosoma is largely immobile and cheliceralmovement is not coupled with that of the palpalcoxae33 Rutellacorniculi hypertrophied setae modified asmouthparts located on the anterior processes of the pal-pal coxae 0 absent 1 present [GEWB 39]

State 1 occurs in the anactinotrichid groups Opilioac-ariformes Holothyrida and Mesostigmata but not Ixo-dida (Van der Hammen 1989) It also occurs inSarcoptiformes (Evans 1992) and certain endeostig-matids (eg Alycus Van der Hammen 1989) but not inProstigmata Comparable structures are unknown inother chelicerate groups34 Terminal segments specially modified in adultmale as a clasper used to engage the female 0 absent1 presentState 1 is known only in extant Xiphosura (YamasakiMakioka amp Saito 1988)35 Robust raptorial 0 absent 1 present [S 17 WH61 GEWB sim59 97]State 1 is present in Amblypygi Thelyphonida andSchizomida (Shear et al 1987) and in Laniatores(Opiliones)36 Orientation of robust raptorial appendage 0operating in subtransverse plane 1 operating in sub-vertical plane - inapplicable coded only for PedipalpiState 0 is present in extant Thelyphonida and mostAmblypygi and State 1 is present in Schizomida (Cok-endolpher amp Reddell 1992) and in the basally diver-gent amblypygid Paracharon caecus (Weygoldt 2000)The orientation of the pedipalps in the fossil Proschi-zomus is uncertain (Dunlop amp Horrocks 19951996)37 Extrinsic muscle attaching to epistome 0 absent1 presentState 1 occurs in Chileogovea and Siro (OpilionesCyphophthalmi) and extant Scorpiones (Vyas 1970Shultz 2007)38 Muscle originating and inserting within coxa 0absent 1 present [GEWB sim232]State 1 occurs is in Amblypygi (Phrynus Shultz 1999Phrynichus Boumlrner 1904) Thelyphonida (Mastigo-proctus Shultz 1993 Thelyphonus Boumlrner 1904) andSchizomida (Stenochrus Shultz unpubl observ)There are two intracoxal muscles in amblypygids andthelyphonids the one coded here and another that isserially homologous with a muscle in the pedal coxae(54)39 Tarsus andor tibia with venom glands 0 absent1 present [GEWB 64]State 1 is known only from extant Pseudoscorpi-ones except the superfamilies Chthonioidea andFeaelloidea (Chamberlin 1931 Harvey 1992) Giri-bet et al (2002) coded this character as present forPseudoscorpiones40 Tarsal grooming organ 0 absent 1 present[GEWB 98]State 1 is a unique synapomorphy of Amblypygi (DelleCave 1975 Weygoldt 2000)41 Scorpionoid chela a large well-developed chelaformed by tibia (manus + fixed finger) and tarsus(moveable finger) 0 absent 1 present [S 16 WH 60GEWB 62]

250 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs throughout Scorpiones andPseudoscorpiones42 Modified in male as copulatory organ 0 absent 1present [WP 35 WH 39 GEWB 67]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)43 Apotele 0 apparently absent not differentiatedexternally from penultimate article 1 present [GEWB69]The apotele exists as a structure (ie dactyl clawempodium) differentiated from the tibiotarsus tarsusor telotarsus and is typically associated with a pair ofantagonistic muscles The apotele occurs as a distinctstructure in Xiphosura (Snodgrass 1948) Eurypter-ida (Clarke amp Ruedemann 1912) Schizomida (Coken-dolpher amp Reddell 1992) Araneae (Foelix 1996)early divergent Amblypygi (Charinus Weygoldt 2000not Phrynus Shultz 1999) Anactinotrichida (CaminClark amp Bourdeau 1956 Lindquist 1984) Ricinulei(Pittard amp Mitchell 1972 Legg 1976) and Solifugae(Roewer 1934) Giribet et al (2002) coded Solifugae asnot having a palpal apotele but it is present as anadhesive organ (45) Dunlop (1999) reconstructedHaptopoda as having a palpal claw but notes in hisdescription that the distal ends were not preserved inthe known specimens44 Apotele (claw) position 0 terminal 1 subtermi-nal - inapplicable coded only for taxa with a distinctapotele (43)State 1 occurs in Parasitiformes (Acari) (Evans 1992)45 Terminal adhesive organ 0 absent 1 present -inapplicable due to lack of an apotele (43) [GEWB 66]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITES IIIndashVI LEGS 2ndash5 IN NON-ARACHNIDS AND LEGS 1ndash4 IN ARACHNIDS

46 Appendage III (= arachnid leg 1) extremely elon-gate antenniform 0 absent 1 present [WP sim24 + 32S 20 WH 14 GEWB sim84]State 1 occurs in Amblypygi Thelyphonida Schizo-mida (Shear et al 1987) and Opilioacariformes (Vander Hammen 1989) State 1 is approximated in Solif-ugae Palpigradi (Roewer 1934) and Haptopoda(Dunlop 1999) and it is coded as uncertain in thesetaxa47 Appendage V (= arachnid leg 3) of male specializedfor sperm transfer 0 absent 1 present [WP 55 GEWB94]State 1 is known only from extant Ricinulei48 Appendages V and VI (= arachnid legs 3 and 4)with femur shorter than patella and with principal siteof flexionextension at patella-tibia joint (lsquoapatellatersquocondition sensu Van der Hammen 1989) 0 absent 1present [S 25 GEWB 86]

State 1 occurs in Pseudoscorpiones and Solifugae(Shultz 1989)49 Appendage VI (= arachnid leg 4) with terminusmodified as flattened blade 0 absent 1 presentState 1 occurs in some Chasmataspidida (DiploapisDunlop et al 2001 Octoberaspis Dunlop 2002a)uncertain in Chasmataspis Dunlop et al 2004 State0 in Loganamaraspis Tetlie amp Braddy 2004) andEurypterida except stylonuroids (Clarke amp Ruede-mann 1912 Waterston 1979 Tollerton 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI COXA AND BODYndashCOXA JOINT

50 Appendage III (= arachnid leg 1) with coxapophy-sis forming floor or wall of preoral chamber 0 absent1 present [S sim14 WH sim58]State 1 occurs in Opiliones (Hansen amp Soslashrensen 1904)and higher scorpions (Jeram 1998) A lobe is presenton the coxa of leg 1 in the fossil scorpion Prearcturusand perhaps Stoermeroscorpio and is coded here as acoxapophysis Coxapophysis of leg 1 together with theepistome and coxapophysis of the palp form a uniquepreoral chamber the stomotheca Weygoldt (1998) andDunlop amp Braddy (2001) reject the homology of cox-apophyses and stomothecae in Opiliones and Scorpi-ones because these structures are not apparent infossils of those scorpions thought to have been aquaticHowever the coxapophyses in Opiliones are formedlargely from soft cuticular lsquolipsrsquo and similar structuresin fossil scorpions would probably not have beenpreserved51 Appendage IV (= arachnid leg 2) with coxapophy-sis 0 absent 1 present [S sim14 WH sim58 GEWB 75]State 1 occurs in extant Scorpiones and most Opil-iones except Dyspnoi (Shultz 1998)52 At least one pair of coxal gnathobases on append-ages III-VI (= arachnid legs 1ndash4) 0 absent 1 present[S sim23 WH sim66 GEWB 78]State 1 occurs in all appropriately preserved Euryp-terida including Baltoeurypterus (Selden 1981) andStylonurus (Clarke amp Ruedemann 1912) extantXiphosura on appendages IIndashVI and apparentlyProscorpius on appendage III (= arachnid leg 1)(Kjellesvig-Waering 1986 Jeram 1998 contra Giribetet al 2002) Fossilized appendages with gnathobaseshave been found in association with Chasmataspis(Chasmataspidida) suggesting that they may havehad State 1 (Dunlop et al 2004)53 Coxae of appendages IIIndashV (= arachnid legs 1ndash3)with jointed moveable endites 0 absent 1 present [Ssim23]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) and in adequately preserved eurypterids(Clarke amp Ruedemann 1912) (BaltoeurypterusSelden 1981)

PHYLOGENY OF THE ARACHNID ORDERS 251

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

54 Intracoxal muscle a muscle arising on anteriorwall of coxa and inserting on posterior wall 0 absent1 present [GEWB 237]State 1 occurs in Thelyphonida (MastigoproctusShultz 1993) and Amblypygi (Phryus Shultz 1999)It is unclear whether the muscle operating the move-able endite (53) in extant Xiphosura (Manton 1964Shultz 2001) is homologous with this muscle and iscoded here as uncertain It is not present in Steno-chrus (Schizomida) (Shultz unpubl observ)55 Coxae of appendages IIndashVI (= arachnid palp andlegs) with dorsal articulation with carapace 0 absent1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)56 Flabellum (exite) on coxa of appendage VI(= arachnid leg 4) 0 absent 1 present [GEWB sim110]State 1 occurs in extant Xiphosura (Yamasaki et al1988) An apparent exite occurs on appendages asso-ciated with Chasmataspis (Chasmataspidida) (Dunlopet al 2004)57 Insertion process of anteromedial tergo-coxal mus-cle 0 weakly developed 1 well to extremely well devel-oped [GEWB 238]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Boumlrner 1904 Shultz 1999)58 Musculi laterales enlarged lateral tergocoxalmuscle with attachment shifted from coxa to adjacentpleural membrane 0 absent 1 present [S 22 WH 65GEWB 242]State 1 occurs in Araneae and Thelyphonida (Shultz1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI TROCHANTER AND COXAndashTROCHANTER JOINT

59 Coxandashtrochanter joint with complex posteriorarticulation composed of two articulating sclerites 0absent 1 present [S 24 WH 67 GEWB 85]State 1 occurs in Araneae Amblypygi Thelyphonidaand Schizomida (Shultz 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI FEMUR AND TROCHANTERndashFEMUR JOINT

60 Depressor muscle (or homologue) of trochanter-femur joint 0 absent 1 present [WH 42 GEWB 220]State 0 appears to be a unique synapomorphy of Ara-neae (Liphistius Shultz 1989 Aphonopelma Ruh-land amp Rathmayer 1978 Hypochilus present studyother examples Clarke 1984 1986)61 Trochanterndashfemur joint with dorsal hinge or pivotoperated by flexor muscles only 0 absent 1 present[WH sim44 GEWB sim91]State 1 occurs in Palpigradi Other chelicerates have abicondylar articulation and are typically equipped

with antagonistic muscles (Van der Hammen 1985Shultz 1989) but see 60 for an exception62 Superior trochanterndashfemur muscle (or homologue)originating broadly in femur inserting on distal mar-gin of trochanter 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 19892001)63 Basifemurndashtelofemur joint of appendages III andIV (= arachnid legs 1 and 2) in adult 0 absent 1present [S 25 WH 68 GEWB 86]The basifemur and telofemur apparently develop aslsquosisterrsquo articles through division of a parental articlethe femur (eg opilioacariform endeostigmatidpalaeosomatid mites) (Coineau amp Van der Hammen1979 Evans 1992) A single animal can have all legswith divided or undivided femora or have some com-bination of the two typically with the more posteriorlegs having the divided femora Thus anterior legsoften tend to be paedomorphic with respect toposterior legs Most chelicerates have undivided fem-ora on some or all legs but muscles homologous withthose of the basifemurndashtelofemur joint may stilldevelop (66) (Shultz 1989) suggesting that the devel-opment of cuticular and muscular components of ajoint have a degree of developmental and evolutionaryindependence

Undivided femora occur in most extant euchelicer-ates including Xiphosura Palpigradi AraneaeAmblypygi Thelyphonida Schizomida ScorpionesOpiliones Pseudoscorpiones higher Sarcoptiformes(Archegozetes Van der Hammen 1989) many Prostig-mata (Tetranychus Evans 1992) and throughout Par-asitiformes Divided femora occur in all legs in manyearly divergent Acariformes including many Endeo-stigmata basally divergent Sarcoptiformes (egPalaeacarus) and Prostigmata (eg AllothrombiumMicrocaeculus) (Van der Hammen 1989 Evans1992) Only leg 4 has basi- and telofemora in Alycus(Van der Hammen 1989) Basi- and telofemora occuron legs 3 and 4 in Solifugae Ricinulei Opilioacari-formes and Eurypterida (Clarke amp Ruedemann 1912Selden 1981 Shultz 1989)

Giribet et al (2002) coded the lsquoannulusrsquo of certainTrigonotarbida (Shear et al 1987) as a basifemur buta similar non-muscularized annulus is present in theposterior legs of extant Thelyphonida In both casesthe condyles of the trochanterndashannulus and annulusndashfemur joints are aligned in such away that they per-form essentially identical movements and appear notto enhance either the range or degree of leg move-ment [This contrasts with the condition in the lasttwo pairs of legs in Baltoeurypterus where condyles ofthe apparent trochanter-basifemur and basifemurndashtelofemur joints have different arrangements (Selden1981)] The annulus in Thelyphonida appears to func-tion in maintaining constant volume of the tro-

252 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

chanterndashfemur joint by preventing ballooning of thearthrodial membrane thereby minimizing the effect ofinternal fluid pressure on movement64 Basifemurndashtelofemur joint of appendage V(= arachnid leg 3) in adult 0 absent 1 present See 6365 Basifemurndashtelofemur joint of appendage VI(= arachnid leg 4) in adult 0 absent 1 present See6366 Cuticular differentiation of basifemur-telofemurjoint absent but muscles present 0 absent 1 present- inapplicable due to presence of joint (63ndash65)State 1 occurs in appendages of postoral somites IIIndashIV (= arachnid legs 1ndash4) in Xiphosura AraneaeAmblypygi Thelyphonida Schizomida Scorpionesand Pseudoscorpiones (Shultz 1989)67 Circumfemoral ring 0 absent 1 presentThe femora of one or more pairs of legs in anactinot-richid Acari have a basal groove or ring associatedwith slit sensilla Acarologists term that portionproximal to the ring the lsquobasifemurrsquo and that portiondistal to the ring the lsquotelofemurrsquo and they refer tothe basifemur as the lsquosecond trochanterrsquo and thetelofemur as the lsquofemurrsquo Despite the confusingterminology the basifemur discussed in 63ndash65 is nothomologous with the lsquobasifemurrsquo of opilioacariformand parasitiform Acari as both a true basifemurndashtelofemur joint and a circumfemoral ring are presentin legs 3 and 4 of Opilioacariformes (Van der Ham-men 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI PATELLA AND FEMURndashPATELLA JOINT

68 Patella of appendage of postoral somite III(= arachnid leg 1) proportionally much longer thanthose of more posterior appendages 0 absent 1presentState 1 occurs in Thelyphonida Schizomida (Hansenamp Soslashrensen 1905 Shultz 1989) and Opilioacari-formes (Van der Hammen 1989)69 Femurndashpatella joint 0 monocondylar several axesof movement and multifunctional muscles 1 bicondy-lar hinge one axis of movement and antagonistic mus-cles 2 hinge one axis of movement flexor muscleswithout muscular antagonists [S sim26 WH sim69 GEWBsim87]State 0 occurs in Solifugae State 1 occurs in Phalang-ida (Opiliones) Scorpiones and PseudoscorpionesState 2 occurs in the remaining taxa (Shultz 1989)70 Patellar plagula 0 absent or with simple medianattachment 1 symmetrical Y-shaped with long prox-imal stem 2 symmetrical U-shaped (= arcuate scler-ite) 3 asymmetrical attaching to patella only atanterior margin [S 27 WH sim70 GEWB 234 = 243]State 1 occurs in extant Xiphosura (Shultz 1989)State 2 occurs in Araneae (Manton 1958 Ruhland amp

Rathmayer 1978 Clarke 1986 Shultz 1989) State 3occurs in Amblypygi Thelyphonida and Schizomida(Shultz 1989) Giribet et al (2002) included this char-acter twice

APPENDAGES OF POSTORAL SOMITES IIIndashVI TIBIA AND PATELLAndashTIBIA JOINT

71 Tibiae divided by one or more joints 0 absent 1presentState 1 occurs in Amblypygi (Weygoldt 2000)72 Patellandashtibia joint 0 monocondylar with or with-out CZY (73) 1 bicondylar hinge one axis of move-ment antagonistic muscles 2 hinge one axis ofmovement muscles without muscular antagonists [Ssim31 WH 74 GEWB sim92] (Shultz 1989 Selden Shearamp Bonamo 1991)73 Patellandashtibia joint with posterior compression zone(lsquoCZYrsquo) 0 absent 1 present - inapplicable due toabsence of monocondylar articulation [GEWB 99]State 1 is a unique synapomorphy of Araneae (Seldenet al 1991)74 Patellandashtibia joint of appendages IIIndashVI(= arachnid legs 1ndash4) with deep-set monocondylar pivotbordered by a pair of tibial processes to which extensormuscles attach 0 absent 1 present - inapplicabledue to absence of monocondylar joint (72)State 1 occurs in extant Xiphosura (Snodgrass 1952Shultz 1989) although a similar arrangement ispresent in the appendage III (= arachnid leg 1) in The-lyphonida (Shultz 1993)75 Patellandashtibia joint largely immobile specializedfor autotomy 0 absent 1 presentState 1 occurs throughout Amblypygi (Weygoldt 19842000)76 Anterior femurndashtibia or femoropatellandashtibia(transpatellar) muscle 0 absent 1 present [S sim30GEWB 222] (Shultz 1989)77 Proximal attachment of posterior femurndashtibia orfemoropatellarndashtibia (transpatellar) muscle 0 muscleabsent 1 dorsoposterior surface of femur andor pos-terior surface of patella 2 distal process of femur mus-cle acting as extensor of femur-patella joint [S sim28 WHsim71 + 72 GEWB 235 = 244 245] (Shultz 1989)78 Distal attachment of posterior femurndashtibia (trans-patellar) muscle 0 posterior 1 dorsal acting as exten-sor of patellandashtibia joint - inapplicable due to absenceof muscle (77) [S 29 GEWB 221] (Shultz 1989)79 Anterior patellandashtibia muscle 0 absent 1 present[S sim32 WH sim73 GEWB 246] (Shultz 1989)80 Posterior patellandashtibia muscle 0 absent 1 present[S 33 WH 76 GEWB 248] (Shultz 1989)81 Patellandashtibia joint spanned by elastic (lsquospringlikersquo)sclerite 0 absent 1 present [GEWB 107]State 1 is a unique synapomorphy of Solifugae (Sens-enig amp Shultz 2003)

PHYLOGENY OF THE ARACHNID ORDERS 253

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

APPENDAGES OF POSTORAL SOMITES IIIndashVI TARSUS AND TIBIAndashTARSUS JOINT

82 Tarsus divided into proximal basitarsus(= metatarsus) and distal telotarsus (= distitarsus orlsquotarsusrsquo) 0 absent 1 presentState 0 occurs in extant Xiphosura (Shultz 1989) andthroughout Acariformes (Lindquist 1984 Evans1992) in the anterior two pairs of legs in chthonioidpseudoscorpions and all legs in lsquomonosphyronidrsquo pseu-doscorpions including Feaelloidea and Cheliferoidea(Chamberlin 1931) A circumtarsal ring (83) associ-ated with slit sensilla in Parasitiformes (Acari) mayrepresent a joint between a telotarsus and basitarsus(Evans 1992)83 Circumtarsal ring 0 absent 1 presentState 1 is an apparent synapomorphy of Parasiti-formes (Evans 1992)84 Telotarsus in adult with two or more tarsomeres 0absent 1 present - inapplicable due to absence of telo-tarsus (82) [GEWB 71]State 1 occurs in Schizomida Thelyphonida Ambly-pygi Palpigradi Ricinulei Solifugae (Shultz 1989)Opilioacariformes (Van der Hammen 1989) and Ple-siosiro (Dunlop 1999)85 Three telotarsomeres on appendages of postoralsomites IV-VI (= arachnid legs 2ndash4) 0 absent 1present - inapplicable due to absence of telotarsus (82)or absence of telotarsomeres (84) [S 34 WH 77 GEWB102]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Shultz 1989)86 Tibiandashtarsus joint spanned by well-developed elas-tic (lsquospringlikersquo) sclerite 0 absent 1 present [GEWB107]State 1 occurs in Scorpiones phalangid Opiliones andSolifugae (Alexander 1967 Shultz 2000 Sensenig ampShultz 2003) Extant Xiphosura is coded for the lastprosomal appendage only (see 88)87 Appendage of postoral somite VI (= arachnid leg 4)with ring of large basally articulated spatulate pro-cesses at tibia-tarsus joint 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)88 Appendages postoral somites IIIndashV (= arachnidlegs 1ndash3) with tibiotarsus (ie tibia and tarsus not dif-ferentiated) 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 1989) butnot apparently in Weinbergina (Moore et al 2005) theonly synxiphosurid in which appendages are suffi-ciently preserved to determine the number ofpodomeres89 Ambulacrum peduncle-like extension of the tarsuswith internal condylophores and terminating distallywith apotele (eg claws) andor pulvillus 0 absent 1present

State 1 is present throughout Acari (Alberti amp Coons1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI APOTELE AND TARSUSndashAPOTELE JOINT

90 Apotele of appendage III (= arachnid leg 1) 0absent or not apparent 1 present [GEWB 101]State 0 occurs in Amblypygi Thelyphonida and Schi-zomida (Shultz 1999) However the apotele is proba-bly present but undifferentiated because muscles thatwould normally insert on the apotele terminate on theend of the lsquotarsusrsquo Dunlop (1999) reconstructed Hap-topoda as lacking an apotelic claw on leg 1 but noted inthe text that this was speculative Dunlop (2002b) hasreviewed apotelic diversity in Chelicerata91 Appendages of postoral somites IIIndashVI (= arachnidlegs) chelate with chela formed from tibiotarsus andapotele or tarsus and apotele 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988) The condition appears to occur in Chasmataspisbased on associated appendages (Dunlop et al 2004)but not in Diploaspis and Octoberaspis (Dunlop et al2001 Dunlop 2002a)92 Apotele with eversible or padlike empodium(= pulvillus) in adult 0 absent although empodialclaw may be present 1 present [S sim38 WH 81 GEWB95]State 1 occurs throughout Pseudoscorpiones (Cham-berlin 1931) Solifugae (Roewer 1934) Opilioacari-formes Holothyrida Ixodida and free-livingMesostigmata (Evans 1992) An eversible lsquopulvillusrsquo isprobably a primitive feature of Amblypygi (eg Char-inus Weygoldt 2000) but is absent in higher groups(eg Phrynus Weygoldt 2000)93 Inferior apotele muscle (= claw depressor) with tib-ial attachment 0 absent 1 present [S 35 WH 78GEWB 249]The character is coded for extant Xiphosura usingappendage of postoral somite VI (= arachnid leg 4) asother legs lack a differentiated tibia (88) State 1appears to occur throughout Arachnida (Shultz 1989)94 Inferior apotele muscle (= claw depressor) withpatellar attachment 0 absent 1 present [S 36 WH 79GEWB 250]State 0 occurs in extant Xiphosura Acari AmblypygiPalpigradi and Ricinulei State 1 occurs in all otherarachnids examined here (Shultz 1989)

OPISTHOSOMA

95 Number of opisthosomal somites in adult 0 five1 eight 2 nine 3 10 4 11 5 12 6 13 [WH sim33GEWB sim190]Chasmataspidids appear to have 13 opisthosomalsomites (Dunlop et al 2004) The synziphosurans

254 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Weinbergina (Moore et al 2005) and Limuloides(Anderson amp Selden 1997) appear to have ten opistho-somal somites Moore et al (2005) found no evidence ofan anterior microtergite in Weinbergina as proposedby Anderson amp Selden (1997) Xiphosurids appear tohave nine opisthosomal somites (Scholl 1977 Ander-son amp Selden 1997 Shultz 2001) Members of the fol-lowing orders have 12 opisthosomal somites Araneae(Millot 1949ac) Amblypygi (Weygoldt 2000) Thely-phonida and Schizomida (Hansen amp Soslashrensen 1905)and Pseudoscorpiones (Chamberlin 1931) Elevensomites are present in Palpigradi Solifugae (Roewer1934) and Trigonotarbida (Gilboarachne Shear et al1987 Palaeocharinus Fayers et al 2004) Plesiosiromay have 11 or 12 somites (Dunlop 1999) Justifica-tion for coding of more controversial taxa is providedbelow

Scorpiones Two principal hypotheses regarding thenumber of opisthosomal somites in scorpions havebeen advocated a 13-somite hypothesis derived fromembryological studies (Brauer 1895 Patten 1912Farley 1999 2005) and a 12-somite hypothesis basedon comparative anatomy of adults (Weygoldt amp Pau-lus 1979) The embryological interpretation is basedon the observation of pregenital genital and pectinalsomites (each with segmental ganglia and paired limbbuds) in early scorpion embryos followed by extremereduction or loss of the pregenital somite in laterembryos According to this view a missing pregenitalsomite should be added to the 12 apparent opisthoso-mal somites of post-embryonic scorpions to achieve afinal number of 13 The anatomy-based hypothesiswas introduced by Weygoldt amp Paulus (1979) whoadvocated a literal interpretation of post-embryonicsegmentation based on opisthosomal tergites Specifi-cally these authors argued that the last pair of dorsalendosternal suspensor muscles of non-scorpionarachnids especially Pedipalpi attach to the first(= pregenital) somite that this condition also occurs inscorpions and that there is no reason to invoke a miss-ing pregenital tergite They proposed that the pectinesbelong to the genital somite and that functional spe-cializations of the nervous system for pectinal functiongive the apparence of an extra neuromere duringembryonic development

I recently dissected the prosoma and anterioropisthosoma of the scorpions Centruroides Hadrurusand Heterometrus and focused on the composition ofthe muscular diaphragm (103) that separates thehaemocoelic compartments of the prosoma andopisthosoma (Shultz 2007) The diaphragm is com-posed of a metameric series of axial muscles fromthree somites the anterior somite corresponds to thelast prosomal somite and the posterior somite corre-sponds to the genital somite The middle elements

insert dorsally along a tranverse tendon attached tothe anterior margin of the first tergite These obser-vations are consistent with the embryological inter-pretation that a pregenital somite is present but itstergite is not expressed It appears likely that the thepregenital somite was compressed longitudinally dur-ing the evolution of the diaphragm I code scorpions ashaving 13 opisthosomal somites

Eurypterida The eurypterid opisthosoma is widelyassumed to have 12 somites (Clarke amp Ruedemann1912 Stoslashrmer 1944) However in a speculative paperon the evolutionary morphology of trilobites and che-licerates Raw (1957) proposed that both scorpions andeurypterids have 15 opisthosomal somites Rawassumed that scorpions have 13 apparent opisthoso-mal somites based on the transient pregenital somiteof scorpion embryology that scorpions and eurypteridsare close relatives and should have the same numberof somites and that olenellid trilobites and chelicer-ates always have somites in multiples of three Rawachieved 15 somites in scorpions by assuming the lastopisthosomal somite to be a diplosomite and that thetelson is a postanal somite even though there is noevidence for either of these proposals He attributedthese features to eurypterids as well He also notedthat the connection between the prosomal carapaceand first opisthosomal tergite in Eurypterida differedstructurally from the connection between adjacentopisthosomal tergites and regarded this as evidencefor a reduced pregenital tergite in the prosomandashopisthosoma junction Rawrsquos speculations were largelyforgotten until Dunlop amp Webster (1999) resurrectedthe proposal that eurypterids have a reduced opistho-somal tergite and therefore share a unique similaritywith scorpions Unfortunately Dunlop amp Bullocktreated Rawrsquos conjecture as if it were based on empir-ical evidence rather than an attempt to force euryp-terid morphology into a peculiar numerical system Inthe absence of convincing evidence to the contrary Ihave coded eurypterids as having 12 opisthosomalsomites

Ricinulei The opisthosoma consists of a membranouspedicel (97) bearing the gonopore ventrally Thefemale gonopore is bordered by an anterior plate and aposterior plate The remainder of the opisthosoma iscomposed of thick sclerites separated by less heavilysclerotized cuticle The first dorsal sclerite is short andfunctions as part of a prosomandashopisthosoma couplingmechanism (96) that may or may not be a specializedcomponent of the following tergite This is followed byfour tergites and sternites and a three-segmentedmetasoma (= lsquopygidiumrsquo) Millot (1945 1949f) rea-soned that the opisthosoma contains ten somites withsomites VIIndashIX incorporated into the pedicel XndashXIIIexpressed as tergites and sternites and the metasoma

PHYLOGENY OF THE ARACHNID ORDERS 255

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

comprising four somites He did not regard the dorsalcoupling sclerite as a separate tergite and his inter-pretation of four rather than three metasomal somiteshas been rejected Pittard amp Mitchell (1972) also pro-posed ten opisthosomal somites but achieved thisnumber by regarding the dorsal coupling sclerite as atergite of somite IX and by recognizing three metaso-mal somites Legg (1976) adopted the system proposedby Pittard amp Mitchell but did not regard the dorsalcoupling sclerite as separate from the following tergite(X) Van der Hammen (1979 1989) reconstructed 13somites The pedicel was regarded as having twosomites (VII + VIII) the following four sets of tergitesand sternites were considered diplosomites (IX + XXI + XII XIII + XIV XV + XVI) and the metasoma hadthree somites (XVIIndashXIX) This interpretation mayhave been biased by Van der Hammenrsquos attempt tounite Ricinulei with anactinotrichid Acari which healso regarded as having a primitive number of 13somites Selden (1992) described a fossil ricinuleidTerpsicroton that shows two pairs of depressions onthe three large premetasomal tergites which con-trasts with the single pair seen in extant species Thisobservation appears to corroborate the diplosomitehypothesis but the evidence does not indicate that thetergite anterior to these is a diplosomite

Dunlop (1996) attempted to homologize theprosomandashopisthosoma coupling mechanisms ofRicinulei and Trigonotarbida a goal that required anovel and rather forced interpretation of the dorsalsclerites He regarded the coupling sclerite as homol-ogous with the first opisthosomal tergite (VII) of trig-onotarbids and then followed Van der Hammenrsquosdiplosomite hypothesis to achieve 12 somites in totalDunloprsquos scheme differs from previous systems in sug-gesting that the pedicel does not contain the dorsalelements of the first and second somites and is in-consistent with Millotrsquos (1945) observation that thepre- and postgenital plates each have dorsoventralmuscles Here I code Ricinulei as having 12 somitesThere are three metasomal somites (XVIndashXVIII)three diplosomites (= six somites) (XndashXV) one cou-pling somite (IX) and two somites in the pedicel (iethe genital and pregenital somites) (VII VIII)

Opiliones Harvestmen have nine opisthosomalsomites and an anal operculum that is traditionallyregarded as the tergite of a tenth somite a tenth ster-nite is lacking (Hansen amp Soslashrensen 1904 Winkler1957) However the anal operculum appears to repre-sent a persistent embryonic telson (Moritz 1957) andis therefore likely to be a postsegmental structurecomparable with the stinger of scorpions flagella ofthelyphonids etc (Shultz 2000)

Acari The number of opisthosomal somites is probe-matic for most mite taxa due to uncertainty about the

location of the prosomandashopisthosoma boundry (5)extensive simplification or loss of metamericallyarranged sclerites and muscle attachments a paucityof developmental studies of engrailed expression andin Acariformes opisthosomal anamorphosis and het-erochronic modification of somite number (Evans1992) Ixodids appear to be the exception developmen-tal studies indicate five somites in the opisthosoma ofticks (Evans 1992)

Some mites retain external evidence of segmenta-tion ndash metameric patterns of furrows muscle attach-ments and slit sensilla ndash and with certainassumptions the number of opisthosomal somites canbe estimated Here it is assumed that the dorsal sur-faces of the last two prosomal somites are present inmites and retain their primitive association with legs3 and 4 (contra Van der Hammen 1989) (see 5 for jus-tification) Given this there appear to be 11 somites inthe opisthosoma of Opilioacariformes a conclusionalso reached by other workers (eg With 1904 Sitni-kova 1978 Klompen 2000) Similar reasoning sug-gests that Alycus has seven opisthosomal somites notnine as advocated by Van der Hammen (1989) Unfor-tunately external evidence is ambiguous for determin-ing the number of somites in other mite lineages andthese are coded as uncertain96 Prosoma-opisthosoma coupling mechanism 0absent 1 present [GEWB sim24]The posterior margin of the carapace and the anteriormargin of the first apparent opisthosomal tergite arespecialized as a coupling mechanism in Ricinulei andTrigonotarbida and Dunlop amp Horrocks (1996) pro-posed that these are synapomorphic for the twogroups However there is uncertainty about thehomology of the anterior opisthosomal somites in Trig-onotarbida and Ricinulei (95)97 Pedicel 0 absent 1 aranean type 2 ricinuleidtype [WP 30 S sim40 WH sim20 GEWB sim126]The body narrows at or near the prosomandashopisthosoma juncture in several arachnid lineages (ieSolifugae Palpigradi Amblypygi Araneae Ricinulei)and this lsquowaistrsquo has often been used as character at theinterordinal level (eg Pocock 1893) However thisinterpretation is rejected here for being subjective andfor uniting non-homologous conditions For exampleAraneae and Amblypygi are often grouped on the basisof a lsquopedicelrsquo yet it is a highly specialized structure inAraneae and its parts are not readily homologizedwith those of Amblypygi In contrast the condition inAmblypygi is a slightly narrower version of the highlymoveable prosomandashopisthosoma juncture in Uropygiwhich is not generally considered a lsquopedicelrsquo Thepedicel in Ricinulei is also unique a weakly sclero-tized stalk containing the genital opening (Cryptocel-lus Pittard amp Mitchell 1972 Ricinoides Legg 1976)See 95

256 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

98 Opisthosoma with three- segmented lsquobucklerrsquo 0absent 1 presentState 1 occurs throughout Chasmatapsidida (Dunlop2002a)99 Thoracetron consolidation of tergites of postgeni-tal somites 0 absent 1 present - inapplicable codedonly for Xiphosura (Anderson amp Selden 1997)100 Fusion of tergites of postoral somites VIII and IX(= opisthosomal somites 2 and 3) only 0 absent 1present [GEWB 146]State 1 appears to be a synapomorphy Trigonotarbida(Shear et al 1987) Ricinuleids also have diplotergites(101) but these appear to encompass a different com-bination of somites (see 95)101 Three diplotergites 0 absent presentState 1 is a unique synapomorphy of Ricinulei (Selden1992) See 95102 Paired opisthosomal defensive glands openingvia ducts on either side of the anus 0 absent 1present [S 46 WH 86 GEWB 122]State 1 occurs in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905)103 Muscular diaphragm separating prosomal andopisthosomal compartments formed by dorsoventralmuscles of postoral somites VIndashVIII and extrinsic mus-cles of leg 4 0 absent 1 presentState 1 is known only in extant scorpions (Lankesteret al 1885 Firstman 1973) The diaphragm is oftentreated as a single structure but recent anatomicalwork (Shultz 2007) has shown that it is a composite ofdorsal endosternal suspensors and extrinsic leg mus-cles (see 95) A diaphragm is also present in Solifugaebut it is located more posteriorly and does not appearto be homologous with that of scorpions (Roewer1934)104 Opisthosomal appendicular chondrites 0absent 1 presentThese cartilage-like columns of mesodermally derivedtissue are associated with each opisthosomal append-age including the chilaria in extant Xiphosura(Patten amp Hazen 1900 Yamasaki et al 1988 Fahr-enbach 1999 Shultz 2001)105 Paired appendages on ventral surface of postoralsomite VII (= opisthosomal somite 1) in adult 0absent 1 present [WP sim13 S 39 WH 82 GEWB 143]State 1 is known in Weinbergina (Moore et al 2005)and extant Xiphosura The bilobed structure of themetastoma in certain eurypterids as well as the cor-responding placement of chilaria in Xiphosura haveled some workers to regard the metastoma as fusedappendages of postoral somite VII (eg Stoslashrmer1955) This hypothesis is problematic given that (i) nospecial explanation is needed to account for bilaterallysymmetrical structures in bilaterally symmetricalorganisms (ii) incorporation of the ventral part of thefirst opisthosomal sternite as a functional element of

the prosoma is typical of arachnids and perhapsEuchelicerata generally and (iii) some workers regardthe metastoma as a sternite (eg Jeram 1998) Thusthe metastoma is coded here as uncertain Dunlop hasproposed that the sternum of scorpions is derivedfrom appendages citing the presence of transient limbbuds in scorpion development (Brauer 1895 Patten1912) but this is not compelling evidence for a persis-tent appendicular contribution to the sternite of theadult106 Megoperculum 0 absent 1 present [S sim41 + 42WH sim83 GEWB 162 + 163]State 1 occurs in Thelyphonida Schizomida Ambly-pygi Araneae (Shultz 1993 1999) Trigonotarbida(Shear et al 1987) and probably Haptopoda (Dunlop1999) and Palpigradi The megoperculum consists ofappendages of opisthosomal somite 2 (= genital somiteor postoral somite VIII) (Shultz 1993 1999 Popadicet al 1998) that have fused medially and displacedthe ventral body wall of the somite anteriorly (oftenrepresented by a small sternite to which dorsoventralmuscles of postoral somite VIII attach) The megoper-culum projects posteriorly to form the ventral surfaceof a pregenital chamber The dorsal surface of thechamber is formed by the ventral body wall of opistho-somal somite 3 (= postoral somite IX) which serves asthe ventral attachment of the paired dorsoventralmuscles of that somite The megoperculum bears book-lungs (where present) and sometimes paired gonopodsthat are probably derived from telopodites A similarbut less developed operculum is present on theopisthosomal somite 3 which may also bear booklungsand eversible vesicles (eg Amblypygi) (see 111) corre-sponding to the booklungs and gonopods of the mego-perculum (Shultz 1999)

I contend here that a megoperculum is present inPalpigradi and that this is particularly evident infemales Specifically a large unpaired lobe projectsposteriorly from the genital somite over the ventroan-terior surface of the first postgenital somite therebyforming a pregenital chamber The dorsoventral mus-cles of the genital somite attach ventrally near theanterior border of postoral somite VIII and those ofpostoral somite IX attach to the upper surface of thepregenital chamber (Rucker 1901 Boumlrner 1902bRoewer 1934 Millot 1949d Van der Hammen 1989)Weygoldt (1998) has questioned this interpretation ofPalpigradi107 Postgenital operculum or lsquosternitersquo 0 sclerotized1 not sclerotized - inapplicable coded only for Pan-tetrapulmonata (Platnick amp Gertsch 1976)State 1 is present in Trigonotarbida Amblypygi The-lyphonida Schizomida and Mesothelae (Araneae) butnot Opisthothelae (Araneae)108 Genital opening of female guarded by four plates(one pregenital one postgenital two laterogenitals)

PHYLOGENY OF THE ARACHNID ORDERS 257

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

genital opening of male guarded by two plates 0absent 1 presentThis an apparent synapomorphy of Holothyrida(Acari) (Van der Hammen 1989)109 Anterior margin of genital opening in male withglands secreting via fusules 0 absent 1 presentState 1 occurs in Araneae as epigastric or lsquoepiandrousrsquoglands (Marples 1967) and in Palpigradi (Condeacute1991b)110 Paired valve-like plates apparently formed fromcomponents of three somites covering triradiate genitalopening 0 absent 1 presentThis is an apparent synapomorphy of Acariformes(Evans 1992 Alberti amp Coons 1999)111 Eversible lsquoappendagesrsquo on the ventral surface ofpostgenital somites 0 absent 1 present [S sim43 WHsim84 GEWB sim127]Paired ventral lsquoappendagesrsquo operated in part byhaemolymph pressure occur in the form of ventralsacs in certain Amblypygi (postoral somite IX) (Wey-goldt 2000) (Charinus Millot 1949b not PhrynusShultz 1999) prokoenenian Palpigradi (postoralsomites XndashXII) (Rucker 1901 Condeacute 1991a) as spin-nerets in Araneae (XndashXI) (Shultz 1987) and as geni-tal papillae in many Acariformes (Alberti amp Coons1999) including Oribatida many Endeostigmata (Aly-cus Van der Hammen 1989) and Prostigmata (Allo-thrombium Saboori amp Kamali 2000 MicrocaeculusEvans 1992) Eversible vesicles similar to those ofamblypygids occur in fossils of the trigonotarbidPalaeocharinus (postoral somite IX) (Fayers et al2004) Two cuticular structures (lsquogenital verrucaersquo)occur anterior to the genital opening in Opilioacari-formes and each covers the opening to a thin-walledinvaginated sac (lsquogenital papillarsquo) of unknown function(Van der Hammen 1989) Van der Hammen proposedthat it is homologous to a genital papilla of Acari-formes but Alberti amp Coons (1999) have questionedthis interpretation on several grounds and refer tothese structures as pregenital capsules112 Opisthosomal silk glands and spinnerets derivedfrom appendages on postoral somites X and XI(= opisthosomal somites 4 and 5) 0 absent 1 present[WP 33 WH 41 GEWB 123 + 142]State 1 is a unique synapomorphy of Araneae (Plat-nick amp Gertsch 1976) although potentially homolo-gous glands and spigot-like setae occur on the ventralopisthosomal surface in Palpigradi (Millot 1943Condeacute 1991a)113 Opisthsosomal spinnerets location 0 near mid-dle of opisthosoma 1 near posterior end of opisthso-som ndash inapplicable coded only for Araneae (Platnickamp Gertsch 1976)114 Anterior medial lsquospinneretsrsquo 0 absent 1 present- inapplicable coded only for Araneae (Platnick ampGertsch 1976)

115 Opisthosomal tergites divided longitudinally intoone median and two lateral plates 0 absent 1 present[GEWB 145]State 1 occurs in non-curculioid Ricinulei (Selden1992) and many Trigonotarbida (Dunlop 1996) Thisfeature or something very like it (eg distinct opistho-somal trilobation) occurs sporadically in severaleuchelicerate groups including Chasmataspis (Chas-mataspidida) (Dunlop et al 2004) most fossil xipho-surans (Anderson amp Selden 1997) and a few derivedEurypterida (especially Mixopteroidea) (Tollerton1989)116 Number of metasomal somites 0 zero 1 two 2three 3 five 4 nine [S sim44 WH sim85 GEWB 128 144]The metasoma is a preanal region comprising multiplesomites that are substantially narrower than the pre-ceding somites or mesosoma Metasomal somitesoften lack pleural membranes and take the form ofsclerotized rings State 1 occurs in TrigonotarbidaState 2 is present in synziphosurid Xiphosura (Ander-son amp Selden 1997) Amblypyi Thelyphonida Schizo-mida and Ricinulei State 3 occurs in Scorpiones andEurypterida and State 4 occurs in Chasmatispidida

Some workers appear to consider a five-segmentedmetasoma as conclusive evidence for the monophylyof a Eurypterida + Scorpiones clade (eg Dunlop ampBraddy 2001) presumably because the character rep-resents a kind of tagmosis and should therefore begiven substantial phylogenetic weight However Wey-goldt (1998) has pointed out that a metasoma contain-ing three somites is probably part of the ground planof Euchelicerata The known diversity of the eucheli-cerate metasoma indicates that this feature can inincrease or decrease its segmental composition in evo-lution or be eliminated entirely without necessarilychanging the total number of opisthosomal somitesConsequently it is problematic to assume that thefive-segmented metasoma has any special immunityto homoplasy117 Postanal structure (telson) 0 absent or not obvi-ously developed 1 presentPostanal structures are considered non-segmentalbecause they are modifications of the embryonicregion posterior to the site of somite addition (embry-onic growth zone) A well-developed postanal struc-ture persists in adults in Xiphosura EurypteridaChasmataspidida Scorpiones Schizomida Thely-phonida and Palpigradi The anal operculum of Opil-iones is generally considered by morphologists to bethe tergite of the tenth opisthosomal or anal somite(eg Hansen amp Soslashrensen 1904) but embryological evi-dence (Moritz 1957) suggests that it is derived fromthe postproliferative zone and thus corresponds to thetelson of other chelicerates (see 95)118 Postanal structure shape 0 caudal spine 1aculeus 2 flagellum 3 anal operculum - inapplica-

258 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ble due to absence of postanal structure (117) [WP sim41S sim45 WH sim24 GEWB sim121 129 + 147]State 0 is present in Xiphosura Eurypterida andChasmataspidida State 1 occurs in Scorpiones State2 occurs in Palpigradi Thelyphonida and SchizomidaState 3 occurs in Opiliones (see 95 117)119 Specialized postanal flagellum in male (see 157ndash159) 0 absent 1 present - inapplicable coded onlyfor taxa with postanal flagellum (118) [GEWB 131]State 1 is a unique synapomorphy of Schizomida(Hansen amp Soslashrensen 1905 Cokendolpher amp Reddell1992 Reddell amp Cokendolpher 1995)

RESPIRATORY SYSTEM

120 Respiratory medium 0 water 1 airAmong the terminal taxa included only fossil scorpi-ons are problematic for this character They are codedhere as uncertain Based on a study of book lungmicrosculpture Scholtz amp Kamenz (2006) have arguedthat arachnids are primitively terrestrial and pulmo-nate (see also Firstman 1973) and have questionedwhether any fossil scorpions were aquatic No positionon this proposal is taken here121 Respiratory lamellae on opisthosomal somite 2(= genital somite postoral somite VIII) 0 absent 1present [WP sim37 S sim51 WH sim22 GWEB sim133ndash137]This character encompasses book gills (= lamellae thatfunction in water) and book lungs (= lamellae thatfunction in air) The distinction between book gills andbook lungs is accommodated here by 120 in combina-tion with 121ndash124 Respiratory lamellae are presenton the genital somite in Trigonotarbida AraneaeAmblypygi Thelyphonida and Schizomida Petrunk-evitch (1949) reconstructed Plesiosiro as having booklungs but Dunlop (1999) could not corroborate thisThis character was used by Dunlop amp Webster (1999)to propose that Xiphosura and Scorpiones are closelyrelated because they both lack respiratory lamellae onthe genital somite Dunlop amp Braddy (2001) alsoargued for the placement of Eurypterida with Xipho-sura and Scorpiones based in part on this characterAt least some eurypterids may have had respiratorylamellae (eg Manning amp Dunlop 1995) but the onlyevidence of their segmental distribution is derivedfrom one specimen (Braddy et al 1999)122 Respiratory lamellae on opisthosomal somite 3(= postoral somite IX) 0 absent 1 present [GEWB136]State 1 occurs in Amblypygi Thelyphonida and Xipho-sura It is a groundplan feature of Araneae (Platnick ampGertsch 1976) and is present in all representativetaxa included here Braddy et al (1999) proposed theexistence of respiratory lamellae on postoral somitesIXndashXII in Eurypterida based on evidence from onespecimen This character is coded as uncertain for the

Eurypterida included here Extant scorpions are codedhere as lacking respiratory lamellae on this somite(= pectinal somite) based on the interpretation of theopisthosomal segmentation discussed in 95123 Respiratory lamellae on opisthosomal somites 4ndash6 (= postoral somites XndashXII) 0 absent 1 presentState 1 is definitely present in extant Xiphosura andScorpiones and may have occurred in Eurypterida124 Respiratory lamellae on opisthosomal somite 7(= postoral somite XIII)State 1 occurs in extant Xiphosura and Scorpionesonly125 Kiemenplatten 0 absent 1 presentThese structures are located on the roof of opercularchambers in Eurypterida (Clarke amp Ruedemann1912) where they take the form of ventrally projectingcones with a distinct cuticular microsculputure (Man-ning amp Dunlop 1995) Dunlop amp Braddy (2001)inferred the existence of Kiemenplatten in all Palaeo-zoic scorpions based on a description and photos of onespecimen of Paraisobuthus duobicarinatus by Kjelles-vig-Waering (1986 pls 16ndash18) The plates depict darkcone-like rearward-pointing denticles distributedwithin a white amorphous material The denticlesappear to lack microsculpture even though the mag-nifications at which the photos were taken (times90ndashtimes330)are comparable with those illustrating the cones ofKiemenplatten (times170ndashtimes350 in Manning amp Dunlop1995 figs 1 2) Given the diversity of cuticular struc-tures present in the atria of booklungs book gills andtracheae the denticles appear to bear no special sim-ilarity to the cones of Kiemenplatten126 Tracheal system 0 absent 1 paired ventral stig-mata on postoral somite VIII (= opisthosomal somite2) 2 paired ventral stigmata on postoral somites IXand X 3 one pair of stigmata opening near legs 3 or 44 paired stigmata associated with chelicerae 5 fourpairs of stigmata on dorsal surface of opisthosoma -inapplicable aquatic (120) [WP 40 sim43 S sim52 + 5354 WH sim23 45 88 GEWB sim138 139]Firstman (1973) and Weygoldt amp Paulus (1979)hypothesized that tracheae are homologous in all tra-cheate arachnids except spiders and subsequentworkers have entertained this hypothesis by includinga character for the presenceabsence of tracheal sys-tems that ignores the diverse arrangement of stig-mata in arachnids (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) However thisapproach assumes that internal tracheal systems areconserved but that tracheal openings (stigmata)appear and disappear on different parts of the bodywith higher evolutionary frequency Here I assumethat stigmata are conserved in evolution and that dif-ferences in their anatomical placement reflect the evo-lution of new tracheae State 1 occurs throughoutOpiliones State 2 occurs throughout Solifugae

PHYLOGENY OF THE ARACHNID ORDERS 259

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

(Roewer 1934) and Pseudoscorpiones (Chamberlin1931) State 3 recognizes the possible homology of stig-mata in Ricinulei Opilioacariformes and Parasiti-formes (Acari) (Lindquist 1984 Van der Hammen1989) Tracheal systems are apparently absent in thegroundplan of Acariformes but Prostigmata typicallyhave tracheae associated with the chelicerae Tra-cheae derived from the posterior pair of book lungs arewidespread in araneomorph Araneae but the booklungs are retained and tracheae are absent in themost basally divergent groups (eg Hypochilus)

BOXndashTRUSS AXIAL MUSCLE SYSTEM (BTAMS)

127 Posterior oblique muscles of BTAMS of postoralsomites IndashVI 0 absent 1 present in one or moresomitesState 0 occurs in Xiphosura and state 1 occurs inPalpigradi Araneae Amblypygi Thelyphonida andSchizomida (Shultz 2001) The condition in AcariRicinulei Opiliones and Solifugae is not known128 Anterior oblique muscles of BTAMS posterior topostoral somite VI 0 absent 1 presentState 1 occurs in Xiphosura which is probably theprimitive condition based on comparison with otherarthropods (Shultz 2001) State 0 occurs in all arach-nids examined thus far129 Ventral attachments of posterior oblique musclesof opisthosomal BTAMS located in prosoma 0 absent1 presentState 1 occurs in extant Xiphosura (Shultz 2001)130 Endosternite fenestrate 0 absent 1 present [S 8WH 54 GEWB 35]State 1 occurs in Thelyphonida and hubbardiid Schi-zomida (Firstman 1973) The condition in protoschi-zomid Schizomida is not known131 Suboral suspensor a tendon that arises from theBTAMS and inserts on the ventral surface of the oralcavity via muscle 0 absent 1 presentState 1 occurs in Palpigradi (Eukoenenia Millot1943) Amblypygi (Shultz 1999) and Thelyphonida(Shultz 1993)132 Perineural vascular membrane in adult 0absent 1 present [WH 28 GEWB 155]Wheeler amp Hayashi (1998) coded extant Xiphosura asunknown although the primary source for this char-acter (Firstman 1973) stated that adult Limulus havea perineural vascular membrane Within Arachnidapresence of a perineural vascular membrane is appar-ently correlated with the presence of tracheae (126)133 Ventral endosternal suspensor attaching on coxaof anteriorly adjacent somite 0 absent 1 present[GEWB 239 253]State 1 occurs in Amblypygi (Shultz 1999) Thely-phonida (Shultz 1993) and Schizomida (Shultzunpubl observ)

134 Posteriormost postoral somite with a pair of dor-soventral muscles 0 VI 1 VII 2 VIII 3 XII 4 XIII5 XIV 6 XV 7 XVIDorsoventral muscles tend to run in a continuousmetameric series beginning in the prosoma Deter-mining the last tergite on which the muscle seriesends is substantially easier than counting especiallygiven frequent anatomical complexities near theprosomandashopisthosoma juncture The dorsoventralmuscle series ends on postoral somite VII inCyphophthalmi (eg Chileogovea) and Laniatores(Opiliones) (Shultz unpubl observ) on postoralsomite VII or VIII in Eupnoi (Leiobunum Shultz2000 unpubl observ) on postoral somite XII in Pal-pigradi (Roewer 1934) Solifugae (Bernard 1896Roewer 1934 Millot amp Vachon 1949) and Araneae(Liphistius Hypochilus Millot 1933 but segmenta-tion reinterpreted here following Shultz 1993 1999)on postoral somite XIII in hubbardiid Schizomida(Cokendolpher amp Reddell 1992) and Scorpiones(Lankester et al 1885) and on postoral somite XIVin extant Xiphosura (Shultz 2001) Thelyphonida(Shultz 1993) protoschizomid Schizomida (Cokendol-pher amp Reddell 1992) and Amblypygi (Shultz 1999)Dunlop (1999) reconstructed Plesiosiro as havingpaired tergal apodemes ending on postoral somite XVbut his figures indicate that the series ends on posto-ral somite XIV The interpretation of opisthosomalsegmentation in Ricinulei coded here (95) indicatesthat the series ends on postoral somite XV in Terpsi-croton and this is treated as the groundplan for theorder The posteriormost dorsoventral muscles inPseudoscorpiones occur on postoral somite XVI(Vachon 1949)

NERVOUS SYSTEM

135 Segmental ganglia 0 consolidated in prosoma1 one or more present in opisthosoma [WH 30 GEWB210 + 216]Adult neuromeres are exclusively prosomal in Ambly-pygi liphistiomorph and araneomorph Araneae Opil-iones Pseudoscorpiones Ricinulei and Acari (Millot1949a)136 Dorsal median eyes 0 absent 1 present [WP14 + 47 S 50 WH sim7 GEWB 1]Median eyes are a groundplan feature of Euchelicer-ata and are retained in extant Xiphosura Eurypteridaand occur in all arachnid orders except PalpigradiSchizomida Ricinulei and Pseudoscorpiones (Paulus1979) Median eyes are also absent in opilioacariformand parasitiform Acari (Lindquist 1984) and perhapsthe fossil whipscorpion Proschizomus (Dunlop amp Hor-rocks 19951996) They are unknown in synziphosu-ran Xiphosura Median eyes are present in a variety ofbasal acariform mites including Endeostigmata (but

260 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

not Alycus) Prostigmata (Microcaeculus) and certainoribatids (Palaeacarus) (Evans 1992 Alberti amp Coons1999) Giribet et al (2002) coded all representativeAcari as lacking median eyes It is not known whetherthe eyes of some cyphophthalmid opilions are medianor lateral evidence from tracheal branching inCyphophthalmus (Janczyk 1956) suggests that theyare median eyes (Shultz amp Pinto da Rocha 2007) andpresence of a tapetum in Stylocellus is consistent withlateral eyes (Shear 1993)137 Retinula cells of dorsal median eyes 0 organizedinto closed rhabdoms 1 organized into network ofrhabdomeres 2 disorganized - inapplicable due toabsence of median eyes (136) [GEWB sim3]State 0 is present in Scorpiones Thelyphonida andAmblypygi State 1 is present in Solifugae and Ara-neae state 2 is present in Xiphosura (Paulus 1979)The retinula cells of median eyes have been studied inseveral prostigmatid Acariformes and are organized ina network in some taxa and in an irregular pattern inothers (Alberti amp Coons 1999) Retinulae in phalangidOpiliones have state 0 proximally and state 2 distally(Schliwa 1979)138 Ventral median eyes 0 absent 1 presentState 1 occurs in early instars of extant Xiphosura(Paulus 1979)139 Lateral eyes 0 absent 1 presentLateral eyes are primitively present in Chelicerataand are absent in Palpigradi and Opiliones (Paulus1979) It is unclear whether the eyes of cyphoph-thalmid opilions are median or lateral (see 136)140 Arrangement and number of lateral eyes 0 com-pound many 1 five or more pairs (includes micro-lenses) 2 three primary pairs (excludes microlenses)3 two pairs 4 one pair - inapplicable due to absenceof lateral eyes (139) [WP sim13 + 18 + 38 + 44 + 52 Ssim49 WH sim10 GEWB sim4 + 5]True compound eyes are present in XiphosuraEurypterida Chasmataspidida and many PalaeozoicScorpiones Among the anactinotrichid Acari Opilioa-cariformes have two (Neocarus) or three (Siamacarus)pairs of lateral eyes (although at least one species ofSiamacarus lacks eyes) the allothryrid Holothyridaand many Ixodida have a single pair of lateral lensesand the remainder apparently lack eyes (Evans1992) Thelyphonida was coded as lsquo12rsquo to reflect fivepairs of lenses comprising three pairs of primarylenses and two pairs of small accessory lenses (notsimply three pairs as coded by Giribet et al 2002)Trigonotarbida is also coded as lsquo12rsquo to reflect threepairs of primary lenses and multiple small accessorylenses Dunlop (1999) illustrated Haptopoda as havingpaired triads of lateral eyes but noted that there wasactually no evidence of this in the fossils The lateraleyes of fossil ricinuleids have two pairs of lenses(Selden 1992) One pair of eyes or eyespots are

present in extant Ricinulei and many Schizomidaalthough five genera of hubbardiid Schizomida have apair of lenses (Reddell amp Cokendolpher 1995) State 1is the groundplan for extant Scorpiones and State 3occurs in the pseudoscorpions Chthonius FeaellaNeobisium (but State 4 in Chelifer) and Solifugae(Paulus 1979)141 Lateral eyes with closed rhabdoms 0 absent 1present - inapplicable due to absence of lateral eyes(139) [WP 21 WH 12 GEWB 6]Presence of closed rhabdoms is probably primitive forChelicerata and is retained in extant Xiphosura andScorpiones Retinula cells form a network of rhab-domeres in other extant chelicerates (Paulus 1979)but these networks can differ substantially in detailWeygoldt (1998) and other workers have given sub-stantial weight to the network character in unitingnon-scorpion arachnids but these authors seem not togrant comparable phylogenetic significance to theanalogous character of the median eyes shared bySolifugae and Araneae (137)142 Slit sensilla 0 absent 1 present [WP 19 S 47WH 11 GEWB 209]State 1 occurs in all arachnid orders except Palpigradi(Shultz 1990) The proposal that lsquoprimitiversquo slit sen-silla were present in Eurypterida (Dunlop amp Braddy1997) appears to be based on the over-interpretation ofa comparatively large notch that occurs in Baltoeu-rypterus at the terminus of the tibia (= podomere 7)(see also Edgecombe et al 2000 Giribet et al 2002)The neural construction of slit sensilla like function-ally similar campaniform sensilla of hexapods is sim-ilar to that of trichoid sensilla (Chapman 1998Klompen 2000) Thus these cuticular stress receptorsmay represent modified bases of sensory setae a viewsupported by the replacement of setae by slit sensilladuring post-embryonic development in some Opilioac-ariformes (Klompen 2000)143 Trichobothria 0 absent 1 present [GEWB 213]Trichobothria are present in extant Scorpiones(Jeram 1998) Pseudoscorpiones (Chamberlin 1931)some endeostigmatid (Alycus Van der Hammen 1989)and prostigmatid Acariformes and most Oribatida(Lindquist 1984) They occur on the ventral surface ofthe pedal femora in the opilioacariform Siamacarus(Leclerc 1989) They are also present in Araneae (Foe-lix 1996) Amblypygi (Weygoldt 2000) Schizomidaand Thelyphonida (Hansen amp Soslashrensen 1905) Theyare apparently absent in all non-arachnid chelicer-ates Solifugae Ricinulei Opiliones (Reissland ampGoumlrner 1985) and Parasitiformes (Acari) (Lindquist1984)144 Tibial trichobothria with 2-1-1-1 pattern onappendages IIIndashVI (= arachnid legs 1ndash4) 0 absent 1present - inapplicable due to absence of trichobothria(143) [S 48 WH 87 GEWB 88]

PHYLOGENY OF THE ARACHNID ORDERS 261

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs only in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905) Note that Shultz (1990)erroneously described this character as a 2-2-1-1 pat-tern and that this error was repeated by Giribet et al(2002) and described as 2-1-1 by Wheeler amp Hayashi(1998)145 Paired trichobothria on dorsal surface ofprosoma 0 absent 1 present - inapplicable due toabsence of trichobothria (143)State 1 is an apparent synapomorphy of Acariformesand is present in all acariforms included in this study(Alberti amp Coons 1999)146 Malleoli 0 absent 1 present [WP 45 GEWB 96]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)147 Pectines 0 absent 1 present [WP 20 WH 38GEWB 120]State 1 is a unique synapomorphy of Scorpiones148 Intercheliceral median organ 0 absent 1present [GEWB 212]State 1 is a unique synapomorphy of Palpigradi(Roewer 1934)149 Tarsal organ on appendage of postoral somite III(= arachnid leg 1) (= Hallerrsquos organ) sensilla containedwithin a cuticular depression on the superior surface ofthe tarsus of appendage III (= arachnid leg 1) 0absent 1 present [GEWB sim100]Klompen (2000) has noted that State 1 occurs in Opil-ioacariformes Parasitiformes (except Mesostigmata)and Ricinulei where it also occurs on leg 2 (Talaricoet al 2005) State 1 occurs on all legs in Araneae andperhaps Scorpiones (Foelix 1985)150 Tarsal organ on appendage of postoral somite IV(= arachnid leg 2) 0 absent 1 present (See 149)

REPRODUCTION

151 Gonads 0 primarily prosomal 1 primarilyopisthosomalState 0 is limited to Xiphosura (Sekiguchi 1988)State 1 occurs throughout Arachnida (Millot1949a)152 Ladder-like opisthosomal gonadsaccessoryglands (see 153) 0 absent 1 present [WH sim37 GEWBsim158]Giribet et al (2002) followed Wheeler amp Hayashi(1998) who followed Clarke (1979) in coding gonads asreticulate (Xiphosura) ladder-like (Scorpiones Thely-phonida Schizomida) or lsquosaccularrsquo (all remainingArachnida) However comparisons between Xipho-sura and Arachnida are problematic given that thexiphosuran gonads are primarily prosomal and thoseof arachnids are primarily opisthosomal (151) Fur-ther the reticulate pattern in Limulus (Xiphosura)but not other extant xiphosurans contains a distinctlyladder-like component The lsquosaccularrsquo state is probably

artificial as it encompasses a wide variety of pairedand unpaired structures153 Male gonads in two distinct parts one producingsperm and another (tubular gland) producing a holo-crine secretion similar to degenerate sperm 0 absent1 presentState 1 occurs in Thelyphonida Schizomida andAmblypygi although the holocrine material is pro-duced by ventral organs in Amblypygi and dorsalorgans in Thelyphonida and Schizomida (Alberti2005)154 Number of gonopores 0 two 1 oneExtant xiphosurans have two small genital openingson the base of the genital telopodite and all extantarachnids have a single opening The condition inEurypterida is not known Clarke amp Ruedemann(1912) located a pair of openings near the base of themedian organ (161) that are the outlets of the lsquohornorgansrsquo but it is unclear whether these are genitalducts or accessory structures Braddy amp Dunlop (1997)have developed numerous speculations about thesestructures and extended their arguments far beyondthe available evidence155 Genital opening (ie gonopore or gonostome)appearing to open in prosomal region (ie between legcoxae or anterior to posterior carapacal margin) 0absent 1 present [WP sim50 WH 26 GEWB 166]The genital opening in Euchelicerata is located onpostoral somite VIII (= opisthosomal somite 2) but ithas shifted anterior to the posterior margin of thecarapace or between the coxae of the last pair of legsin most Scorpiones (but not in PalaeoscorpiusKjellesvig-Waering 1986) and Opiliones The geni-tal opening occurs near or anterior to the last coxaein the opilioacariform and parasitiform Acari repre-sented here It is variable in Acariformes but islocated posterior to the coxae in all representativetaxa156 Ovipositor 0 absent 1 present [WP sim51 S sim60WH sim91 GEWB sim172]An ovipositor with a trilobed terminus is an apparentgroundplan character of Oribatida (Lindquist 1984Alberti amp Coons 1999) An ovipositor is also present inOpilioacariformes and Opiliones (Van der Hammen1989)157 Stalked spermatophore attached to substratum0 absent 1 present [S 57]State 1 occurs in Scorpiones PseudoscorpionesAmblypygi Thelyphonida and Schizomida The mech-anism of sperm transfer is unknown in PalpigradiOpilioacariformes and Holothryrida158 Male turns posterior end toward female duringmating behaviour 0 absent 1 presentState 1 occurs in Amblypygi (Weygoldt 2000) The-lyphonida and Schizomida (Weygoldt amp Paulus1979)

262 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

159 Female grasps male opisthosoma during matingbehaviour 0 absent 1 present [WP 29 S 58 WH 19GEWB 188]State 1 occurs in Thelyphonida and Schizomida (Wey-goldt amp Paulus 1979) Evolution of this behaviour wasprobably facilitated by 158160 Penis 0 absent 1 present [WP sim51 GEWB 167]State 1 is a unique synapomorphy of Opiliones A truepenis occurs in Phalangida A clearly homologousstructure is present in Cyphophthalmi (Opiliones) andapparently functions in depositing a spermatophore inthe femalersquos genital chamber (Karaman 2005) A truepenis may occur in some mites but it does not appearto be a groundplan feature of any major group (Evans1992) The lsquopenisrsquo in Oribatida is really a spermatop-ositor it functions in construction of a spermatophore(Alberti amp Coons 1999)161 Median organ 0 absent 1 presentState 1 occurs throughout Eurypterida (Clarke amp Rue-demann 1912) and has been observed in diploaspididChasmataspidida (Dunlop 2002a also Loganamaras-pis Tetlie amp Braddy 2004)

SPERM MORPHOLOGY

162 Nucleus with manchette of microtubules 0absent 1 present [S 54 WH 89 GEWB 193]State 1 occurs in Araneae Amblypygi Thely-phonida Schizomida Ricinulei and gonyleptid Lani-atores (Opiliones) (Alberti 1995 Giribet et al2002)163 Axoneme 0 absent 1 present [WP sim49 S sim55WH sim25 GEWB 195]State 1 occurs in Xiphosura Scorpiones Cyphoph-thalmi (Opiliones) Pseudoscorpiones RicinuleiAraneae Amblypygi Thelyphonida and Schizomida(Alberti 1995)164 Coiled axoneme 0 absent 1 present - inappli-cable due to absence of axoneme (163) [WP 22 S sim55WH 13 GEWB 196]State 1 occurs in Pseudoscorpiones Ricinulei Ara-neae Amblypygi Thelyphonida and Schizomida(Alberti 1995)165 Microtubule arrangement in axoneme 0 9 + 0 19 + 1 2 9 + 2 3 9 + 3 - inapplicable due to absence ofaxoneme (163) [WP sim26 S sim56 WH sim16 GEWB 198](Alberti 1995)166 Helical or corkscrew shaped nucleus 0 absent 1present [GEWB sim204]State 1 occurs in Araneae Amblypygi ThelyphonidaSchizomida and certain Scorpiones (eg Hadrurus)(Alberti 1995)167 Vacuolated-type sperm 0 absent 1 present[GEWB 205]State 1 is unique to Opilioacariformes and Parasiti-formes (Acari) (Alberti 1995)

168 Sperm aggregates 0 absent 1 present [GEWB206] (Alberti 1995)

DEVELOPMENT

169 Yolk in early embryo 0 concentrated (centroleci-thal or telolecithal) 1 evenly distributed (isolecithal)[WH sim34 GEWB sim191]Wheeler amp Hayashi (1998) and Giribet et al (2002)coded this character as lsquo0 isolecithal or telolecithal 1centrolecithalrsquo based on information presented byYoshikura (1975) Their coding appears to highlightseparate conditions found in Scorpiones (ie isoleci-thal and telolecithal) rather than a property intrinsicto the character itself170 Embryonic nutrition other than yolk 0 absent 1present [WP sim49]State 1 occurs in extant Scorpiones (except buthidsand chaerilids) and Pseudoscorpiones (Weygoldt1969)171 Embryological growth zone 0 initiating segmentaddition within prosoma 1 initiating segment addi-tion posterior to prosoma [GEWB sim192]Giribet et al (2002) followed Dunlop amp Webster (1999)in stating that only Xiphosura and Scorpiones havestate 0 among Chelicerata However the last prosomalsomite develops from the growth zone in Ixodida(Anderson 1973 Evans 1992) A similar processapparently occurs in the acariform mites Tyroglyphus(Sarcoptiformes Astigmata) (Yoshikura 1975) andArchegozetes (Sarcoptiformes Oribatida) (Thomas ampTelford 1999) but not apparently in Tetranychus(Prostigmata) (Dearden Donly amp Grbic 2002)172 Eggsembryos maintained in external attachedbrood sac secreted by genital glands 0 absent 1present [WP sim46 S 59 WH 90 GEWB 208 = 219]State 1 occurs throughout Amblypygi SchizomidaThelyphonida and Pseudoscorpions Giribet et al(2002) followed Shear et al (1987) in homologizingspinneret-derived silken egg sacs of spiders with gen-ital-gland-derived brood sacs of Pedipalpi However asthese structures are not homologous in secretory ori-gin construction or composition (Shultz 1987) theyare not regarded here as homologous Giribet et al(2002) miscoded the character as absent in Pseudo-scorpiones and without justification in PalpigradiFemale palpigrades have well-developed glands asso-ciated with postgenital somites and an array of setaefusules and other structures (112) (Boumlrner 1902b Mil-lot 1942 Condeacute 1991a) which are consistent withsome sort of brood care and Palpigradi should there-fore be coded as unknown for this character173 Embryonic and early postembryonic lsquolateralrsquo orClaparegravede organs associated with coxa of postoralsomite IV (= arachnid leg 2) 0 absent 1 present [WHsim31 GEWB sim211]

PHYLOGENY OF THE ARACHNID ORDERS 263

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bilaterally paired rounded protuberances locatedbetween coxae of appendages of postoral somites IIIand IV (arachnid legs 1 and 2) in prelarval and larvalinstars appear to be a primitive feature of Acariformes(Lindquist 1984 Evans 1992 Alberti amp Coons 1999)It is known to be derived embryologically from thecoxa of leg 2 in the oribatid Archegozetes (Thomas ampTelford 1999) Structures apparently homologouswith the acariform Claparegravede organ are present inembryonic Amblypygi (Weygoldt 2000) and Thely-phonida (Yoshikura 1975) and in embryonic and earlypost-embryonic stages of Solifugae (Roewer 1934)where they are termed lateral organs The so-calledlsquolateral organrsquo of embryonic Xiphosura (174) occurs onthe lateral surface of the carapace and does not appearto be homologous with the coxa-associated structuressharing their name In fact the xiphosuran lateralorgan develops even when the coxa of postoral somiteIV has been excised (Sekiguchi 1988) In contrastWheeler amp Hayashi (1998) and Giribet et al (2002)followed Yoshikura (1975) in regarding all lsquolateralorgansrsquo as homologous and in miscoding this characteras absent in Acariformes Van der Hammen (1989)suggested that the sternal verrucae of Opilioacari-formes are homologous with the Claparegravede organ butit differs in gross structure and ontogenetic timing174 Embryonic lateral organ associated with cara-pace 0 absent 1 present [WH sim31 GEWB sim211]State 1 is known only from embryonic stages of extantXiphosura (Yoshikura 1975) In contrast to previousinterpretations this feature does not appear to behomologous with the lsquolateralrsquo or Clarapegravede organ ofcertain arachnids (173)175 Live birth 0 absent 1 presentState 1 occurs in extant Scorpiones and has evolvedseveral times in a few mites (Evans 1992)176 Hexapodal larva and 1ndash3 nymphal stages 0absent 1 present [WP 53 S sim61 WH 27 GEWB 186]State 1 occurs in Acari and Ricinulei (Lindquist1984)177 Hexapodal prelarva 0 absent 1 present [S sim61WH 47]State 1 occurs in Opilioacariformes (NeocarusKlompen 2000) and many Acariformes (Evans 1992)

EXCRETORYOSMOREGULATORY SYSTEM

178 Malpighian tubules 0 absent 1 present [WP 17S 62 WH 9 GEWB 153]State 1 occurs in all major arachnid groups except Pal-pigradi Pseudoscorpiones Opiliones Oribatida (Sar-coptiformes) and Prostigmata Within Acari State 1occurs in all major anactinotrichid lineages (Opilioac-ariformes Holothyrida Mesostigmata Ixodida) andsimilar structures are present in certain Astigmata(Evans 1992 Alberti amp Coons 1999)

179 Dorsomedian excretory organ 0 absent 1presentThis is a specialized excretoryosmoregulatory organformed by the postventriculus and proctodeum inProstigmata (Alberti amp Coons 1999)180 Adult coxal organ opening on or near coxa ofappendage III (= arachnid leg 1) 0 absent 1 present[S 64 WH 93 GEWB 90]State 1 occurs throughout Acari (Evans 1992 Albertiamp Coons 1999) Ricinulei (Pittard amp Mitchell 1972Legg 1976) Palpigradi (Millot 1942) AraneaeAmblypygi Thelyphonida and Schizomida (Buxton1913 1917) There is evidence that the second lsquoozo-porersquo of gonyleptids (Opiliones Laniatores) (Hara ampGnaspini 2003) is a persistent opening to the coxalorgan associated with leg 1 (Soslashrensen 1879) whichalso occurs in embryonic Phalangium (Opiliones Eup-noi) (Moritz 1959)181 Adult coxal organ opening on or near coxa ofappendage V (= arachnid leg 3) 0 absent 1 present [S63 WH 92 GEWB 89]State 1 occurs in extant Xiphosura (Yamasaki et al1988) Scorpiones mygalomorph (Buxton 1913 1917)and mesothele Araneae (Liphistius J Millot in Bris-towe 1932) State 1 occurs in basal Amblypygi(Charon Charinus Buxton 1913 1917) but degener-ates prior to the adult stage in higher groups(Phrynus Weygoldt 2000) Adequately preservedeurypterids show a small submarginal opening on thecoxa of appendage V (Baltoeurypterus Selden 1981also Eurypterus Hughmilleria Clarke amp Ruedemann1912) and this is interpreted here as the opening to acoxal organ Giribet et al (2002) coded this characteras uncertain for Eurypterida182 Adult coxal organ opening on coxa of appendageII (= arachnid palp) 0 absent 1 present [GEWB 65]Buxton (1913 1917) noted that the coxal organ ofSolifugae opens on the prolateral surface of the palpalcoxa The proximal end of the duct leading to the ori-fice is associated with glands and apparently theglandular secretion andor coxal fluid serve as saliva(Alberti 1979) (see also 14 15)183 Genital papillae 0 absent 1 present - inappli-cable coded only for those taxa with eversible lsquoappend-agesrsquo (111)The genital papillae are associated with the genitalopening in postlarval Acariformes (Evans 1992Alberti amp Coons 1999) These structures are probablyserially homologous with the prosomal Claparegravedeorgan (173) and both structures probably function inwater and ion regulation

DIGESTIVE SYSTEM

184 Ingestion 0 solid food 1 primarily liquid foodwith or without preoral digestion [WP sim16 WH sim8GEWB sim218]

264 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Extant xiphosurans ingest solid food and members ofmost extant arachnid orders ingest fluids Opilionesand Acari are the principal exceptions The phalangidOpiliones ingest solids but Shultz (2000 unpublobserv) has shown that the precerebral pharyngealapparatus of Cyphophthalmi is very similar to that ofthe fluid-feeding Scorpiones and differs substantiallyfrom phalangid Opiliones Examination of gut con-tents of the cyphophthalmids Siro acaroides S exilisand Chileogovea oedipus have failed to reveal solidparticles typically found in the guts of phalangids(Shultz unpubl observ) the character is coded asunknown for Cyphophthalmi Within Acari consump-tion of particles has been documented in Opilioacari-formes (Neocarus Van der Hammen 1989unidentified Walter amp Proctor 1998) Holothyridanon-parasitic Sarcoptiformes (Evans 1992 Walter ampProctor 1998 Alberti amp Coons 1999) and many endeo-stigmatids including the nematophagous Alycusroseus (Walter 1988)185 Mouth 0 directed posteroventrally 1 directedanteroventrally [S 9 WH 55 GEWB 28]State 1 occurs throughout Arachnida (Shultz 1990)186 Oral cavity dilated by muscles arising from coxaeand constricted by large circular sphincter 0 absent1 presentState 1 occurs in extant Xiphosura (Manton 1964Scholl 1977 Shultz 2001)187 Palate plate 0 absent 1 present [GEWB 159]State 1 is a unique synapomorphy of Araneae (Dunlop1994)188 Lateral walls of epistome broadly fused to medialwalls of palpal coxae opposite sides connected by well-developed transverse epistomal muscle 0 absent 1presentState 1 occurs throughout Opiliones and Scorpiones(Shultz 2000 unpubl observ)189 Epistome with a pair of lateral arms that projectsposteriorly into the prosoma on either side of the phar-ynx 0 absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)190 Epistome with four pairs of suspensor musclesattaching to the carapace 0 absent 1 presentState 1 occurs in extant Scorpiones (Lankester et al1885 Vyas 1970 Shultz unpubl observ) and appar-ently in Solifugae (Roewer 1934)191 Intercheliceral epipharyngeal sclerite 0 absent1 present [GEWB 31]State 1 occurs in Palpigradi (Boumlrner 1904) Trigo-notarbida (Palaeocharinus Dunlop 1994) Araneae(Firstman 1954 Marples 1983) Amblypygi (Mil-lot 1949b Shultz 1999) Schizomida (Millot1949e) Thelyphonida (Millot 1949e Shultz 1993)and Phalangida (Opiliones) (Shultz 2000 unpublobserv)

192 Epipharyngeal sclerite large projecting posteri-orly 0 absent 1 present - inapplicable due to absenceof sclerite (191) [GEWB 32]State 1 occurs in Amblypygi (Millot 1949b Shultz1999) Schizomida (Millot 1949e) Thelyphonida (Mil-lot 1949e Shultz 1993) some Araneae (Heptathela ndashcoded as Liphistius ndash and Hypochilus Marples 1983not Aphonopelma Firstman 1954) and perhaps Trig-onotarbida (Palaeocharinus Dunlop 1994)193 Dorsal dilator muscle of precerebral pharynxattaching to intercheliceral septum or associatedepipharyngeal sclerite (191) 0 absent 1 presentState 1 occurs in Palpigradi (Millot 1943) Araneae(Marples 1983) Amblypygi Schizomida and Thely-phonida (Millot 1949e Shultz 1993 1999)Extremely fine muscles have been documented insome Scorpiones (Shultz 2007) and in Leiobunum(Opiliones) (Shultz 2000) Lankester et al (1885)reported a muscle in Limulus that would appear tocorrespond to this muscle but the muscle does notexist (Manton 1964 Shultz 2001)194 Dorsal dilator muscle of precerebral pharynxattaching to dorsal surface of prosoma 0 absent 1present [GEWB 228]State 1 occurs in Palpigradi (Roewer 1934) and mostAraneae (Palmgren 1978 Aphonopelma Firstman1973) but not in Heptathela (coded as Liphistius) orHypochilus (Marples 1983)195 Lateral dilator muscle of precerebral pharynxattaching to endosternite 0 absent 1 present [GEWBsim226]State 1 is a unique feature of Araneae (Marples 1983)196 Lateral dilator muscle of precerebral pharynxattaching to lateral surface of epistomal processes 0absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)197 Lateral dilator muscle of precerebral pharynxattaching to medial process of coxa of appendage II(= arachnid palp) 0 absent 1 present [GEWB sim226]State 1 occurs in Amblypygi Thelyphonida and Schi-zomida (Boumlrner 1904 Shultz 1993 1999)198 Dilator muscle of precerebral pharynx andorpreoral cavity attaching to ventral surface of prosoma0 absent 1 present [GEWB 227]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) Araneae (Marples 1983) Palpigradi(Millot 1943) and Solifugae (Roewer 1934)199 Postcerebral pharynx 0 absent 1 present [WPsim31 S sim5 WH sim21 GEWB sim154]The cuticle-lined foregut passes through the centralnervous system in many arthropods That portion pos-terior to the CNS is here termed the postcerebralpharynx regardless of the details of its morphology orfunctional specialization A substantial postcerebralpharynx is known in extant Xiphosura (Lankester

PHYLOGENY OF THE ARACHNID ORDERS 265

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

et al 1885 Manton 1964 Yamasaki et al 1988)Scorpiones (Centruroides Hadrurus HeterometrusShultz 2007 also Androctonus Abd el-Wahab 1952)Solifugae (Millot amp Vachon 1949) Araneae Ambly-pygi Thelyphonida and Schizomida (Millot 1949andasheShultz 1993) Weygoldt amp Paulus (1979) applied theterm lsquopostcerebral pharynxrsquo only to the lsquosucking stom-achrsquo of Araneae and Amblypygi but the sucking stom-ach is here regarded as a complex of three characters199ndash201

200 Dilator muscle of postcerebral pharynx attach-ing to endosternite 0 absent 1 present - inapplicabledue to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Lankester et al1885 Manton 1964 Shultz 2001) Scorpiones (Cen-truroides Hadrurus Heterometrus Shultz 2007 alsoAndroctonus Abd el-Wahab 1952) Araneae (First-

man 1954 Palmgren 1978) Amblypygi (Millot1949b Shultz 1999) and apparently the palpigradeProkoenenia wheeleri (Rucker 1901) but not inEukoenenia miriabilis (Millot 1943 contra Boumlrner1904)201 Dilator muscle of postcerebral pharynx attachingto dorsal surface of prosoma 0 absent 1 present -inapplicable due to absence of postcerebral pharynx(199)State 1 occurs throughout Araneae (Millot 1949cFirstman 1954 Palmgren 1978) and Amblypygi(Charinus Millot 1949b Phrynus Shultz 1999)202 Crop and gizzard 0 absent 1 present - inap-plicable due to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Sekiguchi 1988)but not in extant Arachnida

PHYLOGENY OF THE ARACHNID ORDERS

225

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Cryptocellus

00000-0000 1211000100 00-00001 0000-000 0010001000 000000001 0001100020 02--000-11 0101000001 00105120-0 100000-000 00--120--1 000003 1500-014 10-000011 10010000 011120000 0010101 00-1100000 0-0000000- --

Ricinoides

00000-0000 1211000100 00-00001 0000-000 0010001000 000000001 0001100020 02--000-11 0101000001 00105120-0 100000-000 00--120--1 000003 1500-014 10-000011 10010000 011120000 0010101 00-1100000 0-0000000- --

Neocarus

00001-0000 0011000020 0000-00001 10100-000 0010010000 00000001 000111120 02--00 0101000011 0114000-0 000000-000 00--000--1 000005 00-013 010--00010 1001110 000--01000 011101 00-0100000 0-0000000- --

Siamacarus

00001-0000 0011000020 0000-00001 1100-000 0010010000 0000000 0011112 02--00 10100011 014000-0 000000-000 00--000--1 000005 00-012 110000010 101110 00--10 01101 -0100000 0-0 --

Australothyrus

01000-0100 0001000020 0000-00001 10100-000 0011000000 0000000 00001020 02--00 010-00011 01000-0 000000-100 00--000--1 000003 00-014 10--00010 1001100 000--01000 0101 00-1100000 0-000 --

Allothyrus

01000-0100 0011000020 0000-00001 10100-000 0011000000 0000000 00001020 02--00 010-00011 01000-0 000000-100 00--000--1 000003 00-00- -10--00010 101100 00--10 0101 00-1100000 0-0 --

Glyptholaspis01000-0000 0011000020 000-00001 10100-000 0011000000 000000001 0000001020 02--011011 010-00011 0110000-0 000000-000 00--000--1 00000300 00-00- -10--00000 1001100000 000--01000 000010101 00-1100000 0-0000000- --

Amblyomma00000-0000 0000-00020 0000-00001 10000-000 000--00000 000000001 0000001020 02--011011 010-00011 01100000-0 000000-000 00--000--1 00000300 100-014 010--00010 1001100000 000--01000 0000010101 00-1100000 0-0000000- --

Argas00000-0000 0000-00020 0000-00001 10000-000 000--00000 000000001 0000001020 02--011011 010-00011 00100000-0 000000-000 00--000--1 00000300 100-014 010--00010 101100000 00--10 0000010101 00-1100000 0-00000- --

Alycus000001000 000110020 000-00001 10100-000 000--00000 000000001 000010020 02--0011 000--00011 00101000-0 000000-001 10--000--1 000000 00-013 110100000 10100 00 0111 0010100000 0-000- --

Allothrombium0000000 000110020 00-00001 10000-000 000--00000 000000001 00111-0020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000004 00-0130110100000 1001001000 000--00000 010011011 0011100000 0-0000000- --

Microcaeculus00000000 000110020 00-00001 10000-000 000--00000 000000001 00111-0020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000004 0120130110100000 10010010 00--0000 01001011 001100000 0-0000000- --

Palaeacarus000011000 040110020 000-00001 10100-000 000--00000 000000001 00111-0020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000000 010 110100000 101011000 00--00 010011001 0010100000 0-00000- --

Table 1 Continued

226 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Archegozetes000011000 040110020 000-00001 10100-000 000--00000 000000001 0000000020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000000 00-00- -110100000 1001011000 000--00000 0010011001 0010100000 0-0000000- --

Cyphophthalmus0000100100 0100-00000 100-00000 -0000-1000 0010000001 1000000001 0000000020 01--010-11 0100-00001 00112000-0 000000-000 00--0013-1 000001000 010100 10--00000 1001110001 001100010 0000000 10-100110 0-0001000- --

Chileogovea0000100100 0100-00000 1010-00000 -0000-1000 0010000001 1000000001 0000000020 01--010-11 0100-00001 00112000-0 000000-000 00--0013-1 000001000 010100 10--00000 1001110001 0000 0000000 10-100110 0-0001000- --

Caddo0000100100 0000-00000 100-00000 -0000-000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 01010100--10--00000 10-1110001 000000 0000000 10-0100110 10001000- --

Leiobunum0000100100 0000-00000 1010-00000 -0000-0000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 010B01C00- -10--00000 1001110001 000--00000 1000000000 10-0100110 101001000- --

Sclerobunus0000000100 0000-00000 100-00000 -0001-000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 010001C00- -10--00000 1001110001 00--00000 1000000000 10-0100110 10001000- --

Gonyleptes0000000100 0000-00000 1010-00000 -0001-000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 010001C00- -10--00000 1001110001 010--00000 100000000 10-0100110 10001000- --

Centruroides0000100000 0100-10000 0111-00000 -0000-1000 100--00001 1000000001 0000010010 01--012110 0100-10001 00116000-0 001000-000 00--0311-1 0011001000 0005110011 1110001011 1101101000 0010110100 0000100100 10-1100111 0-10010011 00

Hadrurus0000100000 0100-10000 0111-00000 -0000-1000 100--00001 1000000001 0000010010 01--012110 0100-10001 00116000-0 001000-000 00--0311-1 0011001000 0005110011 1110001011 1101101000 0010110111 0000100100 10-1100111 0-10010011 00

Heterometrus0000100000 0100-10000 0111-00000 -0000-1000 100--00001 1000000001 0000010010 01--012110 0100-10001 00116000-0 001000-000 00--0311-1 0011001000 00051100111110001011 1101101000 001020111 0000100100 10-1100111 0-10010011 00

Prearcturus000010000 010 -0 -000- 100--001 000 000-0 00- --0311- 0 0

Palaeoscorpius00000000 0D0 -0 -000- 100--0000 00 00000 0 10-0001 06000-0 000-00 --0311- 1 01 0 0

Stoermeroscorpio00000000 00000 -0 -000- 100--0000 000 00000 0 10-0001 06000-0 000-00 --0311- 110 01 1 0

Table 1 Continued

PHYLOGENY OF THE ARACHNID ORDERS 227

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

the Ordovician to the Permian (Tollerton 1989) Phy-logenetic relationships among the eurypterids havenot been rigorously addressed (but see Beall amp Laban-deira 1990) However members of Stylonuroideaappear to be plesiomorphic in retaining relativelyunspecialized legs and were represented by Stylonu-rus (eg Clarke amp Ruedemann 1912 Waterston1979) In the remaining eurypterids the last leg ispaddle-like (Tollerton 1989) The non-stylonuroideurypterids were represented in the matrix by thewell-preserved and well-studied Baltoeurypterus tet-ragonophthalmus especially as described by Selden(1981)

Chasmataspidida Members of this fossil aquaticgroup (OrdovicianndashDevonian six genera) have a four-segmented mesosoma and a nine-segmented meta-soma (Dunlop Poschmann amp Anderson 2001 DunlopAnderson amp Braddy 2004 Dunlop 2002a) Three gen-era were included Chasmataspis from the monotypicfamily Chasmataspididae as well as Diploaspis andOctoberaspis from Diploaspididae

Haptopoda This is a fossil terrestrial group (Carbon-iferous) containing one known species Plesiosiromadeleyi The known specimens have been re-examined by Dunlop (1999)

Palpigradi The extant palpigrades (two families sixgenera ~50 spp) (Harvey 2002) are a poorly studiedgroup of small-bodied terrestrial (some semi-aquatic)arachnids The two most well-studied generaEukoenenia (Eukoeneniidae) especially E mirabilisand Prokoenenia (Prokoeneniidae) especiallyP wheeleri were included in the matrix The oneknown fossil species Paleokoenenia mordax is rela-tively recent (Pliocene) and to the extent known isfairly similar to extant palpigrades (Rowland amp Sis-som 1980) It was not included in the analysis

Trigonotarbida This is a fossil terrestrial order(Upper Silurian to Lower Permian ~ten families ~50spp) The genera coded here were drawn from twoLower Devonian Lagerstatten Gilboarachne from theGilboa Formation (New York) (Shear et al 1987) and

Proscorpius00000000 00000 -0 -000- 100--0000 010 00000 0 10-0001 06000-0 000-00 --0311- 110 01 1 0

Chthonius0000100000 0310-00120 0000-10100 -1000-0000 100--00100 0000000001 0000010010 01--012100 0E00-00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Neobisium0000100000 0310-00121 0000-10100 -1000-0010 100--00100 0000000001 0000010010 01--012100 0100-00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Feaella0000100000 0310-00120 0000-10100 -1000-0000 100--00100 0000000001 0000010010 01--012100 000--00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Chelifer0000100000 0310-00121 0000-10100 -1000-0010 100--00100 0000000001 0000010010 01--012100 000--00001 01115000-0 000000-000 00--000--1 0000021000 01700-014 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Eremocosta0000110000 0310-00121 0000-00010 -1000-0000 001010100 0000000001 00011-0000 02--010-10 1101010001 01114000-0 000000-000 00--000--1 00000200 13111013 010--10000 1001000000 000--00100 1010000100 01-1100001 0-00000110 00

Galeodes0000110000 0310-00121 0000-00010 -1000-0000 001010100 0000000001 00011-0000 02--010-10 1101010001 01114000-0 000000-000 00--000--1 00000200 13111013 010--10000 1001000000 000--00100 1010000100 011100001 0-00000110 00

Table 1 Continued

228 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Palaeocharinus from the Rhynie Chert (Scotland)(Dunlop 1994 Fayers Dunlop amp Trewin 2004)

Araneae (spiders) This is a very large order (110 fam-ilies ~3600 genera ~39 000 extant spp) (Platnick2005) represented here by one genus from each of thethree principal extant lineages (Platnick amp Gertsch1976 Coddington amp Levi 1991 Coddington 2005)Liphistius (Mesothelae) Aphonopelma (Mygalomor-phae) and Hypochilus (Araneomorphae) The sampleis small with respect to known diversity but the basalphylogeny and relevant groundplan states of the orderare well established and states derived from the rep-resented taxa are consistent with them

Amblypygi (whipspiders) This is a small extant ter-restrial order (five families 17 genera 136 spp)(Harvey 2002) and was represented in the matrix bytwo fairly well-studied genera Charinus and Phry-nus Charinus retains several features that appear tobe plesiomorphic for the order (eg adult pedal pul-villi coxal glands associated with leg 3 eversiblevesicles) Weygoldt (1996) has examined the phylog-eny of the order and has provided an important sum-mary of morphology and general biology (Weygoldt2000)

Thelyphonida or Uropygi ss (whipscorpions vinega-roons) Extant whipscorpions (16 genera 106 spp)(Harvey 2002) were represented here by one well-studied species Mastigoproctus giganteus althoughadditional information was drawn from Typopeltis andThelyphonus The morphology of the group is highlyconserved there have been no modern studies ofintraordinal phylogeny A controversial fossil speciesfrom the late Carboniferous Proschizomus (Dunlop ampHorrocks 19951996) was also included but less well-preserved Carboniferous forms were not

Schizomida This small order (two extant families~34 genera ~200 spp) (Harvey 2002) is widelyregarded as the sister group of Thelyphonida Thetaxon sample included one representative from eachextant family Protoschizomus (Protoschizomidae) andStenochrus (Hubbardiidae) especially S portoricen-sis There are three recent (Pliocene) fossil speciesall from the same locality these were not includedhere

Ricinulei The extant ricinuleids (three genera 55spp) (Harvey 2002) were represented by two generaCryptocellus and Ricinoides Detailed studies of grosscuticular anatomy and post-embryonic developmentare available for a representative of each genus (Cryp-tocellus Pittard amp Mitchell 1972 Ricinoides Legg1976) and some information on internal anatomy isavailable for Ricinoides (Millot 1945) and this hasbeen extrapolated to Cryptocellus in the matrix There

are two basic fossil types (Selden 1992) one resem-bling modern taxa and another with a unique opistho-soma that superficially resembles the closed elytra of abeetle (ie curculioids) Two fossils from the formergroup were included Terpsicroton and Poliochera asthey are reasonably well preserved and show impor-tant characters (eg two pairs of eyes or evidence ofopisthosomal diplosegmentation) not expressed inextant forms No curculioid ricinuleids were includedas they appear to offer no additional information rel-evant to resolving ordinal relationships

Opilioacariformes (= Opilioacarida Notostigmata)The opilioacariform mites (nine genera 20 spp) (Har-vey 2002) are generally regarded as plesiomorphicAcari and are fairly conserved in their morphologyThe group was represented by two species Neocarustexanus and Siamacarus withi N texanus has a typi-cal opilioacariform morphology and its external anat-omy has been particularly well studied (eg Van derHammen 1989 Klompen 2000) S withi differs frommost other opilioacariforms in having trichobothriaand three rather than two pairs of lateral eyes (the lat-ter also in Paracarus) features that are potentiallysignificant for assessing ordinal relationships Someinformation on internal morphology of Opilioacaruswas taken from With (1904) and extrapolated to Neo-carus and Siamacarus

Parasitiformes This large group of mites comprisesthree clades Holothyrida Ixodida (ticks) and Mesos-tigmata (= Gamasida) Holothyrida (three familiesfive genera lt 30 spp) (Walter amp Proctor 1999) wasrepresented here by two species Australothyrus ocel-latus and Allothyrus constrictus Ixodids (three fami-lies 22 genera ~860 spp) (Kierans amp Robbins 1999)were represented by Amblyomma and Argas andmesostigmatids (~70 families ~10 000 spp) were rep-resented by Glyptholaspis confusa

Acariformes (= Actinotrichida) This is a very largediverse group of mites that includes the Endeostig-mata Sarcoptiformes and Prostigmata Endeostig-mata is an apparently paraphyletic assemblage of~ten families A morphologically generalized speciesAlycus roseus (Bimichaeliidae) was included herewith character states extracted largely from Van derHammen (1989) Sarcoptiformes (gt 200 families) iseffectively synonymous with Oribatida with Astig-mata being a large derived lsquooribatidrsquo clade (Norton1998 Maraun et al 2004) The group was representedby an early divergent plesiomorphic genus Palaeac-arus (Palaeosomata) and a more derived and well-studied genus Archegozetes (especially A longisetusand A magnus) The Prostigmata (~120 families~7000 spp) (Walter amp Proctor 1999) was representedby two genera Allothrombium and Microcaeculus

PHYLOGENY OF THE ARACHNID ORDERS 229

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Pseudoscorpiones The pseudoscorpions (24 families425 genera ~3200 spp) (Harvey 2002) were repre-sented in the matrix by four terminal genera Chtho-nius (Chthoniidae) was drawn from the basallydivergent superfamily Chthonioidea (Harvey 1992)Feaella (Feaellidae) represented the enigmatic super-family Feaelloidea which has an unusual combinationof characters and may be an early divergent lineageclose to Chthonioidea (Harvey 1992) Neobisium (Neo-bisiidae) is morphologically intermediate between theearly divergent families and the higher families rep-resented by Chelifer (Cheliferidae)

Solifugae (= Solpugida) (sun spiders) The order issmall (~1000 spp) but the absence of modern phylo-genetic treatments precludes meaningful estimates ofgenera and families (Harvey 2002) The taxon sampleincluded two genera one from the New World Eremo-costa and one from the Old World Galeodes

Opiliones (harvestmen) Extant harvestmen (~25 fam-ilies ~500 genera ~6000 spp) are divided into twosuborders Cyphophthalmi and Phalangida Cyphoph-thalmids were represented here by Chileogovea oedi-pus (Petallidae) which is currently under study by theauthor and Cyphophthamus duricorius (Sironidae)which is probably the most well-studied member of thesuborder (eg Janczyk 1956) Phalangida includesthree major groups Laniatores Eupnoi and Dyspnoibut there is disagreement as to whether Dyspnoi is thesister group to Laniatores or to Eupnoi (Shultz ampRegier 2001 Giribet et al 2002) Laniatores was rep-resented by Sclerobunus and Gonyleptes and Eupnoiby Leiobunum and Caddo

Scorpiones (scorpions) The extant scorpions (16 fam-ilies 155 genera 1279 spp) (Fet et al 2000) wererepresented by three taxa Centruroides vittatus(Buthidae) Hadrurus arizonensis (Iuridae) and Heter-ometrus spinifer (Scorpionidae) Buthidae is widelyregarded as the sister group to other extant lineagesand similarities among the represented terminals arelikely to be ground plan features of extant scorpionsgenerally The morphology of fossil scorpions is sub-stantially more diverse (Kjellesvig-Waering 1986)but their phylogenetic relationships are unclear (butsee Jeram 1998) The fossil taxa used in this study(Proscorpius Stoermeroscorpio Palaeoscorpius Pre-arcturus) were chosen for quality of preservation andor presence of a phylogenetically significant constella-tion of characters

CHARACTER CODING

Most character states were determined from directobservation the primary literature and authoritativereviews In some cases states were assigned to termi-

nals based on observations from related species asnoted in the Appendix In the matrix state lsquo-rsquo indicatesthat the character is inapplicable because the taxonlacks a more general character For example if a taxonlacks eyes then special features of the eyes (numberposition retinal configuration etc) are inapplicable tothat taxon A lsquorsquo indicates that the state is unknown oruncertain An entry with multiple states (eg 01 345) should be treated as an ambiguity code not a poly-morphism it indicates that two or more interpreta-tions of homology are applicable and that assignmentwas established analytically by character concor-dance Although lsquo-rsquo lsquorsquo and lsquo01rsquo are analyticallyidentical when applied to binary characters theyprovide information about the empirical status of thecharacter state in specific taxa Characters are citedthroughout the text as italicized numbers in parenthe-ses and are discussed in the Appendix

PHYLOGENETIC ANALYSIS

Phylogenetic analysis of the full data matrix wasperformed using the program Tree Analysis usingNew Technologies (TNT) ver 1 (Goloboff Farris ampNixon 2000) using lsquotraditionalrsquo search based on1000 replicates using TBR branch swapping Resultswere compared to those obtaining using the ratchetalgorithm (Nixon 1999) to determine any differencedue to analytical approach Nodal support for theminimal-length topology was evaluated by boot-strap (Felsenstein 1985) and Bremer support(Bremer 1994) Bootstrap analysis was conducted inTNT and based on 1000 pseudoreplicates each anal-ysed by ten random-addition replicates using TBRbranch swapping A nexus file containing the result-ing 1000 trees was imported into PAUP ver 10(Swofford 2002) to obtain bootstrap frequenciesBremer support was determined in TNT by con-straining specific nodes in the minimal-length topol-ogy and then determining the shortest tree that didnot recover the specified clade The difference inlength between the unconstrained and constrainedminimal-length trees is the Bremer support Theeffect of homoplasy on results was explored by con-ducting implied weights analysis (Goloboff 1993) inTNT Six analyses were conducted each with con-stant of concavity (k) set to a different integer valueof 1ndash6 where 1 is weighted most severely againsthomoplasious characters Each implied weightsanalysis was conducted using lsquotraditionalrsquo searchbased on 1000 replicates using TBR branch swap-ping The same procedures were used in analysing amatrix that included only extant taxa Howevercharacters rendered uninformative by removal offossil taxa (ie 49 98 99 100 113 114 125 161)were excluded prior to analysis

230 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

COMPARISONS OF ALTERNATIVE HYPOTHESES

Phylogenetic hypotheses proposed in the recent liter-ature were also evaluated including those thatattempted to resolve arachnid phylogeny completely(Fig 2) their hypothesized subclades (Fig 3) andhypotheses that proposed only specific nodes (Figs 34) The fully resolved topologies of Weygoldt amp Paulus(1979) Van der Hammen (1989) Shultz (1990)Wheeler amp Hayashi (1998) and Giribet et al (2002)were compared with the optimal topology using theTempleton test (Templeton 1983) as implementedin PAUP Internal relationships of multisampledorders were constrained to match those of the optimaltopology unless the original authors explicitlyfavoured an alternative Node-specific hypotheseswere evaluated by determining the frequency withwhich the node was recovered in bootstrap analysisThey were also evaluated by constraining parsimonyanalysis in TNT to recover the shortest tree contain-ing each specific node and then assessing the effect onrelative tree length and overall resolution takingnote of effects on otherwise stable or well-supportedclades The entire matrix was used to assess the twocases where node-specific hypotheses involved fossiltaxa (Fig 4)

RESULTS

EXTANT TAXA

Unweighted analysis of extant taxa produced two min-imal-length trees (length 383 CI 057) (Fig 5A) withconflicts limited to relationships within Pseudoscorpi-ones Multiply sampled orders were recovered asmonophyletic with high nodal support as measured bybootstrap percentage (BP gt 80) except Acari (BP 30)There were few well-supported interordinal groupsexcept Uropygi (BP 100) Pedipalpi (BP 100) Tetra-pulmonata (BP 93) and ArachnidaXiphosura (BP100) All other interordinal relationships were recov-ered with BP less than 60

Results from implied weights analysis (IWA) indi-cated a significant effect of homoplasy on resolution ofthe deepest interordinal relationships although theinterordinal clades Stomotheca Haplocnemata Acar-omorpha Tetrapulmonata Pedipalpi and Uropygiwere recovered under all values of k IWA with k = 1 (1tree best score = 5124) recovered Megoperculata(= Palpigradi + Tetrapulmonata) and Micrura(= Megoperculata + Acaromorpha) with Micrura beingthe sister group to Haplocnemata (= Pseudoscorpiones+ Solifugae) and with Stomothecata (= Opiliones +Scorpiones) being the sister group to all other arach-nids (Fig 5B) With k = 2 (1 tree best score = 3807)and k = 3 (1 tree best score = 3054) Megoperculatawas recovered and Stomotheca remained the sister

group to other arachnids but the relative positions ofHaplocnemata and Acaromorpha changed (Fig 5C)IWA using k = 4 (1 tree best score = 2559) k = 5 (1tree best score = 2205) and k = 6 (1 tree bestscore = 1938) recovered topologies consistent with theminimal-length topology derived from unweightedcharacters (Fig 5A)

EXTANT AND FOSSIL TAXA

Unweighted analysis of extant and fossil taxa pro-duced 18 minimal-length trees (length 426 CI 056)(Fig 6A) with conflicts limited to relationships withinPseudoscorpiones Scorpiones and Araneae Multiplysampled orders were again recovered as monophyleticwith high nodal support (BP gt 80) except Acari whichwas reconstructed as diphyletic with Anactinotrichidabeing the sister group to Ricinulei (BP 33) Therewere no well-supported interordinal groupings (ieBP gt 80) except Uropygi (BP 89) Pedipalpi (BP 92)and Arachnida (BP 90)

Results from IWA indicated a significant effect ofhomoplasy on resolution of the deepest interordinalrelationships in Arachnida although StomothecataHaplocnemata Acaromorpha Cryptognomae Pantet-rapulmonata Tetrapulmonata Schizotarsata Pedi-palpi and Uropygi were recovered under all values ofk IWA with k = 1 (7 trees best score = 5675) recov-ered Megoperculata and Micrura with Micrura beingthe sister group to Haplocnemata Stomothecata wasnot recovered in the strict consensus (Fig 6B) Withk = 2 (3 trees best score = 4243) IWA recovered thetopology shown in Fig 6C with Palpigradi being thesister group to all other arachnids The strict consen-sus topologies reconstructed using k = 3ndash6 were iden-tical to those obtained with k = 2 (k = 3 3 trees bestscore = 3419 k = 4 3 trees best score = 2874 k = 53 trees best score = 2483 k = 6 3 trees bestscore = 2188)

COMPARISONS OF ALTERNATIVE HYPOTHESES

The hypothesis of Weygoldt amp Paulus (1979) (Fig 2)was 398 steps (15 steps longer than the unconstrainedminimal-length tree) Van der Hammen (1989) was400 steps (17 steps longer) Shultz (1990) was 399steps (six steps longer) Wheeler amp Hayashi (1998)was 391 steps (eight steps longer) and Giribet et al(2002) was 394 steps (11 steps longer) Templetontests conducted at the 005 significance level rejectedthe hypothesis of no difference between the optimaltopology and those of Weygoldt amp Paulus (P = 00002)Van der Hammen (P = 00011) Wheeler amp Hayashi(P = 00018) and Giribet et al (P = 00343) The Shultztopology was not significantly different (P = 01444)from the optimal topology

PHYLOGENY OF THE ARACHNID ORDERS 231

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 2 Fully resolved phylogenetic hypotheses of extant euchelicerate groups proposed in the recent literature Note thesimilarity in the topologies of the parsimony-based analyses by Shultz (1990) Wheeler amp Hayashi (1998) and Giribet et al(2002) The Giribet et al topology is based on neontological data (morphology and molecules) and the original lsquoROOTrsquo maybe an artefact from use of the highly divergent pycnogonids as an outgroup

Xip

hosu

raS

corp

ione

sO

pilio

nes

Ara

neae

Am

blyp

ygi

Uro

pygi

Ric

inul

eiA

nact

inot

richi

daS

olifu

gae

Pse

udos

corp

ione

sP

alpi

gra

diA

carif

orm

es

Xip

hosu

raS

corp

ione

sA

rane

aeA

mbl

ypyg

iU

ropy

giP

alpi

gra

diS

olifu

gae

Pse

udos

corp

ione

sO

pilio

nes

Ric

inul

eiA

cari

Lipoctena

Arachnida

Apulmonata

Holotracheata

Cryptoperculata

Acaromorpha(ldquoAcarinomorphardquo)

Tetrapulmonata(rdquoMegoperculatardquo)

LabellataHaplo-cnemata

Weygoldt amp Paulus (1979)

Labellata

Tetrapulmonata(rdquoArachnidardquo)

Haplocnemata(rdquoApatellatardquo)

Van der Hammen (1989)

Xip

hosu

raP

alpi

gra

diA

rane

aeA

mbl

ypyg

iU

ropy

giR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Pedipalpi Haplocnemata

Novogenuata

Dromopoda

Shultz (1990)

Acaro-morpha

Micrura

Xip

hosu

raP

alpi

gra

diU

ropy

giA

mbl

ypyg

iA

rane

aeR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Labellata Haplocnemata

Novogenuata

Dromopoda

Wheeler amp Hayashi (1998)

Acaro-morpha

Micrura

Xip

hosu

raA

cari

Pal

pig

radi

Ric

inul

eiA

rane

aeA

mbl

ypyg

iU

ropy

giO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Tetrapulmonata

PedipalpiHaplocnemata

Novogenuata

Dromopoda

Giribet et al (2002)

ROOT

Micrura

Crypto-gnomae

Sternocoxata

Epimerata

RostrosomataMyliosomata

Neosternata

232 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 3 Consensus trees produced by parsimony analysis of neontological data constrained to produce relationships pro-posed in the recent literature The constrained (target) node is indicated by a black dot and the taxa encompassed by theconstraint are enclosed in a box Numbers below each tree represent the number of minimal-length constrained treeslength of minimal-length trees difference in the length of the unconstrained minimal-length tree and the constrained min-imal-length tree and percentage unconstrained bootstrap trees in which the target node was recovered respectively Thesetrees indicate the effect on branch length imposed by specific hypotheses and impact of constraining target nodes on non-target nodes

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Lipoctena4 385 2 lt5

Xi

Sc

Op

Ps

So

Pa

Ur

Am

Ar

Ri

An

Ac

Labellata6 397 14 lt5

Xi

Sc

Ar

Am

Ur

Op

Pa

Ps

So

Ri

An

Ac

Apulmonata2 390 7 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

So

Ri

An

Ac

Holotracheata5 390 7 lt5

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptoperculata2 388 5 lt5

Xi

Sc

Op

Pa

So

Ps

Ar

Am

Ur

Ri

An

Ac

Rostrosomata6 384 1 25

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Neosternata3 388 5 lt5

Xi

Sc

Op

Ar

Am

Ur

Ri

An

Ps

So

Pa

Ac

Coxisternata2 388 5 lt5

Epimerata7 387 4 11

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptognomae6 384 1 24

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Micrura3 385 2 7

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Megoperculata3 384 1 28

Xi

Sc

Ps

So

Op

Pa

Ar

Am

Ur

Ri

An

Ac

Novogenuata10 388 5 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

Ri

So

An

Ac

Solifugae + Acari4 388 5 lt5

Shultz (1990) Alberti amp Peretti (2002)

Van der Hammen (1989)

Weygoldt amp Paulus (1979)

Xi

Ar

Am

Ur

So

Ps

Sc

Op

Ri

Pa

An

Ac

PHYLOGENY OF THE ARACHNID ORDERS 233

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Results from analyses of node-specific hypothesesare summarized in Figures 3 4 In most cases pro-posed relationships were recovered in fewer than 5of bootstrap pseudoreplicates The notable exceptionswere Rostrosomata (25) Cryptognomae (24) andMegoperculata (28) and the strict-consensus con-strained minimal-length tree for each was only onestep longer than the unconstrained optimal tree Inmost cases constrained topologies did not add manysteps most were less than five steps longer and nonewas greater than 14 steps longer than the un-constrained minimal-length tree It is noteworthyhowever that even constrained nodes that imposerelatively few extra steps sometimes supportedimprobable relationships or eliminated clades thatwere well supported or stable in the unconstrainedanalysis Specific examples and their implications arediscussed below

DISCUSSION

RECENT ISSUES IN ARACHNID PHYLOGENY

Available morphological evidence consistently re-solves some interordinal relationships and fails to

resolve others (Fig 1) Continued progress depends onthe ability of arachnologists to discover new charac-ters and to assess the evidence critically Whilereviewing the recent arachnological literature severalaspects of phylogenetic practice emerged that seemedcounterproductive to both the perception and theactual rate of progress toward resolving arachnid phy-logeny Some of these are summarized here withspecific examples given in the remainder of theDiscussion and in the Appendix

There is a tendency to portray arachnid ordinal phy-logeny as more poorly resolved and contentious than isactually the case (Coddington et al 2004) Phyloge-netic hypotheses generated during different historicalperiods and using differing standards of evidence areoften cited as examples of current disagreement (egSelden 1993 Dunlop 1996 Selden amp Dunlop 1998Wheeler amp Hayashi 1998) In fact recent parsimony-based analyses of morphology have tended to convergeon topologies with internal structures congruent withthose found here (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) (Fig 2) Arachnidphylogeny is not fully resolved especially at its deep-est levels (Fig 1) but this does not mean that allaspects of arachnid phylogeny are controversial orpoorly supported by the available evidence

Matrices are sometimes constructed by uncriticallsquorecyclingrsquo of erroneous or problematic charactersbased on diverse secondary sources (Jenner 2001)Conclusions derived from mixtures of valid invalidspeculative and redundant characters are sometimesportrayed as the phylogenetic signal provided by mor-phology Data recycling can perpetuate errors (see 1320 39 52 136 140 144 152 169 171ndash174) legiti-mize speculations (see 13 32 63 95) or create dupli-cate or non-independent characters (see 13 30 70 77172) The lsquolateral organrsquo is a notable exampleYoshikura (1975) equated the embryonicearly post-embryonic lsquolateral organsrsquo of Amblypygi Thelyphonidaand Solifugae with the dissimilar lsquolateral organrsquo ofXiphosura but failed to note the very similar Cla-paregravede organ of Acariformes (see 173 174) This cod-ing was recycled by Wheeler amp Hayashi (1998) andthen by Giribet et al (2002) Error is probably inevi-table when assembling a large matrix from morphol-ogy including the one presented here but this can beminimized by making original observations and byconsulting primary sources

Some workers advocate weighting characters a pri-ori on the basis of structural or functional complexity(eg Kraus 1998 Dunlop amp Braddy 2001) and dismissphylogenetic conclusions derived from equal-weightsparsimony In short these workers criticize parsimonyfor emphasizing data quantity over data quality Thiscriticism ignores the intense debate in systematic biol-ogy that eventually led to widespread adoption of

Figure 4 Consensus trees produced by parsimony analy-sis of the full data matrix constrained to produce relation-ships proposed in the recent literature See legend toFigure 3 for details

Xi

Eu

Sc

Op

Pa

Ac

Ri

An

So

Ps

Tr

Ar

Ha

Am

UrEurypterida slat + Scorpiones12 435 9 lt5

Xi

Eu

Sc

Op

Pa

Ac

An

So

Ps

Ri

Tr

Ar

Ha

Am

UrTrigonotarbida+ Ricinulei28 434 8 lt5

Dunlop amp Braddy (2002)et al

Dunlop (1996)

234 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 5 Results from analysis of neontological data A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology AFor B and C relationships within terminal clades are the same as those shown in A

LimulusTachypleusProkoeneniaEukoeneniaCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAllothyrusAustralothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesLiphistiusAphonopelmaHypochilusCharinusPhrynusStenochrusProtoschizomusMastigoproctusCyphophthalmusChileogoveaCaddoLeiobunumSclerobunusGonyleptesCentruroidesHadrurusHeterometrusChthoniusFeaellaNeobisiumCheliferEremocostaGaleodes

10031

10018451

501561

733

723

944

633

10010

832

901

943

903

522

341

1007

1008

Acaromorpha

Haplocnemata

10012

Stomothecata

Opiliones

Scorpiones

Pseudo-scorpiones

Solifugae

Acariformes

Parasitiformes

Opilioacariformes

Ricinulei

Xiphosura

Palpigradi

Thelyphonida

Schizomida

Araneae

Amblypygi

10011

936853

100141005

1009

Tetrapulmonata Pedipalpi

Uropygi

692

522572

968

331

603

81

511

161

101

Acari

Dromopoda

Arachnida

A

Xiphosura

Stomothecata

Haplocnemata

Acaromorpha

Palpigradi

TetrapulmonataMicrura

Megoperculata

BXiphosura

Stomothecata

Acaromorpha

Haplocnemata

Palpigradi

TetrapulmonataMegoperculata

C

PHYLOGENY OF THE ARACHNID ORDERS 235

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 6 Results from analysis of the full data matrix A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology CFor B and C relationships within terminal clades are the same as those shown in A

WeinberginaLimuloidesEuproopsTachypleusLimulusStylonurusBaltoeurypterusChasmataspisDiploaspisOctoberaspisCyphophthamusChileogoveaCaddoLeiobunumSclerobunusGonyleptesProscorpiusPalaeoscorpiusStoermeroscorpioPrearcturusCentruroidesHadrurusHeterometrusProkoeneniaEukoeneniaTerpsicrotonPoliocheraCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAustralothyrusAllothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesEremocostaGaleodesChthoniusFeaellaNeobisiumCheliferGilboarachnePalaeocharinusLiphistiusAphonopelmaHypochilusPlesiosiroCharinusPhrynusStenochrusProtoschizomusProschizomusMastigoproctus

631

893

923

984

984

671

341

671

986

10011

423

974

Xiphosura

Eurypterida sstr

Chasmataspidida

Opiliones

Scorpiones

Palpigradi

Ricinulei

Opilioacariformes

Parasitiformes

Acariformes

Solifugae

Pseudoscorpiones

Trigonotarbida

Araneae

Haptopoda

Amblypygi

Schizomida

Thelyphonida

842

251331

361311

552

832

882

925

913

573

561291

443

976

541

954

743644

732521

571

331

833

663591

966

573

71

221

905

832

351

Uropygi

Pedipalpi

Haplocnemata

Tetrapulmonata

Anactinotrichida

Cryptognomae

Acaromorpha

Stomothecata

Pantetrapulmonata

Arachnida

Eurypterida slat

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatProscorpiusStoermeroscorpioPalaeoscorpiusPrearcturusCentruroidesHadrurusHeterometrusOpilionesHaplocnemataRicinuleiAnactinotrichidaAcariformesPalpigradiPantetrapulmonata

Megoperculata

Micrura

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatPalpigradiStomothecataHaplocnemataPantetrapulmonataAcaromorpha

A

B

C

Schizotarsata

236 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

parsimony and instead advocates a return to thespeculative and subjective approaches of the late 19thand early 20th century that once threatened thescientific legitimacy of the discipline (Bowler 1996)Furthermore it misrepresents the properties of parsi-mony-based analyses Specifically characters withfunctional significance are readily encompassed byparsimony analyses and in fact characters derivedfrom locomotor systems (eg 46ndash94) have played animportant role in developing current ideas aboutarachnid phylogeny (eg Shultz 1989 1991) In addi-tion lsquocomplex charactersrsquo can be viewed as compositesof several characters such that morphological com-plexity is effectively weighted by the number of inde-pendent lsquosubcharactersrsquo it contains For example thelsquosucking stomachrsquo once considered a synapomorphy ofLabellata (= Araneae + Amblypygi Fig 3) is codedhere as a composite of three characters (199ndash201) TheLabellata hypothesis was not corroborated in thepresent analysis (Figs 1 3ndash5) and in fact was highlydisfavoured (Fig 3) but this result cannot be dis-missed as a failure to acknowledge the complexity ofthe character

Some workers support specific (target) clades withone or more similarities without exploring the impacton overall homoplasy or relationships among non-target clades Each character offers its own phyloge-netic hypothesis which may or may not be consistentwith relationships implied by other characters It isexceedingly rare for all characters to be perfectly com-patible in the phylogenetic hypotheses they supportand criteria such as parsimony have been developed todiscover those hypotheses that minimize the conflict-ing phylogenetic signals of different characters Eventhough character conflict (homoplasy) is a virtuallyinescapable phenomenon in comparative biology it isnot uncommon for workers to discover one or morecharacters and to promote their phylogeneticimplications over alternative hypotheses even thosethat otherwise appear to be well supported Thisapproach may have value in highlighting new dataand perspectives but accomplishes this by promotingthe erroneous impression that all aspects of arachnidphylogeny are so tenuous that a single character canfalsify even well-supported hypotheses Several exam-ples of this approach have appeared in the recentarachnological literature

For example Alberti amp Peretti (2002) argued thataflagellate spermatozoa (163) are a compelling syna-pomorphy for a Solifugae + Acari clade and dismissedsome of the characters that support Solifugae + Pseu-doscorpiones as having lsquodebatable valuersquo Yet theirproposal rejects two hypotheses that are consistentlyrecovered in recent phylogenetic analyses that isHaplocnemata (= Solifugae + Pseudoscorpiones) andAcaromorpha (= Acari + Ricinulei) (Figs 1 2) Rejec-

tion of Haplocnemata would require its presumed syn-apomorphies to be reinterpreted as homoplasiesincluding features of the chelicerae (18 19 20) pre-oral chamber (13 32) legs (12 48) and respiratorysystem (126) The same reason applies to Acaromor-pha and its synapomorphies It is noteworthy that aSolifugae + Acari clade was recovered in the presentanalysis in fewer than 5 of bootstrap pseudorepli-cates and that analyses constrained to recover thisclade were five steps longer than the minimal-lengthtree and favoured a problematic clade unitingRicinulei and Pseudoscorpiones (Fig 3)

In another example Dunlop (1996) proposed a closerelationship between Trigonotarbida and Ricinuleibased on two-segmented chelicerae (18) prosomandashopisthosoma coupling mechanism (96) diplotergites(100 101) and longitudinally divided opisthosomaltergites (115) However phylogenetic analyses con-strained to recover this relationship required eightadditional steps eliminated support for Acari and nec-essarily rejected Acaromorpha and forced its synapo-morphies to be reinterpreted as homoplasies Alberti(2005) proposed an interesting hypothesis for the evo-lution of male gonads in tetrapulmonates (153) butchose to accept Labellata (= Araneae + Amblypygi) indeveloping his argument over the much more well-supported Pedipalpi (= Amblypygi + Uropygi) (Shultz1999) a phylogenetic reconfiguration that was recov-ered here in fewer than 5 of bootstrap pseudorepli-cates increased tree length by a minimum of 14 stepsand resulted in the collapse of Stomothecata and Hap-locnemata (Fig 3) Many other examples can be cited

Promoting or defending a specific phylogenetichypothesis via lists of compatible synapomorphies is acommon but problematic approach By restrictingattention to the states of specific characters at one ortwo target nodes one can easily overlook the unin-tended impact of the hypothesis on phylogenetic sig-nal elsewhere and its effect on non-target clades Anode supported by a long list of synapomorphies mayseem convincing taken in isolation but may becomeless acceptable when its full phylogenetic implicationsare explored

SUMMARY OF INTERORDINAL ARACHNID CLADES RECOVERED IN THIS ANALYSIS

Arachnida Lamarck 1801Analyses consistently recovered Arachnida as a mono-phyletic group with high bootstrap support (Figs 1 45) Possible synapomorphies include the loss of thecarapacal pleural doublure (9) cardiac lobe (10) pedalgnathobases (52) and moveable endites (53) and thegain of aerial respiration (120) and an anteriorly oranteroventrally directed mouth (185) Some tradi-tional synapomorphies such as slit sensilla (142) and

PHYLOGENY OF THE ARACHNID ORDERS 237

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

fluid feeding (184) may have appeared later in arach-nid evolution but this can only be decided once theinternal phylogeny of Arachnida has been determined

Some workers regard many proposed arachnid syn-apomorphies as adaptations to terrestrial life andthereby link the hypothesis of arachnid monophyly tothe hypothesis of a single ancestral aquatic-to-terres-trial transition and arachnid polyphyly to multipletransitions (eg Selden amp Jeram 1989 Dunlop 1997Dunlop amp Webster 1999 Dunlop amp Braddy 2001)Given this line of reasoning the existence of appar-ently aquatic scorpions in the fossil record (Kjellesvig-Waering 1986 Jeram 1998) and the inference thatterrestrialization occurred late in scorpion evolutionthese workers conclude that character states support-ing arachnid monophyly are actually convergencesand do not necessarily support arachnid monophyly

However this approach to assessing phylogenetichypotheses is founded on overly simplistic assump-tions such as the ability of the investigator to discrim-inate unerringly between characters that existexclusively in aquatic organisms (including fossils)from those that occur exclusively in terrestrial organ-isms There also appears to be an assumption thathomoplasy can be generated through parallelism (iemultiple aquatic-to-terrestrial events) but not throughterrestrial-to-aquatic reversals Furthermore anassumed dichotomy between exclusively aquatic andterrestrial life histories in ancestral arachnids is sim-plistic as illustrated by the amphibious life cycles ofbasal vertebrates and pterygote hexapods In factthese examples demonstrate that there is no neces-sary inconsistency in basing a hypothesis of arachnidmonophyly on the derived terrestrial features of anamphibious ancestor whose descendants thencompleted terrestrialization once or several timesindependently or even returned to a fully aquaticexistence Workers who link the frequency and direc-tion of aquaticndashterrestrial transitions to the assess-ment of arachnid phylogeny do so by endowingthemselves with substantially greater insight thanseems prudent by ignoring the huge gaps in ourunderstanding of early arachnid evolution and bydenying to arachnids the evolutionary complexityknown to exist in other groups

Several palaeontologists have been particularlyactive during the past decade in proposing new char-acters with the stated goal of removing Scorpionesfrom Arachnida and erecting a Scorpiones + Eurypter-ida clade (Braddy amp Dunlop 1997 Dunlop amp Braddy1997 Dunlop 1998 Braddy et al 1999 Dunlop ampWebster 1999) Dunlop amp Braddy (2001) recentlysummarized this evidence and conducted aparsimony-based analysis of Xiphosura EurypteridaScorpiones Opiliones and Tetrapulmonata (but notHaplocnemata Acaromorpha or Palpigradi) using 33

morphological characters Their analysis produced atopology congruent with those generated here (Figs 15 6) including recovery of Stomothecata (= Scorpiones+ Opiliones) However they rejected this result as aproduct of lsquoempirical cladisticsrsquo because it gives thesame weight to prosomal characters that supportarachnid monophyly and to selected opisthosomalcharacters that support their favoured Eurypterida +Scorpiones clade namely a five-segmented metasoma(116) suppression of opisthosomal tergite 1 (95) lossof lamellate respiratory organs on the postgenitalsomite (122) Kiemenplatten (125) loss of respiratorylamellae on the genital segment (121) and a lsquonon-stainingrsquo exocuticle (but see Grainge amp Pearson 1966for evidence of this in Opiliones see Appendix for com-ments on the other characters)

A Eurypterida + Scorpiones clade was not favouredin the present analysis (Fig 6) the strict consensus ofminimal-length trees constrained to recover this clade(Fig 4) is nine steps longer than the tree recovered byanalysis without this constraint It is also noteworthythat Opiliones was consistently reconstructed as thesister group to Eurypterida + Scorpiones in the mini-mal-length constrained trees a provocative result thatwas probably unintended and unanticipated by sup-porters of the Eurypterida + Scorpiones Given thatDunlop amp Braddy (2001) reject equal weights parsi-mony as an arbiter of phylogenetic hypotheses theywould presumably dismiss these results as irrelevantto their argument just as they dismissed their ownparsimony-based results However if the Eurypterida+ Scorpiones hypothesis is to be credible it must beopen to evaluation and potential falsification usingobjective criteria and the subjective or intuitive a pri-ori weighting of characters advocated by Dunlop ampBraddy clearly does not qualify At present it isunclear how one would objectively evaluate Dunlop ampBraddyrsquos proposal with criteria compatible with thoseused in its original formulation For now the Euryp-terida + Scorpiones concept advocated by Dunlop ampBraddy may persist outside the mainstream of mod-ern systematic practice but it is increasingly problem-atic within it

Stomothecata nomnovOpiliones and Scorpiones were consistently recoveredas a monophyletic group The proposed name acknowl-edges a unique preoral chamber the stomothecaformed by coxapophyses of the palp and leg 1 (50)often with an auxiliary role played by the coxapophy-sis of leg 2 (51) In addition the epistome appears tohave been modified for adduction of the palpal coxaeThe lateral walls of the epistome are fused to themedial surfaces of the palpal coxae and the epistomallumen is spanned by a transverse muscle (188) which

238 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

apparently adducts the palpal coxae thereby constrict-ing the stomothecal chamber Scorpions and opilionsare also unique in having a pair of large epistomalarms projecting rearward into the prosoma andattaching to the endosternite (189) The epistomalarms provide attachment sites for pharyngeal dilatormuscles (196) and extrinsic muscles of anterior proso-mal appendages (eg 37) The chelicera is equippedwith a muscle that arises on the carapace and insertson the ventral margin of the second cheliceral article(23) There is an anteriorly placed genital opening(155)

Several notable similarities were found whilereviewing the literature but information from otherarachnid groups was considered too incomplete toallow them to be included in the matrix For exampleboth orders have apparent haemocytopoeitic organsassociated with major nerves of the anterior opistho-soma These are termed supraneural organs in scorpi-ons (Farley 1999) and perineural organs in opilions(Kaumlstner 1935) Haemocytes develop in the cardiacwall in spiders (Seitz 1972) and perhaps amblypygids(Weygoldt 2000) but haemocytopoeitic organs areunknown in most other arachnid groups Germ cellsdifferentiate very early during embryogenesis in bothscorpions and opilions relative to spiders (Moritz1957 Anderson 1973) Additional research is requiredto determine the phylogenetic utility of thesecharacters

Haplocnemata Boumlrner 1904Several workers recognize a close phylogenetic rela-tionship between Pseudoscorpiones and Solifugae(Weygoldt amp Paulus 1979 Van der Hammen 1989Shultz 1990 Wheeler amp Hayashi 1998 Giribetet al 2002) and this was consistently recovered inthe present analysis Haplocnemata is supported byseveral synapomorphies including (i) a feeding com-plex formed by two-segmented chelicerae (18) with aventrolateral intrinsic articulation (19) and perhapsa dorsolateral articulation with the carapace (20)and (ii) a rostrosoma (32) a preoral apparatus formedby an anteriorly projecting epistome affixed dorsallyto enlarged palpal coxae and bordered by lateral pal-pal projections The epistome forms the dorsal compo-nent of a lsquobeakrsquo with the ventral part formed by amidventral sternapophysis (13) fused between thepalpal coxae The locomotor apparatus consists ofcoxae that meet along the ventral midline (12) elon-gate femur-like patellae (48) and an apotele in theadult modified as an eversible empodium or pulvillus(92) The respiratory system is formed entirely (Pseu-doscorpiones) or in part (Solifugae) by paired trachealstigmata opening on opisthosomal somites 3 and 4(126)

Acaromorpha Dubinin 1957A clade comprising Ricinulei and Acari is consistentlyrecovered in the present analysis It is united here bytwo unique and seven homoplasious synapomorphiesThese include a gnathosoma (31) defined in part bymedial fusion of the palpal coxae (30) although pres-ence of a gnathosoma in Ricinulei is debatable as dis-cussed in the Appendix Acaromorphs also have aunique post-embryonic development consisting of ahexapodal larva and up to three octopodal nymphalinstars (176) The pedal patellandashtibia joints are formedby a bicondylar dorsal hinge rather than a single mid-dorsal condyle (69) all postcheliceral segmental gan-glia are unified in the subesophageal ganglion (130)and a postcerebral pharynx is absent (192) The groupis also tentatively united by presence of differentiatedpedal basi- and telofemora (63 64) and the absence ofa ventral (sternal) pharyngeal dilator muscle (199)but these may be symplesiomorphies erroneouslyreconstructed as a synapomorphies

The internal phylogenetic structure of Acaromorphais controversial with many recent workers favouringa monophyletic Acari (Weygoldt amp Paulus 1979Shultz 1990 argued most thoroughly by Lindquist1984) and others advocating a diphyletic Acari withAcariformes being the sister group to Cryptognomae(= Ricinulei + Anactinotrichida) (especially Van derHammen 1979 1989) (Fig 3) Acari was recovered asmonophyletic when fossils were excluded (Fig 2) butnot when they were included (Fig 3) A lsquomite-centredrsquosurvey of arachnid characters may be needed if mor-phology is to offer a compelling solution to the internalphylogeny of Acari and its placement within Arach-nida These issues are not resolved by the presentanalysis

Pantetrapulmonata nom novPantetrapulmonata includes the extinct orders Trigo-notarbida Haptopoda and the extant orders AraneaeAmblypygi Schizomida and Thelyphonida The cladeis united by cheliceral structure (18 19) a megoper-culum (106) booklungs on the genital and first post-genital somites (121 122) and enlargement of theepipharyngeal sclerite (192) It is important to notehowever that most of these characters were coded asuncertain in Plesiosiro (Haptopoda) Aside from theplacement of Haptopoda the monophyly of Trigono-tarbida and the extant orders was anticipated byShear et al (1987) and is generally regarded as amonophyletic group

Tetrapulmonata Shultz 1990Tetrapulmonata was originally proposed on the basisof neontological analyses and encompassed Araneae

PHYLOGENY OF THE ARACHNID ORDERS 239

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Amblypygi Schizomida and Thelyphonida (Shultz1990) and was also recovered here in analysis of neon-tological data (Fig 5) However results from analysisof all taxa required that the Tetrapulmonata conceptbe expanded to include the fossil order HaptopodaFeatures uniting this group are problematic howeveras many states in Haptopoda were coded as uncertain

Schizotarsata nomnovHaptopoda and Pedipalpi (= Amblypygi + Schizomida+ Thelyphonida) are united here in a group named forpossession of divided pedal telotarsi (84) Synapomor-phies include a pointed anterior carapacal margin (3)and elongation of leg 1 (46) As already noted theplacement of Haptopoda should probably be regardedas tentative because the state of many characters inthis group was coded as unknown The placement ofHaptopoda as the sister group to Pedipalpi was antic-ipated by Dunlop (1999 2002c)

Pedipalpi Boumlrner 1904Pedipalpi encompasses Amblypygi and Uropygi(= Thelyphonida + Schizomida) Although widely rec-ognized in the past Pedipalpi was set aside in the mid-20th century in favour of Labellata (= Amblypygi +Araneae) based largely on the presumed synapomor-phies of a narrow prosomandashopisthosoma juncture (97)and sucking stomach (199ndash201) The situation beganto reverse when Shear et al (1987) again highlightedsimilarities of Amblypygi and Uropygi such as rapto-rial palps (35) and antenniform leg 1 (46) Shultz(1989 1990 1999) subsequently described numerousderived similarities including asymmetrical flexormuscles at the pedal femurndashpatella joint (70) threetelotarsomeres on legs 2ndash4 (85) and modification ofthe palpal coxae to provide support for extrinsic pha-ryngeal muscles (197) Pedipalpi has emerged as oneof the most well-supported interordinal relationshipsin Arachnida

Uropygi Thorell 1882Uropygi has long been accepted as a monophyleticunion of Thelyphonida and Schizomida Synapomor-phies include a unique mating behaviour (159) fusedpalpal coxae (30) 2-1-1-1 arrangement of tibial tricho-bothria (144) posterior defensive glands (102) elon-gated patella of leg 1 (68) and many others

ACKNOWLEDGEMENTS

I thank Nikolaj Scharff and three anonymous review-ers for useful comments This work was supported bythe Maryland Agricultural Experiment Station the

General Research Board of the University of Mary-land and National Science Foundation Grant 0416628to Susan Masta

REFERENCES

Abd el-Wahab A 1952 Notes on the morphology of the scor-pion Buthus quinquestriatus (HE) Publications de lrsquoInsti-tut Fouad 1er du Desert 3 1ndash129

Alberti G 1979 Licht- und elektronenmikroskopische Unter-suchungen an Coxaldruumlsen von Walzenspinnen (ArachnidaSolifugae) Zoologischer Anzeiger 203 48ndash64

Alberti G 1995 Comparative spermatology of Cheliceratareview and perspective Memoires du Museum NationaldrsquoHistoire Naturelle 166 203ndash230

Alberti G 2005 Double spermatogenesis in ChelicerataJournal of Morphology 266 281ndash297

Alberti G Coons LB 1999 Acari mites In Harrison FWFoelix RF eds Microscopic anatomy of invertebratesVol 8C Chelicerate Arthropoda New York Wiley Liss515ndash1215

Alberti G Peretti AV 2002 Fine structure of male genitalsystem and sperm in Solifugae does not support a sister-group relationship with Pseudoscorpiones (Arachnida)Journal of Arachnology 30 268ndash274

Alexander AJ 1967 Problems of limb extension in the scor-pion Opisthophthalmus latimanus Koch Transactions of theRoyal Society of South Africa 37 165ndash181

Anderson DT 1973 Embryology and phylogeny in annelidsand arthropods Oxford Pergamon Press

Anderson LI Selden PA 1997 Opisthosomal fusion andphylogeny of Palaeozoic Xiphosura Lethaia 30 19ndash31

Beall BS Labandeira CC 1990 Macroevolutionary patternsof the Chelicerata and Tracheata In Culver SJ ed Arthro-pod paleobiology Short Courses in Paleontology No 3 Lon-don The Paleontological Society 257ndash284

Bernard HM 1896 The comparative morphology of the Gale-odidae Transactions of the Linnean Society of London Sec-ond Series 6 305ndash417

Boumlrner C 1902a Arachnologische Studien (II und III) Zool-ogischer Anzeiger 25 433ndash466

Boumlrner C 1902b Arachnologische Studien IV Die Genitalor-gane der Pedipalpen Zoologischer Anzeiger 26 81ndash92

Boumlrner C 1904 Beitraumlge zur Morphologie der Arthropoden IEin Beitrag zur Kenntnis der Pedipalpen Zoologica (Stut-tgart) 42 1ndash174

Bowler PJ 1996 Lifersquos splendid drama Chicago Universityof Chicago Press

Braddy SJ Alridge RJ Gabbott SE Theron JN 1999Lamellate book-gills in a late Ordovician eurypterid from theSoom Shale South Africa support for a eurypterid-scorpionclade Lethaia 32 72ndash74

Braddy SJ Dunlop JA 1997 The functional morphology ofmating in the Silurian eurypterid Baltoeurypterus tet-ragonophthalmus (Fischer 1839) Zoological Journal of theLinnean Society 121 435ndash461

Brauer A 1895 Beitrage zur Kenntnis der Entwicklungsge-schichte des Skorpiones II Zeitschrift fuumlr WissenschaftlicheZoologie 59 351ndash433

240 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bremer K 1994 Branch support and tree stability Cladistics10 295ndash304

Bristowe WS 1932 The liphistiid spiders Proceedings of theZoological Society of London Part 4 1016ndash1057

Butt AG Taylor HH 1991 The function of spider coxalorgans effects of feeding and salt-loading on Porrhotheleantipodiana (Mygalomorpha Dipluridae) Journal of Exper-imental Biology 158 439ndash461

Buxton BH 1913 Coxal glands of arachnids ZoologischeJahrbucher (Suppl XIV) 16ndash50

Buxton BH 1917 Notes on the anatomy of arachnids Jour-nal of Morphology 29 1ndash31

Camin JH Clark GM Bourdeau FG 1956 The palpallsquotined setarsquo in the Mesostigmata a homologue of the palpalclaw in the Onychopalpida (Acarina) In Becker EC ed Pro-ceedings of the Tenth International Congress of EntomologyOttawa Mortimer 903ndash908

Chamberlin JC 1931 The arachnid order ChelonethidaStanford University Stanford University Press

Chapman RF 1998 The insects structure and function 4thedn Cambridge Cambridge University Press

Clarke KU 1979 Visceral anatomy and arthropod phylogenyIn Gupta AP ed Arthropod phylogeny New York Van Nos-trand Reinhold 467ndash549

Clarke J 1984 On the relationship between structure andfunction in the leg joints of Heteropoda venatoria (L) (Ara-neae Eusparassidae) Bulletin of the British ArachnologicalSociety 6 181ndash192

Clarke J 1986 The comparative functional morphology of theleg joints and muscles of five spiders Bulletin of the BritishArachnological Society 7 37ndash47

Clarke JM Ruedemann R 1912 The Eurypterida of NewYork New York State Museum Memoir 14 (1ndash2) 1ndash628

Coddington JA 2005 Phylogeny and classification of spi-ders In Ubick D Paquin P Cushing PE Roth V eds Spi-ders of North America an identification manual AmericanArachnological Society 18ndash24

Coddington JA Giribet G Harvey MS Prendini LWalter DE 2004 Arachnida In Cracraft J Donoghue MJeds Assembling the tree of life Oxford Oxford UniversityPress 269ndash318

Coddington JA Levi HW 1991 Systematics and evolutionof spiders (Araneae) Annual Review of Ecology and System-atics 22 565ndash592

Coineau Y Van der Hammen L 1979 The postembryonicdevelopment of Opilioacarida with notes on new taxa and ona general model for the evolution In Piffle E ed Proceed-ings of the 4th International Congress of Acarology Budap-est Akadeacutemiai Kiadoacute 437ndash441

Cokendolpher JC Reddell JR 1992 Revision of the Pro-toschizomidae (Arachnida Schizomida) with notes on thephylogeny of the order Texas Memorial Museum Speleolog-ical Monographs 3 31ndash74

Condeacute B 1991a Le dimorphisme sexuel des Palpigrades Bul-letin de la Socieacuteteacute Neuchacircteloise des Sciences Naturelles 11667ndash73

Condeacute B 1991b Lrsquoaire genitale male des Palpigrades Bulle-tin de la Socieacuteteacute Europeenne drsquoArachnologie 1990 64ndash69

Dearden PK Donly C Grbicccc M 2002 Expression of pair-rule gene homologues in a chelicerate early patterning ofthe two-spotted spider mite Tetranychus urticae Develop-ment 129 5461ndash5472

Delle Cave L 1975 The ldquocleaning brushrdquo in Damon diademaSimon 1876 and Phrynichus jayakari Pocock 1894 (Ambly-pygi Arachnida) Monitore Zoologico Italiano (NS) 9 175ndash184

Dunlop JA 1994 Filtration mechanisms in the mouthparts oftetrapulmonate arachnids (Trigonotarbida Araneae Ambly-pygi Uropygi Schizomida) Bulletin of the British Arachno-logical Society 9 267ndash273

Dunlop JA 1996 Evidence for a sister group relationshipbetween Ricinulei and Trigonotarbida Bulletin of the BritishArachnological Society 10 193ndash204

Dunlop JA 1997 Palaeozoic arachnids and their significancefor arachnid phylogeny In Zabka M ed Proceedings of the16th European colloquium of arachnology Siedlce PolandWyzsza Szkola Rolnicko-Pedagogiczna 65ndash82

Dunlop JA 1998 The origins of tetrapulmonate book lungsand their significance for chelicerate phylogeny In SeldenPA ed Proceedings of the 17th European Colloquium ofArachnology Edinburgh 1997 Burnham Beeches BritishArachnological Society 9ndash16

Dunlop JA 1999 A redescription of the Carboniferous arach-nid Plesiosiro madeleyi Pocock 1911 (Arachnida Hapto-poda) Transactions of the Royal Society of Edinburgh EarthSciences 90 29ndash47

Dunlop JA 2000 The epistomo-labral plate and lateral lips insolifuges pseudoscorpions and mites Ekoloacutegia (Bratislava)19 67ndash78

Dunlop JA 2002a Arthropods from the Lower Devonian Sev-ernaya Zemlya Formation of October Revolution Island(Russia) Geodiversitas 24 349ndash379

Dunlop JA 2002b Character states and evolution of the che-licerate claws In Toft S Scharff N eds European Arach-nology 2000 Aarhus Aarhus University Press 345ndash354

Dunlop JA 2002c Phylogeny of Chelicerata In LlorenteBousquets J Morrone JJ eds Biodiversidad Taxonomiacutea yBiogeografiacutea de Arthroacutepodos de Meacutexico Hacia Una Siacutentesisde Su Conocimiento Vol III Prensas de Ciencias Univer-sidad Nacional Autonoma de Meacutexico 117ndash141

Dunlop JA Anderson LI Braddy SJ 2004 A redescriptionof Chasmataspis laurencii Caster amp Brooks 1956 (Chelicer-ata Chasmataspidida) from the Middle Ordovician of Ten-nessee USA with remarks on chasmataspid phylogenyTransactions of the Royal Society of Edinburgh Earth Sci-ences 94 207ndash225

Dunlop JA Braddy SJ 1997 Slit-like structures on theprosomal appendages of the eurypterid BaltoeurypterusNeues Jahrbuch fuumlr Geologie und Palaumlontologie-Monatshefte1997 31ndash38

Dunlop JA Braddy SJ 2001 Scorpions and their sister-group relationships In Fet V Selden PA eds Scorpions2001 in Memoriam Gary A Polis London British Arachno-logical Society 1ndash24

Dunlop JA Horrocks CA 19951996 A new Upper Carbon-iferous whip scorpion (Arachnida Uropygi Thelyphonida)

PHYLOGENY OF THE ARACHNID ORDERS 241

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

with a revision of the British Carboniferous Uropygi Zoolo-gischer Anzeiger 234 293ndash306

Dunlop JA Poschmann M Anderson L 2001 On theEmsian (early Devonian) arthropods of the Rhenish SlateMountains 3 The chasmataspidid Diploaspis Palaumlon-tolgische Zeitschrift 75 253ndash269

Dunlop JA Webster M 1999 Fossil evidence terrestrializa-tion and arachnid phylogeny Journal of Arachnology 27 86ndash93

Edgecombe GD Wilson GDF Colgan DJ Gray MR Cas-sis G 2000 Arthropod cladistics combined analysis of his-tone H3 and U2 snRNA sequences and morphologyCladistics 16 155ndash203

Evans GO 1992 Principles of acarology Wallingford UKCABI

Fahrenbach WH 1999 Merostomata In Harrison FW Foe-lix RF eds Microscopic anatomy of invertebrates Vol 8AChelicerate arthropods New York Wiley-Liss 21ndash115

Farley RD 1999 Scorpiones In Harrison FW Foelix RF edsMicroscopic anatomy of invertebrates Vol 8A Cheliceratearthropods New York Wiley-Liss 117ndash222

Farley RD 2005 Developmental changes in the embryo pro-nymph and first molt of the scorpion Centruroides vittatus(Scorpiones Buthidae) Journal of Morphology 265 1ndash27

Fayers SR Dunlop JA Trewin NH 2004 A new earlyDevonian trigonotarbid arachnid from the Windyfield ChertRhynie Scotland Journal of Systematic Palaeontology 2269ndash284

Felsenstein J 1985 Confidence limits on phylogenies anapproach using the bootstrap Evolution 39 783ndash791

Fet V Sissom WD Lowe G Braunwalder ME 2000 Cat-alog of the scorpions of the world (1758ndash1998) New YorkNew York Entomological Society

Firstman B 1954 The central nervous system musculatureand segmentation of the cephalothorax of a tarantula (Eu-rypelma californicum Ausserer) Microentomology 87 1ndash40

Firstman B 1973 The relationship of the chelicerate arterialsystem to the evolution of the endosternite Journal ofArachnology 1 1ndash54

Foelix RF 1985 Mechano- and chemoreceptive sensilla InBarth FG ed Neurobiology of arachnids Berlin Spinger-Verlag 118ndash137

Foelix RF 1996 Biology of spiders 2nd edn Oxford OxfordUniversity Press

Giribet G Edgecombe GD Wheeler WC Babbit C 2002Phylogeny and systematic position of Opiliones a combinedanalysis of chelicerate relationships using morphologicaland molecular data Cladistics 18 5ndash70

Goloboff P 1993 Estimating character weights during treesearch Cladistics 9 83ndash91

Goloboff PA Farris JS Nixon K 2000 TNT Tree AnalysisUsing New Technologies Version 1 Program and documen-tation httpwwwzmucdkpublicphylogenytnt

Grainge CA Pearson RG 1966 Cuticular structure in thePhalangida Nature 211 866

Grandjean F 1938 Observations sur les Bdelles (Acariens)Annales de la Socieacuteteacute Entomologique de France Paris 107 1ndash24

Hansen HJ Soslashrensen W 1904 On two orders of ArachnidaCambridge Cambridge University Press

Hansen HJ Soslashrensen W 1905 The Tartarides a tribe of theorder Pedipalpi Arkiv fuumlr Zoologi 2 1ndash78

Hara MR Gnaspini P 2003 Comparative study of the defen-sive behavior and morphology of the gland opening areaamong harvestmen (Arachnida Opiliones Gonyleptidae)under a phylogenetic perspective Arthropod Structure andDevelopment 32 257ndash275

Harvey MS 1992 The phylogeny and classification of thePseudoscorpionida (Chelicerata Arachnida) InvertebrateTaxonomy 6 1373ndash1435

Harvey MS 2002 The neglected cousins what do we knowabout the smaller arachnid orders Journal of Arachnology30 357ndash372

Haupt J Song D 1996 Revision of East Asian whip scorpions(Arachnida Uropygi Theyphonida) I China and JapanArthropoda Selecta 5 43ndash52

Hjelle JT 1990 Anatomy and morphology In Polis GA edThe biology of scorpions Stanford University Stanford Uni-versity Press 9ndash63

Janczyk FSW 1956 Anatomie von Siro duricorius Joseph imVergleich mit anderen Opilioniden Sitzungsberichte Oster-reichischen Akademie der Wissenschaften Mathematisch-Natruwissenschaftliche Klasse 165 474ndash522

Jenner RA 2001 Bilaterian phylogeny and uncritical recy-cling of morphological data sets Systematic Biology 50 730ndash742

Jeram AJ 1998 Phylogeny classification and evolution of Sil-urian and Devonian scorpions In Selden PA ed Proceed-ings of the 17th European Colloquium of ArachnologyEdinburgh 1997 Burnham Beeches UK British Arachno-logical Society 15ndash31

Karaman IM 2005 Evidence of spermatophores in Cyphoph-thalmi (Arachnida Opiliones) Revue Suisse de Zoologie 1123ndash11

Kaumlstner A 1935 Die Funktion der sogenannten sympathis-chen Ganglien und die Exkretion bei den PhalangiidenZoologischer Anzeiger 109 273ndash288

Kierans JE Robbins RG 1999 A world checklist of generasubgenera and species of ticks (Acari Ixodida) publishedfrom 1973 to 1997 Journal of Vector Ecology 24 115ndash129

Kjellesvig-Waering EN 1986 A restudy of the fossil Scorpi-onida of the world Palaeontographica Americana 55 1ndash287

Klompen JSH 2000 Prelarva and larva of Opilioacarus (Neo-carus) texanus (Chamberlin and Mulaik) (Acari Opilioacar-ida) with notes on the patterns of setation and lyrifissuresJournal of Natural History 34 1977ndash1992

Kraus O 1998 Elucidating the historical process of phylog-eny phylogenetic systematics versus cladistic techniquesIn Selden PA ed Proceedings of the 17th European Collo-quium of Arachnology Edinburgh 1997 Burnham BeechesUK British Arachnological Society 1ndash7

Kraus O Kraus M 1993 Divergent transformation of cheli-cerae and original arrangement of eyes in spiders (Arach-nida Araneae) Memoirs of the Queensland Museum 33579ndash584

Lankester ER Benham WBS Beck EJ 1885 On the mus-

242 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

cular and endoskeletal systems of Limulus and Scorpio withsome notes on the anatomy and generic characters of scor-pions Transactions of the Zoological Society of London 11311ndash384

Leclerc P 1989 Considerations paleobiogeographiques a pro-pos la decouverte en Thailande drsquoopilioacariens nouveaux(Acari ndash Notostigmata) Compte Rendu des Seances de laSociete de Biogeographie 65 162ndash174

Legg G 1976 The external morphology of a new species ofricinuleid (Arachnida) from Sierra Leone Zoological Journalof the Linnean Society 59 1ndash58

Lindquist EE 1984 Current theories on the evolution ofmajor groups of Acari and on their relationships with othergroups of Arachnida with consequent implications for theirclassificaton In Griffiths DA Bowman CE eds AcarologyVI Vol 1 New York John Wiley amp Sons 28ndash62

Manning PL Dunlop JA 1995 The respiratory organs ofeurypterids Palaeontology 38 287ndash297

Manton SM 1958 Hydrostatic pressure and leg exten-sion in arthropods with special reference to arachnidsAnnals and Magazine of Natural History Series 13 1161ndash182

Manton SM 1964 Mandibular mechanisms and the evolutionof arthropods Philosophical Transactions of the Royal Soci-ety of London B 7 1ndash183

Maraun M Heethoff M Schneider K Scheu S WeigmannG Cianciolo J Thomas RH Norton RA 2004 Molecularphylogeny of oribatid mites (Oribatida Acari) evidence formultiple radiations of parthenogenetic lineages Experimen-tal and Applied Acarology 33 183ndash201

Marples BJ 1967 The spinnerets and epiandrous glands ofspiders Journal of the Linnean Society (Zoology) 46 209ndash222

Marples BJ 1983 Observations on the structure of the fore-gut of spiders Bulletin of the British Arachnological Society8 46ndash52

Millot J 1933 Notes compleacutementaires sur lrsquoanatomie desliphistiides et des hypchilides a propos drsquoun travail reacutecent deA Petrunkevitch Bulletin de la Socieacuteteacute Zoologique de France68 217ndash235

Millot J 1942 Sur lrsquoanatomie et lrsquohistophysiologie deKoenenia mirabilis Grassi (Arachnida Palpigradi) RevueFranccedilaise drsquoEntomologie 9 33ndash51

Millot J 1943 Notes complementaires sur lrsquoanatomielrsquohistologie et la repartition geographique en France deKoenenia mirabilis Grassi Revue Franccedilaise drsquoEntomologie 9127ndash135

Millot J 1945 Lrsquoanatomie interne des Ricinulei (Arachnides)Annales des Sciences Naturelles (Zoologie) Series 11 7 1ndash29

Millot J 1949a Morphologie generale et anatomie interne InGrasseacute P-P ed Traite de zoologie VI Paris Masson et Cie263ndash319

Millot J 1949b Ordre des Amblypyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 563ndash588

Millot J 1949c Ordre des Araneacuteides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 589ndash743

Millot J 1949d Ordre des Palpigrades In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 520ndash532

Millot J 1949e Ordre des Uropyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 533ndash562

Millot J 1949f Ordre des Ricinuleides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 744ndash760

Millot J Vachon M 1949 Ordre Des Solifuges In Grasseacute P-P ed Traite de zoologie VI Paris Masson et Cie 482ndash519

Moore RA Briggs DEG Bartels C 2005 A new specimen ofWeinbergina opitzi (Chelicerata Xiphosura) from the LowerDevonian Huumlnsruck Slate Germany PalaumlontologischeZeitschrift 79 399ndash408

Moritz M 1957 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) unter besonderer Beruumlcksichtigungder ausseren Morphologie der Bildung des Mitteldarmesund der Genitalanlage Zoologische Jahrbuumlcher Abteilungfuumlr Anatomie und Ontogenie der Tiere 76 331ndash370

Moritz M 1959 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) II Die Anlage und Entwicklung derCoxaldruumlse bei Phalangium opilio L Zoologische Jahr-buumlcher Abteilung fuumlr Anatomie und Ontogenie der Tiere 77229ndash240

Nixon K 1999 The parsimony ratchet a new method forrapid parsimony analysis Cladistics 15 407ndash414

Norton RA 1998 Morphological evidence for the evolutionaryorigin of Astigmata (Acari Acariformes) Experimental andApplied Acarology 22 559ndash594

Palmgren P 1978 On the muscular anatomy of spiders ActaZoologica Fennica 155 1ndash41

Patten W 1912 The evolution of the vertebrates and their kinPhiladelphia P Blakistonrsquos Son

Patten W Hazen AP 1900 The development of the coxalgland branchial cartilages and genital ducts of Limuluspolyphemus Journal of Morphology 16 459ndash490

Paulus HF 1979 Eye structure and the monophyly of theArthropoda In Gupta AP ed Arthropod phylogeny NewYork Van Nostrand Reinhold 299ndash383

Petrunkevitch A 1949 A study of Palaeozoic ArachnidaTransactions of the Connecticut Academy of Arts and Sci-ences 37 69ndash315

Pittard K Mitchell RW 1972 Comparative morphology ofthe life stages of Cryptocellus pelaezi (Arachnida Ricinulei)Graduate Studies Texas Tech University 1 1ndash77

Platnick NI 2005 The world spider catalog Version55 American Museum of Natural History httpresearchamnhorgentomologyspiderscatalogindexhtm

Platnick NI Gertsch WJ 1976 The suborders of spiders acladistic analysis (Arachnida Araneae) American MuseumNovitates 2607 1ndash15

Pocock RI 1893 On some points in the morphology of theArachnida (ss) with notes on the classification of the groupAnnals and Magazine of Natural History Series 6 11 1ndash19

Pocock RI 1902 Some points in the morphology and classi-fication of the Opiliones Annals and Magazine of NaturalHistory Series 7 10 504ndash516

Popadicccc A Panganiban G Rusch D Shear WA KaufmanTC 1998 Molecular evidence for the gnathobasic derivationof arthropod mandibles and for the appendicular origin ofthe labrum and other structures Development Genes andEvolution 208 142ndash150

PHYLOGENY OF THE ARACHNID ORDERS 243

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Raw F 1957 Origin of chelicerates Journal of Paleontology31 139ndash192

Reddell JR Cokendolpher JC 1995 Catalogue bibliogra-phy and generic revision of the order Schizomida (Arach-nida) Texas Memorial Museum Speleological Monographs4 1ndash170

Reissland A Goumlrner P 1985 Trichobothria In Barth FGed Neurobiology of arachnids Berlin Springer-Verlag 138ndash161

Roewer C-F 1934 Solifugae Palpigradi Bronns Klassen undOrdnungen ses Tierreich 5 IV 4 1ndash713

Rowland JM Cooke JAL 1973 Systematics of the arachnidorder Uropygida (=Thelyphonida) Journal of Arachnology 155ndash71

Rowland JM Sissom WD 1980 Report on a fossil palpi-grade from the Tertiary of Arizona and a review of themorphology and systematics of the order (ArachnidaPalpigradida) Journal of Arachnology 8 69ndash86

Rucker A 1901 The Texan Koenenia American Naturalist35 615ndash630

Ruhland M Rathmayer W 1978 Die Beinmuskulatur undihre Innervation bei der Vogelspinne Dugesiella henzti (Ch)(Araneae Aviculariidae) Zoomorphologie 89 33ndash46

Saboori A Kamali K 2000 Description of Allothrombiumtriticium adult (Acari Trombidiidae) from Iran Systematicand Applied Acarology 5 207ndash208

Schliwa M 1979 The retina of the phalangid Opilio raven-nae with particular reference to arhabdomeric cells Celland Tissue Research 204 473ndash495

Scholl G 1977 Beitrage zur Embryonalentwicklung von Lim-ulus polyphemus L (Chelicerata Xiphosura) Zoomorpholo-gie 86 99ndash154

Scholtz G Kamenz C 2006 The book lungs of Scorpionesand Tetrapulmonata (Chelicerata Arachnida) evidence forhomology and a single terrestrialization event of a commonarachnid ancestor Zoology 109 2ndash13

Seitz KA 1972 Zur Histologie und Feinstruktur des Herzensund der Haumlmocyten von Cupiennius salei Keys (AraneaeCtenidae) I Herzwandung Bildung und Differenzierungder Haumlmocyten Zoologische Jahrbuumlcher Abteilung fuumlr Anat-omie und Ontogenie der Tiere 89 351ndash384

Sekiguchi K ed 1988 Biology of horseshoe crabs Tokyo Sci-ence House

Selden PA 1981 Functional morphology of the prosoma ofBaltoeurypterus tetragonophthalmus (Fischer) (ChelicerataEurypterida) Transactions of the Royal Society of Edin-burgh Earth Sciences 72 9ndash48

Selden P 1992 Revision of the fossil ricinuleids Transactionsof the Royal Society of Edinburgh Earth Sciences 83 595ndash634

Selden PA 1993 Fossil arachnids ndash recent advances andfuture prospects Memoirs of the Queensland Museum 33389ndash400

Selden PA Dunlop JA 1998 Fossil taxa and relationshipsof chelicerates In Edgecombe GD ed Arthropod fossilsand phylogeny New York Columbia University Press 303ndash331

Selden PA Jeram AJ 1989 Palaeophysiology of terrestrial-

isation in the Chelicerata Transactions of the Royal Societyof Edinburgh Earth Sciences 80 303ndash310

Selden PA Shear WA Bonamo PM 1991 A spider and otherarachnids from the Devonian of New York and reinterpre-tations of Devonian Araneae Palaeontology 34 241ndash281

Sensenig A Shultz JW 2003 Mechanics of cuticular elasticenergy storage in leg joints lacking extensor muscles inarachnids Journal of Experimental Biology 206 771ndash784

Shear WA 1993 New species in the opilionid genus Stylocel-lus from Malaysia Indonesia and the Philippines (OpilionesCyphophthalmi Stylocellidae) Bulletin of the British Arach-nological Society 9 174ndash188

Shear WA Selden PA Rolfe WDI Bonamo PM GriersonJD 1987 New terrestrial arachnids from the Devonian ofGilboa New York (Arachnida Trigonotarbida) AmericanMuseum Novitates 2901 1ndash74

Shultz JW 1987 The origin of the spinning apparatus in spi-ders Biological Reviews of the Cambridge PhilosophicalSociety 62 89ndash113

Shultz JW 1989 Morphology of locomotor appendages inArachnida evolutionary trends and phylogenetic implica-tions Zoological Journal of the Linnean Society 97 1ndash56

Shultz JW 1990 Evolutionary morphology and phylogeny ofArachnida Cladistics 6 1ndash31

Shultz JW 1991 Evolution of locomotion in Arachnida thehydraulic pressure pump of the giant whipscorpion Masti-goproctus giganteus (Uropygi) Journal of Morphology 21013ndash31

Shultz JW 1993 Muscular anatomy of the giant whipscor-pion Mastigoproctus giganteus (Arachnida Uropygi) and itsevolutionary significance Zoological Journal of the LinneanSociety 108 335ndash365

Shultz JW 1998 Phylogeny of Opiliones (Arachnida) anassessment of the lsquoCyphopalpatoresrsquo concept Journal ofArachnology 26 257ndash272

Shultz JW 1999 Muscular anatomy of a whipspider Phrynuslongipes (Amblypygi) and its evolutionary significance Zoo-logical Journal of the Linnean Society 126 81ndash116

Shultz JW 2000 Skeletomuscular anatomy of the harvest-man Leiobunum aldrichi (Weed 1893) (Arachnida Opil-iones Palpatores) and its evolutionary significanceZoological Journal of the Linnean Society 128 401ndash438

Shultz JW 2001 Gross muscular anatomy of Limuluspolyphemus (Chelicerata Xiphosura) and its bearing on evo-lution in the Arachnida Journal of Arachnology 29 283ndash303

Shultz JW 2007 Morphology of the prosomal exoskeleton ofScorpiones (Arachnida) and a new hypothesis for the evolu-tion of cuticular cephalic endoskeletons in arthropodsArthropod Structure amp Development 36 77ndash102

Shultz JW Pinto da Rocha R 2007 Morphology and func-tional anatomy In Pinto da Rocha R Machado G Giribet Geds The harvestmen the biology of Opiliones CambridgeMA Harvard University Press 14ndash61

Shultz JW Regier JC 2001 Phylogenetic analysis of Pha-langida (Arachnida Opiliones) using two nuclear protein-encoding genes supports monophyly of Palpatores Journalof Arachnology 29 189ndash200

Sitnikova LG 1978 The main evolutionary trends of the

244 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Acari and the problem of their monophyletism Entomologis-cheskoe Obozrenie 57 431ndash457

Snodgrass RE 1948 The feeding organs of Arachnidaincluding mites and ticks Smithsonian Miscellaneous Col-lections 110 (10) 1ndash94

Snodgrass RE 1952 A textbook of arthropod anatomy Ith-aca NY Comstock Publishing

Soslashrensen W 1879 Om Bygningen af Gonyleptiderne en Typeaf Arachnidernes Classe Naturhistorisk Tidsskrifter Series3 8 97ndash222

Stockwell SA 1989 Revision of the phylogeny and higherclassification of scorpions (Chelicerata) Unpublished doc-toral dissertation University of California at Berkeley

Stoslashrmer L 1944 On the relationships and phylogeny of fossiland Recent Arachnomorpha Skrifter utgitt av Det NorskeVidenskaps-Akademi i Oslo I 5 1ndash158

Stoslashrmer L 1955 Merostomata In Moore RC ed Treatise oninvertebrate paleontology Part P Arthropoda 2 LawrenceKS University of Kansas Press and Geological Society ofAmerica 4ndash41

Swofford DL 2002 PAUP Phylogenetic analysis using par-simony (and other methods) Version Beta 10 SunderlandMA Sinauer Associates

Talarico G Palacios-Vargas JG Fuentes Silva M AlbertiG 2005 First ultrastructural observations of the tarsal poreorgan of Pseudocellus pearsei and P boneti (ArachnidaRicinulei) Journal of Arachnology 33 604ndash612

Templeton AR 1983 Phylogenetic inference from restrictionendonuclease cleavage site maps with articular reference tothe evolution of humans and the apes Evolution 37 221ndash244

Tetlie OE Braddy SJ 2004 The first Silurian chasmataspidLoganamaraspis dunlopi gen et sp nov (Chelicerata Chas-mataspidida) from Lesmahagow Scotland and its implica-tions for eurypterid phylogeny Transactions of the RoyalSociety of Edinburgh Earth Sciences 94 227ndash234

Thomas RH Telford MJ 1999 Appendage development inembryos of the oribatid mite Archegozetes longisetosus(Acari Oribatei Trhypochthoniidae) Acta Zoologica (Stock-holm) 80 193ndash200

Tollerton VP Jr 1989 Morphology taxonomy and classifica-tion of the order Eurypterida Burmeister 1843 Journal ofPaleontology 63 642ndash657

Vachon M 1949 Ordre des Pseudoscorpions In Grasseacute P-Ped Traite de zoologie VI Paris Masson et Cie 436ndash481

Van der Hammen L 1979 Comparative studies in Chelicer-ata I The Cryptognomae (Ricinulei Architarbi and Anacti-notrichida) Zoologische Verhandelingen 174 1ndash62

Van der Hammen L 1982 Comparative studies in Chelicer-ata II Epimerata (Palpigradi and Actinotrichida) Zoolo-gische Verhandelingen 196 1ndash70

Van der Hammen L 1985 Functional morphology and affin-ities of extant Chelicerata in evolutionary perspectiveTransactions of the Royal Society of Edinburgh 76 137ndash146

Van der Hammen L 1989 An introduction to comparativearachnology Leiden SPB Academic Publishing

Vyas AB 1970 Studies in myology of the scorpion Heter-ometrus fulvipes Koch Unpublished PhD DissertationDepartment of Zoology Gujarat University India

Vyas AB 1974 The cheliceral muscles of the scorpion Heter-ometrus fulvipes Bulletin Southern California Academy ofSciences 73 9ndash14

Walter DE 1988 Predation and mycophagy by endeostigma-tid mites (Acariformes Prostigmata) Experimental andApplied Acarology 4 159ndash166

Walter DE Proctor HC 1998 Feeding behaviour and phy-logeny observations on early derivative Acari Experimentaland Applied Acarology 22 39ndash50

Walter DE Proctor HC 1999 Mites ecology evolution andbehaviour Sydney University of New South Wales

Waterston CD 1979 Problems of functional morphology andclassification in stylonuroid eurypterids (Chelierata Meros-tomata) with observations on the Scottish Silurian Sty-lonuroidea Transactions of the Royal Society of EdinburghEarth Science 70 251ndash322

Weygoldt P 1969 Biology of pseudoscorpions CambridgeMA Harvard University Press

Weygoldt P 1984 Lrsquoautotomie chez les Amblypyges RevueArachnologique 5 321ndash327

Weygoldt P 1996 Evolutionary morphology of whip spiderstowards a phylogenetic system (Chelicerata ArachnidaAmblypygi) Journal of Zoological Systematics and Olution-ary Research 34 185ndash202

Weygoldt P 1998 Evolution and systematics of the Chelicer-ata Experimental and Applied Acarology 22 63ndash79

Weygoldt P 2000 Whip spiders (Chelicerata Amblypygi)their biology morphology and systematics StenstrupDenmark Apollo Books

Weygoldt P Paulus HF 1979 Untersuchungen zur Morphol-ogie Taxonomie und Phylogenie der Chelicerata Zeitschriftfuumlr Zoologische Systematik und Evolutionsforschung 17 85ndash116 177ndash200

Wheeler WC Hayashi CY 1998 The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192

Winkler D 1957 Die Entwicklung der aumlusseren Koumlrperge-stalt bei den Phalangiidae (Opiliones) Mitteilungen aus demZoologischen Museum in Berlin 33 355ndash389

With CJ 1904 The Notostigmata a new suborder of AcariVidenskabelige Meddelelser fra Dansk Naturhistorisk Foren-ingi Kjobenhaus 1904 137ndash192

Yamasaki T Makioka T Saito J 1988 Morphology InSekiguchi K ed Biology of horseshoe crabs Tokyo ScienceHouse 69ndash132

Yoshikura M 1975 Comparative embryology and phylogeny ofArachnida Kumamoto Journal of Science Biology 12 71ndash142

APPENDIX

Each character used in this analysis is numbereddefined and cross-referenced with characters used infour previous studies that are abbreviated in bracketsspecifically WP = Weygoldt amp Paulus (1979)S = Shultz (1990) WH = Wheeler amp Hayashi (1998)GEWB = Giribet et al (2002) Descriptions includecorrections of errors discussions of controversies orambiguities and justifications for state assignmentsto lsquoproblemrsquo taxa

PHYLOGENY OF THE ARACHNID ORDERS 245

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

PROSOMA

1 Dorsal sclerite formed by fusion of the prosomal car-apace the dorsal portion of the first opisthosomalsomite and the dorsomedial (axial) portion of the sec-ond opisthosomal somite 0 absent 1 presentState 1 is present in extant xiphosurids and close fos-sil relatives (Scholl 1977 Anderson amp Selden 1997Shultz 2001) The fossil record suggests that thisstructure originated with the disappearance ofthe already reduced first opisthosomal tergite(= microtergite) and axial portion of the secondopisthosomal tergite (Anderson amp Selden 1997)2 Single dorsal sclerite covering entire dorsal surfaceof body no lines indicating original segmentation 0absent 1 presentState 1 occurs throughout Holothyrida (Acari) (Vander Hammen 1989) and in many Mesostigmata(Alberti amp Coons 1999) including Glyptholaspis (Vander Hammen 1989)3 Anterior end of dorsal prosoma with medianmarginal or submarginal pointed process 0 absent1 presentState 1 occurs throughout Schizomida (Protoschizo-mus Stenochrus Reddell amp Cokendolpher 1995) The-lyphonida (Rowland amp Cooke 1973 Shultz unpublobs Proschizomus Dunlop amp Horrocks 19951996)Amblypygi (Charinus Millot 1949b Phrynus Shultz1999) and Plesiosiro (Dunlop 1999) Citing a figure inRowland amp Cooke (1973) Dunlop amp Horrocks assumedthat the anterior process was limited to lsquohypoctonidrsquothelyphonids and united them with Proschizomus andSchizomida If valid this would make Thelyphonidaparaphyletic However the doubtful monophyly oflsquohypoctonidsrsquo (Haupt amp Song 1996) and widespreadpresence of a submarginal anterior process in Pedi-palpi weaken this proposal (Harvey 2002)4 Ophthalmic ridges pair of longitudinal crests inter-secting or passing near the region of the lateral eyes orcomparable region where lateral eyes are absent 0absent 1 presentState 1 is present throughout Xiphosura (Anderson ampSelden 1997) Similar structures are present in Plesi-osiro (Dunlop 1999) and lsquonon-hypoctonidrsquo Thely-phonida (Mastigoproctus Rowland amp Cooke 1973)Dunlop (2002a) described apparent ophthalmic ridgesin the chasmataspidid Octoberaspis but these werenot observed in Diploaspis (Dunlop et al 2001) eitherdue to absence or inadequate preservation of speci-mens nor in Chasmataspis (Dunlop et al 2004)5 Carapace with demarcations (eg grooves scleritesphragmata) between pro- meso- or metapeltidia 0absent 1 present [S 1 WH 48 GEWB 29]State 1 is present in extant Scorpiones Cyphoph-thalmi (Hansen amp Soslashrensen 1904) and many Palpa-tores (Opiliones) including Caddo and Leiobunum

and most Pseudoscorpiones (Chamberlin 1931) Giri-bet et al (2002) coded this character as absent inCyphophthalmi (Opiliones) but a transverseprocurved groove is present throughout the group(Hansen amp Soslashrensen 1904) and represents the meso-metapeltidial border as indicated by attachment ofthe dorsal endosternal suspensor of postoral somite VIin Chileogovea oedipus and Siro exilis (Shultz unpublobserv) Kjellesvig-Waering (1986) did not explicitlydescribe such grooves in fossil scorpions although heillustrated corresponding structures in PrearcturusArchaeophonus Archaeoctonus Palaeophonus andother Palaeozoic scorpions (Kjellesvig-Waering 1986figs 14A 21E 27D 59) (coded as uncertain here) butnot in Proscorpius (contra Giribet et al 2002) Dis-tinct carapacal sclerites are present throughout Palpi-gradi Schizomida and Solifugae (6)

The condition in Acari is controversial but is codedhere as State 1 Van der Hammen (1989) consideredthe tergal region of the leg-bearing somites (postoralsomites IIIndashVI) or podosoma to be reduced andreplaced by posterior migration of the aspidosoma (ietergal region assumed to be associated with append-ages of the gnathosoma) and anterior migration of theopisthosomal tergal region to form the hysterosomaThe dorsal proterosomal and hysterosomal elementsmeet at a transverse sejugal furrow that continues lat-erally and passes ventrally between the coxae of legs 2and 3 According to this scheme the prosomandashopisthosoma border is expressed as the disjugal fur-row which passes from the ventral posterior margin ofthe podosoma anterodorsally to join the dorsal part ofthe sejugal furrow Based on the arrangement of setaeand slit sensilla (= lyrifissures) Van der Hammen(1989) interpreted the region above coxae 3 and 4 (ieregion C) as a fusion of the dorsal parts of the first twoopisthosomal somites (= postoral somites VII and VIII)in early divergent Anactinotrichida (ie Opilioacari-formes) and Acariformes (eg Alycus) These interpre-tations have been followed by many acarologistsalthough its speculative aspects are widely acknowl-edged (Evans 1992 Alberti amp Coons 1999)

From the standpoint of a general arachnologist theVan der Hammen system seems unnecessarily compli-cated it appears to have been formulated to explainbroad morphological themes in Oribatida but was thenextrapolated to other mites A more conservativelsquoarachnologicalrsquo scheme adopted here equates the pro-podosoma with the pars cephalica (= propeltidium) ofthe arachnid prosoma and region C with the pars tho-racica (= mesopeltidium + metapetidium) In fact theearliest and most recent treatments of Opilioacari-formes (With 1904 Klompen 2000) reached similarconclusions This interpretation is also consistent withevidence from Acariformes For example the regiondorsal to the coxa of legs 3 and 4 in Alycus is as readily

246 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

explained by the persistence of primitive prosomal ter-gal elements as by the supposition that these disap-peared and were replaced by two opisthosomalelements Furthermore several diverse lineages ofbasally divergent Acariformes display features that donot fit readily within the GrandjeanndashVan der Ham-men system but seem to correspond to the cephalicathoracica division of the prosoma They have a sejugalfurrow that may have demarcated the cephalicatho-racica border and a postpodosomal furrow thatappears to represent the prosomandashopisthosoma junc-tion there is no disjugal furrow Examples includepalaeosomate (especially aphelacarid) and pediculoch-elid Oribatida alicorhagiid and mircopsammidlsquoEndeostigmatarsquo and paratydeid Prostigmata (Albertiamp Coons 1999 fig 42) This organization may be sym-plesiomorphic for Acariformes but could be a conver-gence brought about by selection for enhancedflexibility6 Carapace with distinct pro- meso- or metapeltidialsclerites 0 absent 1 present - inapplicable due toabsence of pro- meso- or metapeltidial demarcations(5)State 1 occurs throughout Palpigradi Schizomida andSolifugae It also appears sporadically within Opil-iones (eg Leiobunum flavum Shultz unpubl observ)and Pseudoscorpiones (eg Pseudochiridium) (Cham-berlin 1931)7 Sejugal furrow circumferential zone of body flexi-bility that passes between the coxae of legs 2 and 3 0absent 1 presentState 1 is probably synapomorphic for Acariformes asit occurs in most lsquoendeostigmatidsrsquo including Alycus(Van der Hammen 1989) and is variously developedthroughout Sarcoptiformes (Alberti amp Coons 1999) Itis probably the primitive condition for Prostigmatabut is either absent or weakly expressed in represen-tatives included here See 5 for alternative morpholog-ical interpretations of this body region in Acari8 Prosomal ozopores 0 absent 1 present [WH 46GEWB 12]State 1 occurs throughout Opiliones and Holothyrida(Van der Hammen 1989)9 Carapacal pleural doublure 0 absent 1 present[WP 12 S 2 WH 49 GEWB 27]A carapacal pleural doublure occurs throughoutXiphosura (Stoslashrmer 1944) Eurypterida (Clarke ampRuedemann 1912 Selden 1981) and Chasmataspid-ida (Dunlop et al 2001 2004 Dunlop 2002a) and isabsent in all known Arachnida10 Cardiac lobe a longitudinal axial elevation of thecarapace 0 absent 1 presentState 1 occurs throughout Xiphosura (Anderson ampSelden 1997) and Eurypterida (Clarke amp Ruedemann1912 Selden 1981) Carapaces are not preserved suf-ficiently to determine this condition in the chasmatas-

pidids Diploaspis (Dunlop et al 2001) andOctoberaspis (Dunlop 2002a) but an apparent cardiaclobe occurs in the Chasmataspis (Dunlop et al 2004)A cardiac lobe is absent in Arachnida11 Moveable cucullus 0 absent 1 present [WP 54GEWB 22]State 1 is a synapomorphy of Ricinulei (Selden 1992)12 Medial intercoxal lsquosternalrsquo region 0 all pedalcoxae separated medially 1 anterior pedal coxae abut-ting medially posterior coxae separated 2 anteriorpedal coxae separated medially posterior coxae abut-ting 3 all pedal coxae abutting medially 4 epimeracoxae undifferentiated medially from ventral body wall[S sim3 WH sim50 GEWB sim17]Prosomal sternites appear to be those portions of theventral body wall not occupied by coxae Despite earlyattempts to assess metamerism of the ventralprosoma (protosternum deutosternum tritosternumetc) (Boumlrner 1902a Millot 1949a) no apparent mor-phological features such as borders of sclerites ormuscle attachments reliably demarcate the ventralbody wall of one somite from that of an adjacentsomite Thus coding schemes that focus on describingmetameric components of the sternal region are prob-lematic Shultz (1990) attempted to code the shape ofthe entire intercoxal region and subsequent workershave adopted this approach All pedal coxae are sepa-rated medially in Plesiosiro (Dunlop 1999) Palpigradi(Roewer 1934) Araneae (Millot 1949c) Amblypygi(Weygoldt 2000) and many Acari including Opilioac-ariformes Holothryrida Mesostigmata Ixodida Aly-cus and many Prostigmata (Van der Hammen 1989Evans 1992 Alberti amp Coons 1999) The anteriorcoxae are separated in extant Thelyphonida Schizo-mida (Hansen amp Soslashrensen 1905 Millot 1949e) andRicinulei (Pittard amp Mitchell 1972 Legg 1976) Thecoxae are fused to the ventral body wall in Sarcopti-formes (Van der Hammen 1989 Evans 1992)

Interpreting this character is complicated for theanterior coxae of Scorpiones and Opiliones and allcoxae in Xiphosura and Eurypterida due to the pres-ence of coxapophyses endites andor gnathobaseswhich form components of the preoral chambers inthese groups Thus coxae may abut medially whenthese structures are considered part of the coxae butare separate medially if these structures are ignoredAs coxapophyses etc are here coded as separate char-acters (50ndash53) I have chosen to code this character asif these structures did not exist Consequently pha-langid Opiliones are coded as having all coxae sepa-rated (Pocock 1902 Hansen amp Soslashrensen 1904) Fossilscorpions also show substantial variation in this char-acter All pairs of pedal coxae in Palaeoscorpius abutmedially but also have gnathobases (Jeram 1998)and this condition is coded here as uncertain A ster-num separates all pairs of pedal coxae medially in

PHYLOGENY OF THE ARACHNID ORDERS

247

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Proscorpius

and

Stoermeroscorpio

(State 0) and

Pre-arcturus

approximates modern scorpions (State 1)(Kjellesvig-Waering 1986 Jeram 1998) The coxalbases of Eurypterida and Xiphosura are separated bya small sternite the endostoma (Selden 1981 Shultz2001)

13 Postoral sternapophysis (

=

tritosternum labium)a cuticular evagination of the ventral body wall poste-riorly adjacent to the palpal coxae forming the poste-rior border of the preoral chamber in some taxa ordisplaced posteriorly by fusion of the palpal coxae inothers 0 absent 1 present

[S

sim

10 WH

sim

56 GEWB19

=

33]State 1 is present throughout Palpigradi AraneaeAmblypygi (Snodgrass 1948) Schizomida (Hansen ampSoslashrensen 1905 Van der Hammen 1989) Thely-phonida (Van der Hammen 1989) extant Ricinulei(Pittard amp Mitchell 1972 Legg 1976) Opilioacari-formes and Mesostigmata (Van der Hammen 1989Alberti amp Coons 1999)

Allothyrus constrictus

is theonly holothyrid mite known to have sternapophysesAn apparent sternapophysis has been described in apalaeocharinid trigonotarbid (Dunlop 1994) Giribet

et al

(2002) miscoded Ixodida and Acariformes as hav-ing a lsquolabium (

=

tritosternum)rsquo The lophognath ofpseudoscorpions (Chamberlin 1931) and lsquolabiumrsquo ofSolifugae (Roewer 1934) also appear to be sternapo-physes that have been incorporated into the ros-trosoma (

32

) The labium of phalangid Opiliones(Pocock 1902 Hansen amp Soslashrensen 1904) is a ster-napophysis but is associated with leg 1 rather thepedipalp (Winkler 1957) A small anterior sclerite inthe fossil scorpions

Proscorpius

Waeringoscorpio

and

Labriscorpio

has been interpreted as a lsquolabiumrsquo homol-ogous with the sternapophyses of other arachnids(Weygoldt 1998) despite absence of any indication ofthe relationship of this sclerite to the mouth and pos-sible alternative interpretations (eg

16

) Giribet

et al

coded this sclerite as a sternapophysis but the char-acter is coded here as uncertain in fossil scorpions

14 Channels on the body surface linking openings ofcoxal organs to preoral chamber 0 absent 1 present

Van der Hammen (1989) described bilaterally pairedcuticular tracts (lsquotaenidiarsquo) connecting orifices of coxalorgans to the preoral chamber in some arachnids andor a ventromedian groove (

=

intercoxal or subcapitulargutter) that leads in turn to the preoral chamber inother taxa Taenidia and intercoxal gutters arepresent in Opilioacariformes (Van der Hammen1989) Holothyrida (Van der Hammen 1989) Mesos-tigmata (

Glyptholapis

Van der Hammen 1989)Ricinulei (Pittard amp Mitchell 1972 Van der Hammen1989) Araneae (

Heptathela

Aphonopelma

Hypochi-lus

Shultz unpubl observ also

Porrhothele

(Mygalo-morphae Dipluridae) Butt amp Taylor 1991

Segestria

(Araneomorphae Segestriidae) Van der Hammen

1989) Amblypygi (Van der Hammen 1989) Schizo-mida (

Stenochrus

Shultz unpubl observ unidenti-fied Hubbardiinae Van der Hammen 1989) andThelyphonida (

Mastigoproctus

Shultz unpublobserv also

Tetrabalius

Van der Hammen 1989) Theconduction of large volumes of fluid from orifices of thecoxal organs to the preoral chamber has been observedin the mygalomorph spider

Porrhothele

(Dipluridae)(Butt amp Taylor 1991) and

Mastigoproctus

(Thely-phonida) (Shultz unpubl observ) The podocephaliccanal (

15

) is present in Acariformes (Acari) (Lindquist1984) and appears to have evolved as an invaginatedsupracoxal channel (Grandjean 1938 G Alberti inEvans 1992) Van der Hammen (1982) could not dis-cern with certainty whether a channel exists in

Eukoenenia

(Palpigradi) but none has been reportedby previous authors and they have not been observedby the present author Palpigradi is coded as havingState 0

15 Podocephalic canal cuticular channel andorduct draining multiple glands and opening nearmouthparts 0 absent 1 present - inapplicable due toabsence of channel (14)

State 1 occurs throughout Acariformes (Acari)(Lindquist 1984)

16 Heavily sclerotized suboral sclerite serving inpart as basal pivot point for coxae of appendages ofpostoral somites IIndashIV (

=

arachnid palp and legs 1 and2) 0 absent 1 present

The sclerite forms the posterior wall of the true mouthand is heavily sclerotized in extant scorpions It is notvisible externally due to the tightly fitting coxae of theanterior prosomal appendages (Shultz 2007) It maycorrespond to the so-called labium (Weygoldt 1998) orlabrum (Kjellesvig-Waering 1986) of certain fossilscorpions (see

13

)

17 Genal angles 0 rounded 1 pointed

Anderson amp Selden (1997) originally coded a pointedgenal angle and genal spine as separate charactersbut they are combined here State 1 is present in manyXiphosura (Anderson amp Selden 1997) and Chas-mataspidida (Dunlop

et al

2001 2004 Dunlop2002a)

A

PPENDAGES

OF

POSTORAL

SOMITE

I

CHELICERAE

18 Cheliceral segmentation 0 three articles 1 twoarticles

[WP

sim

25

+

42 WH

sim

15 GEWB

sim

44]State 0 is present in extant Xiphosura Eurypterida(Clarke amp Ruedemann 1912 Selden 1981) Palpi-gradi (Roewer 1934) Opiliones (Hansen amp Soslashrensen1904) Scorpiones (Hjelle 1990) and throughout non-acariform Acari (Evans 1992) State 1 is presentthroughout known Trigonotarbida Araneae Ambly-pygi Thelyphonida Schizomida Pseudoscorpionesand Solifugae Acariform mites have two unambigu-

248 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ous cheliceral articles but may have a reduced thirdarticle (Van der Hammen 1989 Evans 1992) acari-form mites are thus coded as uncertain Chelicerae arenot known or their condition is unclear in several fos-sil taxa including Xiphosura Chasmataspidida (Dun-lop et al 2001 2004 Dunlop 2002a) Riciniulei(Selden 1992) Plesiosiro (Dunlop 1999) Proschizo-mus (Dunlop amp Horrocks 19951996) and the fossilscorpions Praearcturus and Palaeoscorpius (Kjelles-vig-Waering 1986) Kjellesvig-Waering (1986) consis-tently interpreted chelicerae of fossil scorpions ashaving four articles but this is probably a misinter-pretation of the basal article (see also Stockwell1989) The basal article in extant scorpions consists ofa distal collar of cuticle and a large proximal processand this arrangement would probably appear as twoarticles in compressed specimens In any event thereare at least three cheliceral articles in Stoermeroscor-pio and Proscorpius (Kjellesvig-Waering 1986)19 Terminal cheliceral joint 0 laterally placedbicondylar hinge 1 dorsally placed bicondylar hinge2 ventrally placed bicondylar hinge [WP 25 + 42 Ssim11 + 12 WH sim15 GEWB sim44]State 0 is present in Xiphosura (Shultz 2001)extant Scorpiones (Millot amp Vachon 1949) Opil-iones (Hansen amp Soslashrensen 1904) Palpigradi(Boumlrner 1904 Millot 1949d) Ricinulei (Pittard ampMitchell 1972 Legg 1976) and Eurypterida includ-ing Baltoeurypterus (Selden 1981) and Stylonurus(Clarke amp Ruedemann 1912) State 0 is present inthe fossil scorpions Proscorpius and Stoermeroscor-pio but the state of the chelicerae is unclear in theother fossil scorpions represented here (Kjellesvig-Waering 1986) State 1 is present in Trigonotarbida(Shear et al 1987) mygalomorph Araneae Ambly-pygi Schizomida and Theylphonida (Weygoldt ampPaulus 1979) State 2 is present in Pseudoscorpi-ones (Chamberlin 1931) and Solifugae (Roewer1934) and appears to be a groundplan feature for allmajor groups of mites (eg Van der Hammen 1989Alberti amp Coons 1999) The difference between theorthognathy of mygalomorph Araneae and lsquoplagiog-nathyrsquo mesothele (Liphistius Heptathela) and palae-ocribellate Araneae (Hypochilus) (Kraus amp Kraus1993) is small compared with the variation thatoccurs throughout arachnids and plagiognathy iscoded as State 1 here20 Chelicera articulating with carapace at anterolat-eral pivot 0 absent 1 present [S 13 WH 57 GEWB45]State 1 occurs in Solifugae (Roewer 1934) and Pseu-doscorpiones except Chthonioidea and Feaelloidea(Chamberlin 1931) Giribet et al (2002) coded thecharacter as present in Pseudoscorpiones21 Chelicera pivoting on dorsal protuberance of epis-tome 0 absent 1 present

State 1 appears to be a unique synapomorphy of Opil-iones (Shultz 2000)22 Extrinsic cheliceral muscle arising on carapaceand inserting on dorsal margin of nonbasal cheliceralarticle 0 absent 1 present (Note Characters 22 and23 refer to two different muscles)State 1 is known only in extant Scorpiones (LankesterBenham amp Beck 1885 Vyas 1970 1974 Shultz2007) but may occur in acariform mites if their cheli-cerae are composed of three rather than two articles(Evans 1992) (see 18) Acariformes are coded asuncertain23 Extrinsic cheliceral muscle arising on carapaceand inserting on ventral margin of nonbasal cheliceralarticle 0 absent 1 present [GEWB 252] (Note Char-acters 22 and 23 refer to two different muscles)State 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) somemesostigmatid Acari (Evans 1992) and representa-tive Opiliones particularly Leiobunum (Shultz 2000)Chileogovea and Gonyleptes (Shultz unpubl observ)The status of the character in other opilions is not yetknown The character was originally defined as anextrinsic muscle inserting on the second segment ofthree-segmented chelicerae (Shultz 2000 Giribetet al 2002) but it has been redefined here to make itapplicable to arachnids with two-segmented cheli-cerae24 Extrinsic cheliceral muscles attaching to epistome0 absent 1 presentState 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) but mayoccur in Prostigmata if the sigmoid piece to which thecheliceral protractors attach (Evans 1992) is a modi-fication of the epistome25 Lateral tergocheliceral muscle with three heads 0absent 1 present - coded only for extant tetrapulmo-nates and the palpigrade Eukoenenia homology isunclear in other taxa [GEWB 230]State 1 is present in Pedipalpi (Shultz 1999)26 Cheliceral silk glands 0 absent 1 present [GEWB47]State 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931) The silk gland of spider mites(eg Tetranychus) are associated with the pedipalps(Evans 1992)27 Cheliceral venom glands 0 absent 1 present [WP34 WH 40 GEWB 46]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)28 Cheliceral serrula interior and exterior 0 absent1 presentState 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931 Harvey 1992)29 Cheliceral lsquoflagellumrsquo in male 0 absent 1 present[GEWB 48]

PHYLOGENY OF THE ARACHNID ORDERS 249

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITE II LEG 1 IN NONARACHNIDS AND PALP IN ARACHNIDS

30 Palpal coxae fused ventromedially and formingposterior wall of preoral chamber 0 absent 1 present[WP 27 S 18 WH 17 = 62 GEWB 33 + 63]State 1 occurs in Schizomida Thelyphonida Acari andRicinulei31 Gnathosoma 0 absent 1 present - inapplicablepalpal coxae not fused (30) [WP 56 S 19 WH sim63GEWB sim36 + 37]The gnathosoma is a functional complex comprisingan epistome attached to the dorsal surface of mediallyfused palpal coxae (= infra- or subcapitulum) (see 30)the chelicerae and in some taxa a supracheliceral tec-tum These components are tightly integrated andmove in unison relative to the body The gnathosomahas traditionally been considered a feature of AcariHowever Van der Hammen (1989) argued that a gna-thosoma is present in Ricinulei and this interpreta-tion has received wide acceptance (eg Lindquist1984) However it is not clear that the relationship ofthe chelicerae and subcapitulum is as intimate inRicinulei as in Acari and the ricinuleid subcapitulumdoes not appear to be as moveable In fact the lsquosubca-pitulumrsquo in Ricinulei is similar in many respects to thecondition in Schizomida and Thelyphonida (Hansen ampSoslashrensen 1904) which has never been regarded as agnathosoma Consequently I have coded the conditionof Ricinulei as uncertain32 Rostrosoma long narrow subcylindrical epistomeprojecting anteriorly with base fixed to dorsal surfaceof palpal coxae bordered laterally by lobes projectingfrom palpal coxae ventral wall of preoral chamberformed by anterior element of prosoma (sternapophy-sis) 0 absent 1 present [S 15 WH 59 GEWB 20 + 36]State 1 is present in Pseudoscorpiones and Solifugae(Chamberlin 1931 Roewer 1934 Van der Hammen1989) Dunlop (2000) attempted to homologize com-ponents of the rostrosoma and gnathosoma (31) anargument accepted by Giribet et al (2002) Howeverfusion of the palpal coxae (30) apparently evolvedindependently in the two structures Coxal fusion inthe gnathosoma is complete ventrally and excludesthe suboral sternapophysis (13) while fusion isincomplete ventrally in the rostrosoma and incorpo-rates the sternapophysis Unlike the gnathosomathe rostrosoma is largely immobile and cheliceralmovement is not coupled with that of the palpalcoxae33 Rutellacorniculi hypertrophied setae modified asmouthparts located on the anterior processes of the pal-pal coxae 0 absent 1 present [GEWB 39]

State 1 occurs in the anactinotrichid groups Opilioac-ariformes Holothyrida and Mesostigmata but not Ixo-dida (Van der Hammen 1989) It also occurs inSarcoptiformes (Evans 1992) and certain endeostig-matids (eg Alycus Van der Hammen 1989) but not inProstigmata Comparable structures are unknown inother chelicerate groups34 Terminal segments specially modified in adultmale as a clasper used to engage the female 0 absent1 presentState 1 is known only in extant Xiphosura (YamasakiMakioka amp Saito 1988)35 Robust raptorial 0 absent 1 present [S 17 WH61 GEWB sim59 97]State 1 is present in Amblypygi Thelyphonida andSchizomida (Shear et al 1987) and in Laniatores(Opiliones)36 Orientation of robust raptorial appendage 0operating in subtransverse plane 1 operating in sub-vertical plane - inapplicable coded only for PedipalpiState 0 is present in extant Thelyphonida and mostAmblypygi and State 1 is present in Schizomida (Cok-endolpher amp Reddell 1992) and in the basally diver-gent amblypygid Paracharon caecus (Weygoldt 2000)The orientation of the pedipalps in the fossil Proschi-zomus is uncertain (Dunlop amp Horrocks 19951996)37 Extrinsic muscle attaching to epistome 0 absent1 presentState 1 occurs in Chileogovea and Siro (OpilionesCyphophthalmi) and extant Scorpiones (Vyas 1970Shultz 2007)38 Muscle originating and inserting within coxa 0absent 1 present [GEWB sim232]State 1 occurs is in Amblypygi (Phrynus Shultz 1999Phrynichus Boumlrner 1904) Thelyphonida (Mastigo-proctus Shultz 1993 Thelyphonus Boumlrner 1904) andSchizomida (Stenochrus Shultz unpubl observ)There are two intracoxal muscles in amblypygids andthelyphonids the one coded here and another that isserially homologous with a muscle in the pedal coxae(54)39 Tarsus andor tibia with venom glands 0 absent1 present [GEWB 64]State 1 is known only from extant Pseudoscorpi-ones except the superfamilies Chthonioidea andFeaelloidea (Chamberlin 1931 Harvey 1992) Giri-bet et al (2002) coded this character as present forPseudoscorpiones40 Tarsal grooming organ 0 absent 1 present[GEWB 98]State 1 is a unique synapomorphy of Amblypygi (DelleCave 1975 Weygoldt 2000)41 Scorpionoid chela a large well-developed chelaformed by tibia (manus + fixed finger) and tarsus(moveable finger) 0 absent 1 present [S 16 WH 60GEWB 62]

250 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs throughout Scorpiones andPseudoscorpiones42 Modified in male as copulatory organ 0 absent 1present [WP 35 WH 39 GEWB 67]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)43 Apotele 0 apparently absent not differentiatedexternally from penultimate article 1 present [GEWB69]The apotele exists as a structure (ie dactyl clawempodium) differentiated from the tibiotarsus tarsusor telotarsus and is typically associated with a pair ofantagonistic muscles The apotele occurs as a distinctstructure in Xiphosura (Snodgrass 1948) Eurypter-ida (Clarke amp Ruedemann 1912) Schizomida (Coken-dolpher amp Reddell 1992) Araneae (Foelix 1996)early divergent Amblypygi (Charinus Weygoldt 2000not Phrynus Shultz 1999) Anactinotrichida (CaminClark amp Bourdeau 1956 Lindquist 1984) Ricinulei(Pittard amp Mitchell 1972 Legg 1976) and Solifugae(Roewer 1934) Giribet et al (2002) coded Solifugae asnot having a palpal apotele but it is present as anadhesive organ (45) Dunlop (1999) reconstructedHaptopoda as having a palpal claw but notes in hisdescription that the distal ends were not preserved inthe known specimens44 Apotele (claw) position 0 terminal 1 subtermi-nal - inapplicable coded only for taxa with a distinctapotele (43)State 1 occurs in Parasitiformes (Acari) (Evans 1992)45 Terminal adhesive organ 0 absent 1 present -inapplicable due to lack of an apotele (43) [GEWB 66]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITES IIIndashVI LEGS 2ndash5 IN NON-ARACHNIDS AND LEGS 1ndash4 IN ARACHNIDS

46 Appendage III (= arachnid leg 1) extremely elon-gate antenniform 0 absent 1 present [WP sim24 + 32S 20 WH 14 GEWB sim84]State 1 occurs in Amblypygi Thelyphonida Schizo-mida (Shear et al 1987) and Opilioacariformes (Vander Hammen 1989) State 1 is approximated in Solif-ugae Palpigradi (Roewer 1934) and Haptopoda(Dunlop 1999) and it is coded as uncertain in thesetaxa47 Appendage V (= arachnid leg 3) of male specializedfor sperm transfer 0 absent 1 present [WP 55 GEWB94]State 1 is known only from extant Ricinulei48 Appendages V and VI (= arachnid legs 3 and 4)with femur shorter than patella and with principal siteof flexionextension at patella-tibia joint (lsquoapatellatersquocondition sensu Van der Hammen 1989) 0 absent 1present [S 25 GEWB 86]

State 1 occurs in Pseudoscorpiones and Solifugae(Shultz 1989)49 Appendage VI (= arachnid leg 4) with terminusmodified as flattened blade 0 absent 1 presentState 1 occurs in some Chasmataspidida (DiploapisDunlop et al 2001 Octoberaspis Dunlop 2002a)uncertain in Chasmataspis Dunlop et al 2004 State0 in Loganamaraspis Tetlie amp Braddy 2004) andEurypterida except stylonuroids (Clarke amp Ruede-mann 1912 Waterston 1979 Tollerton 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI COXA AND BODYndashCOXA JOINT

50 Appendage III (= arachnid leg 1) with coxapophy-sis forming floor or wall of preoral chamber 0 absent1 present [S sim14 WH sim58]State 1 occurs in Opiliones (Hansen amp Soslashrensen 1904)and higher scorpions (Jeram 1998) A lobe is presenton the coxa of leg 1 in the fossil scorpion Prearcturusand perhaps Stoermeroscorpio and is coded here as acoxapophysis Coxapophysis of leg 1 together with theepistome and coxapophysis of the palp form a uniquepreoral chamber the stomotheca Weygoldt (1998) andDunlop amp Braddy (2001) reject the homology of cox-apophyses and stomothecae in Opiliones and Scorpi-ones because these structures are not apparent infossils of those scorpions thought to have been aquaticHowever the coxapophyses in Opiliones are formedlargely from soft cuticular lsquolipsrsquo and similar structuresin fossil scorpions would probably not have beenpreserved51 Appendage IV (= arachnid leg 2) with coxapophy-sis 0 absent 1 present [S sim14 WH sim58 GEWB 75]State 1 occurs in extant Scorpiones and most Opil-iones except Dyspnoi (Shultz 1998)52 At least one pair of coxal gnathobases on append-ages III-VI (= arachnid legs 1ndash4) 0 absent 1 present[S sim23 WH sim66 GEWB 78]State 1 occurs in all appropriately preserved Euryp-terida including Baltoeurypterus (Selden 1981) andStylonurus (Clarke amp Ruedemann 1912) extantXiphosura on appendages IIndashVI and apparentlyProscorpius on appendage III (= arachnid leg 1)(Kjellesvig-Waering 1986 Jeram 1998 contra Giribetet al 2002) Fossilized appendages with gnathobaseshave been found in association with Chasmataspis(Chasmataspidida) suggesting that they may havehad State 1 (Dunlop et al 2004)53 Coxae of appendages IIIndashV (= arachnid legs 1ndash3)with jointed moveable endites 0 absent 1 present [Ssim23]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) and in adequately preserved eurypterids(Clarke amp Ruedemann 1912) (BaltoeurypterusSelden 1981)

PHYLOGENY OF THE ARACHNID ORDERS 251

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

54 Intracoxal muscle a muscle arising on anteriorwall of coxa and inserting on posterior wall 0 absent1 present [GEWB 237]State 1 occurs in Thelyphonida (MastigoproctusShultz 1993) and Amblypygi (Phryus Shultz 1999)It is unclear whether the muscle operating the move-able endite (53) in extant Xiphosura (Manton 1964Shultz 2001) is homologous with this muscle and iscoded here as uncertain It is not present in Steno-chrus (Schizomida) (Shultz unpubl observ)55 Coxae of appendages IIndashVI (= arachnid palp andlegs) with dorsal articulation with carapace 0 absent1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)56 Flabellum (exite) on coxa of appendage VI(= arachnid leg 4) 0 absent 1 present [GEWB sim110]State 1 occurs in extant Xiphosura (Yamasaki et al1988) An apparent exite occurs on appendages asso-ciated with Chasmataspis (Chasmataspidida) (Dunlopet al 2004)57 Insertion process of anteromedial tergo-coxal mus-cle 0 weakly developed 1 well to extremely well devel-oped [GEWB 238]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Boumlrner 1904 Shultz 1999)58 Musculi laterales enlarged lateral tergocoxalmuscle with attachment shifted from coxa to adjacentpleural membrane 0 absent 1 present [S 22 WH 65GEWB 242]State 1 occurs in Araneae and Thelyphonida (Shultz1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI TROCHANTER AND COXAndashTROCHANTER JOINT

59 Coxandashtrochanter joint with complex posteriorarticulation composed of two articulating sclerites 0absent 1 present [S 24 WH 67 GEWB 85]State 1 occurs in Araneae Amblypygi Thelyphonidaand Schizomida (Shultz 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI FEMUR AND TROCHANTERndashFEMUR JOINT

60 Depressor muscle (or homologue) of trochanter-femur joint 0 absent 1 present [WH 42 GEWB 220]State 0 appears to be a unique synapomorphy of Ara-neae (Liphistius Shultz 1989 Aphonopelma Ruh-land amp Rathmayer 1978 Hypochilus present studyother examples Clarke 1984 1986)61 Trochanterndashfemur joint with dorsal hinge or pivotoperated by flexor muscles only 0 absent 1 present[WH sim44 GEWB sim91]State 1 occurs in Palpigradi Other chelicerates have abicondylar articulation and are typically equipped

with antagonistic muscles (Van der Hammen 1985Shultz 1989) but see 60 for an exception62 Superior trochanterndashfemur muscle (or homologue)originating broadly in femur inserting on distal mar-gin of trochanter 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 19892001)63 Basifemurndashtelofemur joint of appendages III andIV (= arachnid legs 1 and 2) in adult 0 absent 1present [S 25 WH 68 GEWB 86]The basifemur and telofemur apparently develop aslsquosisterrsquo articles through division of a parental articlethe femur (eg opilioacariform endeostigmatidpalaeosomatid mites) (Coineau amp Van der Hammen1979 Evans 1992) A single animal can have all legswith divided or undivided femora or have some com-bination of the two typically with the more posteriorlegs having the divided femora Thus anterior legsoften tend to be paedomorphic with respect toposterior legs Most chelicerates have undivided fem-ora on some or all legs but muscles homologous withthose of the basifemurndashtelofemur joint may stilldevelop (66) (Shultz 1989) suggesting that the devel-opment of cuticular and muscular components of ajoint have a degree of developmental and evolutionaryindependence

Undivided femora occur in most extant euchelicer-ates including Xiphosura Palpigradi AraneaeAmblypygi Thelyphonida Schizomida ScorpionesOpiliones Pseudoscorpiones higher Sarcoptiformes(Archegozetes Van der Hammen 1989) many Prostig-mata (Tetranychus Evans 1992) and throughout Par-asitiformes Divided femora occur in all legs in manyearly divergent Acariformes including many Endeo-stigmata basally divergent Sarcoptiformes (egPalaeacarus) and Prostigmata (eg AllothrombiumMicrocaeculus) (Van der Hammen 1989 Evans1992) Only leg 4 has basi- and telofemora in Alycus(Van der Hammen 1989) Basi- and telofemora occuron legs 3 and 4 in Solifugae Ricinulei Opilioacari-formes and Eurypterida (Clarke amp Ruedemann 1912Selden 1981 Shultz 1989)

Giribet et al (2002) coded the lsquoannulusrsquo of certainTrigonotarbida (Shear et al 1987) as a basifemur buta similar non-muscularized annulus is present in theposterior legs of extant Thelyphonida In both casesthe condyles of the trochanterndashannulus and annulusndashfemur joints are aligned in such away that they per-form essentially identical movements and appear notto enhance either the range or degree of leg move-ment [This contrasts with the condition in the lasttwo pairs of legs in Baltoeurypterus where condyles ofthe apparent trochanter-basifemur and basifemurndashtelofemur joints have different arrangements (Selden1981)] The annulus in Thelyphonida appears to func-tion in maintaining constant volume of the tro-

252 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

chanterndashfemur joint by preventing ballooning of thearthrodial membrane thereby minimizing the effect ofinternal fluid pressure on movement64 Basifemurndashtelofemur joint of appendage V(= arachnid leg 3) in adult 0 absent 1 present See 6365 Basifemurndashtelofemur joint of appendage VI(= arachnid leg 4) in adult 0 absent 1 present See6366 Cuticular differentiation of basifemur-telofemurjoint absent but muscles present 0 absent 1 present- inapplicable due to presence of joint (63ndash65)State 1 occurs in appendages of postoral somites IIIndashIV (= arachnid legs 1ndash4) in Xiphosura AraneaeAmblypygi Thelyphonida Schizomida Scorpionesand Pseudoscorpiones (Shultz 1989)67 Circumfemoral ring 0 absent 1 presentThe femora of one or more pairs of legs in anactinot-richid Acari have a basal groove or ring associatedwith slit sensilla Acarologists term that portionproximal to the ring the lsquobasifemurrsquo and that portiondistal to the ring the lsquotelofemurrsquo and they refer tothe basifemur as the lsquosecond trochanterrsquo and thetelofemur as the lsquofemurrsquo Despite the confusingterminology the basifemur discussed in 63ndash65 is nothomologous with the lsquobasifemurrsquo of opilioacariformand parasitiform Acari as both a true basifemurndashtelofemur joint and a circumfemoral ring are presentin legs 3 and 4 of Opilioacariformes (Van der Ham-men 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI PATELLA AND FEMURndashPATELLA JOINT

68 Patella of appendage of postoral somite III(= arachnid leg 1) proportionally much longer thanthose of more posterior appendages 0 absent 1presentState 1 occurs in Thelyphonida Schizomida (Hansenamp Soslashrensen 1905 Shultz 1989) and Opilioacari-formes (Van der Hammen 1989)69 Femurndashpatella joint 0 monocondylar several axesof movement and multifunctional muscles 1 bicondy-lar hinge one axis of movement and antagonistic mus-cles 2 hinge one axis of movement flexor muscleswithout muscular antagonists [S sim26 WH sim69 GEWBsim87]State 0 occurs in Solifugae State 1 occurs in Phalang-ida (Opiliones) Scorpiones and PseudoscorpionesState 2 occurs in the remaining taxa (Shultz 1989)70 Patellar plagula 0 absent or with simple medianattachment 1 symmetrical Y-shaped with long prox-imal stem 2 symmetrical U-shaped (= arcuate scler-ite) 3 asymmetrical attaching to patella only atanterior margin [S 27 WH sim70 GEWB 234 = 243]State 1 occurs in extant Xiphosura (Shultz 1989)State 2 occurs in Araneae (Manton 1958 Ruhland amp

Rathmayer 1978 Clarke 1986 Shultz 1989) State 3occurs in Amblypygi Thelyphonida and Schizomida(Shultz 1989) Giribet et al (2002) included this char-acter twice

APPENDAGES OF POSTORAL SOMITES IIIndashVI TIBIA AND PATELLAndashTIBIA JOINT

71 Tibiae divided by one or more joints 0 absent 1presentState 1 occurs in Amblypygi (Weygoldt 2000)72 Patellandashtibia joint 0 monocondylar with or with-out CZY (73) 1 bicondylar hinge one axis of move-ment antagonistic muscles 2 hinge one axis ofmovement muscles without muscular antagonists [Ssim31 WH 74 GEWB sim92] (Shultz 1989 Selden Shearamp Bonamo 1991)73 Patellandashtibia joint with posterior compression zone(lsquoCZYrsquo) 0 absent 1 present - inapplicable due toabsence of monocondylar articulation [GEWB 99]State 1 is a unique synapomorphy of Araneae (Seldenet al 1991)74 Patellandashtibia joint of appendages IIIndashVI(= arachnid legs 1ndash4) with deep-set monocondylar pivotbordered by a pair of tibial processes to which extensormuscles attach 0 absent 1 present - inapplicabledue to absence of monocondylar joint (72)State 1 occurs in extant Xiphosura (Snodgrass 1952Shultz 1989) although a similar arrangement ispresent in the appendage III (= arachnid leg 1) in The-lyphonida (Shultz 1993)75 Patellandashtibia joint largely immobile specializedfor autotomy 0 absent 1 presentState 1 occurs throughout Amblypygi (Weygoldt 19842000)76 Anterior femurndashtibia or femoropatellandashtibia(transpatellar) muscle 0 absent 1 present [S sim30GEWB 222] (Shultz 1989)77 Proximal attachment of posterior femurndashtibia orfemoropatellarndashtibia (transpatellar) muscle 0 muscleabsent 1 dorsoposterior surface of femur andor pos-terior surface of patella 2 distal process of femur mus-cle acting as extensor of femur-patella joint [S sim28 WHsim71 + 72 GEWB 235 = 244 245] (Shultz 1989)78 Distal attachment of posterior femurndashtibia (trans-patellar) muscle 0 posterior 1 dorsal acting as exten-sor of patellandashtibia joint - inapplicable due to absenceof muscle (77) [S 29 GEWB 221] (Shultz 1989)79 Anterior patellandashtibia muscle 0 absent 1 present[S sim32 WH sim73 GEWB 246] (Shultz 1989)80 Posterior patellandashtibia muscle 0 absent 1 present[S 33 WH 76 GEWB 248] (Shultz 1989)81 Patellandashtibia joint spanned by elastic (lsquospringlikersquo)sclerite 0 absent 1 present [GEWB 107]State 1 is a unique synapomorphy of Solifugae (Sens-enig amp Shultz 2003)

PHYLOGENY OF THE ARACHNID ORDERS 253

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

APPENDAGES OF POSTORAL SOMITES IIIndashVI TARSUS AND TIBIAndashTARSUS JOINT

82 Tarsus divided into proximal basitarsus(= metatarsus) and distal telotarsus (= distitarsus orlsquotarsusrsquo) 0 absent 1 presentState 0 occurs in extant Xiphosura (Shultz 1989) andthroughout Acariformes (Lindquist 1984 Evans1992) in the anterior two pairs of legs in chthonioidpseudoscorpions and all legs in lsquomonosphyronidrsquo pseu-doscorpions including Feaelloidea and Cheliferoidea(Chamberlin 1931) A circumtarsal ring (83) associ-ated with slit sensilla in Parasitiformes (Acari) mayrepresent a joint between a telotarsus and basitarsus(Evans 1992)83 Circumtarsal ring 0 absent 1 presentState 1 is an apparent synapomorphy of Parasiti-formes (Evans 1992)84 Telotarsus in adult with two or more tarsomeres 0absent 1 present - inapplicable due to absence of telo-tarsus (82) [GEWB 71]State 1 occurs in Schizomida Thelyphonida Ambly-pygi Palpigradi Ricinulei Solifugae (Shultz 1989)Opilioacariformes (Van der Hammen 1989) and Ple-siosiro (Dunlop 1999)85 Three telotarsomeres on appendages of postoralsomites IV-VI (= arachnid legs 2ndash4) 0 absent 1present - inapplicable due to absence of telotarsus (82)or absence of telotarsomeres (84) [S 34 WH 77 GEWB102]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Shultz 1989)86 Tibiandashtarsus joint spanned by well-developed elas-tic (lsquospringlikersquo) sclerite 0 absent 1 present [GEWB107]State 1 occurs in Scorpiones phalangid Opiliones andSolifugae (Alexander 1967 Shultz 2000 Sensenig ampShultz 2003) Extant Xiphosura is coded for the lastprosomal appendage only (see 88)87 Appendage of postoral somite VI (= arachnid leg 4)with ring of large basally articulated spatulate pro-cesses at tibia-tarsus joint 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)88 Appendages postoral somites IIIndashV (= arachnidlegs 1ndash3) with tibiotarsus (ie tibia and tarsus not dif-ferentiated) 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 1989) butnot apparently in Weinbergina (Moore et al 2005) theonly synxiphosurid in which appendages are suffi-ciently preserved to determine the number ofpodomeres89 Ambulacrum peduncle-like extension of the tarsuswith internal condylophores and terminating distallywith apotele (eg claws) andor pulvillus 0 absent 1present

State 1 is present throughout Acari (Alberti amp Coons1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI APOTELE AND TARSUSndashAPOTELE JOINT

90 Apotele of appendage III (= arachnid leg 1) 0absent or not apparent 1 present [GEWB 101]State 0 occurs in Amblypygi Thelyphonida and Schi-zomida (Shultz 1999) However the apotele is proba-bly present but undifferentiated because muscles thatwould normally insert on the apotele terminate on theend of the lsquotarsusrsquo Dunlop (1999) reconstructed Hap-topoda as lacking an apotelic claw on leg 1 but noted inthe text that this was speculative Dunlop (2002b) hasreviewed apotelic diversity in Chelicerata91 Appendages of postoral somites IIIndashVI (= arachnidlegs) chelate with chela formed from tibiotarsus andapotele or tarsus and apotele 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988) The condition appears to occur in Chasmataspisbased on associated appendages (Dunlop et al 2004)but not in Diploaspis and Octoberaspis (Dunlop et al2001 Dunlop 2002a)92 Apotele with eversible or padlike empodium(= pulvillus) in adult 0 absent although empodialclaw may be present 1 present [S sim38 WH 81 GEWB95]State 1 occurs throughout Pseudoscorpiones (Cham-berlin 1931) Solifugae (Roewer 1934) Opilioacari-formes Holothyrida Ixodida and free-livingMesostigmata (Evans 1992) An eversible lsquopulvillusrsquo isprobably a primitive feature of Amblypygi (eg Char-inus Weygoldt 2000) but is absent in higher groups(eg Phrynus Weygoldt 2000)93 Inferior apotele muscle (= claw depressor) with tib-ial attachment 0 absent 1 present [S 35 WH 78GEWB 249]The character is coded for extant Xiphosura usingappendage of postoral somite VI (= arachnid leg 4) asother legs lack a differentiated tibia (88) State 1appears to occur throughout Arachnida (Shultz 1989)94 Inferior apotele muscle (= claw depressor) withpatellar attachment 0 absent 1 present [S 36 WH 79GEWB 250]State 0 occurs in extant Xiphosura Acari AmblypygiPalpigradi and Ricinulei State 1 occurs in all otherarachnids examined here (Shultz 1989)

OPISTHOSOMA

95 Number of opisthosomal somites in adult 0 five1 eight 2 nine 3 10 4 11 5 12 6 13 [WH sim33GEWB sim190]Chasmataspidids appear to have 13 opisthosomalsomites (Dunlop et al 2004) The synziphosurans

254 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Weinbergina (Moore et al 2005) and Limuloides(Anderson amp Selden 1997) appear to have ten opistho-somal somites Moore et al (2005) found no evidence ofan anterior microtergite in Weinbergina as proposedby Anderson amp Selden (1997) Xiphosurids appear tohave nine opisthosomal somites (Scholl 1977 Ander-son amp Selden 1997 Shultz 2001) Members of the fol-lowing orders have 12 opisthosomal somites Araneae(Millot 1949ac) Amblypygi (Weygoldt 2000) Thely-phonida and Schizomida (Hansen amp Soslashrensen 1905)and Pseudoscorpiones (Chamberlin 1931) Elevensomites are present in Palpigradi Solifugae (Roewer1934) and Trigonotarbida (Gilboarachne Shear et al1987 Palaeocharinus Fayers et al 2004) Plesiosiromay have 11 or 12 somites (Dunlop 1999) Justifica-tion for coding of more controversial taxa is providedbelow

Scorpiones Two principal hypotheses regarding thenumber of opisthosomal somites in scorpions havebeen advocated a 13-somite hypothesis derived fromembryological studies (Brauer 1895 Patten 1912Farley 1999 2005) and a 12-somite hypothesis basedon comparative anatomy of adults (Weygoldt amp Pau-lus 1979) The embryological interpretation is basedon the observation of pregenital genital and pectinalsomites (each with segmental ganglia and paired limbbuds) in early scorpion embryos followed by extremereduction or loss of the pregenital somite in laterembryos According to this view a missing pregenitalsomite should be added to the 12 apparent opisthoso-mal somites of post-embryonic scorpions to achieve afinal number of 13 The anatomy-based hypothesiswas introduced by Weygoldt amp Paulus (1979) whoadvocated a literal interpretation of post-embryonicsegmentation based on opisthosomal tergites Specifi-cally these authors argued that the last pair of dorsalendosternal suspensor muscles of non-scorpionarachnids especially Pedipalpi attach to the first(= pregenital) somite that this condition also occurs inscorpions and that there is no reason to invoke a miss-ing pregenital tergite They proposed that the pectinesbelong to the genital somite and that functional spe-cializations of the nervous system for pectinal functiongive the apparence of an extra neuromere duringembryonic development

I recently dissected the prosoma and anterioropisthosoma of the scorpions Centruroides Hadrurusand Heterometrus and focused on the composition ofthe muscular diaphragm (103) that separates thehaemocoelic compartments of the prosoma andopisthosoma (Shultz 2007) The diaphragm is com-posed of a metameric series of axial muscles fromthree somites the anterior somite corresponds to thelast prosomal somite and the posterior somite corre-sponds to the genital somite The middle elements

insert dorsally along a tranverse tendon attached tothe anterior margin of the first tergite These obser-vations are consistent with the embryological inter-pretation that a pregenital somite is present but itstergite is not expressed It appears likely that the thepregenital somite was compressed longitudinally dur-ing the evolution of the diaphragm I code scorpions ashaving 13 opisthosomal somites

Eurypterida The eurypterid opisthosoma is widelyassumed to have 12 somites (Clarke amp Ruedemann1912 Stoslashrmer 1944) However in a speculative paperon the evolutionary morphology of trilobites and che-licerates Raw (1957) proposed that both scorpions andeurypterids have 15 opisthosomal somites Rawassumed that scorpions have 13 apparent opisthoso-mal somites based on the transient pregenital somiteof scorpion embryology that scorpions and eurypteridsare close relatives and should have the same numberof somites and that olenellid trilobites and chelicer-ates always have somites in multiples of three Rawachieved 15 somites in scorpions by assuming the lastopisthosomal somite to be a diplosomite and that thetelson is a postanal somite even though there is noevidence for either of these proposals He attributedthese features to eurypterids as well He also notedthat the connection between the prosomal carapaceand first opisthosomal tergite in Eurypterida differedstructurally from the connection between adjacentopisthosomal tergites and regarded this as evidencefor a reduced pregenital tergite in the prosomandashopisthosoma junction Rawrsquos speculations were largelyforgotten until Dunlop amp Webster (1999) resurrectedthe proposal that eurypterids have a reduced opistho-somal tergite and therefore share a unique similaritywith scorpions Unfortunately Dunlop amp Bullocktreated Rawrsquos conjecture as if it were based on empir-ical evidence rather than an attempt to force euryp-terid morphology into a peculiar numerical system Inthe absence of convincing evidence to the contrary Ihave coded eurypterids as having 12 opisthosomalsomites

Ricinulei The opisthosoma consists of a membranouspedicel (97) bearing the gonopore ventrally Thefemale gonopore is bordered by an anterior plate and aposterior plate The remainder of the opisthosoma iscomposed of thick sclerites separated by less heavilysclerotized cuticle The first dorsal sclerite is short andfunctions as part of a prosomandashopisthosoma couplingmechanism (96) that may or may not be a specializedcomponent of the following tergite This is followed byfour tergites and sternites and a three-segmentedmetasoma (= lsquopygidiumrsquo) Millot (1945 1949f) rea-soned that the opisthosoma contains ten somites withsomites VIIndashIX incorporated into the pedicel XndashXIIIexpressed as tergites and sternites and the metasoma

PHYLOGENY OF THE ARACHNID ORDERS 255

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

comprising four somites He did not regard the dorsalcoupling sclerite as a separate tergite and his inter-pretation of four rather than three metasomal somiteshas been rejected Pittard amp Mitchell (1972) also pro-posed ten opisthosomal somites but achieved thisnumber by regarding the dorsal coupling sclerite as atergite of somite IX and by recognizing three metaso-mal somites Legg (1976) adopted the system proposedby Pittard amp Mitchell but did not regard the dorsalcoupling sclerite as separate from the following tergite(X) Van der Hammen (1979 1989) reconstructed 13somites The pedicel was regarded as having twosomites (VII + VIII) the following four sets of tergitesand sternites were considered diplosomites (IX + XXI + XII XIII + XIV XV + XVI) and the metasoma hadthree somites (XVIIndashXIX) This interpretation mayhave been biased by Van der Hammenrsquos attempt tounite Ricinulei with anactinotrichid Acari which healso regarded as having a primitive number of 13somites Selden (1992) described a fossil ricinuleidTerpsicroton that shows two pairs of depressions onthe three large premetasomal tergites which con-trasts with the single pair seen in extant species Thisobservation appears to corroborate the diplosomitehypothesis but the evidence does not indicate that thetergite anterior to these is a diplosomite

Dunlop (1996) attempted to homologize theprosomandashopisthosoma coupling mechanisms ofRicinulei and Trigonotarbida a goal that required anovel and rather forced interpretation of the dorsalsclerites He regarded the coupling sclerite as homol-ogous with the first opisthosomal tergite (VII) of trig-onotarbids and then followed Van der Hammenrsquosdiplosomite hypothesis to achieve 12 somites in totalDunloprsquos scheme differs from previous systems in sug-gesting that the pedicel does not contain the dorsalelements of the first and second somites and is in-consistent with Millotrsquos (1945) observation that thepre- and postgenital plates each have dorsoventralmuscles Here I code Ricinulei as having 12 somitesThere are three metasomal somites (XVIndashXVIII)three diplosomites (= six somites) (XndashXV) one cou-pling somite (IX) and two somites in the pedicel (iethe genital and pregenital somites) (VII VIII)

Opiliones Harvestmen have nine opisthosomalsomites and an anal operculum that is traditionallyregarded as the tergite of a tenth somite a tenth ster-nite is lacking (Hansen amp Soslashrensen 1904 Winkler1957) However the anal operculum appears to repre-sent a persistent embryonic telson (Moritz 1957) andis therefore likely to be a postsegmental structurecomparable with the stinger of scorpions flagella ofthelyphonids etc (Shultz 2000)

Acari The number of opisthosomal somites is probe-matic for most mite taxa due to uncertainty about the

location of the prosomandashopisthosoma boundry (5)extensive simplification or loss of metamericallyarranged sclerites and muscle attachments a paucityof developmental studies of engrailed expression andin Acariformes opisthosomal anamorphosis and het-erochronic modification of somite number (Evans1992) Ixodids appear to be the exception developmen-tal studies indicate five somites in the opisthosoma ofticks (Evans 1992)

Some mites retain external evidence of segmenta-tion ndash metameric patterns of furrows muscle attach-ments and slit sensilla ndash and with certainassumptions the number of opisthosomal somites canbe estimated Here it is assumed that the dorsal sur-faces of the last two prosomal somites are present inmites and retain their primitive association with legs3 and 4 (contra Van der Hammen 1989) (see 5 for jus-tification) Given this there appear to be 11 somites inthe opisthosoma of Opilioacariformes a conclusionalso reached by other workers (eg With 1904 Sitni-kova 1978 Klompen 2000) Similar reasoning sug-gests that Alycus has seven opisthosomal somites notnine as advocated by Van der Hammen (1989) Unfor-tunately external evidence is ambiguous for determin-ing the number of somites in other mite lineages andthese are coded as uncertain96 Prosoma-opisthosoma coupling mechanism 0absent 1 present [GEWB sim24]The posterior margin of the carapace and the anteriormargin of the first apparent opisthosomal tergite arespecialized as a coupling mechanism in Ricinulei andTrigonotarbida and Dunlop amp Horrocks (1996) pro-posed that these are synapomorphic for the twogroups However there is uncertainty about thehomology of the anterior opisthosomal somites in Trig-onotarbida and Ricinulei (95)97 Pedicel 0 absent 1 aranean type 2 ricinuleidtype [WP 30 S sim40 WH sim20 GEWB sim126]The body narrows at or near the prosomandashopisthosoma juncture in several arachnid lineages (ieSolifugae Palpigradi Amblypygi Araneae Ricinulei)and this lsquowaistrsquo has often been used as character at theinterordinal level (eg Pocock 1893) However thisinterpretation is rejected here for being subjective andfor uniting non-homologous conditions For exampleAraneae and Amblypygi are often grouped on the basisof a lsquopedicelrsquo yet it is a highly specialized structure inAraneae and its parts are not readily homologizedwith those of Amblypygi In contrast the condition inAmblypygi is a slightly narrower version of the highlymoveable prosomandashopisthosoma juncture in Uropygiwhich is not generally considered a lsquopedicelrsquo Thepedicel in Ricinulei is also unique a weakly sclero-tized stalk containing the genital opening (Cryptocel-lus Pittard amp Mitchell 1972 Ricinoides Legg 1976)See 95

256 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

98 Opisthosoma with three- segmented lsquobucklerrsquo 0absent 1 presentState 1 occurs throughout Chasmatapsidida (Dunlop2002a)99 Thoracetron consolidation of tergites of postgeni-tal somites 0 absent 1 present - inapplicable codedonly for Xiphosura (Anderson amp Selden 1997)100 Fusion of tergites of postoral somites VIII and IX(= opisthosomal somites 2 and 3) only 0 absent 1present [GEWB 146]State 1 appears to be a synapomorphy Trigonotarbida(Shear et al 1987) Ricinuleids also have diplotergites(101) but these appear to encompass a different com-bination of somites (see 95)101 Three diplotergites 0 absent presentState 1 is a unique synapomorphy of Ricinulei (Selden1992) See 95102 Paired opisthosomal defensive glands openingvia ducts on either side of the anus 0 absent 1present [S 46 WH 86 GEWB 122]State 1 occurs in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905)103 Muscular diaphragm separating prosomal andopisthosomal compartments formed by dorsoventralmuscles of postoral somites VIndashVIII and extrinsic mus-cles of leg 4 0 absent 1 presentState 1 is known only in extant scorpions (Lankesteret al 1885 Firstman 1973) The diaphragm is oftentreated as a single structure but recent anatomicalwork (Shultz 2007) has shown that it is a composite ofdorsal endosternal suspensors and extrinsic leg mus-cles (see 95) A diaphragm is also present in Solifugaebut it is located more posteriorly and does not appearto be homologous with that of scorpions (Roewer1934)104 Opisthosomal appendicular chondrites 0absent 1 presentThese cartilage-like columns of mesodermally derivedtissue are associated with each opisthosomal append-age including the chilaria in extant Xiphosura(Patten amp Hazen 1900 Yamasaki et al 1988 Fahr-enbach 1999 Shultz 2001)105 Paired appendages on ventral surface of postoralsomite VII (= opisthosomal somite 1) in adult 0absent 1 present [WP sim13 S 39 WH 82 GEWB 143]State 1 is known in Weinbergina (Moore et al 2005)and extant Xiphosura The bilobed structure of themetastoma in certain eurypterids as well as the cor-responding placement of chilaria in Xiphosura haveled some workers to regard the metastoma as fusedappendages of postoral somite VII (eg Stoslashrmer1955) This hypothesis is problematic given that (i) nospecial explanation is needed to account for bilaterallysymmetrical structures in bilaterally symmetricalorganisms (ii) incorporation of the ventral part of thefirst opisthosomal sternite as a functional element of

the prosoma is typical of arachnids and perhapsEuchelicerata generally and (iii) some workers regardthe metastoma as a sternite (eg Jeram 1998) Thusthe metastoma is coded here as uncertain Dunlop hasproposed that the sternum of scorpions is derivedfrom appendages citing the presence of transient limbbuds in scorpion development (Brauer 1895 Patten1912) but this is not compelling evidence for a persis-tent appendicular contribution to the sternite of theadult106 Megoperculum 0 absent 1 present [S sim41 + 42WH sim83 GEWB 162 + 163]State 1 occurs in Thelyphonida Schizomida Ambly-pygi Araneae (Shultz 1993 1999) Trigonotarbida(Shear et al 1987) and probably Haptopoda (Dunlop1999) and Palpigradi The megoperculum consists ofappendages of opisthosomal somite 2 (= genital somiteor postoral somite VIII) (Shultz 1993 1999 Popadicet al 1998) that have fused medially and displacedthe ventral body wall of the somite anteriorly (oftenrepresented by a small sternite to which dorsoventralmuscles of postoral somite VIII attach) The megoper-culum projects posteriorly to form the ventral surfaceof a pregenital chamber The dorsal surface of thechamber is formed by the ventral body wall of opistho-somal somite 3 (= postoral somite IX) which serves asthe ventral attachment of the paired dorsoventralmuscles of that somite The megoperculum bears book-lungs (where present) and sometimes paired gonopodsthat are probably derived from telopodites A similarbut less developed operculum is present on theopisthosomal somite 3 which may also bear booklungsand eversible vesicles (eg Amblypygi) (see 111) corre-sponding to the booklungs and gonopods of the mego-perculum (Shultz 1999)

I contend here that a megoperculum is present inPalpigradi and that this is particularly evident infemales Specifically a large unpaired lobe projectsposteriorly from the genital somite over the ventroan-terior surface of the first postgenital somite therebyforming a pregenital chamber The dorsoventral mus-cles of the genital somite attach ventrally near theanterior border of postoral somite VIII and those ofpostoral somite IX attach to the upper surface of thepregenital chamber (Rucker 1901 Boumlrner 1902bRoewer 1934 Millot 1949d Van der Hammen 1989)Weygoldt (1998) has questioned this interpretation ofPalpigradi107 Postgenital operculum or lsquosternitersquo 0 sclerotized1 not sclerotized - inapplicable coded only for Pan-tetrapulmonata (Platnick amp Gertsch 1976)State 1 is present in Trigonotarbida Amblypygi The-lyphonida Schizomida and Mesothelae (Araneae) butnot Opisthothelae (Araneae)108 Genital opening of female guarded by four plates(one pregenital one postgenital two laterogenitals)

PHYLOGENY OF THE ARACHNID ORDERS 257

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

genital opening of male guarded by two plates 0absent 1 presentThis an apparent synapomorphy of Holothyrida(Acari) (Van der Hammen 1989)109 Anterior margin of genital opening in male withglands secreting via fusules 0 absent 1 presentState 1 occurs in Araneae as epigastric or lsquoepiandrousrsquoglands (Marples 1967) and in Palpigradi (Condeacute1991b)110 Paired valve-like plates apparently formed fromcomponents of three somites covering triradiate genitalopening 0 absent 1 presentThis is an apparent synapomorphy of Acariformes(Evans 1992 Alberti amp Coons 1999)111 Eversible lsquoappendagesrsquo on the ventral surface ofpostgenital somites 0 absent 1 present [S sim43 WHsim84 GEWB sim127]Paired ventral lsquoappendagesrsquo operated in part byhaemolymph pressure occur in the form of ventralsacs in certain Amblypygi (postoral somite IX) (Wey-goldt 2000) (Charinus Millot 1949b not PhrynusShultz 1999) prokoenenian Palpigradi (postoralsomites XndashXII) (Rucker 1901 Condeacute 1991a) as spin-nerets in Araneae (XndashXI) (Shultz 1987) and as geni-tal papillae in many Acariformes (Alberti amp Coons1999) including Oribatida many Endeostigmata (Aly-cus Van der Hammen 1989) and Prostigmata (Allo-thrombium Saboori amp Kamali 2000 MicrocaeculusEvans 1992) Eversible vesicles similar to those ofamblypygids occur in fossils of the trigonotarbidPalaeocharinus (postoral somite IX) (Fayers et al2004) Two cuticular structures (lsquogenital verrucaersquo)occur anterior to the genital opening in Opilioacari-formes and each covers the opening to a thin-walledinvaginated sac (lsquogenital papillarsquo) of unknown function(Van der Hammen 1989) Van der Hammen proposedthat it is homologous to a genital papilla of Acari-formes but Alberti amp Coons (1999) have questionedthis interpretation on several grounds and refer tothese structures as pregenital capsules112 Opisthosomal silk glands and spinnerets derivedfrom appendages on postoral somites X and XI(= opisthosomal somites 4 and 5) 0 absent 1 present[WP 33 WH 41 GEWB 123 + 142]State 1 is a unique synapomorphy of Araneae (Plat-nick amp Gertsch 1976) although potentially homolo-gous glands and spigot-like setae occur on the ventralopisthosomal surface in Palpigradi (Millot 1943Condeacute 1991a)113 Opisthsosomal spinnerets location 0 near mid-dle of opisthosoma 1 near posterior end of opisthso-som ndash inapplicable coded only for Araneae (Platnickamp Gertsch 1976)114 Anterior medial lsquospinneretsrsquo 0 absent 1 present- inapplicable coded only for Araneae (Platnick ampGertsch 1976)

115 Opisthosomal tergites divided longitudinally intoone median and two lateral plates 0 absent 1 present[GEWB 145]State 1 occurs in non-curculioid Ricinulei (Selden1992) and many Trigonotarbida (Dunlop 1996) Thisfeature or something very like it (eg distinct opistho-somal trilobation) occurs sporadically in severaleuchelicerate groups including Chasmataspis (Chas-mataspidida) (Dunlop et al 2004) most fossil xipho-surans (Anderson amp Selden 1997) and a few derivedEurypterida (especially Mixopteroidea) (Tollerton1989)116 Number of metasomal somites 0 zero 1 two 2three 3 five 4 nine [S sim44 WH sim85 GEWB 128 144]The metasoma is a preanal region comprising multiplesomites that are substantially narrower than the pre-ceding somites or mesosoma Metasomal somitesoften lack pleural membranes and take the form ofsclerotized rings State 1 occurs in TrigonotarbidaState 2 is present in synziphosurid Xiphosura (Ander-son amp Selden 1997) Amblypyi Thelyphonida Schizo-mida and Ricinulei State 3 occurs in Scorpiones andEurypterida and State 4 occurs in Chasmatispidida

Some workers appear to consider a five-segmentedmetasoma as conclusive evidence for the monophylyof a Eurypterida + Scorpiones clade (eg Dunlop ampBraddy 2001) presumably because the character rep-resents a kind of tagmosis and should therefore begiven substantial phylogenetic weight However Wey-goldt (1998) has pointed out that a metasoma contain-ing three somites is probably part of the ground planof Euchelicerata The known diversity of the eucheli-cerate metasoma indicates that this feature can inincrease or decrease its segmental composition in evo-lution or be eliminated entirely without necessarilychanging the total number of opisthosomal somitesConsequently it is problematic to assume that thefive-segmented metasoma has any special immunityto homoplasy117 Postanal structure (telson) 0 absent or not obvi-ously developed 1 presentPostanal structures are considered non-segmentalbecause they are modifications of the embryonicregion posterior to the site of somite addition (embry-onic growth zone) A well-developed postanal struc-ture persists in adults in Xiphosura EurypteridaChasmataspidida Scorpiones Schizomida Thely-phonida and Palpigradi The anal operculum of Opil-iones is generally considered by morphologists to bethe tergite of the tenth opisthosomal or anal somite(eg Hansen amp Soslashrensen 1904) but embryological evi-dence (Moritz 1957) suggests that it is derived fromthe postproliferative zone and thus corresponds to thetelson of other chelicerates (see 95)118 Postanal structure shape 0 caudal spine 1aculeus 2 flagellum 3 anal operculum - inapplica-

258 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ble due to absence of postanal structure (117) [WP sim41S sim45 WH sim24 GEWB sim121 129 + 147]State 0 is present in Xiphosura Eurypterida andChasmataspidida State 1 occurs in Scorpiones State2 occurs in Palpigradi Thelyphonida and SchizomidaState 3 occurs in Opiliones (see 95 117)119 Specialized postanal flagellum in male (see 157ndash159) 0 absent 1 present - inapplicable coded onlyfor taxa with postanal flagellum (118) [GEWB 131]State 1 is a unique synapomorphy of Schizomida(Hansen amp Soslashrensen 1905 Cokendolpher amp Reddell1992 Reddell amp Cokendolpher 1995)

RESPIRATORY SYSTEM

120 Respiratory medium 0 water 1 airAmong the terminal taxa included only fossil scorpi-ons are problematic for this character They are codedhere as uncertain Based on a study of book lungmicrosculpture Scholtz amp Kamenz (2006) have arguedthat arachnids are primitively terrestrial and pulmo-nate (see also Firstman 1973) and have questionedwhether any fossil scorpions were aquatic No positionon this proposal is taken here121 Respiratory lamellae on opisthosomal somite 2(= genital somite postoral somite VIII) 0 absent 1present [WP sim37 S sim51 WH sim22 GWEB sim133ndash137]This character encompasses book gills (= lamellae thatfunction in water) and book lungs (= lamellae thatfunction in air) The distinction between book gills andbook lungs is accommodated here by 120 in combina-tion with 121ndash124 Respiratory lamellae are presenton the genital somite in Trigonotarbida AraneaeAmblypygi Thelyphonida and Schizomida Petrunk-evitch (1949) reconstructed Plesiosiro as having booklungs but Dunlop (1999) could not corroborate thisThis character was used by Dunlop amp Webster (1999)to propose that Xiphosura and Scorpiones are closelyrelated because they both lack respiratory lamellae onthe genital somite Dunlop amp Braddy (2001) alsoargued for the placement of Eurypterida with Xipho-sura and Scorpiones based in part on this characterAt least some eurypterids may have had respiratorylamellae (eg Manning amp Dunlop 1995) but the onlyevidence of their segmental distribution is derivedfrom one specimen (Braddy et al 1999)122 Respiratory lamellae on opisthosomal somite 3(= postoral somite IX) 0 absent 1 present [GEWB136]State 1 occurs in Amblypygi Thelyphonida and Xipho-sura It is a groundplan feature of Araneae (Platnick ampGertsch 1976) and is present in all representativetaxa included here Braddy et al (1999) proposed theexistence of respiratory lamellae on postoral somitesIXndashXII in Eurypterida based on evidence from onespecimen This character is coded as uncertain for the

Eurypterida included here Extant scorpions are codedhere as lacking respiratory lamellae on this somite(= pectinal somite) based on the interpretation of theopisthosomal segmentation discussed in 95123 Respiratory lamellae on opisthosomal somites 4ndash6 (= postoral somites XndashXII) 0 absent 1 presentState 1 is definitely present in extant Xiphosura andScorpiones and may have occurred in Eurypterida124 Respiratory lamellae on opisthosomal somite 7(= postoral somite XIII)State 1 occurs in extant Xiphosura and Scorpionesonly125 Kiemenplatten 0 absent 1 presentThese structures are located on the roof of opercularchambers in Eurypterida (Clarke amp Ruedemann1912) where they take the form of ventrally projectingcones with a distinct cuticular microsculputure (Man-ning amp Dunlop 1995) Dunlop amp Braddy (2001)inferred the existence of Kiemenplatten in all Palaeo-zoic scorpions based on a description and photos of onespecimen of Paraisobuthus duobicarinatus by Kjelles-vig-Waering (1986 pls 16ndash18) The plates depict darkcone-like rearward-pointing denticles distributedwithin a white amorphous material The denticlesappear to lack microsculpture even though the mag-nifications at which the photos were taken (times90ndashtimes330)are comparable with those illustrating the cones ofKiemenplatten (times170ndashtimes350 in Manning amp Dunlop1995 figs 1 2) Given the diversity of cuticular struc-tures present in the atria of booklungs book gills andtracheae the denticles appear to bear no special sim-ilarity to the cones of Kiemenplatten126 Tracheal system 0 absent 1 paired ventral stig-mata on postoral somite VIII (= opisthosomal somite2) 2 paired ventral stigmata on postoral somites IXand X 3 one pair of stigmata opening near legs 3 or 44 paired stigmata associated with chelicerae 5 fourpairs of stigmata on dorsal surface of opisthosoma -inapplicable aquatic (120) [WP 40 sim43 S sim52 + 5354 WH sim23 45 88 GEWB sim138 139]Firstman (1973) and Weygoldt amp Paulus (1979)hypothesized that tracheae are homologous in all tra-cheate arachnids except spiders and subsequentworkers have entertained this hypothesis by includinga character for the presenceabsence of tracheal sys-tems that ignores the diverse arrangement of stig-mata in arachnids (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) However thisapproach assumes that internal tracheal systems areconserved but that tracheal openings (stigmata)appear and disappear on different parts of the bodywith higher evolutionary frequency Here I assumethat stigmata are conserved in evolution and that dif-ferences in their anatomical placement reflect the evo-lution of new tracheae State 1 occurs throughoutOpiliones State 2 occurs throughout Solifugae

PHYLOGENY OF THE ARACHNID ORDERS 259

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

(Roewer 1934) and Pseudoscorpiones (Chamberlin1931) State 3 recognizes the possible homology of stig-mata in Ricinulei Opilioacariformes and Parasiti-formes (Acari) (Lindquist 1984 Van der Hammen1989) Tracheal systems are apparently absent in thegroundplan of Acariformes but Prostigmata typicallyhave tracheae associated with the chelicerae Tra-cheae derived from the posterior pair of book lungs arewidespread in araneomorph Araneae but the booklungs are retained and tracheae are absent in themost basally divergent groups (eg Hypochilus)

BOXndashTRUSS AXIAL MUSCLE SYSTEM (BTAMS)

127 Posterior oblique muscles of BTAMS of postoralsomites IndashVI 0 absent 1 present in one or moresomitesState 0 occurs in Xiphosura and state 1 occurs inPalpigradi Araneae Amblypygi Thelyphonida andSchizomida (Shultz 2001) The condition in AcariRicinulei Opiliones and Solifugae is not known128 Anterior oblique muscles of BTAMS posterior topostoral somite VI 0 absent 1 presentState 1 occurs in Xiphosura which is probably theprimitive condition based on comparison with otherarthropods (Shultz 2001) State 0 occurs in all arach-nids examined thus far129 Ventral attachments of posterior oblique musclesof opisthosomal BTAMS located in prosoma 0 absent1 presentState 1 occurs in extant Xiphosura (Shultz 2001)130 Endosternite fenestrate 0 absent 1 present [S 8WH 54 GEWB 35]State 1 occurs in Thelyphonida and hubbardiid Schi-zomida (Firstman 1973) The condition in protoschi-zomid Schizomida is not known131 Suboral suspensor a tendon that arises from theBTAMS and inserts on the ventral surface of the oralcavity via muscle 0 absent 1 presentState 1 occurs in Palpigradi (Eukoenenia Millot1943) Amblypygi (Shultz 1999) and Thelyphonida(Shultz 1993)132 Perineural vascular membrane in adult 0absent 1 present [WH 28 GEWB 155]Wheeler amp Hayashi (1998) coded extant Xiphosura asunknown although the primary source for this char-acter (Firstman 1973) stated that adult Limulus havea perineural vascular membrane Within Arachnidapresence of a perineural vascular membrane is appar-ently correlated with the presence of tracheae (126)133 Ventral endosternal suspensor attaching on coxaof anteriorly adjacent somite 0 absent 1 present[GEWB 239 253]State 1 occurs in Amblypygi (Shultz 1999) Thely-phonida (Shultz 1993) and Schizomida (Shultzunpubl observ)

134 Posteriormost postoral somite with a pair of dor-soventral muscles 0 VI 1 VII 2 VIII 3 XII 4 XIII5 XIV 6 XV 7 XVIDorsoventral muscles tend to run in a continuousmetameric series beginning in the prosoma Deter-mining the last tergite on which the muscle seriesends is substantially easier than counting especiallygiven frequent anatomical complexities near theprosomandashopisthosoma juncture The dorsoventralmuscle series ends on postoral somite VII inCyphophthalmi (eg Chileogovea) and Laniatores(Opiliones) (Shultz unpubl observ) on postoralsomite VII or VIII in Eupnoi (Leiobunum Shultz2000 unpubl observ) on postoral somite XII in Pal-pigradi (Roewer 1934) Solifugae (Bernard 1896Roewer 1934 Millot amp Vachon 1949) and Araneae(Liphistius Hypochilus Millot 1933 but segmenta-tion reinterpreted here following Shultz 1993 1999)on postoral somite XIII in hubbardiid Schizomida(Cokendolpher amp Reddell 1992) and Scorpiones(Lankester et al 1885) and on postoral somite XIVin extant Xiphosura (Shultz 2001) Thelyphonida(Shultz 1993) protoschizomid Schizomida (Cokendol-pher amp Reddell 1992) and Amblypygi (Shultz 1999)Dunlop (1999) reconstructed Plesiosiro as havingpaired tergal apodemes ending on postoral somite XVbut his figures indicate that the series ends on posto-ral somite XIV The interpretation of opisthosomalsegmentation in Ricinulei coded here (95) indicatesthat the series ends on postoral somite XV in Terpsi-croton and this is treated as the groundplan for theorder The posteriormost dorsoventral muscles inPseudoscorpiones occur on postoral somite XVI(Vachon 1949)

NERVOUS SYSTEM

135 Segmental ganglia 0 consolidated in prosoma1 one or more present in opisthosoma [WH 30 GEWB210 + 216]Adult neuromeres are exclusively prosomal in Ambly-pygi liphistiomorph and araneomorph Araneae Opil-iones Pseudoscorpiones Ricinulei and Acari (Millot1949a)136 Dorsal median eyes 0 absent 1 present [WP14 + 47 S 50 WH sim7 GEWB 1]Median eyes are a groundplan feature of Euchelicer-ata and are retained in extant Xiphosura Eurypteridaand occur in all arachnid orders except PalpigradiSchizomida Ricinulei and Pseudoscorpiones (Paulus1979) Median eyes are also absent in opilioacariformand parasitiform Acari (Lindquist 1984) and perhapsthe fossil whipscorpion Proschizomus (Dunlop amp Hor-rocks 19951996) They are unknown in synziphosu-ran Xiphosura Median eyes are present in a variety ofbasal acariform mites including Endeostigmata (but

260 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

not Alycus) Prostigmata (Microcaeculus) and certainoribatids (Palaeacarus) (Evans 1992 Alberti amp Coons1999) Giribet et al (2002) coded all representativeAcari as lacking median eyes It is not known whetherthe eyes of some cyphophthalmid opilions are medianor lateral evidence from tracheal branching inCyphophthalmus (Janczyk 1956) suggests that theyare median eyes (Shultz amp Pinto da Rocha 2007) andpresence of a tapetum in Stylocellus is consistent withlateral eyes (Shear 1993)137 Retinula cells of dorsal median eyes 0 organizedinto closed rhabdoms 1 organized into network ofrhabdomeres 2 disorganized - inapplicable due toabsence of median eyes (136) [GEWB sim3]State 0 is present in Scorpiones Thelyphonida andAmblypygi State 1 is present in Solifugae and Ara-neae state 2 is present in Xiphosura (Paulus 1979)The retinula cells of median eyes have been studied inseveral prostigmatid Acariformes and are organized ina network in some taxa and in an irregular pattern inothers (Alberti amp Coons 1999) Retinulae in phalangidOpiliones have state 0 proximally and state 2 distally(Schliwa 1979)138 Ventral median eyes 0 absent 1 presentState 1 occurs in early instars of extant Xiphosura(Paulus 1979)139 Lateral eyes 0 absent 1 presentLateral eyes are primitively present in Chelicerataand are absent in Palpigradi and Opiliones (Paulus1979) It is unclear whether the eyes of cyphoph-thalmid opilions are median or lateral (see 136)140 Arrangement and number of lateral eyes 0 com-pound many 1 five or more pairs (includes micro-lenses) 2 three primary pairs (excludes microlenses)3 two pairs 4 one pair - inapplicable due to absenceof lateral eyes (139) [WP sim13 + 18 + 38 + 44 + 52 Ssim49 WH sim10 GEWB sim4 + 5]True compound eyes are present in XiphosuraEurypterida Chasmataspidida and many PalaeozoicScorpiones Among the anactinotrichid Acari Opilioa-cariformes have two (Neocarus) or three (Siamacarus)pairs of lateral eyes (although at least one species ofSiamacarus lacks eyes) the allothryrid Holothyridaand many Ixodida have a single pair of lateral lensesand the remainder apparently lack eyes (Evans1992) Thelyphonida was coded as lsquo12rsquo to reflect fivepairs of lenses comprising three pairs of primarylenses and two pairs of small accessory lenses (notsimply three pairs as coded by Giribet et al 2002)Trigonotarbida is also coded as lsquo12rsquo to reflect threepairs of primary lenses and multiple small accessorylenses Dunlop (1999) illustrated Haptopoda as havingpaired triads of lateral eyes but noted that there wasactually no evidence of this in the fossils The lateraleyes of fossil ricinuleids have two pairs of lenses(Selden 1992) One pair of eyes or eyespots are

present in extant Ricinulei and many Schizomidaalthough five genera of hubbardiid Schizomida have apair of lenses (Reddell amp Cokendolpher 1995) State 1is the groundplan for extant Scorpiones and State 3occurs in the pseudoscorpions Chthonius FeaellaNeobisium (but State 4 in Chelifer) and Solifugae(Paulus 1979)141 Lateral eyes with closed rhabdoms 0 absent 1present - inapplicable due to absence of lateral eyes(139) [WP 21 WH 12 GEWB 6]Presence of closed rhabdoms is probably primitive forChelicerata and is retained in extant Xiphosura andScorpiones Retinula cells form a network of rhab-domeres in other extant chelicerates (Paulus 1979)but these networks can differ substantially in detailWeygoldt (1998) and other workers have given sub-stantial weight to the network character in unitingnon-scorpion arachnids but these authors seem not togrant comparable phylogenetic significance to theanalogous character of the median eyes shared bySolifugae and Araneae (137)142 Slit sensilla 0 absent 1 present [WP 19 S 47WH 11 GEWB 209]State 1 occurs in all arachnid orders except Palpigradi(Shultz 1990) The proposal that lsquoprimitiversquo slit sen-silla were present in Eurypterida (Dunlop amp Braddy1997) appears to be based on the over-interpretation ofa comparatively large notch that occurs in Baltoeu-rypterus at the terminus of the tibia (= podomere 7)(see also Edgecombe et al 2000 Giribet et al 2002)The neural construction of slit sensilla like function-ally similar campaniform sensilla of hexapods is sim-ilar to that of trichoid sensilla (Chapman 1998Klompen 2000) Thus these cuticular stress receptorsmay represent modified bases of sensory setae a viewsupported by the replacement of setae by slit sensilladuring post-embryonic development in some Opilioac-ariformes (Klompen 2000)143 Trichobothria 0 absent 1 present [GEWB 213]Trichobothria are present in extant Scorpiones(Jeram 1998) Pseudoscorpiones (Chamberlin 1931)some endeostigmatid (Alycus Van der Hammen 1989)and prostigmatid Acariformes and most Oribatida(Lindquist 1984) They occur on the ventral surface ofthe pedal femora in the opilioacariform Siamacarus(Leclerc 1989) They are also present in Araneae (Foe-lix 1996) Amblypygi (Weygoldt 2000) Schizomidaand Thelyphonida (Hansen amp Soslashrensen 1905) Theyare apparently absent in all non-arachnid chelicer-ates Solifugae Ricinulei Opiliones (Reissland ampGoumlrner 1985) and Parasitiformes (Acari) (Lindquist1984)144 Tibial trichobothria with 2-1-1-1 pattern onappendages IIIndashVI (= arachnid legs 1ndash4) 0 absent 1present - inapplicable due to absence of trichobothria(143) [S 48 WH 87 GEWB 88]

PHYLOGENY OF THE ARACHNID ORDERS 261

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs only in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905) Note that Shultz (1990)erroneously described this character as a 2-2-1-1 pat-tern and that this error was repeated by Giribet et al(2002) and described as 2-1-1 by Wheeler amp Hayashi(1998)145 Paired trichobothria on dorsal surface ofprosoma 0 absent 1 present - inapplicable due toabsence of trichobothria (143)State 1 is an apparent synapomorphy of Acariformesand is present in all acariforms included in this study(Alberti amp Coons 1999)146 Malleoli 0 absent 1 present [WP 45 GEWB 96]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)147 Pectines 0 absent 1 present [WP 20 WH 38GEWB 120]State 1 is a unique synapomorphy of Scorpiones148 Intercheliceral median organ 0 absent 1present [GEWB 212]State 1 is a unique synapomorphy of Palpigradi(Roewer 1934)149 Tarsal organ on appendage of postoral somite III(= arachnid leg 1) (= Hallerrsquos organ) sensilla containedwithin a cuticular depression on the superior surface ofthe tarsus of appendage III (= arachnid leg 1) 0absent 1 present [GEWB sim100]Klompen (2000) has noted that State 1 occurs in Opil-ioacariformes Parasitiformes (except Mesostigmata)and Ricinulei where it also occurs on leg 2 (Talaricoet al 2005) State 1 occurs on all legs in Araneae andperhaps Scorpiones (Foelix 1985)150 Tarsal organ on appendage of postoral somite IV(= arachnid leg 2) 0 absent 1 present (See 149)

REPRODUCTION

151 Gonads 0 primarily prosomal 1 primarilyopisthosomalState 0 is limited to Xiphosura (Sekiguchi 1988)State 1 occurs throughout Arachnida (Millot1949a)152 Ladder-like opisthosomal gonadsaccessoryglands (see 153) 0 absent 1 present [WH sim37 GEWBsim158]Giribet et al (2002) followed Wheeler amp Hayashi(1998) who followed Clarke (1979) in coding gonads asreticulate (Xiphosura) ladder-like (Scorpiones Thely-phonida Schizomida) or lsquosaccularrsquo (all remainingArachnida) However comparisons between Xipho-sura and Arachnida are problematic given that thexiphosuran gonads are primarily prosomal and thoseof arachnids are primarily opisthosomal (151) Fur-ther the reticulate pattern in Limulus (Xiphosura)but not other extant xiphosurans contains a distinctlyladder-like component The lsquosaccularrsquo state is probably

artificial as it encompasses a wide variety of pairedand unpaired structures153 Male gonads in two distinct parts one producingsperm and another (tubular gland) producing a holo-crine secretion similar to degenerate sperm 0 absent1 presentState 1 occurs in Thelyphonida Schizomida andAmblypygi although the holocrine material is pro-duced by ventral organs in Amblypygi and dorsalorgans in Thelyphonida and Schizomida (Alberti2005)154 Number of gonopores 0 two 1 oneExtant xiphosurans have two small genital openingson the base of the genital telopodite and all extantarachnids have a single opening The condition inEurypterida is not known Clarke amp Ruedemann(1912) located a pair of openings near the base of themedian organ (161) that are the outlets of the lsquohornorgansrsquo but it is unclear whether these are genitalducts or accessory structures Braddy amp Dunlop (1997)have developed numerous speculations about thesestructures and extended their arguments far beyondthe available evidence155 Genital opening (ie gonopore or gonostome)appearing to open in prosomal region (ie between legcoxae or anterior to posterior carapacal margin) 0absent 1 present [WP sim50 WH 26 GEWB 166]The genital opening in Euchelicerata is located onpostoral somite VIII (= opisthosomal somite 2) but ithas shifted anterior to the posterior margin of thecarapace or between the coxae of the last pair of legsin most Scorpiones (but not in PalaeoscorpiusKjellesvig-Waering 1986) and Opiliones The geni-tal opening occurs near or anterior to the last coxaein the opilioacariform and parasitiform Acari repre-sented here It is variable in Acariformes but islocated posterior to the coxae in all representativetaxa156 Ovipositor 0 absent 1 present [WP sim51 S sim60WH sim91 GEWB sim172]An ovipositor with a trilobed terminus is an apparentgroundplan character of Oribatida (Lindquist 1984Alberti amp Coons 1999) An ovipositor is also present inOpilioacariformes and Opiliones (Van der Hammen1989)157 Stalked spermatophore attached to substratum0 absent 1 present [S 57]State 1 occurs in Scorpiones PseudoscorpionesAmblypygi Thelyphonida and Schizomida The mech-anism of sperm transfer is unknown in PalpigradiOpilioacariformes and Holothryrida158 Male turns posterior end toward female duringmating behaviour 0 absent 1 presentState 1 occurs in Amblypygi (Weygoldt 2000) The-lyphonida and Schizomida (Weygoldt amp Paulus1979)

262 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

159 Female grasps male opisthosoma during matingbehaviour 0 absent 1 present [WP 29 S 58 WH 19GEWB 188]State 1 occurs in Thelyphonida and Schizomida (Wey-goldt amp Paulus 1979) Evolution of this behaviour wasprobably facilitated by 158160 Penis 0 absent 1 present [WP sim51 GEWB 167]State 1 is a unique synapomorphy of Opiliones A truepenis occurs in Phalangida A clearly homologousstructure is present in Cyphophthalmi (Opiliones) andapparently functions in depositing a spermatophore inthe femalersquos genital chamber (Karaman 2005) A truepenis may occur in some mites but it does not appearto be a groundplan feature of any major group (Evans1992) The lsquopenisrsquo in Oribatida is really a spermatop-ositor it functions in construction of a spermatophore(Alberti amp Coons 1999)161 Median organ 0 absent 1 presentState 1 occurs throughout Eurypterida (Clarke amp Rue-demann 1912) and has been observed in diploaspididChasmataspidida (Dunlop 2002a also Loganamaras-pis Tetlie amp Braddy 2004)

SPERM MORPHOLOGY

162 Nucleus with manchette of microtubules 0absent 1 present [S 54 WH 89 GEWB 193]State 1 occurs in Araneae Amblypygi Thely-phonida Schizomida Ricinulei and gonyleptid Lani-atores (Opiliones) (Alberti 1995 Giribet et al2002)163 Axoneme 0 absent 1 present [WP sim49 S sim55WH sim25 GEWB 195]State 1 occurs in Xiphosura Scorpiones Cyphoph-thalmi (Opiliones) Pseudoscorpiones RicinuleiAraneae Amblypygi Thelyphonida and Schizomida(Alberti 1995)164 Coiled axoneme 0 absent 1 present - inappli-cable due to absence of axoneme (163) [WP 22 S sim55WH 13 GEWB 196]State 1 occurs in Pseudoscorpiones Ricinulei Ara-neae Amblypygi Thelyphonida and Schizomida(Alberti 1995)165 Microtubule arrangement in axoneme 0 9 + 0 19 + 1 2 9 + 2 3 9 + 3 - inapplicable due to absence ofaxoneme (163) [WP sim26 S sim56 WH sim16 GEWB 198](Alberti 1995)166 Helical or corkscrew shaped nucleus 0 absent 1present [GEWB sim204]State 1 occurs in Araneae Amblypygi ThelyphonidaSchizomida and certain Scorpiones (eg Hadrurus)(Alberti 1995)167 Vacuolated-type sperm 0 absent 1 present[GEWB 205]State 1 is unique to Opilioacariformes and Parasiti-formes (Acari) (Alberti 1995)

168 Sperm aggregates 0 absent 1 present [GEWB206] (Alberti 1995)

DEVELOPMENT

169 Yolk in early embryo 0 concentrated (centroleci-thal or telolecithal) 1 evenly distributed (isolecithal)[WH sim34 GEWB sim191]Wheeler amp Hayashi (1998) and Giribet et al (2002)coded this character as lsquo0 isolecithal or telolecithal 1centrolecithalrsquo based on information presented byYoshikura (1975) Their coding appears to highlightseparate conditions found in Scorpiones (ie isoleci-thal and telolecithal) rather than a property intrinsicto the character itself170 Embryonic nutrition other than yolk 0 absent 1present [WP sim49]State 1 occurs in extant Scorpiones (except buthidsand chaerilids) and Pseudoscorpiones (Weygoldt1969)171 Embryological growth zone 0 initiating segmentaddition within prosoma 1 initiating segment addi-tion posterior to prosoma [GEWB sim192]Giribet et al (2002) followed Dunlop amp Webster (1999)in stating that only Xiphosura and Scorpiones havestate 0 among Chelicerata However the last prosomalsomite develops from the growth zone in Ixodida(Anderson 1973 Evans 1992) A similar processapparently occurs in the acariform mites Tyroglyphus(Sarcoptiformes Astigmata) (Yoshikura 1975) andArchegozetes (Sarcoptiformes Oribatida) (Thomas ampTelford 1999) but not apparently in Tetranychus(Prostigmata) (Dearden Donly amp Grbic 2002)172 Eggsembryos maintained in external attachedbrood sac secreted by genital glands 0 absent 1present [WP sim46 S 59 WH 90 GEWB 208 = 219]State 1 occurs throughout Amblypygi SchizomidaThelyphonida and Pseudoscorpions Giribet et al(2002) followed Shear et al (1987) in homologizingspinneret-derived silken egg sacs of spiders with gen-ital-gland-derived brood sacs of Pedipalpi However asthese structures are not homologous in secretory ori-gin construction or composition (Shultz 1987) theyare not regarded here as homologous Giribet et al(2002) miscoded the character as absent in Pseudo-scorpiones and without justification in PalpigradiFemale palpigrades have well-developed glands asso-ciated with postgenital somites and an array of setaefusules and other structures (112) (Boumlrner 1902b Mil-lot 1942 Condeacute 1991a) which are consistent withsome sort of brood care and Palpigradi should there-fore be coded as unknown for this character173 Embryonic and early postembryonic lsquolateralrsquo orClaparegravede organs associated with coxa of postoralsomite IV (= arachnid leg 2) 0 absent 1 present [WHsim31 GEWB sim211]

PHYLOGENY OF THE ARACHNID ORDERS 263

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bilaterally paired rounded protuberances locatedbetween coxae of appendages of postoral somites IIIand IV (arachnid legs 1 and 2) in prelarval and larvalinstars appear to be a primitive feature of Acariformes(Lindquist 1984 Evans 1992 Alberti amp Coons 1999)It is known to be derived embryologically from thecoxa of leg 2 in the oribatid Archegozetes (Thomas ampTelford 1999) Structures apparently homologouswith the acariform Claparegravede organ are present inembryonic Amblypygi (Weygoldt 2000) and Thely-phonida (Yoshikura 1975) and in embryonic and earlypost-embryonic stages of Solifugae (Roewer 1934)where they are termed lateral organs The so-calledlsquolateral organrsquo of embryonic Xiphosura (174) occurs onthe lateral surface of the carapace and does not appearto be homologous with the coxa-associated structuressharing their name In fact the xiphosuran lateralorgan develops even when the coxa of postoral somiteIV has been excised (Sekiguchi 1988) In contrastWheeler amp Hayashi (1998) and Giribet et al (2002)followed Yoshikura (1975) in regarding all lsquolateralorgansrsquo as homologous and in miscoding this characteras absent in Acariformes Van der Hammen (1989)suggested that the sternal verrucae of Opilioacari-formes are homologous with the Claparegravede organ butit differs in gross structure and ontogenetic timing174 Embryonic lateral organ associated with cara-pace 0 absent 1 present [WH sim31 GEWB sim211]State 1 is known only from embryonic stages of extantXiphosura (Yoshikura 1975) In contrast to previousinterpretations this feature does not appear to behomologous with the lsquolateralrsquo or Clarapegravede organ ofcertain arachnids (173)175 Live birth 0 absent 1 presentState 1 occurs in extant Scorpiones and has evolvedseveral times in a few mites (Evans 1992)176 Hexapodal larva and 1ndash3 nymphal stages 0absent 1 present [WP 53 S sim61 WH 27 GEWB 186]State 1 occurs in Acari and Ricinulei (Lindquist1984)177 Hexapodal prelarva 0 absent 1 present [S sim61WH 47]State 1 occurs in Opilioacariformes (NeocarusKlompen 2000) and many Acariformes (Evans 1992)

EXCRETORYOSMOREGULATORY SYSTEM

178 Malpighian tubules 0 absent 1 present [WP 17S 62 WH 9 GEWB 153]State 1 occurs in all major arachnid groups except Pal-pigradi Pseudoscorpiones Opiliones Oribatida (Sar-coptiformes) and Prostigmata Within Acari State 1occurs in all major anactinotrichid lineages (Opilioac-ariformes Holothyrida Mesostigmata Ixodida) andsimilar structures are present in certain Astigmata(Evans 1992 Alberti amp Coons 1999)

179 Dorsomedian excretory organ 0 absent 1presentThis is a specialized excretoryosmoregulatory organformed by the postventriculus and proctodeum inProstigmata (Alberti amp Coons 1999)180 Adult coxal organ opening on or near coxa ofappendage III (= arachnid leg 1) 0 absent 1 present[S 64 WH 93 GEWB 90]State 1 occurs throughout Acari (Evans 1992 Albertiamp Coons 1999) Ricinulei (Pittard amp Mitchell 1972Legg 1976) Palpigradi (Millot 1942) AraneaeAmblypygi Thelyphonida and Schizomida (Buxton1913 1917) There is evidence that the second lsquoozo-porersquo of gonyleptids (Opiliones Laniatores) (Hara ampGnaspini 2003) is a persistent opening to the coxalorgan associated with leg 1 (Soslashrensen 1879) whichalso occurs in embryonic Phalangium (Opiliones Eup-noi) (Moritz 1959)181 Adult coxal organ opening on or near coxa ofappendage V (= arachnid leg 3) 0 absent 1 present [S63 WH 92 GEWB 89]State 1 occurs in extant Xiphosura (Yamasaki et al1988) Scorpiones mygalomorph (Buxton 1913 1917)and mesothele Araneae (Liphistius J Millot in Bris-towe 1932) State 1 occurs in basal Amblypygi(Charon Charinus Buxton 1913 1917) but degener-ates prior to the adult stage in higher groups(Phrynus Weygoldt 2000) Adequately preservedeurypterids show a small submarginal opening on thecoxa of appendage V (Baltoeurypterus Selden 1981also Eurypterus Hughmilleria Clarke amp Ruedemann1912) and this is interpreted here as the opening to acoxal organ Giribet et al (2002) coded this characteras uncertain for Eurypterida182 Adult coxal organ opening on coxa of appendageII (= arachnid palp) 0 absent 1 present [GEWB 65]Buxton (1913 1917) noted that the coxal organ ofSolifugae opens on the prolateral surface of the palpalcoxa The proximal end of the duct leading to the ori-fice is associated with glands and apparently theglandular secretion andor coxal fluid serve as saliva(Alberti 1979) (see also 14 15)183 Genital papillae 0 absent 1 present - inappli-cable coded only for those taxa with eversible lsquoappend-agesrsquo (111)The genital papillae are associated with the genitalopening in postlarval Acariformes (Evans 1992Alberti amp Coons 1999) These structures are probablyserially homologous with the prosomal Claparegravedeorgan (173) and both structures probably function inwater and ion regulation

DIGESTIVE SYSTEM

184 Ingestion 0 solid food 1 primarily liquid foodwith or without preoral digestion [WP sim16 WH sim8GEWB sim218]

264 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Extant xiphosurans ingest solid food and members ofmost extant arachnid orders ingest fluids Opilionesand Acari are the principal exceptions The phalangidOpiliones ingest solids but Shultz (2000 unpublobserv) has shown that the precerebral pharyngealapparatus of Cyphophthalmi is very similar to that ofthe fluid-feeding Scorpiones and differs substantiallyfrom phalangid Opiliones Examination of gut con-tents of the cyphophthalmids Siro acaroides S exilisand Chileogovea oedipus have failed to reveal solidparticles typically found in the guts of phalangids(Shultz unpubl observ) the character is coded asunknown for Cyphophthalmi Within Acari consump-tion of particles has been documented in Opilioacari-formes (Neocarus Van der Hammen 1989unidentified Walter amp Proctor 1998) Holothyridanon-parasitic Sarcoptiformes (Evans 1992 Walter ampProctor 1998 Alberti amp Coons 1999) and many endeo-stigmatids including the nematophagous Alycusroseus (Walter 1988)185 Mouth 0 directed posteroventrally 1 directedanteroventrally [S 9 WH 55 GEWB 28]State 1 occurs throughout Arachnida (Shultz 1990)186 Oral cavity dilated by muscles arising from coxaeand constricted by large circular sphincter 0 absent1 presentState 1 occurs in extant Xiphosura (Manton 1964Scholl 1977 Shultz 2001)187 Palate plate 0 absent 1 present [GEWB 159]State 1 is a unique synapomorphy of Araneae (Dunlop1994)188 Lateral walls of epistome broadly fused to medialwalls of palpal coxae opposite sides connected by well-developed transverse epistomal muscle 0 absent 1presentState 1 occurs throughout Opiliones and Scorpiones(Shultz 2000 unpubl observ)189 Epistome with a pair of lateral arms that projectsposteriorly into the prosoma on either side of the phar-ynx 0 absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)190 Epistome with four pairs of suspensor musclesattaching to the carapace 0 absent 1 presentState 1 occurs in extant Scorpiones (Lankester et al1885 Vyas 1970 Shultz unpubl observ) and appar-ently in Solifugae (Roewer 1934)191 Intercheliceral epipharyngeal sclerite 0 absent1 present [GEWB 31]State 1 occurs in Palpigradi (Boumlrner 1904) Trigo-notarbida (Palaeocharinus Dunlop 1994) Araneae(Firstman 1954 Marples 1983) Amblypygi (Mil-lot 1949b Shultz 1999) Schizomida (Millot1949e) Thelyphonida (Millot 1949e Shultz 1993)and Phalangida (Opiliones) (Shultz 2000 unpublobserv)

192 Epipharyngeal sclerite large projecting posteri-orly 0 absent 1 present - inapplicable due to absenceof sclerite (191) [GEWB 32]State 1 occurs in Amblypygi (Millot 1949b Shultz1999) Schizomida (Millot 1949e) Thelyphonida (Mil-lot 1949e Shultz 1993) some Araneae (Heptathela ndashcoded as Liphistius ndash and Hypochilus Marples 1983not Aphonopelma Firstman 1954) and perhaps Trig-onotarbida (Palaeocharinus Dunlop 1994)193 Dorsal dilator muscle of precerebral pharynxattaching to intercheliceral septum or associatedepipharyngeal sclerite (191) 0 absent 1 presentState 1 occurs in Palpigradi (Millot 1943) Araneae(Marples 1983) Amblypygi Schizomida and Thely-phonida (Millot 1949e Shultz 1993 1999)Extremely fine muscles have been documented insome Scorpiones (Shultz 2007) and in Leiobunum(Opiliones) (Shultz 2000) Lankester et al (1885)reported a muscle in Limulus that would appear tocorrespond to this muscle but the muscle does notexist (Manton 1964 Shultz 2001)194 Dorsal dilator muscle of precerebral pharynxattaching to dorsal surface of prosoma 0 absent 1present [GEWB 228]State 1 occurs in Palpigradi (Roewer 1934) and mostAraneae (Palmgren 1978 Aphonopelma Firstman1973) but not in Heptathela (coded as Liphistius) orHypochilus (Marples 1983)195 Lateral dilator muscle of precerebral pharynxattaching to endosternite 0 absent 1 present [GEWBsim226]State 1 is a unique feature of Araneae (Marples 1983)196 Lateral dilator muscle of precerebral pharynxattaching to lateral surface of epistomal processes 0absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)197 Lateral dilator muscle of precerebral pharynxattaching to medial process of coxa of appendage II(= arachnid palp) 0 absent 1 present [GEWB sim226]State 1 occurs in Amblypygi Thelyphonida and Schi-zomida (Boumlrner 1904 Shultz 1993 1999)198 Dilator muscle of precerebral pharynx andorpreoral cavity attaching to ventral surface of prosoma0 absent 1 present [GEWB 227]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) Araneae (Marples 1983) Palpigradi(Millot 1943) and Solifugae (Roewer 1934)199 Postcerebral pharynx 0 absent 1 present [WPsim31 S sim5 WH sim21 GEWB sim154]The cuticle-lined foregut passes through the centralnervous system in many arthropods That portion pos-terior to the CNS is here termed the postcerebralpharynx regardless of the details of its morphology orfunctional specialization A substantial postcerebralpharynx is known in extant Xiphosura (Lankester

PHYLOGENY OF THE ARACHNID ORDERS 265

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

et al 1885 Manton 1964 Yamasaki et al 1988)Scorpiones (Centruroides Hadrurus HeterometrusShultz 2007 also Androctonus Abd el-Wahab 1952)Solifugae (Millot amp Vachon 1949) Araneae Ambly-pygi Thelyphonida and Schizomida (Millot 1949andasheShultz 1993) Weygoldt amp Paulus (1979) applied theterm lsquopostcerebral pharynxrsquo only to the lsquosucking stom-achrsquo of Araneae and Amblypygi but the sucking stom-ach is here regarded as a complex of three characters199ndash201

200 Dilator muscle of postcerebral pharynx attach-ing to endosternite 0 absent 1 present - inapplicabledue to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Lankester et al1885 Manton 1964 Shultz 2001) Scorpiones (Cen-truroides Hadrurus Heterometrus Shultz 2007 alsoAndroctonus Abd el-Wahab 1952) Araneae (First-

man 1954 Palmgren 1978) Amblypygi (Millot1949b Shultz 1999) and apparently the palpigradeProkoenenia wheeleri (Rucker 1901) but not inEukoenenia miriabilis (Millot 1943 contra Boumlrner1904)201 Dilator muscle of postcerebral pharynx attachingto dorsal surface of prosoma 0 absent 1 present -inapplicable due to absence of postcerebral pharynx(199)State 1 occurs throughout Araneae (Millot 1949cFirstman 1954 Palmgren 1978) and Amblypygi(Charinus Millot 1949b Phrynus Shultz 1999)202 Crop and gizzard 0 absent 1 present - inap-plicable due to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Sekiguchi 1988)but not in extant Arachnida

226 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Archegozetes000011000 040110020 000-00001 10100-000 000--00000 000000001 0000000020 02--011011 000--00011 0010000-0 000000-001 10--000--1 000000 00-00- -110100000 1001011000 000--00000 0010011001 0010100000 0-0000000- --

Cyphophthalmus0000100100 0100-00000 100-00000 -0000-1000 0010000001 1000000001 0000000020 01--010-11 0100-00001 00112000-0 000000-000 00--0013-1 000001000 010100 10--00000 1001110001 001100010 0000000 10-100110 0-0001000- --

Chileogovea0000100100 0100-00000 1010-00000 -0000-1000 0010000001 1000000001 0000000020 01--010-11 0100-00001 00112000-0 000000-000 00--0013-1 000001000 010100 10--00000 1001110001 0000 0000000 10-100110 0-0001000- --

Caddo0000100100 0000-00000 100-00000 -0000-000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 01010100--10--00000 10-1110001 000000 0000000 10-0100110 10001000- --

Leiobunum0000100100 0000-00000 1010-00000 -0000-0000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 010B01C00- -10--00000 1001110001 000--00000 1000000000 10-0100110 101001000- --

Sclerobunus0000000100 0000-00000 100-00000 -0001-000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 010001C00- -10--00000 1001110001 00--00000 1000000000 10-0100110 10001000- --

Gonyleptes0000000100 0000-00000 1010-00000 -0001-000 0010000001 1000000001 0000000010 01--012011 0101010001 00112000-0 000000-000 00--0013-1 000001000 010001C00- -10--00000 1001110001 010--00000 100000000 10-0100110 10001000- --

Centruroides0000100000 0100-10000 0111-00000 -0000-1000 100--00001 1000000001 0000010010 01--012110 0100-10001 00116000-0 001000-000 00--0311-1 0011001000 0005110011 1110001011 1101101000 0010110100 0000100100 10-1100111 0-10010011 00

Hadrurus0000100000 0100-10000 0111-00000 -0000-1000 100--00001 1000000001 0000010010 01--012110 0100-10001 00116000-0 001000-000 00--0311-1 0011001000 0005110011 1110001011 1101101000 0010110111 0000100100 10-1100111 0-10010011 00

Heterometrus0000100000 0100-10000 0111-00000 -0000-1000 100--00001 1000000001 0000010010 01--012110 0100-10001 00116000-0 001000-000 00--0311-1 0011001000 00051100111110001011 1101101000 001020111 0000100100 10-1100111 0-10010011 00

Prearcturus000010000 010 -0 -000- 100--001 000 000-0 00- --0311- 0 0

Palaeoscorpius00000000 0D0 -0 -000- 100--0000 00 00000 0 10-0001 06000-0 000-00 --0311- 1 01 0 0

Stoermeroscorpio00000000 00000 -0 -000- 100--0000 000 00000 0 10-0001 06000-0 000-00 --0311- 110 01 1 0

Table 1 Continued

PHYLOGENY OF THE ARACHNID ORDERS 227

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

the Ordovician to the Permian (Tollerton 1989) Phy-logenetic relationships among the eurypterids havenot been rigorously addressed (but see Beall amp Laban-deira 1990) However members of Stylonuroideaappear to be plesiomorphic in retaining relativelyunspecialized legs and were represented by Stylonu-rus (eg Clarke amp Ruedemann 1912 Waterston1979) In the remaining eurypterids the last leg ispaddle-like (Tollerton 1989) The non-stylonuroideurypterids were represented in the matrix by thewell-preserved and well-studied Baltoeurypterus tet-ragonophthalmus especially as described by Selden(1981)

Chasmataspidida Members of this fossil aquaticgroup (OrdovicianndashDevonian six genera) have a four-segmented mesosoma and a nine-segmented meta-soma (Dunlop Poschmann amp Anderson 2001 DunlopAnderson amp Braddy 2004 Dunlop 2002a) Three gen-era were included Chasmataspis from the monotypicfamily Chasmataspididae as well as Diploaspis andOctoberaspis from Diploaspididae

Haptopoda This is a fossil terrestrial group (Carbon-iferous) containing one known species Plesiosiromadeleyi The known specimens have been re-examined by Dunlop (1999)

Palpigradi The extant palpigrades (two families sixgenera ~50 spp) (Harvey 2002) are a poorly studiedgroup of small-bodied terrestrial (some semi-aquatic)arachnids The two most well-studied generaEukoenenia (Eukoeneniidae) especially E mirabilisand Prokoenenia (Prokoeneniidae) especiallyP wheeleri were included in the matrix The oneknown fossil species Paleokoenenia mordax is rela-tively recent (Pliocene) and to the extent known isfairly similar to extant palpigrades (Rowland amp Sis-som 1980) It was not included in the analysis

Trigonotarbida This is a fossil terrestrial order(Upper Silurian to Lower Permian ~ten families ~50spp) The genera coded here were drawn from twoLower Devonian Lagerstatten Gilboarachne from theGilboa Formation (New York) (Shear et al 1987) and

Proscorpius00000000 00000 -0 -000- 100--0000 010 00000 0 10-0001 06000-0 000-00 --0311- 110 01 1 0

Chthonius0000100000 0310-00120 0000-10100 -1000-0000 100--00100 0000000001 0000010010 01--012100 0E00-00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Neobisium0000100000 0310-00121 0000-10100 -1000-0010 100--00100 0000000001 0000010010 01--012100 0100-00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Feaella0000100000 0310-00120 0000-10100 -1000-0000 100--00100 0000000001 0000010010 01--012100 000--00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Chelifer0000100000 0310-00121 0000-10100 -1000-0010 100--00100 0000000001 0000010010 01--012100 000--00001 01115000-0 000000-000 00--000--1 0000021000 01700-014 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Eremocosta0000110000 0310-00121 0000-00010 -1000-0000 001010100 0000000001 00011-0000 02--010-10 1101010001 01114000-0 000000-000 00--000--1 00000200 13111013 010--10000 1001000000 000--00100 1010000100 01-1100001 0-00000110 00

Galeodes0000110000 0310-00121 0000-00010 -1000-0000 001010100 0000000001 00011-0000 02--010-10 1101010001 01114000-0 000000-000 00--000--1 00000200 13111013 010--10000 1001000000 000--00100 1010000100 011100001 0-00000110 00

Table 1 Continued

228 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Palaeocharinus from the Rhynie Chert (Scotland)(Dunlop 1994 Fayers Dunlop amp Trewin 2004)

Araneae (spiders) This is a very large order (110 fam-ilies ~3600 genera ~39 000 extant spp) (Platnick2005) represented here by one genus from each of thethree principal extant lineages (Platnick amp Gertsch1976 Coddington amp Levi 1991 Coddington 2005)Liphistius (Mesothelae) Aphonopelma (Mygalomor-phae) and Hypochilus (Araneomorphae) The sampleis small with respect to known diversity but the basalphylogeny and relevant groundplan states of the orderare well established and states derived from the rep-resented taxa are consistent with them

Amblypygi (whipspiders) This is a small extant ter-restrial order (five families 17 genera 136 spp)(Harvey 2002) and was represented in the matrix bytwo fairly well-studied genera Charinus and Phry-nus Charinus retains several features that appear tobe plesiomorphic for the order (eg adult pedal pul-villi coxal glands associated with leg 3 eversiblevesicles) Weygoldt (1996) has examined the phylog-eny of the order and has provided an important sum-mary of morphology and general biology (Weygoldt2000)

Thelyphonida or Uropygi ss (whipscorpions vinega-roons) Extant whipscorpions (16 genera 106 spp)(Harvey 2002) were represented here by one well-studied species Mastigoproctus giganteus althoughadditional information was drawn from Typopeltis andThelyphonus The morphology of the group is highlyconserved there have been no modern studies ofintraordinal phylogeny A controversial fossil speciesfrom the late Carboniferous Proschizomus (Dunlop ampHorrocks 19951996) was also included but less well-preserved Carboniferous forms were not

Schizomida This small order (two extant families~34 genera ~200 spp) (Harvey 2002) is widelyregarded as the sister group of Thelyphonida Thetaxon sample included one representative from eachextant family Protoschizomus (Protoschizomidae) andStenochrus (Hubbardiidae) especially S portoricen-sis There are three recent (Pliocene) fossil speciesall from the same locality these were not includedhere

Ricinulei The extant ricinuleids (three genera 55spp) (Harvey 2002) were represented by two generaCryptocellus and Ricinoides Detailed studies of grosscuticular anatomy and post-embryonic developmentare available for a representative of each genus (Cryp-tocellus Pittard amp Mitchell 1972 Ricinoides Legg1976) and some information on internal anatomy isavailable for Ricinoides (Millot 1945) and this hasbeen extrapolated to Cryptocellus in the matrix There

are two basic fossil types (Selden 1992) one resem-bling modern taxa and another with a unique opistho-soma that superficially resembles the closed elytra of abeetle (ie curculioids) Two fossils from the formergroup were included Terpsicroton and Poliochera asthey are reasonably well preserved and show impor-tant characters (eg two pairs of eyes or evidence ofopisthosomal diplosegmentation) not expressed inextant forms No curculioid ricinuleids were includedas they appear to offer no additional information rel-evant to resolving ordinal relationships

Opilioacariformes (= Opilioacarida Notostigmata)The opilioacariform mites (nine genera 20 spp) (Har-vey 2002) are generally regarded as plesiomorphicAcari and are fairly conserved in their morphologyThe group was represented by two species Neocarustexanus and Siamacarus withi N texanus has a typi-cal opilioacariform morphology and its external anat-omy has been particularly well studied (eg Van derHammen 1989 Klompen 2000) S withi differs frommost other opilioacariforms in having trichobothriaand three rather than two pairs of lateral eyes (the lat-ter also in Paracarus) features that are potentiallysignificant for assessing ordinal relationships Someinformation on internal morphology of Opilioacaruswas taken from With (1904) and extrapolated to Neo-carus and Siamacarus

Parasitiformes This large group of mites comprisesthree clades Holothyrida Ixodida (ticks) and Mesos-tigmata (= Gamasida) Holothyrida (three familiesfive genera lt 30 spp) (Walter amp Proctor 1999) wasrepresented here by two species Australothyrus ocel-latus and Allothyrus constrictus Ixodids (three fami-lies 22 genera ~860 spp) (Kierans amp Robbins 1999)were represented by Amblyomma and Argas andmesostigmatids (~70 families ~10 000 spp) were rep-resented by Glyptholaspis confusa

Acariformes (= Actinotrichida) This is a very largediverse group of mites that includes the Endeostig-mata Sarcoptiformes and Prostigmata Endeostig-mata is an apparently paraphyletic assemblage of~ten families A morphologically generalized speciesAlycus roseus (Bimichaeliidae) was included herewith character states extracted largely from Van derHammen (1989) Sarcoptiformes (gt 200 families) iseffectively synonymous with Oribatida with Astig-mata being a large derived lsquooribatidrsquo clade (Norton1998 Maraun et al 2004) The group was representedby an early divergent plesiomorphic genus Palaeac-arus (Palaeosomata) and a more derived and well-studied genus Archegozetes (especially A longisetusand A magnus) The Prostigmata (~120 families~7000 spp) (Walter amp Proctor 1999) was representedby two genera Allothrombium and Microcaeculus

PHYLOGENY OF THE ARACHNID ORDERS 229

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Pseudoscorpiones The pseudoscorpions (24 families425 genera ~3200 spp) (Harvey 2002) were repre-sented in the matrix by four terminal genera Chtho-nius (Chthoniidae) was drawn from the basallydivergent superfamily Chthonioidea (Harvey 1992)Feaella (Feaellidae) represented the enigmatic super-family Feaelloidea which has an unusual combinationof characters and may be an early divergent lineageclose to Chthonioidea (Harvey 1992) Neobisium (Neo-bisiidae) is morphologically intermediate between theearly divergent families and the higher families rep-resented by Chelifer (Cheliferidae)

Solifugae (= Solpugida) (sun spiders) The order issmall (~1000 spp) but the absence of modern phylo-genetic treatments precludes meaningful estimates ofgenera and families (Harvey 2002) The taxon sampleincluded two genera one from the New World Eremo-costa and one from the Old World Galeodes

Opiliones (harvestmen) Extant harvestmen (~25 fam-ilies ~500 genera ~6000 spp) are divided into twosuborders Cyphophthalmi and Phalangida Cyphoph-thalmids were represented here by Chileogovea oedi-pus (Petallidae) which is currently under study by theauthor and Cyphophthamus duricorius (Sironidae)which is probably the most well-studied member of thesuborder (eg Janczyk 1956) Phalangida includesthree major groups Laniatores Eupnoi and Dyspnoibut there is disagreement as to whether Dyspnoi is thesister group to Laniatores or to Eupnoi (Shultz ampRegier 2001 Giribet et al 2002) Laniatores was rep-resented by Sclerobunus and Gonyleptes and Eupnoiby Leiobunum and Caddo

Scorpiones (scorpions) The extant scorpions (16 fam-ilies 155 genera 1279 spp) (Fet et al 2000) wererepresented by three taxa Centruroides vittatus(Buthidae) Hadrurus arizonensis (Iuridae) and Heter-ometrus spinifer (Scorpionidae) Buthidae is widelyregarded as the sister group to other extant lineagesand similarities among the represented terminals arelikely to be ground plan features of extant scorpionsgenerally The morphology of fossil scorpions is sub-stantially more diverse (Kjellesvig-Waering 1986)but their phylogenetic relationships are unclear (butsee Jeram 1998) The fossil taxa used in this study(Proscorpius Stoermeroscorpio Palaeoscorpius Pre-arcturus) were chosen for quality of preservation andor presence of a phylogenetically significant constella-tion of characters

CHARACTER CODING

Most character states were determined from directobservation the primary literature and authoritativereviews In some cases states were assigned to termi-

nals based on observations from related species asnoted in the Appendix In the matrix state lsquo-rsquo indicatesthat the character is inapplicable because the taxonlacks a more general character For example if a taxonlacks eyes then special features of the eyes (numberposition retinal configuration etc) are inapplicable tothat taxon A lsquorsquo indicates that the state is unknown oruncertain An entry with multiple states (eg 01 345) should be treated as an ambiguity code not a poly-morphism it indicates that two or more interpreta-tions of homology are applicable and that assignmentwas established analytically by character concor-dance Although lsquo-rsquo lsquorsquo and lsquo01rsquo are analyticallyidentical when applied to binary characters theyprovide information about the empirical status of thecharacter state in specific taxa Characters are citedthroughout the text as italicized numbers in parenthe-ses and are discussed in the Appendix

PHYLOGENETIC ANALYSIS

Phylogenetic analysis of the full data matrix wasperformed using the program Tree Analysis usingNew Technologies (TNT) ver 1 (Goloboff Farris ampNixon 2000) using lsquotraditionalrsquo search based on1000 replicates using TBR branch swapping Resultswere compared to those obtaining using the ratchetalgorithm (Nixon 1999) to determine any differencedue to analytical approach Nodal support for theminimal-length topology was evaluated by boot-strap (Felsenstein 1985) and Bremer support(Bremer 1994) Bootstrap analysis was conducted inTNT and based on 1000 pseudoreplicates each anal-ysed by ten random-addition replicates using TBRbranch swapping A nexus file containing the result-ing 1000 trees was imported into PAUP ver 10(Swofford 2002) to obtain bootstrap frequenciesBremer support was determined in TNT by con-straining specific nodes in the minimal-length topol-ogy and then determining the shortest tree that didnot recover the specified clade The difference inlength between the unconstrained and constrainedminimal-length trees is the Bremer support Theeffect of homoplasy on results was explored by con-ducting implied weights analysis (Goloboff 1993) inTNT Six analyses were conducted each with con-stant of concavity (k) set to a different integer valueof 1ndash6 where 1 is weighted most severely againsthomoplasious characters Each implied weightsanalysis was conducted using lsquotraditionalrsquo searchbased on 1000 replicates using TBR branch swap-ping The same procedures were used in analysing amatrix that included only extant taxa Howevercharacters rendered uninformative by removal offossil taxa (ie 49 98 99 100 113 114 125 161)were excluded prior to analysis

230 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

COMPARISONS OF ALTERNATIVE HYPOTHESES

Phylogenetic hypotheses proposed in the recent liter-ature were also evaluated including those thatattempted to resolve arachnid phylogeny completely(Fig 2) their hypothesized subclades (Fig 3) andhypotheses that proposed only specific nodes (Figs 34) The fully resolved topologies of Weygoldt amp Paulus(1979) Van der Hammen (1989) Shultz (1990)Wheeler amp Hayashi (1998) and Giribet et al (2002)were compared with the optimal topology using theTempleton test (Templeton 1983) as implementedin PAUP Internal relationships of multisampledorders were constrained to match those of the optimaltopology unless the original authors explicitlyfavoured an alternative Node-specific hypotheseswere evaluated by determining the frequency withwhich the node was recovered in bootstrap analysisThey were also evaluated by constraining parsimonyanalysis in TNT to recover the shortest tree contain-ing each specific node and then assessing the effect onrelative tree length and overall resolution takingnote of effects on otherwise stable or well-supportedclades The entire matrix was used to assess the twocases where node-specific hypotheses involved fossiltaxa (Fig 4)

RESULTS

EXTANT TAXA

Unweighted analysis of extant taxa produced two min-imal-length trees (length 383 CI 057) (Fig 5A) withconflicts limited to relationships within Pseudoscorpi-ones Multiply sampled orders were recovered asmonophyletic with high nodal support as measured bybootstrap percentage (BP gt 80) except Acari (BP 30)There were few well-supported interordinal groupsexcept Uropygi (BP 100) Pedipalpi (BP 100) Tetra-pulmonata (BP 93) and ArachnidaXiphosura (BP100) All other interordinal relationships were recov-ered with BP less than 60

Results from implied weights analysis (IWA) indi-cated a significant effect of homoplasy on resolution ofthe deepest interordinal relationships although theinterordinal clades Stomotheca Haplocnemata Acar-omorpha Tetrapulmonata Pedipalpi and Uropygiwere recovered under all values of k IWA with k = 1 (1tree best score = 5124) recovered Megoperculata(= Palpigradi + Tetrapulmonata) and Micrura(= Megoperculata + Acaromorpha) with Micrura beingthe sister group to Haplocnemata (= Pseudoscorpiones+ Solifugae) and with Stomothecata (= Opiliones +Scorpiones) being the sister group to all other arach-nids (Fig 5B) With k = 2 (1 tree best score = 3807)and k = 3 (1 tree best score = 3054) Megoperculatawas recovered and Stomotheca remained the sister

group to other arachnids but the relative positions ofHaplocnemata and Acaromorpha changed (Fig 5C)IWA using k = 4 (1 tree best score = 2559) k = 5 (1tree best score = 2205) and k = 6 (1 tree bestscore = 1938) recovered topologies consistent with theminimal-length topology derived from unweightedcharacters (Fig 5A)

EXTANT AND FOSSIL TAXA

Unweighted analysis of extant and fossil taxa pro-duced 18 minimal-length trees (length 426 CI 056)(Fig 6A) with conflicts limited to relationships withinPseudoscorpiones Scorpiones and Araneae Multiplysampled orders were again recovered as monophyleticwith high nodal support (BP gt 80) except Acari whichwas reconstructed as diphyletic with Anactinotrichidabeing the sister group to Ricinulei (BP 33) Therewere no well-supported interordinal groupings (ieBP gt 80) except Uropygi (BP 89) Pedipalpi (BP 92)and Arachnida (BP 90)

Results from IWA indicated a significant effect ofhomoplasy on resolution of the deepest interordinalrelationships in Arachnida although StomothecataHaplocnemata Acaromorpha Cryptognomae Pantet-rapulmonata Tetrapulmonata Schizotarsata Pedi-palpi and Uropygi were recovered under all values ofk IWA with k = 1 (7 trees best score = 5675) recov-ered Megoperculata and Micrura with Micrura beingthe sister group to Haplocnemata Stomothecata wasnot recovered in the strict consensus (Fig 6B) Withk = 2 (3 trees best score = 4243) IWA recovered thetopology shown in Fig 6C with Palpigradi being thesister group to all other arachnids The strict consen-sus topologies reconstructed using k = 3ndash6 were iden-tical to those obtained with k = 2 (k = 3 3 trees bestscore = 3419 k = 4 3 trees best score = 2874 k = 53 trees best score = 2483 k = 6 3 trees bestscore = 2188)

COMPARISONS OF ALTERNATIVE HYPOTHESES

The hypothesis of Weygoldt amp Paulus (1979) (Fig 2)was 398 steps (15 steps longer than the unconstrainedminimal-length tree) Van der Hammen (1989) was400 steps (17 steps longer) Shultz (1990) was 399steps (six steps longer) Wheeler amp Hayashi (1998)was 391 steps (eight steps longer) and Giribet et al(2002) was 394 steps (11 steps longer) Templetontests conducted at the 005 significance level rejectedthe hypothesis of no difference between the optimaltopology and those of Weygoldt amp Paulus (P = 00002)Van der Hammen (P = 00011) Wheeler amp Hayashi(P = 00018) and Giribet et al (P = 00343) The Shultztopology was not significantly different (P = 01444)from the optimal topology

PHYLOGENY OF THE ARACHNID ORDERS 231

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 2 Fully resolved phylogenetic hypotheses of extant euchelicerate groups proposed in the recent literature Note thesimilarity in the topologies of the parsimony-based analyses by Shultz (1990) Wheeler amp Hayashi (1998) and Giribet et al(2002) The Giribet et al topology is based on neontological data (morphology and molecules) and the original lsquoROOTrsquo maybe an artefact from use of the highly divergent pycnogonids as an outgroup

Xip

hosu

raS

corp

ione

sO

pilio

nes

Ara

neae

Am

blyp

ygi

Uro

pygi

Ric

inul

eiA

nact

inot

richi

daS

olifu

gae

Pse

udos

corp

ione

sP

alpi

gra

diA

carif

orm

es

Xip

hosu

raS

corp

ione

sA

rane

aeA

mbl

ypyg

iU

ropy

giP

alpi

gra

diS

olifu

gae

Pse

udos

corp

ione

sO

pilio

nes

Ric

inul

eiA

cari

Lipoctena

Arachnida

Apulmonata

Holotracheata

Cryptoperculata

Acaromorpha(ldquoAcarinomorphardquo)

Tetrapulmonata(rdquoMegoperculatardquo)

LabellataHaplo-cnemata

Weygoldt amp Paulus (1979)

Labellata

Tetrapulmonata(rdquoArachnidardquo)

Haplocnemata(rdquoApatellatardquo)

Van der Hammen (1989)

Xip

hosu

raP

alpi

gra

diA

rane

aeA

mbl

ypyg

iU

ropy

giR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Pedipalpi Haplocnemata

Novogenuata

Dromopoda

Shultz (1990)

Acaro-morpha

Micrura

Xip

hosu

raP

alpi

gra

diU

ropy

giA

mbl

ypyg

iA

rane

aeR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Labellata Haplocnemata

Novogenuata

Dromopoda

Wheeler amp Hayashi (1998)

Acaro-morpha

Micrura

Xip

hosu

raA

cari

Pal

pig

radi

Ric

inul

eiA

rane

aeA

mbl

ypyg

iU

ropy

giO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Tetrapulmonata

PedipalpiHaplocnemata

Novogenuata

Dromopoda

Giribet et al (2002)

ROOT

Micrura

Crypto-gnomae

Sternocoxata

Epimerata

RostrosomataMyliosomata

Neosternata

232 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 3 Consensus trees produced by parsimony analysis of neontological data constrained to produce relationships pro-posed in the recent literature The constrained (target) node is indicated by a black dot and the taxa encompassed by theconstraint are enclosed in a box Numbers below each tree represent the number of minimal-length constrained treeslength of minimal-length trees difference in the length of the unconstrained minimal-length tree and the constrained min-imal-length tree and percentage unconstrained bootstrap trees in which the target node was recovered respectively Thesetrees indicate the effect on branch length imposed by specific hypotheses and impact of constraining target nodes on non-target nodes

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Lipoctena4 385 2 lt5

Xi

Sc

Op

Ps

So

Pa

Ur

Am

Ar

Ri

An

Ac

Labellata6 397 14 lt5

Xi

Sc

Ar

Am

Ur

Op

Pa

Ps

So

Ri

An

Ac

Apulmonata2 390 7 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

So

Ri

An

Ac

Holotracheata5 390 7 lt5

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptoperculata2 388 5 lt5

Xi

Sc

Op

Pa

So

Ps

Ar

Am

Ur

Ri

An

Ac

Rostrosomata6 384 1 25

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Neosternata3 388 5 lt5

Xi

Sc

Op

Ar

Am

Ur

Ri

An

Ps

So

Pa

Ac

Coxisternata2 388 5 lt5

Epimerata7 387 4 11

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptognomae6 384 1 24

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Micrura3 385 2 7

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Megoperculata3 384 1 28

Xi

Sc

Ps

So

Op

Pa

Ar

Am

Ur

Ri

An

Ac

Novogenuata10 388 5 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

Ri

So

An

Ac

Solifugae + Acari4 388 5 lt5

Shultz (1990) Alberti amp Peretti (2002)

Van der Hammen (1989)

Weygoldt amp Paulus (1979)

Xi

Ar

Am

Ur

So

Ps

Sc

Op

Ri

Pa

An

Ac

PHYLOGENY OF THE ARACHNID ORDERS 233

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Results from analyses of node-specific hypothesesare summarized in Figures 3 4 In most cases pro-posed relationships were recovered in fewer than 5of bootstrap pseudoreplicates The notable exceptionswere Rostrosomata (25) Cryptognomae (24) andMegoperculata (28) and the strict-consensus con-strained minimal-length tree for each was only onestep longer than the unconstrained optimal tree Inmost cases constrained topologies did not add manysteps most were less than five steps longer and nonewas greater than 14 steps longer than the un-constrained minimal-length tree It is noteworthyhowever that even constrained nodes that imposerelatively few extra steps sometimes supportedimprobable relationships or eliminated clades thatwere well supported or stable in the unconstrainedanalysis Specific examples and their implications arediscussed below

DISCUSSION

RECENT ISSUES IN ARACHNID PHYLOGENY

Available morphological evidence consistently re-solves some interordinal relationships and fails to

resolve others (Fig 1) Continued progress depends onthe ability of arachnologists to discover new charac-ters and to assess the evidence critically Whilereviewing the recent arachnological literature severalaspects of phylogenetic practice emerged that seemedcounterproductive to both the perception and theactual rate of progress toward resolving arachnid phy-logeny Some of these are summarized here withspecific examples given in the remainder of theDiscussion and in the Appendix

There is a tendency to portray arachnid ordinal phy-logeny as more poorly resolved and contentious than isactually the case (Coddington et al 2004) Phyloge-netic hypotheses generated during different historicalperiods and using differing standards of evidence areoften cited as examples of current disagreement (egSelden 1993 Dunlop 1996 Selden amp Dunlop 1998Wheeler amp Hayashi 1998) In fact recent parsimony-based analyses of morphology have tended to convergeon topologies with internal structures congruent withthose found here (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) (Fig 2) Arachnidphylogeny is not fully resolved especially at its deep-est levels (Fig 1) but this does not mean that allaspects of arachnid phylogeny are controversial orpoorly supported by the available evidence

Matrices are sometimes constructed by uncriticallsquorecyclingrsquo of erroneous or problematic charactersbased on diverse secondary sources (Jenner 2001)Conclusions derived from mixtures of valid invalidspeculative and redundant characters are sometimesportrayed as the phylogenetic signal provided by mor-phology Data recycling can perpetuate errors (see 1320 39 52 136 140 144 152 169 171ndash174) legiti-mize speculations (see 13 32 63 95) or create dupli-cate or non-independent characters (see 13 30 70 77172) The lsquolateral organrsquo is a notable exampleYoshikura (1975) equated the embryonicearly post-embryonic lsquolateral organsrsquo of Amblypygi Thelyphonidaand Solifugae with the dissimilar lsquolateral organrsquo ofXiphosura but failed to note the very similar Cla-paregravede organ of Acariformes (see 173 174) This cod-ing was recycled by Wheeler amp Hayashi (1998) andthen by Giribet et al (2002) Error is probably inevi-table when assembling a large matrix from morphol-ogy including the one presented here but this can beminimized by making original observations and byconsulting primary sources

Some workers advocate weighting characters a pri-ori on the basis of structural or functional complexity(eg Kraus 1998 Dunlop amp Braddy 2001) and dismissphylogenetic conclusions derived from equal-weightsparsimony In short these workers criticize parsimonyfor emphasizing data quantity over data quality Thiscriticism ignores the intense debate in systematic biol-ogy that eventually led to widespread adoption of

Figure 4 Consensus trees produced by parsimony analy-sis of the full data matrix constrained to produce relation-ships proposed in the recent literature See legend toFigure 3 for details

Xi

Eu

Sc

Op

Pa

Ac

Ri

An

So

Ps

Tr

Ar

Ha

Am

UrEurypterida slat + Scorpiones12 435 9 lt5

Xi

Eu

Sc

Op

Pa

Ac

An

So

Ps

Ri

Tr

Ar

Ha

Am

UrTrigonotarbida+ Ricinulei28 434 8 lt5

Dunlop amp Braddy (2002)et al

Dunlop (1996)

234 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 5 Results from analysis of neontological data A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology AFor B and C relationships within terminal clades are the same as those shown in A

LimulusTachypleusProkoeneniaEukoeneniaCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAllothyrusAustralothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesLiphistiusAphonopelmaHypochilusCharinusPhrynusStenochrusProtoschizomusMastigoproctusCyphophthalmusChileogoveaCaddoLeiobunumSclerobunusGonyleptesCentruroidesHadrurusHeterometrusChthoniusFeaellaNeobisiumCheliferEremocostaGaleodes

10031

10018451

501561

733

723

944

633

10010

832

901

943

903

522

341

1007

1008

Acaromorpha

Haplocnemata

10012

Stomothecata

Opiliones

Scorpiones

Pseudo-scorpiones

Solifugae

Acariformes

Parasitiformes

Opilioacariformes

Ricinulei

Xiphosura

Palpigradi

Thelyphonida

Schizomida

Araneae

Amblypygi

10011

936853

100141005

1009

Tetrapulmonata Pedipalpi

Uropygi

692

522572

968

331

603

81

511

161

101

Acari

Dromopoda

Arachnida

A

Xiphosura

Stomothecata

Haplocnemata

Acaromorpha

Palpigradi

TetrapulmonataMicrura

Megoperculata

BXiphosura

Stomothecata

Acaromorpha

Haplocnemata

Palpigradi

TetrapulmonataMegoperculata

C

PHYLOGENY OF THE ARACHNID ORDERS 235

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 6 Results from analysis of the full data matrix A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology CFor B and C relationships within terminal clades are the same as those shown in A

WeinberginaLimuloidesEuproopsTachypleusLimulusStylonurusBaltoeurypterusChasmataspisDiploaspisOctoberaspisCyphophthamusChileogoveaCaddoLeiobunumSclerobunusGonyleptesProscorpiusPalaeoscorpiusStoermeroscorpioPrearcturusCentruroidesHadrurusHeterometrusProkoeneniaEukoeneniaTerpsicrotonPoliocheraCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAustralothyrusAllothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesEremocostaGaleodesChthoniusFeaellaNeobisiumCheliferGilboarachnePalaeocharinusLiphistiusAphonopelmaHypochilusPlesiosiroCharinusPhrynusStenochrusProtoschizomusProschizomusMastigoproctus

631

893

923

984

984

671

341

671

986

10011

423

974

Xiphosura

Eurypterida sstr

Chasmataspidida

Opiliones

Scorpiones

Palpigradi

Ricinulei

Opilioacariformes

Parasitiformes

Acariformes

Solifugae

Pseudoscorpiones

Trigonotarbida

Araneae

Haptopoda

Amblypygi

Schizomida

Thelyphonida

842

251331

361311

552

832

882

925

913

573

561291

443

976

541

954

743644

732521

571

331

833

663591

966

573

71

221

905

832

351

Uropygi

Pedipalpi

Haplocnemata

Tetrapulmonata

Anactinotrichida

Cryptognomae

Acaromorpha

Stomothecata

Pantetrapulmonata

Arachnida

Eurypterida slat

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatProscorpiusStoermeroscorpioPalaeoscorpiusPrearcturusCentruroidesHadrurusHeterometrusOpilionesHaplocnemataRicinuleiAnactinotrichidaAcariformesPalpigradiPantetrapulmonata

Megoperculata

Micrura

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatPalpigradiStomothecataHaplocnemataPantetrapulmonataAcaromorpha

A

B

C

Schizotarsata

236 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

parsimony and instead advocates a return to thespeculative and subjective approaches of the late 19thand early 20th century that once threatened thescientific legitimacy of the discipline (Bowler 1996)Furthermore it misrepresents the properties of parsi-mony-based analyses Specifically characters withfunctional significance are readily encompassed byparsimony analyses and in fact characters derivedfrom locomotor systems (eg 46ndash94) have played animportant role in developing current ideas aboutarachnid phylogeny (eg Shultz 1989 1991) In addi-tion lsquocomplex charactersrsquo can be viewed as compositesof several characters such that morphological com-plexity is effectively weighted by the number of inde-pendent lsquosubcharactersrsquo it contains For example thelsquosucking stomachrsquo once considered a synapomorphy ofLabellata (= Araneae + Amblypygi Fig 3) is codedhere as a composite of three characters (199ndash201) TheLabellata hypothesis was not corroborated in thepresent analysis (Figs 1 3ndash5) and in fact was highlydisfavoured (Fig 3) but this result cannot be dis-missed as a failure to acknowledge the complexity ofthe character

Some workers support specific (target) clades withone or more similarities without exploring the impacton overall homoplasy or relationships among non-target clades Each character offers its own phyloge-netic hypothesis which may or may not be consistentwith relationships implied by other characters It isexceedingly rare for all characters to be perfectly com-patible in the phylogenetic hypotheses they supportand criteria such as parsimony have been developed todiscover those hypotheses that minimize the conflict-ing phylogenetic signals of different characters Eventhough character conflict (homoplasy) is a virtuallyinescapable phenomenon in comparative biology it isnot uncommon for workers to discover one or morecharacters and to promote their phylogeneticimplications over alternative hypotheses even thosethat otherwise appear to be well supported Thisapproach may have value in highlighting new dataand perspectives but accomplishes this by promotingthe erroneous impression that all aspects of arachnidphylogeny are so tenuous that a single character canfalsify even well-supported hypotheses Several exam-ples of this approach have appeared in the recentarachnological literature

For example Alberti amp Peretti (2002) argued thataflagellate spermatozoa (163) are a compelling syna-pomorphy for a Solifugae + Acari clade and dismissedsome of the characters that support Solifugae + Pseu-doscorpiones as having lsquodebatable valuersquo Yet theirproposal rejects two hypotheses that are consistentlyrecovered in recent phylogenetic analyses that isHaplocnemata (= Solifugae + Pseudoscorpiones) andAcaromorpha (= Acari + Ricinulei) (Figs 1 2) Rejec-

tion of Haplocnemata would require its presumed syn-apomorphies to be reinterpreted as homoplasiesincluding features of the chelicerae (18 19 20) pre-oral chamber (13 32) legs (12 48) and respiratorysystem (126) The same reason applies to Acaromor-pha and its synapomorphies It is noteworthy that aSolifugae + Acari clade was recovered in the presentanalysis in fewer than 5 of bootstrap pseudorepli-cates and that analyses constrained to recover thisclade were five steps longer than the minimal-lengthtree and favoured a problematic clade unitingRicinulei and Pseudoscorpiones (Fig 3)

In another example Dunlop (1996) proposed a closerelationship between Trigonotarbida and Ricinuleibased on two-segmented chelicerae (18) prosomandashopisthosoma coupling mechanism (96) diplotergites(100 101) and longitudinally divided opisthosomaltergites (115) However phylogenetic analyses con-strained to recover this relationship required eightadditional steps eliminated support for Acari and nec-essarily rejected Acaromorpha and forced its synapo-morphies to be reinterpreted as homoplasies Alberti(2005) proposed an interesting hypothesis for the evo-lution of male gonads in tetrapulmonates (153) butchose to accept Labellata (= Araneae + Amblypygi) indeveloping his argument over the much more well-supported Pedipalpi (= Amblypygi + Uropygi) (Shultz1999) a phylogenetic reconfiguration that was recov-ered here in fewer than 5 of bootstrap pseudorepli-cates increased tree length by a minimum of 14 stepsand resulted in the collapse of Stomothecata and Hap-locnemata (Fig 3) Many other examples can be cited

Promoting or defending a specific phylogenetichypothesis via lists of compatible synapomorphies is acommon but problematic approach By restrictingattention to the states of specific characters at one ortwo target nodes one can easily overlook the unin-tended impact of the hypothesis on phylogenetic sig-nal elsewhere and its effect on non-target clades Anode supported by a long list of synapomorphies mayseem convincing taken in isolation but may becomeless acceptable when its full phylogenetic implicationsare explored

SUMMARY OF INTERORDINAL ARACHNID CLADES RECOVERED IN THIS ANALYSIS

Arachnida Lamarck 1801Analyses consistently recovered Arachnida as a mono-phyletic group with high bootstrap support (Figs 1 45) Possible synapomorphies include the loss of thecarapacal pleural doublure (9) cardiac lobe (10) pedalgnathobases (52) and moveable endites (53) and thegain of aerial respiration (120) and an anteriorly oranteroventrally directed mouth (185) Some tradi-tional synapomorphies such as slit sensilla (142) and

PHYLOGENY OF THE ARACHNID ORDERS 237

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

fluid feeding (184) may have appeared later in arach-nid evolution but this can only be decided once theinternal phylogeny of Arachnida has been determined

Some workers regard many proposed arachnid syn-apomorphies as adaptations to terrestrial life andthereby link the hypothesis of arachnid monophyly tothe hypothesis of a single ancestral aquatic-to-terres-trial transition and arachnid polyphyly to multipletransitions (eg Selden amp Jeram 1989 Dunlop 1997Dunlop amp Webster 1999 Dunlop amp Braddy 2001)Given this line of reasoning the existence of appar-ently aquatic scorpions in the fossil record (Kjellesvig-Waering 1986 Jeram 1998) and the inference thatterrestrialization occurred late in scorpion evolutionthese workers conclude that character states support-ing arachnid monophyly are actually convergencesand do not necessarily support arachnid monophyly

However this approach to assessing phylogenetichypotheses is founded on overly simplistic assump-tions such as the ability of the investigator to discrim-inate unerringly between characters that existexclusively in aquatic organisms (including fossils)from those that occur exclusively in terrestrial organ-isms There also appears to be an assumption thathomoplasy can be generated through parallelism (iemultiple aquatic-to-terrestrial events) but not throughterrestrial-to-aquatic reversals Furthermore anassumed dichotomy between exclusively aquatic andterrestrial life histories in ancestral arachnids is sim-plistic as illustrated by the amphibious life cycles ofbasal vertebrates and pterygote hexapods In factthese examples demonstrate that there is no neces-sary inconsistency in basing a hypothesis of arachnidmonophyly on the derived terrestrial features of anamphibious ancestor whose descendants thencompleted terrestrialization once or several timesindependently or even returned to a fully aquaticexistence Workers who link the frequency and direc-tion of aquaticndashterrestrial transitions to the assess-ment of arachnid phylogeny do so by endowingthemselves with substantially greater insight thanseems prudent by ignoring the huge gaps in ourunderstanding of early arachnid evolution and bydenying to arachnids the evolutionary complexityknown to exist in other groups

Several palaeontologists have been particularlyactive during the past decade in proposing new char-acters with the stated goal of removing Scorpionesfrom Arachnida and erecting a Scorpiones + Eurypter-ida clade (Braddy amp Dunlop 1997 Dunlop amp Braddy1997 Dunlop 1998 Braddy et al 1999 Dunlop ampWebster 1999) Dunlop amp Braddy (2001) recentlysummarized this evidence and conducted aparsimony-based analysis of Xiphosura EurypteridaScorpiones Opiliones and Tetrapulmonata (but notHaplocnemata Acaromorpha or Palpigradi) using 33

morphological characters Their analysis produced atopology congruent with those generated here (Figs 15 6) including recovery of Stomothecata (= Scorpiones+ Opiliones) However they rejected this result as aproduct of lsquoempirical cladisticsrsquo because it gives thesame weight to prosomal characters that supportarachnid monophyly and to selected opisthosomalcharacters that support their favoured Eurypterida +Scorpiones clade namely a five-segmented metasoma(116) suppression of opisthosomal tergite 1 (95) lossof lamellate respiratory organs on the postgenitalsomite (122) Kiemenplatten (125) loss of respiratorylamellae on the genital segment (121) and a lsquonon-stainingrsquo exocuticle (but see Grainge amp Pearson 1966for evidence of this in Opiliones see Appendix for com-ments on the other characters)

A Eurypterida + Scorpiones clade was not favouredin the present analysis (Fig 6) the strict consensus ofminimal-length trees constrained to recover this clade(Fig 4) is nine steps longer than the tree recovered byanalysis without this constraint It is also noteworthythat Opiliones was consistently reconstructed as thesister group to Eurypterida + Scorpiones in the mini-mal-length constrained trees a provocative result thatwas probably unintended and unanticipated by sup-porters of the Eurypterida + Scorpiones Given thatDunlop amp Braddy (2001) reject equal weights parsi-mony as an arbiter of phylogenetic hypotheses theywould presumably dismiss these results as irrelevantto their argument just as they dismissed their ownparsimony-based results However if the Eurypterida+ Scorpiones hypothesis is to be credible it must beopen to evaluation and potential falsification usingobjective criteria and the subjective or intuitive a pri-ori weighting of characters advocated by Dunlop ampBraddy clearly does not qualify At present it isunclear how one would objectively evaluate Dunlop ampBraddyrsquos proposal with criteria compatible with thoseused in its original formulation For now the Euryp-terida + Scorpiones concept advocated by Dunlop ampBraddy may persist outside the mainstream of mod-ern systematic practice but it is increasingly problem-atic within it

Stomothecata nomnovOpiliones and Scorpiones were consistently recoveredas a monophyletic group The proposed name acknowl-edges a unique preoral chamber the stomothecaformed by coxapophyses of the palp and leg 1 (50)often with an auxiliary role played by the coxapophy-sis of leg 2 (51) In addition the epistome appears tohave been modified for adduction of the palpal coxaeThe lateral walls of the epistome are fused to themedial surfaces of the palpal coxae and the epistomallumen is spanned by a transverse muscle (188) which

238 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

apparently adducts the palpal coxae thereby constrict-ing the stomothecal chamber Scorpions and opilionsare also unique in having a pair of large epistomalarms projecting rearward into the prosoma andattaching to the endosternite (189) The epistomalarms provide attachment sites for pharyngeal dilatormuscles (196) and extrinsic muscles of anterior proso-mal appendages (eg 37) The chelicera is equippedwith a muscle that arises on the carapace and insertson the ventral margin of the second cheliceral article(23) There is an anteriorly placed genital opening(155)

Several notable similarities were found whilereviewing the literature but information from otherarachnid groups was considered too incomplete toallow them to be included in the matrix For exampleboth orders have apparent haemocytopoeitic organsassociated with major nerves of the anterior opistho-soma These are termed supraneural organs in scorpi-ons (Farley 1999) and perineural organs in opilions(Kaumlstner 1935) Haemocytes develop in the cardiacwall in spiders (Seitz 1972) and perhaps amblypygids(Weygoldt 2000) but haemocytopoeitic organs areunknown in most other arachnid groups Germ cellsdifferentiate very early during embryogenesis in bothscorpions and opilions relative to spiders (Moritz1957 Anderson 1973) Additional research is requiredto determine the phylogenetic utility of thesecharacters

Haplocnemata Boumlrner 1904Several workers recognize a close phylogenetic rela-tionship between Pseudoscorpiones and Solifugae(Weygoldt amp Paulus 1979 Van der Hammen 1989Shultz 1990 Wheeler amp Hayashi 1998 Giribetet al 2002) and this was consistently recovered inthe present analysis Haplocnemata is supported byseveral synapomorphies including (i) a feeding com-plex formed by two-segmented chelicerae (18) with aventrolateral intrinsic articulation (19) and perhapsa dorsolateral articulation with the carapace (20)and (ii) a rostrosoma (32) a preoral apparatus formedby an anteriorly projecting epistome affixed dorsallyto enlarged palpal coxae and bordered by lateral pal-pal projections The epistome forms the dorsal compo-nent of a lsquobeakrsquo with the ventral part formed by amidventral sternapophysis (13) fused between thepalpal coxae The locomotor apparatus consists ofcoxae that meet along the ventral midline (12) elon-gate femur-like patellae (48) and an apotele in theadult modified as an eversible empodium or pulvillus(92) The respiratory system is formed entirely (Pseu-doscorpiones) or in part (Solifugae) by paired trachealstigmata opening on opisthosomal somites 3 and 4(126)

Acaromorpha Dubinin 1957A clade comprising Ricinulei and Acari is consistentlyrecovered in the present analysis It is united here bytwo unique and seven homoplasious synapomorphiesThese include a gnathosoma (31) defined in part bymedial fusion of the palpal coxae (30) although pres-ence of a gnathosoma in Ricinulei is debatable as dis-cussed in the Appendix Acaromorphs also have aunique post-embryonic development consisting of ahexapodal larva and up to three octopodal nymphalinstars (176) The pedal patellandashtibia joints are formedby a bicondylar dorsal hinge rather than a single mid-dorsal condyle (69) all postcheliceral segmental gan-glia are unified in the subesophageal ganglion (130)and a postcerebral pharynx is absent (192) The groupis also tentatively united by presence of differentiatedpedal basi- and telofemora (63 64) and the absence ofa ventral (sternal) pharyngeal dilator muscle (199)but these may be symplesiomorphies erroneouslyreconstructed as a synapomorphies

The internal phylogenetic structure of Acaromorphais controversial with many recent workers favouringa monophyletic Acari (Weygoldt amp Paulus 1979Shultz 1990 argued most thoroughly by Lindquist1984) and others advocating a diphyletic Acari withAcariformes being the sister group to Cryptognomae(= Ricinulei + Anactinotrichida) (especially Van derHammen 1979 1989) (Fig 3) Acari was recovered asmonophyletic when fossils were excluded (Fig 2) butnot when they were included (Fig 3) A lsquomite-centredrsquosurvey of arachnid characters may be needed if mor-phology is to offer a compelling solution to the internalphylogeny of Acari and its placement within Arach-nida These issues are not resolved by the presentanalysis

Pantetrapulmonata nom novPantetrapulmonata includes the extinct orders Trigo-notarbida Haptopoda and the extant orders AraneaeAmblypygi Schizomida and Thelyphonida The cladeis united by cheliceral structure (18 19) a megoper-culum (106) booklungs on the genital and first post-genital somites (121 122) and enlargement of theepipharyngeal sclerite (192) It is important to notehowever that most of these characters were coded asuncertain in Plesiosiro (Haptopoda) Aside from theplacement of Haptopoda the monophyly of Trigono-tarbida and the extant orders was anticipated byShear et al (1987) and is generally regarded as amonophyletic group

Tetrapulmonata Shultz 1990Tetrapulmonata was originally proposed on the basisof neontological analyses and encompassed Araneae

PHYLOGENY OF THE ARACHNID ORDERS 239

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Amblypygi Schizomida and Thelyphonida (Shultz1990) and was also recovered here in analysis of neon-tological data (Fig 5) However results from analysisof all taxa required that the Tetrapulmonata conceptbe expanded to include the fossil order HaptopodaFeatures uniting this group are problematic howeveras many states in Haptopoda were coded as uncertain

Schizotarsata nomnovHaptopoda and Pedipalpi (= Amblypygi + Schizomida+ Thelyphonida) are united here in a group named forpossession of divided pedal telotarsi (84) Synapomor-phies include a pointed anterior carapacal margin (3)and elongation of leg 1 (46) As already noted theplacement of Haptopoda should probably be regardedas tentative because the state of many characters inthis group was coded as unknown The placement ofHaptopoda as the sister group to Pedipalpi was antic-ipated by Dunlop (1999 2002c)

Pedipalpi Boumlrner 1904Pedipalpi encompasses Amblypygi and Uropygi(= Thelyphonida + Schizomida) Although widely rec-ognized in the past Pedipalpi was set aside in the mid-20th century in favour of Labellata (= Amblypygi +Araneae) based largely on the presumed synapomor-phies of a narrow prosomandashopisthosoma juncture (97)and sucking stomach (199ndash201) The situation beganto reverse when Shear et al (1987) again highlightedsimilarities of Amblypygi and Uropygi such as rapto-rial palps (35) and antenniform leg 1 (46) Shultz(1989 1990 1999) subsequently described numerousderived similarities including asymmetrical flexormuscles at the pedal femurndashpatella joint (70) threetelotarsomeres on legs 2ndash4 (85) and modification ofthe palpal coxae to provide support for extrinsic pha-ryngeal muscles (197) Pedipalpi has emerged as oneof the most well-supported interordinal relationshipsin Arachnida

Uropygi Thorell 1882Uropygi has long been accepted as a monophyleticunion of Thelyphonida and Schizomida Synapomor-phies include a unique mating behaviour (159) fusedpalpal coxae (30) 2-1-1-1 arrangement of tibial tricho-bothria (144) posterior defensive glands (102) elon-gated patella of leg 1 (68) and many others

ACKNOWLEDGEMENTS

I thank Nikolaj Scharff and three anonymous review-ers for useful comments This work was supported bythe Maryland Agricultural Experiment Station the

General Research Board of the University of Mary-land and National Science Foundation Grant 0416628to Susan Masta

REFERENCES

Abd el-Wahab A 1952 Notes on the morphology of the scor-pion Buthus quinquestriatus (HE) Publications de lrsquoInsti-tut Fouad 1er du Desert 3 1ndash129

Alberti G 1979 Licht- und elektronenmikroskopische Unter-suchungen an Coxaldruumlsen von Walzenspinnen (ArachnidaSolifugae) Zoologischer Anzeiger 203 48ndash64

Alberti G 1995 Comparative spermatology of Cheliceratareview and perspective Memoires du Museum NationaldrsquoHistoire Naturelle 166 203ndash230

Alberti G 2005 Double spermatogenesis in ChelicerataJournal of Morphology 266 281ndash297

Alberti G Coons LB 1999 Acari mites In Harrison FWFoelix RF eds Microscopic anatomy of invertebratesVol 8C Chelicerate Arthropoda New York Wiley Liss515ndash1215

Alberti G Peretti AV 2002 Fine structure of male genitalsystem and sperm in Solifugae does not support a sister-group relationship with Pseudoscorpiones (Arachnida)Journal of Arachnology 30 268ndash274

Alexander AJ 1967 Problems of limb extension in the scor-pion Opisthophthalmus latimanus Koch Transactions of theRoyal Society of South Africa 37 165ndash181

Anderson DT 1973 Embryology and phylogeny in annelidsand arthropods Oxford Pergamon Press

Anderson LI Selden PA 1997 Opisthosomal fusion andphylogeny of Palaeozoic Xiphosura Lethaia 30 19ndash31

Beall BS Labandeira CC 1990 Macroevolutionary patternsof the Chelicerata and Tracheata In Culver SJ ed Arthro-pod paleobiology Short Courses in Paleontology No 3 Lon-don The Paleontological Society 257ndash284

Bernard HM 1896 The comparative morphology of the Gale-odidae Transactions of the Linnean Society of London Sec-ond Series 6 305ndash417

Boumlrner C 1902a Arachnologische Studien (II und III) Zool-ogischer Anzeiger 25 433ndash466

Boumlrner C 1902b Arachnologische Studien IV Die Genitalor-gane der Pedipalpen Zoologischer Anzeiger 26 81ndash92

Boumlrner C 1904 Beitraumlge zur Morphologie der Arthropoden IEin Beitrag zur Kenntnis der Pedipalpen Zoologica (Stut-tgart) 42 1ndash174

Bowler PJ 1996 Lifersquos splendid drama Chicago Universityof Chicago Press

Braddy SJ Alridge RJ Gabbott SE Theron JN 1999Lamellate book-gills in a late Ordovician eurypterid from theSoom Shale South Africa support for a eurypterid-scorpionclade Lethaia 32 72ndash74

Braddy SJ Dunlop JA 1997 The functional morphology ofmating in the Silurian eurypterid Baltoeurypterus tet-ragonophthalmus (Fischer 1839) Zoological Journal of theLinnean Society 121 435ndash461

Brauer A 1895 Beitrage zur Kenntnis der Entwicklungsge-schichte des Skorpiones II Zeitschrift fuumlr WissenschaftlicheZoologie 59 351ndash433

240 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bremer K 1994 Branch support and tree stability Cladistics10 295ndash304

Bristowe WS 1932 The liphistiid spiders Proceedings of theZoological Society of London Part 4 1016ndash1057

Butt AG Taylor HH 1991 The function of spider coxalorgans effects of feeding and salt-loading on Porrhotheleantipodiana (Mygalomorpha Dipluridae) Journal of Exper-imental Biology 158 439ndash461

Buxton BH 1913 Coxal glands of arachnids ZoologischeJahrbucher (Suppl XIV) 16ndash50

Buxton BH 1917 Notes on the anatomy of arachnids Jour-nal of Morphology 29 1ndash31

Camin JH Clark GM Bourdeau FG 1956 The palpallsquotined setarsquo in the Mesostigmata a homologue of the palpalclaw in the Onychopalpida (Acarina) In Becker EC ed Pro-ceedings of the Tenth International Congress of EntomologyOttawa Mortimer 903ndash908

Chamberlin JC 1931 The arachnid order ChelonethidaStanford University Stanford University Press

Chapman RF 1998 The insects structure and function 4thedn Cambridge Cambridge University Press

Clarke KU 1979 Visceral anatomy and arthropod phylogenyIn Gupta AP ed Arthropod phylogeny New York Van Nos-trand Reinhold 467ndash549

Clarke J 1984 On the relationship between structure andfunction in the leg joints of Heteropoda venatoria (L) (Ara-neae Eusparassidae) Bulletin of the British ArachnologicalSociety 6 181ndash192

Clarke J 1986 The comparative functional morphology of theleg joints and muscles of five spiders Bulletin of the BritishArachnological Society 7 37ndash47

Clarke JM Ruedemann R 1912 The Eurypterida of NewYork New York State Museum Memoir 14 (1ndash2) 1ndash628

Coddington JA 2005 Phylogeny and classification of spi-ders In Ubick D Paquin P Cushing PE Roth V eds Spi-ders of North America an identification manual AmericanArachnological Society 18ndash24

Coddington JA Giribet G Harvey MS Prendini LWalter DE 2004 Arachnida In Cracraft J Donoghue MJeds Assembling the tree of life Oxford Oxford UniversityPress 269ndash318

Coddington JA Levi HW 1991 Systematics and evolutionof spiders (Araneae) Annual Review of Ecology and System-atics 22 565ndash592

Coineau Y Van der Hammen L 1979 The postembryonicdevelopment of Opilioacarida with notes on new taxa and ona general model for the evolution In Piffle E ed Proceed-ings of the 4th International Congress of Acarology Budap-est Akadeacutemiai Kiadoacute 437ndash441

Cokendolpher JC Reddell JR 1992 Revision of the Pro-toschizomidae (Arachnida Schizomida) with notes on thephylogeny of the order Texas Memorial Museum Speleolog-ical Monographs 3 31ndash74

Condeacute B 1991a Le dimorphisme sexuel des Palpigrades Bul-letin de la Socieacuteteacute Neuchacircteloise des Sciences Naturelles 11667ndash73

Condeacute B 1991b Lrsquoaire genitale male des Palpigrades Bulle-tin de la Socieacuteteacute Europeenne drsquoArachnologie 1990 64ndash69

Dearden PK Donly C Grbicccc M 2002 Expression of pair-rule gene homologues in a chelicerate early patterning ofthe two-spotted spider mite Tetranychus urticae Develop-ment 129 5461ndash5472

Delle Cave L 1975 The ldquocleaning brushrdquo in Damon diademaSimon 1876 and Phrynichus jayakari Pocock 1894 (Ambly-pygi Arachnida) Monitore Zoologico Italiano (NS) 9 175ndash184

Dunlop JA 1994 Filtration mechanisms in the mouthparts oftetrapulmonate arachnids (Trigonotarbida Araneae Ambly-pygi Uropygi Schizomida) Bulletin of the British Arachno-logical Society 9 267ndash273

Dunlop JA 1996 Evidence for a sister group relationshipbetween Ricinulei and Trigonotarbida Bulletin of the BritishArachnological Society 10 193ndash204

Dunlop JA 1997 Palaeozoic arachnids and their significancefor arachnid phylogeny In Zabka M ed Proceedings of the16th European colloquium of arachnology Siedlce PolandWyzsza Szkola Rolnicko-Pedagogiczna 65ndash82

Dunlop JA 1998 The origins of tetrapulmonate book lungsand their significance for chelicerate phylogeny In SeldenPA ed Proceedings of the 17th European Colloquium ofArachnology Edinburgh 1997 Burnham Beeches BritishArachnological Society 9ndash16

Dunlop JA 1999 A redescription of the Carboniferous arach-nid Plesiosiro madeleyi Pocock 1911 (Arachnida Hapto-poda) Transactions of the Royal Society of Edinburgh EarthSciences 90 29ndash47

Dunlop JA 2000 The epistomo-labral plate and lateral lips insolifuges pseudoscorpions and mites Ekoloacutegia (Bratislava)19 67ndash78

Dunlop JA 2002a Arthropods from the Lower Devonian Sev-ernaya Zemlya Formation of October Revolution Island(Russia) Geodiversitas 24 349ndash379

Dunlop JA 2002b Character states and evolution of the che-licerate claws In Toft S Scharff N eds European Arach-nology 2000 Aarhus Aarhus University Press 345ndash354

Dunlop JA 2002c Phylogeny of Chelicerata In LlorenteBousquets J Morrone JJ eds Biodiversidad Taxonomiacutea yBiogeografiacutea de Arthroacutepodos de Meacutexico Hacia Una Siacutentesisde Su Conocimiento Vol III Prensas de Ciencias Univer-sidad Nacional Autonoma de Meacutexico 117ndash141

Dunlop JA Anderson LI Braddy SJ 2004 A redescriptionof Chasmataspis laurencii Caster amp Brooks 1956 (Chelicer-ata Chasmataspidida) from the Middle Ordovician of Ten-nessee USA with remarks on chasmataspid phylogenyTransactions of the Royal Society of Edinburgh Earth Sci-ences 94 207ndash225

Dunlop JA Braddy SJ 1997 Slit-like structures on theprosomal appendages of the eurypterid BaltoeurypterusNeues Jahrbuch fuumlr Geologie und Palaumlontologie-Monatshefte1997 31ndash38

Dunlop JA Braddy SJ 2001 Scorpions and their sister-group relationships In Fet V Selden PA eds Scorpions2001 in Memoriam Gary A Polis London British Arachno-logical Society 1ndash24

Dunlop JA Horrocks CA 19951996 A new Upper Carbon-iferous whip scorpion (Arachnida Uropygi Thelyphonida)

PHYLOGENY OF THE ARACHNID ORDERS 241

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

with a revision of the British Carboniferous Uropygi Zoolo-gischer Anzeiger 234 293ndash306

Dunlop JA Poschmann M Anderson L 2001 On theEmsian (early Devonian) arthropods of the Rhenish SlateMountains 3 The chasmataspidid Diploaspis Palaumlon-tolgische Zeitschrift 75 253ndash269

Dunlop JA Webster M 1999 Fossil evidence terrestrializa-tion and arachnid phylogeny Journal of Arachnology 27 86ndash93

Edgecombe GD Wilson GDF Colgan DJ Gray MR Cas-sis G 2000 Arthropod cladistics combined analysis of his-tone H3 and U2 snRNA sequences and morphologyCladistics 16 155ndash203

Evans GO 1992 Principles of acarology Wallingford UKCABI

Fahrenbach WH 1999 Merostomata In Harrison FW Foe-lix RF eds Microscopic anatomy of invertebrates Vol 8AChelicerate arthropods New York Wiley-Liss 21ndash115

Farley RD 1999 Scorpiones In Harrison FW Foelix RF edsMicroscopic anatomy of invertebrates Vol 8A Cheliceratearthropods New York Wiley-Liss 117ndash222

Farley RD 2005 Developmental changes in the embryo pro-nymph and first molt of the scorpion Centruroides vittatus(Scorpiones Buthidae) Journal of Morphology 265 1ndash27

Fayers SR Dunlop JA Trewin NH 2004 A new earlyDevonian trigonotarbid arachnid from the Windyfield ChertRhynie Scotland Journal of Systematic Palaeontology 2269ndash284

Felsenstein J 1985 Confidence limits on phylogenies anapproach using the bootstrap Evolution 39 783ndash791

Fet V Sissom WD Lowe G Braunwalder ME 2000 Cat-alog of the scorpions of the world (1758ndash1998) New YorkNew York Entomological Society

Firstman B 1954 The central nervous system musculatureand segmentation of the cephalothorax of a tarantula (Eu-rypelma californicum Ausserer) Microentomology 87 1ndash40

Firstman B 1973 The relationship of the chelicerate arterialsystem to the evolution of the endosternite Journal ofArachnology 1 1ndash54

Foelix RF 1985 Mechano- and chemoreceptive sensilla InBarth FG ed Neurobiology of arachnids Berlin Spinger-Verlag 118ndash137

Foelix RF 1996 Biology of spiders 2nd edn Oxford OxfordUniversity Press

Giribet G Edgecombe GD Wheeler WC Babbit C 2002Phylogeny and systematic position of Opiliones a combinedanalysis of chelicerate relationships using morphologicaland molecular data Cladistics 18 5ndash70

Goloboff P 1993 Estimating character weights during treesearch Cladistics 9 83ndash91

Goloboff PA Farris JS Nixon K 2000 TNT Tree AnalysisUsing New Technologies Version 1 Program and documen-tation httpwwwzmucdkpublicphylogenytnt

Grainge CA Pearson RG 1966 Cuticular structure in thePhalangida Nature 211 866

Grandjean F 1938 Observations sur les Bdelles (Acariens)Annales de la Socieacuteteacute Entomologique de France Paris 107 1ndash24

Hansen HJ Soslashrensen W 1904 On two orders of ArachnidaCambridge Cambridge University Press

Hansen HJ Soslashrensen W 1905 The Tartarides a tribe of theorder Pedipalpi Arkiv fuumlr Zoologi 2 1ndash78

Hara MR Gnaspini P 2003 Comparative study of the defen-sive behavior and morphology of the gland opening areaamong harvestmen (Arachnida Opiliones Gonyleptidae)under a phylogenetic perspective Arthropod Structure andDevelopment 32 257ndash275

Harvey MS 1992 The phylogeny and classification of thePseudoscorpionida (Chelicerata Arachnida) InvertebrateTaxonomy 6 1373ndash1435

Harvey MS 2002 The neglected cousins what do we knowabout the smaller arachnid orders Journal of Arachnology30 357ndash372

Haupt J Song D 1996 Revision of East Asian whip scorpions(Arachnida Uropygi Theyphonida) I China and JapanArthropoda Selecta 5 43ndash52

Hjelle JT 1990 Anatomy and morphology In Polis GA edThe biology of scorpions Stanford University Stanford Uni-versity Press 9ndash63

Janczyk FSW 1956 Anatomie von Siro duricorius Joseph imVergleich mit anderen Opilioniden Sitzungsberichte Oster-reichischen Akademie der Wissenschaften Mathematisch-Natruwissenschaftliche Klasse 165 474ndash522

Jenner RA 2001 Bilaterian phylogeny and uncritical recy-cling of morphological data sets Systematic Biology 50 730ndash742

Jeram AJ 1998 Phylogeny classification and evolution of Sil-urian and Devonian scorpions In Selden PA ed Proceed-ings of the 17th European Colloquium of ArachnologyEdinburgh 1997 Burnham Beeches UK British Arachno-logical Society 15ndash31

Karaman IM 2005 Evidence of spermatophores in Cyphoph-thalmi (Arachnida Opiliones) Revue Suisse de Zoologie 1123ndash11

Kaumlstner A 1935 Die Funktion der sogenannten sympathis-chen Ganglien und die Exkretion bei den PhalangiidenZoologischer Anzeiger 109 273ndash288

Kierans JE Robbins RG 1999 A world checklist of generasubgenera and species of ticks (Acari Ixodida) publishedfrom 1973 to 1997 Journal of Vector Ecology 24 115ndash129

Kjellesvig-Waering EN 1986 A restudy of the fossil Scorpi-onida of the world Palaeontographica Americana 55 1ndash287

Klompen JSH 2000 Prelarva and larva of Opilioacarus (Neo-carus) texanus (Chamberlin and Mulaik) (Acari Opilioacar-ida) with notes on the patterns of setation and lyrifissuresJournal of Natural History 34 1977ndash1992

Kraus O 1998 Elucidating the historical process of phylog-eny phylogenetic systematics versus cladistic techniquesIn Selden PA ed Proceedings of the 17th European Collo-quium of Arachnology Edinburgh 1997 Burnham BeechesUK British Arachnological Society 1ndash7

Kraus O Kraus M 1993 Divergent transformation of cheli-cerae and original arrangement of eyes in spiders (Arach-nida Araneae) Memoirs of the Queensland Museum 33579ndash584

Lankester ER Benham WBS Beck EJ 1885 On the mus-

242 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

cular and endoskeletal systems of Limulus and Scorpio withsome notes on the anatomy and generic characters of scor-pions Transactions of the Zoological Society of London 11311ndash384

Leclerc P 1989 Considerations paleobiogeographiques a pro-pos la decouverte en Thailande drsquoopilioacariens nouveaux(Acari ndash Notostigmata) Compte Rendu des Seances de laSociete de Biogeographie 65 162ndash174

Legg G 1976 The external morphology of a new species ofricinuleid (Arachnida) from Sierra Leone Zoological Journalof the Linnean Society 59 1ndash58

Lindquist EE 1984 Current theories on the evolution ofmajor groups of Acari and on their relationships with othergroups of Arachnida with consequent implications for theirclassificaton In Griffiths DA Bowman CE eds AcarologyVI Vol 1 New York John Wiley amp Sons 28ndash62

Manning PL Dunlop JA 1995 The respiratory organs ofeurypterids Palaeontology 38 287ndash297

Manton SM 1958 Hydrostatic pressure and leg exten-sion in arthropods with special reference to arachnidsAnnals and Magazine of Natural History Series 13 1161ndash182

Manton SM 1964 Mandibular mechanisms and the evolutionof arthropods Philosophical Transactions of the Royal Soci-ety of London B 7 1ndash183

Maraun M Heethoff M Schneider K Scheu S WeigmannG Cianciolo J Thomas RH Norton RA 2004 Molecularphylogeny of oribatid mites (Oribatida Acari) evidence formultiple radiations of parthenogenetic lineages Experimen-tal and Applied Acarology 33 183ndash201

Marples BJ 1967 The spinnerets and epiandrous glands ofspiders Journal of the Linnean Society (Zoology) 46 209ndash222

Marples BJ 1983 Observations on the structure of the fore-gut of spiders Bulletin of the British Arachnological Society8 46ndash52

Millot J 1933 Notes compleacutementaires sur lrsquoanatomie desliphistiides et des hypchilides a propos drsquoun travail reacutecent deA Petrunkevitch Bulletin de la Socieacuteteacute Zoologique de France68 217ndash235

Millot J 1942 Sur lrsquoanatomie et lrsquohistophysiologie deKoenenia mirabilis Grassi (Arachnida Palpigradi) RevueFranccedilaise drsquoEntomologie 9 33ndash51

Millot J 1943 Notes complementaires sur lrsquoanatomielrsquohistologie et la repartition geographique en France deKoenenia mirabilis Grassi Revue Franccedilaise drsquoEntomologie 9127ndash135

Millot J 1945 Lrsquoanatomie interne des Ricinulei (Arachnides)Annales des Sciences Naturelles (Zoologie) Series 11 7 1ndash29

Millot J 1949a Morphologie generale et anatomie interne InGrasseacute P-P ed Traite de zoologie VI Paris Masson et Cie263ndash319

Millot J 1949b Ordre des Amblypyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 563ndash588

Millot J 1949c Ordre des Araneacuteides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 589ndash743

Millot J 1949d Ordre des Palpigrades In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 520ndash532

Millot J 1949e Ordre des Uropyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 533ndash562

Millot J 1949f Ordre des Ricinuleides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 744ndash760

Millot J Vachon M 1949 Ordre Des Solifuges In Grasseacute P-P ed Traite de zoologie VI Paris Masson et Cie 482ndash519

Moore RA Briggs DEG Bartels C 2005 A new specimen ofWeinbergina opitzi (Chelicerata Xiphosura) from the LowerDevonian Huumlnsruck Slate Germany PalaumlontologischeZeitschrift 79 399ndash408

Moritz M 1957 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) unter besonderer Beruumlcksichtigungder ausseren Morphologie der Bildung des Mitteldarmesund der Genitalanlage Zoologische Jahrbuumlcher Abteilungfuumlr Anatomie und Ontogenie der Tiere 76 331ndash370

Moritz M 1959 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) II Die Anlage und Entwicklung derCoxaldruumlse bei Phalangium opilio L Zoologische Jahr-buumlcher Abteilung fuumlr Anatomie und Ontogenie der Tiere 77229ndash240

Nixon K 1999 The parsimony ratchet a new method forrapid parsimony analysis Cladistics 15 407ndash414

Norton RA 1998 Morphological evidence for the evolutionaryorigin of Astigmata (Acari Acariformes) Experimental andApplied Acarology 22 559ndash594

Palmgren P 1978 On the muscular anatomy of spiders ActaZoologica Fennica 155 1ndash41

Patten W 1912 The evolution of the vertebrates and their kinPhiladelphia P Blakistonrsquos Son

Patten W Hazen AP 1900 The development of the coxalgland branchial cartilages and genital ducts of Limuluspolyphemus Journal of Morphology 16 459ndash490

Paulus HF 1979 Eye structure and the monophyly of theArthropoda In Gupta AP ed Arthropod phylogeny NewYork Van Nostrand Reinhold 299ndash383

Petrunkevitch A 1949 A study of Palaeozoic ArachnidaTransactions of the Connecticut Academy of Arts and Sci-ences 37 69ndash315

Pittard K Mitchell RW 1972 Comparative morphology ofthe life stages of Cryptocellus pelaezi (Arachnida Ricinulei)Graduate Studies Texas Tech University 1 1ndash77

Platnick NI 2005 The world spider catalog Version55 American Museum of Natural History httpresearchamnhorgentomologyspiderscatalogindexhtm

Platnick NI Gertsch WJ 1976 The suborders of spiders acladistic analysis (Arachnida Araneae) American MuseumNovitates 2607 1ndash15

Pocock RI 1893 On some points in the morphology of theArachnida (ss) with notes on the classification of the groupAnnals and Magazine of Natural History Series 6 11 1ndash19

Pocock RI 1902 Some points in the morphology and classi-fication of the Opiliones Annals and Magazine of NaturalHistory Series 7 10 504ndash516

Popadicccc A Panganiban G Rusch D Shear WA KaufmanTC 1998 Molecular evidence for the gnathobasic derivationof arthropod mandibles and for the appendicular origin ofthe labrum and other structures Development Genes andEvolution 208 142ndash150

PHYLOGENY OF THE ARACHNID ORDERS 243

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Raw F 1957 Origin of chelicerates Journal of Paleontology31 139ndash192

Reddell JR Cokendolpher JC 1995 Catalogue bibliogra-phy and generic revision of the order Schizomida (Arach-nida) Texas Memorial Museum Speleological Monographs4 1ndash170

Reissland A Goumlrner P 1985 Trichobothria In Barth FGed Neurobiology of arachnids Berlin Springer-Verlag 138ndash161

Roewer C-F 1934 Solifugae Palpigradi Bronns Klassen undOrdnungen ses Tierreich 5 IV 4 1ndash713

Rowland JM Cooke JAL 1973 Systematics of the arachnidorder Uropygida (=Thelyphonida) Journal of Arachnology 155ndash71

Rowland JM Sissom WD 1980 Report on a fossil palpi-grade from the Tertiary of Arizona and a review of themorphology and systematics of the order (ArachnidaPalpigradida) Journal of Arachnology 8 69ndash86

Rucker A 1901 The Texan Koenenia American Naturalist35 615ndash630

Ruhland M Rathmayer W 1978 Die Beinmuskulatur undihre Innervation bei der Vogelspinne Dugesiella henzti (Ch)(Araneae Aviculariidae) Zoomorphologie 89 33ndash46

Saboori A Kamali K 2000 Description of Allothrombiumtriticium adult (Acari Trombidiidae) from Iran Systematicand Applied Acarology 5 207ndash208

Schliwa M 1979 The retina of the phalangid Opilio raven-nae with particular reference to arhabdomeric cells Celland Tissue Research 204 473ndash495

Scholl G 1977 Beitrage zur Embryonalentwicklung von Lim-ulus polyphemus L (Chelicerata Xiphosura) Zoomorpholo-gie 86 99ndash154

Scholtz G Kamenz C 2006 The book lungs of Scorpionesand Tetrapulmonata (Chelicerata Arachnida) evidence forhomology and a single terrestrialization event of a commonarachnid ancestor Zoology 109 2ndash13

Seitz KA 1972 Zur Histologie und Feinstruktur des Herzensund der Haumlmocyten von Cupiennius salei Keys (AraneaeCtenidae) I Herzwandung Bildung und Differenzierungder Haumlmocyten Zoologische Jahrbuumlcher Abteilung fuumlr Anat-omie und Ontogenie der Tiere 89 351ndash384

Sekiguchi K ed 1988 Biology of horseshoe crabs Tokyo Sci-ence House

Selden PA 1981 Functional morphology of the prosoma ofBaltoeurypterus tetragonophthalmus (Fischer) (ChelicerataEurypterida) Transactions of the Royal Society of Edin-burgh Earth Sciences 72 9ndash48

Selden P 1992 Revision of the fossil ricinuleids Transactionsof the Royal Society of Edinburgh Earth Sciences 83 595ndash634

Selden PA 1993 Fossil arachnids ndash recent advances andfuture prospects Memoirs of the Queensland Museum 33389ndash400

Selden PA Dunlop JA 1998 Fossil taxa and relationshipsof chelicerates In Edgecombe GD ed Arthropod fossilsand phylogeny New York Columbia University Press 303ndash331

Selden PA Jeram AJ 1989 Palaeophysiology of terrestrial-

isation in the Chelicerata Transactions of the Royal Societyof Edinburgh Earth Sciences 80 303ndash310

Selden PA Shear WA Bonamo PM 1991 A spider and otherarachnids from the Devonian of New York and reinterpre-tations of Devonian Araneae Palaeontology 34 241ndash281

Sensenig A Shultz JW 2003 Mechanics of cuticular elasticenergy storage in leg joints lacking extensor muscles inarachnids Journal of Experimental Biology 206 771ndash784

Shear WA 1993 New species in the opilionid genus Stylocel-lus from Malaysia Indonesia and the Philippines (OpilionesCyphophthalmi Stylocellidae) Bulletin of the British Arach-nological Society 9 174ndash188

Shear WA Selden PA Rolfe WDI Bonamo PM GriersonJD 1987 New terrestrial arachnids from the Devonian ofGilboa New York (Arachnida Trigonotarbida) AmericanMuseum Novitates 2901 1ndash74

Shultz JW 1987 The origin of the spinning apparatus in spi-ders Biological Reviews of the Cambridge PhilosophicalSociety 62 89ndash113

Shultz JW 1989 Morphology of locomotor appendages inArachnida evolutionary trends and phylogenetic implica-tions Zoological Journal of the Linnean Society 97 1ndash56

Shultz JW 1990 Evolutionary morphology and phylogeny ofArachnida Cladistics 6 1ndash31

Shultz JW 1991 Evolution of locomotion in Arachnida thehydraulic pressure pump of the giant whipscorpion Masti-goproctus giganteus (Uropygi) Journal of Morphology 21013ndash31

Shultz JW 1993 Muscular anatomy of the giant whipscor-pion Mastigoproctus giganteus (Arachnida Uropygi) and itsevolutionary significance Zoological Journal of the LinneanSociety 108 335ndash365

Shultz JW 1998 Phylogeny of Opiliones (Arachnida) anassessment of the lsquoCyphopalpatoresrsquo concept Journal ofArachnology 26 257ndash272

Shultz JW 1999 Muscular anatomy of a whipspider Phrynuslongipes (Amblypygi) and its evolutionary significance Zoo-logical Journal of the Linnean Society 126 81ndash116

Shultz JW 2000 Skeletomuscular anatomy of the harvest-man Leiobunum aldrichi (Weed 1893) (Arachnida Opil-iones Palpatores) and its evolutionary significanceZoological Journal of the Linnean Society 128 401ndash438

Shultz JW 2001 Gross muscular anatomy of Limuluspolyphemus (Chelicerata Xiphosura) and its bearing on evo-lution in the Arachnida Journal of Arachnology 29 283ndash303

Shultz JW 2007 Morphology of the prosomal exoskeleton ofScorpiones (Arachnida) and a new hypothesis for the evolu-tion of cuticular cephalic endoskeletons in arthropodsArthropod Structure amp Development 36 77ndash102

Shultz JW Pinto da Rocha R 2007 Morphology and func-tional anatomy In Pinto da Rocha R Machado G Giribet Geds The harvestmen the biology of Opiliones CambridgeMA Harvard University Press 14ndash61

Shultz JW Regier JC 2001 Phylogenetic analysis of Pha-langida (Arachnida Opiliones) using two nuclear protein-encoding genes supports monophyly of Palpatores Journalof Arachnology 29 189ndash200

Sitnikova LG 1978 The main evolutionary trends of the

244 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Acari and the problem of their monophyletism Entomologis-cheskoe Obozrenie 57 431ndash457

Snodgrass RE 1948 The feeding organs of Arachnidaincluding mites and ticks Smithsonian Miscellaneous Col-lections 110 (10) 1ndash94

Snodgrass RE 1952 A textbook of arthropod anatomy Ith-aca NY Comstock Publishing

Soslashrensen W 1879 Om Bygningen af Gonyleptiderne en Typeaf Arachnidernes Classe Naturhistorisk Tidsskrifter Series3 8 97ndash222

Stockwell SA 1989 Revision of the phylogeny and higherclassification of scorpions (Chelicerata) Unpublished doc-toral dissertation University of California at Berkeley

Stoslashrmer L 1944 On the relationships and phylogeny of fossiland Recent Arachnomorpha Skrifter utgitt av Det NorskeVidenskaps-Akademi i Oslo I 5 1ndash158

Stoslashrmer L 1955 Merostomata In Moore RC ed Treatise oninvertebrate paleontology Part P Arthropoda 2 LawrenceKS University of Kansas Press and Geological Society ofAmerica 4ndash41

Swofford DL 2002 PAUP Phylogenetic analysis using par-simony (and other methods) Version Beta 10 SunderlandMA Sinauer Associates

Talarico G Palacios-Vargas JG Fuentes Silva M AlbertiG 2005 First ultrastructural observations of the tarsal poreorgan of Pseudocellus pearsei and P boneti (ArachnidaRicinulei) Journal of Arachnology 33 604ndash612

Templeton AR 1983 Phylogenetic inference from restrictionendonuclease cleavage site maps with articular reference tothe evolution of humans and the apes Evolution 37 221ndash244

Tetlie OE Braddy SJ 2004 The first Silurian chasmataspidLoganamaraspis dunlopi gen et sp nov (Chelicerata Chas-mataspidida) from Lesmahagow Scotland and its implica-tions for eurypterid phylogeny Transactions of the RoyalSociety of Edinburgh Earth Sciences 94 227ndash234

Thomas RH Telford MJ 1999 Appendage development inembryos of the oribatid mite Archegozetes longisetosus(Acari Oribatei Trhypochthoniidae) Acta Zoologica (Stock-holm) 80 193ndash200

Tollerton VP Jr 1989 Morphology taxonomy and classifica-tion of the order Eurypterida Burmeister 1843 Journal ofPaleontology 63 642ndash657

Vachon M 1949 Ordre des Pseudoscorpions In Grasseacute P-Ped Traite de zoologie VI Paris Masson et Cie 436ndash481

Van der Hammen L 1979 Comparative studies in Chelicer-ata I The Cryptognomae (Ricinulei Architarbi and Anacti-notrichida) Zoologische Verhandelingen 174 1ndash62

Van der Hammen L 1982 Comparative studies in Chelicer-ata II Epimerata (Palpigradi and Actinotrichida) Zoolo-gische Verhandelingen 196 1ndash70

Van der Hammen L 1985 Functional morphology and affin-ities of extant Chelicerata in evolutionary perspectiveTransactions of the Royal Society of Edinburgh 76 137ndash146

Van der Hammen L 1989 An introduction to comparativearachnology Leiden SPB Academic Publishing

Vyas AB 1970 Studies in myology of the scorpion Heter-ometrus fulvipes Koch Unpublished PhD DissertationDepartment of Zoology Gujarat University India

Vyas AB 1974 The cheliceral muscles of the scorpion Heter-ometrus fulvipes Bulletin Southern California Academy ofSciences 73 9ndash14

Walter DE 1988 Predation and mycophagy by endeostigma-tid mites (Acariformes Prostigmata) Experimental andApplied Acarology 4 159ndash166

Walter DE Proctor HC 1998 Feeding behaviour and phy-logeny observations on early derivative Acari Experimentaland Applied Acarology 22 39ndash50

Walter DE Proctor HC 1999 Mites ecology evolution andbehaviour Sydney University of New South Wales

Waterston CD 1979 Problems of functional morphology andclassification in stylonuroid eurypterids (Chelierata Meros-tomata) with observations on the Scottish Silurian Sty-lonuroidea Transactions of the Royal Society of EdinburghEarth Science 70 251ndash322

Weygoldt P 1969 Biology of pseudoscorpions CambridgeMA Harvard University Press

Weygoldt P 1984 Lrsquoautotomie chez les Amblypyges RevueArachnologique 5 321ndash327

Weygoldt P 1996 Evolutionary morphology of whip spiderstowards a phylogenetic system (Chelicerata ArachnidaAmblypygi) Journal of Zoological Systematics and Olution-ary Research 34 185ndash202

Weygoldt P 1998 Evolution and systematics of the Chelicer-ata Experimental and Applied Acarology 22 63ndash79

Weygoldt P 2000 Whip spiders (Chelicerata Amblypygi)their biology morphology and systematics StenstrupDenmark Apollo Books

Weygoldt P Paulus HF 1979 Untersuchungen zur Morphol-ogie Taxonomie und Phylogenie der Chelicerata Zeitschriftfuumlr Zoologische Systematik und Evolutionsforschung 17 85ndash116 177ndash200

Wheeler WC Hayashi CY 1998 The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192

Winkler D 1957 Die Entwicklung der aumlusseren Koumlrperge-stalt bei den Phalangiidae (Opiliones) Mitteilungen aus demZoologischen Museum in Berlin 33 355ndash389

With CJ 1904 The Notostigmata a new suborder of AcariVidenskabelige Meddelelser fra Dansk Naturhistorisk Foren-ingi Kjobenhaus 1904 137ndash192

Yamasaki T Makioka T Saito J 1988 Morphology InSekiguchi K ed Biology of horseshoe crabs Tokyo ScienceHouse 69ndash132

Yoshikura M 1975 Comparative embryology and phylogeny ofArachnida Kumamoto Journal of Science Biology 12 71ndash142

APPENDIX

Each character used in this analysis is numbereddefined and cross-referenced with characters used infour previous studies that are abbreviated in bracketsspecifically WP = Weygoldt amp Paulus (1979)S = Shultz (1990) WH = Wheeler amp Hayashi (1998)GEWB = Giribet et al (2002) Descriptions includecorrections of errors discussions of controversies orambiguities and justifications for state assignmentsto lsquoproblemrsquo taxa

PHYLOGENY OF THE ARACHNID ORDERS 245

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

PROSOMA

1 Dorsal sclerite formed by fusion of the prosomal car-apace the dorsal portion of the first opisthosomalsomite and the dorsomedial (axial) portion of the sec-ond opisthosomal somite 0 absent 1 presentState 1 is present in extant xiphosurids and close fos-sil relatives (Scholl 1977 Anderson amp Selden 1997Shultz 2001) The fossil record suggests that thisstructure originated with the disappearance ofthe already reduced first opisthosomal tergite(= microtergite) and axial portion of the secondopisthosomal tergite (Anderson amp Selden 1997)2 Single dorsal sclerite covering entire dorsal surfaceof body no lines indicating original segmentation 0absent 1 presentState 1 occurs throughout Holothyrida (Acari) (Vander Hammen 1989) and in many Mesostigmata(Alberti amp Coons 1999) including Glyptholaspis (Vander Hammen 1989)3 Anterior end of dorsal prosoma with medianmarginal or submarginal pointed process 0 absent1 presentState 1 occurs throughout Schizomida (Protoschizo-mus Stenochrus Reddell amp Cokendolpher 1995) The-lyphonida (Rowland amp Cooke 1973 Shultz unpublobs Proschizomus Dunlop amp Horrocks 19951996)Amblypygi (Charinus Millot 1949b Phrynus Shultz1999) and Plesiosiro (Dunlop 1999) Citing a figure inRowland amp Cooke (1973) Dunlop amp Horrocks assumedthat the anterior process was limited to lsquohypoctonidrsquothelyphonids and united them with Proschizomus andSchizomida If valid this would make Thelyphonidaparaphyletic However the doubtful monophyly oflsquohypoctonidsrsquo (Haupt amp Song 1996) and widespreadpresence of a submarginal anterior process in Pedi-palpi weaken this proposal (Harvey 2002)4 Ophthalmic ridges pair of longitudinal crests inter-secting or passing near the region of the lateral eyes orcomparable region where lateral eyes are absent 0absent 1 presentState 1 is present throughout Xiphosura (Anderson ampSelden 1997) Similar structures are present in Plesi-osiro (Dunlop 1999) and lsquonon-hypoctonidrsquo Thely-phonida (Mastigoproctus Rowland amp Cooke 1973)Dunlop (2002a) described apparent ophthalmic ridgesin the chasmataspidid Octoberaspis but these werenot observed in Diploaspis (Dunlop et al 2001) eitherdue to absence or inadequate preservation of speci-mens nor in Chasmataspis (Dunlop et al 2004)5 Carapace with demarcations (eg grooves scleritesphragmata) between pro- meso- or metapeltidia 0absent 1 present [S 1 WH 48 GEWB 29]State 1 is present in extant Scorpiones Cyphoph-thalmi (Hansen amp Soslashrensen 1904) and many Palpa-tores (Opiliones) including Caddo and Leiobunum

and most Pseudoscorpiones (Chamberlin 1931) Giri-bet et al (2002) coded this character as absent inCyphophthalmi (Opiliones) but a transverseprocurved groove is present throughout the group(Hansen amp Soslashrensen 1904) and represents the meso-metapeltidial border as indicated by attachment ofthe dorsal endosternal suspensor of postoral somite VIin Chileogovea oedipus and Siro exilis (Shultz unpublobserv) Kjellesvig-Waering (1986) did not explicitlydescribe such grooves in fossil scorpions although heillustrated corresponding structures in PrearcturusArchaeophonus Archaeoctonus Palaeophonus andother Palaeozoic scorpions (Kjellesvig-Waering 1986figs 14A 21E 27D 59) (coded as uncertain here) butnot in Proscorpius (contra Giribet et al 2002) Dis-tinct carapacal sclerites are present throughout Palpi-gradi Schizomida and Solifugae (6)

The condition in Acari is controversial but is codedhere as State 1 Van der Hammen (1989) consideredthe tergal region of the leg-bearing somites (postoralsomites IIIndashVI) or podosoma to be reduced andreplaced by posterior migration of the aspidosoma (ietergal region assumed to be associated with append-ages of the gnathosoma) and anterior migration of theopisthosomal tergal region to form the hysterosomaThe dorsal proterosomal and hysterosomal elementsmeet at a transverse sejugal furrow that continues lat-erally and passes ventrally between the coxae of legs 2and 3 According to this scheme the prosomandashopisthosoma border is expressed as the disjugal fur-row which passes from the ventral posterior margin ofthe podosoma anterodorsally to join the dorsal part ofthe sejugal furrow Based on the arrangement of setaeand slit sensilla (= lyrifissures) Van der Hammen(1989) interpreted the region above coxae 3 and 4 (ieregion C) as a fusion of the dorsal parts of the first twoopisthosomal somites (= postoral somites VII and VIII)in early divergent Anactinotrichida (ie Opilioacari-formes) and Acariformes (eg Alycus) These interpre-tations have been followed by many acarologistsalthough its speculative aspects are widely acknowl-edged (Evans 1992 Alberti amp Coons 1999)

From the standpoint of a general arachnologist theVan der Hammen system seems unnecessarily compli-cated it appears to have been formulated to explainbroad morphological themes in Oribatida but was thenextrapolated to other mites A more conservativelsquoarachnologicalrsquo scheme adopted here equates the pro-podosoma with the pars cephalica (= propeltidium) ofthe arachnid prosoma and region C with the pars tho-racica (= mesopeltidium + metapetidium) In fact theearliest and most recent treatments of Opilioacari-formes (With 1904 Klompen 2000) reached similarconclusions This interpretation is also consistent withevidence from Acariformes For example the regiondorsal to the coxa of legs 3 and 4 in Alycus is as readily

246 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

explained by the persistence of primitive prosomal ter-gal elements as by the supposition that these disap-peared and were replaced by two opisthosomalelements Furthermore several diverse lineages ofbasally divergent Acariformes display features that donot fit readily within the GrandjeanndashVan der Ham-men system but seem to correspond to the cephalicathoracica division of the prosoma They have a sejugalfurrow that may have demarcated the cephalicatho-racica border and a postpodosomal furrow thatappears to represent the prosomandashopisthosoma junc-tion there is no disjugal furrow Examples includepalaeosomate (especially aphelacarid) and pediculoch-elid Oribatida alicorhagiid and mircopsammidlsquoEndeostigmatarsquo and paratydeid Prostigmata (Albertiamp Coons 1999 fig 42) This organization may be sym-plesiomorphic for Acariformes but could be a conver-gence brought about by selection for enhancedflexibility6 Carapace with distinct pro- meso- or metapeltidialsclerites 0 absent 1 present - inapplicable due toabsence of pro- meso- or metapeltidial demarcations(5)State 1 occurs throughout Palpigradi Schizomida andSolifugae It also appears sporadically within Opil-iones (eg Leiobunum flavum Shultz unpubl observ)and Pseudoscorpiones (eg Pseudochiridium) (Cham-berlin 1931)7 Sejugal furrow circumferential zone of body flexi-bility that passes between the coxae of legs 2 and 3 0absent 1 presentState 1 is probably synapomorphic for Acariformes asit occurs in most lsquoendeostigmatidsrsquo including Alycus(Van der Hammen 1989) and is variously developedthroughout Sarcoptiformes (Alberti amp Coons 1999) Itis probably the primitive condition for Prostigmatabut is either absent or weakly expressed in represen-tatives included here See 5 for alternative morpholog-ical interpretations of this body region in Acari8 Prosomal ozopores 0 absent 1 present [WH 46GEWB 12]State 1 occurs throughout Opiliones and Holothyrida(Van der Hammen 1989)9 Carapacal pleural doublure 0 absent 1 present[WP 12 S 2 WH 49 GEWB 27]A carapacal pleural doublure occurs throughoutXiphosura (Stoslashrmer 1944) Eurypterida (Clarke ampRuedemann 1912 Selden 1981) and Chasmataspid-ida (Dunlop et al 2001 2004 Dunlop 2002a) and isabsent in all known Arachnida10 Cardiac lobe a longitudinal axial elevation of thecarapace 0 absent 1 presentState 1 occurs throughout Xiphosura (Anderson ampSelden 1997) and Eurypterida (Clarke amp Ruedemann1912 Selden 1981) Carapaces are not preserved suf-ficiently to determine this condition in the chasmatas-

pidids Diploaspis (Dunlop et al 2001) andOctoberaspis (Dunlop 2002a) but an apparent cardiaclobe occurs in the Chasmataspis (Dunlop et al 2004)A cardiac lobe is absent in Arachnida11 Moveable cucullus 0 absent 1 present [WP 54GEWB 22]State 1 is a synapomorphy of Ricinulei (Selden 1992)12 Medial intercoxal lsquosternalrsquo region 0 all pedalcoxae separated medially 1 anterior pedal coxae abut-ting medially posterior coxae separated 2 anteriorpedal coxae separated medially posterior coxae abut-ting 3 all pedal coxae abutting medially 4 epimeracoxae undifferentiated medially from ventral body wall[S sim3 WH sim50 GEWB sim17]Prosomal sternites appear to be those portions of theventral body wall not occupied by coxae Despite earlyattempts to assess metamerism of the ventralprosoma (protosternum deutosternum tritosternumetc) (Boumlrner 1902a Millot 1949a) no apparent mor-phological features such as borders of sclerites ormuscle attachments reliably demarcate the ventralbody wall of one somite from that of an adjacentsomite Thus coding schemes that focus on describingmetameric components of the sternal region are prob-lematic Shultz (1990) attempted to code the shape ofthe entire intercoxal region and subsequent workershave adopted this approach All pedal coxae are sepa-rated medially in Plesiosiro (Dunlop 1999) Palpigradi(Roewer 1934) Araneae (Millot 1949c) Amblypygi(Weygoldt 2000) and many Acari including Opilioac-ariformes Holothryrida Mesostigmata Ixodida Aly-cus and many Prostigmata (Van der Hammen 1989Evans 1992 Alberti amp Coons 1999) The anteriorcoxae are separated in extant Thelyphonida Schizo-mida (Hansen amp Soslashrensen 1905 Millot 1949e) andRicinulei (Pittard amp Mitchell 1972 Legg 1976) Thecoxae are fused to the ventral body wall in Sarcopti-formes (Van der Hammen 1989 Evans 1992)

Interpreting this character is complicated for theanterior coxae of Scorpiones and Opiliones and allcoxae in Xiphosura and Eurypterida due to the pres-ence of coxapophyses endites andor gnathobaseswhich form components of the preoral chambers inthese groups Thus coxae may abut medially whenthese structures are considered part of the coxae butare separate medially if these structures are ignoredAs coxapophyses etc are here coded as separate char-acters (50ndash53) I have chosen to code this character asif these structures did not exist Consequently pha-langid Opiliones are coded as having all coxae sepa-rated (Pocock 1902 Hansen amp Soslashrensen 1904) Fossilscorpions also show substantial variation in this char-acter All pairs of pedal coxae in Palaeoscorpius abutmedially but also have gnathobases (Jeram 1998)and this condition is coded here as uncertain A ster-num separates all pairs of pedal coxae medially in

PHYLOGENY OF THE ARACHNID ORDERS

247

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Proscorpius

and

Stoermeroscorpio

(State 0) and

Pre-arcturus

approximates modern scorpions (State 1)(Kjellesvig-Waering 1986 Jeram 1998) The coxalbases of Eurypterida and Xiphosura are separated bya small sternite the endostoma (Selden 1981 Shultz2001)

13 Postoral sternapophysis (

=

tritosternum labium)a cuticular evagination of the ventral body wall poste-riorly adjacent to the palpal coxae forming the poste-rior border of the preoral chamber in some taxa ordisplaced posteriorly by fusion of the palpal coxae inothers 0 absent 1 present

[S

sim

10 WH

sim

56 GEWB19

=

33]State 1 is present throughout Palpigradi AraneaeAmblypygi (Snodgrass 1948) Schizomida (Hansen ampSoslashrensen 1905 Van der Hammen 1989) Thely-phonida (Van der Hammen 1989) extant Ricinulei(Pittard amp Mitchell 1972 Legg 1976) Opilioacari-formes and Mesostigmata (Van der Hammen 1989Alberti amp Coons 1999)

Allothyrus constrictus

is theonly holothyrid mite known to have sternapophysesAn apparent sternapophysis has been described in apalaeocharinid trigonotarbid (Dunlop 1994) Giribet

et al

(2002) miscoded Ixodida and Acariformes as hav-ing a lsquolabium (

=

tritosternum)rsquo The lophognath ofpseudoscorpions (Chamberlin 1931) and lsquolabiumrsquo ofSolifugae (Roewer 1934) also appear to be sternapo-physes that have been incorporated into the ros-trosoma (

32

) The labium of phalangid Opiliones(Pocock 1902 Hansen amp Soslashrensen 1904) is a ster-napophysis but is associated with leg 1 rather thepedipalp (Winkler 1957) A small anterior sclerite inthe fossil scorpions

Proscorpius

Waeringoscorpio

and

Labriscorpio

has been interpreted as a lsquolabiumrsquo homol-ogous with the sternapophyses of other arachnids(Weygoldt 1998) despite absence of any indication ofthe relationship of this sclerite to the mouth and pos-sible alternative interpretations (eg

16

) Giribet

et al

coded this sclerite as a sternapophysis but the char-acter is coded here as uncertain in fossil scorpions

14 Channels on the body surface linking openings ofcoxal organs to preoral chamber 0 absent 1 present

Van der Hammen (1989) described bilaterally pairedcuticular tracts (lsquotaenidiarsquo) connecting orifices of coxalorgans to the preoral chamber in some arachnids andor a ventromedian groove (

=

intercoxal or subcapitulargutter) that leads in turn to the preoral chamber inother taxa Taenidia and intercoxal gutters arepresent in Opilioacariformes (Van der Hammen1989) Holothyrida (Van der Hammen 1989) Mesos-tigmata (

Glyptholapis

Van der Hammen 1989)Ricinulei (Pittard amp Mitchell 1972 Van der Hammen1989) Araneae (

Heptathela

Aphonopelma

Hypochi-lus

Shultz unpubl observ also

Porrhothele

(Mygalo-morphae Dipluridae) Butt amp Taylor 1991

Segestria

(Araneomorphae Segestriidae) Van der Hammen

1989) Amblypygi (Van der Hammen 1989) Schizo-mida (

Stenochrus

Shultz unpubl observ unidenti-fied Hubbardiinae Van der Hammen 1989) andThelyphonida (

Mastigoproctus

Shultz unpublobserv also

Tetrabalius

Van der Hammen 1989) Theconduction of large volumes of fluid from orifices of thecoxal organs to the preoral chamber has been observedin the mygalomorph spider

Porrhothele

(Dipluridae)(Butt amp Taylor 1991) and

Mastigoproctus

(Thely-phonida) (Shultz unpubl observ) The podocephaliccanal (

15

) is present in Acariformes (Acari) (Lindquist1984) and appears to have evolved as an invaginatedsupracoxal channel (Grandjean 1938 G Alberti inEvans 1992) Van der Hammen (1982) could not dis-cern with certainty whether a channel exists in

Eukoenenia

(Palpigradi) but none has been reportedby previous authors and they have not been observedby the present author Palpigradi is coded as havingState 0

15 Podocephalic canal cuticular channel andorduct draining multiple glands and opening nearmouthparts 0 absent 1 present - inapplicable due toabsence of channel (14)

State 1 occurs throughout Acariformes (Acari)(Lindquist 1984)

16 Heavily sclerotized suboral sclerite serving inpart as basal pivot point for coxae of appendages ofpostoral somites IIndashIV (

=

arachnid palp and legs 1 and2) 0 absent 1 present

The sclerite forms the posterior wall of the true mouthand is heavily sclerotized in extant scorpions It is notvisible externally due to the tightly fitting coxae of theanterior prosomal appendages (Shultz 2007) It maycorrespond to the so-called labium (Weygoldt 1998) orlabrum (Kjellesvig-Waering 1986) of certain fossilscorpions (see

13

)

17 Genal angles 0 rounded 1 pointed

Anderson amp Selden (1997) originally coded a pointedgenal angle and genal spine as separate charactersbut they are combined here State 1 is present in manyXiphosura (Anderson amp Selden 1997) and Chas-mataspidida (Dunlop

et al

2001 2004 Dunlop2002a)

A

PPENDAGES

OF

POSTORAL

SOMITE

I

CHELICERAE

18 Cheliceral segmentation 0 three articles 1 twoarticles

[WP

sim

25

+

42 WH

sim

15 GEWB

sim

44]State 0 is present in extant Xiphosura Eurypterida(Clarke amp Ruedemann 1912 Selden 1981) Palpi-gradi (Roewer 1934) Opiliones (Hansen amp Soslashrensen1904) Scorpiones (Hjelle 1990) and throughout non-acariform Acari (Evans 1992) State 1 is presentthroughout known Trigonotarbida Araneae Ambly-pygi Thelyphonida Schizomida Pseudoscorpionesand Solifugae Acariform mites have two unambigu-

248 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ous cheliceral articles but may have a reduced thirdarticle (Van der Hammen 1989 Evans 1992) acari-form mites are thus coded as uncertain Chelicerae arenot known or their condition is unclear in several fos-sil taxa including Xiphosura Chasmataspidida (Dun-lop et al 2001 2004 Dunlop 2002a) Riciniulei(Selden 1992) Plesiosiro (Dunlop 1999) Proschizo-mus (Dunlop amp Horrocks 19951996) and the fossilscorpions Praearcturus and Palaeoscorpius (Kjelles-vig-Waering 1986) Kjellesvig-Waering (1986) consis-tently interpreted chelicerae of fossil scorpions ashaving four articles but this is probably a misinter-pretation of the basal article (see also Stockwell1989) The basal article in extant scorpions consists ofa distal collar of cuticle and a large proximal processand this arrangement would probably appear as twoarticles in compressed specimens In any event thereare at least three cheliceral articles in Stoermeroscor-pio and Proscorpius (Kjellesvig-Waering 1986)19 Terminal cheliceral joint 0 laterally placedbicondylar hinge 1 dorsally placed bicondylar hinge2 ventrally placed bicondylar hinge [WP 25 + 42 Ssim11 + 12 WH sim15 GEWB sim44]State 0 is present in Xiphosura (Shultz 2001)extant Scorpiones (Millot amp Vachon 1949) Opil-iones (Hansen amp Soslashrensen 1904) Palpigradi(Boumlrner 1904 Millot 1949d) Ricinulei (Pittard ampMitchell 1972 Legg 1976) and Eurypterida includ-ing Baltoeurypterus (Selden 1981) and Stylonurus(Clarke amp Ruedemann 1912) State 0 is present inthe fossil scorpions Proscorpius and Stoermeroscor-pio but the state of the chelicerae is unclear in theother fossil scorpions represented here (Kjellesvig-Waering 1986) State 1 is present in Trigonotarbida(Shear et al 1987) mygalomorph Araneae Ambly-pygi Schizomida and Theylphonida (Weygoldt ampPaulus 1979) State 2 is present in Pseudoscorpi-ones (Chamberlin 1931) and Solifugae (Roewer1934) and appears to be a groundplan feature for allmajor groups of mites (eg Van der Hammen 1989Alberti amp Coons 1999) The difference between theorthognathy of mygalomorph Araneae and lsquoplagiog-nathyrsquo mesothele (Liphistius Heptathela) and palae-ocribellate Araneae (Hypochilus) (Kraus amp Kraus1993) is small compared with the variation thatoccurs throughout arachnids and plagiognathy iscoded as State 1 here20 Chelicera articulating with carapace at anterolat-eral pivot 0 absent 1 present [S 13 WH 57 GEWB45]State 1 occurs in Solifugae (Roewer 1934) and Pseu-doscorpiones except Chthonioidea and Feaelloidea(Chamberlin 1931) Giribet et al (2002) coded thecharacter as present in Pseudoscorpiones21 Chelicera pivoting on dorsal protuberance of epis-tome 0 absent 1 present

State 1 appears to be a unique synapomorphy of Opil-iones (Shultz 2000)22 Extrinsic cheliceral muscle arising on carapaceand inserting on dorsal margin of nonbasal cheliceralarticle 0 absent 1 present (Note Characters 22 and23 refer to two different muscles)State 1 is known only in extant Scorpiones (LankesterBenham amp Beck 1885 Vyas 1970 1974 Shultz2007) but may occur in acariform mites if their cheli-cerae are composed of three rather than two articles(Evans 1992) (see 18) Acariformes are coded asuncertain23 Extrinsic cheliceral muscle arising on carapaceand inserting on ventral margin of nonbasal cheliceralarticle 0 absent 1 present [GEWB 252] (Note Char-acters 22 and 23 refer to two different muscles)State 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) somemesostigmatid Acari (Evans 1992) and representa-tive Opiliones particularly Leiobunum (Shultz 2000)Chileogovea and Gonyleptes (Shultz unpubl observ)The status of the character in other opilions is not yetknown The character was originally defined as anextrinsic muscle inserting on the second segment ofthree-segmented chelicerae (Shultz 2000 Giribetet al 2002) but it has been redefined here to make itapplicable to arachnids with two-segmented cheli-cerae24 Extrinsic cheliceral muscles attaching to epistome0 absent 1 presentState 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) but mayoccur in Prostigmata if the sigmoid piece to which thecheliceral protractors attach (Evans 1992) is a modi-fication of the epistome25 Lateral tergocheliceral muscle with three heads 0absent 1 present - coded only for extant tetrapulmo-nates and the palpigrade Eukoenenia homology isunclear in other taxa [GEWB 230]State 1 is present in Pedipalpi (Shultz 1999)26 Cheliceral silk glands 0 absent 1 present [GEWB47]State 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931) The silk gland of spider mites(eg Tetranychus) are associated with the pedipalps(Evans 1992)27 Cheliceral venom glands 0 absent 1 present [WP34 WH 40 GEWB 46]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)28 Cheliceral serrula interior and exterior 0 absent1 presentState 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931 Harvey 1992)29 Cheliceral lsquoflagellumrsquo in male 0 absent 1 present[GEWB 48]

PHYLOGENY OF THE ARACHNID ORDERS 249

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITE II LEG 1 IN NONARACHNIDS AND PALP IN ARACHNIDS

30 Palpal coxae fused ventromedially and formingposterior wall of preoral chamber 0 absent 1 present[WP 27 S 18 WH 17 = 62 GEWB 33 + 63]State 1 occurs in Schizomida Thelyphonida Acari andRicinulei31 Gnathosoma 0 absent 1 present - inapplicablepalpal coxae not fused (30) [WP 56 S 19 WH sim63GEWB sim36 + 37]The gnathosoma is a functional complex comprisingan epistome attached to the dorsal surface of mediallyfused palpal coxae (= infra- or subcapitulum) (see 30)the chelicerae and in some taxa a supracheliceral tec-tum These components are tightly integrated andmove in unison relative to the body The gnathosomahas traditionally been considered a feature of AcariHowever Van der Hammen (1989) argued that a gna-thosoma is present in Ricinulei and this interpreta-tion has received wide acceptance (eg Lindquist1984) However it is not clear that the relationship ofthe chelicerae and subcapitulum is as intimate inRicinulei as in Acari and the ricinuleid subcapitulumdoes not appear to be as moveable In fact the lsquosubca-pitulumrsquo in Ricinulei is similar in many respects to thecondition in Schizomida and Thelyphonida (Hansen ampSoslashrensen 1904) which has never been regarded as agnathosoma Consequently I have coded the conditionof Ricinulei as uncertain32 Rostrosoma long narrow subcylindrical epistomeprojecting anteriorly with base fixed to dorsal surfaceof palpal coxae bordered laterally by lobes projectingfrom palpal coxae ventral wall of preoral chamberformed by anterior element of prosoma (sternapophy-sis) 0 absent 1 present [S 15 WH 59 GEWB 20 + 36]State 1 is present in Pseudoscorpiones and Solifugae(Chamberlin 1931 Roewer 1934 Van der Hammen1989) Dunlop (2000) attempted to homologize com-ponents of the rostrosoma and gnathosoma (31) anargument accepted by Giribet et al (2002) Howeverfusion of the palpal coxae (30) apparently evolvedindependently in the two structures Coxal fusion inthe gnathosoma is complete ventrally and excludesthe suboral sternapophysis (13) while fusion isincomplete ventrally in the rostrosoma and incorpo-rates the sternapophysis Unlike the gnathosomathe rostrosoma is largely immobile and cheliceralmovement is not coupled with that of the palpalcoxae33 Rutellacorniculi hypertrophied setae modified asmouthparts located on the anterior processes of the pal-pal coxae 0 absent 1 present [GEWB 39]

State 1 occurs in the anactinotrichid groups Opilioac-ariformes Holothyrida and Mesostigmata but not Ixo-dida (Van der Hammen 1989) It also occurs inSarcoptiformes (Evans 1992) and certain endeostig-matids (eg Alycus Van der Hammen 1989) but not inProstigmata Comparable structures are unknown inother chelicerate groups34 Terminal segments specially modified in adultmale as a clasper used to engage the female 0 absent1 presentState 1 is known only in extant Xiphosura (YamasakiMakioka amp Saito 1988)35 Robust raptorial 0 absent 1 present [S 17 WH61 GEWB sim59 97]State 1 is present in Amblypygi Thelyphonida andSchizomida (Shear et al 1987) and in Laniatores(Opiliones)36 Orientation of robust raptorial appendage 0operating in subtransverse plane 1 operating in sub-vertical plane - inapplicable coded only for PedipalpiState 0 is present in extant Thelyphonida and mostAmblypygi and State 1 is present in Schizomida (Cok-endolpher amp Reddell 1992) and in the basally diver-gent amblypygid Paracharon caecus (Weygoldt 2000)The orientation of the pedipalps in the fossil Proschi-zomus is uncertain (Dunlop amp Horrocks 19951996)37 Extrinsic muscle attaching to epistome 0 absent1 presentState 1 occurs in Chileogovea and Siro (OpilionesCyphophthalmi) and extant Scorpiones (Vyas 1970Shultz 2007)38 Muscle originating and inserting within coxa 0absent 1 present [GEWB sim232]State 1 occurs is in Amblypygi (Phrynus Shultz 1999Phrynichus Boumlrner 1904) Thelyphonida (Mastigo-proctus Shultz 1993 Thelyphonus Boumlrner 1904) andSchizomida (Stenochrus Shultz unpubl observ)There are two intracoxal muscles in amblypygids andthelyphonids the one coded here and another that isserially homologous with a muscle in the pedal coxae(54)39 Tarsus andor tibia with venom glands 0 absent1 present [GEWB 64]State 1 is known only from extant Pseudoscorpi-ones except the superfamilies Chthonioidea andFeaelloidea (Chamberlin 1931 Harvey 1992) Giri-bet et al (2002) coded this character as present forPseudoscorpiones40 Tarsal grooming organ 0 absent 1 present[GEWB 98]State 1 is a unique synapomorphy of Amblypygi (DelleCave 1975 Weygoldt 2000)41 Scorpionoid chela a large well-developed chelaformed by tibia (manus + fixed finger) and tarsus(moveable finger) 0 absent 1 present [S 16 WH 60GEWB 62]

250 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs throughout Scorpiones andPseudoscorpiones42 Modified in male as copulatory organ 0 absent 1present [WP 35 WH 39 GEWB 67]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)43 Apotele 0 apparently absent not differentiatedexternally from penultimate article 1 present [GEWB69]The apotele exists as a structure (ie dactyl clawempodium) differentiated from the tibiotarsus tarsusor telotarsus and is typically associated with a pair ofantagonistic muscles The apotele occurs as a distinctstructure in Xiphosura (Snodgrass 1948) Eurypter-ida (Clarke amp Ruedemann 1912) Schizomida (Coken-dolpher amp Reddell 1992) Araneae (Foelix 1996)early divergent Amblypygi (Charinus Weygoldt 2000not Phrynus Shultz 1999) Anactinotrichida (CaminClark amp Bourdeau 1956 Lindquist 1984) Ricinulei(Pittard amp Mitchell 1972 Legg 1976) and Solifugae(Roewer 1934) Giribet et al (2002) coded Solifugae asnot having a palpal apotele but it is present as anadhesive organ (45) Dunlop (1999) reconstructedHaptopoda as having a palpal claw but notes in hisdescription that the distal ends were not preserved inthe known specimens44 Apotele (claw) position 0 terminal 1 subtermi-nal - inapplicable coded only for taxa with a distinctapotele (43)State 1 occurs in Parasitiformes (Acari) (Evans 1992)45 Terminal adhesive organ 0 absent 1 present -inapplicable due to lack of an apotele (43) [GEWB 66]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITES IIIndashVI LEGS 2ndash5 IN NON-ARACHNIDS AND LEGS 1ndash4 IN ARACHNIDS

46 Appendage III (= arachnid leg 1) extremely elon-gate antenniform 0 absent 1 present [WP sim24 + 32S 20 WH 14 GEWB sim84]State 1 occurs in Amblypygi Thelyphonida Schizo-mida (Shear et al 1987) and Opilioacariformes (Vander Hammen 1989) State 1 is approximated in Solif-ugae Palpigradi (Roewer 1934) and Haptopoda(Dunlop 1999) and it is coded as uncertain in thesetaxa47 Appendage V (= arachnid leg 3) of male specializedfor sperm transfer 0 absent 1 present [WP 55 GEWB94]State 1 is known only from extant Ricinulei48 Appendages V and VI (= arachnid legs 3 and 4)with femur shorter than patella and with principal siteof flexionextension at patella-tibia joint (lsquoapatellatersquocondition sensu Van der Hammen 1989) 0 absent 1present [S 25 GEWB 86]

State 1 occurs in Pseudoscorpiones and Solifugae(Shultz 1989)49 Appendage VI (= arachnid leg 4) with terminusmodified as flattened blade 0 absent 1 presentState 1 occurs in some Chasmataspidida (DiploapisDunlop et al 2001 Octoberaspis Dunlop 2002a)uncertain in Chasmataspis Dunlop et al 2004 State0 in Loganamaraspis Tetlie amp Braddy 2004) andEurypterida except stylonuroids (Clarke amp Ruede-mann 1912 Waterston 1979 Tollerton 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI COXA AND BODYndashCOXA JOINT

50 Appendage III (= arachnid leg 1) with coxapophy-sis forming floor or wall of preoral chamber 0 absent1 present [S sim14 WH sim58]State 1 occurs in Opiliones (Hansen amp Soslashrensen 1904)and higher scorpions (Jeram 1998) A lobe is presenton the coxa of leg 1 in the fossil scorpion Prearcturusand perhaps Stoermeroscorpio and is coded here as acoxapophysis Coxapophysis of leg 1 together with theepistome and coxapophysis of the palp form a uniquepreoral chamber the stomotheca Weygoldt (1998) andDunlop amp Braddy (2001) reject the homology of cox-apophyses and stomothecae in Opiliones and Scorpi-ones because these structures are not apparent infossils of those scorpions thought to have been aquaticHowever the coxapophyses in Opiliones are formedlargely from soft cuticular lsquolipsrsquo and similar structuresin fossil scorpions would probably not have beenpreserved51 Appendage IV (= arachnid leg 2) with coxapophy-sis 0 absent 1 present [S sim14 WH sim58 GEWB 75]State 1 occurs in extant Scorpiones and most Opil-iones except Dyspnoi (Shultz 1998)52 At least one pair of coxal gnathobases on append-ages III-VI (= arachnid legs 1ndash4) 0 absent 1 present[S sim23 WH sim66 GEWB 78]State 1 occurs in all appropriately preserved Euryp-terida including Baltoeurypterus (Selden 1981) andStylonurus (Clarke amp Ruedemann 1912) extantXiphosura on appendages IIndashVI and apparentlyProscorpius on appendage III (= arachnid leg 1)(Kjellesvig-Waering 1986 Jeram 1998 contra Giribetet al 2002) Fossilized appendages with gnathobaseshave been found in association with Chasmataspis(Chasmataspidida) suggesting that they may havehad State 1 (Dunlop et al 2004)53 Coxae of appendages IIIndashV (= arachnid legs 1ndash3)with jointed moveable endites 0 absent 1 present [Ssim23]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) and in adequately preserved eurypterids(Clarke amp Ruedemann 1912) (BaltoeurypterusSelden 1981)

PHYLOGENY OF THE ARACHNID ORDERS 251

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

54 Intracoxal muscle a muscle arising on anteriorwall of coxa and inserting on posterior wall 0 absent1 present [GEWB 237]State 1 occurs in Thelyphonida (MastigoproctusShultz 1993) and Amblypygi (Phryus Shultz 1999)It is unclear whether the muscle operating the move-able endite (53) in extant Xiphosura (Manton 1964Shultz 2001) is homologous with this muscle and iscoded here as uncertain It is not present in Steno-chrus (Schizomida) (Shultz unpubl observ)55 Coxae of appendages IIndashVI (= arachnid palp andlegs) with dorsal articulation with carapace 0 absent1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)56 Flabellum (exite) on coxa of appendage VI(= arachnid leg 4) 0 absent 1 present [GEWB sim110]State 1 occurs in extant Xiphosura (Yamasaki et al1988) An apparent exite occurs on appendages asso-ciated with Chasmataspis (Chasmataspidida) (Dunlopet al 2004)57 Insertion process of anteromedial tergo-coxal mus-cle 0 weakly developed 1 well to extremely well devel-oped [GEWB 238]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Boumlrner 1904 Shultz 1999)58 Musculi laterales enlarged lateral tergocoxalmuscle with attachment shifted from coxa to adjacentpleural membrane 0 absent 1 present [S 22 WH 65GEWB 242]State 1 occurs in Araneae and Thelyphonida (Shultz1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI TROCHANTER AND COXAndashTROCHANTER JOINT

59 Coxandashtrochanter joint with complex posteriorarticulation composed of two articulating sclerites 0absent 1 present [S 24 WH 67 GEWB 85]State 1 occurs in Araneae Amblypygi Thelyphonidaand Schizomida (Shultz 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI FEMUR AND TROCHANTERndashFEMUR JOINT

60 Depressor muscle (or homologue) of trochanter-femur joint 0 absent 1 present [WH 42 GEWB 220]State 0 appears to be a unique synapomorphy of Ara-neae (Liphistius Shultz 1989 Aphonopelma Ruh-land amp Rathmayer 1978 Hypochilus present studyother examples Clarke 1984 1986)61 Trochanterndashfemur joint with dorsal hinge or pivotoperated by flexor muscles only 0 absent 1 present[WH sim44 GEWB sim91]State 1 occurs in Palpigradi Other chelicerates have abicondylar articulation and are typically equipped

with antagonistic muscles (Van der Hammen 1985Shultz 1989) but see 60 for an exception62 Superior trochanterndashfemur muscle (or homologue)originating broadly in femur inserting on distal mar-gin of trochanter 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 19892001)63 Basifemurndashtelofemur joint of appendages III andIV (= arachnid legs 1 and 2) in adult 0 absent 1present [S 25 WH 68 GEWB 86]The basifemur and telofemur apparently develop aslsquosisterrsquo articles through division of a parental articlethe femur (eg opilioacariform endeostigmatidpalaeosomatid mites) (Coineau amp Van der Hammen1979 Evans 1992) A single animal can have all legswith divided or undivided femora or have some com-bination of the two typically with the more posteriorlegs having the divided femora Thus anterior legsoften tend to be paedomorphic with respect toposterior legs Most chelicerates have undivided fem-ora on some or all legs but muscles homologous withthose of the basifemurndashtelofemur joint may stilldevelop (66) (Shultz 1989) suggesting that the devel-opment of cuticular and muscular components of ajoint have a degree of developmental and evolutionaryindependence

Undivided femora occur in most extant euchelicer-ates including Xiphosura Palpigradi AraneaeAmblypygi Thelyphonida Schizomida ScorpionesOpiliones Pseudoscorpiones higher Sarcoptiformes(Archegozetes Van der Hammen 1989) many Prostig-mata (Tetranychus Evans 1992) and throughout Par-asitiformes Divided femora occur in all legs in manyearly divergent Acariformes including many Endeo-stigmata basally divergent Sarcoptiformes (egPalaeacarus) and Prostigmata (eg AllothrombiumMicrocaeculus) (Van der Hammen 1989 Evans1992) Only leg 4 has basi- and telofemora in Alycus(Van der Hammen 1989) Basi- and telofemora occuron legs 3 and 4 in Solifugae Ricinulei Opilioacari-formes and Eurypterida (Clarke amp Ruedemann 1912Selden 1981 Shultz 1989)

Giribet et al (2002) coded the lsquoannulusrsquo of certainTrigonotarbida (Shear et al 1987) as a basifemur buta similar non-muscularized annulus is present in theposterior legs of extant Thelyphonida In both casesthe condyles of the trochanterndashannulus and annulusndashfemur joints are aligned in such away that they per-form essentially identical movements and appear notto enhance either the range or degree of leg move-ment [This contrasts with the condition in the lasttwo pairs of legs in Baltoeurypterus where condyles ofthe apparent trochanter-basifemur and basifemurndashtelofemur joints have different arrangements (Selden1981)] The annulus in Thelyphonida appears to func-tion in maintaining constant volume of the tro-

252 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

chanterndashfemur joint by preventing ballooning of thearthrodial membrane thereby minimizing the effect ofinternal fluid pressure on movement64 Basifemurndashtelofemur joint of appendage V(= arachnid leg 3) in adult 0 absent 1 present See 6365 Basifemurndashtelofemur joint of appendage VI(= arachnid leg 4) in adult 0 absent 1 present See6366 Cuticular differentiation of basifemur-telofemurjoint absent but muscles present 0 absent 1 present- inapplicable due to presence of joint (63ndash65)State 1 occurs in appendages of postoral somites IIIndashIV (= arachnid legs 1ndash4) in Xiphosura AraneaeAmblypygi Thelyphonida Schizomida Scorpionesand Pseudoscorpiones (Shultz 1989)67 Circumfemoral ring 0 absent 1 presentThe femora of one or more pairs of legs in anactinot-richid Acari have a basal groove or ring associatedwith slit sensilla Acarologists term that portionproximal to the ring the lsquobasifemurrsquo and that portiondistal to the ring the lsquotelofemurrsquo and they refer tothe basifemur as the lsquosecond trochanterrsquo and thetelofemur as the lsquofemurrsquo Despite the confusingterminology the basifemur discussed in 63ndash65 is nothomologous with the lsquobasifemurrsquo of opilioacariformand parasitiform Acari as both a true basifemurndashtelofemur joint and a circumfemoral ring are presentin legs 3 and 4 of Opilioacariformes (Van der Ham-men 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI PATELLA AND FEMURndashPATELLA JOINT

68 Patella of appendage of postoral somite III(= arachnid leg 1) proportionally much longer thanthose of more posterior appendages 0 absent 1presentState 1 occurs in Thelyphonida Schizomida (Hansenamp Soslashrensen 1905 Shultz 1989) and Opilioacari-formes (Van der Hammen 1989)69 Femurndashpatella joint 0 monocondylar several axesof movement and multifunctional muscles 1 bicondy-lar hinge one axis of movement and antagonistic mus-cles 2 hinge one axis of movement flexor muscleswithout muscular antagonists [S sim26 WH sim69 GEWBsim87]State 0 occurs in Solifugae State 1 occurs in Phalang-ida (Opiliones) Scorpiones and PseudoscorpionesState 2 occurs in the remaining taxa (Shultz 1989)70 Patellar plagula 0 absent or with simple medianattachment 1 symmetrical Y-shaped with long prox-imal stem 2 symmetrical U-shaped (= arcuate scler-ite) 3 asymmetrical attaching to patella only atanterior margin [S 27 WH sim70 GEWB 234 = 243]State 1 occurs in extant Xiphosura (Shultz 1989)State 2 occurs in Araneae (Manton 1958 Ruhland amp

Rathmayer 1978 Clarke 1986 Shultz 1989) State 3occurs in Amblypygi Thelyphonida and Schizomida(Shultz 1989) Giribet et al (2002) included this char-acter twice

APPENDAGES OF POSTORAL SOMITES IIIndashVI TIBIA AND PATELLAndashTIBIA JOINT

71 Tibiae divided by one or more joints 0 absent 1presentState 1 occurs in Amblypygi (Weygoldt 2000)72 Patellandashtibia joint 0 monocondylar with or with-out CZY (73) 1 bicondylar hinge one axis of move-ment antagonistic muscles 2 hinge one axis ofmovement muscles without muscular antagonists [Ssim31 WH 74 GEWB sim92] (Shultz 1989 Selden Shearamp Bonamo 1991)73 Patellandashtibia joint with posterior compression zone(lsquoCZYrsquo) 0 absent 1 present - inapplicable due toabsence of monocondylar articulation [GEWB 99]State 1 is a unique synapomorphy of Araneae (Seldenet al 1991)74 Patellandashtibia joint of appendages IIIndashVI(= arachnid legs 1ndash4) with deep-set monocondylar pivotbordered by a pair of tibial processes to which extensormuscles attach 0 absent 1 present - inapplicabledue to absence of monocondylar joint (72)State 1 occurs in extant Xiphosura (Snodgrass 1952Shultz 1989) although a similar arrangement ispresent in the appendage III (= arachnid leg 1) in The-lyphonida (Shultz 1993)75 Patellandashtibia joint largely immobile specializedfor autotomy 0 absent 1 presentState 1 occurs throughout Amblypygi (Weygoldt 19842000)76 Anterior femurndashtibia or femoropatellandashtibia(transpatellar) muscle 0 absent 1 present [S sim30GEWB 222] (Shultz 1989)77 Proximal attachment of posterior femurndashtibia orfemoropatellarndashtibia (transpatellar) muscle 0 muscleabsent 1 dorsoposterior surface of femur andor pos-terior surface of patella 2 distal process of femur mus-cle acting as extensor of femur-patella joint [S sim28 WHsim71 + 72 GEWB 235 = 244 245] (Shultz 1989)78 Distal attachment of posterior femurndashtibia (trans-patellar) muscle 0 posterior 1 dorsal acting as exten-sor of patellandashtibia joint - inapplicable due to absenceof muscle (77) [S 29 GEWB 221] (Shultz 1989)79 Anterior patellandashtibia muscle 0 absent 1 present[S sim32 WH sim73 GEWB 246] (Shultz 1989)80 Posterior patellandashtibia muscle 0 absent 1 present[S 33 WH 76 GEWB 248] (Shultz 1989)81 Patellandashtibia joint spanned by elastic (lsquospringlikersquo)sclerite 0 absent 1 present [GEWB 107]State 1 is a unique synapomorphy of Solifugae (Sens-enig amp Shultz 2003)

PHYLOGENY OF THE ARACHNID ORDERS 253

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

APPENDAGES OF POSTORAL SOMITES IIIndashVI TARSUS AND TIBIAndashTARSUS JOINT

82 Tarsus divided into proximal basitarsus(= metatarsus) and distal telotarsus (= distitarsus orlsquotarsusrsquo) 0 absent 1 presentState 0 occurs in extant Xiphosura (Shultz 1989) andthroughout Acariformes (Lindquist 1984 Evans1992) in the anterior two pairs of legs in chthonioidpseudoscorpions and all legs in lsquomonosphyronidrsquo pseu-doscorpions including Feaelloidea and Cheliferoidea(Chamberlin 1931) A circumtarsal ring (83) associ-ated with slit sensilla in Parasitiformes (Acari) mayrepresent a joint between a telotarsus and basitarsus(Evans 1992)83 Circumtarsal ring 0 absent 1 presentState 1 is an apparent synapomorphy of Parasiti-formes (Evans 1992)84 Telotarsus in adult with two or more tarsomeres 0absent 1 present - inapplicable due to absence of telo-tarsus (82) [GEWB 71]State 1 occurs in Schizomida Thelyphonida Ambly-pygi Palpigradi Ricinulei Solifugae (Shultz 1989)Opilioacariformes (Van der Hammen 1989) and Ple-siosiro (Dunlop 1999)85 Three telotarsomeres on appendages of postoralsomites IV-VI (= arachnid legs 2ndash4) 0 absent 1present - inapplicable due to absence of telotarsus (82)or absence of telotarsomeres (84) [S 34 WH 77 GEWB102]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Shultz 1989)86 Tibiandashtarsus joint spanned by well-developed elas-tic (lsquospringlikersquo) sclerite 0 absent 1 present [GEWB107]State 1 occurs in Scorpiones phalangid Opiliones andSolifugae (Alexander 1967 Shultz 2000 Sensenig ampShultz 2003) Extant Xiphosura is coded for the lastprosomal appendage only (see 88)87 Appendage of postoral somite VI (= arachnid leg 4)with ring of large basally articulated spatulate pro-cesses at tibia-tarsus joint 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)88 Appendages postoral somites IIIndashV (= arachnidlegs 1ndash3) with tibiotarsus (ie tibia and tarsus not dif-ferentiated) 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 1989) butnot apparently in Weinbergina (Moore et al 2005) theonly synxiphosurid in which appendages are suffi-ciently preserved to determine the number ofpodomeres89 Ambulacrum peduncle-like extension of the tarsuswith internal condylophores and terminating distallywith apotele (eg claws) andor pulvillus 0 absent 1present

State 1 is present throughout Acari (Alberti amp Coons1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI APOTELE AND TARSUSndashAPOTELE JOINT

90 Apotele of appendage III (= arachnid leg 1) 0absent or not apparent 1 present [GEWB 101]State 0 occurs in Amblypygi Thelyphonida and Schi-zomida (Shultz 1999) However the apotele is proba-bly present but undifferentiated because muscles thatwould normally insert on the apotele terminate on theend of the lsquotarsusrsquo Dunlop (1999) reconstructed Hap-topoda as lacking an apotelic claw on leg 1 but noted inthe text that this was speculative Dunlop (2002b) hasreviewed apotelic diversity in Chelicerata91 Appendages of postoral somites IIIndashVI (= arachnidlegs) chelate with chela formed from tibiotarsus andapotele or tarsus and apotele 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988) The condition appears to occur in Chasmataspisbased on associated appendages (Dunlop et al 2004)but not in Diploaspis and Octoberaspis (Dunlop et al2001 Dunlop 2002a)92 Apotele with eversible or padlike empodium(= pulvillus) in adult 0 absent although empodialclaw may be present 1 present [S sim38 WH 81 GEWB95]State 1 occurs throughout Pseudoscorpiones (Cham-berlin 1931) Solifugae (Roewer 1934) Opilioacari-formes Holothyrida Ixodida and free-livingMesostigmata (Evans 1992) An eversible lsquopulvillusrsquo isprobably a primitive feature of Amblypygi (eg Char-inus Weygoldt 2000) but is absent in higher groups(eg Phrynus Weygoldt 2000)93 Inferior apotele muscle (= claw depressor) with tib-ial attachment 0 absent 1 present [S 35 WH 78GEWB 249]The character is coded for extant Xiphosura usingappendage of postoral somite VI (= arachnid leg 4) asother legs lack a differentiated tibia (88) State 1appears to occur throughout Arachnida (Shultz 1989)94 Inferior apotele muscle (= claw depressor) withpatellar attachment 0 absent 1 present [S 36 WH 79GEWB 250]State 0 occurs in extant Xiphosura Acari AmblypygiPalpigradi and Ricinulei State 1 occurs in all otherarachnids examined here (Shultz 1989)

OPISTHOSOMA

95 Number of opisthosomal somites in adult 0 five1 eight 2 nine 3 10 4 11 5 12 6 13 [WH sim33GEWB sim190]Chasmataspidids appear to have 13 opisthosomalsomites (Dunlop et al 2004) The synziphosurans

254 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Weinbergina (Moore et al 2005) and Limuloides(Anderson amp Selden 1997) appear to have ten opistho-somal somites Moore et al (2005) found no evidence ofan anterior microtergite in Weinbergina as proposedby Anderson amp Selden (1997) Xiphosurids appear tohave nine opisthosomal somites (Scholl 1977 Ander-son amp Selden 1997 Shultz 2001) Members of the fol-lowing orders have 12 opisthosomal somites Araneae(Millot 1949ac) Amblypygi (Weygoldt 2000) Thely-phonida and Schizomida (Hansen amp Soslashrensen 1905)and Pseudoscorpiones (Chamberlin 1931) Elevensomites are present in Palpigradi Solifugae (Roewer1934) and Trigonotarbida (Gilboarachne Shear et al1987 Palaeocharinus Fayers et al 2004) Plesiosiromay have 11 or 12 somites (Dunlop 1999) Justifica-tion for coding of more controversial taxa is providedbelow

Scorpiones Two principal hypotheses regarding thenumber of opisthosomal somites in scorpions havebeen advocated a 13-somite hypothesis derived fromembryological studies (Brauer 1895 Patten 1912Farley 1999 2005) and a 12-somite hypothesis basedon comparative anatomy of adults (Weygoldt amp Pau-lus 1979) The embryological interpretation is basedon the observation of pregenital genital and pectinalsomites (each with segmental ganglia and paired limbbuds) in early scorpion embryos followed by extremereduction or loss of the pregenital somite in laterembryos According to this view a missing pregenitalsomite should be added to the 12 apparent opisthoso-mal somites of post-embryonic scorpions to achieve afinal number of 13 The anatomy-based hypothesiswas introduced by Weygoldt amp Paulus (1979) whoadvocated a literal interpretation of post-embryonicsegmentation based on opisthosomal tergites Specifi-cally these authors argued that the last pair of dorsalendosternal suspensor muscles of non-scorpionarachnids especially Pedipalpi attach to the first(= pregenital) somite that this condition also occurs inscorpions and that there is no reason to invoke a miss-ing pregenital tergite They proposed that the pectinesbelong to the genital somite and that functional spe-cializations of the nervous system for pectinal functiongive the apparence of an extra neuromere duringembryonic development

I recently dissected the prosoma and anterioropisthosoma of the scorpions Centruroides Hadrurusand Heterometrus and focused on the composition ofthe muscular diaphragm (103) that separates thehaemocoelic compartments of the prosoma andopisthosoma (Shultz 2007) The diaphragm is com-posed of a metameric series of axial muscles fromthree somites the anterior somite corresponds to thelast prosomal somite and the posterior somite corre-sponds to the genital somite The middle elements

insert dorsally along a tranverse tendon attached tothe anterior margin of the first tergite These obser-vations are consistent with the embryological inter-pretation that a pregenital somite is present but itstergite is not expressed It appears likely that the thepregenital somite was compressed longitudinally dur-ing the evolution of the diaphragm I code scorpions ashaving 13 opisthosomal somites

Eurypterida The eurypterid opisthosoma is widelyassumed to have 12 somites (Clarke amp Ruedemann1912 Stoslashrmer 1944) However in a speculative paperon the evolutionary morphology of trilobites and che-licerates Raw (1957) proposed that both scorpions andeurypterids have 15 opisthosomal somites Rawassumed that scorpions have 13 apparent opisthoso-mal somites based on the transient pregenital somiteof scorpion embryology that scorpions and eurypteridsare close relatives and should have the same numberof somites and that olenellid trilobites and chelicer-ates always have somites in multiples of three Rawachieved 15 somites in scorpions by assuming the lastopisthosomal somite to be a diplosomite and that thetelson is a postanal somite even though there is noevidence for either of these proposals He attributedthese features to eurypterids as well He also notedthat the connection between the prosomal carapaceand first opisthosomal tergite in Eurypterida differedstructurally from the connection between adjacentopisthosomal tergites and regarded this as evidencefor a reduced pregenital tergite in the prosomandashopisthosoma junction Rawrsquos speculations were largelyforgotten until Dunlop amp Webster (1999) resurrectedthe proposal that eurypterids have a reduced opistho-somal tergite and therefore share a unique similaritywith scorpions Unfortunately Dunlop amp Bullocktreated Rawrsquos conjecture as if it were based on empir-ical evidence rather than an attempt to force euryp-terid morphology into a peculiar numerical system Inthe absence of convincing evidence to the contrary Ihave coded eurypterids as having 12 opisthosomalsomites

Ricinulei The opisthosoma consists of a membranouspedicel (97) bearing the gonopore ventrally Thefemale gonopore is bordered by an anterior plate and aposterior plate The remainder of the opisthosoma iscomposed of thick sclerites separated by less heavilysclerotized cuticle The first dorsal sclerite is short andfunctions as part of a prosomandashopisthosoma couplingmechanism (96) that may or may not be a specializedcomponent of the following tergite This is followed byfour tergites and sternites and a three-segmentedmetasoma (= lsquopygidiumrsquo) Millot (1945 1949f) rea-soned that the opisthosoma contains ten somites withsomites VIIndashIX incorporated into the pedicel XndashXIIIexpressed as tergites and sternites and the metasoma

PHYLOGENY OF THE ARACHNID ORDERS 255

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

comprising four somites He did not regard the dorsalcoupling sclerite as a separate tergite and his inter-pretation of four rather than three metasomal somiteshas been rejected Pittard amp Mitchell (1972) also pro-posed ten opisthosomal somites but achieved thisnumber by regarding the dorsal coupling sclerite as atergite of somite IX and by recognizing three metaso-mal somites Legg (1976) adopted the system proposedby Pittard amp Mitchell but did not regard the dorsalcoupling sclerite as separate from the following tergite(X) Van der Hammen (1979 1989) reconstructed 13somites The pedicel was regarded as having twosomites (VII + VIII) the following four sets of tergitesand sternites were considered diplosomites (IX + XXI + XII XIII + XIV XV + XVI) and the metasoma hadthree somites (XVIIndashXIX) This interpretation mayhave been biased by Van der Hammenrsquos attempt tounite Ricinulei with anactinotrichid Acari which healso regarded as having a primitive number of 13somites Selden (1992) described a fossil ricinuleidTerpsicroton that shows two pairs of depressions onthe three large premetasomal tergites which con-trasts with the single pair seen in extant species Thisobservation appears to corroborate the diplosomitehypothesis but the evidence does not indicate that thetergite anterior to these is a diplosomite

Dunlop (1996) attempted to homologize theprosomandashopisthosoma coupling mechanisms ofRicinulei and Trigonotarbida a goal that required anovel and rather forced interpretation of the dorsalsclerites He regarded the coupling sclerite as homol-ogous with the first opisthosomal tergite (VII) of trig-onotarbids and then followed Van der Hammenrsquosdiplosomite hypothesis to achieve 12 somites in totalDunloprsquos scheme differs from previous systems in sug-gesting that the pedicel does not contain the dorsalelements of the first and second somites and is in-consistent with Millotrsquos (1945) observation that thepre- and postgenital plates each have dorsoventralmuscles Here I code Ricinulei as having 12 somitesThere are three metasomal somites (XVIndashXVIII)three diplosomites (= six somites) (XndashXV) one cou-pling somite (IX) and two somites in the pedicel (iethe genital and pregenital somites) (VII VIII)

Opiliones Harvestmen have nine opisthosomalsomites and an anal operculum that is traditionallyregarded as the tergite of a tenth somite a tenth ster-nite is lacking (Hansen amp Soslashrensen 1904 Winkler1957) However the anal operculum appears to repre-sent a persistent embryonic telson (Moritz 1957) andis therefore likely to be a postsegmental structurecomparable with the stinger of scorpions flagella ofthelyphonids etc (Shultz 2000)

Acari The number of opisthosomal somites is probe-matic for most mite taxa due to uncertainty about the

location of the prosomandashopisthosoma boundry (5)extensive simplification or loss of metamericallyarranged sclerites and muscle attachments a paucityof developmental studies of engrailed expression andin Acariformes opisthosomal anamorphosis and het-erochronic modification of somite number (Evans1992) Ixodids appear to be the exception developmen-tal studies indicate five somites in the opisthosoma ofticks (Evans 1992)

Some mites retain external evidence of segmenta-tion ndash metameric patterns of furrows muscle attach-ments and slit sensilla ndash and with certainassumptions the number of opisthosomal somites canbe estimated Here it is assumed that the dorsal sur-faces of the last two prosomal somites are present inmites and retain their primitive association with legs3 and 4 (contra Van der Hammen 1989) (see 5 for jus-tification) Given this there appear to be 11 somites inthe opisthosoma of Opilioacariformes a conclusionalso reached by other workers (eg With 1904 Sitni-kova 1978 Klompen 2000) Similar reasoning sug-gests that Alycus has seven opisthosomal somites notnine as advocated by Van der Hammen (1989) Unfor-tunately external evidence is ambiguous for determin-ing the number of somites in other mite lineages andthese are coded as uncertain96 Prosoma-opisthosoma coupling mechanism 0absent 1 present [GEWB sim24]The posterior margin of the carapace and the anteriormargin of the first apparent opisthosomal tergite arespecialized as a coupling mechanism in Ricinulei andTrigonotarbida and Dunlop amp Horrocks (1996) pro-posed that these are synapomorphic for the twogroups However there is uncertainty about thehomology of the anterior opisthosomal somites in Trig-onotarbida and Ricinulei (95)97 Pedicel 0 absent 1 aranean type 2 ricinuleidtype [WP 30 S sim40 WH sim20 GEWB sim126]The body narrows at or near the prosomandashopisthosoma juncture in several arachnid lineages (ieSolifugae Palpigradi Amblypygi Araneae Ricinulei)and this lsquowaistrsquo has often been used as character at theinterordinal level (eg Pocock 1893) However thisinterpretation is rejected here for being subjective andfor uniting non-homologous conditions For exampleAraneae and Amblypygi are often grouped on the basisof a lsquopedicelrsquo yet it is a highly specialized structure inAraneae and its parts are not readily homologizedwith those of Amblypygi In contrast the condition inAmblypygi is a slightly narrower version of the highlymoveable prosomandashopisthosoma juncture in Uropygiwhich is not generally considered a lsquopedicelrsquo Thepedicel in Ricinulei is also unique a weakly sclero-tized stalk containing the genital opening (Cryptocel-lus Pittard amp Mitchell 1972 Ricinoides Legg 1976)See 95

256 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

98 Opisthosoma with three- segmented lsquobucklerrsquo 0absent 1 presentState 1 occurs throughout Chasmatapsidida (Dunlop2002a)99 Thoracetron consolidation of tergites of postgeni-tal somites 0 absent 1 present - inapplicable codedonly for Xiphosura (Anderson amp Selden 1997)100 Fusion of tergites of postoral somites VIII and IX(= opisthosomal somites 2 and 3) only 0 absent 1present [GEWB 146]State 1 appears to be a synapomorphy Trigonotarbida(Shear et al 1987) Ricinuleids also have diplotergites(101) but these appear to encompass a different com-bination of somites (see 95)101 Three diplotergites 0 absent presentState 1 is a unique synapomorphy of Ricinulei (Selden1992) See 95102 Paired opisthosomal defensive glands openingvia ducts on either side of the anus 0 absent 1present [S 46 WH 86 GEWB 122]State 1 occurs in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905)103 Muscular diaphragm separating prosomal andopisthosomal compartments formed by dorsoventralmuscles of postoral somites VIndashVIII and extrinsic mus-cles of leg 4 0 absent 1 presentState 1 is known only in extant scorpions (Lankesteret al 1885 Firstman 1973) The diaphragm is oftentreated as a single structure but recent anatomicalwork (Shultz 2007) has shown that it is a composite ofdorsal endosternal suspensors and extrinsic leg mus-cles (see 95) A diaphragm is also present in Solifugaebut it is located more posteriorly and does not appearto be homologous with that of scorpions (Roewer1934)104 Opisthosomal appendicular chondrites 0absent 1 presentThese cartilage-like columns of mesodermally derivedtissue are associated with each opisthosomal append-age including the chilaria in extant Xiphosura(Patten amp Hazen 1900 Yamasaki et al 1988 Fahr-enbach 1999 Shultz 2001)105 Paired appendages on ventral surface of postoralsomite VII (= opisthosomal somite 1) in adult 0absent 1 present [WP sim13 S 39 WH 82 GEWB 143]State 1 is known in Weinbergina (Moore et al 2005)and extant Xiphosura The bilobed structure of themetastoma in certain eurypterids as well as the cor-responding placement of chilaria in Xiphosura haveled some workers to regard the metastoma as fusedappendages of postoral somite VII (eg Stoslashrmer1955) This hypothesis is problematic given that (i) nospecial explanation is needed to account for bilaterallysymmetrical structures in bilaterally symmetricalorganisms (ii) incorporation of the ventral part of thefirst opisthosomal sternite as a functional element of

the prosoma is typical of arachnids and perhapsEuchelicerata generally and (iii) some workers regardthe metastoma as a sternite (eg Jeram 1998) Thusthe metastoma is coded here as uncertain Dunlop hasproposed that the sternum of scorpions is derivedfrom appendages citing the presence of transient limbbuds in scorpion development (Brauer 1895 Patten1912) but this is not compelling evidence for a persis-tent appendicular contribution to the sternite of theadult106 Megoperculum 0 absent 1 present [S sim41 + 42WH sim83 GEWB 162 + 163]State 1 occurs in Thelyphonida Schizomida Ambly-pygi Araneae (Shultz 1993 1999) Trigonotarbida(Shear et al 1987) and probably Haptopoda (Dunlop1999) and Palpigradi The megoperculum consists ofappendages of opisthosomal somite 2 (= genital somiteor postoral somite VIII) (Shultz 1993 1999 Popadicet al 1998) that have fused medially and displacedthe ventral body wall of the somite anteriorly (oftenrepresented by a small sternite to which dorsoventralmuscles of postoral somite VIII attach) The megoper-culum projects posteriorly to form the ventral surfaceof a pregenital chamber The dorsal surface of thechamber is formed by the ventral body wall of opistho-somal somite 3 (= postoral somite IX) which serves asthe ventral attachment of the paired dorsoventralmuscles of that somite The megoperculum bears book-lungs (where present) and sometimes paired gonopodsthat are probably derived from telopodites A similarbut less developed operculum is present on theopisthosomal somite 3 which may also bear booklungsand eversible vesicles (eg Amblypygi) (see 111) corre-sponding to the booklungs and gonopods of the mego-perculum (Shultz 1999)

I contend here that a megoperculum is present inPalpigradi and that this is particularly evident infemales Specifically a large unpaired lobe projectsposteriorly from the genital somite over the ventroan-terior surface of the first postgenital somite therebyforming a pregenital chamber The dorsoventral mus-cles of the genital somite attach ventrally near theanterior border of postoral somite VIII and those ofpostoral somite IX attach to the upper surface of thepregenital chamber (Rucker 1901 Boumlrner 1902bRoewer 1934 Millot 1949d Van der Hammen 1989)Weygoldt (1998) has questioned this interpretation ofPalpigradi107 Postgenital operculum or lsquosternitersquo 0 sclerotized1 not sclerotized - inapplicable coded only for Pan-tetrapulmonata (Platnick amp Gertsch 1976)State 1 is present in Trigonotarbida Amblypygi The-lyphonida Schizomida and Mesothelae (Araneae) butnot Opisthothelae (Araneae)108 Genital opening of female guarded by four plates(one pregenital one postgenital two laterogenitals)

PHYLOGENY OF THE ARACHNID ORDERS 257

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

genital opening of male guarded by two plates 0absent 1 presentThis an apparent synapomorphy of Holothyrida(Acari) (Van der Hammen 1989)109 Anterior margin of genital opening in male withglands secreting via fusules 0 absent 1 presentState 1 occurs in Araneae as epigastric or lsquoepiandrousrsquoglands (Marples 1967) and in Palpigradi (Condeacute1991b)110 Paired valve-like plates apparently formed fromcomponents of three somites covering triradiate genitalopening 0 absent 1 presentThis is an apparent synapomorphy of Acariformes(Evans 1992 Alberti amp Coons 1999)111 Eversible lsquoappendagesrsquo on the ventral surface ofpostgenital somites 0 absent 1 present [S sim43 WHsim84 GEWB sim127]Paired ventral lsquoappendagesrsquo operated in part byhaemolymph pressure occur in the form of ventralsacs in certain Amblypygi (postoral somite IX) (Wey-goldt 2000) (Charinus Millot 1949b not PhrynusShultz 1999) prokoenenian Palpigradi (postoralsomites XndashXII) (Rucker 1901 Condeacute 1991a) as spin-nerets in Araneae (XndashXI) (Shultz 1987) and as geni-tal papillae in many Acariformes (Alberti amp Coons1999) including Oribatida many Endeostigmata (Aly-cus Van der Hammen 1989) and Prostigmata (Allo-thrombium Saboori amp Kamali 2000 MicrocaeculusEvans 1992) Eversible vesicles similar to those ofamblypygids occur in fossils of the trigonotarbidPalaeocharinus (postoral somite IX) (Fayers et al2004) Two cuticular structures (lsquogenital verrucaersquo)occur anterior to the genital opening in Opilioacari-formes and each covers the opening to a thin-walledinvaginated sac (lsquogenital papillarsquo) of unknown function(Van der Hammen 1989) Van der Hammen proposedthat it is homologous to a genital papilla of Acari-formes but Alberti amp Coons (1999) have questionedthis interpretation on several grounds and refer tothese structures as pregenital capsules112 Opisthosomal silk glands and spinnerets derivedfrom appendages on postoral somites X and XI(= opisthosomal somites 4 and 5) 0 absent 1 present[WP 33 WH 41 GEWB 123 + 142]State 1 is a unique synapomorphy of Araneae (Plat-nick amp Gertsch 1976) although potentially homolo-gous glands and spigot-like setae occur on the ventralopisthosomal surface in Palpigradi (Millot 1943Condeacute 1991a)113 Opisthsosomal spinnerets location 0 near mid-dle of opisthosoma 1 near posterior end of opisthso-som ndash inapplicable coded only for Araneae (Platnickamp Gertsch 1976)114 Anterior medial lsquospinneretsrsquo 0 absent 1 present- inapplicable coded only for Araneae (Platnick ampGertsch 1976)

115 Opisthosomal tergites divided longitudinally intoone median and two lateral plates 0 absent 1 present[GEWB 145]State 1 occurs in non-curculioid Ricinulei (Selden1992) and many Trigonotarbida (Dunlop 1996) Thisfeature or something very like it (eg distinct opistho-somal trilobation) occurs sporadically in severaleuchelicerate groups including Chasmataspis (Chas-mataspidida) (Dunlop et al 2004) most fossil xipho-surans (Anderson amp Selden 1997) and a few derivedEurypterida (especially Mixopteroidea) (Tollerton1989)116 Number of metasomal somites 0 zero 1 two 2three 3 five 4 nine [S sim44 WH sim85 GEWB 128 144]The metasoma is a preanal region comprising multiplesomites that are substantially narrower than the pre-ceding somites or mesosoma Metasomal somitesoften lack pleural membranes and take the form ofsclerotized rings State 1 occurs in TrigonotarbidaState 2 is present in synziphosurid Xiphosura (Ander-son amp Selden 1997) Amblypyi Thelyphonida Schizo-mida and Ricinulei State 3 occurs in Scorpiones andEurypterida and State 4 occurs in Chasmatispidida

Some workers appear to consider a five-segmentedmetasoma as conclusive evidence for the monophylyof a Eurypterida + Scorpiones clade (eg Dunlop ampBraddy 2001) presumably because the character rep-resents a kind of tagmosis and should therefore begiven substantial phylogenetic weight However Wey-goldt (1998) has pointed out that a metasoma contain-ing three somites is probably part of the ground planof Euchelicerata The known diversity of the eucheli-cerate metasoma indicates that this feature can inincrease or decrease its segmental composition in evo-lution or be eliminated entirely without necessarilychanging the total number of opisthosomal somitesConsequently it is problematic to assume that thefive-segmented metasoma has any special immunityto homoplasy117 Postanal structure (telson) 0 absent or not obvi-ously developed 1 presentPostanal structures are considered non-segmentalbecause they are modifications of the embryonicregion posterior to the site of somite addition (embry-onic growth zone) A well-developed postanal struc-ture persists in adults in Xiphosura EurypteridaChasmataspidida Scorpiones Schizomida Thely-phonida and Palpigradi The anal operculum of Opil-iones is generally considered by morphologists to bethe tergite of the tenth opisthosomal or anal somite(eg Hansen amp Soslashrensen 1904) but embryological evi-dence (Moritz 1957) suggests that it is derived fromthe postproliferative zone and thus corresponds to thetelson of other chelicerates (see 95)118 Postanal structure shape 0 caudal spine 1aculeus 2 flagellum 3 anal operculum - inapplica-

258 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ble due to absence of postanal structure (117) [WP sim41S sim45 WH sim24 GEWB sim121 129 + 147]State 0 is present in Xiphosura Eurypterida andChasmataspidida State 1 occurs in Scorpiones State2 occurs in Palpigradi Thelyphonida and SchizomidaState 3 occurs in Opiliones (see 95 117)119 Specialized postanal flagellum in male (see 157ndash159) 0 absent 1 present - inapplicable coded onlyfor taxa with postanal flagellum (118) [GEWB 131]State 1 is a unique synapomorphy of Schizomida(Hansen amp Soslashrensen 1905 Cokendolpher amp Reddell1992 Reddell amp Cokendolpher 1995)

RESPIRATORY SYSTEM

120 Respiratory medium 0 water 1 airAmong the terminal taxa included only fossil scorpi-ons are problematic for this character They are codedhere as uncertain Based on a study of book lungmicrosculpture Scholtz amp Kamenz (2006) have arguedthat arachnids are primitively terrestrial and pulmo-nate (see also Firstman 1973) and have questionedwhether any fossil scorpions were aquatic No positionon this proposal is taken here121 Respiratory lamellae on opisthosomal somite 2(= genital somite postoral somite VIII) 0 absent 1present [WP sim37 S sim51 WH sim22 GWEB sim133ndash137]This character encompasses book gills (= lamellae thatfunction in water) and book lungs (= lamellae thatfunction in air) The distinction between book gills andbook lungs is accommodated here by 120 in combina-tion with 121ndash124 Respiratory lamellae are presenton the genital somite in Trigonotarbida AraneaeAmblypygi Thelyphonida and Schizomida Petrunk-evitch (1949) reconstructed Plesiosiro as having booklungs but Dunlop (1999) could not corroborate thisThis character was used by Dunlop amp Webster (1999)to propose that Xiphosura and Scorpiones are closelyrelated because they both lack respiratory lamellae onthe genital somite Dunlop amp Braddy (2001) alsoargued for the placement of Eurypterida with Xipho-sura and Scorpiones based in part on this characterAt least some eurypterids may have had respiratorylamellae (eg Manning amp Dunlop 1995) but the onlyevidence of their segmental distribution is derivedfrom one specimen (Braddy et al 1999)122 Respiratory lamellae on opisthosomal somite 3(= postoral somite IX) 0 absent 1 present [GEWB136]State 1 occurs in Amblypygi Thelyphonida and Xipho-sura It is a groundplan feature of Araneae (Platnick ampGertsch 1976) and is present in all representativetaxa included here Braddy et al (1999) proposed theexistence of respiratory lamellae on postoral somitesIXndashXII in Eurypterida based on evidence from onespecimen This character is coded as uncertain for the

Eurypterida included here Extant scorpions are codedhere as lacking respiratory lamellae on this somite(= pectinal somite) based on the interpretation of theopisthosomal segmentation discussed in 95123 Respiratory lamellae on opisthosomal somites 4ndash6 (= postoral somites XndashXII) 0 absent 1 presentState 1 is definitely present in extant Xiphosura andScorpiones and may have occurred in Eurypterida124 Respiratory lamellae on opisthosomal somite 7(= postoral somite XIII)State 1 occurs in extant Xiphosura and Scorpionesonly125 Kiemenplatten 0 absent 1 presentThese structures are located on the roof of opercularchambers in Eurypterida (Clarke amp Ruedemann1912) where they take the form of ventrally projectingcones with a distinct cuticular microsculputure (Man-ning amp Dunlop 1995) Dunlop amp Braddy (2001)inferred the existence of Kiemenplatten in all Palaeo-zoic scorpions based on a description and photos of onespecimen of Paraisobuthus duobicarinatus by Kjelles-vig-Waering (1986 pls 16ndash18) The plates depict darkcone-like rearward-pointing denticles distributedwithin a white amorphous material The denticlesappear to lack microsculpture even though the mag-nifications at which the photos were taken (times90ndashtimes330)are comparable with those illustrating the cones ofKiemenplatten (times170ndashtimes350 in Manning amp Dunlop1995 figs 1 2) Given the diversity of cuticular struc-tures present in the atria of booklungs book gills andtracheae the denticles appear to bear no special sim-ilarity to the cones of Kiemenplatten126 Tracheal system 0 absent 1 paired ventral stig-mata on postoral somite VIII (= opisthosomal somite2) 2 paired ventral stigmata on postoral somites IXand X 3 one pair of stigmata opening near legs 3 or 44 paired stigmata associated with chelicerae 5 fourpairs of stigmata on dorsal surface of opisthosoma -inapplicable aquatic (120) [WP 40 sim43 S sim52 + 5354 WH sim23 45 88 GEWB sim138 139]Firstman (1973) and Weygoldt amp Paulus (1979)hypothesized that tracheae are homologous in all tra-cheate arachnids except spiders and subsequentworkers have entertained this hypothesis by includinga character for the presenceabsence of tracheal sys-tems that ignores the diverse arrangement of stig-mata in arachnids (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) However thisapproach assumes that internal tracheal systems areconserved but that tracheal openings (stigmata)appear and disappear on different parts of the bodywith higher evolutionary frequency Here I assumethat stigmata are conserved in evolution and that dif-ferences in their anatomical placement reflect the evo-lution of new tracheae State 1 occurs throughoutOpiliones State 2 occurs throughout Solifugae

PHYLOGENY OF THE ARACHNID ORDERS 259

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

(Roewer 1934) and Pseudoscorpiones (Chamberlin1931) State 3 recognizes the possible homology of stig-mata in Ricinulei Opilioacariformes and Parasiti-formes (Acari) (Lindquist 1984 Van der Hammen1989) Tracheal systems are apparently absent in thegroundplan of Acariformes but Prostigmata typicallyhave tracheae associated with the chelicerae Tra-cheae derived from the posterior pair of book lungs arewidespread in araneomorph Araneae but the booklungs are retained and tracheae are absent in themost basally divergent groups (eg Hypochilus)

BOXndashTRUSS AXIAL MUSCLE SYSTEM (BTAMS)

127 Posterior oblique muscles of BTAMS of postoralsomites IndashVI 0 absent 1 present in one or moresomitesState 0 occurs in Xiphosura and state 1 occurs inPalpigradi Araneae Amblypygi Thelyphonida andSchizomida (Shultz 2001) The condition in AcariRicinulei Opiliones and Solifugae is not known128 Anterior oblique muscles of BTAMS posterior topostoral somite VI 0 absent 1 presentState 1 occurs in Xiphosura which is probably theprimitive condition based on comparison with otherarthropods (Shultz 2001) State 0 occurs in all arach-nids examined thus far129 Ventral attachments of posterior oblique musclesof opisthosomal BTAMS located in prosoma 0 absent1 presentState 1 occurs in extant Xiphosura (Shultz 2001)130 Endosternite fenestrate 0 absent 1 present [S 8WH 54 GEWB 35]State 1 occurs in Thelyphonida and hubbardiid Schi-zomida (Firstman 1973) The condition in protoschi-zomid Schizomida is not known131 Suboral suspensor a tendon that arises from theBTAMS and inserts on the ventral surface of the oralcavity via muscle 0 absent 1 presentState 1 occurs in Palpigradi (Eukoenenia Millot1943) Amblypygi (Shultz 1999) and Thelyphonida(Shultz 1993)132 Perineural vascular membrane in adult 0absent 1 present [WH 28 GEWB 155]Wheeler amp Hayashi (1998) coded extant Xiphosura asunknown although the primary source for this char-acter (Firstman 1973) stated that adult Limulus havea perineural vascular membrane Within Arachnidapresence of a perineural vascular membrane is appar-ently correlated with the presence of tracheae (126)133 Ventral endosternal suspensor attaching on coxaof anteriorly adjacent somite 0 absent 1 present[GEWB 239 253]State 1 occurs in Amblypygi (Shultz 1999) Thely-phonida (Shultz 1993) and Schizomida (Shultzunpubl observ)

134 Posteriormost postoral somite with a pair of dor-soventral muscles 0 VI 1 VII 2 VIII 3 XII 4 XIII5 XIV 6 XV 7 XVIDorsoventral muscles tend to run in a continuousmetameric series beginning in the prosoma Deter-mining the last tergite on which the muscle seriesends is substantially easier than counting especiallygiven frequent anatomical complexities near theprosomandashopisthosoma juncture The dorsoventralmuscle series ends on postoral somite VII inCyphophthalmi (eg Chileogovea) and Laniatores(Opiliones) (Shultz unpubl observ) on postoralsomite VII or VIII in Eupnoi (Leiobunum Shultz2000 unpubl observ) on postoral somite XII in Pal-pigradi (Roewer 1934) Solifugae (Bernard 1896Roewer 1934 Millot amp Vachon 1949) and Araneae(Liphistius Hypochilus Millot 1933 but segmenta-tion reinterpreted here following Shultz 1993 1999)on postoral somite XIII in hubbardiid Schizomida(Cokendolpher amp Reddell 1992) and Scorpiones(Lankester et al 1885) and on postoral somite XIVin extant Xiphosura (Shultz 2001) Thelyphonida(Shultz 1993) protoschizomid Schizomida (Cokendol-pher amp Reddell 1992) and Amblypygi (Shultz 1999)Dunlop (1999) reconstructed Plesiosiro as havingpaired tergal apodemes ending on postoral somite XVbut his figures indicate that the series ends on posto-ral somite XIV The interpretation of opisthosomalsegmentation in Ricinulei coded here (95) indicatesthat the series ends on postoral somite XV in Terpsi-croton and this is treated as the groundplan for theorder The posteriormost dorsoventral muscles inPseudoscorpiones occur on postoral somite XVI(Vachon 1949)

NERVOUS SYSTEM

135 Segmental ganglia 0 consolidated in prosoma1 one or more present in opisthosoma [WH 30 GEWB210 + 216]Adult neuromeres are exclusively prosomal in Ambly-pygi liphistiomorph and araneomorph Araneae Opil-iones Pseudoscorpiones Ricinulei and Acari (Millot1949a)136 Dorsal median eyes 0 absent 1 present [WP14 + 47 S 50 WH sim7 GEWB 1]Median eyes are a groundplan feature of Euchelicer-ata and are retained in extant Xiphosura Eurypteridaand occur in all arachnid orders except PalpigradiSchizomida Ricinulei and Pseudoscorpiones (Paulus1979) Median eyes are also absent in opilioacariformand parasitiform Acari (Lindquist 1984) and perhapsthe fossil whipscorpion Proschizomus (Dunlop amp Hor-rocks 19951996) They are unknown in synziphosu-ran Xiphosura Median eyes are present in a variety ofbasal acariform mites including Endeostigmata (but

260 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

not Alycus) Prostigmata (Microcaeculus) and certainoribatids (Palaeacarus) (Evans 1992 Alberti amp Coons1999) Giribet et al (2002) coded all representativeAcari as lacking median eyes It is not known whetherthe eyes of some cyphophthalmid opilions are medianor lateral evidence from tracheal branching inCyphophthalmus (Janczyk 1956) suggests that theyare median eyes (Shultz amp Pinto da Rocha 2007) andpresence of a tapetum in Stylocellus is consistent withlateral eyes (Shear 1993)137 Retinula cells of dorsal median eyes 0 organizedinto closed rhabdoms 1 organized into network ofrhabdomeres 2 disorganized - inapplicable due toabsence of median eyes (136) [GEWB sim3]State 0 is present in Scorpiones Thelyphonida andAmblypygi State 1 is present in Solifugae and Ara-neae state 2 is present in Xiphosura (Paulus 1979)The retinula cells of median eyes have been studied inseveral prostigmatid Acariformes and are organized ina network in some taxa and in an irregular pattern inothers (Alberti amp Coons 1999) Retinulae in phalangidOpiliones have state 0 proximally and state 2 distally(Schliwa 1979)138 Ventral median eyes 0 absent 1 presentState 1 occurs in early instars of extant Xiphosura(Paulus 1979)139 Lateral eyes 0 absent 1 presentLateral eyes are primitively present in Chelicerataand are absent in Palpigradi and Opiliones (Paulus1979) It is unclear whether the eyes of cyphoph-thalmid opilions are median or lateral (see 136)140 Arrangement and number of lateral eyes 0 com-pound many 1 five or more pairs (includes micro-lenses) 2 three primary pairs (excludes microlenses)3 two pairs 4 one pair - inapplicable due to absenceof lateral eyes (139) [WP sim13 + 18 + 38 + 44 + 52 Ssim49 WH sim10 GEWB sim4 + 5]True compound eyes are present in XiphosuraEurypterida Chasmataspidida and many PalaeozoicScorpiones Among the anactinotrichid Acari Opilioa-cariformes have two (Neocarus) or three (Siamacarus)pairs of lateral eyes (although at least one species ofSiamacarus lacks eyes) the allothryrid Holothyridaand many Ixodida have a single pair of lateral lensesand the remainder apparently lack eyes (Evans1992) Thelyphonida was coded as lsquo12rsquo to reflect fivepairs of lenses comprising three pairs of primarylenses and two pairs of small accessory lenses (notsimply three pairs as coded by Giribet et al 2002)Trigonotarbida is also coded as lsquo12rsquo to reflect threepairs of primary lenses and multiple small accessorylenses Dunlop (1999) illustrated Haptopoda as havingpaired triads of lateral eyes but noted that there wasactually no evidence of this in the fossils The lateraleyes of fossil ricinuleids have two pairs of lenses(Selden 1992) One pair of eyes or eyespots are

present in extant Ricinulei and many Schizomidaalthough five genera of hubbardiid Schizomida have apair of lenses (Reddell amp Cokendolpher 1995) State 1is the groundplan for extant Scorpiones and State 3occurs in the pseudoscorpions Chthonius FeaellaNeobisium (but State 4 in Chelifer) and Solifugae(Paulus 1979)141 Lateral eyes with closed rhabdoms 0 absent 1present - inapplicable due to absence of lateral eyes(139) [WP 21 WH 12 GEWB 6]Presence of closed rhabdoms is probably primitive forChelicerata and is retained in extant Xiphosura andScorpiones Retinula cells form a network of rhab-domeres in other extant chelicerates (Paulus 1979)but these networks can differ substantially in detailWeygoldt (1998) and other workers have given sub-stantial weight to the network character in unitingnon-scorpion arachnids but these authors seem not togrant comparable phylogenetic significance to theanalogous character of the median eyes shared bySolifugae and Araneae (137)142 Slit sensilla 0 absent 1 present [WP 19 S 47WH 11 GEWB 209]State 1 occurs in all arachnid orders except Palpigradi(Shultz 1990) The proposal that lsquoprimitiversquo slit sen-silla were present in Eurypterida (Dunlop amp Braddy1997) appears to be based on the over-interpretation ofa comparatively large notch that occurs in Baltoeu-rypterus at the terminus of the tibia (= podomere 7)(see also Edgecombe et al 2000 Giribet et al 2002)The neural construction of slit sensilla like function-ally similar campaniform sensilla of hexapods is sim-ilar to that of trichoid sensilla (Chapman 1998Klompen 2000) Thus these cuticular stress receptorsmay represent modified bases of sensory setae a viewsupported by the replacement of setae by slit sensilladuring post-embryonic development in some Opilioac-ariformes (Klompen 2000)143 Trichobothria 0 absent 1 present [GEWB 213]Trichobothria are present in extant Scorpiones(Jeram 1998) Pseudoscorpiones (Chamberlin 1931)some endeostigmatid (Alycus Van der Hammen 1989)and prostigmatid Acariformes and most Oribatida(Lindquist 1984) They occur on the ventral surface ofthe pedal femora in the opilioacariform Siamacarus(Leclerc 1989) They are also present in Araneae (Foe-lix 1996) Amblypygi (Weygoldt 2000) Schizomidaand Thelyphonida (Hansen amp Soslashrensen 1905) Theyare apparently absent in all non-arachnid chelicer-ates Solifugae Ricinulei Opiliones (Reissland ampGoumlrner 1985) and Parasitiformes (Acari) (Lindquist1984)144 Tibial trichobothria with 2-1-1-1 pattern onappendages IIIndashVI (= arachnid legs 1ndash4) 0 absent 1present - inapplicable due to absence of trichobothria(143) [S 48 WH 87 GEWB 88]

PHYLOGENY OF THE ARACHNID ORDERS 261

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs only in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905) Note that Shultz (1990)erroneously described this character as a 2-2-1-1 pat-tern and that this error was repeated by Giribet et al(2002) and described as 2-1-1 by Wheeler amp Hayashi(1998)145 Paired trichobothria on dorsal surface ofprosoma 0 absent 1 present - inapplicable due toabsence of trichobothria (143)State 1 is an apparent synapomorphy of Acariformesand is present in all acariforms included in this study(Alberti amp Coons 1999)146 Malleoli 0 absent 1 present [WP 45 GEWB 96]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)147 Pectines 0 absent 1 present [WP 20 WH 38GEWB 120]State 1 is a unique synapomorphy of Scorpiones148 Intercheliceral median organ 0 absent 1present [GEWB 212]State 1 is a unique synapomorphy of Palpigradi(Roewer 1934)149 Tarsal organ on appendage of postoral somite III(= arachnid leg 1) (= Hallerrsquos organ) sensilla containedwithin a cuticular depression on the superior surface ofthe tarsus of appendage III (= arachnid leg 1) 0absent 1 present [GEWB sim100]Klompen (2000) has noted that State 1 occurs in Opil-ioacariformes Parasitiformes (except Mesostigmata)and Ricinulei where it also occurs on leg 2 (Talaricoet al 2005) State 1 occurs on all legs in Araneae andperhaps Scorpiones (Foelix 1985)150 Tarsal organ on appendage of postoral somite IV(= arachnid leg 2) 0 absent 1 present (See 149)

REPRODUCTION

151 Gonads 0 primarily prosomal 1 primarilyopisthosomalState 0 is limited to Xiphosura (Sekiguchi 1988)State 1 occurs throughout Arachnida (Millot1949a)152 Ladder-like opisthosomal gonadsaccessoryglands (see 153) 0 absent 1 present [WH sim37 GEWBsim158]Giribet et al (2002) followed Wheeler amp Hayashi(1998) who followed Clarke (1979) in coding gonads asreticulate (Xiphosura) ladder-like (Scorpiones Thely-phonida Schizomida) or lsquosaccularrsquo (all remainingArachnida) However comparisons between Xipho-sura and Arachnida are problematic given that thexiphosuran gonads are primarily prosomal and thoseof arachnids are primarily opisthosomal (151) Fur-ther the reticulate pattern in Limulus (Xiphosura)but not other extant xiphosurans contains a distinctlyladder-like component The lsquosaccularrsquo state is probably

artificial as it encompasses a wide variety of pairedand unpaired structures153 Male gonads in two distinct parts one producingsperm and another (tubular gland) producing a holo-crine secretion similar to degenerate sperm 0 absent1 presentState 1 occurs in Thelyphonida Schizomida andAmblypygi although the holocrine material is pro-duced by ventral organs in Amblypygi and dorsalorgans in Thelyphonida and Schizomida (Alberti2005)154 Number of gonopores 0 two 1 oneExtant xiphosurans have two small genital openingson the base of the genital telopodite and all extantarachnids have a single opening The condition inEurypterida is not known Clarke amp Ruedemann(1912) located a pair of openings near the base of themedian organ (161) that are the outlets of the lsquohornorgansrsquo but it is unclear whether these are genitalducts or accessory structures Braddy amp Dunlop (1997)have developed numerous speculations about thesestructures and extended their arguments far beyondthe available evidence155 Genital opening (ie gonopore or gonostome)appearing to open in prosomal region (ie between legcoxae or anterior to posterior carapacal margin) 0absent 1 present [WP sim50 WH 26 GEWB 166]The genital opening in Euchelicerata is located onpostoral somite VIII (= opisthosomal somite 2) but ithas shifted anterior to the posterior margin of thecarapace or between the coxae of the last pair of legsin most Scorpiones (but not in PalaeoscorpiusKjellesvig-Waering 1986) and Opiliones The geni-tal opening occurs near or anterior to the last coxaein the opilioacariform and parasitiform Acari repre-sented here It is variable in Acariformes but islocated posterior to the coxae in all representativetaxa156 Ovipositor 0 absent 1 present [WP sim51 S sim60WH sim91 GEWB sim172]An ovipositor with a trilobed terminus is an apparentgroundplan character of Oribatida (Lindquist 1984Alberti amp Coons 1999) An ovipositor is also present inOpilioacariformes and Opiliones (Van der Hammen1989)157 Stalked spermatophore attached to substratum0 absent 1 present [S 57]State 1 occurs in Scorpiones PseudoscorpionesAmblypygi Thelyphonida and Schizomida The mech-anism of sperm transfer is unknown in PalpigradiOpilioacariformes and Holothryrida158 Male turns posterior end toward female duringmating behaviour 0 absent 1 presentState 1 occurs in Amblypygi (Weygoldt 2000) The-lyphonida and Schizomida (Weygoldt amp Paulus1979)

262 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

159 Female grasps male opisthosoma during matingbehaviour 0 absent 1 present [WP 29 S 58 WH 19GEWB 188]State 1 occurs in Thelyphonida and Schizomida (Wey-goldt amp Paulus 1979) Evolution of this behaviour wasprobably facilitated by 158160 Penis 0 absent 1 present [WP sim51 GEWB 167]State 1 is a unique synapomorphy of Opiliones A truepenis occurs in Phalangida A clearly homologousstructure is present in Cyphophthalmi (Opiliones) andapparently functions in depositing a spermatophore inthe femalersquos genital chamber (Karaman 2005) A truepenis may occur in some mites but it does not appearto be a groundplan feature of any major group (Evans1992) The lsquopenisrsquo in Oribatida is really a spermatop-ositor it functions in construction of a spermatophore(Alberti amp Coons 1999)161 Median organ 0 absent 1 presentState 1 occurs throughout Eurypterida (Clarke amp Rue-demann 1912) and has been observed in diploaspididChasmataspidida (Dunlop 2002a also Loganamaras-pis Tetlie amp Braddy 2004)

SPERM MORPHOLOGY

162 Nucleus with manchette of microtubules 0absent 1 present [S 54 WH 89 GEWB 193]State 1 occurs in Araneae Amblypygi Thely-phonida Schizomida Ricinulei and gonyleptid Lani-atores (Opiliones) (Alberti 1995 Giribet et al2002)163 Axoneme 0 absent 1 present [WP sim49 S sim55WH sim25 GEWB 195]State 1 occurs in Xiphosura Scorpiones Cyphoph-thalmi (Opiliones) Pseudoscorpiones RicinuleiAraneae Amblypygi Thelyphonida and Schizomida(Alberti 1995)164 Coiled axoneme 0 absent 1 present - inappli-cable due to absence of axoneme (163) [WP 22 S sim55WH 13 GEWB 196]State 1 occurs in Pseudoscorpiones Ricinulei Ara-neae Amblypygi Thelyphonida and Schizomida(Alberti 1995)165 Microtubule arrangement in axoneme 0 9 + 0 19 + 1 2 9 + 2 3 9 + 3 - inapplicable due to absence ofaxoneme (163) [WP sim26 S sim56 WH sim16 GEWB 198](Alberti 1995)166 Helical or corkscrew shaped nucleus 0 absent 1present [GEWB sim204]State 1 occurs in Araneae Amblypygi ThelyphonidaSchizomida and certain Scorpiones (eg Hadrurus)(Alberti 1995)167 Vacuolated-type sperm 0 absent 1 present[GEWB 205]State 1 is unique to Opilioacariformes and Parasiti-formes (Acari) (Alberti 1995)

168 Sperm aggregates 0 absent 1 present [GEWB206] (Alberti 1995)

DEVELOPMENT

169 Yolk in early embryo 0 concentrated (centroleci-thal or telolecithal) 1 evenly distributed (isolecithal)[WH sim34 GEWB sim191]Wheeler amp Hayashi (1998) and Giribet et al (2002)coded this character as lsquo0 isolecithal or telolecithal 1centrolecithalrsquo based on information presented byYoshikura (1975) Their coding appears to highlightseparate conditions found in Scorpiones (ie isoleci-thal and telolecithal) rather than a property intrinsicto the character itself170 Embryonic nutrition other than yolk 0 absent 1present [WP sim49]State 1 occurs in extant Scorpiones (except buthidsand chaerilids) and Pseudoscorpiones (Weygoldt1969)171 Embryological growth zone 0 initiating segmentaddition within prosoma 1 initiating segment addi-tion posterior to prosoma [GEWB sim192]Giribet et al (2002) followed Dunlop amp Webster (1999)in stating that only Xiphosura and Scorpiones havestate 0 among Chelicerata However the last prosomalsomite develops from the growth zone in Ixodida(Anderson 1973 Evans 1992) A similar processapparently occurs in the acariform mites Tyroglyphus(Sarcoptiformes Astigmata) (Yoshikura 1975) andArchegozetes (Sarcoptiformes Oribatida) (Thomas ampTelford 1999) but not apparently in Tetranychus(Prostigmata) (Dearden Donly amp Grbic 2002)172 Eggsembryos maintained in external attachedbrood sac secreted by genital glands 0 absent 1present [WP sim46 S 59 WH 90 GEWB 208 = 219]State 1 occurs throughout Amblypygi SchizomidaThelyphonida and Pseudoscorpions Giribet et al(2002) followed Shear et al (1987) in homologizingspinneret-derived silken egg sacs of spiders with gen-ital-gland-derived brood sacs of Pedipalpi However asthese structures are not homologous in secretory ori-gin construction or composition (Shultz 1987) theyare not regarded here as homologous Giribet et al(2002) miscoded the character as absent in Pseudo-scorpiones and without justification in PalpigradiFemale palpigrades have well-developed glands asso-ciated with postgenital somites and an array of setaefusules and other structures (112) (Boumlrner 1902b Mil-lot 1942 Condeacute 1991a) which are consistent withsome sort of brood care and Palpigradi should there-fore be coded as unknown for this character173 Embryonic and early postembryonic lsquolateralrsquo orClaparegravede organs associated with coxa of postoralsomite IV (= arachnid leg 2) 0 absent 1 present [WHsim31 GEWB sim211]

PHYLOGENY OF THE ARACHNID ORDERS 263

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bilaterally paired rounded protuberances locatedbetween coxae of appendages of postoral somites IIIand IV (arachnid legs 1 and 2) in prelarval and larvalinstars appear to be a primitive feature of Acariformes(Lindquist 1984 Evans 1992 Alberti amp Coons 1999)It is known to be derived embryologically from thecoxa of leg 2 in the oribatid Archegozetes (Thomas ampTelford 1999) Structures apparently homologouswith the acariform Claparegravede organ are present inembryonic Amblypygi (Weygoldt 2000) and Thely-phonida (Yoshikura 1975) and in embryonic and earlypost-embryonic stages of Solifugae (Roewer 1934)where they are termed lateral organs The so-calledlsquolateral organrsquo of embryonic Xiphosura (174) occurs onthe lateral surface of the carapace and does not appearto be homologous with the coxa-associated structuressharing their name In fact the xiphosuran lateralorgan develops even when the coxa of postoral somiteIV has been excised (Sekiguchi 1988) In contrastWheeler amp Hayashi (1998) and Giribet et al (2002)followed Yoshikura (1975) in regarding all lsquolateralorgansrsquo as homologous and in miscoding this characteras absent in Acariformes Van der Hammen (1989)suggested that the sternal verrucae of Opilioacari-formes are homologous with the Claparegravede organ butit differs in gross structure and ontogenetic timing174 Embryonic lateral organ associated with cara-pace 0 absent 1 present [WH sim31 GEWB sim211]State 1 is known only from embryonic stages of extantXiphosura (Yoshikura 1975) In contrast to previousinterpretations this feature does not appear to behomologous with the lsquolateralrsquo or Clarapegravede organ ofcertain arachnids (173)175 Live birth 0 absent 1 presentState 1 occurs in extant Scorpiones and has evolvedseveral times in a few mites (Evans 1992)176 Hexapodal larva and 1ndash3 nymphal stages 0absent 1 present [WP 53 S sim61 WH 27 GEWB 186]State 1 occurs in Acari and Ricinulei (Lindquist1984)177 Hexapodal prelarva 0 absent 1 present [S sim61WH 47]State 1 occurs in Opilioacariformes (NeocarusKlompen 2000) and many Acariformes (Evans 1992)

EXCRETORYOSMOREGULATORY SYSTEM

178 Malpighian tubules 0 absent 1 present [WP 17S 62 WH 9 GEWB 153]State 1 occurs in all major arachnid groups except Pal-pigradi Pseudoscorpiones Opiliones Oribatida (Sar-coptiformes) and Prostigmata Within Acari State 1occurs in all major anactinotrichid lineages (Opilioac-ariformes Holothyrida Mesostigmata Ixodida) andsimilar structures are present in certain Astigmata(Evans 1992 Alberti amp Coons 1999)

179 Dorsomedian excretory organ 0 absent 1presentThis is a specialized excretoryosmoregulatory organformed by the postventriculus and proctodeum inProstigmata (Alberti amp Coons 1999)180 Adult coxal organ opening on or near coxa ofappendage III (= arachnid leg 1) 0 absent 1 present[S 64 WH 93 GEWB 90]State 1 occurs throughout Acari (Evans 1992 Albertiamp Coons 1999) Ricinulei (Pittard amp Mitchell 1972Legg 1976) Palpigradi (Millot 1942) AraneaeAmblypygi Thelyphonida and Schizomida (Buxton1913 1917) There is evidence that the second lsquoozo-porersquo of gonyleptids (Opiliones Laniatores) (Hara ampGnaspini 2003) is a persistent opening to the coxalorgan associated with leg 1 (Soslashrensen 1879) whichalso occurs in embryonic Phalangium (Opiliones Eup-noi) (Moritz 1959)181 Adult coxal organ opening on or near coxa ofappendage V (= arachnid leg 3) 0 absent 1 present [S63 WH 92 GEWB 89]State 1 occurs in extant Xiphosura (Yamasaki et al1988) Scorpiones mygalomorph (Buxton 1913 1917)and mesothele Araneae (Liphistius J Millot in Bris-towe 1932) State 1 occurs in basal Amblypygi(Charon Charinus Buxton 1913 1917) but degener-ates prior to the adult stage in higher groups(Phrynus Weygoldt 2000) Adequately preservedeurypterids show a small submarginal opening on thecoxa of appendage V (Baltoeurypterus Selden 1981also Eurypterus Hughmilleria Clarke amp Ruedemann1912) and this is interpreted here as the opening to acoxal organ Giribet et al (2002) coded this characteras uncertain for Eurypterida182 Adult coxal organ opening on coxa of appendageII (= arachnid palp) 0 absent 1 present [GEWB 65]Buxton (1913 1917) noted that the coxal organ ofSolifugae opens on the prolateral surface of the palpalcoxa The proximal end of the duct leading to the ori-fice is associated with glands and apparently theglandular secretion andor coxal fluid serve as saliva(Alberti 1979) (see also 14 15)183 Genital papillae 0 absent 1 present - inappli-cable coded only for those taxa with eversible lsquoappend-agesrsquo (111)The genital papillae are associated with the genitalopening in postlarval Acariformes (Evans 1992Alberti amp Coons 1999) These structures are probablyserially homologous with the prosomal Claparegravedeorgan (173) and both structures probably function inwater and ion regulation

DIGESTIVE SYSTEM

184 Ingestion 0 solid food 1 primarily liquid foodwith or without preoral digestion [WP sim16 WH sim8GEWB sim218]

264 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Extant xiphosurans ingest solid food and members ofmost extant arachnid orders ingest fluids Opilionesand Acari are the principal exceptions The phalangidOpiliones ingest solids but Shultz (2000 unpublobserv) has shown that the precerebral pharyngealapparatus of Cyphophthalmi is very similar to that ofthe fluid-feeding Scorpiones and differs substantiallyfrom phalangid Opiliones Examination of gut con-tents of the cyphophthalmids Siro acaroides S exilisand Chileogovea oedipus have failed to reveal solidparticles typically found in the guts of phalangids(Shultz unpubl observ) the character is coded asunknown for Cyphophthalmi Within Acari consump-tion of particles has been documented in Opilioacari-formes (Neocarus Van der Hammen 1989unidentified Walter amp Proctor 1998) Holothyridanon-parasitic Sarcoptiformes (Evans 1992 Walter ampProctor 1998 Alberti amp Coons 1999) and many endeo-stigmatids including the nematophagous Alycusroseus (Walter 1988)185 Mouth 0 directed posteroventrally 1 directedanteroventrally [S 9 WH 55 GEWB 28]State 1 occurs throughout Arachnida (Shultz 1990)186 Oral cavity dilated by muscles arising from coxaeand constricted by large circular sphincter 0 absent1 presentState 1 occurs in extant Xiphosura (Manton 1964Scholl 1977 Shultz 2001)187 Palate plate 0 absent 1 present [GEWB 159]State 1 is a unique synapomorphy of Araneae (Dunlop1994)188 Lateral walls of epistome broadly fused to medialwalls of palpal coxae opposite sides connected by well-developed transverse epistomal muscle 0 absent 1presentState 1 occurs throughout Opiliones and Scorpiones(Shultz 2000 unpubl observ)189 Epistome with a pair of lateral arms that projectsposteriorly into the prosoma on either side of the phar-ynx 0 absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)190 Epistome with four pairs of suspensor musclesattaching to the carapace 0 absent 1 presentState 1 occurs in extant Scorpiones (Lankester et al1885 Vyas 1970 Shultz unpubl observ) and appar-ently in Solifugae (Roewer 1934)191 Intercheliceral epipharyngeal sclerite 0 absent1 present [GEWB 31]State 1 occurs in Palpigradi (Boumlrner 1904) Trigo-notarbida (Palaeocharinus Dunlop 1994) Araneae(Firstman 1954 Marples 1983) Amblypygi (Mil-lot 1949b Shultz 1999) Schizomida (Millot1949e) Thelyphonida (Millot 1949e Shultz 1993)and Phalangida (Opiliones) (Shultz 2000 unpublobserv)

192 Epipharyngeal sclerite large projecting posteri-orly 0 absent 1 present - inapplicable due to absenceof sclerite (191) [GEWB 32]State 1 occurs in Amblypygi (Millot 1949b Shultz1999) Schizomida (Millot 1949e) Thelyphonida (Mil-lot 1949e Shultz 1993) some Araneae (Heptathela ndashcoded as Liphistius ndash and Hypochilus Marples 1983not Aphonopelma Firstman 1954) and perhaps Trig-onotarbida (Palaeocharinus Dunlop 1994)193 Dorsal dilator muscle of precerebral pharynxattaching to intercheliceral septum or associatedepipharyngeal sclerite (191) 0 absent 1 presentState 1 occurs in Palpigradi (Millot 1943) Araneae(Marples 1983) Amblypygi Schizomida and Thely-phonida (Millot 1949e Shultz 1993 1999)Extremely fine muscles have been documented insome Scorpiones (Shultz 2007) and in Leiobunum(Opiliones) (Shultz 2000) Lankester et al (1885)reported a muscle in Limulus that would appear tocorrespond to this muscle but the muscle does notexist (Manton 1964 Shultz 2001)194 Dorsal dilator muscle of precerebral pharynxattaching to dorsal surface of prosoma 0 absent 1present [GEWB 228]State 1 occurs in Palpigradi (Roewer 1934) and mostAraneae (Palmgren 1978 Aphonopelma Firstman1973) but not in Heptathela (coded as Liphistius) orHypochilus (Marples 1983)195 Lateral dilator muscle of precerebral pharynxattaching to endosternite 0 absent 1 present [GEWBsim226]State 1 is a unique feature of Araneae (Marples 1983)196 Lateral dilator muscle of precerebral pharynxattaching to lateral surface of epistomal processes 0absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)197 Lateral dilator muscle of precerebral pharynxattaching to medial process of coxa of appendage II(= arachnid palp) 0 absent 1 present [GEWB sim226]State 1 occurs in Amblypygi Thelyphonida and Schi-zomida (Boumlrner 1904 Shultz 1993 1999)198 Dilator muscle of precerebral pharynx andorpreoral cavity attaching to ventral surface of prosoma0 absent 1 present [GEWB 227]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) Araneae (Marples 1983) Palpigradi(Millot 1943) and Solifugae (Roewer 1934)199 Postcerebral pharynx 0 absent 1 present [WPsim31 S sim5 WH sim21 GEWB sim154]The cuticle-lined foregut passes through the centralnervous system in many arthropods That portion pos-terior to the CNS is here termed the postcerebralpharynx regardless of the details of its morphology orfunctional specialization A substantial postcerebralpharynx is known in extant Xiphosura (Lankester

PHYLOGENY OF THE ARACHNID ORDERS 265

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

et al 1885 Manton 1964 Yamasaki et al 1988)Scorpiones (Centruroides Hadrurus HeterometrusShultz 2007 also Androctonus Abd el-Wahab 1952)Solifugae (Millot amp Vachon 1949) Araneae Ambly-pygi Thelyphonida and Schizomida (Millot 1949andasheShultz 1993) Weygoldt amp Paulus (1979) applied theterm lsquopostcerebral pharynxrsquo only to the lsquosucking stom-achrsquo of Araneae and Amblypygi but the sucking stom-ach is here regarded as a complex of three characters199ndash201

200 Dilator muscle of postcerebral pharynx attach-ing to endosternite 0 absent 1 present - inapplicabledue to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Lankester et al1885 Manton 1964 Shultz 2001) Scorpiones (Cen-truroides Hadrurus Heterometrus Shultz 2007 alsoAndroctonus Abd el-Wahab 1952) Araneae (First-

man 1954 Palmgren 1978) Amblypygi (Millot1949b Shultz 1999) and apparently the palpigradeProkoenenia wheeleri (Rucker 1901) but not inEukoenenia miriabilis (Millot 1943 contra Boumlrner1904)201 Dilator muscle of postcerebral pharynx attachingto dorsal surface of prosoma 0 absent 1 present -inapplicable due to absence of postcerebral pharynx(199)State 1 occurs throughout Araneae (Millot 1949cFirstman 1954 Palmgren 1978) and Amblypygi(Charinus Millot 1949b Phrynus Shultz 1999)202 Crop and gizzard 0 absent 1 present - inap-plicable due to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Sekiguchi 1988)but not in extant Arachnida

PHYLOGENY OF THE ARACHNID ORDERS 227

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

the Ordovician to the Permian (Tollerton 1989) Phy-logenetic relationships among the eurypterids havenot been rigorously addressed (but see Beall amp Laban-deira 1990) However members of Stylonuroideaappear to be plesiomorphic in retaining relativelyunspecialized legs and were represented by Stylonu-rus (eg Clarke amp Ruedemann 1912 Waterston1979) In the remaining eurypterids the last leg ispaddle-like (Tollerton 1989) The non-stylonuroideurypterids were represented in the matrix by thewell-preserved and well-studied Baltoeurypterus tet-ragonophthalmus especially as described by Selden(1981)

Chasmataspidida Members of this fossil aquaticgroup (OrdovicianndashDevonian six genera) have a four-segmented mesosoma and a nine-segmented meta-soma (Dunlop Poschmann amp Anderson 2001 DunlopAnderson amp Braddy 2004 Dunlop 2002a) Three gen-era were included Chasmataspis from the monotypicfamily Chasmataspididae as well as Diploaspis andOctoberaspis from Diploaspididae

Haptopoda This is a fossil terrestrial group (Carbon-iferous) containing one known species Plesiosiromadeleyi The known specimens have been re-examined by Dunlop (1999)

Palpigradi The extant palpigrades (two families sixgenera ~50 spp) (Harvey 2002) are a poorly studiedgroup of small-bodied terrestrial (some semi-aquatic)arachnids The two most well-studied generaEukoenenia (Eukoeneniidae) especially E mirabilisand Prokoenenia (Prokoeneniidae) especiallyP wheeleri were included in the matrix The oneknown fossil species Paleokoenenia mordax is rela-tively recent (Pliocene) and to the extent known isfairly similar to extant palpigrades (Rowland amp Sis-som 1980) It was not included in the analysis

Trigonotarbida This is a fossil terrestrial order(Upper Silurian to Lower Permian ~ten families ~50spp) The genera coded here were drawn from twoLower Devonian Lagerstatten Gilboarachne from theGilboa Formation (New York) (Shear et al 1987) and

Proscorpius00000000 00000 -0 -000- 100--0000 010 00000 0 10-0001 06000-0 000-00 --0311- 110 01 1 0

Chthonius0000100000 0310-00120 0000-10100 -1000-0000 100--00100 0000000001 0000010010 01--012100 0E00-00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Neobisium0000100000 0310-00121 0000-10100 -1000-0010 100--00100 0000000001 0000010010 01--012100 0100-00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Feaella0000100000 0310-00120 0000-10100 -1000-0000 100--00100 0000000001 0000010010 01--012100 000--00001 01115000-0 000000-000 00--000--1 0000021000 01700-013 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Chelifer0000100000 0310-00121 0000-10100 -1000-0010 100--00100 0000000001 0000010010 01--012100 000--00001 01115000-0 000000-000 00--000--1 0000021000 01700-014 0110000000 1001001000 0011200011 1100000000 10-1100000 0-0000000- --

Eremocosta0000110000 0310-00121 0000-00010 -1000-0000 001010100 0000000001 00011-0000 02--010-10 1101010001 01114000-0 000000-000 00--000--1 00000200 13111013 010--10000 1001000000 000--00100 1010000100 01-1100001 0-00000110 00

Galeodes0000110000 0310-00121 0000-00010 -1000-0000 001010100 0000000001 00011-0000 02--010-10 1101010001 01114000-0 000000-000 00--000--1 00000200 13111013 010--10000 1001000000 000--00100 1010000100 011100001 0-00000110 00

Table 1 Continued

228 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Palaeocharinus from the Rhynie Chert (Scotland)(Dunlop 1994 Fayers Dunlop amp Trewin 2004)

Araneae (spiders) This is a very large order (110 fam-ilies ~3600 genera ~39 000 extant spp) (Platnick2005) represented here by one genus from each of thethree principal extant lineages (Platnick amp Gertsch1976 Coddington amp Levi 1991 Coddington 2005)Liphistius (Mesothelae) Aphonopelma (Mygalomor-phae) and Hypochilus (Araneomorphae) The sampleis small with respect to known diversity but the basalphylogeny and relevant groundplan states of the orderare well established and states derived from the rep-resented taxa are consistent with them

Amblypygi (whipspiders) This is a small extant ter-restrial order (five families 17 genera 136 spp)(Harvey 2002) and was represented in the matrix bytwo fairly well-studied genera Charinus and Phry-nus Charinus retains several features that appear tobe plesiomorphic for the order (eg adult pedal pul-villi coxal glands associated with leg 3 eversiblevesicles) Weygoldt (1996) has examined the phylog-eny of the order and has provided an important sum-mary of morphology and general biology (Weygoldt2000)

Thelyphonida or Uropygi ss (whipscorpions vinega-roons) Extant whipscorpions (16 genera 106 spp)(Harvey 2002) were represented here by one well-studied species Mastigoproctus giganteus althoughadditional information was drawn from Typopeltis andThelyphonus The morphology of the group is highlyconserved there have been no modern studies ofintraordinal phylogeny A controversial fossil speciesfrom the late Carboniferous Proschizomus (Dunlop ampHorrocks 19951996) was also included but less well-preserved Carboniferous forms were not

Schizomida This small order (two extant families~34 genera ~200 spp) (Harvey 2002) is widelyregarded as the sister group of Thelyphonida Thetaxon sample included one representative from eachextant family Protoschizomus (Protoschizomidae) andStenochrus (Hubbardiidae) especially S portoricen-sis There are three recent (Pliocene) fossil speciesall from the same locality these were not includedhere

Ricinulei The extant ricinuleids (three genera 55spp) (Harvey 2002) were represented by two generaCryptocellus and Ricinoides Detailed studies of grosscuticular anatomy and post-embryonic developmentare available for a representative of each genus (Cryp-tocellus Pittard amp Mitchell 1972 Ricinoides Legg1976) and some information on internal anatomy isavailable for Ricinoides (Millot 1945) and this hasbeen extrapolated to Cryptocellus in the matrix There

are two basic fossil types (Selden 1992) one resem-bling modern taxa and another with a unique opistho-soma that superficially resembles the closed elytra of abeetle (ie curculioids) Two fossils from the formergroup were included Terpsicroton and Poliochera asthey are reasonably well preserved and show impor-tant characters (eg two pairs of eyes or evidence ofopisthosomal diplosegmentation) not expressed inextant forms No curculioid ricinuleids were includedas they appear to offer no additional information rel-evant to resolving ordinal relationships

Opilioacariformes (= Opilioacarida Notostigmata)The opilioacariform mites (nine genera 20 spp) (Har-vey 2002) are generally regarded as plesiomorphicAcari and are fairly conserved in their morphologyThe group was represented by two species Neocarustexanus and Siamacarus withi N texanus has a typi-cal opilioacariform morphology and its external anat-omy has been particularly well studied (eg Van derHammen 1989 Klompen 2000) S withi differs frommost other opilioacariforms in having trichobothriaand three rather than two pairs of lateral eyes (the lat-ter also in Paracarus) features that are potentiallysignificant for assessing ordinal relationships Someinformation on internal morphology of Opilioacaruswas taken from With (1904) and extrapolated to Neo-carus and Siamacarus

Parasitiformes This large group of mites comprisesthree clades Holothyrida Ixodida (ticks) and Mesos-tigmata (= Gamasida) Holothyrida (three familiesfive genera lt 30 spp) (Walter amp Proctor 1999) wasrepresented here by two species Australothyrus ocel-latus and Allothyrus constrictus Ixodids (three fami-lies 22 genera ~860 spp) (Kierans amp Robbins 1999)were represented by Amblyomma and Argas andmesostigmatids (~70 families ~10 000 spp) were rep-resented by Glyptholaspis confusa

Acariformes (= Actinotrichida) This is a very largediverse group of mites that includes the Endeostig-mata Sarcoptiformes and Prostigmata Endeostig-mata is an apparently paraphyletic assemblage of~ten families A morphologically generalized speciesAlycus roseus (Bimichaeliidae) was included herewith character states extracted largely from Van derHammen (1989) Sarcoptiformes (gt 200 families) iseffectively synonymous with Oribatida with Astig-mata being a large derived lsquooribatidrsquo clade (Norton1998 Maraun et al 2004) The group was representedby an early divergent plesiomorphic genus Palaeac-arus (Palaeosomata) and a more derived and well-studied genus Archegozetes (especially A longisetusand A magnus) The Prostigmata (~120 families~7000 spp) (Walter amp Proctor 1999) was representedby two genera Allothrombium and Microcaeculus

PHYLOGENY OF THE ARACHNID ORDERS 229

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Pseudoscorpiones The pseudoscorpions (24 families425 genera ~3200 spp) (Harvey 2002) were repre-sented in the matrix by four terminal genera Chtho-nius (Chthoniidae) was drawn from the basallydivergent superfamily Chthonioidea (Harvey 1992)Feaella (Feaellidae) represented the enigmatic super-family Feaelloidea which has an unusual combinationof characters and may be an early divergent lineageclose to Chthonioidea (Harvey 1992) Neobisium (Neo-bisiidae) is morphologically intermediate between theearly divergent families and the higher families rep-resented by Chelifer (Cheliferidae)

Solifugae (= Solpugida) (sun spiders) The order issmall (~1000 spp) but the absence of modern phylo-genetic treatments precludes meaningful estimates ofgenera and families (Harvey 2002) The taxon sampleincluded two genera one from the New World Eremo-costa and one from the Old World Galeodes

Opiliones (harvestmen) Extant harvestmen (~25 fam-ilies ~500 genera ~6000 spp) are divided into twosuborders Cyphophthalmi and Phalangida Cyphoph-thalmids were represented here by Chileogovea oedi-pus (Petallidae) which is currently under study by theauthor and Cyphophthamus duricorius (Sironidae)which is probably the most well-studied member of thesuborder (eg Janczyk 1956) Phalangida includesthree major groups Laniatores Eupnoi and Dyspnoibut there is disagreement as to whether Dyspnoi is thesister group to Laniatores or to Eupnoi (Shultz ampRegier 2001 Giribet et al 2002) Laniatores was rep-resented by Sclerobunus and Gonyleptes and Eupnoiby Leiobunum and Caddo

Scorpiones (scorpions) The extant scorpions (16 fam-ilies 155 genera 1279 spp) (Fet et al 2000) wererepresented by three taxa Centruroides vittatus(Buthidae) Hadrurus arizonensis (Iuridae) and Heter-ometrus spinifer (Scorpionidae) Buthidae is widelyregarded as the sister group to other extant lineagesand similarities among the represented terminals arelikely to be ground plan features of extant scorpionsgenerally The morphology of fossil scorpions is sub-stantially more diverse (Kjellesvig-Waering 1986)but their phylogenetic relationships are unclear (butsee Jeram 1998) The fossil taxa used in this study(Proscorpius Stoermeroscorpio Palaeoscorpius Pre-arcturus) were chosen for quality of preservation andor presence of a phylogenetically significant constella-tion of characters

CHARACTER CODING

Most character states were determined from directobservation the primary literature and authoritativereviews In some cases states were assigned to termi-

nals based on observations from related species asnoted in the Appendix In the matrix state lsquo-rsquo indicatesthat the character is inapplicable because the taxonlacks a more general character For example if a taxonlacks eyes then special features of the eyes (numberposition retinal configuration etc) are inapplicable tothat taxon A lsquorsquo indicates that the state is unknown oruncertain An entry with multiple states (eg 01 345) should be treated as an ambiguity code not a poly-morphism it indicates that two or more interpreta-tions of homology are applicable and that assignmentwas established analytically by character concor-dance Although lsquo-rsquo lsquorsquo and lsquo01rsquo are analyticallyidentical when applied to binary characters theyprovide information about the empirical status of thecharacter state in specific taxa Characters are citedthroughout the text as italicized numbers in parenthe-ses and are discussed in the Appendix

PHYLOGENETIC ANALYSIS

Phylogenetic analysis of the full data matrix wasperformed using the program Tree Analysis usingNew Technologies (TNT) ver 1 (Goloboff Farris ampNixon 2000) using lsquotraditionalrsquo search based on1000 replicates using TBR branch swapping Resultswere compared to those obtaining using the ratchetalgorithm (Nixon 1999) to determine any differencedue to analytical approach Nodal support for theminimal-length topology was evaluated by boot-strap (Felsenstein 1985) and Bremer support(Bremer 1994) Bootstrap analysis was conducted inTNT and based on 1000 pseudoreplicates each anal-ysed by ten random-addition replicates using TBRbranch swapping A nexus file containing the result-ing 1000 trees was imported into PAUP ver 10(Swofford 2002) to obtain bootstrap frequenciesBremer support was determined in TNT by con-straining specific nodes in the minimal-length topol-ogy and then determining the shortest tree that didnot recover the specified clade The difference inlength between the unconstrained and constrainedminimal-length trees is the Bremer support Theeffect of homoplasy on results was explored by con-ducting implied weights analysis (Goloboff 1993) inTNT Six analyses were conducted each with con-stant of concavity (k) set to a different integer valueof 1ndash6 where 1 is weighted most severely againsthomoplasious characters Each implied weightsanalysis was conducted using lsquotraditionalrsquo searchbased on 1000 replicates using TBR branch swap-ping The same procedures were used in analysing amatrix that included only extant taxa Howevercharacters rendered uninformative by removal offossil taxa (ie 49 98 99 100 113 114 125 161)were excluded prior to analysis

230 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

COMPARISONS OF ALTERNATIVE HYPOTHESES

Phylogenetic hypotheses proposed in the recent liter-ature were also evaluated including those thatattempted to resolve arachnid phylogeny completely(Fig 2) their hypothesized subclades (Fig 3) andhypotheses that proposed only specific nodes (Figs 34) The fully resolved topologies of Weygoldt amp Paulus(1979) Van der Hammen (1989) Shultz (1990)Wheeler amp Hayashi (1998) and Giribet et al (2002)were compared with the optimal topology using theTempleton test (Templeton 1983) as implementedin PAUP Internal relationships of multisampledorders were constrained to match those of the optimaltopology unless the original authors explicitlyfavoured an alternative Node-specific hypotheseswere evaluated by determining the frequency withwhich the node was recovered in bootstrap analysisThey were also evaluated by constraining parsimonyanalysis in TNT to recover the shortest tree contain-ing each specific node and then assessing the effect onrelative tree length and overall resolution takingnote of effects on otherwise stable or well-supportedclades The entire matrix was used to assess the twocases where node-specific hypotheses involved fossiltaxa (Fig 4)

RESULTS

EXTANT TAXA

Unweighted analysis of extant taxa produced two min-imal-length trees (length 383 CI 057) (Fig 5A) withconflicts limited to relationships within Pseudoscorpi-ones Multiply sampled orders were recovered asmonophyletic with high nodal support as measured bybootstrap percentage (BP gt 80) except Acari (BP 30)There were few well-supported interordinal groupsexcept Uropygi (BP 100) Pedipalpi (BP 100) Tetra-pulmonata (BP 93) and ArachnidaXiphosura (BP100) All other interordinal relationships were recov-ered with BP less than 60

Results from implied weights analysis (IWA) indi-cated a significant effect of homoplasy on resolution ofthe deepest interordinal relationships although theinterordinal clades Stomotheca Haplocnemata Acar-omorpha Tetrapulmonata Pedipalpi and Uropygiwere recovered under all values of k IWA with k = 1 (1tree best score = 5124) recovered Megoperculata(= Palpigradi + Tetrapulmonata) and Micrura(= Megoperculata + Acaromorpha) with Micrura beingthe sister group to Haplocnemata (= Pseudoscorpiones+ Solifugae) and with Stomothecata (= Opiliones +Scorpiones) being the sister group to all other arach-nids (Fig 5B) With k = 2 (1 tree best score = 3807)and k = 3 (1 tree best score = 3054) Megoperculatawas recovered and Stomotheca remained the sister

group to other arachnids but the relative positions ofHaplocnemata and Acaromorpha changed (Fig 5C)IWA using k = 4 (1 tree best score = 2559) k = 5 (1tree best score = 2205) and k = 6 (1 tree bestscore = 1938) recovered topologies consistent with theminimal-length topology derived from unweightedcharacters (Fig 5A)

EXTANT AND FOSSIL TAXA

Unweighted analysis of extant and fossil taxa pro-duced 18 minimal-length trees (length 426 CI 056)(Fig 6A) with conflicts limited to relationships withinPseudoscorpiones Scorpiones and Araneae Multiplysampled orders were again recovered as monophyleticwith high nodal support (BP gt 80) except Acari whichwas reconstructed as diphyletic with Anactinotrichidabeing the sister group to Ricinulei (BP 33) Therewere no well-supported interordinal groupings (ieBP gt 80) except Uropygi (BP 89) Pedipalpi (BP 92)and Arachnida (BP 90)

Results from IWA indicated a significant effect ofhomoplasy on resolution of the deepest interordinalrelationships in Arachnida although StomothecataHaplocnemata Acaromorpha Cryptognomae Pantet-rapulmonata Tetrapulmonata Schizotarsata Pedi-palpi and Uropygi were recovered under all values ofk IWA with k = 1 (7 trees best score = 5675) recov-ered Megoperculata and Micrura with Micrura beingthe sister group to Haplocnemata Stomothecata wasnot recovered in the strict consensus (Fig 6B) Withk = 2 (3 trees best score = 4243) IWA recovered thetopology shown in Fig 6C with Palpigradi being thesister group to all other arachnids The strict consen-sus topologies reconstructed using k = 3ndash6 were iden-tical to those obtained with k = 2 (k = 3 3 trees bestscore = 3419 k = 4 3 trees best score = 2874 k = 53 trees best score = 2483 k = 6 3 trees bestscore = 2188)

COMPARISONS OF ALTERNATIVE HYPOTHESES

The hypothesis of Weygoldt amp Paulus (1979) (Fig 2)was 398 steps (15 steps longer than the unconstrainedminimal-length tree) Van der Hammen (1989) was400 steps (17 steps longer) Shultz (1990) was 399steps (six steps longer) Wheeler amp Hayashi (1998)was 391 steps (eight steps longer) and Giribet et al(2002) was 394 steps (11 steps longer) Templetontests conducted at the 005 significance level rejectedthe hypothesis of no difference between the optimaltopology and those of Weygoldt amp Paulus (P = 00002)Van der Hammen (P = 00011) Wheeler amp Hayashi(P = 00018) and Giribet et al (P = 00343) The Shultztopology was not significantly different (P = 01444)from the optimal topology

PHYLOGENY OF THE ARACHNID ORDERS 231

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 2 Fully resolved phylogenetic hypotheses of extant euchelicerate groups proposed in the recent literature Note thesimilarity in the topologies of the parsimony-based analyses by Shultz (1990) Wheeler amp Hayashi (1998) and Giribet et al(2002) The Giribet et al topology is based on neontological data (morphology and molecules) and the original lsquoROOTrsquo maybe an artefact from use of the highly divergent pycnogonids as an outgroup

Xip

hosu

raS

corp

ione

sO

pilio

nes

Ara

neae

Am

blyp

ygi

Uro

pygi

Ric

inul

eiA

nact

inot

richi

daS

olifu

gae

Pse

udos

corp

ione

sP

alpi

gra

diA

carif

orm

es

Xip

hosu

raS

corp

ione

sA

rane

aeA

mbl

ypyg

iU

ropy

giP

alpi

gra

diS

olifu

gae

Pse

udos

corp

ione

sO

pilio

nes

Ric

inul

eiA

cari

Lipoctena

Arachnida

Apulmonata

Holotracheata

Cryptoperculata

Acaromorpha(ldquoAcarinomorphardquo)

Tetrapulmonata(rdquoMegoperculatardquo)

LabellataHaplo-cnemata

Weygoldt amp Paulus (1979)

Labellata

Tetrapulmonata(rdquoArachnidardquo)

Haplocnemata(rdquoApatellatardquo)

Van der Hammen (1989)

Xip

hosu

raP

alpi

gra

diA

rane

aeA

mbl

ypyg

iU

ropy

giR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Pedipalpi Haplocnemata

Novogenuata

Dromopoda

Shultz (1990)

Acaro-morpha

Micrura

Xip

hosu

raP

alpi

gra

diU

ropy

giA

mbl

ypyg

iA

rane

aeR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Labellata Haplocnemata

Novogenuata

Dromopoda

Wheeler amp Hayashi (1998)

Acaro-morpha

Micrura

Xip

hosu

raA

cari

Pal

pig

radi

Ric

inul

eiA

rane

aeA

mbl

ypyg

iU

ropy

giO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Tetrapulmonata

PedipalpiHaplocnemata

Novogenuata

Dromopoda

Giribet et al (2002)

ROOT

Micrura

Crypto-gnomae

Sternocoxata

Epimerata

RostrosomataMyliosomata

Neosternata

232 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 3 Consensus trees produced by parsimony analysis of neontological data constrained to produce relationships pro-posed in the recent literature The constrained (target) node is indicated by a black dot and the taxa encompassed by theconstraint are enclosed in a box Numbers below each tree represent the number of minimal-length constrained treeslength of minimal-length trees difference in the length of the unconstrained minimal-length tree and the constrained min-imal-length tree and percentage unconstrained bootstrap trees in which the target node was recovered respectively Thesetrees indicate the effect on branch length imposed by specific hypotheses and impact of constraining target nodes on non-target nodes

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Lipoctena4 385 2 lt5

Xi

Sc

Op

Ps

So

Pa

Ur

Am

Ar

Ri

An

Ac

Labellata6 397 14 lt5

Xi

Sc

Ar

Am

Ur

Op

Pa

Ps

So

Ri

An

Ac

Apulmonata2 390 7 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

So

Ri

An

Ac

Holotracheata5 390 7 lt5

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptoperculata2 388 5 lt5

Xi

Sc

Op

Pa

So

Ps

Ar

Am

Ur

Ri

An

Ac

Rostrosomata6 384 1 25

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Neosternata3 388 5 lt5

Xi

Sc

Op

Ar

Am

Ur

Ri

An

Ps

So

Pa

Ac

Coxisternata2 388 5 lt5

Epimerata7 387 4 11

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptognomae6 384 1 24

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Micrura3 385 2 7

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Megoperculata3 384 1 28

Xi

Sc

Ps

So

Op

Pa

Ar

Am

Ur

Ri

An

Ac

Novogenuata10 388 5 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

Ri

So

An

Ac

Solifugae + Acari4 388 5 lt5

Shultz (1990) Alberti amp Peretti (2002)

Van der Hammen (1989)

Weygoldt amp Paulus (1979)

Xi

Ar

Am

Ur

So

Ps

Sc

Op

Ri

Pa

An

Ac

PHYLOGENY OF THE ARACHNID ORDERS 233

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Results from analyses of node-specific hypothesesare summarized in Figures 3 4 In most cases pro-posed relationships were recovered in fewer than 5of bootstrap pseudoreplicates The notable exceptionswere Rostrosomata (25) Cryptognomae (24) andMegoperculata (28) and the strict-consensus con-strained minimal-length tree for each was only onestep longer than the unconstrained optimal tree Inmost cases constrained topologies did not add manysteps most were less than five steps longer and nonewas greater than 14 steps longer than the un-constrained minimal-length tree It is noteworthyhowever that even constrained nodes that imposerelatively few extra steps sometimes supportedimprobable relationships or eliminated clades thatwere well supported or stable in the unconstrainedanalysis Specific examples and their implications arediscussed below

DISCUSSION

RECENT ISSUES IN ARACHNID PHYLOGENY

Available morphological evidence consistently re-solves some interordinal relationships and fails to

resolve others (Fig 1) Continued progress depends onthe ability of arachnologists to discover new charac-ters and to assess the evidence critically Whilereviewing the recent arachnological literature severalaspects of phylogenetic practice emerged that seemedcounterproductive to both the perception and theactual rate of progress toward resolving arachnid phy-logeny Some of these are summarized here withspecific examples given in the remainder of theDiscussion and in the Appendix

There is a tendency to portray arachnid ordinal phy-logeny as more poorly resolved and contentious than isactually the case (Coddington et al 2004) Phyloge-netic hypotheses generated during different historicalperiods and using differing standards of evidence areoften cited as examples of current disagreement (egSelden 1993 Dunlop 1996 Selden amp Dunlop 1998Wheeler amp Hayashi 1998) In fact recent parsimony-based analyses of morphology have tended to convergeon topologies with internal structures congruent withthose found here (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) (Fig 2) Arachnidphylogeny is not fully resolved especially at its deep-est levels (Fig 1) but this does not mean that allaspects of arachnid phylogeny are controversial orpoorly supported by the available evidence

Matrices are sometimes constructed by uncriticallsquorecyclingrsquo of erroneous or problematic charactersbased on diverse secondary sources (Jenner 2001)Conclusions derived from mixtures of valid invalidspeculative and redundant characters are sometimesportrayed as the phylogenetic signal provided by mor-phology Data recycling can perpetuate errors (see 1320 39 52 136 140 144 152 169 171ndash174) legiti-mize speculations (see 13 32 63 95) or create dupli-cate or non-independent characters (see 13 30 70 77172) The lsquolateral organrsquo is a notable exampleYoshikura (1975) equated the embryonicearly post-embryonic lsquolateral organsrsquo of Amblypygi Thelyphonidaand Solifugae with the dissimilar lsquolateral organrsquo ofXiphosura but failed to note the very similar Cla-paregravede organ of Acariformes (see 173 174) This cod-ing was recycled by Wheeler amp Hayashi (1998) andthen by Giribet et al (2002) Error is probably inevi-table when assembling a large matrix from morphol-ogy including the one presented here but this can beminimized by making original observations and byconsulting primary sources

Some workers advocate weighting characters a pri-ori on the basis of structural or functional complexity(eg Kraus 1998 Dunlop amp Braddy 2001) and dismissphylogenetic conclusions derived from equal-weightsparsimony In short these workers criticize parsimonyfor emphasizing data quantity over data quality Thiscriticism ignores the intense debate in systematic biol-ogy that eventually led to widespread adoption of

Figure 4 Consensus trees produced by parsimony analy-sis of the full data matrix constrained to produce relation-ships proposed in the recent literature See legend toFigure 3 for details

Xi

Eu

Sc

Op

Pa

Ac

Ri

An

So

Ps

Tr

Ar

Ha

Am

UrEurypterida slat + Scorpiones12 435 9 lt5

Xi

Eu

Sc

Op

Pa

Ac

An

So

Ps

Ri

Tr

Ar

Ha

Am

UrTrigonotarbida+ Ricinulei28 434 8 lt5

Dunlop amp Braddy (2002)et al

Dunlop (1996)

234 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 5 Results from analysis of neontological data A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology AFor B and C relationships within terminal clades are the same as those shown in A

LimulusTachypleusProkoeneniaEukoeneniaCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAllothyrusAustralothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesLiphistiusAphonopelmaHypochilusCharinusPhrynusStenochrusProtoschizomusMastigoproctusCyphophthalmusChileogoveaCaddoLeiobunumSclerobunusGonyleptesCentruroidesHadrurusHeterometrusChthoniusFeaellaNeobisiumCheliferEremocostaGaleodes

10031

10018451

501561

733

723

944

633

10010

832

901

943

903

522

341

1007

1008

Acaromorpha

Haplocnemata

10012

Stomothecata

Opiliones

Scorpiones

Pseudo-scorpiones

Solifugae

Acariformes

Parasitiformes

Opilioacariformes

Ricinulei

Xiphosura

Palpigradi

Thelyphonida

Schizomida

Araneae

Amblypygi

10011

936853

100141005

1009

Tetrapulmonata Pedipalpi

Uropygi

692

522572

968

331

603

81

511

161

101

Acari

Dromopoda

Arachnida

A

Xiphosura

Stomothecata

Haplocnemata

Acaromorpha

Palpigradi

TetrapulmonataMicrura

Megoperculata

BXiphosura

Stomothecata

Acaromorpha

Haplocnemata

Palpigradi

TetrapulmonataMegoperculata

C

PHYLOGENY OF THE ARACHNID ORDERS 235

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 6 Results from analysis of the full data matrix A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology CFor B and C relationships within terminal clades are the same as those shown in A

WeinberginaLimuloidesEuproopsTachypleusLimulusStylonurusBaltoeurypterusChasmataspisDiploaspisOctoberaspisCyphophthamusChileogoveaCaddoLeiobunumSclerobunusGonyleptesProscorpiusPalaeoscorpiusStoermeroscorpioPrearcturusCentruroidesHadrurusHeterometrusProkoeneniaEukoeneniaTerpsicrotonPoliocheraCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAustralothyrusAllothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesEremocostaGaleodesChthoniusFeaellaNeobisiumCheliferGilboarachnePalaeocharinusLiphistiusAphonopelmaHypochilusPlesiosiroCharinusPhrynusStenochrusProtoschizomusProschizomusMastigoproctus

631

893

923

984

984

671

341

671

986

10011

423

974

Xiphosura

Eurypterida sstr

Chasmataspidida

Opiliones

Scorpiones

Palpigradi

Ricinulei

Opilioacariformes

Parasitiformes

Acariformes

Solifugae

Pseudoscorpiones

Trigonotarbida

Araneae

Haptopoda

Amblypygi

Schizomida

Thelyphonida

842

251331

361311

552

832

882

925

913

573

561291

443

976

541

954

743644

732521

571

331

833

663591

966

573

71

221

905

832

351

Uropygi

Pedipalpi

Haplocnemata

Tetrapulmonata

Anactinotrichida

Cryptognomae

Acaromorpha

Stomothecata

Pantetrapulmonata

Arachnida

Eurypterida slat

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatProscorpiusStoermeroscorpioPalaeoscorpiusPrearcturusCentruroidesHadrurusHeterometrusOpilionesHaplocnemataRicinuleiAnactinotrichidaAcariformesPalpigradiPantetrapulmonata

Megoperculata

Micrura

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatPalpigradiStomothecataHaplocnemataPantetrapulmonataAcaromorpha

A

B

C

Schizotarsata

236 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

parsimony and instead advocates a return to thespeculative and subjective approaches of the late 19thand early 20th century that once threatened thescientific legitimacy of the discipline (Bowler 1996)Furthermore it misrepresents the properties of parsi-mony-based analyses Specifically characters withfunctional significance are readily encompassed byparsimony analyses and in fact characters derivedfrom locomotor systems (eg 46ndash94) have played animportant role in developing current ideas aboutarachnid phylogeny (eg Shultz 1989 1991) In addi-tion lsquocomplex charactersrsquo can be viewed as compositesof several characters such that morphological com-plexity is effectively weighted by the number of inde-pendent lsquosubcharactersrsquo it contains For example thelsquosucking stomachrsquo once considered a synapomorphy ofLabellata (= Araneae + Amblypygi Fig 3) is codedhere as a composite of three characters (199ndash201) TheLabellata hypothesis was not corroborated in thepresent analysis (Figs 1 3ndash5) and in fact was highlydisfavoured (Fig 3) but this result cannot be dis-missed as a failure to acknowledge the complexity ofthe character

Some workers support specific (target) clades withone or more similarities without exploring the impacton overall homoplasy or relationships among non-target clades Each character offers its own phyloge-netic hypothesis which may or may not be consistentwith relationships implied by other characters It isexceedingly rare for all characters to be perfectly com-patible in the phylogenetic hypotheses they supportand criteria such as parsimony have been developed todiscover those hypotheses that minimize the conflict-ing phylogenetic signals of different characters Eventhough character conflict (homoplasy) is a virtuallyinescapable phenomenon in comparative biology it isnot uncommon for workers to discover one or morecharacters and to promote their phylogeneticimplications over alternative hypotheses even thosethat otherwise appear to be well supported Thisapproach may have value in highlighting new dataand perspectives but accomplishes this by promotingthe erroneous impression that all aspects of arachnidphylogeny are so tenuous that a single character canfalsify even well-supported hypotheses Several exam-ples of this approach have appeared in the recentarachnological literature

For example Alberti amp Peretti (2002) argued thataflagellate spermatozoa (163) are a compelling syna-pomorphy for a Solifugae + Acari clade and dismissedsome of the characters that support Solifugae + Pseu-doscorpiones as having lsquodebatable valuersquo Yet theirproposal rejects two hypotheses that are consistentlyrecovered in recent phylogenetic analyses that isHaplocnemata (= Solifugae + Pseudoscorpiones) andAcaromorpha (= Acari + Ricinulei) (Figs 1 2) Rejec-

tion of Haplocnemata would require its presumed syn-apomorphies to be reinterpreted as homoplasiesincluding features of the chelicerae (18 19 20) pre-oral chamber (13 32) legs (12 48) and respiratorysystem (126) The same reason applies to Acaromor-pha and its synapomorphies It is noteworthy that aSolifugae + Acari clade was recovered in the presentanalysis in fewer than 5 of bootstrap pseudorepli-cates and that analyses constrained to recover thisclade were five steps longer than the minimal-lengthtree and favoured a problematic clade unitingRicinulei and Pseudoscorpiones (Fig 3)

In another example Dunlop (1996) proposed a closerelationship between Trigonotarbida and Ricinuleibased on two-segmented chelicerae (18) prosomandashopisthosoma coupling mechanism (96) diplotergites(100 101) and longitudinally divided opisthosomaltergites (115) However phylogenetic analyses con-strained to recover this relationship required eightadditional steps eliminated support for Acari and nec-essarily rejected Acaromorpha and forced its synapo-morphies to be reinterpreted as homoplasies Alberti(2005) proposed an interesting hypothesis for the evo-lution of male gonads in tetrapulmonates (153) butchose to accept Labellata (= Araneae + Amblypygi) indeveloping his argument over the much more well-supported Pedipalpi (= Amblypygi + Uropygi) (Shultz1999) a phylogenetic reconfiguration that was recov-ered here in fewer than 5 of bootstrap pseudorepli-cates increased tree length by a minimum of 14 stepsand resulted in the collapse of Stomothecata and Hap-locnemata (Fig 3) Many other examples can be cited

Promoting or defending a specific phylogenetichypothesis via lists of compatible synapomorphies is acommon but problematic approach By restrictingattention to the states of specific characters at one ortwo target nodes one can easily overlook the unin-tended impact of the hypothesis on phylogenetic sig-nal elsewhere and its effect on non-target clades Anode supported by a long list of synapomorphies mayseem convincing taken in isolation but may becomeless acceptable when its full phylogenetic implicationsare explored

SUMMARY OF INTERORDINAL ARACHNID CLADES RECOVERED IN THIS ANALYSIS

Arachnida Lamarck 1801Analyses consistently recovered Arachnida as a mono-phyletic group with high bootstrap support (Figs 1 45) Possible synapomorphies include the loss of thecarapacal pleural doublure (9) cardiac lobe (10) pedalgnathobases (52) and moveable endites (53) and thegain of aerial respiration (120) and an anteriorly oranteroventrally directed mouth (185) Some tradi-tional synapomorphies such as slit sensilla (142) and

PHYLOGENY OF THE ARACHNID ORDERS 237

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

fluid feeding (184) may have appeared later in arach-nid evolution but this can only be decided once theinternal phylogeny of Arachnida has been determined

Some workers regard many proposed arachnid syn-apomorphies as adaptations to terrestrial life andthereby link the hypothesis of arachnid monophyly tothe hypothesis of a single ancestral aquatic-to-terres-trial transition and arachnid polyphyly to multipletransitions (eg Selden amp Jeram 1989 Dunlop 1997Dunlop amp Webster 1999 Dunlop amp Braddy 2001)Given this line of reasoning the existence of appar-ently aquatic scorpions in the fossil record (Kjellesvig-Waering 1986 Jeram 1998) and the inference thatterrestrialization occurred late in scorpion evolutionthese workers conclude that character states support-ing arachnid monophyly are actually convergencesand do not necessarily support arachnid monophyly

However this approach to assessing phylogenetichypotheses is founded on overly simplistic assump-tions such as the ability of the investigator to discrim-inate unerringly between characters that existexclusively in aquatic organisms (including fossils)from those that occur exclusively in terrestrial organ-isms There also appears to be an assumption thathomoplasy can be generated through parallelism (iemultiple aquatic-to-terrestrial events) but not throughterrestrial-to-aquatic reversals Furthermore anassumed dichotomy between exclusively aquatic andterrestrial life histories in ancestral arachnids is sim-plistic as illustrated by the amphibious life cycles ofbasal vertebrates and pterygote hexapods In factthese examples demonstrate that there is no neces-sary inconsistency in basing a hypothesis of arachnidmonophyly on the derived terrestrial features of anamphibious ancestor whose descendants thencompleted terrestrialization once or several timesindependently or even returned to a fully aquaticexistence Workers who link the frequency and direc-tion of aquaticndashterrestrial transitions to the assess-ment of arachnid phylogeny do so by endowingthemselves with substantially greater insight thanseems prudent by ignoring the huge gaps in ourunderstanding of early arachnid evolution and bydenying to arachnids the evolutionary complexityknown to exist in other groups

Several palaeontologists have been particularlyactive during the past decade in proposing new char-acters with the stated goal of removing Scorpionesfrom Arachnida and erecting a Scorpiones + Eurypter-ida clade (Braddy amp Dunlop 1997 Dunlop amp Braddy1997 Dunlop 1998 Braddy et al 1999 Dunlop ampWebster 1999) Dunlop amp Braddy (2001) recentlysummarized this evidence and conducted aparsimony-based analysis of Xiphosura EurypteridaScorpiones Opiliones and Tetrapulmonata (but notHaplocnemata Acaromorpha or Palpigradi) using 33

morphological characters Their analysis produced atopology congruent with those generated here (Figs 15 6) including recovery of Stomothecata (= Scorpiones+ Opiliones) However they rejected this result as aproduct of lsquoempirical cladisticsrsquo because it gives thesame weight to prosomal characters that supportarachnid monophyly and to selected opisthosomalcharacters that support their favoured Eurypterida +Scorpiones clade namely a five-segmented metasoma(116) suppression of opisthosomal tergite 1 (95) lossof lamellate respiratory organs on the postgenitalsomite (122) Kiemenplatten (125) loss of respiratorylamellae on the genital segment (121) and a lsquonon-stainingrsquo exocuticle (but see Grainge amp Pearson 1966for evidence of this in Opiliones see Appendix for com-ments on the other characters)

A Eurypterida + Scorpiones clade was not favouredin the present analysis (Fig 6) the strict consensus ofminimal-length trees constrained to recover this clade(Fig 4) is nine steps longer than the tree recovered byanalysis without this constraint It is also noteworthythat Opiliones was consistently reconstructed as thesister group to Eurypterida + Scorpiones in the mini-mal-length constrained trees a provocative result thatwas probably unintended and unanticipated by sup-porters of the Eurypterida + Scorpiones Given thatDunlop amp Braddy (2001) reject equal weights parsi-mony as an arbiter of phylogenetic hypotheses theywould presumably dismiss these results as irrelevantto their argument just as they dismissed their ownparsimony-based results However if the Eurypterida+ Scorpiones hypothesis is to be credible it must beopen to evaluation and potential falsification usingobjective criteria and the subjective or intuitive a pri-ori weighting of characters advocated by Dunlop ampBraddy clearly does not qualify At present it isunclear how one would objectively evaluate Dunlop ampBraddyrsquos proposal with criteria compatible with thoseused in its original formulation For now the Euryp-terida + Scorpiones concept advocated by Dunlop ampBraddy may persist outside the mainstream of mod-ern systematic practice but it is increasingly problem-atic within it

Stomothecata nomnovOpiliones and Scorpiones were consistently recoveredas a monophyletic group The proposed name acknowl-edges a unique preoral chamber the stomothecaformed by coxapophyses of the palp and leg 1 (50)often with an auxiliary role played by the coxapophy-sis of leg 2 (51) In addition the epistome appears tohave been modified for adduction of the palpal coxaeThe lateral walls of the epistome are fused to themedial surfaces of the palpal coxae and the epistomallumen is spanned by a transverse muscle (188) which

238 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

apparently adducts the palpal coxae thereby constrict-ing the stomothecal chamber Scorpions and opilionsare also unique in having a pair of large epistomalarms projecting rearward into the prosoma andattaching to the endosternite (189) The epistomalarms provide attachment sites for pharyngeal dilatormuscles (196) and extrinsic muscles of anterior proso-mal appendages (eg 37) The chelicera is equippedwith a muscle that arises on the carapace and insertson the ventral margin of the second cheliceral article(23) There is an anteriorly placed genital opening(155)

Several notable similarities were found whilereviewing the literature but information from otherarachnid groups was considered too incomplete toallow them to be included in the matrix For exampleboth orders have apparent haemocytopoeitic organsassociated with major nerves of the anterior opistho-soma These are termed supraneural organs in scorpi-ons (Farley 1999) and perineural organs in opilions(Kaumlstner 1935) Haemocytes develop in the cardiacwall in spiders (Seitz 1972) and perhaps amblypygids(Weygoldt 2000) but haemocytopoeitic organs areunknown in most other arachnid groups Germ cellsdifferentiate very early during embryogenesis in bothscorpions and opilions relative to spiders (Moritz1957 Anderson 1973) Additional research is requiredto determine the phylogenetic utility of thesecharacters

Haplocnemata Boumlrner 1904Several workers recognize a close phylogenetic rela-tionship between Pseudoscorpiones and Solifugae(Weygoldt amp Paulus 1979 Van der Hammen 1989Shultz 1990 Wheeler amp Hayashi 1998 Giribetet al 2002) and this was consistently recovered inthe present analysis Haplocnemata is supported byseveral synapomorphies including (i) a feeding com-plex formed by two-segmented chelicerae (18) with aventrolateral intrinsic articulation (19) and perhapsa dorsolateral articulation with the carapace (20)and (ii) a rostrosoma (32) a preoral apparatus formedby an anteriorly projecting epistome affixed dorsallyto enlarged palpal coxae and bordered by lateral pal-pal projections The epistome forms the dorsal compo-nent of a lsquobeakrsquo with the ventral part formed by amidventral sternapophysis (13) fused between thepalpal coxae The locomotor apparatus consists ofcoxae that meet along the ventral midline (12) elon-gate femur-like patellae (48) and an apotele in theadult modified as an eversible empodium or pulvillus(92) The respiratory system is formed entirely (Pseu-doscorpiones) or in part (Solifugae) by paired trachealstigmata opening on opisthosomal somites 3 and 4(126)

Acaromorpha Dubinin 1957A clade comprising Ricinulei and Acari is consistentlyrecovered in the present analysis It is united here bytwo unique and seven homoplasious synapomorphiesThese include a gnathosoma (31) defined in part bymedial fusion of the palpal coxae (30) although pres-ence of a gnathosoma in Ricinulei is debatable as dis-cussed in the Appendix Acaromorphs also have aunique post-embryonic development consisting of ahexapodal larva and up to three octopodal nymphalinstars (176) The pedal patellandashtibia joints are formedby a bicondylar dorsal hinge rather than a single mid-dorsal condyle (69) all postcheliceral segmental gan-glia are unified in the subesophageal ganglion (130)and a postcerebral pharynx is absent (192) The groupis also tentatively united by presence of differentiatedpedal basi- and telofemora (63 64) and the absence ofa ventral (sternal) pharyngeal dilator muscle (199)but these may be symplesiomorphies erroneouslyreconstructed as a synapomorphies

The internal phylogenetic structure of Acaromorphais controversial with many recent workers favouringa monophyletic Acari (Weygoldt amp Paulus 1979Shultz 1990 argued most thoroughly by Lindquist1984) and others advocating a diphyletic Acari withAcariformes being the sister group to Cryptognomae(= Ricinulei + Anactinotrichida) (especially Van derHammen 1979 1989) (Fig 3) Acari was recovered asmonophyletic when fossils were excluded (Fig 2) butnot when they were included (Fig 3) A lsquomite-centredrsquosurvey of arachnid characters may be needed if mor-phology is to offer a compelling solution to the internalphylogeny of Acari and its placement within Arach-nida These issues are not resolved by the presentanalysis

Pantetrapulmonata nom novPantetrapulmonata includes the extinct orders Trigo-notarbida Haptopoda and the extant orders AraneaeAmblypygi Schizomida and Thelyphonida The cladeis united by cheliceral structure (18 19) a megoper-culum (106) booklungs on the genital and first post-genital somites (121 122) and enlargement of theepipharyngeal sclerite (192) It is important to notehowever that most of these characters were coded asuncertain in Plesiosiro (Haptopoda) Aside from theplacement of Haptopoda the monophyly of Trigono-tarbida and the extant orders was anticipated byShear et al (1987) and is generally regarded as amonophyletic group

Tetrapulmonata Shultz 1990Tetrapulmonata was originally proposed on the basisof neontological analyses and encompassed Araneae

PHYLOGENY OF THE ARACHNID ORDERS 239

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Amblypygi Schizomida and Thelyphonida (Shultz1990) and was also recovered here in analysis of neon-tological data (Fig 5) However results from analysisof all taxa required that the Tetrapulmonata conceptbe expanded to include the fossil order HaptopodaFeatures uniting this group are problematic howeveras many states in Haptopoda were coded as uncertain

Schizotarsata nomnovHaptopoda and Pedipalpi (= Amblypygi + Schizomida+ Thelyphonida) are united here in a group named forpossession of divided pedal telotarsi (84) Synapomor-phies include a pointed anterior carapacal margin (3)and elongation of leg 1 (46) As already noted theplacement of Haptopoda should probably be regardedas tentative because the state of many characters inthis group was coded as unknown The placement ofHaptopoda as the sister group to Pedipalpi was antic-ipated by Dunlop (1999 2002c)

Pedipalpi Boumlrner 1904Pedipalpi encompasses Amblypygi and Uropygi(= Thelyphonida + Schizomida) Although widely rec-ognized in the past Pedipalpi was set aside in the mid-20th century in favour of Labellata (= Amblypygi +Araneae) based largely on the presumed synapomor-phies of a narrow prosomandashopisthosoma juncture (97)and sucking stomach (199ndash201) The situation beganto reverse when Shear et al (1987) again highlightedsimilarities of Amblypygi and Uropygi such as rapto-rial palps (35) and antenniform leg 1 (46) Shultz(1989 1990 1999) subsequently described numerousderived similarities including asymmetrical flexormuscles at the pedal femurndashpatella joint (70) threetelotarsomeres on legs 2ndash4 (85) and modification ofthe palpal coxae to provide support for extrinsic pha-ryngeal muscles (197) Pedipalpi has emerged as oneof the most well-supported interordinal relationshipsin Arachnida

Uropygi Thorell 1882Uropygi has long been accepted as a monophyleticunion of Thelyphonida and Schizomida Synapomor-phies include a unique mating behaviour (159) fusedpalpal coxae (30) 2-1-1-1 arrangement of tibial tricho-bothria (144) posterior defensive glands (102) elon-gated patella of leg 1 (68) and many others

ACKNOWLEDGEMENTS

I thank Nikolaj Scharff and three anonymous review-ers for useful comments This work was supported bythe Maryland Agricultural Experiment Station the

General Research Board of the University of Mary-land and National Science Foundation Grant 0416628to Susan Masta

REFERENCES

Abd el-Wahab A 1952 Notes on the morphology of the scor-pion Buthus quinquestriatus (HE) Publications de lrsquoInsti-tut Fouad 1er du Desert 3 1ndash129

Alberti G 1979 Licht- und elektronenmikroskopische Unter-suchungen an Coxaldruumlsen von Walzenspinnen (ArachnidaSolifugae) Zoologischer Anzeiger 203 48ndash64

Alberti G 1995 Comparative spermatology of Cheliceratareview and perspective Memoires du Museum NationaldrsquoHistoire Naturelle 166 203ndash230

Alberti G 2005 Double spermatogenesis in ChelicerataJournal of Morphology 266 281ndash297

Alberti G Coons LB 1999 Acari mites In Harrison FWFoelix RF eds Microscopic anatomy of invertebratesVol 8C Chelicerate Arthropoda New York Wiley Liss515ndash1215

Alberti G Peretti AV 2002 Fine structure of male genitalsystem and sperm in Solifugae does not support a sister-group relationship with Pseudoscorpiones (Arachnida)Journal of Arachnology 30 268ndash274

Alexander AJ 1967 Problems of limb extension in the scor-pion Opisthophthalmus latimanus Koch Transactions of theRoyal Society of South Africa 37 165ndash181

Anderson DT 1973 Embryology and phylogeny in annelidsand arthropods Oxford Pergamon Press

Anderson LI Selden PA 1997 Opisthosomal fusion andphylogeny of Palaeozoic Xiphosura Lethaia 30 19ndash31

Beall BS Labandeira CC 1990 Macroevolutionary patternsof the Chelicerata and Tracheata In Culver SJ ed Arthro-pod paleobiology Short Courses in Paleontology No 3 Lon-don The Paleontological Society 257ndash284

Bernard HM 1896 The comparative morphology of the Gale-odidae Transactions of the Linnean Society of London Sec-ond Series 6 305ndash417

Boumlrner C 1902a Arachnologische Studien (II und III) Zool-ogischer Anzeiger 25 433ndash466

Boumlrner C 1902b Arachnologische Studien IV Die Genitalor-gane der Pedipalpen Zoologischer Anzeiger 26 81ndash92

Boumlrner C 1904 Beitraumlge zur Morphologie der Arthropoden IEin Beitrag zur Kenntnis der Pedipalpen Zoologica (Stut-tgart) 42 1ndash174

Bowler PJ 1996 Lifersquos splendid drama Chicago Universityof Chicago Press

Braddy SJ Alridge RJ Gabbott SE Theron JN 1999Lamellate book-gills in a late Ordovician eurypterid from theSoom Shale South Africa support for a eurypterid-scorpionclade Lethaia 32 72ndash74

Braddy SJ Dunlop JA 1997 The functional morphology ofmating in the Silurian eurypterid Baltoeurypterus tet-ragonophthalmus (Fischer 1839) Zoological Journal of theLinnean Society 121 435ndash461

Brauer A 1895 Beitrage zur Kenntnis der Entwicklungsge-schichte des Skorpiones II Zeitschrift fuumlr WissenschaftlicheZoologie 59 351ndash433

240 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bremer K 1994 Branch support and tree stability Cladistics10 295ndash304

Bristowe WS 1932 The liphistiid spiders Proceedings of theZoological Society of London Part 4 1016ndash1057

Butt AG Taylor HH 1991 The function of spider coxalorgans effects of feeding and salt-loading on Porrhotheleantipodiana (Mygalomorpha Dipluridae) Journal of Exper-imental Biology 158 439ndash461

Buxton BH 1913 Coxal glands of arachnids ZoologischeJahrbucher (Suppl XIV) 16ndash50

Buxton BH 1917 Notes on the anatomy of arachnids Jour-nal of Morphology 29 1ndash31

Camin JH Clark GM Bourdeau FG 1956 The palpallsquotined setarsquo in the Mesostigmata a homologue of the palpalclaw in the Onychopalpida (Acarina) In Becker EC ed Pro-ceedings of the Tenth International Congress of EntomologyOttawa Mortimer 903ndash908

Chamberlin JC 1931 The arachnid order ChelonethidaStanford University Stanford University Press

Chapman RF 1998 The insects structure and function 4thedn Cambridge Cambridge University Press

Clarke KU 1979 Visceral anatomy and arthropod phylogenyIn Gupta AP ed Arthropod phylogeny New York Van Nos-trand Reinhold 467ndash549

Clarke J 1984 On the relationship between structure andfunction in the leg joints of Heteropoda venatoria (L) (Ara-neae Eusparassidae) Bulletin of the British ArachnologicalSociety 6 181ndash192

Clarke J 1986 The comparative functional morphology of theleg joints and muscles of five spiders Bulletin of the BritishArachnological Society 7 37ndash47

Clarke JM Ruedemann R 1912 The Eurypterida of NewYork New York State Museum Memoir 14 (1ndash2) 1ndash628

Coddington JA 2005 Phylogeny and classification of spi-ders In Ubick D Paquin P Cushing PE Roth V eds Spi-ders of North America an identification manual AmericanArachnological Society 18ndash24

Coddington JA Giribet G Harvey MS Prendini LWalter DE 2004 Arachnida In Cracraft J Donoghue MJeds Assembling the tree of life Oxford Oxford UniversityPress 269ndash318

Coddington JA Levi HW 1991 Systematics and evolutionof spiders (Araneae) Annual Review of Ecology and System-atics 22 565ndash592

Coineau Y Van der Hammen L 1979 The postembryonicdevelopment of Opilioacarida with notes on new taxa and ona general model for the evolution In Piffle E ed Proceed-ings of the 4th International Congress of Acarology Budap-est Akadeacutemiai Kiadoacute 437ndash441

Cokendolpher JC Reddell JR 1992 Revision of the Pro-toschizomidae (Arachnida Schizomida) with notes on thephylogeny of the order Texas Memorial Museum Speleolog-ical Monographs 3 31ndash74

Condeacute B 1991a Le dimorphisme sexuel des Palpigrades Bul-letin de la Socieacuteteacute Neuchacircteloise des Sciences Naturelles 11667ndash73

Condeacute B 1991b Lrsquoaire genitale male des Palpigrades Bulle-tin de la Socieacuteteacute Europeenne drsquoArachnologie 1990 64ndash69

Dearden PK Donly C Grbicccc M 2002 Expression of pair-rule gene homologues in a chelicerate early patterning ofthe two-spotted spider mite Tetranychus urticae Develop-ment 129 5461ndash5472

Delle Cave L 1975 The ldquocleaning brushrdquo in Damon diademaSimon 1876 and Phrynichus jayakari Pocock 1894 (Ambly-pygi Arachnida) Monitore Zoologico Italiano (NS) 9 175ndash184

Dunlop JA 1994 Filtration mechanisms in the mouthparts oftetrapulmonate arachnids (Trigonotarbida Araneae Ambly-pygi Uropygi Schizomida) Bulletin of the British Arachno-logical Society 9 267ndash273

Dunlop JA 1996 Evidence for a sister group relationshipbetween Ricinulei and Trigonotarbida Bulletin of the BritishArachnological Society 10 193ndash204

Dunlop JA 1997 Palaeozoic arachnids and their significancefor arachnid phylogeny In Zabka M ed Proceedings of the16th European colloquium of arachnology Siedlce PolandWyzsza Szkola Rolnicko-Pedagogiczna 65ndash82

Dunlop JA 1998 The origins of tetrapulmonate book lungsand their significance for chelicerate phylogeny In SeldenPA ed Proceedings of the 17th European Colloquium ofArachnology Edinburgh 1997 Burnham Beeches BritishArachnological Society 9ndash16

Dunlop JA 1999 A redescription of the Carboniferous arach-nid Plesiosiro madeleyi Pocock 1911 (Arachnida Hapto-poda) Transactions of the Royal Society of Edinburgh EarthSciences 90 29ndash47

Dunlop JA 2000 The epistomo-labral plate and lateral lips insolifuges pseudoscorpions and mites Ekoloacutegia (Bratislava)19 67ndash78

Dunlop JA 2002a Arthropods from the Lower Devonian Sev-ernaya Zemlya Formation of October Revolution Island(Russia) Geodiversitas 24 349ndash379

Dunlop JA 2002b Character states and evolution of the che-licerate claws In Toft S Scharff N eds European Arach-nology 2000 Aarhus Aarhus University Press 345ndash354

Dunlop JA 2002c Phylogeny of Chelicerata In LlorenteBousquets J Morrone JJ eds Biodiversidad Taxonomiacutea yBiogeografiacutea de Arthroacutepodos de Meacutexico Hacia Una Siacutentesisde Su Conocimiento Vol III Prensas de Ciencias Univer-sidad Nacional Autonoma de Meacutexico 117ndash141

Dunlop JA Anderson LI Braddy SJ 2004 A redescriptionof Chasmataspis laurencii Caster amp Brooks 1956 (Chelicer-ata Chasmataspidida) from the Middle Ordovician of Ten-nessee USA with remarks on chasmataspid phylogenyTransactions of the Royal Society of Edinburgh Earth Sci-ences 94 207ndash225

Dunlop JA Braddy SJ 1997 Slit-like structures on theprosomal appendages of the eurypterid BaltoeurypterusNeues Jahrbuch fuumlr Geologie und Palaumlontologie-Monatshefte1997 31ndash38

Dunlop JA Braddy SJ 2001 Scorpions and their sister-group relationships In Fet V Selden PA eds Scorpions2001 in Memoriam Gary A Polis London British Arachno-logical Society 1ndash24

Dunlop JA Horrocks CA 19951996 A new Upper Carbon-iferous whip scorpion (Arachnida Uropygi Thelyphonida)

PHYLOGENY OF THE ARACHNID ORDERS 241

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

with a revision of the British Carboniferous Uropygi Zoolo-gischer Anzeiger 234 293ndash306

Dunlop JA Poschmann M Anderson L 2001 On theEmsian (early Devonian) arthropods of the Rhenish SlateMountains 3 The chasmataspidid Diploaspis Palaumlon-tolgische Zeitschrift 75 253ndash269

Dunlop JA Webster M 1999 Fossil evidence terrestrializa-tion and arachnid phylogeny Journal of Arachnology 27 86ndash93

Edgecombe GD Wilson GDF Colgan DJ Gray MR Cas-sis G 2000 Arthropod cladistics combined analysis of his-tone H3 and U2 snRNA sequences and morphologyCladistics 16 155ndash203

Evans GO 1992 Principles of acarology Wallingford UKCABI

Fahrenbach WH 1999 Merostomata In Harrison FW Foe-lix RF eds Microscopic anatomy of invertebrates Vol 8AChelicerate arthropods New York Wiley-Liss 21ndash115

Farley RD 1999 Scorpiones In Harrison FW Foelix RF edsMicroscopic anatomy of invertebrates Vol 8A Cheliceratearthropods New York Wiley-Liss 117ndash222

Farley RD 2005 Developmental changes in the embryo pro-nymph and first molt of the scorpion Centruroides vittatus(Scorpiones Buthidae) Journal of Morphology 265 1ndash27

Fayers SR Dunlop JA Trewin NH 2004 A new earlyDevonian trigonotarbid arachnid from the Windyfield ChertRhynie Scotland Journal of Systematic Palaeontology 2269ndash284

Felsenstein J 1985 Confidence limits on phylogenies anapproach using the bootstrap Evolution 39 783ndash791

Fet V Sissom WD Lowe G Braunwalder ME 2000 Cat-alog of the scorpions of the world (1758ndash1998) New YorkNew York Entomological Society

Firstman B 1954 The central nervous system musculatureand segmentation of the cephalothorax of a tarantula (Eu-rypelma californicum Ausserer) Microentomology 87 1ndash40

Firstman B 1973 The relationship of the chelicerate arterialsystem to the evolution of the endosternite Journal ofArachnology 1 1ndash54

Foelix RF 1985 Mechano- and chemoreceptive sensilla InBarth FG ed Neurobiology of arachnids Berlin Spinger-Verlag 118ndash137

Foelix RF 1996 Biology of spiders 2nd edn Oxford OxfordUniversity Press

Giribet G Edgecombe GD Wheeler WC Babbit C 2002Phylogeny and systematic position of Opiliones a combinedanalysis of chelicerate relationships using morphologicaland molecular data Cladistics 18 5ndash70

Goloboff P 1993 Estimating character weights during treesearch Cladistics 9 83ndash91

Goloboff PA Farris JS Nixon K 2000 TNT Tree AnalysisUsing New Technologies Version 1 Program and documen-tation httpwwwzmucdkpublicphylogenytnt

Grainge CA Pearson RG 1966 Cuticular structure in thePhalangida Nature 211 866

Grandjean F 1938 Observations sur les Bdelles (Acariens)Annales de la Socieacuteteacute Entomologique de France Paris 107 1ndash24

Hansen HJ Soslashrensen W 1904 On two orders of ArachnidaCambridge Cambridge University Press

Hansen HJ Soslashrensen W 1905 The Tartarides a tribe of theorder Pedipalpi Arkiv fuumlr Zoologi 2 1ndash78

Hara MR Gnaspini P 2003 Comparative study of the defen-sive behavior and morphology of the gland opening areaamong harvestmen (Arachnida Opiliones Gonyleptidae)under a phylogenetic perspective Arthropod Structure andDevelopment 32 257ndash275

Harvey MS 1992 The phylogeny and classification of thePseudoscorpionida (Chelicerata Arachnida) InvertebrateTaxonomy 6 1373ndash1435

Harvey MS 2002 The neglected cousins what do we knowabout the smaller arachnid orders Journal of Arachnology30 357ndash372

Haupt J Song D 1996 Revision of East Asian whip scorpions(Arachnida Uropygi Theyphonida) I China and JapanArthropoda Selecta 5 43ndash52

Hjelle JT 1990 Anatomy and morphology In Polis GA edThe biology of scorpions Stanford University Stanford Uni-versity Press 9ndash63

Janczyk FSW 1956 Anatomie von Siro duricorius Joseph imVergleich mit anderen Opilioniden Sitzungsberichte Oster-reichischen Akademie der Wissenschaften Mathematisch-Natruwissenschaftliche Klasse 165 474ndash522

Jenner RA 2001 Bilaterian phylogeny and uncritical recy-cling of morphological data sets Systematic Biology 50 730ndash742

Jeram AJ 1998 Phylogeny classification and evolution of Sil-urian and Devonian scorpions In Selden PA ed Proceed-ings of the 17th European Colloquium of ArachnologyEdinburgh 1997 Burnham Beeches UK British Arachno-logical Society 15ndash31

Karaman IM 2005 Evidence of spermatophores in Cyphoph-thalmi (Arachnida Opiliones) Revue Suisse de Zoologie 1123ndash11

Kaumlstner A 1935 Die Funktion der sogenannten sympathis-chen Ganglien und die Exkretion bei den PhalangiidenZoologischer Anzeiger 109 273ndash288

Kierans JE Robbins RG 1999 A world checklist of generasubgenera and species of ticks (Acari Ixodida) publishedfrom 1973 to 1997 Journal of Vector Ecology 24 115ndash129

Kjellesvig-Waering EN 1986 A restudy of the fossil Scorpi-onida of the world Palaeontographica Americana 55 1ndash287

Klompen JSH 2000 Prelarva and larva of Opilioacarus (Neo-carus) texanus (Chamberlin and Mulaik) (Acari Opilioacar-ida) with notes on the patterns of setation and lyrifissuresJournal of Natural History 34 1977ndash1992

Kraus O 1998 Elucidating the historical process of phylog-eny phylogenetic systematics versus cladistic techniquesIn Selden PA ed Proceedings of the 17th European Collo-quium of Arachnology Edinburgh 1997 Burnham BeechesUK British Arachnological Society 1ndash7

Kraus O Kraus M 1993 Divergent transformation of cheli-cerae and original arrangement of eyes in spiders (Arach-nida Araneae) Memoirs of the Queensland Museum 33579ndash584

Lankester ER Benham WBS Beck EJ 1885 On the mus-

242 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

cular and endoskeletal systems of Limulus and Scorpio withsome notes on the anatomy and generic characters of scor-pions Transactions of the Zoological Society of London 11311ndash384

Leclerc P 1989 Considerations paleobiogeographiques a pro-pos la decouverte en Thailande drsquoopilioacariens nouveaux(Acari ndash Notostigmata) Compte Rendu des Seances de laSociete de Biogeographie 65 162ndash174

Legg G 1976 The external morphology of a new species ofricinuleid (Arachnida) from Sierra Leone Zoological Journalof the Linnean Society 59 1ndash58

Lindquist EE 1984 Current theories on the evolution ofmajor groups of Acari and on their relationships with othergroups of Arachnida with consequent implications for theirclassificaton In Griffiths DA Bowman CE eds AcarologyVI Vol 1 New York John Wiley amp Sons 28ndash62

Manning PL Dunlop JA 1995 The respiratory organs ofeurypterids Palaeontology 38 287ndash297

Manton SM 1958 Hydrostatic pressure and leg exten-sion in arthropods with special reference to arachnidsAnnals and Magazine of Natural History Series 13 1161ndash182

Manton SM 1964 Mandibular mechanisms and the evolutionof arthropods Philosophical Transactions of the Royal Soci-ety of London B 7 1ndash183

Maraun M Heethoff M Schneider K Scheu S WeigmannG Cianciolo J Thomas RH Norton RA 2004 Molecularphylogeny of oribatid mites (Oribatida Acari) evidence formultiple radiations of parthenogenetic lineages Experimen-tal and Applied Acarology 33 183ndash201

Marples BJ 1967 The spinnerets and epiandrous glands ofspiders Journal of the Linnean Society (Zoology) 46 209ndash222

Marples BJ 1983 Observations on the structure of the fore-gut of spiders Bulletin of the British Arachnological Society8 46ndash52

Millot J 1933 Notes compleacutementaires sur lrsquoanatomie desliphistiides et des hypchilides a propos drsquoun travail reacutecent deA Petrunkevitch Bulletin de la Socieacuteteacute Zoologique de France68 217ndash235

Millot J 1942 Sur lrsquoanatomie et lrsquohistophysiologie deKoenenia mirabilis Grassi (Arachnida Palpigradi) RevueFranccedilaise drsquoEntomologie 9 33ndash51

Millot J 1943 Notes complementaires sur lrsquoanatomielrsquohistologie et la repartition geographique en France deKoenenia mirabilis Grassi Revue Franccedilaise drsquoEntomologie 9127ndash135

Millot J 1945 Lrsquoanatomie interne des Ricinulei (Arachnides)Annales des Sciences Naturelles (Zoologie) Series 11 7 1ndash29

Millot J 1949a Morphologie generale et anatomie interne InGrasseacute P-P ed Traite de zoologie VI Paris Masson et Cie263ndash319

Millot J 1949b Ordre des Amblypyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 563ndash588

Millot J 1949c Ordre des Araneacuteides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 589ndash743

Millot J 1949d Ordre des Palpigrades In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 520ndash532

Millot J 1949e Ordre des Uropyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 533ndash562

Millot J 1949f Ordre des Ricinuleides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 744ndash760

Millot J Vachon M 1949 Ordre Des Solifuges In Grasseacute P-P ed Traite de zoologie VI Paris Masson et Cie 482ndash519

Moore RA Briggs DEG Bartels C 2005 A new specimen ofWeinbergina opitzi (Chelicerata Xiphosura) from the LowerDevonian Huumlnsruck Slate Germany PalaumlontologischeZeitschrift 79 399ndash408

Moritz M 1957 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) unter besonderer Beruumlcksichtigungder ausseren Morphologie der Bildung des Mitteldarmesund der Genitalanlage Zoologische Jahrbuumlcher Abteilungfuumlr Anatomie und Ontogenie der Tiere 76 331ndash370

Moritz M 1959 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) II Die Anlage und Entwicklung derCoxaldruumlse bei Phalangium opilio L Zoologische Jahr-buumlcher Abteilung fuumlr Anatomie und Ontogenie der Tiere 77229ndash240

Nixon K 1999 The parsimony ratchet a new method forrapid parsimony analysis Cladistics 15 407ndash414

Norton RA 1998 Morphological evidence for the evolutionaryorigin of Astigmata (Acari Acariformes) Experimental andApplied Acarology 22 559ndash594

Palmgren P 1978 On the muscular anatomy of spiders ActaZoologica Fennica 155 1ndash41

Patten W 1912 The evolution of the vertebrates and their kinPhiladelphia P Blakistonrsquos Son

Patten W Hazen AP 1900 The development of the coxalgland branchial cartilages and genital ducts of Limuluspolyphemus Journal of Morphology 16 459ndash490

Paulus HF 1979 Eye structure and the monophyly of theArthropoda In Gupta AP ed Arthropod phylogeny NewYork Van Nostrand Reinhold 299ndash383

Petrunkevitch A 1949 A study of Palaeozoic ArachnidaTransactions of the Connecticut Academy of Arts and Sci-ences 37 69ndash315

Pittard K Mitchell RW 1972 Comparative morphology ofthe life stages of Cryptocellus pelaezi (Arachnida Ricinulei)Graduate Studies Texas Tech University 1 1ndash77

Platnick NI 2005 The world spider catalog Version55 American Museum of Natural History httpresearchamnhorgentomologyspiderscatalogindexhtm

Platnick NI Gertsch WJ 1976 The suborders of spiders acladistic analysis (Arachnida Araneae) American MuseumNovitates 2607 1ndash15

Pocock RI 1893 On some points in the morphology of theArachnida (ss) with notes on the classification of the groupAnnals and Magazine of Natural History Series 6 11 1ndash19

Pocock RI 1902 Some points in the morphology and classi-fication of the Opiliones Annals and Magazine of NaturalHistory Series 7 10 504ndash516

Popadicccc A Panganiban G Rusch D Shear WA KaufmanTC 1998 Molecular evidence for the gnathobasic derivationof arthropod mandibles and for the appendicular origin ofthe labrum and other structures Development Genes andEvolution 208 142ndash150

PHYLOGENY OF THE ARACHNID ORDERS 243

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Raw F 1957 Origin of chelicerates Journal of Paleontology31 139ndash192

Reddell JR Cokendolpher JC 1995 Catalogue bibliogra-phy and generic revision of the order Schizomida (Arach-nida) Texas Memorial Museum Speleological Monographs4 1ndash170

Reissland A Goumlrner P 1985 Trichobothria In Barth FGed Neurobiology of arachnids Berlin Springer-Verlag 138ndash161

Roewer C-F 1934 Solifugae Palpigradi Bronns Klassen undOrdnungen ses Tierreich 5 IV 4 1ndash713

Rowland JM Cooke JAL 1973 Systematics of the arachnidorder Uropygida (=Thelyphonida) Journal of Arachnology 155ndash71

Rowland JM Sissom WD 1980 Report on a fossil palpi-grade from the Tertiary of Arizona and a review of themorphology and systematics of the order (ArachnidaPalpigradida) Journal of Arachnology 8 69ndash86

Rucker A 1901 The Texan Koenenia American Naturalist35 615ndash630

Ruhland M Rathmayer W 1978 Die Beinmuskulatur undihre Innervation bei der Vogelspinne Dugesiella henzti (Ch)(Araneae Aviculariidae) Zoomorphologie 89 33ndash46

Saboori A Kamali K 2000 Description of Allothrombiumtriticium adult (Acari Trombidiidae) from Iran Systematicand Applied Acarology 5 207ndash208

Schliwa M 1979 The retina of the phalangid Opilio raven-nae with particular reference to arhabdomeric cells Celland Tissue Research 204 473ndash495

Scholl G 1977 Beitrage zur Embryonalentwicklung von Lim-ulus polyphemus L (Chelicerata Xiphosura) Zoomorpholo-gie 86 99ndash154

Scholtz G Kamenz C 2006 The book lungs of Scorpionesand Tetrapulmonata (Chelicerata Arachnida) evidence forhomology and a single terrestrialization event of a commonarachnid ancestor Zoology 109 2ndash13

Seitz KA 1972 Zur Histologie und Feinstruktur des Herzensund der Haumlmocyten von Cupiennius salei Keys (AraneaeCtenidae) I Herzwandung Bildung und Differenzierungder Haumlmocyten Zoologische Jahrbuumlcher Abteilung fuumlr Anat-omie und Ontogenie der Tiere 89 351ndash384

Sekiguchi K ed 1988 Biology of horseshoe crabs Tokyo Sci-ence House

Selden PA 1981 Functional morphology of the prosoma ofBaltoeurypterus tetragonophthalmus (Fischer) (ChelicerataEurypterida) Transactions of the Royal Society of Edin-burgh Earth Sciences 72 9ndash48

Selden P 1992 Revision of the fossil ricinuleids Transactionsof the Royal Society of Edinburgh Earth Sciences 83 595ndash634

Selden PA 1993 Fossil arachnids ndash recent advances andfuture prospects Memoirs of the Queensland Museum 33389ndash400

Selden PA Dunlop JA 1998 Fossil taxa and relationshipsof chelicerates In Edgecombe GD ed Arthropod fossilsand phylogeny New York Columbia University Press 303ndash331

Selden PA Jeram AJ 1989 Palaeophysiology of terrestrial-

isation in the Chelicerata Transactions of the Royal Societyof Edinburgh Earth Sciences 80 303ndash310

Selden PA Shear WA Bonamo PM 1991 A spider and otherarachnids from the Devonian of New York and reinterpre-tations of Devonian Araneae Palaeontology 34 241ndash281

Sensenig A Shultz JW 2003 Mechanics of cuticular elasticenergy storage in leg joints lacking extensor muscles inarachnids Journal of Experimental Biology 206 771ndash784

Shear WA 1993 New species in the opilionid genus Stylocel-lus from Malaysia Indonesia and the Philippines (OpilionesCyphophthalmi Stylocellidae) Bulletin of the British Arach-nological Society 9 174ndash188

Shear WA Selden PA Rolfe WDI Bonamo PM GriersonJD 1987 New terrestrial arachnids from the Devonian ofGilboa New York (Arachnida Trigonotarbida) AmericanMuseum Novitates 2901 1ndash74

Shultz JW 1987 The origin of the spinning apparatus in spi-ders Biological Reviews of the Cambridge PhilosophicalSociety 62 89ndash113

Shultz JW 1989 Morphology of locomotor appendages inArachnida evolutionary trends and phylogenetic implica-tions Zoological Journal of the Linnean Society 97 1ndash56

Shultz JW 1990 Evolutionary morphology and phylogeny ofArachnida Cladistics 6 1ndash31

Shultz JW 1991 Evolution of locomotion in Arachnida thehydraulic pressure pump of the giant whipscorpion Masti-goproctus giganteus (Uropygi) Journal of Morphology 21013ndash31

Shultz JW 1993 Muscular anatomy of the giant whipscor-pion Mastigoproctus giganteus (Arachnida Uropygi) and itsevolutionary significance Zoological Journal of the LinneanSociety 108 335ndash365

Shultz JW 1998 Phylogeny of Opiliones (Arachnida) anassessment of the lsquoCyphopalpatoresrsquo concept Journal ofArachnology 26 257ndash272

Shultz JW 1999 Muscular anatomy of a whipspider Phrynuslongipes (Amblypygi) and its evolutionary significance Zoo-logical Journal of the Linnean Society 126 81ndash116

Shultz JW 2000 Skeletomuscular anatomy of the harvest-man Leiobunum aldrichi (Weed 1893) (Arachnida Opil-iones Palpatores) and its evolutionary significanceZoological Journal of the Linnean Society 128 401ndash438

Shultz JW 2001 Gross muscular anatomy of Limuluspolyphemus (Chelicerata Xiphosura) and its bearing on evo-lution in the Arachnida Journal of Arachnology 29 283ndash303

Shultz JW 2007 Morphology of the prosomal exoskeleton ofScorpiones (Arachnida) and a new hypothesis for the evolu-tion of cuticular cephalic endoskeletons in arthropodsArthropod Structure amp Development 36 77ndash102

Shultz JW Pinto da Rocha R 2007 Morphology and func-tional anatomy In Pinto da Rocha R Machado G Giribet Geds The harvestmen the biology of Opiliones CambridgeMA Harvard University Press 14ndash61

Shultz JW Regier JC 2001 Phylogenetic analysis of Pha-langida (Arachnida Opiliones) using two nuclear protein-encoding genes supports monophyly of Palpatores Journalof Arachnology 29 189ndash200

Sitnikova LG 1978 The main evolutionary trends of the

244 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Acari and the problem of their monophyletism Entomologis-cheskoe Obozrenie 57 431ndash457

Snodgrass RE 1948 The feeding organs of Arachnidaincluding mites and ticks Smithsonian Miscellaneous Col-lections 110 (10) 1ndash94

Snodgrass RE 1952 A textbook of arthropod anatomy Ith-aca NY Comstock Publishing

Soslashrensen W 1879 Om Bygningen af Gonyleptiderne en Typeaf Arachnidernes Classe Naturhistorisk Tidsskrifter Series3 8 97ndash222

Stockwell SA 1989 Revision of the phylogeny and higherclassification of scorpions (Chelicerata) Unpublished doc-toral dissertation University of California at Berkeley

Stoslashrmer L 1944 On the relationships and phylogeny of fossiland Recent Arachnomorpha Skrifter utgitt av Det NorskeVidenskaps-Akademi i Oslo I 5 1ndash158

Stoslashrmer L 1955 Merostomata In Moore RC ed Treatise oninvertebrate paleontology Part P Arthropoda 2 LawrenceKS University of Kansas Press and Geological Society ofAmerica 4ndash41

Swofford DL 2002 PAUP Phylogenetic analysis using par-simony (and other methods) Version Beta 10 SunderlandMA Sinauer Associates

Talarico G Palacios-Vargas JG Fuentes Silva M AlbertiG 2005 First ultrastructural observations of the tarsal poreorgan of Pseudocellus pearsei and P boneti (ArachnidaRicinulei) Journal of Arachnology 33 604ndash612

Templeton AR 1983 Phylogenetic inference from restrictionendonuclease cleavage site maps with articular reference tothe evolution of humans and the apes Evolution 37 221ndash244

Tetlie OE Braddy SJ 2004 The first Silurian chasmataspidLoganamaraspis dunlopi gen et sp nov (Chelicerata Chas-mataspidida) from Lesmahagow Scotland and its implica-tions for eurypterid phylogeny Transactions of the RoyalSociety of Edinburgh Earth Sciences 94 227ndash234

Thomas RH Telford MJ 1999 Appendage development inembryos of the oribatid mite Archegozetes longisetosus(Acari Oribatei Trhypochthoniidae) Acta Zoologica (Stock-holm) 80 193ndash200

Tollerton VP Jr 1989 Morphology taxonomy and classifica-tion of the order Eurypterida Burmeister 1843 Journal ofPaleontology 63 642ndash657

Vachon M 1949 Ordre des Pseudoscorpions In Grasseacute P-Ped Traite de zoologie VI Paris Masson et Cie 436ndash481

Van der Hammen L 1979 Comparative studies in Chelicer-ata I The Cryptognomae (Ricinulei Architarbi and Anacti-notrichida) Zoologische Verhandelingen 174 1ndash62

Van der Hammen L 1982 Comparative studies in Chelicer-ata II Epimerata (Palpigradi and Actinotrichida) Zoolo-gische Verhandelingen 196 1ndash70

Van der Hammen L 1985 Functional morphology and affin-ities of extant Chelicerata in evolutionary perspectiveTransactions of the Royal Society of Edinburgh 76 137ndash146

Van der Hammen L 1989 An introduction to comparativearachnology Leiden SPB Academic Publishing

Vyas AB 1970 Studies in myology of the scorpion Heter-ometrus fulvipes Koch Unpublished PhD DissertationDepartment of Zoology Gujarat University India

Vyas AB 1974 The cheliceral muscles of the scorpion Heter-ometrus fulvipes Bulletin Southern California Academy ofSciences 73 9ndash14

Walter DE 1988 Predation and mycophagy by endeostigma-tid mites (Acariformes Prostigmata) Experimental andApplied Acarology 4 159ndash166

Walter DE Proctor HC 1998 Feeding behaviour and phy-logeny observations on early derivative Acari Experimentaland Applied Acarology 22 39ndash50

Walter DE Proctor HC 1999 Mites ecology evolution andbehaviour Sydney University of New South Wales

Waterston CD 1979 Problems of functional morphology andclassification in stylonuroid eurypterids (Chelierata Meros-tomata) with observations on the Scottish Silurian Sty-lonuroidea Transactions of the Royal Society of EdinburghEarth Science 70 251ndash322

Weygoldt P 1969 Biology of pseudoscorpions CambridgeMA Harvard University Press

Weygoldt P 1984 Lrsquoautotomie chez les Amblypyges RevueArachnologique 5 321ndash327

Weygoldt P 1996 Evolutionary morphology of whip spiderstowards a phylogenetic system (Chelicerata ArachnidaAmblypygi) Journal of Zoological Systematics and Olution-ary Research 34 185ndash202

Weygoldt P 1998 Evolution and systematics of the Chelicer-ata Experimental and Applied Acarology 22 63ndash79

Weygoldt P 2000 Whip spiders (Chelicerata Amblypygi)their biology morphology and systematics StenstrupDenmark Apollo Books

Weygoldt P Paulus HF 1979 Untersuchungen zur Morphol-ogie Taxonomie und Phylogenie der Chelicerata Zeitschriftfuumlr Zoologische Systematik und Evolutionsforschung 17 85ndash116 177ndash200

Wheeler WC Hayashi CY 1998 The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192

Winkler D 1957 Die Entwicklung der aumlusseren Koumlrperge-stalt bei den Phalangiidae (Opiliones) Mitteilungen aus demZoologischen Museum in Berlin 33 355ndash389

With CJ 1904 The Notostigmata a new suborder of AcariVidenskabelige Meddelelser fra Dansk Naturhistorisk Foren-ingi Kjobenhaus 1904 137ndash192

Yamasaki T Makioka T Saito J 1988 Morphology InSekiguchi K ed Biology of horseshoe crabs Tokyo ScienceHouse 69ndash132

Yoshikura M 1975 Comparative embryology and phylogeny ofArachnida Kumamoto Journal of Science Biology 12 71ndash142

APPENDIX

Each character used in this analysis is numbereddefined and cross-referenced with characters used infour previous studies that are abbreviated in bracketsspecifically WP = Weygoldt amp Paulus (1979)S = Shultz (1990) WH = Wheeler amp Hayashi (1998)GEWB = Giribet et al (2002) Descriptions includecorrections of errors discussions of controversies orambiguities and justifications for state assignmentsto lsquoproblemrsquo taxa

PHYLOGENY OF THE ARACHNID ORDERS 245

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

PROSOMA

1 Dorsal sclerite formed by fusion of the prosomal car-apace the dorsal portion of the first opisthosomalsomite and the dorsomedial (axial) portion of the sec-ond opisthosomal somite 0 absent 1 presentState 1 is present in extant xiphosurids and close fos-sil relatives (Scholl 1977 Anderson amp Selden 1997Shultz 2001) The fossil record suggests that thisstructure originated with the disappearance ofthe already reduced first opisthosomal tergite(= microtergite) and axial portion of the secondopisthosomal tergite (Anderson amp Selden 1997)2 Single dorsal sclerite covering entire dorsal surfaceof body no lines indicating original segmentation 0absent 1 presentState 1 occurs throughout Holothyrida (Acari) (Vander Hammen 1989) and in many Mesostigmata(Alberti amp Coons 1999) including Glyptholaspis (Vander Hammen 1989)3 Anterior end of dorsal prosoma with medianmarginal or submarginal pointed process 0 absent1 presentState 1 occurs throughout Schizomida (Protoschizo-mus Stenochrus Reddell amp Cokendolpher 1995) The-lyphonida (Rowland amp Cooke 1973 Shultz unpublobs Proschizomus Dunlop amp Horrocks 19951996)Amblypygi (Charinus Millot 1949b Phrynus Shultz1999) and Plesiosiro (Dunlop 1999) Citing a figure inRowland amp Cooke (1973) Dunlop amp Horrocks assumedthat the anterior process was limited to lsquohypoctonidrsquothelyphonids and united them with Proschizomus andSchizomida If valid this would make Thelyphonidaparaphyletic However the doubtful monophyly oflsquohypoctonidsrsquo (Haupt amp Song 1996) and widespreadpresence of a submarginal anterior process in Pedi-palpi weaken this proposal (Harvey 2002)4 Ophthalmic ridges pair of longitudinal crests inter-secting or passing near the region of the lateral eyes orcomparable region where lateral eyes are absent 0absent 1 presentState 1 is present throughout Xiphosura (Anderson ampSelden 1997) Similar structures are present in Plesi-osiro (Dunlop 1999) and lsquonon-hypoctonidrsquo Thely-phonida (Mastigoproctus Rowland amp Cooke 1973)Dunlop (2002a) described apparent ophthalmic ridgesin the chasmataspidid Octoberaspis but these werenot observed in Diploaspis (Dunlop et al 2001) eitherdue to absence or inadequate preservation of speci-mens nor in Chasmataspis (Dunlop et al 2004)5 Carapace with demarcations (eg grooves scleritesphragmata) between pro- meso- or metapeltidia 0absent 1 present [S 1 WH 48 GEWB 29]State 1 is present in extant Scorpiones Cyphoph-thalmi (Hansen amp Soslashrensen 1904) and many Palpa-tores (Opiliones) including Caddo and Leiobunum

and most Pseudoscorpiones (Chamberlin 1931) Giri-bet et al (2002) coded this character as absent inCyphophthalmi (Opiliones) but a transverseprocurved groove is present throughout the group(Hansen amp Soslashrensen 1904) and represents the meso-metapeltidial border as indicated by attachment ofthe dorsal endosternal suspensor of postoral somite VIin Chileogovea oedipus and Siro exilis (Shultz unpublobserv) Kjellesvig-Waering (1986) did not explicitlydescribe such grooves in fossil scorpions although heillustrated corresponding structures in PrearcturusArchaeophonus Archaeoctonus Palaeophonus andother Palaeozoic scorpions (Kjellesvig-Waering 1986figs 14A 21E 27D 59) (coded as uncertain here) butnot in Proscorpius (contra Giribet et al 2002) Dis-tinct carapacal sclerites are present throughout Palpi-gradi Schizomida and Solifugae (6)

The condition in Acari is controversial but is codedhere as State 1 Van der Hammen (1989) consideredthe tergal region of the leg-bearing somites (postoralsomites IIIndashVI) or podosoma to be reduced andreplaced by posterior migration of the aspidosoma (ietergal region assumed to be associated with append-ages of the gnathosoma) and anterior migration of theopisthosomal tergal region to form the hysterosomaThe dorsal proterosomal and hysterosomal elementsmeet at a transverse sejugal furrow that continues lat-erally and passes ventrally between the coxae of legs 2and 3 According to this scheme the prosomandashopisthosoma border is expressed as the disjugal fur-row which passes from the ventral posterior margin ofthe podosoma anterodorsally to join the dorsal part ofthe sejugal furrow Based on the arrangement of setaeand slit sensilla (= lyrifissures) Van der Hammen(1989) interpreted the region above coxae 3 and 4 (ieregion C) as a fusion of the dorsal parts of the first twoopisthosomal somites (= postoral somites VII and VIII)in early divergent Anactinotrichida (ie Opilioacari-formes) and Acariformes (eg Alycus) These interpre-tations have been followed by many acarologistsalthough its speculative aspects are widely acknowl-edged (Evans 1992 Alberti amp Coons 1999)

From the standpoint of a general arachnologist theVan der Hammen system seems unnecessarily compli-cated it appears to have been formulated to explainbroad morphological themes in Oribatida but was thenextrapolated to other mites A more conservativelsquoarachnologicalrsquo scheme adopted here equates the pro-podosoma with the pars cephalica (= propeltidium) ofthe arachnid prosoma and region C with the pars tho-racica (= mesopeltidium + metapetidium) In fact theearliest and most recent treatments of Opilioacari-formes (With 1904 Klompen 2000) reached similarconclusions This interpretation is also consistent withevidence from Acariformes For example the regiondorsal to the coxa of legs 3 and 4 in Alycus is as readily

246 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

explained by the persistence of primitive prosomal ter-gal elements as by the supposition that these disap-peared and were replaced by two opisthosomalelements Furthermore several diverse lineages ofbasally divergent Acariformes display features that donot fit readily within the GrandjeanndashVan der Ham-men system but seem to correspond to the cephalicathoracica division of the prosoma They have a sejugalfurrow that may have demarcated the cephalicatho-racica border and a postpodosomal furrow thatappears to represent the prosomandashopisthosoma junc-tion there is no disjugal furrow Examples includepalaeosomate (especially aphelacarid) and pediculoch-elid Oribatida alicorhagiid and mircopsammidlsquoEndeostigmatarsquo and paratydeid Prostigmata (Albertiamp Coons 1999 fig 42) This organization may be sym-plesiomorphic for Acariformes but could be a conver-gence brought about by selection for enhancedflexibility6 Carapace with distinct pro- meso- or metapeltidialsclerites 0 absent 1 present - inapplicable due toabsence of pro- meso- or metapeltidial demarcations(5)State 1 occurs throughout Palpigradi Schizomida andSolifugae It also appears sporadically within Opil-iones (eg Leiobunum flavum Shultz unpubl observ)and Pseudoscorpiones (eg Pseudochiridium) (Cham-berlin 1931)7 Sejugal furrow circumferential zone of body flexi-bility that passes between the coxae of legs 2 and 3 0absent 1 presentState 1 is probably synapomorphic for Acariformes asit occurs in most lsquoendeostigmatidsrsquo including Alycus(Van der Hammen 1989) and is variously developedthroughout Sarcoptiformes (Alberti amp Coons 1999) Itis probably the primitive condition for Prostigmatabut is either absent or weakly expressed in represen-tatives included here See 5 for alternative morpholog-ical interpretations of this body region in Acari8 Prosomal ozopores 0 absent 1 present [WH 46GEWB 12]State 1 occurs throughout Opiliones and Holothyrida(Van der Hammen 1989)9 Carapacal pleural doublure 0 absent 1 present[WP 12 S 2 WH 49 GEWB 27]A carapacal pleural doublure occurs throughoutXiphosura (Stoslashrmer 1944) Eurypterida (Clarke ampRuedemann 1912 Selden 1981) and Chasmataspid-ida (Dunlop et al 2001 2004 Dunlop 2002a) and isabsent in all known Arachnida10 Cardiac lobe a longitudinal axial elevation of thecarapace 0 absent 1 presentState 1 occurs throughout Xiphosura (Anderson ampSelden 1997) and Eurypterida (Clarke amp Ruedemann1912 Selden 1981) Carapaces are not preserved suf-ficiently to determine this condition in the chasmatas-

pidids Diploaspis (Dunlop et al 2001) andOctoberaspis (Dunlop 2002a) but an apparent cardiaclobe occurs in the Chasmataspis (Dunlop et al 2004)A cardiac lobe is absent in Arachnida11 Moveable cucullus 0 absent 1 present [WP 54GEWB 22]State 1 is a synapomorphy of Ricinulei (Selden 1992)12 Medial intercoxal lsquosternalrsquo region 0 all pedalcoxae separated medially 1 anterior pedal coxae abut-ting medially posterior coxae separated 2 anteriorpedal coxae separated medially posterior coxae abut-ting 3 all pedal coxae abutting medially 4 epimeracoxae undifferentiated medially from ventral body wall[S sim3 WH sim50 GEWB sim17]Prosomal sternites appear to be those portions of theventral body wall not occupied by coxae Despite earlyattempts to assess metamerism of the ventralprosoma (protosternum deutosternum tritosternumetc) (Boumlrner 1902a Millot 1949a) no apparent mor-phological features such as borders of sclerites ormuscle attachments reliably demarcate the ventralbody wall of one somite from that of an adjacentsomite Thus coding schemes that focus on describingmetameric components of the sternal region are prob-lematic Shultz (1990) attempted to code the shape ofthe entire intercoxal region and subsequent workershave adopted this approach All pedal coxae are sepa-rated medially in Plesiosiro (Dunlop 1999) Palpigradi(Roewer 1934) Araneae (Millot 1949c) Amblypygi(Weygoldt 2000) and many Acari including Opilioac-ariformes Holothryrida Mesostigmata Ixodida Aly-cus and many Prostigmata (Van der Hammen 1989Evans 1992 Alberti amp Coons 1999) The anteriorcoxae are separated in extant Thelyphonida Schizo-mida (Hansen amp Soslashrensen 1905 Millot 1949e) andRicinulei (Pittard amp Mitchell 1972 Legg 1976) Thecoxae are fused to the ventral body wall in Sarcopti-formes (Van der Hammen 1989 Evans 1992)

Interpreting this character is complicated for theanterior coxae of Scorpiones and Opiliones and allcoxae in Xiphosura and Eurypterida due to the pres-ence of coxapophyses endites andor gnathobaseswhich form components of the preoral chambers inthese groups Thus coxae may abut medially whenthese structures are considered part of the coxae butare separate medially if these structures are ignoredAs coxapophyses etc are here coded as separate char-acters (50ndash53) I have chosen to code this character asif these structures did not exist Consequently pha-langid Opiliones are coded as having all coxae sepa-rated (Pocock 1902 Hansen amp Soslashrensen 1904) Fossilscorpions also show substantial variation in this char-acter All pairs of pedal coxae in Palaeoscorpius abutmedially but also have gnathobases (Jeram 1998)and this condition is coded here as uncertain A ster-num separates all pairs of pedal coxae medially in

PHYLOGENY OF THE ARACHNID ORDERS

247

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Proscorpius

and

Stoermeroscorpio

(State 0) and

Pre-arcturus

approximates modern scorpions (State 1)(Kjellesvig-Waering 1986 Jeram 1998) The coxalbases of Eurypterida and Xiphosura are separated bya small sternite the endostoma (Selden 1981 Shultz2001)

13 Postoral sternapophysis (

=

tritosternum labium)a cuticular evagination of the ventral body wall poste-riorly adjacent to the palpal coxae forming the poste-rior border of the preoral chamber in some taxa ordisplaced posteriorly by fusion of the palpal coxae inothers 0 absent 1 present

[S

sim

10 WH

sim

56 GEWB19

=

33]State 1 is present throughout Palpigradi AraneaeAmblypygi (Snodgrass 1948) Schizomida (Hansen ampSoslashrensen 1905 Van der Hammen 1989) Thely-phonida (Van der Hammen 1989) extant Ricinulei(Pittard amp Mitchell 1972 Legg 1976) Opilioacari-formes and Mesostigmata (Van der Hammen 1989Alberti amp Coons 1999)

Allothyrus constrictus

is theonly holothyrid mite known to have sternapophysesAn apparent sternapophysis has been described in apalaeocharinid trigonotarbid (Dunlop 1994) Giribet

et al

(2002) miscoded Ixodida and Acariformes as hav-ing a lsquolabium (

=

tritosternum)rsquo The lophognath ofpseudoscorpions (Chamberlin 1931) and lsquolabiumrsquo ofSolifugae (Roewer 1934) also appear to be sternapo-physes that have been incorporated into the ros-trosoma (

32

) The labium of phalangid Opiliones(Pocock 1902 Hansen amp Soslashrensen 1904) is a ster-napophysis but is associated with leg 1 rather thepedipalp (Winkler 1957) A small anterior sclerite inthe fossil scorpions

Proscorpius

Waeringoscorpio

and

Labriscorpio

has been interpreted as a lsquolabiumrsquo homol-ogous with the sternapophyses of other arachnids(Weygoldt 1998) despite absence of any indication ofthe relationship of this sclerite to the mouth and pos-sible alternative interpretations (eg

16

) Giribet

et al

coded this sclerite as a sternapophysis but the char-acter is coded here as uncertain in fossil scorpions

14 Channels on the body surface linking openings ofcoxal organs to preoral chamber 0 absent 1 present

Van der Hammen (1989) described bilaterally pairedcuticular tracts (lsquotaenidiarsquo) connecting orifices of coxalorgans to the preoral chamber in some arachnids andor a ventromedian groove (

=

intercoxal or subcapitulargutter) that leads in turn to the preoral chamber inother taxa Taenidia and intercoxal gutters arepresent in Opilioacariformes (Van der Hammen1989) Holothyrida (Van der Hammen 1989) Mesos-tigmata (

Glyptholapis

Van der Hammen 1989)Ricinulei (Pittard amp Mitchell 1972 Van der Hammen1989) Araneae (

Heptathela

Aphonopelma

Hypochi-lus

Shultz unpubl observ also

Porrhothele

(Mygalo-morphae Dipluridae) Butt amp Taylor 1991

Segestria

(Araneomorphae Segestriidae) Van der Hammen

1989) Amblypygi (Van der Hammen 1989) Schizo-mida (

Stenochrus

Shultz unpubl observ unidenti-fied Hubbardiinae Van der Hammen 1989) andThelyphonida (

Mastigoproctus

Shultz unpublobserv also

Tetrabalius

Van der Hammen 1989) Theconduction of large volumes of fluid from orifices of thecoxal organs to the preoral chamber has been observedin the mygalomorph spider

Porrhothele

(Dipluridae)(Butt amp Taylor 1991) and

Mastigoproctus

(Thely-phonida) (Shultz unpubl observ) The podocephaliccanal (

15

) is present in Acariformes (Acari) (Lindquist1984) and appears to have evolved as an invaginatedsupracoxal channel (Grandjean 1938 G Alberti inEvans 1992) Van der Hammen (1982) could not dis-cern with certainty whether a channel exists in

Eukoenenia

(Palpigradi) but none has been reportedby previous authors and they have not been observedby the present author Palpigradi is coded as havingState 0

15 Podocephalic canal cuticular channel andorduct draining multiple glands and opening nearmouthparts 0 absent 1 present - inapplicable due toabsence of channel (14)

State 1 occurs throughout Acariformes (Acari)(Lindquist 1984)

16 Heavily sclerotized suboral sclerite serving inpart as basal pivot point for coxae of appendages ofpostoral somites IIndashIV (

=

arachnid palp and legs 1 and2) 0 absent 1 present

The sclerite forms the posterior wall of the true mouthand is heavily sclerotized in extant scorpions It is notvisible externally due to the tightly fitting coxae of theanterior prosomal appendages (Shultz 2007) It maycorrespond to the so-called labium (Weygoldt 1998) orlabrum (Kjellesvig-Waering 1986) of certain fossilscorpions (see

13

)

17 Genal angles 0 rounded 1 pointed

Anderson amp Selden (1997) originally coded a pointedgenal angle and genal spine as separate charactersbut they are combined here State 1 is present in manyXiphosura (Anderson amp Selden 1997) and Chas-mataspidida (Dunlop

et al

2001 2004 Dunlop2002a)

A

PPENDAGES

OF

POSTORAL

SOMITE

I

CHELICERAE

18 Cheliceral segmentation 0 three articles 1 twoarticles

[WP

sim

25

+

42 WH

sim

15 GEWB

sim

44]State 0 is present in extant Xiphosura Eurypterida(Clarke amp Ruedemann 1912 Selden 1981) Palpi-gradi (Roewer 1934) Opiliones (Hansen amp Soslashrensen1904) Scorpiones (Hjelle 1990) and throughout non-acariform Acari (Evans 1992) State 1 is presentthroughout known Trigonotarbida Araneae Ambly-pygi Thelyphonida Schizomida Pseudoscorpionesand Solifugae Acariform mites have two unambigu-

248 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ous cheliceral articles but may have a reduced thirdarticle (Van der Hammen 1989 Evans 1992) acari-form mites are thus coded as uncertain Chelicerae arenot known or their condition is unclear in several fos-sil taxa including Xiphosura Chasmataspidida (Dun-lop et al 2001 2004 Dunlop 2002a) Riciniulei(Selden 1992) Plesiosiro (Dunlop 1999) Proschizo-mus (Dunlop amp Horrocks 19951996) and the fossilscorpions Praearcturus and Palaeoscorpius (Kjelles-vig-Waering 1986) Kjellesvig-Waering (1986) consis-tently interpreted chelicerae of fossil scorpions ashaving four articles but this is probably a misinter-pretation of the basal article (see also Stockwell1989) The basal article in extant scorpions consists ofa distal collar of cuticle and a large proximal processand this arrangement would probably appear as twoarticles in compressed specimens In any event thereare at least three cheliceral articles in Stoermeroscor-pio and Proscorpius (Kjellesvig-Waering 1986)19 Terminal cheliceral joint 0 laterally placedbicondylar hinge 1 dorsally placed bicondylar hinge2 ventrally placed bicondylar hinge [WP 25 + 42 Ssim11 + 12 WH sim15 GEWB sim44]State 0 is present in Xiphosura (Shultz 2001)extant Scorpiones (Millot amp Vachon 1949) Opil-iones (Hansen amp Soslashrensen 1904) Palpigradi(Boumlrner 1904 Millot 1949d) Ricinulei (Pittard ampMitchell 1972 Legg 1976) and Eurypterida includ-ing Baltoeurypterus (Selden 1981) and Stylonurus(Clarke amp Ruedemann 1912) State 0 is present inthe fossil scorpions Proscorpius and Stoermeroscor-pio but the state of the chelicerae is unclear in theother fossil scorpions represented here (Kjellesvig-Waering 1986) State 1 is present in Trigonotarbida(Shear et al 1987) mygalomorph Araneae Ambly-pygi Schizomida and Theylphonida (Weygoldt ampPaulus 1979) State 2 is present in Pseudoscorpi-ones (Chamberlin 1931) and Solifugae (Roewer1934) and appears to be a groundplan feature for allmajor groups of mites (eg Van der Hammen 1989Alberti amp Coons 1999) The difference between theorthognathy of mygalomorph Araneae and lsquoplagiog-nathyrsquo mesothele (Liphistius Heptathela) and palae-ocribellate Araneae (Hypochilus) (Kraus amp Kraus1993) is small compared with the variation thatoccurs throughout arachnids and plagiognathy iscoded as State 1 here20 Chelicera articulating with carapace at anterolat-eral pivot 0 absent 1 present [S 13 WH 57 GEWB45]State 1 occurs in Solifugae (Roewer 1934) and Pseu-doscorpiones except Chthonioidea and Feaelloidea(Chamberlin 1931) Giribet et al (2002) coded thecharacter as present in Pseudoscorpiones21 Chelicera pivoting on dorsal protuberance of epis-tome 0 absent 1 present

State 1 appears to be a unique synapomorphy of Opil-iones (Shultz 2000)22 Extrinsic cheliceral muscle arising on carapaceand inserting on dorsal margin of nonbasal cheliceralarticle 0 absent 1 present (Note Characters 22 and23 refer to two different muscles)State 1 is known only in extant Scorpiones (LankesterBenham amp Beck 1885 Vyas 1970 1974 Shultz2007) but may occur in acariform mites if their cheli-cerae are composed of three rather than two articles(Evans 1992) (see 18) Acariformes are coded asuncertain23 Extrinsic cheliceral muscle arising on carapaceand inserting on ventral margin of nonbasal cheliceralarticle 0 absent 1 present [GEWB 252] (Note Char-acters 22 and 23 refer to two different muscles)State 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) somemesostigmatid Acari (Evans 1992) and representa-tive Opiliones particularly Leiobunum (Shultz 2000)Chileogovea and Gonyleptes (Shultz unpubl observ)The status of the character in other opilions is not yetknown The character was originally defined as anextrinsic muscle inserting on the second segment ofthree-segmented chelicerae (Shultz 2000 Giribetet al 2002) but it has been redefined here to make itapplicable to arachnids with two-segmented cheli-cerae24 Extrinsic cheliceral muscles attaching to epistome0 absent 1 presentState 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) but mayoccur in Prostigmata if the sigmoid piece to which thecheliceral protractors attach (Evans 1992) is a modi-fication of the epistome25 Lateral tergocheliceral muscle with three heads 0absent 1 present - coded only for extant tetrapulmo-nates and the palpigrade Eukoenenia homology isunclear in other taxa [GEWB 230]State 1 is present in Pedipalpi (Shultz 1999)26 Cheliceral silk glands 0 absent 1 present [GEWB47]State 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931) The silk gland of spider mites(eg Tetranychus) are associated with the pedipalps(Evans 1992)27 Cheliceral venom glands 0 absent 1 present [WP34 WH 40 GEWB 46]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)28 Cheliceral serrula interior and exterior 0 absent1 presentState 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931 Harvey 1992)29 Cheliceral lsquoflagellumrsquo in male 0 absent 1 present[GEWB 48]

PHYLOGENY OF THE ARACHNID ORDERS 249

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITE II LEG 1 IN NONARACHNIDS AND PALP IN ARACHNIDS

30 Palpal coxae fused ventromedially and formingposterior wall of preoral chamber 0 absent 1 present[WP 27 S 18 WH 17 = 62 GEWB 33 + 63]State 1 occurs in Schizomida Thelyphonida Acari andRicinulei31 Gnathosoma 0 absent 1 present - inapplicablepalpal coxae not fused (30) [WP 56 S 19 WH sim63GEWB sim36 + 37]The gnathosoma is a functional complex comprisingan epistome attached to the dorsal surface of mediallyfused palpal coxae (= infra- or subcapitulum) (see 30)the chelicerae and in some taxa a supracheliceral tec-tum These components are tightly integrated andmove in unison relative to the body The gnathosomahas traditionally been considered a feature of AcariHowever Van der Hammen (1989) argued that a gna-thosoma is present in Ricinulei and this interpreta-tion has received wide acceptance (eg Lindquist1984) However it is not clear that the relationship ofthe chelicerae and subcapitulum is as intimate inRicinulei as in Acari and the ricinuleid subcapitulumdoes not appear to be as moveable In fact the lsquosubca-pitulumrsquo in Ricinulei is similar in many respects to thecondition in Schizomida and Thelyphonida (Hansen ampSoslashrensen 1904) which has never been regarded as agnathosoma Consequently I have coded the conditionof Ricinulei as uncertain32 Rostrosoma long narrow subcylindrical epistomeprojecting anteriorly with base fixed to dorsal surfaceof palpal coxae bordered laterally by lobes projectingfrom palpal coxae ventral wall of preoral chamberformed by anterior element of prosoma (sternapophy-sis) 0 absent 1 present [S 15 WH 59 GEWB 20 + 36]State 1 is present in Pseudoscorpiones and Solifugae(Chamberlin 1931 Roewer 1934 Van der Hammen1989) Dunlop (2000) attempted to homologize com-ponents of the rostrosoma and gnathosoma (31) anargument accepted by Giribet et al (2002) Howeverfusion of the palpal coxae (30) apparently evolvedindependently in the two structures Coxal fusion inthe gnathosoma is complete ventrally and excludesthe suboral sternapophysis (13) while fusion isincomplete ventrally in the rostrosoma and incorpo-rates the sternapophysis Unlike the gnathosomathe rostrosoma is largely immobile and cheliceralmovement is not coupled with that of the palpalcoxae33 Rutellacorniculi hypertrophied setae modified asmouthparts located on the anterior processes of the pal-pal coxae 0 absent 1 present [GEWB 39]

State 1 occurs in the anactinotrichid groups Opilioac-ariformes Holothyrida and Mesostigmata but not Ixo-dida (Van der Hammen 1989) It also occurs inSarcoptiformes (Evans 1992) and certain endeostig-matids (eg Alycus Van der Hammen 1989) but not inProstigmata Comparable structures are unknown inother chelicerate groups34 Terminal segments specially modified in adultmale as a clasper used to engage the female 0 absent1 presentState 1 is known only in extant Xiphosura (YamasakiMakioka amp Saito 1988)35 Robust raptorial 0 absent 1 present [S 17 WH61 GEWB sim59 97]State 1 is present in Amblypygi Thelyphonida andSchizomida (Shear et al 1987) and in Laniatores(Opiliones)36 Orientation of robust raptorial appendage 0operating in subtransverse plane 1 operating in sub-vertical plane - inapplicable coded only for PedipalpiState 0 is present in extant Thelyphonida and mostAmblypygi and State 1 is present in Schizomida (Cok-endolpher amp Reddell 1992) and in the basally diver-gent amblypygid Paracharon caecus (Weygoldt 2000)The orientation of the pedipalps in the fossil Proschi-zomus is uncertain (Dunlop amp Horrocks 19951996)37 Extrinsic muscle attaching to epistome 0 absent1 presentState 1 occurs in Chileogovea and Siro (OpilionesCyphophthalmi) and extant Scorpiones (Vyas 1970Shultz 2007)38 Muscle originating and inserting within coxa 0absent 1 present [GEWB sim232]State 1 occurs is in Amblypygi (Phrynus Shultz 1999Phrynichus Boumlrner 1904) Thelyphonida (Mastigo-proctus Shultz 1993 Thelyphonus Boumlrner 1904) andSchizomida (Stenochrus Shultz unpubl observ)There are two intracoxal muscles in amblypygids andthelyphonids the one coded here and another that isserially homologous with a muscle in the pedal coxae(54)39 Tarsus andor tibia with venom glands 0 absent1 present [GEWB 64]State 1 is known only from extant Pseudoscorpi-ones except the superfamilies Chthonioidea andFeaelloidea (Chamberlin 1931 Harvey 1992) Giri-bet et al (2002) coded this character as present forPseudoscorpiones40 Tarsal grooming organ 0 absent 1 present[GEWB 98]State 1 is a unique synapomorphy of Amblypygi (DelleCave 1975 Weygoldt 2000)41 Scorpionoid chela a large well-developed chelaformed by tibia (manus + fixed finger) and tarsus(moveable finger) 0 absent 1 present [S 16 WH 60GEWB 62]

250 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs throughout Scorpiones andPseudoscorpiones42 Modified in male as copulatory organ 0 absent 1present [WP 35 WH 39 GEWB 67]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)43 Apotele 0 apparently absent not differentiatedexternally from penultimate article 1 present [GEWB69]The apotele exists as a structure (ie dactyl clawempodium) differentiated from the tibiotarsus tarsusor telotarsus and is typically associated with a pair ofantagonistic muscles The apotele occurs as a distinctstructure in Xiphosura (Snodgrass 1948) Eurypter-ida (Clarke amp Ruedemann 1912) Schizomida (Coken-dolpher amp Reddell 1992) Araneae (Foelix 1996)early divergent Amblypygi (Charinus Weygoldt 2000not Phrynus Shultz 1999) Anactinotrichida (CaminClark amp Bourdeau 1956 Lindquist 1984) Ricinulei(Pittard amp Mitchell 1972 Legg 1976) and Solifugae(Roewer 1934) Giribet et al (2002) coded Solifugae asnot having a palpal apotele but it is present as anadhesive organ (45) Dunlop (1999) reconstructedHaptopoda as having a palpal claw but notes in hisdescription that the distal ends were not preserved inthe known specimens44 Apotele (claw) position 0 terminal 1 subtermi-nal - inapplicable coded only for taxa with a distinctapotele (43)State 1 occurs in Parasitiformes (Acari) (Evans 1992)45 Terminal adhesive organ 0 absent 1 present -inapplicable due to lack of an apotele (43) [GEWB 66]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITES IIIndashVI LEGS 2ndash5 IN NON-ARACHNIDS AND LEGS 1ndash4 IN ARACHNIDS

46 Appendage III (= arachnid leg 1) extremely elon-gate antenniform 0 absent 1 present [WP sim24 + 32S 20 WH 14 GEWB sim84]State 1 occurs in Amblypygi Thelyphonida Schizo-mida (Shear et al 1987) and Opilioacariformes (Vander Hammen 1989) State 1 is approximated in Solif-ugae Palpigradi (Roewer 1934) and Haptopoda(Dunlop 1999) and it is coded as uncertain in thesetaxa47 Appendage V (= arachnid leg 3) of male specializedfor sperm transfer 0 absent 1 present [WP 55 GEWB94]State 1 is known only from extant Ricinulei48 Appendages V and VI (= arachnid legs 3 and 4)with femur shorter than patella and with principal siteof flexionextension at patella-tibia joint (lsquoapatellatersquocondition sensu Van der Hammen 1989) 0 absent 1present [S 25 GEWB 86]

State 1 occurs in Pseudoscorpiones and Solifugae(Shultz 1989)49 Appendage VI (= arachnid leg 4) with terminusmodified as flattened blade 0 absent 1 presentState 1 occurs in some Chasmataspidida (DiploapisDunlop et al 2001 Octoberaspis Dunlop 2002a)uncertain in Chasmataspis Dunlop et al 2004 State0 in Loganamaraspis Tetlie amp Braddy 2004) andEurypterida except stylonuroids (Clarke amp Ruede-mann 1912 Waterston 1979 Tollerton 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI COXA AND BODYndashCOXA JOINT

50 Appendage III (= arachnid leg 1) with coxapophy-sis forming floor or wall of preoral chamber 0 absent1 present [S sim14 WH sim58]State 1 occurs in Opiliones (Hansen amp Soslashrensen 1904)and higher scorpions (Jeram 1998) A lobe is presenton the coxa of leg 1 in the fossil scorpion Prearcturusand perhaps Stoermeroscorpio and is coded here as acoxapophysis Coxapophysis of leg 1 together with theepistome and coxapophysis of the palp form a uniquepreoral chamber the stomotheca Weygoldt (1998) andDunlop amp Braddy (2001) reject the homology of cox-apophyses and stomothecae in Opiliones and Scorpi-ones because these structures are not apparent infossils of those scorpions thought to have been aquaticHowever the coxapophyses in Opiliones are formedlargely from soft cuticular lsquolipsrsquo and similar structuresin fossil scorpions would probably not have beenpreserved51 Appendage IV (= arachnid leg 2) with coxapophy-sis 0 absent 1 present [S sim14 WH sim58 GEWB 75]State 1 occurs in extant Scorpiones and most Opil-iones except Dyspnoi (Shultz 1998)52 At least one pair of coxal gnathobases on append-ages III-VI (= arachnid legs 1ndash4) 0 absent 1 present[S sim23 WH sim66 GEWB 78]State 1 occurs in all appropriately preserved Euryp-terida including Baltoeurypterus (Selden 1981) andStylonurus (Clarke amp Ruedemann 1912) extantXiphosura on appendages IIndashVI and apparentlyProscorpius on appendage III (= arachnid leg 1)(Kjellesvig-Waering 1986 Jeram 1998 contra Giribetet al 2002) Fossilized appendages with gnathobaseshave been found in association with Chasmataspis(Chasmataspidida) suggesting that they may havehad State 1 (Dunlop et al 2004)53 Coxae of appendages IIIndashV (= arachnid legs 1ndash3)with jointed moveable endites 0 absent 1 present [Ssim23]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) and in adequately preserved eurypterids(Clarke amp Ruedemann 1912) (BaltoeurypterusSelden 1981)

PHYLOGENY OF THE ARACHNID ORDERS 251

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

54 Intracoxal muscle a muscle arising on anteriorwall of coxa and inserting on posterior wall 0 absent1 present [GEWB 237]State 1 occurs in Thelyphonida (MastigoproctusShultz 1993) and Amblypygi (Phryus Shultz 1999)It is unclear whether the muscle operating the move-able endite (53) in extant Xiphosura (Manton 1964Shultz 2001) is homologous with this muscle and iscoded here as uncertain It is not present in Steno-chrus (Schizomida) (Shultz unpubl observ)55 Coxae of appendages IIndashVI (= arachnid palp andlegs) with dorsal articulation with carapace 0 absent1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)56 Flabellum (exite) on coxa of appendage VI(= arachnid leg 4) 0 absent 1 present [GEWB sim110]State 1 occurs in extant Xiphosura (Yamasaki et al1988) An apparent exite occurs on appendages asso-ciated with Chasmataspis (Chasmataspidida) (Dunlopet al 2004)57 Insertion process of anteromedial tergo-coxal mus-cle 0 weakly developed 1 well to extremely well devel-oped [GEWB 238]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Boumlrner 1904 Shultz 1999)58 Musculi laterales enlarged lateral tergocoxalmuscle with attachment shifted from coxa to adjacentpleural membrane 0 absent 1 present [S 22 WH 65GEWB 242]State 1 occurs in Araneae and Thelyphonida (Shultz1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI TROCHANTER AND COXAndashTROCHANTER JOINT

59 Coxandashtrochanter joint with complex posteriorarticulation composed of two articulating sclerites 0absent 1 present [S 24 WH 67 GEWB 85]State 1 occurs in Araneae Amblypygi Thelyphonidaand Schizomida (Shultz 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI FEMUR AND TROCHANTERndashFEMUR JOINT

60 Depressor muscle (or homologue) of trochanter-femur joint 0 absent 1 present [WH 42 GEWB 220]State 0 appears to be a unique synapomorphy of Ara-neae (Liphistius Shultz 1989 Aphonopelma Ruh-land amp Rathmayer 1978 Hypochilus present studyother examples Clarke 1984 1986)61 Trochanterndashfemur joint with dorsal hinge or pivotoperated by flexor muscles only 0 absent 1 present[WH sim44 GEWB sim91]State 1 occurs in Palpigradi Other chelicerates have abicondylar articulation and are typically equipped

with antagonistic muscles (Van der Hammen 1985Shultz 1989) but see 60 for an exception62 Superior trochanterndashfemur muscle (or homologue)originating broadly in femur inserting on distal mar-gin of trochanter 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 19892001)63 Basifemurndashtelofemur joint of appendages III andIV (= arachnid legs 1 and 2) in adult 0 absent 1present [S 25 WH 68 GEWB 86]The basifemur and telofemur apparently develop aslsquosisterrsquo articles through division of a parental articlethe femur (eg opilioacariform endeostigmatidpalaeosomatid mites) (Coineau amp Van der Hammen1979 Evans 1992) A single animal can have all legswith divided or undivided femora or have some com-bination of the two typically with the more posteriorlegs having the divided femora Thus anterior legsoften tend to be paedomorphic with respect toposterior legs Most chelicerates have undivided fem-ora on some or all legs but muscles homologous withthose of the basifemurndashtelofemur joint may stilldevelop (66) (Shultz 1989) suggesting that the devel-opment of cuticular and muscular components of ajoint have a degree of developmental and evolutionaryindependence

Undivided femora occur in most extant euchelicer-ates including Xiphosura Palpigradi AraneaeAmblypygi Thelyphonida Schizomida ScorpionesOpiliones Pseudoscorpiones higher Sarcoptiformes(Archegozetes Van der Hammen 1989) many Prostig-mata (Tetranychus Evans 1992) and throughout Par-asitiformes Divided femora occur in all legs in manyearly divergent Acariformes including many Endeo-stigmata basally divergent Sarcoptiformes (egPalaeacarus) and Prostigmata (eg AllothrombiumMicrocaeculus) (Van der Hammen 1989 Evans1992) Only leg 4 has basi- and telofemora in Alycus(Van der Hammen 1989) Basi- and telofemora occuron legs 3 and 4 in Solifugae Ricinulei Opilioacari-formes and Eurypterida (Clarke amp Ruedemann 1912Selden 1981 Shultz 1989)

Giribet et al (2002) coded the lsquoannulusrsquo of certainTrigonotarbida (Shear et al 1987) as a basifemur buta similar non-muscularized annulus is present in theposterior legs of extant Thelyphonida In both casesthe condyles of the trochanterndashannulus and annulusndashfemur joints are aligned in such away that they per-form essentially identical movements and appear notto enhance either the range or degree of leg move-ment [This contrasts with the condition in the lasttwo pairs of legs in Baltoeurypterus where condyles ofthe apparent trochanter-basifemur and basifemurndashtelofemur joints have different arrangements (Selden1981)] The annulus in Thelyphonida appears to func-tion in maintaining constant volume of the tro-

252 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

chanterndashfemur joint by preventing ballooning of thearthrodial membrane thereby minimizing the effect ofinternal fluid pressure on movement64 Basifemurndashtelofemur joint of appendage V(= arachnid leg 3) in adult 0 absent 1 present See 6365 Basifemurndashtelofemur joint of appendage VI(= arachnid leg 4) in adult 0 absent 1 present See6366 Cuticular differentiation of basifemur-telofemurjoint absent but muscles present 0 absent 1 present- inapplicable due to presence of joint (63ndash65)State 1 occurs in appendages of postoral somites IIIndashIV (= arachnid legs 1ndash4) in Xiphosura AraneaeAmblypygi Thelyphonida Schizomida Scorpionesand Pseudoscorpiones (Shultz 1989)67 Circumfemoral ring 0 absent 1 presentThe femora of one or more pairs of legs in anactinot-richid Acari have a basal groove or ring associatedwith slit sensilla Acarologists term that portionproximal to the ring the lsquobasifemurrsquo and that portiondistal to the ring the lsquotelofemurrsquo and they refer tothe basifemur as the lsquosecond trochanterrsquo and thetelofemur as the lsquofemurrsquo Despite the confusingterminology the basifemur discussed in 63ndash65 is nothomologous with the lsquobasifemurrsquo of opilioacariformand parasitiform Acari as both a true basifemurndashtelofemur joint and a circumfemoral ring are presentin legs 3 and 4 of Opilioacariformes (Van der Ham-men 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI PATELLA AND FEMURndashPATELLA JOINT

68 Patella of appendage of postoral somite III(= arachnid leg 1) proportionally much longer thanthose of more posterior appendages 0 absent 1presentState 1 occurs in Thelyphonida Schizomida (Hansenamp Soslashrensen 1905 Shultz 1989) and Opilioacari-formes (Van der Hammen 1989)69 Femurndashpatella joint 0 monocondylar several axesof movement and multifunctional muscles 1 bicondy-lar hinge one axis of movement and antagonistic mus-cles 2 hinge one axis of movement flexor muscleswithout muscular antagonists [S sim26 WH sim69 GEWBsim87]State 0 occurs in Solifugae State 1 occurs in Phalang-ida (Opiliones) Scorpiones and PseudoscorpionesState 2 occurs in the remaining taxa (Shultz 1989)70 Patellar plagula 0 absent or with simple medianattachment 1 symmetrical Y-shaped with long prox-imal stem 2 symmetrical U-shaped (= arcuate scler-ite) 3 asymmetrical attaching to patella only atanterior margin [S 27 WH sim70 GEWB 234 = 243]State 1 occurs in extant Xiphosura (Shultz 1989)State 2 occurs in Araneae (Manton 1958 Ruhland amp

Rathmayer 1978 Clarke 1986 Shultz 1989) State 3occurs in Amblypygi Thelyphonida and Schizomida(Shultz 1989) Giribet et al (2002) included this char-acter twice

APPENDAGES OF POSTORAL SOMITES IIIndashVI TIBIA AND PATELLAndashTIBIA JOINT

71 Tibiae divided by one or more joints 0 absent 1presentState 1 occurs in Amblypygi (Weygoldt 2000)72 Patellandashtibia joint 0 monocondylar with or with-out CZY (73) 1 bicondylar hinge one axis of move-ment antagonistic muscles 2 hinge one axis ofmovement muscles without muscular antagonists [Ssim31 WH 74 GEWB sim92] (Shultz 1989 Selden Shearamp Bonamo 1991)73 Patellandashtibia joint with posterior compression zone(lsquoCZYrsquo) 0 absent 1 present - inapplicable due toabsence of monocondylar articulation [GEWB 99]State 1 is a unique synapomorphy of Araneae (Seldenet al 1991)74 Patellandashtibia joint of appendages IIIndashVI(= arachnid legs 1ndash4) with deep-set monocondylar pivotbordered by a pair of tibial processes to which extensormuscles attach 0 absent 1 present - inapplicabledue to absence of monocondylar joint (72)State 1 occurs in extant Xiphosura (Snodgrass 1952Shultz 1989) although a similar arrangement ispresent in the appendage III (= arachnid leg 1) in The-lyphonida (Shultz 1993)75 Patellandashtibia joint largely immobile specializedfor autotomy 0 absent 1 presentState 1 occurs throughout Amblypygi (Weygoldt 19842000)76 Anterior femurndashtibia or femoropatellandashtibia(transpatellar) muscle 0 absent 1 present [S sim30GEWB 222] (Shultz 1989)77 Proximal attachment of posterior femurndashtibia orfemoropatellarndashtibia (transpatellar) muscle 0 muscleabsent 1 dorsoposterior surface of femur andor pos-terior surface of patella 2 distal process of femur mus-cle acting as extensor of femur-patella joint [S sim28 WHsim71 + 72 GEWB 235 = 244 245] (Shultz 1989)78 Distal attachment of posterior femurndashtibia (trans-patellar) muscle 0 posterior 1 dorsal acting as exten-sor of patellandashtibia joint - inapplicable due to absenceof muscle (77) [S 29 GEWB 221] (Shultz 1989)79 Anterior patellandashtibia muscle 0 absent 1 present[S sim32 WH sim73 GEWB 246] (Shultz 1989)80 Posterior patellandashtibia muscle 0 absent 1 present[S 33 WH 76 GEWB 248] (Shultz 1989)81 Patellandashtibia joint spanned by elastic (lsquospringlikersquo)sclerite 0 absent 1 present [GEWB 107]State 1 is a unique synapomorphy of Solifugae (Sens-enig amp Shultz 2003)

PHYLOGENY OF THE ARACHNID ORDERS 253

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

APPENDAGES OF POSTORAL SOMITES IIIndashVI TARSUS AND TIBIAndashTARSUS JOINT

82 Tarsus divided into proximal basitarsus(= metatarsus) and distal telotarsus (= distitarsus orlsquotarsusrsquo) 0 absent 1 presentState 0 occurs in extant Xiphosura (Shultz 1989) andthroughout Acariformes (Lindquist 1984 Evans1992) in the anterior two pairs of legs in chthonioidpseudoscorpions and all legs in lsquomonosphyronidrsquo pseu-doscorpions including Feaelloidea and Cheliferoidea(Chamberlin 1931) A circumtarsal ring (83) associ-ated with slit sensilla in Parasitiformes (Acari) mayrepresent a joint between a telotarsus and basitarsus(Evans 1992)83 Circumtarsal ring 0 absent 1 presentState 1 is an apparent synapomorphy of Parasiti-formes (Evans 1992)84 Telotarsus in adult with two or more tarsomeres 0absent 1 present - inapplicable due to absence of telo-tarsus (82) [GEWB 71]State 1 occurs in Schizomida Thelyphonida Ambly-pygi Palpigradi Ricinulei Solifugae (Shultz 1989)Opilioacariformes (Van der Hammen 1989) and Ple-siosiro (Dunlop 1999)85 Three telotarsomeres on appendages of postoralsomites IV-VI (= arachnid legs 2ndash4) 0 absent 1present - inapplicable due to absence of telotarsus (82)or absence of telotarsomeres (84) [S 34 WH 77 GEWB102]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Shultz 1989)86 Tibiandashtarsus joint spanned by well-developed elas-tic (lsquospringlikersquo) sclerite 0 absent 1 present [GEWB107]State 1 occurs in Scorpiones phalangid Opiliones andSolifugae (Alexander 1967 Shultz 2000 Sensenig ampShultz 2003) Extant Xiphosura is coded for the lastprosomal appendage only (see 88)87 Appendage of postoral somite VI (= arachnid leg 4)with ring of large basally articulated spatulate pro-cesses at tibia-tarsus joint 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)88 Appendages postoral somites IIIndashV (= arachnidlegs 1ndash3) with tibiotarsus (ie tibia and tarsus not dif-ferentiated) 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 1989) butnot apparently in Weinbergina (Moore et al 2005) theonly synxiphosurid in which appendages are suffi-ciently preserved to determine the number ofpodomeres89 Ambulacrum peduncle-like extension of the tarsuswith internal condylophores and terminating distallywith apotele (eg claws) andor pulvillus 0 absent 1present

State 1 is present throughout Acari (Alberti amp Coons1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI APOTELE AND TARSUSndashAPOTELE JOINT

90 Apotele of appendage III (= arachnid leg 1) 0absent or not apparent 1 present [GEWB 101]State 0 occurs in Amblypygi Thelyphonida and Schi-zomida (Shultz 1999) However the apotele is proba-bly present but undifferentiated because muscles thatwould normally insert on the apotele terminate on theend of the lsquotarsusrsquo Dunlop (1999) reconstructed Hap-topoda as lacking an apotelic claw on leg 1 but noted inthe text that this was speculative Dunlop (2002b) hasreviewed apotelic diversity in Chelicerata91 Appendages of postoral somites IIIndashVI (= arachnidlegs) chelate with chela formed from tibiotarsus andapotele or tarsus and apotele 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988) The condition appears to occur in Chasmataspisbased on associated appendages (Dunlop et al 2004)but not in Diploaspis and Octoberaspis (Dunlop et al2001 Dunlop 2002a)92 Apotele with eversible or padlike empodium(= pulvillus) in adult 0 absent although empodialclaw may be present 1 present [S sim38 WH 81 GEWB95]State 1 occurs throughout Pseudoscorpiones (Cham-berlin 1931) Solifugae (Roewer 1934) Opilioacari-formes Holothyrida Ixodida and free-livingMesostigmata (Evans 1992) An eversible lsquopulvillusrsquo isprobably a primitive feature of Amblypygi (eg Char-inus Weygoldt 2000) but is absent in higher groups(eg Phrynus Weygoldt 2000)93 Inferior apotele muscle (= claw depressor) with tib-ial attachment 0 absent 1 present [S 35 WH 78GEWB 249]The character is coded for extant Xiphosura usingappendage of postoral somite VI (= arachnid leg 4) asother legs lack a differentiated tibia (88) State 1appears to occur throughout Arachnida (Shultz 1989)94 Inferior apotele muscle (= claw depressor) withpatellar attachment 0 absent 1 present [S 36 WH 79GEWB 250]State 0 occurs in extant Xiphosura Acari AmblypygiPalpigradi and Ricinulei State 1 occurs in all otherarachnids examined here (Shultz 1989)

OPISTHOSOMA

95 Number of opisthosomal somites in adult 0 five1 eight 2 nine 3 10 4 11 5 12 6 13 [WH sim33GEWB sim190]Chasmataspidids appear to have 13 opisthosomalsomites (Dunlop et al 2004) The synziphosurans

254 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Weinbergina (Moore et al 2005) and Limuloides(Anderson amp Selden 1997) appear to have ten opistho-somal somites Moore et al (2005) found no evidence ofan anterior microtergite in Weinbergina as proposedby Anderson amp Selden (1997) Xiphosurids appear tohave nine opisthosomal somites (Scholl 1977 Ander-son amp Selden 1997 Shultz 2001) Members of the fol-lowing orders have 12 opisthosomal somites Araneae(Millot 1949ac) Amblypygi (Weygoldt 2000) Thely-phonida and Schizomida (Hansen amp Soslashrensen 1905)and Pseudoscorpiones (Chamberlin 1931) Elevensomites are present in Palpigradi Solifugae (Roewer1934) and Trigonotarbida (Gilboarachne Shear et al1987 Palaeocharinus Fayers et al 2004) Plesiosiromay have 11 or 12 somites (Dunlop 1999) Justifica-tion for coding of more controversial taxa is providedbelow

Scorpiones Two principal hypotheses regarding thenumber of opisthosomal somites in scorpions havebeen advocated a 13-somite hypothesis derived fromembryological studies (Brauer 1895 Patten 1912Farley 1999 2005) and a 12-somite hypothesis basedon comparative anatomy of adults (Weygoldt amp Pau-lus 1979) The embryological interpretation is basedon the observation of pregenital genital and pectinalsomites (each with segmental ganglia and paired limbbuds) in early scorpion embryos followed by extremereduction or loss of the pregenital somite in laterembryos According to this view a missing pregenitalsomite should be added to the 12 apparent opisthoso-mal somites of post-embryonic scorpions to achieve afinal number of 13 The anatomy-based hypothesiswas introduced by Weygoldt amp Paulus (1979) whoadvocated a literal interpretation of post-embryonicsegmentation based on opisthosomal tergites Specifi-cally these authors argued that the last pair of dorsalendosternal suspensor muscles of non-scorpionarachnids especially Pedipalpi attach to the first(= pregenital) somite that this condition also occurs inscorpions and that there is no reason to invoke a miss-ing pregenital tergite They proposed that the pectinesbelong to the genital somite and that functional spe-cializations of the nervous system for pectinal functiongive the apparence of an extra neuromere duringembryonic development

I recently dissected the prosoma and anterioropisthosoma of the scorpions Centruroides Hadrurusand Heterometrus and focused on the composition ofthe muscular diaphragm (103) that separates thehaemocoelic compartments of the prosoma andopisthosoma (Shultz 2007) The diaphragm is com-posed of a metameric series of axial muscles fromthree somites the anterior somite corresponds to thelast prosomal somite and the posterior somite corre-sponds to the genital somite The middle elements

insert dorsally along a tranverse tendon attached tothe anterior margin of the first tergite These obser-vations are consistent with the embryological inter-pretation that a pregenital somite is present but itstergite is not expressed It appears likely that the thepregenital somite was compressed longitudinally dur-ing the evolution of the diaphragm I code scorpions ashaving 13 opisthosomal somites

Eurypterida The eurypterid opisthosoma is widelyassumed to have 12 somites (Clarke amp Ruedemann1912 Stoslashrmer 1944) However in a speculative paperon the evolutionary morphology of trilobites and che-licerates Raw (1957) proposed that both scorpions andeurypterids have 15 opisthosomal somites Rawassumed that scorpions have 13 apparent opisthoso-mal somites based on the transient pregenital somiteof scorpion embryology that scorpions and eurypteridsare close relatives and should have the same numberof somites and that olenellid trilobites and chelicer-ates always have somites in multiples of three Rawachieved 15 somites in scorpions by assuming the lastopisthosomal somite to be a diplosomite and that thetelson is a postanal somite even though there is noevidence for either of these proposals He attributedthese features to eurypterids as well He also notedthat the connection between the prosomal carapaceand first opisthosomal tergite in Eurypterida differedstructurally from the connection between adjacentopisthosomal tergites and regarded this as evidencefor a reduced pregenital tergite in the prosomandashopisthosoma junction Rawrsquos speculations were largelyforgotten until Dunlop amp Webster (1999) resurrectedthe proposal that eurypterids have a reduced opistho-somal tergite and therefore share a unique similaritywith scorpions Unfortunately Dunlop amp Bullocktreated Rawrsquos conjecture as if it were based on empir-ical evidence rather than an attempt to force euryp-terid morphology into a peculiar numerical system Inthe absence of convincing evidence to the contrary Ihave coded eurypterids as having 12 opisthosomalsomites

Ricinulei The opisthosoma consists of a membranouspedicel (97) bearing the gonopore ventrally Thefemale gonopore is bordered by an anterior plate and aposterior plate The remainder of the opisthosoma iscomposed of thick sclerites separated by less heavilysclerotized cuticle The first dorsal sclerite is short andfunctions as part of a prosomandashopisthosoma couplingmechanism (96) that may or may not be a specializedcomponent of the following tergite This is followed byfour tergites and sternites and a three-segmentedmetasoma (= lsquopygidiumrsquo) Millot (1945 1949f) rea-soned that the opisthosoma contains ten somites withsomites VIIndashIX incorporated into the pedicel XndashXIIIexpressed as tergites and sternites and the metasoma

PHYLOGENY OF THE ARACHNID ORDERS 255

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

comprising four somites He did not regard the dorsalcoupling sclerite as a separate tergite and his inter-pretation of four rather than three metasomal somiteshas been rejected Pittard amp Mitchell (1972) also pro-posed ten opisthosomal somites but achieved thisnumber by regarding the dorsal coupling sclerite as atergite of somite IX and by recognizing three metaso-mal somites Legg (1976) adopted the system proposedby Pittard amp Mitchell but did not regard the dorsalcoupling sclerite as separate from the following tergite(X) Van der Hammen (1979 1989) reconstructed 13somites The pedicel was regarded as having twosomites (VII + VIII) the following four sets of tergitesand sternites were considered diplosomites (IX + XXI + XII XIII + XIV XV + XVI) and the metasoma hadthree somites (XVIIndashXIX) This interpretation mayhave been biased by Van der Hammenrsquos attempt tounite Ricinulei with anactinotrichid Acari which healso regarded as having a primitive number of 13somites Selden (1992) described a fossil ricinuleidTerpsicroton that shows two pairs of depressions onthe three large premetasomal tergites which con-trasts with the single pair seen in extant species Thisobservation appears to corroborate the diplosomitehypothesis but the evidence does not indicate that thetergite anterior to these is a diplosomite

Dunlop (1996) attempted to homologize theprosomandashopisthosoma coupling mechanisms ofRicinulei and Trigonotarbida a goal that required anovel and rather forced interpretation of the dorsalsclerites He regarded the coupling sclerite as homol-ogous with the first opisthosomal tergite (VII) of trig-onotarbids and then followed Van der Hammenrsquosdiplosomite hypothesis to achieve 12 somites in totalDunloprsquos scheme differs from previous systems in sug-gesting that the pedicel does not contain the dorsalelements of the first and second somites and is in-consistent with Millotrsquos (1945) observation that thepre- and postgenital plates each have dorsoventralmuscles Here I code Ricinulei as having 12 somitesThere are three metasomal somites (XVIndashXVIII)three diplosomites (= six somites) (XndashXV) one cou-pling somite (IX) and two somites in the pedicel (iethe genital and pregenital somites) (VII VIII)

Opiliones Harvestmen have nine opisthosomalsomites and an anal operculum that is traditionallyregarded as the tergite of a tenth somite a tenth ster-nite is lacking (Hansen amp Soslashrensen 1904 Winkler1957) However the anal operculum appears to repre-sent a persistent embryonic telson (Moritz 1957) andis therefore likely to be a postsegmental structurecomparable with the stinger of scorpions flagella ofthelyphonids etc (Shultz 2000)

Acari The number of opisthosomal somites is probe-matic for most mite taxa due to uncertainty about the

location of the prosomandashopisthosoma boundry (5)extensive simplification or loss of metamericallyarranged sclerites and muscle attachments a paucityof developmental studies of engrailed expression andin Acariformes opisthosomal anamorphosis and het-erochronic modification of somite number (Evans1992) Ixodids appear to be the exception developmen-tal studies indicate five somites in the opisthosoma ofticks (Evans 1992)

Some mites retain external evidence of segmenta-tion ndash metameric patterns of furrows muscle attach-ments and slit sensilla ndash and with certainassumptions the number of opisthosomal somites canbe estimated Here it is assumed that the dorsal sur-faces of the last two prosomal somites are present inmites and retain their primitive association with legs3 and 4 (contra Van der Hammen 1989) (see 5 for jus-tification) Given this there appear to be 11 somites inthe opisthosoma of Opilioacariformes a conclusionalso reached by other workers (eg With 1904 Sitni-kova 1978 Klompen 2000) Similar reasoning sug-gests that Alycus has seven opisthosomal somites notnine as advocated by Van der Hammen (1989) Unfor-tunately external evidence is ambiguous for determin-ing the number of somites in other mite lineages andthese are coded as uncertain96 Prosoma-opisthosoma coupling mechanism 0absent 1 present [GEWB sim24]The posterior margin of the carapace and the anteriormargin of the first apparent opisthosomal tergite arespecialized as a coupling mechanism in Ricinulei andTrigonotarbida and Dunlop amp Horrocks (1996) pro-posed that these are synapomorphic for the twogroups However there is uncertainty about thehomology of the anterior opisthosomal somites in Trig-onotarbida and Ricinulei (95)97 Pedicel 0 absent 1 aranean type 2 ricinuleidtype [WP 30 S sim40 WH sim20 GEWB sim126]The body narrows at or near the prosomandashopisthosoma juncture in several arachnid lineages (ieSolifugae Palpigradi Amblypygi Araneae Ricinulei)and this lsquowaistrsquo has often been used as character at theinterordinal level (eg Pocock 1893) However thisinterpretation is rejected here for being subjective andfor uniting non-homologous conditions For exampleAraneae and Amblypygi are often grouped on the basisof a lsquopedicelrsquo yet it is a highly specialized structure inAraneae and its parts are not readily homologizedwith those of Amblypygi In contrast the condition inAmblypygi is a slightly narrower version of the highlymoveable prosomandashopisthosoma juncture in Uropygiwhich is not generally considered a lsquopedicelrsquo Thepedicel in Ricinulei is also unique a weakly sclero-tized stalk containing the genital opening (Cryptocel-lus Pittard amp Mitchell 1972 Ricinoides Legg 1976)See 95

256 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

98 Opisthosoma with three- segmented lsquobucklerrsquo 0absent 1 presentState 1 occurs throughout Chasmatapsidida (Dunlop2002a)99 Thoracetron consolidation of tergites of postgeni-tal somites 0 absent 1 present - inapplicable codedonly for Xiphosura (Anderson amp Selden 1997)100 Fusion of tergites of postoral somites VIII and IX(= opisthosomal somites 2 and 3) only 0 absent 1present [GEWB 146]State 1 appears to be a synapomorphy Trigonotarbida(Shear et al 1987) Ricinuleids also have diplotergites(101) but these appear to encompass a different com-bination of somites (see 95)101 Three diplotergites 0 absent presentState 1 is a unique synapomorphy of Ricinulei (Selden1992) See 95102 Paired opisthosomal defensive glands openingvia ducts on either side of the anus 0 absent 1present [S 46 WH 86 GEWB 122]State 1 occurs in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905)103 Muscular diaphragm separating prosomal andopisthosomal compartments formed by dorsoventralmuscles of postoral somites VIndashVIII and extrinsic mus-cles of leg 4 0 absent 1 presentState 1 is known only in extant scorpions (Lankesteret al 1885 Firstman 1973) The diaphragm is oftentreated as a single structure but recent anatomicalwork (Shultz 2007) has shown that it is a composite ofdorsal endosternal suspensors and extrinsic leg mus-cles (see 95) A diaphragm is also present in Solifugaebut it is located more posteriorly and does not appearto be homologous with that of scorpions (Roewer1934)104 Opisthosomal appendicular chondrites 0absent 1 presentThese cartilage-like columns of mesodermally derivedtissue are associated with each opisthosomal append-age including the chilaria in extant Xiphosura(Patten amp Hazen 1900 Yamasaki et al 1988 Fahr-enbach 1999 Shultz 2001)105 Paired appendages on ventral surface of postoralsomite VII (= opisthosomal somite 1) in adult 0absent 1 present [WP sim13 S 39 WH 82 GEWB 143]State 1 is known in Weinbergina (Moore et al 2005)and extant Xiphosura The bilobed structure of themetastoma in certain eurypterids as well as the cor-responding placement of chilaria in Xiphosura haveled some workers to regard the metastoma as fusedappendages of postoral somite VII (eg Stoslashrmer1955) This hypothesis is problematic given that (i) nospecial explanation is needed to account for bilaterallysymmetrical structures in bilaterally symmetricalorganisms (ii) incorporation of the ventral part of thefirst opisthosomal sternite as a functional element of

the prosoma is typical of arachnids and perhapsEuchelicerata generally and (iii) some workers regardthe metastoma as a sternite (eg Jeram 1998) Thusthe metastoma is coded here as uncertain Dunlop hasproposed that the sternum of scorpions is derivedfrom appendages citing the presence of transient limbbuds in scorpion development (Brauer 1895 Patten1912) but this is not compelling evidence for a persis-tent appendicular contribution to the sternite of theadult106 Megoperculum 0 absent 1 present [S sim41 + 42WH sim83 GEWB 162 + 163]State 1 occurs in Thelyphonida Schizomida Ambly-pygi Araneae (Shultz 1993 1999) Trigonotarbida(Shear et al 1987) and probably Haptopoda (Dunlop1999) and Palpigradi The megoperculum consists ofappendages of opisthosomal somite 2 (= genital somiteor postoral somite VIII) (Shultz 1993 1999 Popadicet al 1998) that have fused medially and displacedthe ventral body wall of the somite anteriorly (oftenrepresented by a small sternite to which dorsoventralmuscles of postoral somite VIII attach) The megoper-culum projects posteriorly to form the ventral surfaceof a pregenital chamber The dorsal surface of thechamber is formed by the ventral body wall of opistho-somal somite 3 (= postoral somite IX) which serves asthe ventral attachment of the paired dorsoventralmuscles of that somite The megoperculum bears book-lungs (where present) and sometimes paired gonopodsthat are probably derived from telopodites A similarbut less developed operculum is present on theopisthosomal somite 3 which may also bear booklungsand eversible vesicles (eg Amblypygi) (see 111) corre-sponding to the booklungs and gonopods of the mego-perculum (Shultz 1999)

I contend here that a megoperculum is present inPalpigradi and that this is particularly evident infemales Specifically a large unpaired lobe projectsposteriorly from the genital somite over the ventroan-terior surface of the first postgenital somite therebyforming a pregenital chamber The dorsoventral mus-cles of the genital somite attach ventrally near theanterior border of postoral somite VIII and those ofpostoral somite IX attach to the upper surface of thepregenital chamber (Rucker 1901 Boumlrner 1902bRoewer 1934 Millot 1949d Van der Hammen 1989)Weygoldt (1998) has questioned this interpretation ofPalpigradi107 Postgenital operculum or lsquosternitersquo 0 sclerotized1 not sclerotized - inapplicable coded only for Pan-tetrapulmonata (Platnick amp Gertsch 1976)State 1 is present in Trigonotarbida Amblypygi The-lyphonida Schizomida and Mesothelae (Araneae) butnot Opisthothelae (Araneae)108 Genital opening of female guarded by four plates(one pregenital one postgenital two laterogenitals)

PHYLOGENY OF THE ARACHNID ORDERS 257

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

genital opening of male guarded by two plates 0absent 1 presentThis an apparent synapomorphy of Holothyrida(Acari) (Van der Hammen 1989)109 Anterior margin of genital opening in male withglands secreting via fusules 0 absent 1 presentState 1 occurs in Araneae as epigastric or lsquoepiandrousrsquoglands (Marples 1967) and in Palpigradi (Condeacute1991b)110 Paired valve-like plates apparently formed fromcomponents of three somites covering triradiate genitalopening 0 absent 1 presentThis is an apparent synapomorphy of Acariformes(Evans 1992 Alberti amp Coons 1999)111 Eversible lsquoappendagesrsquo on the ventral surface ofpostgenital somites 0 absent 1 present [S sim43 WHsim84 GEWB sim127]Paired ventral lsquoappendagesrsquo operated in part byhaemolymph pressure occur in the form of ventralsacs in certain Amblypygi (postoral somite IX) (Wey-goldt 2000) (Charinus Millot 1949b not PhrynusShultz 1999) prokoenenian Palpigradi (postoralsomites XndashXII) (Rucker 1901 Condeacute 1991a) as spin-nerets in Araneae (XndashXI) (Shultz 1987) and as geni-tal papillae in many Acariformes (Alberti amp Coons1999) including Oribatida many Endeostigmata (Aly-cus Van der Hammen 1989) and Prostigmata (Allo-thrombium Saboori amp Kamali 2000 MicrocaeculusEvans 1992) Eversible vesicles similar to those ofamblypygids occur in fossils of the trigonotarbidPalaeocharinus (postoral somite IX) (Fayers et al2004) Two cuticular structures (lsquogenital verrucaersquo)occur anterior to the genital opening in Opilioacari-formes and each covers the opening to a thin-walledinvaginated sac (lsquogenital papillarsquo) of unknown function(Van der Hammen 1989) Van der Hammen proposedthat it is homologous to a genital papilla of Acari-formes but Alberti amp Coons (1999) have questionedthis interpretation on several grounds and refer tothese structures as pregenital capsules112 Opisthosomal silk glands and spinnerets derivedfrom appendages on postoral somites X and XI(= opisthosomal somites 4 and 5) 0 absent 1 present[WP 33 WH 41 GEWB 123 + 142]State 1 is a unique synapomorphy of Araneae (Plat-nick amp Gertsch 1976) although potentially homolo-gous glands and spigot-like setae occur on the ventralopisthosomal surface in Palpigradi (Millot 1943Condeacute 1991a)113 Opisthsosomal spinnerets location 0 near mid-dle of opisthosoma 1 near posterior end of opisthso-som ndash inapplicable coded only for Araneae (Platnickamp Gertsch 1976)114 Anterior medial lsquospinneretsrsquo 0 absent 1 present- inapplicable coded only for Araneae (Platnick ampGertsch 1976)

115 Opisthosomal tergites divided longitudinally intoone median and two lateral plates 0 absent 1 present[GEWB 145]State 1 occurs in non-curculioid Ricinulei (Selden1992) and many Trigonotarbida (Dunlop 1996) Thisfeature or something very like it (eg distinct opistho-somal trilobation) occurs sporadically in severaleuchelicerate groups including Chasmataspis (Chas-mataspidida) (Dunlop et al 2004) most fossil xipho-surans (Anderson amp Selden 1997) and a few derivedEurypterida (especially Mixopteroidea) (Tollerton1989)116 Number of metasomal somites 0 zero 1 two 2three 3 five 4 nine [S sim44 WH sim85 GEWB 128 144]The metasoma is a preanal region comprising multiplesomites that are substantially narrower than the pre-ceding somites or mesosoma Metasomal somitesoften lack pleural membranes and take the form ofsclerotized rings State 1 occurs in TrigonotarbidaState 2 is present in synziphosurid Xiphosura (Ander-son amp Selden 1997) Amblypyi Thelyphonida Schizo-mida and Ricinulei State 3 occurs in Scorpiones andEurypterida and State 4 occurs in Chasmatispidida

Some workers appear to consider a five-segmentedmetasoma as conclusive evidence for the monophylyof a Eurypterida + Scorpiones clade (eg Dunlop ampBraddy 2001) presumably because the character rep-resents a kind of tagmosis and should therefore begiven substantial phylogenetic weight However Wey-goldt (1998) has pointed out that a metasoma contain-ing three somites is probably part of the ground planof Euchelicerata The known diversity of the eucheli-cerate metasoma indicates that this feature can inincrease or decrease its segmental composition in evo-lution or be eliminated entirely without necessarilychanging the total number of opisthosomal somitesConsequently it is problematic to assume that thefive-segmented metasoma has any special immunityto homoplasy117 Postanal structure (telson) 0 absent or not obvi-ously developed 1 presentPostanal structures are considered non-segmentalbecause they are modifications of the embryonicregion posterior to the site of somite addition (embry-onic growth zone) A well-developed postanal struc-ture persists in adults in Xiphosura EurypteridaChasmataspidida Scorpiones Schizomida Thely-phonida and Palpigradi The anal operculum of Opil-iones is generally considered by morphologists to bethe tergite of the tenth opisthosomal or anal somite(eg Hansen amp Soslashrensen 1904) but embryological evi-dence (Moritz 1957) suggests that it is derived fromthe postproliferative zone and thus corresponds to thetelson of other chelicerates (see 95)118 Postanal structure shape 0 caudal spine 1aculeus 2 flagellum 3 anal operculum - inapplica-

258 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ble due to absence of postanal structure (117) [WP sim41S sim45 WH sim24 GEWB sim121 129 + 147]State 0 is present in Xiphosura Eurypterida andChasmataspidida State 1 occurs in Scorpiones State2 occurs in Palpigradi Thelyphonida and SchizomidaState 3 occurs in Opiliones (see 95 117)119 Specialized postanal flagellum in male (see 157ndash159) 0 absent 1 present - inapplicable coded onlyfor taxa with postanal flagellum (118) [GEWB 131]State 1 is a unique synapomorphy of Schizomida(Hansen amp Soslashrensen 1905 Cokendolpher amp Reddell1992 Reddell amp Cokendolpher 1995)

RESPIRATORY SYSTEM

120 Respiratory medium 0 water 1 airAmong the terminal taxa included only fossil scorpi-ons are problematic for this character They are codedhere as uncertain Based on a study of book lungmicrosculpture Scholtz amp Kamenz (2006) have arguedthat arachnids are primitively terrestrial and pulmo-nate (see also Firstman 1973) and have questionedwhether any fossil scorpions were aquatic No positionon this proposal is taken here121 Respiratory lamellae on opisthosomal somite 2(= genital somite postoral somite VIII) 0 absent 1present [WP sim37 S sim51 WH sim22 GWEB sim133ndash137]This character encompasses book gills (= lamellae thatfunction in water) and book lungs (= lamellae thatfunction in air) The distinction between book gills andbook lungs is accommodated here by 120 in combina-tion with 121ndash124 Respiratory lamellae are presenton the genital somite in Trigonotarbida AraneaeAmblypygi Thelyphonida and Schizomida Petrunk-evitch (1949) reconstructed Plesiosiro as having booklungs but Dunlop (1999) could not corroborate thisThis character was used by Dunlop amp Webster (1999)to propose that Xiphosura and Scorpiones are closelyrelated because they both lack respiratory lamellae onthe genital somite Dunlop amp Braddy (2001) alsoargued for the placement of Eurypterida with Xipho-sura and Scorpiones based in part on this characterAt least some eurypterids may have had respiratorylamellae (eg Manning amp Dunlop 1995) but the onlyevidence of their segmental distribution is derivedfrom one specimen (Braddy et al 1999)122 Respiratory lamellae on opisthosomal somite 3(= postoral somite IX) 0 absent 1 present [GEWB136]State 1 occurs in Amblypygi Thelyphonida and Xipho-sura It is a groundplan feature of Araneae (Platnick ampGertsch 1976) and is present in all representativetaxa included here Braddy et al (1999) proposed theexistence of respiratory lamellae on postoral somitesIXndashXII in Eurypterida based on evidence from onespecimen This character is coded as uncertain for the

Eurypterida included here Extant scorpions are codedhere as lacking respiratory lamellae on this somite(= pectinal somite) based on the interpretation of theopisthosomal segmentation discussed in 95123 Respiratory lamellae on opisthosomal somites 4ndash6 (= postoral somites XndashXII) 0 absent 1 presentState 1 is definitely present in extant Xiphosura andScorpiones and may have occurred in Eurypterida124 Respiratory lamellae on opisthosomal somite 7(= postoral somite XIII)State 1 occurs in extant Xiphosura and Scorpionesonly125 Kiemenplatten 0 absent 1 presentThese structures are located on the roof of opercularchambers in Eurypterida (Clarke amp Ruedemann1912) where they take the form of ventrally projectingcones with a distinct cuticular microsculputure (Man-ning amp Dunlop 1995) Dunlop amp Braddy (2001)inferred the existence of Kiemenplatten in all Palaeo-zoic scorpions based on a description and photos of onespecimen of Paraisobuthus duobicarinatus by Kjelles-vig-Waering (1986 pls 16ndash18) The plates depict darkcone-like rearward-pointing denticles distributedwithin a white amorphous material The denticlesappear to lack microsculpture even though the mag-nifications at which the photos were taken (times90ndashtimes330)are comparable with those illustrating the cones ofKiemenplatten (times170ndashtimes350 in Manning amp Dunlop1995 figs 1 2) Given the diversity of cuticular struc-tures present in the atria of booklungs book gills andtracheae the denticles appear to bear no special sim-ilarity to the cones of Kiemenplatten126 Tracheal system 0 absent 1 paired ventral stig-mata on postoral somite VIII (= opisthosomal somite2) 2 paired ventral stigmata on postoral somites IXand X 3 one pair of stigmata opening near legs 3 or 44 paired stigmata associated with chelicerae 5 fourpairs of stigmata on dorsal surface of opisthosoma -inapplicable aquatic (120) [WP 40 sim43 S sim52 + 5354 WH sim23 45 88 GEWB sim138 139]Firstman (1973) and Weygoldt amp Paulus (1979)hypothesized that tracheae are homologous in all tra-cheate arachnids except spiders and subsequentworkers have entertained this hypothesis by includinga character for the presenceabsence of tracheal sys-tems that ignores the diverse arrangement of stig-mata in arachnids (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) However thisapproach assumes that internal tracheal systems areconserved but that tracheal openings (stigmata)appear and disappear on different parts of the bodywith higher evolutionary frequency Here I assumethat stigmata are conserved in evolution and that dif-ferences in their anatomical placement reflect the evo-lution of new tracheae State 1 occurs throughoutOpiliones State 2 occurs throughout Solifugae

PHYLOGENY OF THE ARACHNID ORDERS 259

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

(Roewer 1934) and Pseudoscorpiones (Chamberlin1931) State 3 recognizes the possible homology of stig-mata in Ricinulei Opilioacariformes and Parasiti-formes (Acari) (Lindquist 1984 Van der Hammen1989) Tracheal systems are apparently absent in thegroundplan of Acariformes but Prostigmata typicallyhave tracheae associated with the chelicerae Tra-cheae derived from the posterior pair of book lungs arewidespread in araneomorph Araneae but the booklungs are retained and tracheae are absent in themost basally divergent groups (eg Hypochilus)

BOXndashTRUSS AXIAL MUSCLE SYSTEM (BTAMS)

127 Posterior oblique muscles of BTAMS of postoralsomites IndashVI 0 absent 1 present in one or moresomitesState 0 occurs in Xiphosura and state 1 occurs inPalpigradi Araneae Amblypygi Thelyphonida andSchizomida (Shultz 2001) The condition in AcariRicinulei Opiliones and Solifugae is not known128 Anterior oblique muscles of BTAMS posterior topostoral somite VI 0 absent 1 presentState 1 occurs in Xiphosura which is probably theprimitive condition based on comparison with otherarthropods (Shultz 2001) State 0 occurs in all arach-nids examined thus far129 Ventral attachments of posterior oblique musclesof opisthosomal BTAMS located in prosoma 0 absent1 presentState 1 occurs in extant Xiphosura (Shultz 2001)130 Endosternite fenestrate 0 absent 1 present [S 8WH 54 GEWB 35]State 1 occurs in Thelyphonida and hubbardiid Schi-zomida (Firstman 1973) The condition in protoschi-zomid Schizomida is not known131 Suboral suspensor a tendon that arises from theBTAMS and inserts on the ventral surface of the oralcavity via muscle 0 absent 1 presentState 1 occurs in Palpigradi (Eukoenenia Millot1943) Amblypygi (Shultz 1999) and Thelyphonida(Shultz 1993)132 Perineural vascular membrane in adult 0absent 1 present [WH 28 GEWB 155]Wheeler amp Hayashi (1998) coded extant Xiphosura asunknown although the primary source for this char-acter (Firstman 1973) stated that adult Limulus havea perineural vascular membrane Within Arachnidapresence of a perineural vascular membrane is appar-ently correlated with the presence of tracheae (126)133 Ventral endosternal suspensor attaching on coxaof anteriorly adjacent somite 0 absent 1 present[GEWB 239 253]State 1 occurs in Amblypygi (Shultz 1999) Thely-phonida (Shultz 1993) and Schizomida (Shultzunpubl observ)

134 Posteriormost postoral somite with a pair of dor-soventral muscles 0 VI 1 VII 2 VIII 3 XII 4 XIII5 XIV 6 XV 7 XVIDorsoventral muscles tend to run in a continuousmetameric series beginning in the prosoma Deter-mining the last tergite on which the muscle seriesends is substantially easier than counting especiallygiven frequent anatomical complexities near theprosomandashopisthosoma juncture The dorsoventralmuscle series ends on postoral somite VII inCyphophthalmi (eg Chileogovea) and Laniatores(Opiliones) (Shultz unpubl observ) on postoralsomite VII or VIII in Eupnoi (Leiobunum Shultz2000 unpubl observ) on postoral somite XII in Pal-pigradi (Roewer 1934) Solifugae (Bernard 1896Roewer 1934 Millot amp Vachon 1949) and Araneae(Liphistius Hypochilus Millot 1933 but segmenta-tion reinterpreted here following Shultz 1993 1999)on postoral somite XIII in hubbardiid Schizomida(Cokendolpher amp Reddell 1992) and Scorpiones(Lankester et al 1885) and on postoral somite XIVin extant Xiphosura (Shultz 2001) Thelyphonida(Shultz 1993) protoschizomid Schizomida (Cokendol-pher amp Reddell 1992) and Amblypygi (Shultz 1999)Dunlop (1999) reconstructed Plesiosiro as havingpaired tergal apodemes ending on postoral somite XVbut his figures indicate that the series ends on posto-ral somite XIV The interpretation of opisthosomalsegmentation in Ricinulei coded here (95) indicatesthat the series ends on postoral somite XV in Terpsi-croton and this is treated as the groundplan for theorder The posteriormost dorsoventral muscles inPseudoscorpiones occur on postoral somite XVI(Vachon 1949)

NERVOUS SYSTEM

135 Segmental ganglia 0 consolidated in prosoma1 one or more present in opisthosoma [WH 30 GEWB210 + 216]Adult neuromeres are exclusively prosomal in Ambly-pygi liphistiomorph and araneomorph Araneae Opil-iones Pseudoscorpiones Ricinulei and Acari (Millot1949a)136 Dorsal median eyes 0 absent 1 present [WP14 + 47 S 50 WH sim7 GEWB 1]Median eyes are a groundplan feature of Euchelicer-ata and are retained in extant Xiphosura Eurypteridaand occur in all arachnid orders except PalpigradiSchizomida Ricinulei and Pseudoscorpiones (Paulus1979) Median eyes are also absent in opilioacariformand parasitiform Acari (Lindquist 1984) and perhapsthe fossil whipscorpion Proschizomus (Dunlop amp Hor-rocks 19951996) They are unknown in synziphosu-ran Xiphosura Median eyes are present in a variety ofbasal acariform mites including Endeostigmata (but

260 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

not Alycus) Prostigmata (Microcaeculus) and certainoribatids (Palaeacarus) (Evans 1992 Alberti amp Coons1999) Giribet et al (2002) coded all representativeAcari as lacking median eyes It is not known whetherthe eyes of some cyphophthalmid opilions are medianor lateral evidence from tracheal branching inCyphophthalmus (Janczyk 1956) suggests that theyare median eyes (Shultz amp Pinto da Rocha 2007) andpresence of a tapetum in Stylocellus is consistent withlateral eyes (Shear 1993)137 Retinula cells of dorsal median eyes 0 organizedinto closed rhabdoms 1 organized into network ofrhabdomeres 2 disorganized - inapplicable due toabsence of median eyes (136) [GEWB sim3]State 0 is present in Scorpiones Thelyphonida andAmblypygi State 1 is present in Solifugae and Ara-neae state 2 is present in Xiphosura (Paulus 1979)The retinula cells of median eyes have been studied inseveral prostigmatid Acariformes and are organized ina network in some taxa and in an irregular pattern inothers (Alberti amp Coons 1999) Retinulae in phalangidOpiliones have state 0 proximally and state 2 distally(Schliwa 1979)138 Ventral median eyes 0 absent 1 presentState 1 occurs in early instars of extant Xiphosura(Paulus 1979)139 Lateral eyes 0 absent 1 presentLateral eyes are primitively present in Chelicerataand are absent in Palpigradi and Opiliones (Paulus1979) It is unclear whether the eyes of cyphoph-thalmid opilions are median or lateral (see 136)140 Arrangement and number of lateral eyes 0 com-pound many 1 five or more pairs (includes micro-lenses) 2 three primary pairs (excludes microlenses)3 two pairs 4 one pair - inapplicable due to absenceof lateral eyes (139) [WP sim13 + 18 + 38 + 44 + 52 Ssim49 WH sim10 GEWB sim4 + 5]True compound eyes are present in XiphosuraEurypterida Chasmataspidida and many PalaeozoicScorpiones Among the anactinotrichid Acari Opilioa-cariformes have two (Neocarus) or three (Siamacarus)pairs of lateral eyes (although at least one species ofSiamacarus lacks eyes) the allothryrid Holothyridaand many Ixodida have a single pair of lateral lensesand the remainder apparently lack eyes (Evans1992) Thelyphonida was coded as lsquo12rsquo to reflect fivepairs of lenses comprising three pairs of primarylenses and two pairs of small accessory lenses (notsimply three pairs as coded by Giribet et al 2002)Trigonotarbida is also coded as lsquo12rsquo to reflect threepairs of primary lenses and multiple small accessorylenses Dunlop (1999) illustrated Haptopoda as havingpaired triads of lateral eyes but noted that there wasactually no evidence of this in the fossils The lateraleyes of fossil ricinuleids have two pairs of lenses(Selden 1992) One pair of eyes or eyespots are

present in extant Ricinulei and many Schizomidaalthough five genera of hubbardiid Schizomida have apair of lenses (Reddell amp Cokendolpher 1995) State 1is the groundplan for extant Scorpiones and State 3occurs in the pseudoscorpions Chthonius FeaellaNeobisium (but State 4 in Chelifer) and Solifugae(Paulus 1979)141 Lateral eyes with closed rhabdoms 0 absent 1present - inapplicable due to absence of lateral eyes(139) [WP 21 WH 12 GEWB 6]Presence of closed rhabdoms is probably primitive forChelicerata and is retained in extant Xiphosura andScorpiones Retinula cells form a network of rhab-domeres in other extant chelicerates (Paulus 1979)but these networks can differ substantially in detailWeygoldt (1998) and other workers have given sub-stantial weight to the network character in unitingnon-scorpion arachnids but these authors seem not togrant comparable phylogenetic significance to theanalogous character of the median eyes shared bySolifugae and Araneae (137)142 Slit sensilla 0 absent 1 present [WP 19 S 47WH 11 GEWB 209]State 1 occurs in all arachnid orders except Palpigradi(Shultz 1990) The proposal that lsquoprimitiversquo slit sen-silla were present in Eurypterida (Dunlop amp Braddy1997) appears to be based on the over-interpretation ofa comparatively large notch that occurs in Baltoeu-rypterus at the terminus of the tibia (= podomere 7)(see also Edgecombe et al 2000 Giribet et al 2002)The neural construction of slit sensilla like function-ally similar campaniform sensilla of hexapods is sim-ilar to that of trichoid sensilla (Chapman 1998Klompen 2000) Thus these cuticular stress receptorsmay represent modified bases of sensory setae a viewsupported by the replacement of setae by slit sensilladuring post-embryonic development in some Opilioac-ariformes (Klompen 2000)143 Trichobothria 0 absent 1 present [GEWB 213]Trichobothria are present in extant Scorpiones(Jeram 1998) Pseudoscorpiones (Chamberlin 1931)some endeostigmatid (Alycus Van der Hammen 1989)and prostigmatid Acariformes and most Oribatida(Lindquist 1984) They occur on the ventral surface ofthe pedal femora in the opilioacariform Siamacarus(Leclerc 1989) They are also present in Araneae (Foe-lix 1996) Amblypygi (Weygoldt 2000) Schizomidaand Thelyphonida (Hansen amp Soslashrensen 1905) Theyare apparently absent in all non-arachnid chelicer-ates Solifugae Ricinulei Opiliones (Reissland ampGoumlrner 1985) and Parasitiformes (Acari) (Lindquist1984)144 Tibial trichobothria with 2-1-1-1 pattern onappendages IIIndashVI (= arachnid legs 1ndash4) 0 absent 1present - inapplicable due to absence of trichobothria(143) [S 48 WH 87 GEWB 88]

PHYLOGENY OF THE ARACHNID ORDERS 261

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs only in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905) Note that Shultz (1990)erroneously described this character as a 2-2-1-1 pat-tern and that this error was repeated by Giribet et al(2002) and described as 2-1-1 by Wheeler amp Hayashi(1998)145 Paired trichobothria on dorsal surface ofprosoma 0 absent 1 present - inapplicable due toabsence of trichobothria (143)State 1 is an apparent synapomorphy of Acariformesand is present in all acariforms included in this study(Alberti amp Coons 1999)146 Malleoli 0 absent 1 present [WP 45 GEWB 96]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)147 Pectines 0 absent 1 present [WP 20 WH 38GEWB 120]State 1 is a unique synapomorphy of Scorpiones148 Intercheliceral median organ 0 absent 1present [GEWB 212]State 1 is a unique synapomorphy of Palpigradi(Roewer 1934)149 Tarsal organ on appendage of postoral somite III(= arachnid leg 1) (= Hallerrsquos organ) sensilla containedwithin a cuticular depression on the superior surface ofthe tarsus of appendage III (= arachnid leg 1) 0absent 1 present [GEWB sim100]Klompen (2000) has noted that State 1 occurs in Opil-ioacariformes Parasitiformes (except Mesostigmata)and Ricinulei where it also occurs on leg 2 (Talaricoet al 2005) State 1 occurs on all legs in Araneae andperhaps Scorpiones (Foelix 1985)150 Tarsal organ on appendage of postoral somite IV(= arachnid leg 2) 0 absent 1 present (See 149)

REPRODUCTION

151 Gonads 0 primarily prosomal 1 primarilyopisthosomalState 0 is limited to Xiphosura (Sekiguchi 1988)State 1 occurs throughout Arachnida (Millot1949a)152 Ladder-like opisthosomal gonadsaccessoryglands (see 153) 0 absent 1 present [WH sim37 GEWBsim158]Giribet et al (2002) followed Wheeler amp Hayashi(1998) who followed Clarke (1979) in coding gonads asreticulate (Xiphosura) ladder-like (Scorpiones Thely-phonida Schizomida) or lsquosaccularrsquo (all remainingArachnida) However comparisons between Xipho-sura and Arachnida are problematic given that thexiphosuran gonads are primarily prosomal and thoseof arachnids are primarily opisthosomal (151) Fur-ther the reticulate pattern in Limulus (Xiphosura)but not other extant xiphosurans contains a distinctlyladder-like component The lsquosaccularrsquo state is probably

artificial as it encompasses a wide variety of pairedand unpaired structures153 Male gonads in two distinct parts one producingsperm and another (tubular gland) producing a holo-crine secretion similar to degenerate sperm 0 absent1 presentState 1 occurs in Thelyphonida Schizomida andAmblypygi although the holocrine material is pro-duced by ventral organs in Amblypygi and dorsalorgans in Thelyphonida and Schizomida (Alberti2005)154 Number of gonopores 0 two 1 oneExtant xiphosurans have two small genital openingson the base of the genital telopodite and all extantarachnids have a single opening The condition inEurypterida is not known Clarke amp Ruedemann(1912) located a pair of openings near the base of themedian organ (161) that are the outlets of the lsquohornorgansrsquo but it is unclear whether these are genitalducts or accessory structures Braddy amp Dunlop (1997)have developed numerous speculations about thesestructures and extended their arguments far beyondthe available evidence155 Genital opening (ie gonopore or gonostome)appearing to open in prosomal region (ie between legcoxae or anterior to posterior carapacal margin) 0absent 1 present [WP sim50 WH 26 GEWB 166]The genital opening in Euchelicerata is located onpostoral somite VIII (= opisthosomal somite 2) but ithas shifted anterior to the posterior margin of thecarapace or between the coxae of the last pair of legsin most Scorpiones (but not in PalaeoscorpiusKjellesvig-Waering 1986) and Opiliones The geni-tal opening occurs near or anterior to the last coxaein the opilioacariform and parasitiform Acari repre-sented here It is variable in Acariformes but islocated posterior to the coxae in all representativetaxa156 Ovipositor 0 absent 1 present [WP sim51 S sim60WH sim91 GEWB sim172]An ovipositor with a trilobed terminus is an apparentgroundplan character of Oribatida (Lindquist 1984Alberti amp Coons 1999) An ovipositor is also present inOpilioacariformes and Opiliones (Van der Hammen1989)157 Stalked spermatophore attached to substratum0 absent 1 present [S 57]State 1 occurs in Scorpiones PseudoscorpionesAmblypygi Thelyphonida and Schizomida The mech-anism of sperm transfer is unknown in PalpigradiOpilioacariformes and Holothryrida158 Male turns posterior end toward female duringmating behaviour 0 absent 1 presentState 1 occurs in Amblypygi (Weygoldt 2000) The-lyphonida and Schizomida (Weygoldt amp Paulus1979)

262 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

159 Female grasps male opisthosoma during matingbehaviour 0 absent 1 present [WP 29 S 58 WH 19GEWB 188]State 1 occurs in Thelyphonida and Schizomida (Wey-goldt amp Paulus 1979) Evolution of this behaviour wasprobably facilitated by 158160 Penis 0 absent 1 present [WP sim51 GEWB 167]State 1 is a unique synapomorphy of Opiliones A truepenis occurs in Phalangida A clearly homologousstructure is present in Cyphophthalmi (Opiliones) andapparently functions in depositing a spermatophore inthe femalersquos genital chamber (Karaman 2005) A truepenis may occur in some mites but it does not appearto be a groundplan feature of any major group (Evans1992) The lsquopenisrsquo in Oribatida is really a spermatop-ositor it functions in construction of a spermatophore(Alberti amp Coons 1999)161 Median organ 0 absent 1 presentState 1 occurs throughout Eurypterida (Clarke amp Rue-demann 1912) and has been observed in diploaspididChasmataspidida (Dunlop 2002a also Loganamaras-pis Tetlie amp Braddy 2004)

SPERM MORPHOLOGY

162 Nucleus with manchette of microtubules 0absent 1 present [S 54 WH 89 GEWB 193]State 1 occurs in Araneae Amblypygi Thely-phonida Schizomida Ricinulei and gonyleptid Lani-atores (Opiliones) (Alberti 1995 Giribet et al2002)163 Axoneme 0 absent 1 present [WP sim49 S sim55WH sim25 GEWB 195]State 1 occurs in Xiphosura Scorpiones Cyphoph-thalmi (Opiliones) Pseudoscorpiones RicinuleiAraneae Amblypygi Thelyphonida and Schizomida(Alberti 1995)164 Coiled axoneme 0 absent 1 present - inappli-cable due to absence of axoneme (163) [WP 22 S sim55WH 13 GEWB 196]State 1 occurs in Pseudoscorpiones Ricinulei Ara-neae Amblypygi Thelyphonida and Schizomida(Alberti 1995)165 Microtubule arrangement in axoneme 0 9 + 0 19 + 1 2 9 + 2 3 9 + 3 - inapplicable due to absence ofaxoneme (163) [WP sim26 S sim56 WH sim16 GEWB 198](Alberti 1995)166 Helical or corkscrew shaped nucleus 0 absent 1present [GEWB sim204]State 1 occurs in Araneae Amblypygi ThelyphonidaSchizomida and certain Scorpiones (eg Hadrurus)(Alberti 1995)167 Vacuolated-type sperm 0 absent 1 present[GEWB 205]State 1 is unique to Opilioacariformes and Parasiti-formes (Acari) (Alberti 1995)

168 Sperm aggregates 0 absent 1 present [GEWB206] (Alberti 1995)

DEVELOPMENT

169 Yolk in early embryo 0 concentrated (centroleci-thal or telolecithal) 1 evenly distributed (isolecithal)[WH sim34 GEWB sim191]Wheeler amp Hayashi (1998) and Giribet et al (2002)coded this character as lsquo0 isolecithal or telolecithal 1centrolecithalrsquo based on information presented byYoshikura (1975) Their coding appears to highlightseparate conditions found in Scorpiones (ie isoleci-thal and telolecithal) rather than a property intrinsicto the character itself170 Embryonic nutrition other than yolk 0 absent 1present [WP sim49]State 1 occurs in extant Scorpiones (except buthidsand chaerilids) and Pseudoscorpiones (Weygoldt1969)171 Embryological growth zone 0 initiating segmentaddition within prosoma 1 initiating segment addi-tion posterior to prosoma [GEWB sim192]Giribet et al (2002) followed Dunlop amp Webster (1999)in stating that only Xiphosura and Scorpiones havestate 0 among Chelicerata However the last prosomalsomite develops from the growth zone in Ixodida(Anderson 1973 Evans 1992) A similar processapparently occurs in the acariform mites Tyroglyphus(Sarcoptiformes Astigmata) (Yoshikura 1975) andArchegozetes (Sarcoptiformes Oribatida) (Thomas ampTelford 1999) but not apparently in Tetranychus(Prostigmata) (Dearden Donly amp Grbic 2002)172 Eggsembryos maintained in external attachedbrood sac secreted by genital glands 0 absent 1present [WP sim46 S 59 WH 90 GEWB 208 = 219]State 1 occurs throughout Amblypygi SchizomidaThelyphonida and Pseudoscorpions Giribet et al(2002) followed Shear et al (1987) in homologizingspinneret-derived silken egg sacs of spiders with gen-ital-gland-derived brood sacs of Pedipalpi However asthese structures are not homologous in secretory ori-gin construction or composition (Shultz 1987) theyare not regarded here as homologous Giribet et al(2002) miscoded the character as absent in Pseudo-scorpiones and without justification in PalpigradiFemale palpigrades have well-developed glands asso-ciated with postgenital somites and an array of setaefusules and other structures (112) (Boumlrner 1902b Mil-lot 1942 Condeacute 1991a) which are consistent withsome sort of brood care and Palpigradi should there-fore be coded as unknown for this character173 Embryonic and early postembryonic lsquolateralrsquo orClaparegravede organs associated with coxa of postoralsomite IV (= arachnid leg 2) 0 absent 1 present [WHsim31 GEWB sim211]

PHYLOGENY OF THE ARACHNID ORDERS 263

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bilaterally paired rounded protuberances locatedbetween coxae of appendages of postoral somites IIIand IV (arachnid legs 1 and 2) in prelarval and larvalinstars appear to be a primitive feature of Acariformes(Lindquist 1984 Evans 1992 Alberti amp Coons 1999)It is known to be derived embryologically from thecoxa of leg 2 in the oribatid Archegozetes (Thomas ampTelford 1999) Structures apparently homologouswith the acariform Claparegravede organ are present inembryonic Amblypygi (Weygoldt 2000) and Thely-phonida (Yoshikura 1975) and in embryonic and earlypost-embryonic stages of Solifugae (Roewer 1934)where they are termed lateral organs The so-calledlsquolateral organrsquo of embryonic Xiphosura (174) occurs onthe lateral surface of the carapace and does not appearto be homologous with the coxa-associated structuressharing their name In fact the xiphosuran lateralorgan develops even when the coxa of postoral somiteIV has been excised (Sekiguchi 1988) In contrastWheeler amp Hayashi (1998) and Giribet et al (2002)followed Yoshikura (1975) in regarding all lsquolateralorgansrsquo as homologous and in miscoding this characteras absent in Acariformes Van der Hammen (1989)suggested that the sternal verrucae of Opilioacari-formes are homologous with the Claparegravede organ butit differs in gross structure and ontogenetic timing174 Embryonic lateral organ associated with cara-pace 0 absent 1 present [WH sim31 GEWB sim211]State 1 is known only from embryonic stages of extantXiphosura (Yoshikura 1975) In contrast to previousinterpretations this feature does not appear to behomologous with the lsquolateralrsquo or Clarapegravede organ ofcertain arachnids (173)175 Live birth 0 absent 1 presentState 1 occurs in extant Scorpiones and has evolvedseveral times in a few mites (Evans 1992)176 Hexapodal larva and 1ndash3 nymphal stages 0absent 1 present [WP 53 S sim61 WH 27 GEWB 186]State 1 occurs in Acari and Ricinulei (Lindquist1984)177 Hexapodal prelarva 0 absent 1 present [S sim61WH 47]State 1 occurs in Opilioacariformes (NeocarusKlompen 2000) and many Acariformes (Evans 1992)

EXCRETORYOSMOREGULATORY SYSTEM

178 Malpighian tubules 0 absent 1 present [WP 17S 62 WH 9 GEWB 153]State 1 occurs in all major arachnid groups except Pal-pigradi Pseudoscorpiones Opiliones Oribatida (Sar-coptiformes) and Prostigmata Within Acari State 1occurs in all major anactinotrichid lineages (Opilioac-ariformes Holothyrida Mesostigmata Ixodida) andsimilar structures are present in certain Astigmata(Evans 1992 Alberti amp Coons 1999)

179 Dorsomedian excretory organ 0 absent 1presentThis is a specialized excretoryosmoregulatory organformed by the postventriculus and proctodeum inProstigmata (Alberti amp Coons 1999)180 Adult coxal organ opening on or near coxa ofappendage III (= arachnid leg 1) 0 absent 1 present[S 64 WH 93 GEWB 90]State 1 occurs throughout Acari (Evans 1992 Albertiamp Coons 1999) Ricinulei (Pittard amp Mitchell 1972Legg 1976) Palpigradi (Millot 1942) AraneaeAmblypygi Thelyphonida and Schizomida (Buxton1913 1917) There is evidence that the second lsquoozo-porersquo of gonyleptids (Opiliones Laniatores) (Hara ampGnaspini 2003) is a persistent opening to the coxalorgan associated with leg 1 (Soslashrensen 1879) whichalso occurs in embryonic Phalangium (Opiliones Eup-noi) (Moritz 1959)181 Adult coxal organ opening on or near coxa ofappendage V (= arachnid leg 3) 0 absent 1 present [S63 WH 92 GEWB 89]State 1 occurs in extant Xiphosura (Yamasaki et al1988) Scorpiones mygalomorph (Buxton 1913 1917)and mesothele Araneae (Liphistius J Millot in Bris-towe 1932) State 1 occurs in basal Amblypygi(Charon Charinus Buxton 1913 1917) but degener-ates prior to the adult stage in higher groups(Phrynus Weygoldt 2000) Adequately preservedeurypterids show a small submarginal opening on thecoxa of appendage V (Baltoeurypterus Selden 1981also Eurypterus Hughmilleria Clarke amp Ruedemann1912) and this is interpreted here as the opening to acoxal organ Giribet et al (2002) coded this characteras uncertain for Eurypterida182 Adult coxal organ opening on coxa of appendageII (= arachnid palp) 0 absent 1 present [GEWB 65]Buxton (1913 1917) noted that the coxal organ ofSolifugae opens on the prolateral surface of the palpalcoxa The proximal end of the duct leading to the ori-fice is associated with glands and apparently theglandular secretion andor coxal fluid serve as saliva(Alberti 1979) (see also 14 15)183 Genital papillae 0 absent 1 present - inappli-cable coded only for those taxa with eversible lsquoappend-agesrsquo (111)The genital papillae are associated with the genitalopening in postlarval Acariformes (Evans 1992Alberti amp Coons 1999) These structures are probablyserially homologous with the prosomal Claparegravedeorgan (173) and both structures probably function inwater and ion regulation

DIGESTIVE SYSTEM

184 Ingestion 0 solid food 1 primarily liquid foodwith or without preoral digestion [WP sim16 WH sim8GEWB sim218]

264 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Extant xiphosurans ingest solid food and members ofmost extant arachnid orders ingest fluids Opilionesand Acari are the principal exceptions The phalangidOpiliones ingest solids but Shultz (2000 unpublobserv) has shown that the precerebral pharyngealapparatus of Cyphophthalmi is very similar to that ofthe fluid-feeding Scorpiones and differs substantiallyfrom phalangid Opiliones Examination of gut con-tents of the cyphophthalmids Siro acaroides S exilisand Chileogovea oedipus have failed to reveal solidparticles typically found in the guts of phalangids(Shultz unpubl observ) the character is coded asunknown for Cyphophthalmi Within Acari consump-tion of particles has been documented in Opilioacari-formes (Neocarus Van der Hammen 1989unidentified Walter amp Proctor 1998) Holothyridanon-parasitic Sarcoptiformes (Evans 1992 Walter ampProctor 1998 Alberti amp Coons 1999) and many endeo-stigmatids including the nematophagous Alycusroseus (Walter 1988)185 Mouth 0 directed posteroventrally 1 directedanteroventrally [S 9 WH 55 GEWB 28]State 1 occurs throughout Arachnida (Shultz 1990)186 Oral cavity dilated by muscles arising from coxaeand constricted by large circular sphincter 0 absent1 presentState 1 occurs in extant Xiphosura (Manton 1964Scholl 1977 Shultz 2001)187 Palate plate 0 absent 1 present [GEWB 159]State 1 is a unique synapomorphy of Araneae (Dunlop1994)188 Lateral walls of epistome broadly fused to medialwalls of palpal coxae opposite sides connected by well-developed transverse epistomal muscle 0 absent 1presentState 1 occurs throughout Opiliones and Scorpiones(Shultz 2000 unpubl observ)189 Epistome with a pair of lateral arms that projectsposteriorly into the prosoma on either side of the phar-ynx 0 absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)190 Epistome with four pairs of suspensor musclesattaching to the carapace 0 absent 1 presentState 1 occurs in extant Scorpiones (Lankester et al1885 Vyas 1970 Shultz unpubl observ) and appar-ently in Solifugae (Roewer 1934)191 Intercheliceral epipharyngeal sclerite 0 absent1 present [GEWB 31]State 1 occurs in Palpigradi (Boumlrner 1904) Trigo-notarbida (Palaeocharinus Dunlop 1994) Araneae(Firstman 1954 Marples 1983) Amblypygi (Mil-lot 1949b Shultz 1999) Schizomida (Millot1949e) Thelyphonida (Millot 1949e Shultz 1993)and Phalangida (Opiliones) (Shultz 2000 unpublobserv)

192 Epipharyngeal sclerite large projecting posteri-orly 0 absent 1 present - inapplicable due to absenceof sclerite (191) [GEWB 32]State 1 occurs in Amblypygi (Millot 1949b Shultz1999) Schizomida (Millot 1949e) Thelyphonida (Mil-lot 1949e Shultz 1993) some Araneae (Heptathela ndashcoded as Liphistius ndash and Hypochilus Marples 1983not Aphonopelma Firstman 1954) and perhaps Trig-onotarbida (Palaeocharinus Dunlop 1994)193 Dorsal dilator muscle of precerebral pharynxattaching to intercheliceral septum or associatedepipharyngeal sclerite (191) 0 absent 1 presentState 1 occurs in Palpigradi (Millot 1943) Araneae(Marples 1983) Amblypygi Schizomida and Thely-phonida (Millot 1949e Shultz 1993 1999)Extremely fine muscles have been documented insome Scorpiones (Shultz 2007) and in Leiobunum(Opiliones) (Shultz 2000) Lankester et al (1885)reported a muscle in Limulus that would appear tocorrespond to this muscle but the muscle does notexist (Manton 1964 Shultz 2001)194 Dorsal dilator muscle of precerebral pharynxattaching to dorsal surface of prosoma 0 absent 1present [GEWB 228]State 1 occurs in Palpigradi (Roewer 1934) and mostAraneae (Palmgren 1978 Aphonopelma Firstman1973) but not in Heptathela (coded as Liphistius) orHypochilus (Marples 1983)195 Lateral dilator muscle of precerebral pharynxattaching to endosternite 0 absent 1 present [GEWBsim226]State 1 is a unique feature of Araneae (Marples 1983)196 Lateral dilator muscle of precerebral pharynxattaching to lateral surface of epistomal processes 0absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)197 Lateral dilator muscle of precerebral pharynxattaching to medial process of coxa of appendage II(= arachnid palp) 0 absent 1 present [GEWB sim226]State 1 occurs in Amblypygi Thelyphonida and Schi-zomida (Boumlrner 1904 Shultz 1993 1999)198 Dilator muscle of precerebral pharynx andorpreoral cavity attaching to ventral surface of prosoma0 absent 1 present [GEWB 227]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) Araneae (Marples 1983) Palpigradi(Millot 1943) and Solifugae (Roewer 1934)199 Postcerebral pharynx 0 absent 1 present [WPsim31 S sim5 WH sim21 GEWB sim154]The cuticle-lined foregut passes through the centralnervous system in many arthropods That portion pos-terior to the CNS is here termed the postcerebralpharynx regardless of the details of its morphology orfunctional specialization A substantial postcerebralpharynx is known in extant Xiphosura (Lankester

PHYLOGENY OF THE ARACHNID ORDERS 265

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

et al 1885 Manton 1964 Yamasaki et al 1988)Scorpiones (Centruroides Hadrurus HeterometrusShultz 2007 also Androctonus Abd el-Wahab 1952)Solifugae (Millot amp Vachon 1949) Araneae Ambly-pygi Thelyphonida and Schizomida (Millot 1949andasheShultz 1993) Weygoldt amp Paulus (1979) applied theterm lsquopostcerebral pharynxrsquo only to the lsquosucking stom-achrsquo of Araneae and Amblypygi but the sucking stom-ach is here regarded as a complex of three characters199ndash201

200 Dilator muscle of postcerebral pharynx attach-ing to endosternite 0 absent 1 present - inapplicabledue to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Lankester et al1885 Manton 1964 Shultz 2001) Scorpiones (Cen-truroides Hadrurus Heterometrus Shultz 2007 alsoAndroctonus Abd el-Wahab 1952) Araneae (First-

man 1954 Palmgren 1978) Amblypygi (Millot1949b Shultz 1999) and apparently the palpigradeProkoenenia wheeleri (Rucker 1901) but not inEukoenenia miriabilis (Millot 1943 contra Boumlrner1904)201 Dilator muscle of postcerebral pharynx attachingto dorsal surface of prosoma 0 absent 1 present -inapplicable due to absence of postcerebral pharynx(199)State 1 occurs throughout Araneae (Millot 1949cFirstman 1954 Palmgren 1978) and Amblypygi(Charinus Millot 1949b Phrynus Shultz 1999)202 Crop and gizzard 0 absent 1 present - inap-plicable due to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Sekiguchi 1988)but not in extant Arachnida

228 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Palaeocharinus from the Rhynie Chert (Scotland)(Dunlop 1994 Fayers Dunlop amp Trewin 2004)

Araneae (spiders) This is a very large order (110 fam-ilies ~3600 genera ~39 000 extant spp) (Platnick2005) represented here by one genus from each of thethree principal extant lineages (Platnick amp Gertsch1976 Coddington amp Levi 1991 Coddington 2005)Liphistius (Mesothelae) Aphonopelma (Mygalomor-phae) and Hypochilus (Araneomorphae) The sampleis small with respect to known diversity but the basalphylogeny and relevant groundplan states of the orderare well established and states derived from the rep-resented taxa are consistent with them

Amblypygi (whipspiders) This is a small extant ter-restrial order (five families 17 genera 136 spp)(Harvey 2002) and was represented in the matrix bytwo fairly well-studied genera Charinus and Phry-nus Charinus retains several features that appear tobe plesiomorphic for the order (eg adult pedal pul-villi coxal glands associated with leg 3 eversiblevesicles) Weygoldt (1996) has examined the phylog-eny of the order and has provided an important sum-mary of morphology and general biology (Weygoldt2000)

Thelyphonida or Uropygi ss (whipscorpions vinega-roons) Extant whipscorpions (16 genera 106 spp)(Harvey 2002) were represented here by one well-studied species Mastigoproctus giganteus althoughadditional information was drawn from Typopeltis andThelyphonus The morphology of the group is highlyconserved there have been no modern studies ofintraordinal phylogeny A controversial fossil speciesfrom the late Carboniferous Proschizomus (Dunlop ampHorrocks 19951996) was also included but less well-preserved Carboniferous forms were not

Schizomida This small order (two extant families~34 genera ~200 spp) (Harvey 2002) is widelyregarded as the sister group of Thelyphonida Thetaxon sample included one representative from eachextant family Protoschizomus (Protoschizomidae) andStenochrus (Hubbardiidae) especially S portoricen-sis There are three recent (Pliocene) fossil speciesall from the same locality these were not includedhere

Ricinulei The extant ricinuleids (three genera 55spp) (Harvey 2002) were represented by two generaCryptocellus and Ricinoides Detailed studies of grosscuticular anatomy and post-embryonic developmentare available for a representative of each genus (Cryp-tocellus Pittard amp Mitchell 1972 Ricinoides Legg1976) and some information on internal anatomy isavailable for Ricinoides (Millot 1945) and this hasbeen extrapolated to Cryptocellus in the matrix There

are two basic fossil types (Selden 1992) one resem-bling modern taxa and another with a unique opistho-soma that superficially resembles the closed elytra of abeetle (ie curculioids) Two fossils from the formergroup were included Terpsicroton and Poliochera asthey are reasonably well preserved and show impor-tant characters (eg two pairs of eyes or evidence ofopisthosomal diplosegmentation) not expressed inextant forms No curculioid ricinuleids were includedas they appear to offer no additional information rel-evant to resolving ordinal relationships

Opilioacariformes (= Opilioacarida Notostigmata)The opilioacariform mites (nine genera 20 spp) (Har-vey 2002) are generally regarded as plesiomorphicAcari and are fairly conserved in their morphologyThe group was represented by two species Neocarustexanus and Siamacarus withi N texanus has a typi-cal opilioacariform morphology and its external anat-omy has been particularly well studied (eg Van derHammen 1989 Klompen 2000) S withi differs frommost other opilioacariforms in having trichobothriaand three rather than two pairs of lateral eyes (the lat-ter also in Paracarus) features that are potentiallysignificant for assessing ordinal relationships Someinformation on internal morphology of Opilioacaruswas taken from With (1904) and extrapolated to Neo-carus and Siamacarus

Parasitiformes This large group of mites comprisesthree clades Holothyrida Ixodida (ticks) and Mesos-tigmata (= Gamasida) Holothyrida (three familiesfive genera lt 30 spp) (Walter amp Proctor 1999) wasrepresented here by two species Australothyrus ocel-latus and Allothyrus constrictus Ixodids (three fami-lies 22 genera ~860 spp) (Kierans amp Robbins 1999)were represented by Amblyomma and Argas andmesostigmatids (~70 families ~10 000 spp) were rep-resented by Glyptholaspis confusa

Acariformes (= Actinotrichida) This is a very largediverse group of mites that includes the Endeostig-mata Sarcoptiformes and Prostigmata Endeostig-mata is an apparently paraphyletic assemblage of~ten families A morphologically generalized speciesAlycus roseus (Bimichaeliidae) was included herewith character states extracted largely from Van derHammen (1989) Sarcoptiformes (gt 200 families) iseffectively synonymous with Oribatida with Astig-mata being a large derived lsquooribatidrsquo clade (Norton1998 Maraun et al 2004) The group was representedby an early divergent plesiomorphic genus Palaeac-arus (Palaeosomata) and a more derived and well-studied genus Archegozetes (especially A longisetusand A magnus) The Prostigmata (~120 families~7000 spp) (Walter amp Proctor 1999) was representedby two genera Allothrombium and Microcaeculus

PHYLOGENY OF THE ARACHNID ORDERS 229

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Pseudoscorpiones The pseudoscorpions (24 families425 genera ~3200 spp) (Harvey 2002) were repre-sented in the matrix by four terminal genera Chtho-nius (Chthoniidae) was drawn from the basallydivergent superfamily Chthonioidea (Harvey 1992)Feaella (Feaellidae) represented the enigmatic super-family Feaelloidea which has an unusual combinationof characters and may be an early divergent lineageclose to Chthonioidea (Harvey 1992) Neobisium (Neo-bisiidae) is morphologically intermediate between theearly divergent families and the higher families rep-resented by Chelifer (Cheliferidae)

Solifugae (= Solpugida) (sun spiders) The order issmall (~1000 spp) but the absence of modern phylo-genetic treatments precludes meaningful estimates ofgenera and families (Harvey 2002) The taxon sampleincluded two genera one from the New World Eremo-costa and one from the Old World Galeodes

Opiliones (harvestmen) Extant harvestmen (~25 fam-ilies ~500 genera ~6000 spp) are divided into twosuborders Cyphophthalmi and Phalangida Cyphoph-thalmids were represented here by Chileogovea oedi-pus (Petallidae) which is currently under study by theauthor and Cyphophthamus duricorius (Sironidae)which is probably the most well-studied member of thesuborder (eg Janczyk 1956) Phalangida includesthree major groups Laniatores Eupnoi and Dyspnoibut there is disagreement as to whether Dyspnoi is thesister group to Laniatores or to Eupnoi (Shultz ampRegier 2001 Giribet et al 2002) Laniatores was rep-resented by Sclerobunus and Gonyleptes and Eupnoiby Leiobunum and Caddo

Scorpiones (scorpions) The extant scorpions (16 fam-ilies 155 genera 1279 spp) (Fet et al 2000) wererepresented by three taxa Centruroides vittatus(Buthidae) Hadrurus arizonensis (Iuridae) and Heter-ometrus spinifer (Scorpionidae) Buthidae is widelyregarded as the sister group to other extant lineagesand similarities among the represented terminals arelikely to be ground plan features of extant scorpionsgenerally The morphology of fossil scorpions is sub-stantially more diverse (Kjellesvig-Waering 1986)but their phylogenetic relationships are unclear (butsee Jeram 1998) The fossil taxa used in this study(Proscorpius Stoermeroscorpio Palaeoscorpius Pre-arcturus) were chosen for quality of preservation andor presence of a phylogenetically significant constella-tion of characters

CHARACTER CODING

Most character states were determined from directobservation the primary literature and authoritativereviews In some cases states were assigned to termi-

nals based on observations from related species asnoted in the Appendix In the matrix state lsquo-rsquo indicatesthat the character is inapplicable because the taxonlacks a more general character For example if a taxonlacks eyes then special features of the eyes (numberposition retinal configuration etc) are inapplicable tothat taxon A lsquorsquo indicates that the state is unknown oruncertain An entry with multiple states (eg 01 345) should be treated as an ambiguity code not a poly-morphism it indicates that two or more interpreta-tions of homology are applicable and that assignmentwas established analytically by character concor-dance Although lsquo-rsquo lsquorsquo and lsquo01rsquo are analyticallyidentical when applied to binary characters theyprovide information about the empirical status of thecharacter state in specific taxa Characters are citedthroughout the text as italicized numbers in parenthe-ses and are discussed in the Appendix

PHYLOGENETIC ANALYSIS

Phylogenetic analysis of the full data matrix wasperformed using the program Tree Analysis usingNew Technologies (TNT) ver 1 (Goloboff Farris ampNixon 2000) using lsquotraditionalrsquo search based on1000 replicates using TBR branch swapping Resultswere compared to those obtaining using the ratchetalgorithm (Nixon 1999) to determine any differencedue to analytical approach Nodal support for theminimal-length topology was evaluated by boot-strap (Felsenstein 1985) and Bremer support(Bremer 1994) Bootstrap analysis was conducted inTNT and based on 1000 pseudoreplicates each anal-ysed by ten random-addition replicates using TBRbranch swapping A nexus file containing the result-ing 1000 trees was imported into PAUP ver 10(Swofford 2002) to obtain bootstrap frequenciesBremer support was determined in TNT by con-straining specific nodes in the minimal-length topol-ogy and then determining the shortest tree that didnot recover the specified clade The difference inlength between the unconstrained and constrainedminimal-length trees is the Bremer support Theeffect of homoplasy on results was explored by con-ducting implied weights analysis (Goloboff 1993) inTNT Six analyses were conducted each with con-stant of concavity (k) set to a different integer valueof 1ndash6 where 1 is weighted most severely againsthomoplasious characters Each implied weightsanalysis was conducted using lsquotraditionalrsquo searchbased on 1000 replicates using TBR branch swap-ping The same procedures were used in analysing amatrix that included only extant taxa Howevercharacters rendered uninformative by removal offossil taxa (ie 49 98 99 100 113 114 125 161)were excluded prior to analysis

230 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

COMPARISONS OF ALTERNATIVE HYPOTHESES

Phylogenetic hypotheses proposed in the recent liter-ature were also evaluated including those thatattempted to resolve arachnid phylogeny completely(Fig 2) their hypothesized subclades (Fig 3) andhypotheses that proposed only specific nodes (Figs 34) The fully resolved topologies of Weygoldt amp Paulus(1979) Van der Hammen (1989) Shultz (1990)Wheeler amp Hayashi (1998) and Giribet et al (2002)were compared with the optimal topology using theTempleton test (Templeton 1983) as implementedin PAUP Internal relationships of multisampledorders were constrained to match those of the optimaltopology unless the original authors explicitlyfavoured an alternative Node-specific hypotheseswere evaluated by determining the frequency withwhich the node was recovered in bootstrap analysisThey were also evaluated by constraining parsimonyanalysis in TNT to recover the shortest tree contain-ing each specific node and then assessing the effect onrelative tree length and overall resolution takingnote of effects on otherwise stable or well-supportedclades The entire matrix was used to assess the twocases where node-specific hypotheses involved fossiltaxa (Fig 4)

RESULTS

EXTANT TAXA

Unweighted analysis of extant taxa produced two min-imal-length trees (length 383 CI 057) (Fig 5A) withconflicts limited to relationships within Pseudoscorpi-ones Multiply sampled orders were recovered asmonophyletic with high nodal support as measured bybootstrap percentage (BP gt 80) except Acari (BP 30)There were few well-supported interordinal groupsexcept Uropygi (BP 100) Pedipalpi (BP 100) Tetra-pulmonata (BP 93) and ArachnidaXiphosura (BP100) All other interordinal relationships were recov-ered with BP less than 60

Results from implied weights analysis (IWA) indi-cated a significant effect of homoplasy on resolution ofthe deepest interordinal relationships although theinterordinal clades Stomotheca Haplocnemata Acar-omorpha Tetrapulmonata Pedipalpi and Uropygiwere recovered under all values of k IWA with k = 1 (1tree best score = 5124) recovered Megoperculata(= Palpigradi + Tetrapulmonata) and Micrura(= Megoperculata + Acaromorpha) with Micrura beingthe sister group to Haplocnemata (= Pseudoscorpiones+ Solifugae) and with Stomothecata (= Opiliones +Scorpiones) being the sister group to all other arach-nids (Fig 5B) With k = 2 (1 tree best score = 3807)and k = 3 (1 tree best score = 3054) Megoperculatawas recovered and Stomotheca remained the sister

group to other arachnids but the relative positions ofHaplocnemata and Acaromorpha changed (Fig 5C)IWA using k = 4 (1 tree best score = 2559) k = 5 (1tree best score = 2205) and k = 6 (1 tree bestscore = 1938) recovered topologies consistent with theminimal-length topology derived from unweightedcharacters (Fig 5A)

EXTANT AND FOSSIL TAXA

Unweighted analysis of extant and fossil taxa pro-duced 18 minimal-length trees (length 426 CI 056)(Fig 6A) with conflicts limited to relationships withinPseudoscorpiones Scorpiones and Araneae Multiplysampled orders were again recovered as monophyleticwith high nodal support (BP gt 80) except Acari whichwas reconstructed as diphyletic with Anactinotrichidabeing the sister group to Ricinulei (BP 33) Therewere no well-supported interordinal groupings (ieBP gt 80) except Uropygi (BP 89) Pedipalpi (BP 92)and Arachnida (BP 90)

Results from IWA indicated a significant effect ofhomoplasy on resolution of the deepest interordinalrelationships in Arachnida although StomothecataHaplocnemata Acaromorpha Cryptognomae Pantet-rapulmonata Tetrapulmonata Schizotarsata Pedi-palpi and Uropygi were recovered under all values ofk IWA with k = 1 (7 trees best score = 5675) recov-ered Megoperculata and Micrura with Micrura beingthe sister group to Haplocnemata Stomothecata wasnot recovered in the strict consensus (Fig 6B) Withk = 2 (3 trees best score = 4243) IWA recovered thetopology shown in Fig 6C with Palpigradi being thesister group to all other arachnids The strict consen-sus topologies reconstructed using k = 3ndash6 were iden-tical to those obtained with k = 2 (k = 3 3 trees bestscore = 3419 k = 4 3 trees best score = 2874 k = 53 trees best score = 2483 k = 6 3 trees bestscore = 2188)

COMPARISONS OF ALTERNATIVE HYPOTHESES

The hypothesis of Weygoldt amp Paulus (1979) (Fig 2)was 398 steps (15 steps longer than the unconstrainedminimal-length tree) Van der Hammen (1989) was400 steps (17 steps longer) Shultz (1990) was 399steps (six steps longer) Wheeler amp Hayashi (1998)was 391 steps (eight steps longer) and Giribet et al(2002) was 394 steps (11 steps longer) Templetontests conducted at the 005 significance level rejectedthe hypothesis of no difference between the optimaltopology and those of Weygoldt amp Paulus (P = 00002)Van der Hammen (P = 00011) Wheeler amp Hayashi(P = 00018) and Giribet et al (P = 00343) The Shultztopology was not significantly different (P = 01444)from the optimal topology

PHYLOGENY OF THE ARACHNID ORDERS 231

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 2 Fully resolved phylogenetic hypotheses of extant euchelicerate groups proposed in the recent literature Note thesimilarity in the topologies of the parsimony-based analyses by Shultz (1990) Wheeler amp Hayashi (1998) and Giribet et al(2002) The Giribet et al topology is based on neontological data (morphology and molecules) and the original lsquoROOTrsquo maybe an artefact from use of the highly divergent pycnogonids as an outgroup

Xip

hosu

raS

corp

ione

sO

pilio

nes

Ara

neae

Am

blyp

ygi

Uro

pygi

Ric

inul

eiA

nact

inot

richi

daS

olifu

gae

Pse

udos

corp

ione

sP

alpi

gra

diA

carif

orm

es

Xip

hosu

raS

corp

ione

sA

rane

aeA

mbl

ypyg

iU

ropy

giP

alpi

gra

diS

olifu

gae

Pse

udos

corp

ione

sO

pilio

nes

Ric

inul

eiA

cari

Lipoctena

Arachnida

Apulmonata

Holotracheata

Cryptoperculata

Acaromorpha(ldquoAcarinomorphardquo)

Tetrapulmonata(rdquoMegoperculatardquo)

LabellataHaplo-cnemata

Weygoldt amp Paulus (1979)

Labellata

Tetrapulmonata(rdquoArachnidardquo)

Haplocnemata(rdquoApatellatardquo)

Van der Hammen (1989)

Xip

hosu

raP

alpi

gra

diA

rane

aeA

mbl

ypyg

iU

ropy

giR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Pedipalpi Haplocnemata

Novogenuata

Dromopoda

Shultz (1990)

Acaro-morpha

Micrura

Xip

hosu

raP

alpi

gra

diU

ropy

giA

mbl

ypyg

iA

rane

aeR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Labellata Haplocnemata

Novogenuata

Dromopoda

Wheeler amp Hayashi (1998)

Acaro-morpha

Micrura

Xip

hosu

raA

cari

Pal

pig

radi

Ric

inul

eiA

rane

aeA

mbl

ypyg

iU

ropy

giO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Tetrapulmonata

PedipalpiHaplocnemata

Novogenuata

Dromopoda

Giribet et al (2002)

ROOT

Micrura

Crypto-gnomae

Sternocoxata

Epimerata

RostrosomataMyliosomata

Neosternata

232 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 3 Consensus trees produced by parsimony analysis of neontological data constrained to produce relationships pro-posed in the recent literature The constrained (target) node is indicated by a black dot and the taxa encompassed by theconstraint are enclosed in a box Numbers below each tree represent the number of minimal-length constrained treeslength of minimal-length trees difference in the length of the unconstrained minimal-length tree and the constrained min-imal-length tree and percentage unconstrained bootstrap trees in which the target node was recovered respectively Thesetrees indicate the effect on branch length imposed by specific hypotheses and impact of constraining target nodes on non-target nodes

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Lipoctena4 385 2 lt5

Xi

Sc

Op

Ps

So

Pa

Ur

Am

Ar

Ri

An

Ac

Labellata6 397 14 lt5

Xi

Sc

Ar

Am

Ur

Op

Pa

Ps

So

Ri

An

Ac

Apulmonata2 390 7 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

So

Ri

An

Ac

Holotracheata5 390 7 lt5

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptoperculata2 388 5 lt5

Xi

Sc

Op

Pa

So

Ps

Ar

Am

Ur

Ri

An

Ac

Rostrosomata6 384 1 25

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Neosternata3 388 5 lt5

Xi

Sc

Op

Ar

Am

Ur

Ri

An

Ps

So

Pa

Ac

Coxisternata2 388 5 lt5

Epimerata7 387 4 11

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptognomae6 384 1 24

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Micrura3 385 2 7

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Megoperculata3 384 1 28

Xi

Sc

Ps

So

Op

Pa

Ar

Am

Ur

Ri

An

Ac

Novogenuata10 388 5 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

Ri

So

An

Ac

Solifugae + Acari4 388 5 lt5

Shultz (1990) Alberti amp Peretti (2002)

Van der Hammen (1989)

Weygoldt amp Paulus (1979)

Xi

Ar

Am

Ur

So

Ps

Sc

Op

Ri

Pa

An

Ac

PHYLOGENY OF THE ARACHNID ORDERS 233

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Results from analyses of node-specific hypothesesare summarized in Figures 3 4 In most cases pro-posed relationships were recovered in fewer than 5of bootstrap pseudoreplicates The notable exceptionswere Rostrosomata (25) Cryptognomae (24) andMegoperculata (28) and the strict-consensus con-strained minimal-length tree for each was only onestep longer than the unconstrained optimal tree Inmost cases constrained topologies did not add manysteps most were less than five steps longer and nonewas greater than 14 steps longer than the un-constrained minimal-length tree It is noteworthyhowever that even constrained nodes that imposerelatively few extra steps sometimes supportedimprobable relationships or eliminated clades thatwere well supported or stable in the unconstrainedanalysis Specific examples and their implications arediscussed below

DISCUSSION

RECENT ISSUES IN ARACHNID PHYLOGENY

Available morphological evidence consistently re-solves some interordinal relationships and fails to

resolve others (Fig 1) Continued progress depends onthe ability of arachnologists to discover new charac-ters and to assess the evidence critically Whilereviewing the recent arachnological literature severalaspects of phylogenetic practice emerged that seemedcounterproductive to both the perception and theactual rate of progress toward resolving arachnid phy-logeny Some of these are summarized here withspecific examples given in the remainder of theDiscussion and in the Appendix

There is a tendency to portray arachnid ordinal phy-logeny as more poorly resolved and contentious than isactually the case (Coddington et al 2004) Phyloge-netic hypotheses generated during different historicalperiods and using differing standards of evidence areoften cited as examples of current disagreement (egSelden 1993 Dunlop 1996 Selden amp Dunlop 1998Wheeler amp Hayashi 1998) In fact recent parsimony-based analyses of morphology have tended to convergeon topologies with internal structures congruent withthose found here (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) (Fig 2) Arachnidphylogeny is not fully resolved especially at its deep-est levels (Fig 1) but this does not mean that allaspects of arachnid phylogeny are controversial orpoorly supported by the available evidence

Matrices are sometimes constructed by uncriticallsquorecyclingrsquo of erroneous or problematic charactersbased on diverse secondary sources (Jenner 2001)Conclusions derived from mixtures of valid invalidspeculative and redundant characters are sometimesportrayed as the phylogenetic signal provided by mor-phology Data recycling can perpetuate errors (see 1320 39 52 136 140 144 152 169 171ndash174) legiti-mize speculations (see 13 32 63 95) or create dupli-cate or non-independent characters (see 13 30 70 77172) The lsquolateral organrsquo is a notable exampleYoshikura (1975) equated the embryonicearly post-embryonic lsquolateral organsrsquo of Amblypygi Thelyphonidaand Solifugae with the dissimilar lsquolateral organrsquo ofXiphosura but failed to note the very similar Cla-paregravede organ of Acariformes (see 173 174) This cod-ing was recycled by Wheeler amp Hayashi (1998) andthen by Giribet et al (2002) Error is probably inevi-table when assembling a large matrix from morphol-ogy including the one presented here but this can beminimized by making original observations and byconsulting primary sources

Some workers advocate weighting characters a pri-ori on the basis of structural or functional complexity(eg Kraus 1998 Dunlop amp Braddy 2001) and dismissphylogenetic conclusions derived from equal-weightsparsimony In short these workers criticize parsimonyfor emphasizing data quantity over data quality Thiscriticism ignores the intense debate in systematic biol-ogy that eventually led to widespread adoption of

Figure 4 Consensus trees produced by parsimony analy-sis of the full data matrix constrained to produce relation-ships proposed in the recent literature See legend toFigure 3 for details

Xi

Eu

Sc

Op

Pa

Ac

Ri

An

So

Ps

Tr

Ar

Ha

Am

UrEurypterida slat + Scorpiones12 435 9 lt5

Xi

Eu

Sc

Op

Pa

Ac

An

So

Ps

Ri

Tr

Ar

Ha

Am

UrTrigonotarbida+ Ricinulei28 434 8 lt5

Dunlop amp Braddy (2002)et al

Dunlop (1996)

234 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 5 Results from analysis of neontological data A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology AFor B and C relationships within terminal clades are the same as those shown in A

LimulusTachypleusProkoeneniaEukoeneniaCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAllothyrusAustralothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesLiphistiusAphonopelmaHypochilusCharinusPhrynusStenochrusProtoschizomusMastigoproctusCyphophthalmusChileogoveaCaddoLeiobunumSclerobunusGonyleptesCentruroidesHadrurusHeterometrusChthoniusFeaellaNeobisiumCheliferEremocostaGaleodes

10031

10018451

501561

733

723

944

633

10010

832

901

943

903

522

341

1007

1008

Acaromorpha

Haplocnemata

10012

Stomothecata

Opiliones

Scorpiones

Pseudo-scorpiones

Solifugae

Acariformes

Parasitiformes

Opilioacariformes

Ricinulei

Xiphosura

Palpigradi

Thelyphonida

Schizomida

Araneae

Amblypygi

10011

936853

100141005

1009

Tetrapulmonata Pedipalpi

Uropygi

692

522572

968

331

603

81

511

161

101

Acari

Dromopoda

Arachnida

A

Xiphosura

Stomothecata

Haplocnemata

Acaromorpha

Palpigradi

TetrapulmonataMicrura

Megoperculata

BXiphosura

Stomothecata

Acaromorpha

Haplocnemata

Palpigradi

TetrapulmonataMegoperculata

C

PHYLOGENY OF THE ARACHNID ORDERS 235

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 6 Results from analysis of the full data matrix A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology CFor B and C relationships within terminal clades are the same as those shown in A

WeinberginaLimuloidesEuproopsTachypleusLimulusStylonurusBaltoeurypterusChasmataspisDiploaspisOctoberaspisCyphophthamusChileogoveaCaddoLeiobunumSclerobunusGonyleptesProscorpiusPalaeoscorpiusStoermeroscorpioPrearcturusCentruroidesHadrurusHeterometrusProkoeneniaEukoeneniaTerpsicrotonPoliocheraCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAustralothyrusAllothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesEremocostaGaleodesChthoniusFeaellaNeobisiumCheliferGilboarachnePalaeocharinusLiphistiusAphonopelmaHypochilusPlesiosiroCharinusPhrynusStenochrusProtoschizomusProschizomusMastigoproctus

631

893

923

984

984

671

341

671

986

10011

423

974

Xiphosura

Eurypterida sstr

Chasmataspidida

Opiliones

Scorpiones

Palpigradi

Ricinulei

Opilioacariformes

Parasitiformes

Acariformes

Solifugae

Pseudoscorpiones

Trigonotarbida

Araneae

Haptopoda

Amblypygi

Schizomida

Thelyphonida

842

251331

361311

552

832

882

925

913

573

561291

443

976

541

954

743644

732521

571

331

833

663591

966

573

71

221

905

832

351

Uropygi

Pedipalpi

Haplocnemata

Tetrapulmonata

Anactinotrichida

Cryptognomae

Acaromorpha

Stomothecata

Pantetrapulmonata

Arachnida

Eurypterida slat

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatProscorpiusStoermeroscorpioPalaeoscorpiusPrearcturusCentruroidesHadrurusHeterometrusOpilionesHaplocnemataRicinuleiAnactinotrichidaAcariformesPalpigradiPantetrapulmonata

Megoperculata

Micrura

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatPalpigradiStomothecataHaplocnemataPantetrapulmonataAcaromorpha

A

B

C

Schizotarsata

236 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

parsimony and instead advocates a return to thespeculative and subjective approaches of the late 19thand early 20th century that once threatened thescientific legitimacy of the discipline (Bowler 1996)Furthermore it misrepresents the properties of parsi-mony-based analyses Specifically characters withfunctional significance are readily encompassed byparsimony analyses and in fact characters derivedfrom locomotor systems (eg 46ndash94) have played animportant role in developing current ideas aboutarachnid phylogeny (eg Shultz 1989 1991) In addi-tion lsquocomplex charactersrsquo can be viewed as compositesof several characters such that morphological com-plexity is effectively weighted by the number of inde-pendent lsquosubcharactersrsquo it contains For example thelsquosucking stomachrsquo once considered a synapomorphy ofLabellata (= Araneae + Amblypygi Fig 3) is codedhere as a composite of three characters (199ndash201) TheLabellata hypothesis was not corroborated in thepresent analysis (Figs 1 3ndash5) and in fact was highlydisfavoured (Fig 3) but this result cannot be dis-missed as a failure to acknowledge the complexity ofthe character

Some workers support specific (target) clades withone or more similarities without exploring the impacton overall homoplasy or relationships among non-target clades Each character offers its own phyloge-netic hypothesis which may or may not be consistentwith relationships implied by other characters It isexceedingly rare for all characters to be perfectly com-patible in the phylogenetic hypotheses they supportand criteria such as parsimony have been developed todiscover those hypotheses that minimize the conflict-ing phylogenetic signals of different characters Eventhough character conflict (homoplasy) is a virtuallyinescapable phenomenon in comparative biology it isnot uncommon for workers to discover one or morecharacters and to promote their phylogeneticimplications over alternative hypotheses even thosethat otherwise appear to be well supported Thisapproach may have value in highlighting new dataand perspectives but accomplishes this by promotingthe erroneous impression that all aspects of arachnidphylogeny are so tenuous that a single character canfalsify even well-supported hypotheses Several exam-ples of this approach have appeared in the recentarachnological literature

For example Alberti amp Peretti (2002) argued thataflagellate spermatozoa (163) are a compelling syna-pomorphy for a Solifugae + Acari clade and dismissedsome of the characters that support Solifugae + Pseu-doscorpiones as having lsquodebatable valuersquo Yet theirproposal rejects two hypotheses that are consistentlyrecovered in recent phylogenetic analyses that isHaplocnemata (= Solifugae + Pseudoscorpiones) andAcaromorpha (= Acari + Ricinulei) (Figs 1 2) Rejec-

tion of Haplocnemata would require its presumed syn-apomorphies to be reinterpreted as homoplasiesincluding features of the chelicerae (18 19 20) pre-oral chamber (13 32) legs (12 48) and respiratorysystem (126) The same reason applies to Acaromor-pha and its synapomorphies It is noteworthy that aSolifugae + Acari clade was recovered in the presentanalysis in fewer than 5 of bootstrap pseudorepli-cates and that analyses constrained to recover thisclade were five steps longer than the minimal-lengthtree and favoured a problematic clade unitingRicinulei and Pseudoscorpiones (Fig 3)

In another example Dunlop (1996) proposed a closerelationship between Trigonotarbida and Ricinuleibased on two-segmented chelicerae (18) prosomandashopisthosoma coupling mechanism (96) diplotergites(100 101) and longitudinally divided opisthosomaltergites (115) However phylogenetic analyses con-strained to recover this relationship required eightadditional steps eliminated support for Acari and nec-essarily rejected Acaromorpha and forced its synapo-morphies to be reinterpreted as homoplasies Alberti(2005) proposed an interesting hypothesis for the evo-lution of male gonads in tetrapulmonates (153) butchose to accept Labellata (= Araneae + Amblypygi) indeveloping his argument over the much more well-supported Pedipalpi (= Amblypygi + Uropygi) (Shultz1999) a phylogenetic reconfiguration that was recov-ered here in fewer than 5 of bootstrap pseudorepli-cates increased tree length by a minimum of 14 stepsand resulted in the collapse of Stomothecata and Hap-locnemata (Fig 3) Many other examples can be cited

Promoting or defending a specific phylogenetichypothesis via lists of compatible synapomorphies is acommon but problematic approach By restrictingattention to the states of specific characters at one ortwo target nodes one can easily overlook the unin-tended impact of the hypothesis on phylogenetic sig-nal elsewhere and its effect on non-target clades Anode supported by a long list of synapomorphies mayseem convincing taken in isolation but may becomeless acceptable when its full phylogenetic implicationsare explored

SUMMARY OF INTERORDINAL ARACHNID CLADES RECOVERED IN THIS ANALYSIS

Arachnida Lamarck 1801Analyses consistently recovered Arachnida as a mono-phyletic group with high bootstrap support (Figs 1 45) Possible synapomorphies include the loss of thecarapacal pleural doublure (9) cardiac lobe (10) pedalgnathobases (52) and moveable endites (53) and thegain of aerial respiration (120) and an anteriorly oranteroventrally directed mouth (185) Some tradi-tional synapomorphies such as slit sensilla (142) and

PHYLOGENY OF THE ARACHNID ORDERS 237

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

fluid feeding (184) may have appeared later in arach-nid evolution but this can only be decided once theinternal phylogeny of Arachnida has been determined

Some workers regard many proposed arachnid syn-apomorphies as adaptations to terrestrial life andthereby link the hypothesis of arachnid monophyly tothe hypothesis of a single ancestral aquatic-to-terres-trial transition and arachnid polyphyly to multipletransitions (eg Selden amp Jeram 1989 Dunlop 1997Dunlop amp Webster 1999 Dunlop amp Braddy 2001)Given this line of reasoning the existence of appar-ently aquatic scorpions in the fossil record (Kjellesvig-Waering 1986 Jeram 1998) and the inference thatterrestrialization occurred late in scorpion evolutionthese workers conclude that character states support-ing arachnid monophyly are actually convergencesand do not necessarily support arachnid monophyly

However this approach to assessing phylogenetichypotheses is founded on overly simplistic assump-tions such as the ability of the investigator to discrim-inate unerringly between characters that existexclusively in aquatic organisms (including fossils)from those that occur exclusively in terrestrial organ-isms There also appears to be an assumption thathomoplasy can be generated through parallelism (iemultiple aquatic-to-terrestrial events) but not throughterrestrial-to-aquatic reversals Furthermore anassumed dichotomy between exclusively aquatic andterrestrial life histories in ancestral arachnids is sim-plistic as illustrated by the amphibious life cycles ofbasal vertebrates and pterygote hexapods In factthese examples demonstrate that there is no neces-sary inconsistency in basing a hypothesis of arachnidmonophyly on the derived terrestrial features of anamphibious ancestor whose descendants thencompleted terrestrialization once or several timesindependently or even returned to a fully aquaticexistence Workers who link the frequency and direc-tion of aquaticndashterrestrial transitions to the assess-ment of arachnid phylogeny do so by endowingthemselves with substantially greater insight thanseems prudent by ignoring the huge gaps in ourunderstanding of early arachnid evolution and bydenying to arachnids the evolutionary complexityknown to exist in other groups

Several palaeontologists have been particularlyactive during the past decade in proposing new char-acters with the stated goal of removing Scorpionesfrom Arachnida and erecting a Scorpiones + Eurypter-ida clade (Braddy amp Dunlop 1997 Dunlop amp Braddy1997 Dunlop 1998 Braddy et al 1999 Dunlop ampWebster 1999) Dunlop amp Braddy (2001) recentlysummarized this evidence and conducted aparsimony-based analysis of Xiphosura EurypteridaScorpiones Opiliones and Tetrapulmonata (but notHaplocnemata Acaromorpha or Palpigradi) using 33

morphological characters Their analysis produced atopology congruent with those generated here (Figs 15 6) including recovery of Stomothecata (= Scorpiones+ Opiliones) However they rejected this result as aproduct of lsquoempirical cladisticsrsquo because it gives thesame weight to prosomal characters that supportarachnid monophyly and to selected opisthosomalcharacters that support their favoured Eurypterida +Scorpiones clade namely a five-segmented metasoma(116) suppression of opisthosomal tergite 1 (95) lossof lamellate respiratory organs on the postgenitalsomite (122) Kiemenplatten (125) loss of respiratorylamellae on the genital segment (121) and a lsquonon-stainingrsquo exocuticle (but see Grainge amp Pearson 1966for evidence of this in Opiliones see Appendix for com-ments on the other characters)

A Eurypterida + Scorpiones clade was not favouredin the present analysis (Fig 6) the strict consensus ofminimal-length trees constrained to recover this clade(Fig 4) is nine steps longer than the tree recovered byanalysis without this constraint It is also noteworthythat Opiliones was consistently reconstructed as thesister group to Eurypterida + Scorpiones in the mini-mal-length constrained trees a provocative result thatwas probably unintended and unanticipated by sup-porters of the Eurypterida + Scorpiones Given thatDunlop amp Braddy (2001) reject equal weights parsi-mony as an arbiter of phylogenetic hypotheses theywould presumably dismiss these results as irrelevantto their argument just as they dismissed their ownparsimony-based results However if the Eurypterida+ Scorpiones hypothesis is to be credible it must beopen to evaluation and potential falsification usingobjective criteria and the subjective or intuitive a pri-ori weighting of characters advocated by Dunlop ampBraddy clearly does not qualify At present it isunclear how one would objectively evaluate Dunlop ampBraddyrsquos proposal with criteria compatible with thoseused in its original formulation For now the Euryp-terida + Scorpiones concept advocated by Dunlop ampBraddy may persist outside the mainstream of mod-ern systematic practice but it is increasingly problem-atic within it

Stomothecata nomnovOpiliones and Scorpiones were consistently recoveredas a monophyletic group The proposed name acknowl-edges a unique preoral chamber the stomothecaformed by coxapophyses of the palp and leg 1 (50)often with an auxiliary role played by the coxapophy-sis of leg 2 (51) In addition the epistome appears tohave been modified for adduction of the palpal coxaeThe lateral walls of the epistome are fused to themedial surfaces of the palpal coxae and the epistomallumen is spanned by a transverse muscle (188) which

238 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

apparently adducts the palpal coxae thereby constrict-ing the stomothecal chamber Scorpions and opilionsare also unique in having a pair of large epistomalarms projecting rearward into the prosoma andattaching to the endosternite (189) The epistomalarms provide attachment sites for pharyngeal dilatormuscles (196) and extrinsic muscles of anterior proso-mal appendages (eg 37) The chelicera is equippedwith a muscle that arises on the carapace and insertson the ventral margin of the second cheliceral article(23) There is an anteriorly placed genital opening(155)

Several notable similarities were found whilereviewing the literature but information from otherarachnid groups was considered too incomplete toallow them to be included in the matrix For exampleboth orders have apparent haemocytopoeitic organsassociated with major nerves of the anterior opistho-soma These are termed supraneural organs in scorpi-ons (Farley 1999) and perineural organs in opilions(Kaumlstner 1935) Haemocytes develop in the cardiacwall in spiders (Seitz 1972) and perhaps amblypygids(Weygoldt 2000) but haemocytopoeitic organs areunknown in most other arachnid groups Germ cellsdifferentiate very early during embryogenesis in bothscorpions and opilions relative to spiders (Moritz1957 Anderson 1973) Additional research is requiredto determine the phylogenetic utility of thesecharacters

Haplocnemata Boumlrner 1904Several workers recognize a close phylogenetic rela-tionship between Pseudoscorpiones and Solifugae(Weygoldt amp Paulus 1979 Van der Hammen 1989Shultz 1990 Wheeler amp Hayashi 1998 Giribetet al 2002) and this was consistently recovered inthe present analysis Haplocnemata is supported byseveral synapomorphies including (i) a feeding com-plex formed by two-segmented chelicerae (18) with aventrolateral intrinsic articulation (19) and perhapsa dorsolateral articulation with the carapace (20)and (ii) a rostrosoma (32) a preoral apparatus formedby an anteriorly projecting epistome affixed dorsallyto enlarged palpal coxae and bordered by lateral pal-pal projections The epistome forms the dorsal compo-nent of a lsquobeakrsquo with the ventral part formed by amidventral sternapophysis (13) fused between thepalpal coxae The locomotor apparatus consists ofcoxae that meet along the ventral midline (12) elon-gate femur-like patellae (48) and an apotele in theadult modified as an eversible empodium or pulvillus(92) The respiratory system is formed entirely (Pseu-doscorpiones) or in part (Solifugae) by paired trachealstigmata opening on opisthosomal somites 3 and 4(126)

Acaromorpha Dubinin 1957A clade comprising Ricinulei and Acari is consistentlyrecovered in the present analysis It is united here bytwo unique and seven homoplasious synapomorphiesThese include a gnathosoma (31) defined in part bymedial fusion of the palpal coxae (30) although pres-ence of a gnathosoma in Ricinulei is debatable as dis-cussed in the Appendix Acaromorphs also have aunique post-embryonic development consisting of ahexapodal larva and up to three octopodal nymphalinstars (176) The pedal patellandashtibia joints are formedby a bicondylar dorsal hinge rather than a single mid-dorsal condyle (69) all postcheliceral segmental gan-glia are unified in the subesophageal ganglion (130)and a postcerebral pharynx is absent (192) The groupis also tentatively united by presence of differentiatedpedal basi- and telofemora (63 64) and the absence ofa ventral (sternal) pharyngeal dilator muscle (199)but these may be symplesiomorphies erroneouslyreconstructed as a synapomorphies

The internal phylogenetic structure of Acaromorphais controversial with many recent workers favouringa monophyletic Acari (Weygoldt amp Paulus 1979Shultz 1990 argued most thoroughly by Lindquist1984) and others advocating a diphyletic Acari withAcariformes being the sister group to Cryptognomae(= Ricinulei + Anactinotrichida) (especially Van derHammen 1979 1989) (Fig 3) Acari was recovered asmonophyletic when fossils were excluded (Fig 2) butnot when they were included (Fig 3) A lsquomite-centredrsquosurvey of arachnid characters may be needed if mor-phology is to offer a compelling solution to the internalphylogeny of Acari and its placement within Arach-nida These issues are not resolved by the presentanalysis

Pantetrapulmonata nom novPantetrapulmonata includes the extinct orders Trigo-notarbida Haptopoda and the extant orders AraneaeAmblypygi Schizomida and Thelyphonida The cladeis united by cheliceral structure (18 19) a megoper-culum (106) booklungs on the genital and first post-genital somites (121 122) and enlargement of theepipharyngeal sclerite (192) It is important to notehowever that most of these characters were coded asuncertain in Plesiosiro (Haptopoda) Aside from theplacement of Haptopoda the monophyly of Trigono-tarbida and the extant orders was anticipated byShear et al (1987) and is generally regarded as amonophyletic group

Tetrapulmonata Shultz 1990Tetrapulmonata was originally proposed on the basisof neontological analyses and encompassed Araneae

PHYLOGENY OF THE ARACHNID ORDERS 239

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Amblypygi Schizomida and Thelyphonida (Shultz1990) and was also recovered here in analysis of neon-tological data (Fig 5) However results from analysisof all taxa required that the Tetrapulmonata conceptbe expanded to include the fossil order HaptopodaFeatures uniting this group are problematic howeveras many states in Haptopoda were coded as uncertain

Schizotarsata nomnovHaptopoda and Pedipalpi (= Amblypygi + Schizomida+ Thelyphonida) are united here in a group named forpossession of divided pedal telotarsi (84) Synapomor-phies include a pointed anterior carapacal margin (3)and elongation of leg 1 (46) As already noted theplacement of Haptopoda should probably be regardedas tentative because the state of many characters inthis group was coded as unknown The placement ofHaptopoda as the sister group to Pedipalpi was antic-ipated by Dunlop (1999 2002c)

Pedipalpi Boumlrner 1904Pedipalpi encompasses Amblypygi and Uropygi(= Thelyphonida + Schizomida) Although widely rec-ognized in the past Pedipalpi was set aside in the mid-20th century in favour of Labellata (= Amblypygi +Araneae) based largely on the presumed synapomor-phies of a narrow prosomandashopisthosoma juncture (97)and sucking stomach (199ndash201) The situation beganto reverse when Shear et al (1987) again highlightedsimilarities of Amblypygi and Uropygi such as rapto-rial palps (35) and antenniform leg 1 (46) Shultz(1989 1990 1999) subsequently described numerousderived similarities including asymmetrical flexormuscles at the pedal femurndashpatella joint (70) threetelotarsomeres on legs 2ndash4 (85) and modification ofthe palpal coxae to provide support for extrinsic pha-ryngeal muscles (197) Pedipalpi has emerged as oneof the most well-supported interordinal relationshipsin Arachnida

Uropygi Thorell 1882Uropygi has long been accepted as a monophyleticunion of Thelyphonida and Schizomida Synapomor-phies include a unique mating behaviour (159) fusedpalpal coxae (30) 2-1-1-1 arrangement of tibial tricho-bothria (144) posterior defensive glands (102) elon-gated patella of leg 1 (68) and many others

ACKNOWLEDGEMENTS

I thank Nikolaj Scharff and three anonymous review-ers for useful comments This work was supported bythe Maryland Agricultural Experiment Station the

General Research Board of the University of Mary-land and National Science Foundation Grant 0416628to Susan Masta

REFERENCES

Abd el-Wahab A 1952 Notes on the morphology of the scor-pion Buthus quinquestriatus (HE) Publications de lrsquoInsti-tut Fouad 1er du Desert 3 1ndash129

Alberti G 1979 Licht- und elektronenmikroskopische Unter-suchungen an Coxaldruumlsen von Walzenspinnen (ArachnidaSolifugae) Zoologischer Anzeiger 203 48ndash64

Alberti G 1995 Comparative spermatology of Cheliceratareview and perspective Memoires du Museum NationaldrsquoHistoire Naturelle 166 203ndash230

Alberti G 2005 Double spermatogenesis in ChelicerataJournal of Morphology 266 281ndash297

Alberti G Coons LB 1999 Acari mites In Harrison FWFoelix RF eds Microscopic anatomy of invertebratesVol 8C Chelicerate Arthropoda New York Wiley Liss515ndash1215

Alberti G Peretti AV 2002 Fine structure of male genitalsystem and sperm in Solifugae does not support a sister-group relationship with Pseudoscorpiones (Arachnida)Journal of Arachnology 30 268ndash274

Alexander AJ 1967 Problems of limb extension in the scor-pion Opisthophthalmus latimanus Koch Transactions of theRoyal Society of South Africa 37 165ndash181

Anderson DT 1973 Embryology and phylogeny in annelidsand arthropods Oxford Pergamon Press

Anderson LI Selden PA 1997 Opisthosomal fusion andphylogeny of Palaeozoic Xiphosura Lethaia 30 19ndash31

Beall BS Labandeira CC 1990 Macroevolutionary patternsof the Chelicerata and Tracheata In Culver SJ ed Arthro-pod paleobiology Short Courses in Paleontology No 3 Lon-don The Paleontological Society 257ndash284

Bernard HM 1896 The comparative morphology of the Gale-odidae Transactions of the Linnean Society of London Sec-ond Series 6 305ndash417

Boumlrner C 1902a Arachnologische Studien (II und III) Zool-ogischer Anzeiger 25 433ndash466

Boumlrner C 1902b Arachnologische Studien IV Die Genitalor-gane der Pedipalpen Zoologischer Anzeiger 26 81ndash92

Boumlrner C 1904 Beitraumlge zur Morphologie der Arthropoden IEin Beitrag zur Kenntnis der Pedipalpen Zoologica (Stut-tgart) 42 1ndash174

Bowler PJ 1996 Lifersquos splendid drama Chicago Universityof Chicago Press

Braddy SJ Alridge RJ Gabbott SE Theron JN 1999Lamellate book-gills in a late Ordovician eurypterid from theSoom Shale South Africa support for a eurypterid-scorpionclade Lethaia 32 72ndash74

Braddy SJ Dunlop JA 1997 The functional morphology ofmating in the Silurian eurypterid Baltoeurypterus tet-ragonophthalmus (Fischer 1839) Zoological Journal of theLinnean Society 121 435ndash461

Brauer A 1895 Beitrage zur Kenntnis der Entwicklungsge-schichte des Skorpiones II Zeitschrift fuumlr WissenschaftlicheZoologie 59 351ndash433

240 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bremer K 1994 Branch support and tree stability Cladistics10 295ndash304

Bristowe WS 1932 The liphistiid spiders Proceedings of theZoological Society of London Part 4 1016ndash1057

Butt AG Taylor HH 1991 The function of spider coxalorgans effects of feeding and salt-loading on Porrhotheleantipodiana (Mygalomorpha Dipluridae) Journal of Exper-imental Biology 158 439ndash461

Buxton BH 1913 Coxal glands of arachnids ZoologischeJahrbucher (Suppl XIV) 16ndash50

Buxton BH 1917 Notes on the anatomy of arachnids Jour-nal of Morphology 29 1ndash31

Camin JH Clark GM Bourdeau FG 1956 The palpallsquotined setarsquo in the Mesostigmata a homologue of the palpalclaw in the Onychopalpida (Acarina) In Becker EC ed Pro-ceedings of the Tenth International Congress of EntomologyOttawa Mortimer 903ndash908

Chamberlin JC 1931 The arachnid order ChelonethidaStanford University Stanford University Press

Chapman RF 1998 The insects structure and function 4thedn Cambridge Cambridge University Press

Clarke KU 1979 Visceral anatomy and arthropod phylogenyIn Gupta AP ed Arthropod phylogeny New York Van Nos-trand Reinhold 467ndash549

Clarke J 1984 On the relationship between structure andfunction in the leg joints of Heteropoda venatoria (L) (Ara-neae Eusparassidae) Bulletin of the British ArachnologicalSociety 6 181ndash192

Clarke J 1986 The comparative functional morphology of theleg joints and muscles of five spiders Bulletin of the BritishArachnological Society 7 37ndash47

Clarke JM Ruedemann R 1912 The Eurypterida of NewYork New York State Museum Memoir 14 (1ndash2) 1ndash628

Coddington JA 2005 Phylogeny and classification of spi-ders In Ubick D Paquin P Cushing PE Roth V eds Spi-ders of North America an identification manual AmericanArachnological Society 18ndash24

Coddington JA Giribet G Harvey MS Prendini LWalter DE 2004 Arachnida In Cracraft J Donoghue MJeds Assembling the tree of life Oxford Oxford UniversityPress 269ndash318

Coddington JA Levi HW 1991 Systematics and evolutionof spiders (Araneae) Annual Review of Ecology and System-atics 22 565ndash592

Coineau Y Van der Hammen L 1979 The postembryonicdevelopment of Opilioacarida with notes on new taxa and ona general model for the evolution In Piffle E ed Proceed-ings of the 4th International Congress of Acarology Budap-est Akadeacutemiai Kiadoacute 437ndash441

Cokendolpher JC Reddell JR 1992 Revision of the Pro-toschizomidae (Arachnida Schizomida) with notes on thephylogeny of the order Texas Memorial Museum Speleolog-ical Monographs 3 31ndash74

Condeacute B 1991a Le dimorphisme sexuel des Palpigrades Bul-letin de la Socieacuteteacute Neuchacircteloise des Sciences Naturelles 11667ndash73

Condeacute B 1991b Lrsquoaire genitale male des Palpigrades Bulle-tin de la Socieacuteteacute Europeenne drsquoArachnologie 1990 64ndash69

Dearden PK Donly C Grbicccc M 2002 Expression of pair-rule gene homologues in a chelicerate early patterning ofthe two-spotted spider mite Tetranychus urticae Develop-ment 129 5461ndash5472

Delle Cave L 1975 The ldquocleaning brushrdquo in Damon diademaSimon 1876 and Phrynichus jayakari Pocock 1894 (Ambly-pygi Arachnida) Monitore Zoologico Italiano (NS) 9 175ndash184

Dunlop JA 1994 Filtration mechanisms in the mouthparts oftetrapulmonate arachnids (Trigonotarbida Araneae Ambly-pygi Uropygi Schizomida) Bulletin of the British Arachno-logical Society 9 267ndash273

Dunlop JA 1996 Evidence for a sister group relationshipbetween Ricinulei and Trigonotarbida Bulletin of the BritishArachnological Society 10 193ndash204

Dunlop JA 1997 Palaeozoic arachnids and their significancefor arachnid phylogeny In Zabka M ed Proceedings of the16th European colloquium of arachnology Siedlce PolandWyzsza Szkola Rolnicko-Pedagogiczna 65ndash82

Dunlop JA 1998 The origins of tetrapulmonate book lungsand their significance for chelicerate phylogeny In SeldenPA ed Proceedings of the 17th European Colloquium ofArachnology Edinburgh 1997 Burnham Beeches BritishArachnological Society 9ndash16

Dunlop JA 1999 A redescription of the Carboniferous arach-nid Plesiosiro madeleyi Pocock 1911 (Arachnida Hapto-poda) Transactions of the Royal Society of Edinburgh EarthSciences 90 29ndash47

Dunlop JA 2000 The epistomo-labral plate and lateral lips insolifuges pseudoscorpions and mites Ekoloacutegia (Bratislava)19 67ndash78

Dunlop JA 2002a Arthropods from the Lower Devonian Sev-ernaya Zemlya Formation of October Revolution Island(Russia) Geodiversitas 24 349ndash379

Dunlop JA 2002b Character states and evolution of the che-licerate claws In Toft S Scharff N eds European Arach-nology 2000 Aarhus Aarhus University Press 345ndash354

Dunlop JA 2002c Phylogeny of Chelicerata In LlorenteBousquets J Morrone JJ eds Biodiversidad Taxonomiacutea yBiogeografiacutea de Arthroacutepodos de Meacutexico Hacia Una Siacutentesisde Su Conocimiento Vol III Prensas de Ciencias Univer-sidad Nacional Autonoma de Meacutexico 117ndash141

Dunlop JA Anderson LI Braddy SJ 2004 A redescriptionof Chasmataspis laurencii Caster amp Brooks 1956 (Chelicer-ata Chasmataspidida) from the Middle Ordovician of Ten-nessee USA with remarks on chasmataspid phylogenyTransactions of the Royal Society of Edinburgh Earth Sci-ences 94 207ndash225

Dunlop JA Braddy SJ 1997 Slit-like structures on theprosomal appendages of the eurypterid BaltoeurypterusNeues Jahrbuch fuumlr Geologie und Palaumlontologie-Monatshefte1997 31ndash38

Dunlop JA Braddy SJ 2001 Scorpions and their sister-group relationships In Fet V Selden PA eds Scorpions2001 in Memoriam Gary A Polis London British Arachno-logical Society 1ndash24

Dunlop JA Horrocks CA 19951996 A new Upper Carbon-iferous whip scorpion (Arachnida Uropygi Thelyphonida)

PHYLOGENY OF THE ARACHNID ORDERS 241

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

with a revision of the British Carboniferous Uropygi Zoolo-gischer Anzeiger 234 293ndash306

Dunlop JA Poschmann M Anderson L 2001 On theEmsian (early Devonian) arthropods of the Rhenish SlateMountains 3 The chasmataspidid Diploaspis Palaumlon-tolgische Zeitschrift 75 253ndash269

Dunlop JA Webster M 1999 Fossil evidence terrestrializa-tion and arachnid phylogeny Journal of Arachnology 27 86ndash93

Edgecombe GD Wilson GDF Colgan DJ Gray MR Cas-sis G 2000 Arthropod cladistics combined analysis of his-tone H3 and U2 snRNA sequences and morphologyCladistics 16 155ndash203

Evans GO 1992 Principles of acarology Wallingford UKCABI

Fahrenbach WH 1999 Merostomata In Harrison FW Foe-lix RF eds Microscopic anatomy of invertebrates Vol 8AChelicerate arthropods New York Wiley-Liss 21ndash115

Farley RD 1999 Scorpiones In Harrison FW Foelix RF edsMicroscopic anatomy of invertebrates Vol 8A Cheliceratearthropods New York Wiley-Liss 117ndash222

Farley RD 2005 Developmental changes in the embryo pro-nymph and first molt of the scorpion Centruroides vittatus(Scorpiones Buthidae) Journal of Morphology 265 1ndash27

Fayers SR Dunlop JA Trewin NH 2004 A new earlyDevonian trigonotarbid arachnid from the Windyfield ChertRhynie Scotland Journal of Systematic Palaeontology 2269ndash284

Felsenstein J 1985 Confidence limits on phylogenies anapproach using the bootstrap Evolution 39 783ndash791

Fet V Sissom WD Lowe G Braunwalder ME 2000 Cat-alog of the scorpions of the world (1758ndash1998) New YorkNew York Entomological Society

Firstman B 1954 The central nervous system musculatureand segmentation of the cephalothorax of a tarantula (Eu-rypelma californicum Ausserer) Microentomology 87 1ndash40

Firstman B 1973 The relationship of the chelicerate arterialsystem to the evolution of the endosternite Journal ofArachnology 1 1ndash54

Foelix RF 1985 Mechano- and chemoreceptive sensilla InBarth FG ed Neurobiology of arachnids Berlin Spinger-Verlag 118ndash137

Foelix RF 1996 Biology of spiders 2nd edn Oxford OxfordUniversity Press

Giribet G Edgecombe GD Wheeler WC Babbit C 2002Phylogeny and systematic position of Opiliones a combinedanalysis of chelicerate relationships using morphologicaland molecular data Cladistics 18 5ndash70

Goloboff P 1993 Estimating character weights during treesearch Cladistics 9 83ndash91

Goloboff PA Farris JS Nixon K 2000 TNT Tree AnalysisUsing New Technologies Version 1 Program and documen-tation httpwwwzmucdkpublicphylogenytnt

Grainge CA Pearson RG 1966 Cuticular structure in thePhalangida Nature 211 866

Grandjean F 1938 Observations sur les Bdelles (Acariens)Annales de la Socieacuteteacute Entomologique de France Paris 107 1ndash24

Hansen HJ Soslashrensen W 1904 On two orders of ArachnidaCambridge Cambridge University Press

Hansen HJ Soslashrensen W 1905 The Tartarides a tribe of theorder Pedipalpi Arkiv fuumlr Zoologi 2 1ndash78

Hara MR Gnaspini P 2003 Comparative study of the defen-sive behavior and morphology of the gland opening areaamong harvestmen (Arachnida Opiliones Gonyleptidae)under a phylogenetic perspective Arthropod Structure andDevelopment 32 257ndash275

Harvey MS 1992 The phylogeny and classification of thePseudoscorpionida (Chelicerata Arachnida) InvertebrateTaxonomy 6 1373ndash1435

Harvey MS 2002 The neglected cousins what do we knowabout the smaller arachnid orders Journal of Arachnology30 357ndash372

Haupt J Song D 1996 Revision of East Asian whip scorpions(Arachnida Uropygi Theyphonida) I China and JapanArthropoda Selecta 5 43ndash52

Hjelle JT 1990 Anatomy and morphology In Polis GA edThe biology of scorpions Stanford University Stanford Uni-versity Press 9ndash63

Janczyk FSW 1956 Anatomie von Siro duricorius Joseph imVergleich mit anderen Opilioniden Sitzungsberichte Oster-reichischen Akademie der Wissenschaften Mathematisch-Natruwissenschaftliche Klasse 165 474ndash522

Jenner RA 2001 Bilaterian phylogeny and uncritical recy-cling of morphological data sets Systematic Biology 50 730ndash742

Jeram AJ 1998 Phylogeny classification and evolution of Sil-urian and Devonian scorpions In Selden PA ed Proceed-ings of the 17th European Colloquium of ArachnologyEdinburgh 1997 Burnham Beeches UK British Arachno-logical Society 15ndash31

Karaman IM 2005 Evidence of spermatophores in Cyphoph-thalmi (Arachnida Opiliones) Revue Suisse de Zoologie 1123ndash11

Kaumlstner A 1935 Die Funktion der sogenannten sympathis-chen Ganglien und die Exkretion bei den PhalangiidenZoologischer Anzeiger 109 273ndash288

Kierans JE Robbins RG 1999 A world checklist of generasubgenera and species of ticks (Acari Ixodida) publishedfrom 1973 to 1997 Journal of Vector Ecology 24 115ndash129

Kjellesvig-Waering EN 1986 A restudy of the fossil Scorpi-onida of the world Palaeontographica Americana 55 1ndash287

Klompen JSH 2000 Prelarva and larva of Opilioacarus (Neo-carus) texanus (Chamberlin and Mulaik) (Acari Opilioacar-ida) with notes on the patterns of setation and lyrifissuresJournal of Natural History 34 1977ndash1992

Kraus O 1998 Elucidating the historical process of phylog-eny phylogenetic systematics versus cladistic techniquesIn Selden PA ed Proceedings of the 17th European Collo-quium of Arachnology Edinburgh 1997 Burnham BeechesUK British Arachnological Society 1ndash7

Kraus O Kraus M 1993 Divergent transformation of cheli-cerae and original arrangement of eyes in spiders (Arach-nida Araneae) Memoirs of the Queensland Museum 33579ndash584

Lankester ER Benham WBS Beck EJ 1885 On the mus-

242 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

cular and endoskeletal systems of Limulus and Scorpio withsome notes on the anatomy and generic characters of scor-pions Transactions of the Zoological Society of London 11311ndash384

Leclerc P 1989 Considerations paleobiogeographiques a pro-pos la decouverte en Thailande drsquoopilioacariens nouveaux(Acari ndash Notostigmata) Compte Rendu des Seances de laSociete de Biogeographie 65 162ndash174

Legg G 1976 The external morphology of a new species ofricinuleid (Arachnida) from Sierra Leone Zoological Journalof the Linnean Society 59 1ndash58

Lindquist EE 1984 Current theories on the evolution ofmajor groups of Acari and on their relationships with othergroups of Arachnida with consequent implications for theirclassificaton In Griffiths DA Bowman CE eds AcarologyVI Vol 1 New York John Wiley amp Sons 28ndash62

Manning PL Dunlop JA 1995 The respiratory organs ofeurypterids Palaeontology 38 287ndash297

Manton SM 1958 Hydrostatic pressure and leg exten-sion in arthropods with special reference to arachnidsAnnals and Magazine of Natural History Series 13 1161ndash182

Manton SM 1964 Mandibular mechanisms and the evolutionof arthropods Philosophical Transactions of the Royal Soci-ety of London B 7 1ndash183

Maraun M Heethoff M Schneider K Scheu S WeigmannG Cianciolo J Thomas RH Norton RA 2004 Molecularphylogeny of oribatid mites (Oribatida Acari) evidence formultiple radiations of parthenogenetic lineages Experimen-tal and Applied Acarology 33 183ndash201

Marples BJ 1967 The spinnerets and epiandrous glands ofspiders Journal of the Linnean Society (Zoology) 46 209ndash222

Marples BJ 1983 Observations on the structure of the fore-gut of spiders Bulletin of the British Arachnological Society8 46ndash52

Millot J 1933 Notes compleacutementaires sur lrsquoanatomie desliphistiides et des hypchilides a propos drsquoun travail reacutecent deA Petrunkevitch Bulletin de la Socieacuteteacute Zoologique de France68 217ndash235

Millot J 1942 Sur lrsquoanatomie et lrsquohistophysiologie deKoenenia mirabilis Grassi (Arachnida Palpigradi) RevueFranccedilaise drsquoEntomologie 9 33ndash51

Millot J 1943 Notes complementaires sur lrsquoanatomielrsquohistologie et la repartition geographique en France deKoenenia mirabilis Grassi Revue Franccedilaise drsquoEntomologie 9127ndash135

Millot J 1945 Lrsquoanatomie interne des Ricinulei (Arachnides)Annales des Sciences Naturelles (Zoologie) Series 11 7 1ndash29

Millot J 1949a Morphologie generale et anatomie interne InGrasseacute P-P ed Traite de zoologie VI Paris Masson et Cie263ndash319

Millot J 1949b Ordre des Amblypyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 563ndash588

Millot J 1949c Ordre des Araneacuteides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 589ndash743

Millot J 1949d Ordre des Palpigrades In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 520ndash532

Millot J 1949e Ordre des Uropyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 533ndash562

Millot J 1949f Ordre des Ricinuleides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 744ndash760

Millot J Vachon M 1949 Ordre Des Solifuges In Grasseacute P-P ed Traite de zoologie VI Paris Masson et Cie 482ndash519

Moore RA Briggs DEG Bartels C 2005 A new specimen ofWeinbergina opitzi (Chelicerata Xiphosura) from the LowerDevonian Huumlnsruck Slate Germany PalaumlontologischeZeitschrift 79 399ndash408

Moritz M 1957 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) unter besonderer Beruumlcksichtigungder ausseren Morphologie der Bildung des Mitteldarmesund der Genitalanlage Zoologische Jahrbuumlcher Abteilungfuumlr Anatomie und Ontogenie der Tiere 76 331ndash370

Moritz M 1959 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) II Die Anlage und Entwicklung derCoxaldruumlse bei Phalangium opilio L Zoologische Jahr-buumlcher Abteilung fuumlr Anatomie und Ontogenie der Tiere 77229ndash240

Nixon K 1999 The parsimony ratchet a new method forrapid parsimony analysis Cladistics 15 407ndash414

Norton RA 1998 Morphological evidence for the evolutionaryorigin of Astigmata (Acari Acariformes) Experimental andApplied Acarology 22 559ndash594

Palmgren P 1978 On the muscular anatomy of spiders ActaZoologica Fennica 155 1ndash41

Patten W 1912 The evolution of the vertebrates and their kinPhiladelphia P Blakistonrsquos Son

Patten W Hazen AP 1900 The development of the coxalgland branchial cartilages and genital ducts of Limuluspolyphemus Journal of Morphology 16 459ndash490

Paulus HF 1979 Eye structure and the monophyly of theArthropoda In Gupta AP ed Arthropod phylogeny NewYork Van Nostrand Reinhold 299ndash383

Petrunkevitch A 1949 A study of Palaeozoic ArachnidaTransactions of the Connecticut Academy of Arts and Sci-ences 37 69ndash315

Pittard K Mitchell RW 1972 Comparative morphology ofthe life stages of Cryptocellus pelaezi (Arachnida Ricinulei)Graduate Studies Texas Tech University 1 1ndash77

Platnick NI 2005 The world spider catalog Version55 American Museum of Natural History httpresearchamnhorgentomologyspiderscatalogindexhtm

Platnick NI Gertsch WJ 1976 The suborders of spiders acladistic analysis (Arachnida Araneae) American MuseumNovitates 2607 1ndash15

Pocock RI 1893 On some points in the morphology of theArachnida (ss) with notes on the classification of the groupAnnals and Magazine of Natural History Series 6 11 1ndash19

Pocock RI 1902 Some points in the morphology and classi-fication of the Opiliones Annals and Magazine of NaturalHistory Series 7 10 504ndash516

Popadicccc A Panganiban G Rusch D Shear WA KaufmanTC 1998 Molecular evidence for the gnathobasic derivationof arthropod mandibles and for the appendicular origin ofthe labrum and other structures Development Genes andEvolution 208 142ndash150

PHYLOGENY OF THE ARACHNID ORDERS 243

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Raw F 1957 Origin of chelicerates Journal of Paleontology31 139ndash192

Reddell JR Cokendolpher JC 1995 Catalogue bibliogra-phy and generic revision of the order Schizomida (Arach-nida) Texas Memorial Museum Speleological Monographs4 1ndash170

Reissland A Goumlrner P 1985 Trichobothria In Barth FGed Neurobiology of arachnids Berlin Springer-Verlag 138ndash161

Roewer C-F 1934 Solifugae Palpigradi Bronns Klassen undOrdnungen ses Tierreich 5 IV 4 1ndash713

Rowland JM Cooke JAL 1973 Systematics of the arachnidorder Uropygida (=Thelyphonida) Journal of Arachnology 155ndash71

Rowland JM Sissom WD 1980 Report on a fossil palpi-grade from the Tertiary of Arizona and a review of themorphology and systematics of the order (ArachnidaPalpigradida) Journal of Arachnology 8 69ndash86

Rucker A 1901 The Texan Koenenia American Naturalist35 615ndash630

Ruhland M Rathmayer W 1978 Die Beinmuskulatur undihre Innervation bei der Vogelspinne Dugesiella henzti (Ch)(Araneae Aviculariidae) Zoomorphologie 89 33ndash46

Saboori A Kamali K 2000 Description of Allothrombiumtriticium adult (Acari Trombidiidae) from Iran Systematicand Applied Acarology 5 207ndash208

Schliwa M 1979 The retina of the phalangid Opilio raven-nae with particular reference to arhabdomeric cells Celland Tissue Research 204 473ndash495

Scholl G 1977 Beitrage zur Embryonalentwicklung von Lim-ulus polyphemus L (Chelicerata Xiphosura) Zoomorpholo-gie 86 99ndash154

Scholtz G Kamenz C 2006 The book lungs of Scorpionesand Tetrapulmonata (Chelicerata Arachnida) evidence forhomology and a single terrestrialization event of a commonarachnid ancestor Zoology 109 2ndash13

Seitz KA 1972 Zur Histologie und Feinstruktur des Herzensund der Haumlmocyten von Cupiennius salei Keys (AraneaeCtenidae) I Herzwandung Bildung und Differenzierungder Haumlmocyten Zoologische Jahrbuumlcher Abteilung fuumlr Anat-omie und Ontogenie der Tiere 89 351ndash384

Sekiguchi K ed 1988 Biology of horseshoe crabs Tokyo Sci-ence House

Selden PA 1981 Functional morphology of the prosoma ofBaltoeurypterus tetragonophthalmus (Fischer) (ChelicerataEurypterida) Transactions of the Royal Society of Edin-burgh Earth Sciences 72 9ndash48

Selden P 1992 Revision of the fossil ricinuleids Transactionsof the Royal Society of Edinburgh Earth Sciences 83 595ndash634

Selden PA 1993 Fossil arachnids ndash recent advances andfuture prospects Memoirs of the Queensland Museum 33389ndash400

Selden PA Dunlop JA 1998 Fossil taxa and relationshipsof chelicerates In Edgecombe GD ed Arthropod fossilsand phylogeny New York Columbia University Press 303ndash331

Selden PA Jeram AJ 1989 Palaeophysiology of terrestrial-

isation in the Chelicerata Transactions of the Royal Societyof Edinburgh Earth Sciences 80 303ndash310

Selden PA Shear WA Bonamo PM 1991 A spider and otherarachnids from the Devonian of New York and reinterpre-tations of Devonian Araneae Palaeontology 34 241ndash281

Sensenig A Shultz JW 2003 Mechanics of cuticular elasticenergy storage in leg joints lacking extensor muscles inarachnids Journal of Experimental Biology 206 771ndash784

Shear WA 1993 New species in the opilionid genus Stylocel-lus from Malaysia Indonesia and the Philippines (OpilionesCyphophthalmi Stylocellidae) Bulletin of the British Arach-nological Society 9 174ndash188

Shear WA Selden PA Rolfe WDI Bonamo PM GriersonJD 1987 New terrestrial arachnids from the Devonian ofGilboa New York (Arachnida Trigonotarbida) AmericanMuseum Novitates 2901 1ndash74

Shultz JW 1987 The origin of the spinning apparatus in spi-ders Biological Reviews of the Cambridge PhilosophicalSociety 62 89ndash113

Shultz JW 1989 Morphology of locomotor appendages inArachnida evolutionary trends and phylogenetic implica-tions Zoological Journal of the Linnean Society 97 1ndash56

Shultz JW 1990 Evolutionary morphology and phylogeny ofArachnida Cladistics 6 1ndash31

Shultz JW 1991 Evolution of locomotion in Arachnida thehydraulic pressure pump of the giant whipscorpion Masti-goproctus giganteus (Uropygi) Journal of Morphology 21013ndash31

Shultz JW 1993 Muscular anatomy of the giant whipscor-pion Mastigoproctus giganteus (Arachnida Uropygi) and itsevolutionary significance Zoological Journal of the LinneanSociety 108 335ndash365

Shultz JW 1998 Phylogeny of Opiliones (Arachnida) anassessment of the lsquoCyphopalpatoresrsquo concept Journal ofArachnology 26 257ndash272

Shultz JW 1999 Muscular anatomy of a whipspider Phrynuslongipes (Amblypygi) and its evolutionary significance Zoo-logical Journal of the Linnean Society 126 81ndash116

Shultz JW 2000 Skeletomuscular anatomy of the harvest-man Leiobunum aldrichi (Weed 1893) (Arachnida Opil-iones Palpatores) and its evolutionary significanceZoological Journal of the Linnean Society 128 401ndash438

Shultz JW 2001 Gross muscular anatomy of Limuluspolyphemus (Chelicerata Xiphosura) and its bearing on evo-lution in the Arachnida Journal of Arachnology 29 283ndash303

Shultz JW 2007 Morphology of the prosomal exoskeleton ofScorpiones (Arachnida) and a new hypothesis for the evolu-tion of cuticular cephalic endoskeletons in arthropodsArthropod Structure amp Development 36 77ndash102

Shultz JW Pinto da Rocha R 2007 Morphology and func-tional anatomy In Pinto da Rocha R Machado G Giribet Geds The harvestmen the biology of Opiliones CambridgeMA Harvard University Press 14ndash61

Shultz JW Regier JC 2001 Phylogenetic analysis of Pha-langida (Arachnida Opiliones) using two nuclear protein-encoding genes supports monophyly of Palpatores Journalof Arachnology 29 189ndash200

Sitnikova LG 1978 The main evolutionary trends of the

244 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Acari and the problem of their monophyletism Entomologis-cheskoe Obozrenie 57 431ndash457

Snodgrass RE 1948 The feeding organs of Arachnidaincluding mites and ticks Smithsonian Miscellaneous Col-lections 110 (10) 1ndash94

Snodgrass RE 1952 A textbook of arthropod anatomy Ith-aca NY Comstock Publishing

Soslashrensen W 1879 Om Bygningen af Gonyleptiderne en Typeaf Arachnidernes Classe Naturhistorisk Tidsskrifter Series3 8 97ndash222

Stockwell SA 1989 Revision of the phylogeny and higherclassification of scorpions (Chelicerata) Unpublished doc-toral dissertation University of California at Berkeley

Stoslashrmer L 1944 On the relationships and phylogeny of fossiland Recent Arachnomorpha Skrifter utgitt av Det NorskeVidenskaps-Akademi i Oslo I 5 1ndash158

Stoslashrmer L 1955 Merostomata In Moore RC ed Treatise oninvertebrate paleontology Part P Arthropoda 2 LawrenceKS University of Kansas Press and Geological Society ofAmerica 4ndash41

Swofford DL 2002 PAUP Phylogenetic analysis using par-simony (and other methods) Version Beta 10 SunderlandMA Sinauer Associates

Talarico G Palacios-Vargas JG Fuentes Silva M AlbertiG 2005 First ultrastructural observations of the tarsal poreorgan of Pseudocellus pearsei and P boneti (ArachnidaRicinulei) Journal of Arachnology 33 604ndash612

Templeton AR 1983 Phylogenetic inference from restrictionendonuclease cleavage site maps with articular reference tothe evolution of humans and the apes Evolution 37 221ndash244

Tetlie OE Braddy SJ 2004 The first Silurian chasmataspidLoganamaraspis dunlopi gen et sp nov (Chelicerata Chas-mataspidida) from Lesmahagow Scotland and its implica-tions for eurypterid phylogeny Transactions of the RoyalSociety of Edinburgh Earth Sciences 94 227ndash234

Thomas RH Telford MJ 1999 Appendage development inembryos of the oribatid mite Archegozetes longisetosus(Acari Oribatei Trhypochthoniidae) Acta Zoologica (Stock-holm) 80 193ndash200

Tollerton VP Jr 1989 Morphology taxonomy and classifica-tion of the order Eurypterida Burmeister 1843 Journal ofPaleontology 63 642ndash657

Vachon M 1949 Ordre des Pseudoscorpions In Grasseacute P-Ped Traite de zoologie VI Paris Masson et Cie 436ndash481

Van der Hammen L 1979 Comparative studies in Chelicer-ata I The Cryptognomae (Ricinulei Architarbi and Anacti-notrichida) Zoologische Verhandelingen 174 1ndash62

Van der Hammen L 1982 Comparative studies in Chelicer-ata II Epimerata (Palpigradi and Actinotrichida) Zoolo-gische Verhandelingen 196 1ndash70

Van der Hammen L 1985 Functional morphology and affin-ities of extant Chelicerata in evolutionary perspectiveTransactions of the Royal Society of Edinburgh 76 137ndash146

Van der Hammen L 1989 An introduction to comparativearachnology Leiden SPB Academic Publishing

Vyas AB 1970 Studies in myology of the scorpion Heter-ometrus fulvipes Koch Unpublished PhD DissertationDepartment of Zoology Gujarat University India

Vyas AB 1974 The cheliceral muscles of the scorpion Heter-ometrus fulvipes Bulletin Southern California Academy ofSciences 73 9ndash14

Walter DE 1988 Predation and mycophagy by endeostigma-tid mites (Acariformes Prostigmata) Experimental andApplied Acarology 4 159ndash166

Walter DE Proctor HC 1998 Feeding behaviour and phy-logeny observations on early derivative Acari Experimentaland Applied Acarology 22 39ndash50

Walter DE Proctor HC 1999 Mites ecology evolution andbehaviour Sydney University of New South Wales

Waterston CD 1979 Problems of functional morphology andclassification in stylonuroid eurypterids (Chelierata Meros-tomata) with observations on the Scottish Silurian Sty-lonuroidea Transactions of the Royal Society of EdinburghEarth Science 70 251ndash322

Weygoldt P 1969 Biology of pseudoscorpions CambridgeMA Harvard University Press

Weygoldt P 1984 Lrsquoautotomie chez les Amblypyges RevueArachnologique 5 321ndash327

Weygoldt P 1996 Evolutionary morphology of whip spiderstowards a phylogenetic system (Chelicerata ArachnidaAmblypygi) Journal of Zoological Systematics and Olution-ary Research 34 185ndash202

Weygoldt P 1998 Evolution and systematics of the Chelicer-ata Experimental and Applied Acarology 22 63ndash79

Weygoldt P 2000 Whip spiders (Chelicerata Amblypygi)their biology morphology and systematics StenstrupDenmark Apollo Books

Weygoldt P Paulus HF 1979 Untersuchungen zur Morphol-ogie Taxonomie und Phylogenie der Chelicerata Zeitschriftfuumlr Zoologische Systematik und Evolutionsforschung 17 85ndash116 177ndash200

Wheeler WC Hayashi CY 1998 The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192

Winkler D 1957 Die Entwicklung der aumlusseren Koumlrperge-stalt bei den Phalangiidae (Opiliones) Mitteilungen aus demZoologischen Museum in Berlin 33 355ndash389

With CJ 1904 The Notostigmata a new suborder of AcariVidenskabelige Meddelelser fra Dansk Naturhistorisk Foren-ingi Kjobenhaus 1904 137ndash192

Yamasaki T Makioka T Saito J 1988 Morphology InSekiguchi K ed Biology of horseshoe crabs Tokyo ScienceHouse 69ndash132

Yoshikura M 1975 Comparative embryology and phylogeny ofArachnida Kumamoto Journal of Science Biology 12 71ndash142

APPENDIX

Each character used in this analysis is numbereddefined and cross-referenced with characters used infour previous studies that are abbreviated in bracketsspecifically WP = Weygoldt amp Paulus (1979)S = Shultz (1990) WH = Wheeler amp Hayashi (1998)GEWB = Giribet et al (2002) Descriptions includecorrections of errors discussions of controversies orambiguities and justifications for state assignmentsto lsquoproblemrsquo taxa

PHYLOGENY OF THE ARACHNID ORDERS 245

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

PROSOMA

1 Dorsal sclerite formed by fusion of the prosomal car-apace the dorsal portion of the first opisthosomalsomite and the dorsomedial (axial) portion of the sec-ond opisthosomal somite 0 absent 1 presentState 1 is present in extant xiphosurids and close fos-sil relatives (Scholl 1977 Anderson amp Selden 1997Shultz 2001) The fossil record suggests that thisstructure originated with the disappearance ofthe already reduced first opisthosomal tergite(= microtergite) and axial portion of the secondopisthosomal tergite (Anderson amp Selden 1997)2 Single dorsal sclerite covering entire dorsal surfaceof body no lines indicating original segmentation 0absent 1 presentState 1 occurs throughout Holothyrida (Acari) (Vander Hammen 1989) and in many Mesostigmata(Alberti amp Coons 1999) including Glyptholaspis (Vander Hammen 1989)3 Anterior end of dorsal prosoma with medianmarginal or submarginal pointed process 0 absent1 presentState 1 occurs throughout Schizomida (Protoschizo-mus Stenochrus Reddell amp Cokendolpher 1995) The-lyphonida (Rowland amp Cooke 1973 Shultz unpublobs Proschizomus Dunlop amp Horrocks 19951996)Amblypygi (Charinus Millot 1949b Phrynus Shultz1999) and Plesiosiro (Dunlop 1999) Citing a figure inRowland amp Cooke (1973) Dunlop amp Horrocks assumedthat the anterior process was limited to lsquohypoctonidrsquothelyphonids and united them with Proschizomus andSchizomida If valid this would make Thelyphonidaparaphyletic However the doubtful monophyly oflsquohypoctonidsrsquo (Haupt amp Song 1996) and widespreadpresence of a submarginal anterior process in Pedi-palpi weaken this proposal (Harvey 2002)4 Ophthalmic ridges pair of longitudinal crests inter-secting or passing near the region of the lateral eyes orcomparable region where lateral eyes are absent 0absent 1 presentState 1 is present throughout Xiphosura (Anderson ampSelden 1997) Similar structures are present in Plesi-osiro (Dunlop 1999) and lsquonon-hypoctonidrsquo Thely-phonida (Mastigoproctus Rowland amp Cooke 1973)Dunlop (2002a) described apparent ophthalmic ridgesin the chasmataspidid Octoberaspis but these werenot observed in Diploaspis (Dunlop et al 2001) eitherdue to absence or inadequate preservation of speci-mens nor in Chasmataspis (Dunlop et al 2004)5 Carapace with demarcations (eg grooves scleritesphragmata) between pro- meso- or metapeltidia 0absent 1 present [S 1 WH 48 GEWB 29]State 1 is present in extant Scorpiones Cyphoph-thalmi (Hansen amp Soslashrensen 1904) and many Palpa-tores (Opiliones) including Caddo and Leiobunum

and most Pseudoscorpiones (Chamberlin 1931) Giri-bet et al (2002) coded this character as absent inCyphophthalmi (Opiliones) but a transverseprocurved groove is present throughout the group(Hansen amp Soslashrensen 1904) and represents the meso-metapeltidial border as indicated by attachment ofthe dorsal endosternal suspensor of postoral somite VIin Chileogovea oedipus and Siro exilis (Shultz unpublobserv) Kjellesvig-Waering (1986) did not explicitlydescribe such grooves in fossil scorpions although heillustrated corresponding structures in PrearcturusArchaeophonus Archaeoctonus Palaeophonus andother Palaeozoic scorpions (Kjellesvig-Waering 1986figs 14A 21E 27D 59) (coded as uncertain here) butnot in Proscorpius (contra Giribet et al 2002) Dis-tinct carapacal sclerites are present throughout Palpi-gradi Schizomida and Solifugae (6)

The condition in Acari is controversial but is codedhere as State 1 Van der Hammen (1989) consideredthe tergal region of the leg-bearing somites (postoralsomites IIIndashVI) or podosoma to be reduced andreplaced by posterior migration of the aspidosoma (ietergal region assumed to be associated with append-ages of the gnathosoma) and anterior migration of theopisthosomal tergal region to form the hysterosomaThe dorsal proterosomal and hysterosomal elementsmeet at a transverse sejugal furrow that continues lat-erally and passes ventrally between the coxae of legs 2and 3 According to this scheme the prosomandashopisthosoma border is expressed as the disjugal fur-row which passes from the ventral posterior margin ofthe podosoma anterodorsally to join the dorsal part ofthe sejugal furrow Based on the arrangement of setaeand slit sensilla (= lyrifissures) Van der Hammen(1989) interpreted the region above coxae 3 and 4 (ieregion C) as a fusion of the dorsal parts of the first twoopisthosomal somites (= postoral somites VII and VIII)in early divergent Anactinotrichida (ie Opilioacari-formes) and Acariformes (eg Alycus) These interpre-tations have been followed by many acarologistsalthough its speculative aspects are widely acknowl-edged (Evans 1992 Alberti amp Coons 1999)

From the standpoint of a general arachnologist theVan der Hammen system seems unnecessarily compli-cated it appears to have been formulated to explainbroad morphological themes in Oribatida but was thenextrapolated to other mites A more conservativelsquoarachnologicalrsquo scheme adopted here equates the pro-podosoma with the pars cephalica (= propeltidium) ofthe arachnid prosoma and region C with the pars tho-racica (= mesopeltidium + metapetidium) In fact theearliest and most recent treatments of Opilioacari-formes (With 1904 Klompen 2000) reached similarconclusions This interpretation is also consistent withevidence from Acariformes For example the regiondorsal to the coxa of legs 3 and 4 in Alycus is as readily

246 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

explained by the persistence of primitive prosomal ter-gal elements as by the supposition that these disap-peared and were replaced by two opisthosomalelements Furthermore several diverse lineages ofbasally divergent Acariformes display features that donot fit readily within the GrandjeanndashVan der Ham-men system but seem to correspond to the cephalicathoracica division of the prosoma They have a sejugalfurrow that may have demarcated the cephalicatho-racica border and a postpodosomal furrow thatappears to represent the prosomandashopisthosoma junc-tion there is no disjugal furrow Examples includepalaeosomate (especially aphelacarid) and pediculoch-elid Oribatida alicorhagiid and mircopsammidlsquoEndeostigmatarsquo and paratydeid Prostigmata (Albertiamp Coons 1999 fig 42) This organization may be sym-plesiomorphic for Acariformes but could be a conver-gence brought about by selection for enhancedflexibility6 Carapace with distinct pro- meso- or metapeltidialsclerites 0 absent 1 present - inapplicable due toabsence of pro- meso- or metapeltidial demarcations(5)State 1 occurs throughout Palpigradi Schizomida andSolifugae It also appears sporadically within Opil-iones (eg Leiobunum flavum Shultz unpubl observ)and Pseudoscorpiones (eg Pseudochiridium) (Cham-berlin 1931)7 Sejugal furrow circumferential zone of body flexi-bility that passes between the coxae of legs 2 and 3 0absent 1 presentState 1 is probably synapomorphic for Acariformes asit occurs in most lsquoendeostigmatidsrsquo including Alycus(Van der Hammen 1989) and is variously developedthroughout Sarcoptiformes (Alberti amp Coons 1999) Itis probably the primitive condition for Prostigmatabut is either absent or weakly expressed in represen-tatives included here See 5 for alternative morpholog-ical interpretations of this body region in Acari8 Prosomal ozopores 0 absent 1 present [WH 46GEWB 12]State 1 occurs throughout Opiliones and Holothyrida(Van der Hammen 1989)9 Carapacal pleural doublure 0 absent 1 present[WP 12 S 2 WH 49 GEWB 27]A carapacal pleural doublure occurs throughoutXiphosura (Stoslashrmer 1944) Eurypterida (Clarke ampRuedemann 1912 Selden 1981) and Chasmataspid-ida (Dunlop et al 2001 2004 Dunlop 2002a) and isabsent in all known Arachnida10 Cardiac lobe a longitudinal axial elevation of thecarapace 0 absent 1 presentState 1 occurs throughout Xiphosura (Anderson ampSelden 1997) and Eurypterida (Clarke amp Ruedemann1912 Selden 1981) Carapaces are not preserved suf-ficiently to determine this condition in the chasmatas-

pidids Diploaspis (Dunlop et al 2001) andOctoberaspis (Dunlop 2002a) but an apparent cardiaclobe occurs in the Chasmataspis (Dunlop et al 2004)A cardiac lobe is absent in Arachnida11 Moveable cucullus 0 absent 1 present [WP 54GEWB 22]State 1 is a synapomorphy of Ricinulei (Selden 1992)12 Medial intercoxal lsquosternalrsquo region 0 all pedalcoxae separated medially 1 anterior pedal coxae abut-ting medially posterior coxae separated 2 anteriorpedal coxae separated medially posterior coxae abut-ting 3 all pedal coxae abutting medially 4 epimeracoxae undifferentiated medially from ventral body wall[S sim3 WH sim50 GEWB sim17]Prosomal sternites appear to be those portions of theventral body wall not occupied by coxae Despite earlyattempts to assess metamerism of the ventralprosoma (protosternum deutosternum tritosternumetc) (Boumlrner 1902a Millot 1949a) no apparent mor-phological features such as borders of sclerites ormuscle attachments reliably demarcate the ventralbody wall of one somite from that of an adjacentsomite Thus coding schemes that focus on describingmetameric components of the sternal region are prob-lematic Shultz (1990) attempted to code the shape ofthe entire intercoxal region and subsequent workershave adopted this approach All pedal coxae are sepa-rated medially in Plesiosiro (Dunlop 1999) Palpigradi(Roewer 1934) Araneae (Millot 1949c) Amblypygi(Weygoldt 2000) and many Acari including Opilioac-ariformes Holothryrida Mesostigmata Ixodida Aly-cus and many Prostigmata (Van der Hammen 1989Evans 1992 Alberti amp Coons 1999) The anteriorcoxae are separated in extant Thelyphonida Schizo-mida (Hansen amp Soslashrensen 1905 Millot 1949e) andRicinulei (Pittard amp Mitchell 1972 Legg 1976) Thecoxae are fused to the ventral body wall in Sarcopti-formes (Van der Hammen 1989 Evans 1992)

Interpreting this character is complicated for theanterior coxae of Scorpiones and Opiliones and allcoxae in Xiphosura and Eurypterida due to the pres-ence of coxapophyses endites andor gnathobaseswhich form components of the preoral chambers inthese groups Thus coxae may abut medially whenthese structures are considered part of the coxae butare separate medially if these structures are ignoredAs coxapophyses etc are here coded as separate char-acters (50ndash53) I have chosen to code this character asif these structures did not exist Consequently pha-langid Opiliones are coded as having all coxae sepa-rated (Pocock 1902 Hansen amp Soslashrensen 1904) Fossilscorpions also show substantial variation in this char-acter All pairs of pedal coxae in Palaeoscorpius abutmedially but also have gnathobases (Jeram 1998)and this condition is coded here as uncertain A ster-num separates all pairs of pedal coxae medially in

PHYLOGENY OF THE ARACHNID ORDERS

247

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Proscorpius

and

Stoermeroscorpio

(State 0) and

Pre-arcturus

approximates modern scorpions (State 1)(Kjellesvig-Waering 1986 Jeram 1998) The coxalbases of Eurypterida and Xiphosura are separated bya small sternite the endostoma (Selden 1981 Shultz2001)

13 Postoral sternapophysis (

=

tritosternum labium)a cuticular evagination of the ventral body wall poste-riorly adjacent to the palpal coxae forming the poste-rior border of the preoral chamber in some taxa ordisplaced posteriorly by fusion of the palpal coxae inothers 0 absent 1 present

[S

sim

10 WH

sim

56 GEWB19

=

33]State 1 is present throughout Palpigradi AraneaeAmblypygi (Snodgrass 1948) Schizomida (Hansen ampSoslashrensen 1905 Van der Hammen 1989) Thely-phonida (Van der Hammen 1989) extant Ricinulei(Pittard amp Mitchell 1972 Legg 1976) Opilioacari-formes and Mesostigmata (Van der Hammen 1989Alberti amp Coons 1999)

Allothyrus constrictus

is theonly holothyrid mite known to have sternapophysesAn apparent sternapophysis has been described in apalaeocharinid trigonotarbid (Dunlop 1994) Giribet

et al

(2002) miscoded Ixodida and Acariformes as hav-ing a lsquolabium (

=

tritosternum)rsquo The lophognath ofpseudoscorpions (Chamberlin 1931) and lsquolabiumrsquo ofSolifugae (Roewer 1934) also appear to be sternapo-physes that have been incorporated into the ros-trosoma (

32

) The labium of phalangid Opiliones(Pocock 1902 Hansen amp Soslashrensen 1904) is a ster-napophysis but is associated with leg 1 rather thepedipalp (Winkler 1957) A small anterior sclerite inthe fossil scorpions

Proscorpius

Waeringoscorpio

and

Labriscorpio

has been interpreted as a lsquolabiumrsquo homol-ogous with the sternapophyses of other arachnids(Weygoldt 1998) despite absence of any indication ofthe relationship of this sclerite to the mouth and pos-sible alternative interpretations (eg

16

) Giribet

et al

coded this sclerite as a sternapophysis but the char-acter is coded here as uncertain in fossil scorpions

14 Channels on the body surface linking openings ofcoxal organs to preoral chamber 0 absent 1 present

Van der Hammen (1989) described bilaterally pairedcuticular tracts (lsquotaenidiarsquo) connecting orifices of coxalorgans to the preoral chamber in some arachnids andor a ventromedian groove (

=

intercoxal or subcapitulargutter) that leads in turn to the preoral chamber inother taxa Taenidia and intercoxal gutters arepresent in Opilioacariformes (Van der Hammen1989) Holothyrida (Van der Hammen 1989) Mesos-tigmata (

Glyptholapis

Van der Hammen 1989)Ricinulei (Pittard amp Mitchell 1972 Van der Hammen1989) Araneae (

Heptathela

Aphonopelma

Hypochi-lus

Shultz unpubl observ also

Porrhothele

(Mygalo-morphae Dipluridae) Butt amp Taylor 1991

Segestria

(Araneomorphae Segestriidae) Van der Hammen

1989) Amblypygi (Van der Hammen 1989) Schizo-mida (

Stenochrus

Shultz unpubl observ unidenti-fied Hubbardiinae Van der Hammen 1989) andThelyphonida (

Mastigoproctus

Shultz unpublobserv also

Tetrabalius

Van der Hammen 1989) Theconduction of large volumes of fluid from orifices of thecoxal organs to the preoral chamber has been observedin the mygalomorph spider

Porrhothele

(Dipluridae)(Butt amp Taylor 1991) and

Mastigoproctus

(Thely-phonida) (Shultz unpubl observ) The podocephaliccanal (

15

) is present in Acariformes (Acari) (Lindquist1984) and appears to have evolved as an invaginatedsupracoxal channel (Grandjean 1938 G Alberti inEvans 1992) Van der Hammen (1982) could not dis-cern with certainty whether a channel exists in

Eukoenenia

(Palpigradi) but none has been reportedby previous authors and they have not been observedby the present author Palpigradi is coded as havingState 0

15 Podocephalic canal cuticular channel andorduct draining multiple glands and opening nearmouthparts 0 absent 1 present - inapplicable due toabsence of channel (14)

State 1 occurs throughout Acariformes (Acari)(Lindquist 1984)

16 Heavily sclerotized suboral sclerite serving inpart as basal pivot point for coxae of appendages ofpostoral somites IIndashIV (

=

arachnid palp and legs 1 and2) 0 absent 1 present

The sclerite forms the posterior wall of the true mouthand is heavily sclerotized in extant scorpions It is notvisible externally due to the tightly fitting coxae of theanterior prosomal appendages (Shultz 2007) It maycorrespond to the so-called labium (Weygoldt 1998) orlabrum (Kjellesvig-Waering 1986) of certain fossilscorpions (see

13

)

17 Genal angles 0 rounded 1 pointed

Anderson amp Selden (1997) originally coded a pointedgenal angle and genal spine as separate charactersbut they are combined here State 1 is present in manyXiphosura (Anderson amp Selden 1997) and Chas-mataspidida (Dunlop

et al

2001 2004 Dunlop2002a)

A

PPENDAGES

OF

POSTORAL

SOMITE

I

CHELICERAE

18 Cheliceral segmentation 0 three articles 1 twoarticles

[WP

sim

25

+

42 WH

sim

15 GEWB

sim

44]State 0 is present in extant Xiphosura Eurypterida(Clarke amp Ruedemann 1912 Selden 1981) Palpi-gradi (Roewer 1934) Opiliones (Hansen amp Soslashrensen1904) Scorpiones (Hjelle 1990) and throughout non-acariform Acari (Evans 1992) State 1 is presentthroughout known Trigonotarbida Araneae Ambly-pygi Thelyphonida Schizomida Pseudoscorpionesand Solifugae Acariform mites have two unambigu-

248 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ous cheliceral articles but may have a reduced thirdarticle (Van der Hammen 1989 Evans 1992) acari-form mites are thus coded as uncertain Chelicerae arenot known or their condition is unclear in several fos-sil taxa including Xiphosura Chasmataspidida (Dun-lop et al 2001 2004 Dunlop 2002a) Riciniulei(Selden 1992) Plesiosiro (Dunlop 1999) Proschizo-mus (Dunlop amp Horrocks 19951996) and the fossilscorpions Praearcturus and Palaeoscorpius (Kjelles-vig-Waering 1986) Kjellesvig-Waering (1986) consis-tently interpreted chelicerae of fossil scorpions ashaving four articles but this is probably a misinter-pretation of the basal article (see also Stockwell1989) The basal article in extant scorpions consists ofa distal collar of cuticle and a large proximal processand this arrangement would probably appear as twoarticles in compressed specimens In any event thereare at least three cheliceral articles in Stoermeroscor-pio and Proscorpius (Kjellesvig-Waering 1986)19 Terminal cheliceral joint 0 laterally placedbicondylar hinge 1 dorsally placed bicondylar hinge2 ventrally placed bicondylar hinge [WP 25 + 42 Ssim11 + 12 WH sim15 GEWB sim44]State 0 is present in Xiphosura (Shultz 2001)extant Scorpiones (Millot amp Vachon 1949) Opil-iones (Hansen amp Soslashrensen 1904) Palpigradi(Boumlrner 1904 Millot 1949d) Ricinulei (Pittard ampMitchell 1972 Legg 1976) and Eurypterida includ-ing Baltoeurypterus (Selden 1981) and Stylonurus(Clarke amp Ruedemann 1912) State 0 is present inthe fossil scorpions Proscorpius and Stoermeroscor-pio but the state of the chelicerae is unclear in theother fossil scorpions represented here (Kjellesvig-Waering 1986) State 1 is present in Trigonotarbida(Shear et al 1987) mygalomorph Araneae Ambly-pygi Schizomida and Theylphonida (Weygoldt ampPaulus 1979) State 2 is present in Pseudoscorpi-ones (Chamberlin 1931) and Solifugae (Roewer1934) and appears to be a groundplan feature for allmajor groups of mites (eg Van der Hammen 1989Alberti amp Coons 1999) The difference between theorthognathy of mygalomorph Araneae and lsquoplagiog-nathyrsquo mesothele (Liphistius Heptathela) and palae-ocribellate Araneae (Hypochilus) (Kraus amp Kraus1993) is small compared with the variation thatoccurs throughout arachnids and plagiognathy iscoded as State 1 here20 Chelicera articulating with carapace at anterolat-eral pivot 0 absent 1 present [S 13 WH 57 GEWB45]State 1 occurs in Solifugae (Roewer 1934) and Pseu-doscorpiones except Chthonioidea and Feaelloidea(Chamberlin 1931) Giribet et al (2002) coded thecharacter as present in Pseudoscorpiones21 Chelicera pivoting on dorsal protuberance of epis-tome 0 absent 1 present

State 1 appears to be a unique synapomorphy of Opil-iones (Shultz 2000)22 Extrinsic cheliceral muscle arising on carapaceand inserting on dorsal margin of nonbasal cheliceralarticle 0 absent 1 present (Note Characters 22 and23 refer to two different muscles)State 1 is known only in extant Scorpiones (LankesterBenham amp Beck 1885 Vyas 1970 1974 Shultz2007) but may occur in acariform mites if their cheli-cerae are composed of three rather than two articles(Evans 1992) (see 18) Acariformes are coded asuncertain23 Extrinsic cheliceral muscle arising on carapaceand inserting on ventral margin of nonbasal cheliceralarticle 0 absent 1 present [GEWB 252] (Note Char-acters 22 and 23 refer to two different muscles)State 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) somemesostigmatid Acari (Evans 1992) and representa-tive Opiliones particularly Leiobunum (Shultz 2000)Chileogovea and Gonyleptes (Shultz unpubl observ)The status of the character in other opilions is not yetknown The character was originally defined as anextrinsic muscle inserting on the second segment ofthree-segmented chelicerae (Shultz 2000 Giribetet al 2002) but it has been redefined here to make itapplicable to arachnids with two-segmented cheli-cerae24 Extrinsic cheliceral muscles attaching to epistome0 absent 1 presentState 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) but mayoccur in Prostigmata if the sigmoid piece to which thecheliceral protractors attach (Evans 1992) is a modi-fication of the epistome25 Lateral tergocheliceral muscle with three heads 0absent 1 present - coded only for extant tetrapulmo-nates and the palpigrade Eukoenenia homology isunclear in other taxa [GEWB 230]State 1 is present in Pedipalpi (Shultz 1999)26 Cheliceral silk glands 0 absent 1 present [GEWB47]State 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931) The silk gland of spider mites(eg Tetranychus) are associated with the pedipalps(Evans 1992)27 Cheliceral venom glands 0 absent 1 present [WP34 WH 40 GEWB 46]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)28 Cheliceral serrula interior and exterior 0 absent1 presentState 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931 Harvey 1992)29 Cheliceral lsquoflagellumrsquo in male 0 absent 1 present[GEWB 48]

PHYLOGENY OF THE ARACHNID ORDERS 249

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITE II LEG 1 IN NONARACHNIDS AND PALP IN ARACHNIDS

30 Palpal coxae fused ventromedially and formingposterior wall of preoral chamber 0 absent 1 present[WP 27 S 18 WH 17 = 62 GEWB 33 + 63]State 1 occurs in Schizomida Thelyphonida Acari andRicinulei31 Gnathosoma 0 absent 1 present - inapplicablepalpal coxae not fused (30) [WP 56 S 19 WH sim63GEWB sim36 + 37]The gnathosoma is a functional complex comprisingan epistome attached to the dorsal surface of mediallyfused palpal coxae (= infra- or subcapitulum) (see 30)the chelicerae and in some taxa a supracheliceral tec-tum These components are tightly integrated andmove in unison relative to the body The gnathosomahas traditionally been considered a feature of AcariHowever Van der Hammen (1989) argued that a gna-thosoma is present in Ricinulei and this interpreta-tion has received wide acceptance (eg Lindquist1984) However it is not clear that the relationship ofthe chelicerae and subcapitulum is as intimate inRicinulei as in Acari and the ricinuleid subcapitulumdoes not appear to be as moveable In fact the lsquosubca-pitulumrsquo in Ricinulei is similar in many respects to thecondition in Schizomida and Thelyphonida (Hansen ampSoslashrensen 1904) which has never been regarded as agnathosoma Consequently I have coded the conditionof Ricinulei as uncertain32 Rostrosoma long narrow subcylindrical epistomeprojecting anteriorly with base fixed to dorsal surfaceof palpal coxae bordered laterally by lobes projectingfrom palpal coxae ventral wall of preoral chamberformed by anterior element of prosoma (sternapophy-sis) 0 absent 1 present [S 15 WH 59 GEWB 20 + 36]State 1 is present in Pseudoscorpiones and Solifugae(Chamberlin 1931 Roewer 1934 Van der Hammen1989) Dunlop (2000) attempted to homologize com-ponents of the rostrosoma and gnathosoma (31) anargument accepted by Giribet et al (2002) Howeverfusion of the palpal coxae (30) apparently evolvedindependently in the two structures Coxal fusion inthe gnathosoma is complete ventrally and excludesthe suboral sternapophysis (13) while fusion isincomplete ventrally in the rostrosoma and incorpo-rates the sternapophysis Unlike the gnathosomathe rostrosoma is largely immobile and cheliceralmovement is not coupled with that of the palpalcoxae33 Rutellacorniculi hypertrophied setae modified asmouthparts located on the anterior processes of the pal-pal coxae 0 absent 1 present [GEWB 39]

State 1 occurs in the anactinotrichid groups Opilioac-ariformes Holothyrida and Mesostigmata but not Ixo-dida (Van der Hammen 1989) It also occurs inSarcoptiformes (Evans 1992) and certain endeostig-matids (eg Alycus Van der Hammen 1989) but not inProstigmata Comparable structures are unknown inother chelicerate groups34 Terminal segments specially modified in adultmale as a clasper used to engage the female 0 absent1 presentState 1 is known only in extant Xiphosura (YamasakiMakioka amp Saito 1988)35 Robust raptorial 0 absent 1 present [S 17 WH61 GEWB sim59 97]State 1 is present in Amblypygi Thelyphonida andSchizomida (Shear et al 1987) and in Laniatores(Opiliones)36 Orientation of robust raptorial appendage 0operating in subtransverse plane 1 operating in sub-vertical plane - inapplicable coded only for PedipalpiState 0 is present in extant Thelyphonida and mostAmblypygi and State 1 is present in Schizomida (Cok-endolpher amp Reddell 1992) and in the basally diver-gent amblypygid Paracharon caecus (Weygoldt 2000)The orientation of the pedipalps in the fossil Proschi-zomus is uncertain (Dunlop amp Horrocks 19951996)37 Extrinsic muscle attaching to epistome 0 absent1 presentState 1 occurs in Chileogovea and Siro (OpilionesCyphophthalmi) and extant Scorpiones (Vyas 1970Shultz 2007)38 Muscle originating and inserting within coxa 0absent 1 present [GEWB sim232]State 1 occurs is in Amblypygi (Phrynus Shultz 1999Phrynichus Boumlrner 1904) Thelyphonida (Mastigo-proctus Shultz 1993 Thelyphonus Boumlrner 1904) andSchizomida (Stenochrus Shultz unpubl observ)There are two intracoxal muscles in amblypygids andthelyphonids the one coded here and another that isserially homologous with a muscle in the pedal coxae(54)39 Tarsus andor tibia with venom glands 0 absent1 present [GEWB 64]State 1 is known only from extant Pseudoscorpi-ones except the superfamilies Chthonioidea andFeaelloidea (Chamberlin 1931 Harvey 1992) Giri-bet et al (2002) coded this character as present forPseudoscorpiones40 Tarsal grooming organ 0 absent 1 present[GEWB 98]State 1 is a unique synapomorphy of Amblypygi (DelleCave 1975 Weygoldt 2000)41 Scorpionoid chela a large well-developed chelaformed by tibia (manus + fixed finger) and tarsus(moveable finger) 0 absent 1 present [S 16 WH 60GEWB 62]

250 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs throughout Scorpiones andPseudoscorpiones42 Modified in male as copulatory organ 0 absent 1present [WP 35 WH 39 GEWB 67]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)43 Apotele 0 apparently absent not differentiatedexternally from penultimate article 1 present [GEWB69]The apotele exists as a structure (ie dactyl clawempodium) differentiated from the tibiotarsus tarsusor telotarsus and is typically associated with a pair ofantagonistic muscles The apotele occurs as a distinctstructure in Xiphosura (Snodgrass 1948) Eurypter-ida (Clarke amp Ruedemann 1912) Schizomida (Coken-dolpher amp Reddell 1992) Araneae (Foelix 1996)early divergent Amblypygi (Charinus Weygoldt 2000not Phrynus Shultz 1999) Anactinotrichida (CaminClark amp Bourdeau 1956 Lindquist 1984) Ricinulei(Pittard amp Mitchell 1972 Legg 1976) and Solifugae(Roewer 1934) Giribet et al (2002) coded Solifugae asnot having a palpal apotele but it is present as anadhesive organ (45) Dunlop (1999) reconstructedHaptopoda as having a palpal claw but notes in hisdescription that the distal ends were not preserved inthe known specimens44 Apotele (claw) position 0 terminal 1 subtermi-nal - inapplicable coded only for taxa with a distinctapotele (43)State 1 occurs in Parasitiformes (Acari) (Evans 1992)45 Terminal adhesive organ 0 absent 1 present -inapplicable due to lack of an apotele (43) [GEWB 66]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITES IIIndashVI LEGS 2ndash5 IN NON-ARACHNIDS AND LEGS 1ndash4 IN ARACHNIDS

46 Appendage III (= arachnid leg 1) extremely elon-gate antenniform 0 absent 1 present [WP sim24 + 32S 20 WH 14 GEWB sim84]State 1 occurs in Amblypygi Thelyphonida Schizo-mida (Shear et al 1987) and Opilioacariformes (Vander Hammen 1989) State 1 is approximated in Solif-ugae Palpigradi (Roewer 1934) and Haptopoda(Dunlop 1999) and it is coded as uncertain in thesetaxa47 Appendage V (= arachnid leg 3) of male specializedfor sperm transfer 0 absent 1 present [WP 55 GEWB94]State 1 is known only from extant Ricinulei48 Appendages V and VI (= arachnid legs 3 and 4)with femur shorter than patella and with principal siteof flexionextension at patella-tibia joint (lsquoapatellatersquocondition sensu Van der Hammen 1989) 0 absent 1present [S 25 GEWB 86]

State 1 occurs in Pseudoscorpiones and Solifugae(Shultz 1989)49 Appendage VI (= arachnid leg 4) with terminusmodified as flattened blade 0 absent 1 presentState 1 occurs in some Chasmataspidida (DiploapisDunlop et al 2001 Octoberaspis Dunlop 2002a)uncertain in Chasmataspis Dunlop et al 2004 State0 in Loganamaraspis Tetlie amp Braddy 2004) andEurypterida except stylonuroids (Clarke amp Ruede-mann 1912 Waterston 1979 Tollerton 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI COXA AND BODYndashCOXA JOINT

50 Appendage III (= arachnid leg 1) with coxapophy-sis forming floor or wall of preoral chamber 0 absent1 present [S sim14 WH sim58]State 1 occurs in Opiliones (Hansen amp Soslashrensen 1904)and higher scorpions (Jeram 1998) A lobe is presenton the coxa of leg 1 in the fossil scorpion Prearcturusand perhaps Stoermeroscorpio and is coded here as acoxapophysis Coxapophysis of leg 1 together with theepistome and coxapophysis of the palp form a uniquepreoral chamber the stomotheca Weygoldt (1998) andDunlop amp Braddy (2001) reject the homology of cox-apophyses and stomothecae in Opiliones and Scorpi-ones because these structures are not apparent infossils of those scorpions thought to have been aquaticHowever the coxapophyses in Opiliones are formedlargely from soft cuticular lsquolipsrsquo and similar structuresin fossil scorpions would probably not have beenpreserved51 Appendage IV (= arachnid leg 2) with coxapophy-sis 0 absent 1 present [S sim14 WH sim58 GEWB 75]State 1 occurs in extant Scorpiones and most Opil-iones except Dyspnoi (Shultz 1998)52 At least one pair of coxal gnathobases on append-ages III-VI (= arachnid legs 1ndash4) 0 absent 1 present[S sim23 WH sim66 GEWB 78]State 1 occurs in all appropriately preserved Euryp-terida including Baltoeurypterus (Selden 1981) andStylonurus (Clarke amp Ruedemann 1912) extantXiphosura on appendages IIndashVI and apparentlyProscorpius on appendage III (= arachnid leg 1)(Kjellesvig-Waering 1986 Jeram 1998 contra Giribetet al 2002) Fossilized appendages with gnathobaseshave been found in association with Chasmataspis(Chasmataspidida) suggesting that they may havehad State 1 (Dunlop et al 2004)53 Coxae of appendages IIIndashV (= arachnid legs 1ndash3)with jointed moveable endites 0 absent 1 present [Ssim23]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) and in adequately preserved eurypterids(Clarke amp Ruedemann 1912) (BaltoeurypterusSelden 1981)

PHYLOGENY OF THE ARACHNID ORDERS 251

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

54 Intracoxal muscle a muscle arising on anteriorwall of coxa and inserting on posterior wall 0 absent1 present [GEWB 237]State 1 occurs in Thelyphonida (MastigoproctusShultz 1993) and Amblypygi (Phryus Shultz 1999)It is unclear whether the muscle operating the move-able endite (53) in extant Xiphosura (Manton 1964Shultz 2001) is homologous with this muscle and iscoded here as uncertain It is not present in Steno-chrus (Schizomida) (Shultz unpubl observ)55 Coxae of appendages IIndashVI (= arachnid palp andlegs) with dorsal articulation with carapace 0 absent1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)56 Flabellum (exite) on coxa of appendage VI(= arachnid leg 4) 0 absent 1 present [GEWB sim110]State 1 occurs in extant Xiphosura (Yamasaki et al1988) An apparent exite occurs on appendages asso-ciated with Chasmataspis (Chasmataspidida) (Dunlopet al 2004)57 Insertion process of anteromedial tergo-coxal mus-cle 0 weakly developed 1 well to extremely well devel-oped [GEWB 238]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Boumlrner 1904 Shultz 1999)58 Musculi laterales enlarged lateral tergocoxalmuscle with attachment shifted from coxa to adjacentpleural membrane 0 absent 1 present [S 22 WH 65GEWB 242]State 1 occurs in Araneae and Thelyphonida (Shultz1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI TROCHANTER AND COXAndashTROCHANTER JOINT

59 Coxandashtrochanter joint with complex posteriorarticulation composed of two articulating sclerites 0absent 1 present [S 24 WH 67 GEWB 85]State 1 occurs in Araneae Amblypygi Thelyphonidaand Schizomida (Shultz 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI FEMUR AND TROCHANTERndashFEMUR JOINT

60 Depressor muscle (or homologue) of trochanter-femur joint 0 absent 1 present [WH 42 GEWB 220]State 0 appears to be a unique synapomorphy of Ara-neae (Liphistius Shultz 1989 Aphonopelma Ruh-land amp Rathmayer 1978 Hypochilus present studyother examples Clarke 1984 1986)61 Trochanterndashfemur joint with dorsal hinge or pivotoperated by flexor muscles only 0 absent 1 present[WH sim44 GEWB sim91]State 1 occurs in Palpigradi Other chelicerates have abicondylar articulation and are typically equipped

with antagonistic muscles (Van der Hammen 1985Shultz 1989) but see 60 for an exception62 Superior trochanterndashfemur muscle (or homologue)originating broadly in femur inserting on distal mar-gin of trochanter 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 19892001)63 Basifemurndashtelofemur joint of appendages III andIV (= arachnid legs 1 and 2) in adult 0 absent 1present [S 25 WH 68 GEWB 86]The basifemur and telofemur apparently develop aslsquosisterrsquo articles through division of a parental articlethe femur (eg opilioacariform endeostigmatidpalaeosomatid mites) (Coineau amp Van der Hammen1979 Evans 1992) A single animal can have all legswith divided or undivided femora or have some com-bination of the two typically with the more posteriorlegs having the divided femora Thus anterior legsoften tend to be paedomorphic with respect toposterior legs Most chelicerates have undivided fem-ora on some or all legs but muscles homologous withthose of the basifemurndashtelofemur joint may stilldevelop (66) (Shultz 1989) suggesting that the devel-opment of cuticular and muscular components of ajoint have a degree of developmental and evolutionaryindependence

Undivided femora occur in most extant euchelicer-ates including Xiphosura Palpigradi AraneaeAmblypygi Thelyphonida Schizomida ScorpionesOpiliones Pseudoscorpiones higher Sarcoptiformes(Archegozetes Van der Hammen 1989) many Prostig-mata (Tetranychus Evans 1992) and throughout Par-asitiformes Divided femora occur in all legs in manyearly divergent Acariformes including many Endeo-stigmata basally divergent Sarcoptiformes (egPalaeacarus) and Prostigmata (eg AllothrombiumMicrocaeculus) (Van der Hammen 1989 Evans1992) Only leg 4 has basi- and telofemora in Alycus(Van der Hammen 1989) Basi- and telofemora occuron legs 3 and 4 in Solifugae Ricinulei Opilioacari-formes and Eurypterida (Clarke amp Ruedemann 1912Selden 1981 Shultz 1989)

Giribet et al (2002) coded the lsquoannulusrsquo of certainTrigonotarbida (Shear et al 1987) as a basifemur buta similar non-muscularized annulus is present in theposterior legs of extant Thelyphonida In both casesthe condyles of the trochanterndashannulus and annulusndashfemur joints are aligned in such away that they per-form essentially identical movements and appear notto enhance either the range or degree of leg move-ment [This contrasts with the condition in the lasttwo pairs of legs in Baltoeurypterus where condyles ofthe apparent trochanter-basifemur and basifemurndashtelofemur joints have different arrangements (Selden1981)] The annulus in Thelyphonida appears to func-tion in maintaining constant volume of the tro-

252 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

chanterndashfemur joint by preventing ballooning of thearthrodial membrane thereby minimizing the effect ofinternal fluid pressure on movement64 Basifemurndashtelofemur joint of appendage V(= arachnid leg 3) in adult 0 absent 1 present See 6365 Basifemurndashtelofemur joint of appendage VI(= arachnid leg 4) in adult 0 absent 1 present See6366 Cuticular differentiation of basifemur-telofemurjoint absent but muscles present 0 absent 1 present- inapplicable due to presence of joint (63ndash65)State 1 occurs in appendages of postoral somites IIIndashIV (= arachnid legs 1ndash4) in Xiphosura AraneaeAmblypygi Thelyphonida Schizomida Scorpionesand Pseudoscorpiones (Shultz 1989)67 Circumfemoral ring 0 absent 1 presentThe femora of one or more pairs of legs in anactinot-richid Acari have a basal groove or ring associatedwith slit sensilla Acarologists term that portionproximal to the ring the lsquobasifemurrsquo and that portiondistal to the ring the lsquotelofemurrsquo and they refer tothe basifemur as the lsquosecond trochanterrsquo and thetelofemur as the lsquofemurrsquo Despite the confusingterminology the basifemur discussed in 63ndash65 is nothomologous with the lsquobasifemurrsquo of opilioacariformand parasitiform Acari as both a true basifemurndashtelofemur joint and a circumfemoral ring are presentin legs 3 and 4 of Opilioacariformes (Van der Ham-men 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI PATELLA AND FEMURndashPATELLA JOINT

68 Patella of appendage of postoral somite III(= arachnid leg 1) proportionally much longer thanthose of more posterior appendages 0 absent 1presentState 1 occurs in Thelyphonida Schizomida (Hansenamp Soslashrensen 1905 Shultz 1989) and Opilioacari-formes (Van der Hammen 1989)69 Femurndashpatella joint 0 monocondylar several axesof movement and multifunctional muscles 1 bicondy-lar hinge one axis of movement and antagonistic mus-cles 2 hinge one axis of movement flexor muscleswithout muscular antagonists [S sim26 WH sim69 GEWBsim87]State 0 occurs in Solifugae State 1 occurs in Phalang-ida (Opiliones) Scorpiones and PseudoscorpionesState 2 occurs in the remaining taxa (Shultz 1989)70 Patellar plagula 0 absent or with simple medianattachment 1 symmetrical Y-shaped with long prox-imal stem 2 symmetrical U-shaped (= arcuate scler-ite) 3 asymmetrical attaching to patella only atanterior margin [S 27 WH sim70 GEWB 234 = 243]State 1 occurs in extant Xiphosura (Shultz 1989)State 2 occurs in Araneae (Manton 1958 Ruhland amp

Rathmayer 1978 Clarke 1986 Shultz 1989) State 3occurs in Amblypygi Thelyphonida and Schizomida(Shultz 1989) Giribet et al (2002) included this char-acter twice

APPENDAGES OF POSTORAL SOMITES IIIndashVI TIBIA AND PATELLAndashTIBIA JOINT

71 Tibiae divided by one or more joints 0 absent 1presentState 1 occurs in Amblypygi (Weygoldt 2000)72 Patellandashtibia joint 0 monocondylar with or with-out CZY (73) 1 bicondylar hinge one axis of move-ment antagonistic muscles 2 hinge one axis ofmovement muscles without muscular antagonists [Ssim31 WH 74 GEWB sim92] (Shultz 1989 Selden Shearamp Bonamo 1991)73 Patellandashtibia joint with posterior compression zone(lsquoCZYrsquo) 0 absent 1 present - inapplicable due toabsence of monocondylar articulation [GEWB 99]State 1 is a unique synapomorphy of Araneae (Seldenet al 1991)74 Patellandashtibia joint of appendages IIIndashVI(= arachnid legs 1ndash4) with deep-set monocondylar pivotbordered by a pair of tibial processes to which extensormuscles attach 0 absent 1 present - inapplicabledue to absence of monocondylar joint (72)State 1 occurs in extant Xiphosura (Snodgrass 1952Shultz 1989) although a similar arrangement ispresent in the appendage III (= arachnid leg 1) in The-lyphonida (Shultz 1993)75 Patellandashtibia joint largely immobile specializedfor autotomy 0 absent 1 presentState 1 occurs throughout Amblypygi (Weygoldt 19842000)76 Anterior femurndashtibia or femoropatellandashtibia(transpatellar) muscle 0 absent 1 present [S sim30GEWB 222] (Shultz 1989)77 Proximal attachment of posterior femurndashtibia orfemoropatellarndashtibia (transpatellar) muscle 0 muscleabsent 1 dorsoposterior surface of femur andor pos-terior surface of patella 2 distal process of femur mus-cle acting as extensor of femur-patella joint [S sim28 WHsim71 + 72 GEWB 235 = 244 245] (Shultz 1989)78 Distal attachment of posterior femurndashtibia (trans-patellar) muscle 0 posterior 1 dorsal acting as exten-sor of patellandashtibia joint - inapplicable due to absenceof muscle (77) [S 29 GEWB 221] (Shultz 1989)79 Anterior patellandashtibia muscle 0 absent 1 present[S sim32 WH sim73 GEWB 246] (Shultz 1989)80 Posterior patellandashtibia muscle 0 absent 1 present[S 33 WH 76 GEWB 248] (Shultz 1989)81 Patellandashtibia joint spanned by elastic (lsquospringlikersquo)sclerite 0 absent 1 present [GEWB 107]State 1 is a unique synapomorphy of Solifugae (Sens-enig amp Shultz 2003)

PHYLOGENY OF THE ARACHNID ORDERS 253

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

APPENDAGES OF POSTORAL SOMITES IIIndashVI TARSUS AND TIBIAndashTARSUS JOINT

82 Tarsus divided into proximal basitarsus(= metatarsus) and distal telotarsus (= distitarsus orlsquotarsusrsquo) 0 absent 1 presentState 0 occurs in extant Xiphosura (Shultz 1989) andthroughout Acariformes (Lindquist 1984 Evans1992) in the anterior two pairs of legs in chthonioidpseudoscorpions and all legs in lsquomonosphyronidrsquo pseu-doscorpions including Feaelloidea and Cheliferoidea(Chamberlin 1931) A circumtarsal ring (83) associ-ated with slit sensilla in Parasitiformes (Acari) mayrepresent a joint between a telotarsus and basitarsus(Evans 1992)83 Circumtarsal ring 0 absent 1 presentState 1 is an apparent synapomorphy of Parasiti-formes (Evans 1992)84 Telotarsus in adult with two or more tarsomeres 0absent 1 present - inapplicable due to absence of telo-tarsus (82) [GEWB 71]State 1 occurs in Schizomida Thelyphonida Ambly-pygi Palpigradi Ricinulei Solifugae (Shultz 1989)Opilioacariformes (Van der Hammen 1989) and Ple-siosiro (Dunlop 1999)85 Three telotarsomeres on appendages of postoralsomites IV-VI (= arachnid legs 2ndash4) 0 absent 1present - inapplicable due to absence of telotarsus (82)or absence of telotarsomeres (84) [S 34 WH 77 GEWB102]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Shultz 1989)86 Tibiandashtarsus joint spanned by well-developed elas-tic (lsquospringlikersquo) sclerite 0 absent 1 present [GEWB107]State 1 occurs in Scorpiones phalangid Opiliones andSolifugae (Alexander 1967 Shultz 2000 Sensenig ampShultz 2003) Extant Xiphosura is coded for the lastprosomal appendage only (see 88)87 Appendage of postoral somite VI (= arachnid leg 4)with ring of large basally articulated spatulate pro-cesses at tibia-tarsus joint 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)88 Appendages postoral somites IIIndashV (= arachnidlegs 1ndash3) with tibiotarsus (ie tibia and tarsus not dif-ferentiated) 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 1989) butnot apparently in Weinbergina (Moore et al 2005) theonly synxiphosurid in which appendages are suffi-ciently preserved to determine the number ofpodomeres89 Ambulacrum peduncle-like extension of the tarsuswith internal condylophores and terminating distallywith apotele (eg claws) andor pulvillus 0 absent 1present

State 1 is present throughout Acari (Alberti amp Coons1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI APOTELE AND TARSUSndashAPOTELE JOINT

90 Apotele of appendage III (= arachnid leg 1) 0absent or not apparent 1 present [GEWB 101]State 0 occurs in Amblypygi Thelyphonida and Schi-zomida (Shultz 1999) However the apotele is proba-bly present but undifferentiated because muscles thatwould normally insert on the apotele terminate on theend of the lsquotarsusrsquo Dunlop (1999) reconstructed Hap-topoda as lacking an apotelic claw on leg 1 but noted inthe text that this was speculative Dunlop (2002b) hasreviewed apotelic diversity in Chelicerata91 Appendages of postoral somites IIIndashVI (= arachnidlegs) chelate with chela formed from tibiotarsus andapotele or tarsus and apotele 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988) The condition appears to occur in Chasmataspisbased on associated appendages (Dunlop et al 2004)but not in Diploaspis and Octoberaspis (Dunlop et al2001 Dunlop 2002a)92 Apotele with eversible or padlike empodium(= pulvillus) in adult 0 absent although empodialclaw may be present 1 present [S sim38 WH 81 GEWB95]State 1 occurs throughout Pseudoscorpiones (Cham-berlin 1931) Solifugae (Roewer 1934) Opilioacari-formes Holothyrida Ixodida and free-livingMesostigmata (Evans 1992) An eversible lsquopulvillusrsquo isprobably a primitive feature of Amblypygi (eg Char-inus Weygoldt 2000) but is absent in higher groups(eg Phrynus Weygoldt 2000)93 Inferior apotele muscle (= claw depressor) with tib-ial attachment 0 absent 1 present [S 35 WH 78GEWB 249]The character is coded for extant Xiphosura usingappendage of postoral somite VI (= arachnid leg 4) asother legs lack a differentiated tibia (88) State 1appears to occur throughout Arachnida (Shultz 1989)94 Inferior apotele muscle (= claw depressor) withpatellar attachment 0 absent 1 present [S 36 WH 79GEWB 250]State 0 occurs in extant Xiphosura Acari AmblypygiPalpigradi and Ricinulei State 1 occurs in all otherarachnids examined here (Shultz 1989)

OPISTHOSOMA

95 Number of opisthosomal somites in adult 0 five1 eight 2 nine 3 10 4 11 5 12 6 13 [WH sim33GEWB sim190]Chasmataspidids appear to have 13 opisthosomalsomites (Dunlop et al 2004) The synziphosurans

254 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Weinbergina (Moore et al 2005) and Limuloides(Anderson amp Selden 1997) appear to have ten opistho-somal somites Moore et al (2005) found no evidence ofan anterior microtergite in Weinbergina as proposedby Anderson amp Selden (1997) Xiphosurids appear tohave nine opisthosomal somites (Scholl 1977 Ander-son amp Selden 1997 Shultz 2001) Members of the fol-lowing orders have 12 opisthosomal somites Araneae(Millot 1949ac) Amblypygi (Weygoldt 2000) Thely-phonida and Schizomida (Hansen amp Soslashrensen 1905)and Pseudoscorpiones (Chamberlin 1931) Elevensomites are present in Palpigradi Solifugae (Roewer1934) and Trigonotarbida (Gilboarachne Shear et al1987 Palaeocharinus Fayers et al 2004) Plesiosiromay have 11 or 12 somites (Dunlop 1999) Justifica-tion for coding of more controversial taxa is providedbelow

Scorpiones Two principal hypotheses regarding thenumber of opisthosomal somites in scorpions havebeen advocated a 13-somite hypothesis derived fromembryological studies (Brauer 1895 Patten 1912Farley 1999 2005) and a 12-somite hypothesis basedon comparative anatomy of adults (Weygoldt amp Pau-lus 1979) The embryological interpretation is basedon the observation of pregenital genital and pectinalsomites (each with segmental ganglia and paired limbbuds) in early scorpion embryos followed by extremereduction or loss of the pregenital somite in laterembryos According to this view a missing pregenitalsomite should be added to the 12 apparent opisthoso-mal somites of post-embryonic scorpions to achieve afinal number of 13 The anatomy-based hypothesiswas introduced by Weygoldt amp Paulus (1979) whoadvocated a literal interpretation of post-embryonicsegmentation based on opisthosomal tergites Specifi-cally these authors argued that the last pair of dorsalendosternal suspensor muscles of non-scorpionarachnids especially Pedipalpi attach to the first(= pregenital) somite that this condition also occurs inscorpions and that there is no reason to invoke a miss-ing pregenital tergite They proposed that the pectinesbelong to the genital somite and that functional spe-cializations of the nervous system for pectinal functiongive the apparence of an extra neuromere duringembryonic development

I recently dissected the prosoma and anterioropisthosoma of the scorpions Centruroides Hadrurusand Heterometrus and focused on the composition ofthe muscular diaphragm (103) that separates thehaemocoelic compartments of the prosoma andopisthosoma (Shultz 2007) The diaphragm is com-posed of a metameric series of axial muscles fromthree somites the anterior somite corresponds to thelast prosomal somite and the posterior somite corre-sponds to the genital somite The middle elements

insert dorsally along a tranverse tendon attached tothe anterior margin of the first tergite These obser-vations are consistent with the embryological inter-pretation that a pregenital somite is present but itstergite is not expressed It appears likely that the thepregenital somite was compressed longitudinally dur-ing the evolution of the diaphragm I code scorpions ashaving 13 opisthosomal somites

Eurypterida The eurypterid opisthosoma is widelyassumed to have 12 somites (Clarke amp Ruedemann1912 Stoslashrmer 1944) However in a speculative paperon the evolutionary morphology of trilobites and che-licerates Raw (1957) proposed that both scorpions andeurypterids have 15 opisthosomal somites Rawassumed that scorpions have 13 apparent opisthoso-mal somites based on the transient pregenital somiteof scorpion embryology that scorpions and eurypteridsare close relatives and should have the same numberof somites and that olenellid trilobites and chelicer-ates always have somites in multiples of three Rawachieved 15 somites in scorpions by assuming the lastopisthosomal somite to be a diplosomite and that thetelson is a postanal somite even though there is noevidence for either of these proposals He attributedthese features to eurypterids as well He also notedthat the connection between the prosomal carapaceand first opisthosomal tergite in Eurypterida differedstructurally from the connection between adjacentopisthosomal tergites and regarded this as evidencefor a reduced pregenital tergite in the prosomandashopisthosoma junction Rawrsquos speculations were largelyforgotten until Dunlop amp Webster (1999) resurrectedthe proposal that eurypterids have a reduced opistho-somal tergite and therefore share a unique similaritywith scorpions Unfortunately Dunlop amp Bullocktreated Rawrsquos conjecture as if it were based on empir-ical evidence rather than an attempt to force euryp-terid morphology into a peculiar numerical system Inthe absence of convincing evidence to the contrary Ihave coded eurypterids as having 12 opisthosomalsomites

Ricinulei The opisthosoma consists of a membranouspedicel (97) bearing the gonopore ventrally Thefemale gonopore is bordered by an anterior plate and aposterior plate The remainder of the opisthosoma iscomposed of thick sclerites separated by less heavilysclerotized cuticle The first dorsal sclerite is short andfunctions as part of a prosomandashopisthosoma couplingmechanism (96) that may or may not be a specializedcomponent of the following tergite This is followed byfour tergites and sternites and a three-segmentedmetasoma (= lsquopygidiumrsquo) Millot (1945 1949f) rea-soned that the opisthosoma contains ten somites withsomites VIIndashIX incorporated into the pedicel XndashXIIIexpressed as tergites and sternites and the metasoma

PHYLOGENY OF THE ARACHNID ORDERS 255

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

comprising four somites He did not regard the dorsalcoupling sclerite as a separate tergite and his inter-pretation of four rather than three metasomal somiteshas been rejected Pittard amp Mitchell (1972) also pro-posed ten opisthosomal somites but achieved thisnumber by regarding the dorsal coupling sclerite as atergite of somite IX and by recognizing three metaso-mal somites Legg (1976) adopted the system proposedby Pittard amp Mitchell but did not regard the dorsalcoupling sclerite as separate from the following tergite(X) Van der Hammen (1979 1989) reconstructed 13somites The pedicel was regarded as having twosomites (VII + VIII) the following four sets of tergitesand sternites were considered diplosomites (IX + XXI + XII XIII + XIV XV + XVI) and the metasoma hadthree somites (XVIIndashXIX) This interpretation mayhave been biased by Van der Hammenrsquos attempt tounite Ricinulei with anactinotrichid Acari which healso regarded as having a primitive number of 13somites Selden (1992) described a fossil ricinuleidTerpsicroton that shows two pairs of depressions onthe three large premetasomal tergites which con-trasts with the single pair seen in extant species Thisobservation appears to corroborate the diplosomitehypothesis but the evidence does not indicate that thetergite anterior to these is a diplosomite

Dunlop (1996) attempted to homologize theprosomandashopisthosoma coupling mechanisms ofRicinulei and Trigonotarbida a goal that required anovel and rather forced interpretation of the dorsalsclerites He regarded the coupling sclerite as homol-ogous with the first opisthosomal tergite (VII) of trig-onotarbids and then followed Van der Hammenrsquosdiplosomite hypothesis to achieve 12 somites in totalDunloprsquos scheme differs from previous systems in sug-gesting that the pedicel does not contain the dorsalelements of the first and second somites and is in-consistent with Millotrsquos (1945) observation that thepre- and postgenital plates each have dorsoventralmuscles Here I code Ricinulei as having 12 somitesThere are three metasomal somites (XVIndashXVIII)three diplosomites (= six somites) (XndashXV) one cou-pling somite (IX) and two somites in the pedicel (iethe genital and pregenital somites) (VII VIII)

Opiliones Harvestmen have nine opisthosomalsomites and an anal operculum that is traditionallyregarded as the tergite of a tenth somite a tenth ster-nite is lacking (Hansen amp Soslashrensen 1904 Winkler1957) However the anal operculum appears to repre-sent a persistent embryonic telson (Moritz 1957) andis therefore likely to be a postsegmental structurecomparable with the stinger of scorpions flagella ofthelyphonids etc (Shultz 2000)

Acari The number of opisthosomal somites is probe-matic for most mite taxa due to uncertainty about the

location of the prosomandashopisthosoma boundry (5)extensive simplification or loss of metamericallyarranged sclerites and muscle attachments a paucityof developmental studies of engrailed expression andin Acariformes opisthosomal anamorphosis and het-erochronic modification of somite number (Evans1992) Ixodids appear to be the exception developmen-tal studies indicate five somites in the opisthosoma ofticks (Evans 1992)

Some mites retain external evidence of segmenta-tion ndash metameric patterns of furrows muscle attach-ments and slit sensilla ndash and with certainassumptions the number of opisthosomal somites canbe estimated Here it is assumed that the dorsal sur-faces of the last two prosomal somites are present inmites and retain their primitive association with legs3 and 4 (contra Van der Hammen 1989) (see 5 for jus-tification) Given this there appear to be 11 somites inthe opisthosoma of Opilioacariformes a conclusionalso reached by other workers (eg With 1904 Sitni-kova 1978 Klompen 2000) Similar reasoning sug-gests that Alycus has seven opisthosomal somites notnine as advocated by Van der Hammen (1989) Unfor-tunately external evidence is ambiguous for determin-ing the number of somites in other mite lineages andthese are coded as uncertain96 Prosoma-opisthosoma coupling mechanism 0absent 1 present [GEWB sim24]The posterior margin of the carapace and the anteriormargin of the first apparent opisthosomal tergite arespecialized as a coupling mechanism in Ricinulei andTrigonotarbida and Dunlop amp Horrocks (1996) pro-posed that these are synapomorphic for the twogroups However there is uncertainty about thehomology of the anterior opisthosomal somites in Trig-onotarbida and Ricinulei (95)97 Pedicel 0 absent 1 aranean type 2 ricinuleidtype [WP 30 S sim40 WH sim20 GEWB sim126]The body narrows at or near the prosomandashopisthosoma juncture in several arachnid lineages (ieSolifugae Palpigradi Amblypygi Araneae Ricinulei)and this lsquowaistrsquo has often been used as character at theinterordinal level (eg Pocock 1893) However thisinterpretation is rejected here for being subjective andfor uniting non-homologous conditions For exampleAraneae and Amblypygi are often grouped on the basisof a lsquopedicelrsquo yet it is a highly specialized structure inAraneae and its parts are not readily homologizedwith those of Amblypygi In contrast the condition inAmblypygi is a slightly narrower version of the highlymoveable prosomandashopisthosoma juncture in Uropygiwhich is not generally considered a lsquopedicelrsquo Thepedicel in Ricinulei is also unique a weakly sclero-tized stalk containing the genital opening (Cryptocel-lus Pittard amp Mitchell 1972 Ricinoides Legg 1976)See 95

256 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

98 Opisthosoma with three- segmented lsquobucklerrsquo 0absent 1 presentState 1 occurs throughout Chasmatapsidida (Dunlop2002a)99 Thoracetron consolidation of tergites of postgeni-tal somites 0 absent 1 present - inapplicable codedonly for Xiphosura (Anderson amp Selden 1997)100 Fusion of tergites of postoral somites VIII and IX(= opisthosomal somites 2 and 3) only 0 absent 1present [GEWB 146]State 1 appears to be a synapomorphy Trigonotarbida(Shear et al 1987) Ricinuleids also have diplotergites(101) but these appear to encompass a different com-bination of somites (see 95)101 Three diplotergites 0 absent presentState 1 is a unique synapomorphy of Ricinulei (Selden1992) See 95102 Paired opisthosomal defensive glands openingvia ducts on either side of the anus 0 absent 1present [S 46 WH 86 GEWB 122]State 1 occurs in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905)103 Muscular diaphragm separating prosomal andopisthosomal compartments formed by dorsoventralmuscles of postoral somites VIndashVIII and extrinsic mus-cles of leg 4 0 absent 1 presentState 1 is known only in extant scorpions (Lankesteret al 1885 Firstman 1973) The diaphragm is oftentreated as a single structure but recent anatomicalwork (Shultz 2007) has shown that it is a composite ofdorsal endosternal suspensors and extrinsic leg mus-cles (see 95) A diaphragm is also present in Solifugaebut it is located more posteriorly and does not appearto be homologous with that of scorpions (Roewer1934)104 Opisthosomal appendicular chondrites 0absent 1 presentThese cartilage-like columns of mesodermally derivedtissue are associated with each opisthosomal append-age including the chilaria in extant Xiphosura(Patten amp Hazen 1900 Yamasaki et al 1988 Fahr-enbach 1999 Shultz 2001)105 Paired appendages on ventral surface of postoralsomite VII (= opisthosomal somite 1) in adult 0absent 1 present [WP sim13 S 39 WH 82 GEWB 143]State 1 is known in Weinbergina (Moore et al 2005)and extant Xiphosura The bilobed structure of themetastoma in certain eurypterids as well as the cor-responding placement of chilaria in Xiphosura haveled some workers to regard the metastoma as fusedappendages of postoral somite VII (eg Stoslashrmer1955) This hypothesis is problematic given that (i) nospecial explanation is needed to account for bilaterallysymmetrical structures in bilaterally symmetricalorganisms (ii) incorporation of the ventral part of thefirst opisthosomal sternite as a functional element of

the prosoma is typical of arachnids and perhapsEuchelicerata generally and (iii) some workers regardthe metastoma as a sternite (eg Jeram 1998) Thusthe metastoma is coded here as uncertain Dunlop hasproposed that the sternum of scorpions is derivedfrom appendages citing the presence of transient limbbuds in scorpion development (Brauer 1895 Patten1912) but this is not compelling evidence for a persis-tent appendicular contribution to the sternite of theadult106 Megoperculum 0 absent 1 present [S sim41 + 42WH sim83 GEWB 162 + 163]State 1 occurs in Thelyphonida Schizomida Ambly-pygi Araneae (Shultz 1993 1999) Trigonotarbida(Shear et al 1987) and probably Haptopoda (Dunlop1999) and Palpigradi The megoperculum consists ofappendages of opisthosomal somite 2 (= genital somiteor postoral somite VIII) (Shultz 1993 1999 Popadicet al 1998) that have fused medially and displacedthe ventral body wall of the somite anteriorly (oftenrepresented by a small sternite to which dorsoventralmuscles of postoral somite VIII attach) The megoper-culum projects posteriorly to form the ventral surfaceof a pregenital chamber The dorsal surface of thechamber is formed by the ventral body wall of opistho-somal somite 3 (= postoral somite IX) which serves asthe ventral attachment of the paired dorsoventralmuscles of that somite The megoperculum bears book-lungs (where present) and sometimes paired gonopodsthat are probably derived from telopodites A similarbut less developed operculum is present on theopisthosomal somite 3 which may also bear booklungsand eversible vesicles (eg Amblypygi) (see 111) corre-sponding to the booklungs and gonopods of the mego-perculum (Shultz 1999)

I contend here that a megoperculum is present inPalpigradi and that this is particularly evident infemales Specifically a large unpaired lobe projectsposteriorly from the genital somite over the ventroan-terior surface of the first postgenital somite therebyforming a pregenital chamber The dorsoventral mus-cles of the genital somite attach ventrally near theanterior border of postoral somite VIII and those ofpostoral somite IX attach to the upper surface of thepregenital chamber (Rucker 1901 Boumlrner 1902bRoewer 1934 Millot 1949d Van der Hammen 1989)Weygoldt (1998) has questioned this interpretation ofPalpigradi107 Postgenital operculum or lsquosternitersquo 0 sclerotized1 not sclerotized - inapplicable coded only for Pan-tetrapulmonata (Platnick amp Gertsch 1976)State 1 is present in Trigonotarbida Amblypygi The-lyphonida Schizomida and Mesothelae (Araneae) butnot Opisthothelae (Araneae)108 Genital opening of female guarded by four plates(one pregenital one postgenital two laterogenitals)

PHYLOGENY OF THE ARACHNID ORDERS 257

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

genital opening of male guarded by two plates 0absent 1 presentThis an apparent synapomorphy of Holothyrida(Acari) (Van der Hammen 1989)109 Anterior margin of genital opening in male withglands secreting via fusules 0 absent 1 presentState 1 occurs in Araneae as epigastric or lsquoepiandrousrsquoglands (Marples 1967) and in Palpigradi (Condeacute1991b)110 Paired valve-like plates apparently formed fromcomponents of three somites covering triradiate genitalopening 0 absent 1 presentThis is an apparent synapomorphy of Acariformes(Evans 1992 Alberti amp Coons 1999)111 Eversible lsquoappendagesrsquo on the ventral surface ofpostgenital somites 0 absent 1 present [S sim43 WHsim84 GEWB sim127]Paired ventral lsquoappendagesrsquo operated in part byhaemolymph pressure occur in the form of ventralsacs in certain Amblypygi (postoral somite IX) (Wey-goldt 2000) (Charinus Millot 1949b not PhrynusShultz 1999) prokoenenian Palpigradi (postoralsomites XndashXII) (Rucker 1901 Condeacute 1991a) as spin-nerets in Araneae (XndashXI) (Shultz 1987) and as geni-tal papillae in many Acariformes (Alberti amp Coons1999) including Oribatida many Endeostigmata (Aly-cus Van der Hammen 1989) and Prostigmata (Allo-thrombium Saboori amp Kamali 2000 MicrocaeculusEvans 1992) Eversible vesicles similar to those ofamblypygids occur in fossils of the trigonotarbidPalaeocharinus (postoral somite IX) (Fayers et al2004) Two cuticular structures (lsquogenital verrucaersquo)occur anterior to the genital opening in Opilioacari-formes and each covers the opening to a thin-walledinvaginated sac (lsquogenital papillarsquo) of unknown function(Van der Hammen 1989) Van der Hammen proposedthat it is homologous to a genital papilla of Acari-formes but Alberti amp Coons (1999) have questionedthis interpretation on several grounds and refer tothese structures as pregenital capsules112 Opisthosomal silk glands and spinnerets derivedfrom appendages on postoral somites X and XI(= opisthosomal somites 4 and 5) 0 absent 1 present[WP 33 WH 41 GEWB 123 + 142]State 1 is a unique synapomorphy of Araneae (Plat-nick amp Gertsch 1976) although potentially homolo-gous glands and spigot-like setae occur on the ventralopisthosomal surface in Palpigradi (Millot 1943Condeacute 1991a)113 Opisthsosomal spinnerets location 0 near mid-dle of opisthosoma 1 near posterior end of opisthso-som ndash inapplicable coded only for Araneae (Platnickamp Gertsch 1976)114 Anterior medial lsquospinneretsrsquo 0 absent 1 present- inapplicable coded only for Araneae (Platnick ampGertsch 1976)

115 Opisthosomal tergites divided longitudinally intoone median and two lateral plates 0 absent 1 present[GEWB 145]State 1 occurs in non-curculioid Ricinulei (Selden1992) and many Trigonotarbida (Dunlop 1996) Thisfeature or something very like it (eg distinct opistho-somal trilobation) occurs sporadically in severaleuchelicerate groups including Chasmataspis (Chas-mataspidida) (Dunlop et al 2004) most fossil xipho-surans (Anderson amp Selden 1997) and a few derivedEurypterida (especially Mixopteroidea) (Tollerton1989)116 Number of metasomal somites 0 zero 1 two 2three 3 five 4 nine [S sim44 WH sim85 GEWB 128 144]The metasoma is a preanal region comprising multiplesomites that are substantially narrower than the pre-ceding somites or mesosoma Metasomal somitesoften lack pleural membranes and take the form ofsclerotized rings State 1 occurs in TrigonotarbidaState 2 is present in synziphosurid Xiphosura (Ander-son amp Selden 1997) Amblypyi Thelyphonida Schizo-mida and Ricinulei State 3 occurs in Scorpiones andEurypterida and State 4 occurs in Chasmatispidida

Some workers appear to consider a five-segmentedmetasoma as conclusive evidence for the monophylyof a Eurypterida + Scorpiones clade (eg Dunlop ampBraddy 2001) presumably because the character rep-resents a kind of tagmosis and should therefore begiven substantial phylogenetic weight However Wey-goldt (1998) has pointed out that a metasoma contain-ing three somites is probably part of the ground planof Euchelicerata The known diversity of the eucheli-cerate metasoma indicates that this feature can inincrease or decrease its segmental composition in evo-lution or be eliminated entirely without necessarilychanging the total number of opisthosomal somitesConsequently it is problematic to assume that thefive-segmented metasoma has any special immunityto homoplasy117 Postanal structure (telson) 0 absent or not obvi-ously developed 1 presentPostanal structures are considered non-segmentalbecause they are modifications of the embryonicregion posterior to the site of somite addition (embry-onic growth zone) A well-developed postanal struc-ture persists in adults in Xiphosura EurypteridaChasmataspidida Scorpiones Schizomida Thely-phonida and Palpigradi The anal operculum of Opil-iones is generally considered by morphologists to bethe tergite of the tenth opisthosomal or anal somite(eg Hansen amp Soslashrensen 1904) but embryological evi-dence (Moritz 1957) suggests that it is derived fromthe postproliferative zone and thus corresponds to thetelson of other chelicerates (see 95)118 Postanal structure shape 0 caudal spine 1aculeus 2 flagellum 3 anal operculum - inapplica-

258 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ble due to absence of postanal structure (117) [WP sim41S sim45 WH sim24 GEWB sim121 129 + 147]State 0 is present in Xiphosura Eurypterida andChasmataspidida State 1 occurs in Scorpiones State2 occurs in Palpigradi Thelyphonida and SchizomidaState 3 occurs in Opiliones (see 95 117)119 Specialized postanal flagellum in male (see 157ndash159) 0 absent 1 present - inapplicable coded onlyfor taxa with postanal flagellum (118) [GEWB 131]State 1 is a unique synapomorphy of Schizomida(Hansen amp Soslashrensen 1905 Cokendolpher amp Reddell1992 Reddell amp Cokendolpher 1995)

RESPIRATORY SYSTEM

120 Respiratory medium 0 water 1 airAmong the terminal taxa included only fossil scorpi-ons are problematic for this character They are codedhere as uncertain Based on a study of book lungmicrosculpture Scholtz amp Kamenz (2006) have arguedthat arachnids are primitively terrestrial and pulmo-nate (see also Firstman 1973) and have questionedwhether any fossil scorpions were aquatic No positionon this proposal is taken here121 Respiratory lamellae on opisthosomal somite 2(= genital somite postoral somite VIII) 0 absent 1present [WP sim37 S sim51 WH sim22 GWEB sim133ndash137]This character encompasses book gills (= lamellae thatfunction in water) and book lungs (= lamellae thatfunction in air) The distinction between book gills andbook lungs is accommodated here by 120 in combina-tion with 121ndash124 Respiratory lamellae are presenton the genital somite in Trigonotarbida AraneaeAmblypygi Thelyphonida and Schizomida Petrunk-evitch (1949) reconstructed Plesiosiro as having booklungs but Dunlop (1999) could not corroborate thisThis character was used by Dunlop amp Webster (1999)to propose that Xiphosura and Scorpiones are closelyrelated because they both lack respiratory lamellae onthe genital somite Dunlop amp Braddy (2001) alsoargued for the placement of Eurypterida with Xipho-sura and Scorpiones based in part on this characterAt least some eurypterids may have had respiratorylamellae (eg Manning amp Dunlop 1995) but the onlyevidence of their segmental distribution is derivedfrom one specimen (Braddy et al 1999)122 Respiratory lamellae on opisthosomal somite 3(= postoral somite IX) 0 absent 1 present [GEWB136]State 1 occurs in Amblypygi Thelyphonida and Xipho-sura It is a groundplan feature of Araneae (Platnick ampGertsch 1976) and is present in all representativetaxa included here Braddy et al (1999) proposed theexistence of respiratory lamellae on postoral somitesIXndashXII in Eurypterida based on evidence from onespecimen This character is coded as uncertain for the

Eurypterida included here Extant scorpions are codedhere as lacking respiratory lamellae on this somite(= pectinal somite) based on the interpretation of theopisthosomal segmentation discussed in 95123 Respiratory lamellae on opisthosomal somites 4ndash6 (= postoral somites XndashXII) 0 absent 1 presentState 1 is definitely present in extant Xiphosura andScorpiones and may have occurred in Eurypterida124 Respiratory lamellae on opisthosomal somite 7(= postoral somite XIII)State 1 occurs in extant Xiphosura and Scorpionesonly125 Kiemenplatten 0 absent 1 presentThese structures are located on the roof of opercularchambers in Eurypterida (Clarke amp Ruedemann1912) where they take the form of ventrally projectingcones with a distinct cuticular microsculputure (Man-ning amp Dunlop 1995) Dunlop amp Braddy (2001)inferred the existence of Kiemenplatten in all Palaeo-zoic scorpions based on a description and photos of onespecimen of Paraisobuthus duobicarinatus by Kjelles-vig-Waering (1986 pls 16ndash18) The plates depict darkcone-like rearward-pointing denticles distributedwithin a white amorphous material The denticlesappear to lack microsculpture even though the mag-nifications at which the photos were taken (times90ndashtimes330)are comparable with those illustrating the cones ofKiemenplatten (times170ndashtimes350 in Manning amp Dunlop1995 figs 1 2) Given the diversity of cuticular struc-tures present in the atria of booklungs book gills andtracheae the denticles appear to bear no special sim-ilarity to the cones of Kiemenplatten126 Tracheal system 0 absent 1 paired ventral stig-mata on postoral somite VIII (= opisthosomal somite2) 2 paired ventral stigmata on postoral somites IXand X 3 one pair of stigmata opening near legs 3 or 44 paired stigmata associated with chelicerae 5 fourpairs of stigmata on dorsal surface of opisthosoma -inapplicable aquatic (120) [WP 40 sim43 S sim52 + 5354 WH sim23 45 88 GEWB sim138 139]Firstman (1973) and Weygoldt amp Paulus (1979)hypothesized that tracheae are homologous in all tra-cheate arachnids except spiders and subsequentworkers have entertained this hypothesis by includinga character for the presenceabsence of tracheal sys-tems that ignores the diverse arrangement of stig-mata in arachnids (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) However thisapproach assumes that internal tracheal systems areconserved but that tracheal openings (stigmata)appear and disappear on different parts of the bodywith higher evolutionary frequency Here I assumethat stigmata are conserved in evolution and that dif-ferences in their anatomical placement reflect the evo-lution of new tracheae State 1 occurs throughoutOpiliones State 2 occurs throughout Solifugae

PHYLOGENY OF THE ARACHNID ORDERS 259

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

(Roewer 1934) and Pseudoscorpiones (Chamberlin1931) State 3 recognizes the possible homology of stig-mata in Ricinulei Opilioacariformes and Parasiti-formes (Acari) (Lindquist 1984 Van der Hammen1989) Tracheal systems are apparently absent in thegroundplan of Acariformes but Prostigmata typicallyhave tracheae associated with the chelicerae Tra-cheae derived from the posterior pair of book lungs arewidespread in araneomorph Araneae but the booklungs are retained and tracheae are absent in themost basally divergent groups (eg Hypochilus)

BOXndashTRUSS AXIAL MUSCLE SYSTEM (BTAMS)

127 Posterior oblique muscles of BTAMS of postoralsomites IndashVI 0 absent 1 present in one or moresomitesState 0 occurs in Xiphosura and state 1 occurs inPalpigradi Araneae Amblypygi Thelyphonida andSchizomida (Shultz 2001) The condition in AcariRicinulei Opiliones and Solifugae is not known128 Anterior oblique muscles of BTAMS posterior topostoral somite VI 0 absent 1 presentState 1 occurs in Xiphosura which is probably theprimitive condition based on comparison with otherarthropods (Shultz 2001) State 0 occurs in all arach-nids examined thus far129 Ventral attachments of posterior oblique musclesof opisthosomal BTAMS located in prosoma 0 absent1 presentState 1 occurs in extant Xiphosura (Shultz 2001)130 Endosternite fenestrate 0 absent 1 present [S 8WH 54 GEWB 35]State 1 occurs in Thelyphonida and hubbardiid Schi-zomida (Firstman 1973) The condition in protoschi-zomid Schizomida is not known131 Suboral suspensor a tendon that arises from theBTAMS and inserts on the ventral surface of the oralcavity via muscle 0 absent 1 presentState 1 occurs in Palpigradi (Eukoenenia Millot1943) Amblypygi (Shultz 1999) and Thelyphonida(Shultz 1993)132 Perineural vascular membrane in adult 0absent 1 present [WH 28 GEWB 155]Wheeler amp Hayashi (1998) coded extant Xiphosura asunknown although the primary source for this char-acter (Firstman 1973) stated that adult Limulus havea perineural vascular membrane Within Arachnidapresence of a perineural vascular membrane is appar-ently correlated with the presence of tracheae (126)133 Ventral endosternal suspensor attaching on coxaof anteriorly adjacent somite 0 absent 1 present[GEWB 239 253]State 1 occurs in Amblypygi (Shultz 1999) Thely-phonida (Shultz 1993) and Schizomida (Shultzunpubl observ)

134 Posteriormost postoral somite with a pair of dor-soventral muscles 0 VI 1 VII 2 VIII 3 XII 4 XIII5 XIV 6 XV 7 XVIDorsoventral muscles tend to run in a continuousmetameric series beginning in the prosoma Deter-mining the last tergite on which the muscle seriesends is substantially easier than counting especiallygiven frequent anatomical complexities near theprosomandashopisthosoma juncture The dorsoventralmuscle series ends on postoral somite VII inCyphophthalmi (eg Chileogovea) and Laniatores(Opiliones) (Shultz unpubl observ) on postoralsomite VII or VIII in Eupnoi (Leiobunum Shultz2000 unpubl observ) on postoral somite XII in Pal-pigradi (Roewer 1934) Solifugae (Bernard 1896Roewer 1934 Millot amp Vachon 1949) and Araneae(Liphistius Hypochilus Millot 1933 but segmenta-tion reinterpreted here following Shultz 1993 1999)on postoral somite XIII in hubbardiid Schizomida(Cokendolpher amp Reddell 1992) and Scorpiones(Lankester et al 1885) and on postoral somite XIVin extant Xiphosura (Shultz 2001) Thelyphonida(Shultz 1993) protoschizomid Schizomida (Cokendol-pher amp Reddell 1992) and Amblypygi (Shultz 1999)Dunlop (1999) reconstructed Plesiosiro as havingpaired tergal apodemes ending on postoral somite XVbut his figures indicate that the series ends on posto-ral somite XIV The interpretation of opisthosomalsegmentation in Ricinulei coded here (95) indicatesthat the series ends on postoral somite XV in Terpsi-croton and this is treated as the groundplan for theorder The posteriormost dorsoventral muscles inPseudoscorpiones occur on postoral somite XVI(Vachon 1949)

NERVOUS SYSTEM

135 Segmental ganglia 0 consolidated in prosoma1 one or more present in opisthosoma [WH 30 GEWB210 + 216]Adult neuromeres are exclusively prosomal in Ambly-pygi liphistiomorph and araneomorph Araneae Opil-iones Pseudoscorpiones Ricinulei and Acari (Millot1949a)136 Dorsal median eyes 0 absent 1 present [WP14 + 47 S 50 WH sim7 GEWB 1]Median eyes are a groundplan feature of Euchelicer-ata and are retained in extant Xiphosura Eurypteridaand occur in all arachnid orders except PalpigradiSchizomida Ricinulei and Pseudoscorpiones (Paulus1979) Median eyes are also absent in opilioacariformand parasitiform Acari (Lindquist 1984) and perhapsthe fossil whipscorpion Proschizomus (Dunlop amp Hor-rocks 19951996) They are unknown in synziphosu-ran Xiphosura Median eyes are present in a variety ofbasal acariform mites including Endeostigmata (but

260 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

not Alycus) Prostigmata (Microcaeculus) and certainoribatids (Palaeacarus) (Evans 1992 Alberti amp Coons1999) Giribet et al (2002) coded all representativeAcari as lacking median eyes It is not known whetherthe eyes of some cyphophthalmid opilions are medianor lateral evidence from tracheal branching inCyphophthalmus (Janczyk 1956) suggests that theyare median eyes (Shultz amp Pinto da Rocha 2007) andpresence of a tapetum in Stylocellus is consistent withlateral eyes (Shear 1993)137 Retinula cells of dorsal median eyes 0 organizedinto closed rhabdoms 1 organized into network ofrhabdomeres 2 disorganized - inapplicable due toabsence of median eyes (136) [GEWB sim3]State 0 is present in Scorpiones Thelyphonida andAmblypygi State 1 is present in Solifugae and Ara-neae state 2 is present in Xiphosura (Paulus 1979)The retinula cells of median eyes have been studied inseveral prostigmatid Acariformes and are organized ina network in some taxa and in an irregular pattern inothers (Alberti amp Coons 1999) Retinulae in phalangidOpiliones have state 0 proximally and state 2 distally(Schliwa 1979)138 Ventral median eyes 0 absent 1 presentState 1 occurs in early instars of extant Xiphosura(Paulus 1979)139 Lateral eyes 0 absent 1 presentLateral eyes are primitively present in Chelicerataand are absent in Palpigradi and Opiliones (Paulus1979) It is unclear whether the eyes of cyphoph-thalmid opilions are median or lateral (see 136)140 Arrangement and number of lateral eyes 0 com-pound many 1 five or more pairs (includes micro-lenses) 2 three primary pairs (excludes microlenses)3 two pairs 4 one pair - inapplicable due to absenceof lateral eyes (139) [WP sim13 + 18 + 38 + 44 + 52 Ssim49 WH sim10 GEWB sim4 + 5]True compound eyes are present in XiphosuraEurypterida Chasmataspidida and many PalaeozoicScorpiones Among the anactinotrichid Acari Opilioa-cariformes have two (Neocarus) or three (Siamacarus)pairs of lateral eyes (although at least one species ofSiamacarus lacks eyes) the allothryrid Holothyridaand many Ixodida have a single pair of lateral lensesand the remainder apparently lack eyes (Evans1992) Thelyphonida was coded as lsquo12rsquo to reflect fivepairs of lenses comprising three pairs of primarylenses and two pairs of small accessory lenses (notsimply three pairs as coded by Giribet et al 2002)Trigonotarbida is also coded as lsquo12rsquo to reflect threepairs of primary lenses and multiple small accessorylenses Dunlop (1999) illustrated Haptopoda as havingpaired triads of lateral eyes but noted that there wasactually no evidence of this in the fossils The lateraleyes of fossil ricinuleids have two pairs of lenses(Selden 1992) One pair of eyes or eyespots are

present in extant Ricinulei and many Schizomidaalthough five genera of hubbardiid Schizomida have apair of lenses (Reddell amp Cokendolpher 1995) State 1is the groundplan for extant Scorpiones and State 3occurs in the pseudoscorpions Chthonius FeaellaNeobisium (but State 4 in Chelifer) and Solifugae(Paulus 1979)141 Lateral eyes with closed rhabdoms 0 absent 1present - inapplicable due to absence of lateral eyes(139) [WP 21 WH 12 GEWB 6]Presence of closed rhabdoms is probably primitive forChelicerata and is retained in extant Xiphosura andScorpiones Retinula cells form a network of rhab-domeres in other extant chelicerates (Paulus 1979)but these networks can differ substantially in detailWeygoldt (1998) and other workers have given sub-stantial weight to the network character in unitingnon-scorpion arachnids but these authors seem not togrant comparable phylogenetic significance to theanalogous character of the median eyes shared bySolifugae and Araneae (137)142 Slit sensilla 0 absent 1 present [WP 19 S 47WH 11 GEWB 209]State 1 occurs in all arachnid orders except Palpigradi(Shultz 1990) The proposal that lsquoprimitiversquo slit sen-silla were present in Eurypterida (Dunlop amp Braddy1997) appears to be based on the over-interpretation ofa comparatively large notch that occurs in Baltoeu-rypterus at the terminus of the tibia (= podomere 7)(see also Edgecombe et al 2000 Giribet et al 2002)The neural construction of slit sensilla like function-ally similar campaniform sensilla of hexapods is sim-ilar to that of trichoid sensilla (Chapman 1998Klompen 2000) Thus these cuticular stress receptorsmay represent modified bases of sensory setae a viewsupported by the replacement of setae by slit sensilladuring post-embryonic development in some Opilioac-ariformes (Klompen 2000)143 Trichobothria 0 absent 1 present [GEWB 213]Trichobothria are present in extant Scorpiones(Jeram 1998) Pseudoscorpiones (Chamberlin 1931)some endeostigmatid (Alycus Van der Hammen 1989)and prostigmatid Acariformes and most Oribatida(Lindquist 1984) They occur on the ventral surface ofthe pedal femora in the opilioacariform Siamacarus(Leclerc 1989) They are also present in Araneae (Foe-lix 1996) Amblypygi (Weygoldt 2000) Schizomidaand Thelyphonida (Hansen amp Soslashrensen 1905) Theyare apparently absent in all non-arachnid chelicer-ates Solifugae Ricinulei Opiliones (Reissland ampGoumlrner 1985) and Parasitiformes (Acari) (Lindquist1984)144 Tibial trichobothria with 2-1-1-1 pattern onappendages IIIndashVI (= arachnid legs 1ndash4) 0 absent 1present - inapplicable due to absence of trichobothria(143) [S 48 WH 87 GEWB 88]

PHYLOGENY OF THE ARACHNID ORDERS 261

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs only in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905) Note that Shultz (1990)erroneously described this character as a 2-2-1-1 pat-tern and that this error was repeated by Giribet et al(2002) and described as 2-1-1 by Wheeler amp Hayashi(1998)145 Paired trichobothria on dorsal surface ofprosoma 0 absent 1 present - inapplicable due toabsence of trichobothria (143)State 1 is an apparent synapomorphy of Acariformesand is present in all acariforms included in this study(Alberti amp Coons 1999)146 Malleoli 0 absent 1 present [WP 45 GEWB 96]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)147 Pectines 0 absent 1 present [WP 20 WH 38GEWB 120]State 1 is a unique synapomorphy of Scorpiones148 Intercheliceral median organ 0 absent 1present [GEWB 212]State 1 is a unique synapomorphy of Palpigradi(Roewer 1934)149 Tarsal organ on appendage of postoral somite III(= arachnid leg 1) (= Hallerrsquos organ) sensilla containedwithin a cuticular depression on the superior surface ofthe tarsus of appendage III (= arachnid leg 1) 0absent 1 present [GEWB sim100]Klompen (2000) has noted that State 1 occurs in Opil-ioacariformes Parasitiformes (except Mesostigmata)and Ricinulei where it also occurs on leg 2 (Talaricoet al 2005) State 1 occurs on all legs in Araneae andperhaps Scorpiones (Foelix 1985)150 Tarsal organ on appendage of postoral somite IV(= arachnid leg 2) 0 absent 1 present (See 149)

REPRODUCTION

151 Gonads 0 primarily prosomal 1 primarilyopisthosomalState 0 is limited to Xiphosura (Sekiguchi 1988)State 1 occurs throughout Arachnida (Millot1949a)152 Ladder-like opisthosomal gonadsaccessoryglands (see 153) 0 absent 1 present [WH sim37 GEWBsim158]Giribet et al (2002) followed Wheeler amp Hayashi(1998) who followed Clarke (1979) in coding gonads asreticulate (Xiphosura) ladder-like (Scorpiones Thely-phonida Schizomida) or lsquosaccularrsquo (all remainingArachnida) However comparisons between Xipho-sura and Arachnida are problematic given that thexiphosuran gonads are primarily prosomal and thoseof arachnids are primarily opisthosomal (151) Fur-ther the reticulate pattern in Limulus (Xiphosura)but not other extant xiphosurans contains a distinctlyladder-like component The lsquosaccularrsquo state is probably

artificial as it encompasses a wide variety of pairedand unpaired structures153 Male gonads in two distinct parts one producingsperm and another (tubular gland) producing a holo-crine secretion similar to degenerate sperm 0 absent1 presentState 1 occurs in Thelyphonida Schizomida andAmblypygi although the holocrine material is pro-duced by ventral organs in Amblypygi and dorsalorgans in Thelyphonida and Schizomida (Alberti2005)154 Number of gonopores 0 two 1 oneExtant xiphosurans have two small genital openingson the base of the genital telopodite and all extantarachnids have a single opening The condition inEurypterida is not known Clarke amp Ruedemann(1912) located a pair of openings near the base of themedian organ (161) that are the outlets of the lsquohornorgansrsquo but it is unclear whether these are genitalducts or accessory structures Braddy amp Dunlop (1997)have developed numerous speculations about thesestructures and extended their arguments far beyondthe available evidence155 Genital opening (ie gonopore or gonostome)appearing to open in prosomal region (ie between legcoxae or anterior to posterior carapacal margin) 0absent 1 present [WP sim50 WH 26 GEWB 166]The genital opening in Euchelicerata is located onpostoral somite VIII (= opisthosomal somite 2) but ithas shifted anterior to the posterior margin of thecarapace or between the coxae of the last pair of legsin most Scorpiones (but not in PalaeoscorpiusKjellesvig-Waering 1986) and Opiliones The geni-tal opening occurs near or anterior to the last coxaein the opilioacariform and parasitiform Acari repre-sented here It is variable in Acariformes but islocated posterior to the coxae in all representativetaxa156 Ovipositor 0 absent 1 present [WP sim51 S sim60WH sim91 GEWB sim172]An ovipositor with a trilobed terminus is an apparentgroundplan character of Oribatida (Lindquist 1984Alberti amp Coons 1999) An ovipositor is also present inOpilioacariformes and Opiliones (Van der Hammen1989)157 Stalked spermatophore attached to substratum0 absent 1 present [S 57]State 1 occurs in Scorpiones PseudoscorpionesAmblypygi Thelyphonida and Schizomida The mech-anism of sperm transfer is unknown in PalpigradiOpilioacariformes and Holothryrida158 Male turns posterior end toward female duringmating behaviour 0 absent 1 presentState 1 occurs in Amblypygi (Weygoldt 2000) The-lyphonida and Schizomida (Weygoldt amp Paulus1979)

262 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

159 Female grasps male opisthosoma during matingbehaviour 0 absent 1 present [WP 29 S 58 WH 19GEWB 188]State 1 occurs in Thelyphonida and Schizomida (Wey-goldt amp Paulus 1979) Evolution of this behaviour wasprobably facilitated by 158160 Penis 0 absent 1 present [WP sim51 GEWB 167]State 1 is a unique synapomorphy of Opiliones A truepenis occurs in Phalangida A clearly homologousstructure is present in Cyphophthalmi (Opiliones) andapparently functions in depositing a spermatophore inthe femalersquos genital chamber (Karaman 2005) A truepenis may occur in some mites but it does not appearto be a groundplan feature of any major group (Evans1992) The lsquopenisrsquo in Oribatida is really a spermatop-ositor it functions in construction of a spermatophore(Alberti amp Coons 1999)161 Median organ 0 absent 1 presentState 1 occurs throughout Eurypterida (Clarke amp Rue-demann 1912) and has been observed in diploaspididChasmataspidida (Dunlop 2002a also Loganamaras-pis Tetlie amp Braddy 2004)

SPERM MORPHOLOGY

162 Nucleus with manchette of microtubules 0absent 1 present [S 54 WH 89 GEWB 193]State 1 occurs in Araneae Amblypygi Thely-phonida Schizomida Ricinulei and gonyleptid Lani-atores (Opiliones) (Alberti 1995 Giribet et al2002)163 Axoneme 0 absent 1 present [WP sim49 S sim55WH sim25 GEWB 195]State 1 occurs in Xiphosura Scorpiones Cyphoph-thalmi (Opiliones) Pseudoscorpiones RicinuleiAraneae Amblypygi Thelyphonida and Schizomida(Alberti 1995)164 Coiled axoneme 0 absent 1 present - inappli-cable due to absence of axoneme (163) [WP 22 S sim55WH 13 GEWB 196]State 1 occurs in Pseudoscorpiones Ricinulei Ara-neae Amblypygi Thelyphonida and Schizomida(Alberti 1995)165 Microtubule arrangement in axoneme 0 9 + 0 19 + 1 2 9 + 2 3 9 + 3 - inapplicable due to absence ofaxoneme (163) [WP sim26 S sim56 WH sim16 GEWB 198](Alberti 1995)166 Helical or corkscrew shaped nucleus 0 absent 1present [GEWB sim204]State 1 occurs in Araneae Amblypygi ThelyphonidaSchizomida and certain Scorpiones (eg Hadrurus)(Alberti 1995)167 Vacuolated-type sperm 0 absent 1 present[GEWB 205]State 1 is unique to Opilioacariformes and Parasiti-formes (Acari) (Alberti 1995)

168 Sperm aggregates 0 absent 1 present [GEWB206] (Alberti 1995)

DEVELOPMENT

169 Yolk in early embryo 0 concentrated (centroleci-thal or telolecithal) 1 evenly distributed (isolecithal)[WH sim34 GEWB sim191]Wheeler amp Hayashi (1998) and Giribet et al (2002)coded this character as lsquo0 isolecithal or telolecithal 1centrolecithalrsquo based on information presented byYoshikura (1975) Their coding appears to highlightseparate conditions found in Scorpiones (ie isoleci-thal and telolecithal) rather than a property intrinsicto the character itself170 Embryonic nutrition other than yolk 0 absent 1present [WP sim49]State 1 occurs in extant Scorpiones (except buthidsand chaerilids) and Pseudoscorpiones (Weygoldt1969)171 Embryological growth zone 0 initiating segmentaddition within prosoma 1 initiating segment addi-tion posterior to prosoma [GEWB sim192]Giribet et al (2002) followed Dunlop amp Webster (1999)in stating that only Xiphosura and Scorpiones havestate 0 among Chelicerata However the last prosomalsomite develops from the growth zone in Ixodida(Anderson 1973 Evans 1992) A similar processapparently occurs in the acariform mites Tyroglyphus(Sarcoptiformes Astigmata) (Yoshikura 1975) andArchegozetes (Sarcoptiformes Oribatida) (Thomas ampTelford 1999) but not apparently in Tetranychus(Prostigmata) (Dearden Donly amp Grbic 2002)172 Eggsembryos maintained in external attachedbrood sac secreted by genital glands 0 absent 1present [WP sim46 S 59 WH 90 GEWB 208 = 219]State 1 occurs throughout Amblypygi SchizomidaThelyphonida and Pseudoscorpions Giribet et al(2002) followed Shear et al (1987) in homologizingspinneret-derived silken egg sacs of spiders with gen-ital-gland-derived brood sacs of Pedipalpi However asthese structures are not homologous in secretory ori-gin construction or composition (Shultz 1987) theyare not regarded here as homologous Giribet et al(2002) miscoded the character as absent in Pseudo-scorpiones and without justification in PalpigradiFemale palpigrades have well-developed glands asso-ciated with postgenital somites and an array of setaefusules and other structures (112) (Boumlrner 1902b Mil-lot 1942 Condeacute 1991a) which are consistent withsome sort of brood care and Palpigradi should there-fore be coded as unknown for this character173 Embryonic and early postembryonic lsquolateralrsquo orClaparegravede organs associated with coxa of postoralsomite IV (= arachnid leg 2) 0 absent 1 present [WHsim31 GEWB sim211]

PHYLOGENY OF THE ARACHNID ORDERS 263

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bilaterally paired rounded protuberances locatedbetween coxae of appendages of postoral somites IIIand IV (arachnid legs 1 and 2) in prelarval and larvalinstars appear to be a primitive feature of Acariformes(Lindquist 1984 Evans 1992 Alberti amp Coons 1999)It is known to be derived embryologically from thecoxa of leg 2 in the oribatid Archegozetes (Thomas ampTelford 1999) Structures apparently homologouswith the acariform Claparegravede organ are present inembryonic Amblypygi (Weygoldt 2000) and Thely-phonida (Yoshikura 1975) and in embryonic and earlypost-embryonic stages of Solifugae (Roewer 1934)where they are termed lateral organs The so-calledlsquolateral organrsquo of embryonic Xiphosura (174) occurs onthe lateral surface of the carapace and does not appearto be homologous with the coxa-associated structuressharing their name In fact the xiphosuran lateralorgan develops even when the coxa of postoral somiteIV has been excised (Sekiguchi 1988) In contrastWheeler amp Hayashi (1998) and Giribet et al (2002)followed Yoshikura (1975) in regarding all lsquolateralorgansrsquo as homologous and in miscoding this characteras absent in Acariformes Van der Hammen (1989)suggested that the sternal verrucae of Opilioacari-formes are homologous with the Claparegravede organ butit differs in gross structure and ontogenetic timing174 Embryonic lateral organ associated with cara-pace 0 absent 1 present [WH sim31 GEWB sim211]State 1 is known only from embryonic stages of extantXiphosura (Yoshikura 1975) In contrast to previousinterpretations this feature does not appear to behomologous with the lsquolateralrsquo or Clarapegravede organ ofcertain arachnids (173)175 Live birth 0 absent 1 presentState 1 occurs in extant Scorpiones and has evolvedseveral times in a few mites (Evans 1992)176 Hexapodal larva and 1ndash3 nymphal stages 0absent 1 present [WP 53 S sim61 WH 27 GEWB 186]State 1 occurs in Acari and Ricinulei (Lindquist1984)177 Hexapodal prelarva 0 absent 1 present [S sim61WH 47]State 1 occurs in Opilioacariformes (NeocarusKlompen 2000) and many Acariformes (Evans 1992)

EXCRETORYOSMOREGULATORY SYSTEM

178 Malpighian tubules 0 absent 1 present [WP 17S 62 WH 9 GEWB 153]State 1 occurs in all major arachnid groups except Pal-pigradi Pseudoscorpiones Opiliones Oribatida (Sar-coptiformes) and Prostigmata Within Acari State 1occurs in all major anactinotrichid lineages (Opilioac-ariformes Holothyrida Mesostigmata Ixodida) andsimilar structures are present in certain Astigmata(Evans 1992 Alberti amp Coons 1999)

179 Dorsomedian excretory organ 0 absent 1presentThis is a specialized excretoryosmoregulatory organformed by the postventriculus and proctodeum inProstigmata (Alberti amp Coons 1999)180 Adult coxal organ opening on or near coxa ofappendage III (= arachnid leg 1) 0 absent 1 present[S 64 WH 93 GEWB 90]State 1 occurs throughout Acari (Evans 1992 Albertiamp Coons 1999) Ricinulei (Pittard amp Mitchell 1972Legg 1976) Palpigradi (Millot 1942) AraneaeAmblypygi Thelyphonida and Schizomida (Buxton1913 1917) There is evidence that the second lsquoozo-porersquo of gonyleptids (Opiliones Laniatores) (Hara ampGnaspini 2003) is a persistent opening to the coxalorgan associated with leg 1 (Soslashrensen 1879) whichalso occurs in embryonic Phalangium (Opiliones Eup-noi) (Moritz 1959)181 Adult coxal organ opening on or near coxa ofappendage V (= arachnid leg 3) 0 absent 1 present [S63 WH 92 GEWB 89]State 1 occurs in extant Xiphosura (Yamasaki et al1988) Scorpiones mygalomorph (Buxton 1913 1917)and mesothele Araneae (Liphistius J Millot in Bris-towe 1932) State 1 occurs in basal Amblypygi(Charon Charinus Buxton 1913 1917) but degener-ates prior to the adult stage in higher groups(Phrynus Weygoldt 2000) Adequately preservedeurypterids show a small submarginal opening on thecoxa of appendage V (Baltoeurypterus Selden 1981also Eurypterus Hughmilleria Clarke amp Ruedemann1912) and this is interpreted here as the opening to acoxal organ Giribet et al (2002) coded this characteras uncertain for Eurypterida182 Adult coxal organ opening on coxa of appendageII (= arachnid palp) 0 absent 1 present [GEWB 65]Buxton (1913 1917) noted that the coxal organ ofSolifugae opens on the prolateral surface of the palpalcoxa The proximal end of the duct leading to the ori-fice is associated with glands and apparently theglandular secretion andor coxal fluid serve as saliva(Alberti 1979) (see also 14 15)183 Genital papillae 0 absent 1 present - inappli-cable coded only for those taxa with eversible lsquoappend-agesrsquo (111)The genital papillae are associated with the genitalopening in postlarval Acariformes (Evans 1992Alberti amp Coons 1999) These structures are probablyserially homologous with the prosomal Claparegravedeorgan (173) and both structures probably function inwater and ion regulation

DIGESTIVE SYSTEM

184 Ingestion 0 solid food 1 primarily liquid foodwith or without preoral digestion [WP sim16 WH sim8GEWB sim218]

264 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Extant xiphosurans ingest solid food and members ofmost extant arachnid orders ingest fluids Opilionesand Acari are the principal exceptions The phalangidOpiliones ingest solids but Shultz (2000 unpublobserv) has shown that the precerebral pharyngealapparatus of Cyphophthalmi is very similar to that ofthe fluid-feeding Scorpiones and differs substantiallyfrom phalangid Opiliones Examination of gut con-tents of the cyphophthalmids Siro acaroides S exilisand Chileogovea oedipus have failed to reveal solidparticles typically found in the guts of phalangids(Shultz unpubl observ) the character is coded asunknown for Cyphophthalmi Within Acari consump-tion of particles has been documented in Opilioacari-formes (Neocarus Van der Hammen 1989unidentified Walter amp Proctor 1998) Holothyridanon-parasitic Sarcoptiformes (Evans 1992 Walter ampProctor 1998 Alberti amp Coons 1999) and many endeo-stigmatids including the nematophagous Alycusroseus (Walter 1988)185 Mouth 0 directed posteroventrally 1 directedanteroventrally [S 9 WH 55 GEWB 28]State 1 occurs throughout Arachnida (Shultz 1990)186 Oral cavity dilated by muscles arising from coxaeand constricted by large circular sphincter 0 absent1 presentState 1 occurs in extant Xiphosura (Manton 1964Scholl 1977 Shultz 2001)187 Palate plate 0 absent 1 present [GEWB 159]State 1 is a unique synapomorphy of Araneae (Dunlop1994)188 Lateral walls of epistome broadly fused to medialwalls of palpal coxae opposite sides connected by well-developed transverse epistomal muscle 0 absent 1presentState 1 occurs throughout Opiliones and Scorpiones(Shultz 2000 unpubl observ)189 Epistome with a pair of lateral arms that projectsposteriorly into the prosoma on either side of the phar-ynx 0 absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)190 Epistome with four pairs of suspensor musclesattaching to the carapace 0 absent 1 presentState 1 occurs in extant Scorpiones (Lankester et al1885 Vyas 1970 Shultz unpubl observ) and appar-ently in Solifugae (Roewer 1934)191 Intercheliceral epipharyngeal sclerite 0 absent1 present [GEWB 31]State 1 occurs in Palpigradi (Boumlrner 1904) Trigo-notarbida (Palaeocharinus Dunlop 1994) Araneae(Firstman 1954 Marples 1983) Amblypygi (Mil-lot 1949b Shultz 1999) Schizomida (Millot1949e) Thelyphonida (Millot 1949e Shultz 1993)and Phalangida (Opiliones) (Shultz 2000 unpublobserv)

192 Epipharyngeal sclerite large projecting posteri-orly 0 absent 1 present - inapplicable due to absenceof sclerite (191) [GEWB 32]State 1 occurs in Amblypygi (Millot 1949b Shultz1999) Schizomida (Millot 1949e) Thelyphonida (Mil-lot 1949e Shultz 1993) some Araneae (Heptathela ndashcoded as Liphistius ndash and Hypochilus Marples 1983not Aphonopelma Firstman 1954) and perhaps Trig-onotarbida (Palaeocharinus Dunlop 1994)193 Dorsal dilator muscle of precerebral pharynxattaching to intercheliceral septum or associatedepipharyngeal sclerite (191) 0 absent 1 presentState 1 occurs in Palpigradi (Millot 1943) Araneae(Marples 1983) Amblypygi Schizomida and Thely-phonida (Millot 1949e Shultz 1993 1999)Extremely fine muscles have been documented insome Scorpiones (Shultz 2007) and in Leiobunum(Opiliones) (Shultz 2000) Lankester et al (1885)reported a muscle in Limulus that would appear tocorrespond to this muscle but the muscle does notexist (Manton 1964 Shultz 2001)194 Dorsal dilator muscle of precerebral pharynxattaching to dorsal surface of prosoma 0 absent 1present [GEWB 228]State 1 occurs in Palpigradi (Roewer 1934) and mostAraneae (Palmgren 1978 Aphonopelma Firstman1973) but not in Heptathela (coded as Liphistius) orHypochilus (Marples 1983)195 Lateral dilator muscle of precerebral pharynxattaching to endosternite 0 absent 1 present [GEWBsim226]State 1 is a unique feature of Araneae (Marples 1983)196 Lateral dilator muscle of precerebral pharynxattaching to lateral surface of epistomal processes 0absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)197 Lateral dilator muscle of precerebral pharynxattaching to medial process of coxa of appendage II(= arachnid palp) 0 absent 1 present [GEWB sim226]State 1 occurs in Amblypygi Thelyphonida and Schi-zomida (Boumlrner 1904 Shultz 1993 1999)198 Dilator muscle of precerebral pharynx andorpreoral cavity attaching to ventral surface of prosoma0 absent 1 present [GEWB 227]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) Araneae (Marples 1983) Palpigradi(Millot 1943) and Solifugae (Roewer 1934)199 Postcerebral pharynx 0 absent 1 present [WPsim31 S sim5 WH sim21 GEWB sim154]The cuticle-lined foregut passes through the centralnervous system in many arthropods That portion pos-terior to the CNS is here termed the postcerebralpharynx regardless of the details of its morphology orfunctional specialization A substantial postcerebralpharynx is known in extant Xiphosura (Lankester

PHYLOGENY OF THE ARACHNID ORDERS 265

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

et al 1885 Manton 1964 Yamasaki et al 1988)Scorpiones (Centruroides Hadrurus HeterometrusShultz 2007 also Androctonus Abd el-Wahab 1952)Solifugae (Millot amp Vachon 1949) Araneae Ambly-pygi Thelyphonida and Schizomida (Millot 1949andasheShultz 1993) Weygoldt amp Paulus (1979) applied theterm lsquopostcerebral pharynxrsquo only to the lsquosucking stom-achrsquo of Araneae and Amblypygi but the sucking stom-ach is here regarded as a complex of three characters199ndash201

200 Dilator muscle of postcerebral pharynx attach-ing to endosternite 0 absent 1 present - inapplicabledue to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Lankester et al1885 Manton 1964 Shultz 2001) Scorpiones (Cen-truroides Hadrurus Heterometrus Shultz 2007 alsoAndroctonus Abd el-Wahab 1952) Araneae (First-

man 1954 Palmgren 1978) Amblypygi (Millot1949b Shultz 1999) and apparently the palpigradeProkoenenia wheeleri (Rucker 1901) but not inEukoenenia miriabilis (Millot 1943 contra Boumlrner1904)201 Dilator muscle of postcerebral pharynx attachingto dorsal surface of prosoma 0 absent 1 present -inapplicable due to absence of postcerebral pharynx(199)State 1 occurs throughout Araneae (Millot 1949cFirstman 1954 Palmgren 1978) and Amblypygi(Charinus Millot 1949b Phrynus Shultz 1999)202 Crop and gizzard 0 absent 1 present - inap-plicable due to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Sekiguchi 1988)but not in extant Arachnida

PHYLOGENY OF THE ARACHNID ORDERS 229

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Pseudoscorpiones The pseudoscorpions (24 families425 genera ~3200 spp) (Harvey 2002) were repre-sented in the matrix by four terminal genera Chtho-nius (Chthoniidae) was drawn from the basallydivergent superfamily Chthonioidea (Harvey 1992)Feaella (Feaellidae) represented the enigmatic super-family Feaelloidea which has an unusual combinationof characters and may be an early divergent lineageclose to Chthonioidea (Harvey 1992) Neobisium (Neo-bisiidae) is morphologically intermediate between theearly divergent families and the higher families rep-resented by Chelifer (Cheliferidae)

Solifugae (= Solpugida) (sun spiders) The order issmall (~1000 spp) but the absence of modern phylo-genetic treatments precludes meaningful estimates ofgenera and families (Harvey 2002) The taxon sampleincluded two genera one from the New World Eremo-costa and one from the Old World Galeodes

Opiliones (harvestmen) Extant harvestmen (~25 fam-ilies ~500 genera ~6000 spp) are divided into twosuborders Cyphophthalmi and Phalangida Cyphoph-thalmids were represented here by Chileogovea oedi-pus (Petallidae) which is currently under study by theauthor and Cyphophthamus duricorius (Sironidae)which is probably the most well-studied member of thesuborder (eg Janczyk 1956) Phalangida includesthree major groups Laniatores Eupnoi and Dyspnoibut there is disagreement as to whether Dyspnoi is thesister group to Laniatores or to Eupnoi (Shultz ampRegier 2001 Giribet et al 2002) Laniatores was rep-resented by Sclerobunus and Gonyleptes and Eupnoiby Leiobunum and Caddo

Scorpiones (scorpions) The extant scorpions (16 fam-ilies 155 genera 1279 spp) (Fet et al 2000) wererepresented by three taxa Centruroides vittatus(Buthidae) Hadrurus arizonensis (Iuridae) and Heter-ometrus spinifer (Scorpionidae) Buthidae is widelyregarded as the sister group to other extant lineagesand similarities among the represented terminals arelikely to be ground plan features of extant scorpionsgenerally The morphology of fossil scorpions is sub-stantially more diverse (Kjellesvig-Waering 1986)but their phylogenetic relationships are unclear (butsee Jeram 1998) The fossil taxa used in this study(Proscorpius Stoermeroscorpio Palaeoscorpius Pre-arcturus) were chosen for quality of preservation andor presence of a phylogenetically significant constella-tion of characters

CHARACTER CODING

Most character states were determined from directobservation the primary literature and authoritativereviews In some cases states were assigned to termi-

nals based on observations from related species asnoted in the Appendix In the matrix state lsquo-rsquo indicatesthat the character is inapplicable because the taxonlacks a more general character For example if a taxonlacks eyes then special features of the eyes (numberposition retinal configuration etc) are inapplicable tothat taxon A lsquorsquo indicates that the state is unknown oruncertain An entry with multiple states (eg 01 345) should be treated as an ambiguity code not a poly-morphism it indicates that two or more interpreta-tions of homology are applicable and that assignmentwas established analytically by character concor-dance Although lsquo-rsquo lsquorsquo and lsquo01rsquo are analyticallyidentical when applied to binary characters theyprovide information about the empirical status of thecharacter state in specific taxa Characters are citedthroughout the text as italicized numbers in parenthe-ses and are discussed in the Appendix

PHYLOGENETIC ANALYSIS

Phylogenetic analysis of the full data matrix wasperformed using the program Tree Analysis usingNew Technologies (TNT) ver 1 (Goloboff Farris ampNixon 2000) using lsquotraditionalrsquo search based on1000 replicates using TBR branch swapping Resultswere compared to those obtaining using the ratchetalgorithm (Nixon 1999) to determine any differencedue to analytical approach Nodal support for theminimal-length topology was evaluated by boot-strap (Felsenstein 1985) and Bremer support(Bremer 1994) Bootstrap analysis was conducted inTNT and based on 1000 pseudoreplicates each anal-ysed by ten random-addition replicates using TBRbranch swapping A nexus file containing the result-ing 1000 trees was imported into PAUP ver 10(Swofford 2002) to obtain bootstrap frequenciesBremer support was determined in TNT by con-straining specific nodes in the minimal-length topol-ogy and then determining the shortest tree that didnot recover the specified clade The difference inlength between the unconstrained and constrainedminimal-length trees is the Bremer support Theeffect of homoplasy on results was explored by con-ducting implied weights analysis (Goloboff 1993) inTNT Six analyses were conducted each with con-stant of concavity (k) set to a different integer valueof 1ndash6 where 1 is weighted most severely againsthomoplasious characters Each implied weightsanalysis was conducted using lsquotraditionalrsquo searchbased on 1000 replicates using TBR branch swap-ping The same procedures were used in analysing amatrix that included only extant taxa Howevercharacters rendered uninformative by removal offossil taxa (ie 49 98 99 100 113 114 125 161)were excluded prior to analysis

230 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

COMPARISONS OF ALTERNATIVE HYPOTHESES

Phylogenetic hypotheses proposed in the recent liter-ature were also evaluated including those thatattempted to resolve arachnid phylogeny completely(Fig 2) their hypothesized subclades (Fig 3) andhypotheses that proposed only specific nodes (Figs 34) The fully resolved topologies of Weygoldt amp Paulus(1979) Van der Hammen (1989) Shultz (1990)Wheeler amp Hayashi (1998) and Giribet et al (2002)were compared with the optimal topology using theTempleton test (Templeton 1983) as implementedin PAUP Internal relationships of multisampledorders were constrained to match those of the optimaltopology unless the original authors explicitlyfavoured an alternative Node-specific hypotheseswere evaluated by determining the frequency withwhich the node was recovered in bootstrap analysisThey were also evaluated by constraining parsimonyanalysis in TNT to recover the shortest tree contain-ing each specific node and then assessing the effect onrelative tree length and overall resolution takingnote of effects on otherwise stable or well-supportedclades The entire matrix was used to assess the twocases where node-specific hypotheses involved fossiltaxa (Fig 4)

RESULTS

EXTANT TAXA

Unweighted analysis of extant taxa produced two min-imal-length trees (length 383 CI 057) (Fig 5A) withconflicts limited to relationships within Pseudoscorpi-ones Multiply sampled orders were recovered asmonophyletic with high nodal support as measured bybootstrap percentage (BP gt 80) except Acari (BP 30)There were few well-supported interordinal groupsexcept Uropygi (BP 100) Pedipalpi (BP 100) Tetra-pulmonata (BP 93) and ArachnidaXiphosura (BP100) All other interordinal relationships were recov-ered with BP less than 60

Results from implied weights analysis (IWA) indi-cated a significant effect of homoplasy on resolution ofthe deepest interordinal relationships although theinterordinal clades Stomotheca Haplocnemata Acar-omorpha Tetrapulmonata Pedipalpi and Uropygiwere recovered under all values of k IWA with k = 1 (1tree best score = 5124) recovered Megoperculata(= Palpigradi + Tetrapulmonata) and Micrura(= Megoperculata + Acaromorpha) with Micrura beingthe sister group to Haplocnemata (= Pseudoscorpiones+ Solifugae) and with Stomothecata (= Opiliones +Scorpiones) being the sister group to all other arach-nids (Fig 5B) With k = 2 (1 tree best score = 3807)and k = 3 (1 tree best score = 3054) Megoperculatawas recovered and Stomotheca remained the sister

group to other arachnids but the relative positions ofHaplocnemata and Acaromorpha changed (Fig 5C)IWA using k = 4 (1 tree best score = 2559) k = 5 (1tree best score = 2205) and k = 6 (1 tree bestscore = 1938) recovered topologies consistent with theminimal-length topology derived from unweightedcharacters (Fig 5A)

EXTANT AND FOSSIL TAXA

Unweighted analysis of extant and fossil taxa pro-duced 18 minimal-length trees (length 426 CI 056)(Fig 6A) with conflicts limited to relationships withinPseudoscorpiones Scorpiones and Araneae Multiplysampled orders were again recovered as monophyleticwith high nodal support (BP gt 80) except Acari whichwas reconstructed as diphyletic with Anactinotrichidabeing the sister group to Ricinulei (BP 33) Therewere no well-supported interordinal groupings (ieBP gt 80) except Uropygi (BP 89) Pedipalpi (BP 92)and Arachnida (BP 90)

Results from IWA indicated a significant effect ofhomoplasy on resolution of the deepest interordinalrelationships in Arachnida although StomothecataHaplocnemata Acaromorpha Cryptognomae Pantet-rapulmonata Tetrapulmonata Schizotarsata Pedi-palpi and Uropygi were recovered under all values ofk IWA with k = 1 (7 trees best score = 5675) recov-ered Megoperculata and Micrura with Micrura beingthe sister group to Haplocnemata Stomothecata wasnot recovered in the strict consensus (Fig 6B) Withk = 2 (3 trees best score = 4243) IWA recovered thetopology shown in Fig 6C with Palpigradi being thesister group to all other arachnids The strict consen-sus topologies reconstructed using k = 3ndash6 were iden-tical to those obtained with k = 2 (k = 3 3 trees bestscore = 3419 k = 4 3 trees best score = 2874 k = 53 trees best score = 2483 k = 6 3 trees bestscore = 2188)

COMPARISONS OF ALTERNATIVE HYPOTHESES

The hypothesis of Weygoldt amp Paulus (1979) (Fig 2)was 398 steps (15 steps longer than the unconstrainedminimal-length tree) Van der Hammen (1989) was400 steps (17 steps longer) Shultz (1990) was 399steps (six steps longer) Wheeler amp Hayashi (1998)was 391 steps (eight steps longer) and Giribet et al(2002) was 394 steps (11 steps longer) Templetontests conducted at the 005 significance level rejectedthe hypothesis of no difference between the optimaltopology and those of Weygoldt amp Paulus (P = 00002)Van der Hammen (P = 00011) Wheeler amp Hayashi(P = 00018) and Giribet et al (P = 00343) The Shultztopology was not significantly different (P = 01444)from the optimal topology

PHYLOGENY OF THE ARACHNID ORDERS 231

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 2 Fully resolved phylogenetic hypotheses of extant euchelicerate groups proposed in the recent literature Note thesimilarity in the topologies of the parsimony-based analyses by Shultz (1990) Wheeler amp Hayashi (1998) and Giribet et al(2002) The Giribet et al topology is based on neontological data (morphology and molecules) and the original lsquoROOTrsquo maybe an artefact from use of the highly divergent pycnogonids as an outgroup

Xip

hosu

raS

corp

ione

sO

pilio

nes

Ara

neae

Am

blyp

ygi

Uro

pygi

Ric

inul

eiA

nact

inot

richi

daS

olifu

gae

Pse

udos

corp

ione

sP

alpi

gra

diA

carif

orm

es

Xip

hosu

raS

corp

ione

sA

rane

aeA

mbl

ypyg

iU

ropy

giP

alpi

gra

diS

olifu

gae

Pse

udos

corp

ione

sO

pilio

nes

Ric

inul

eiA

cari

Lipoctena

Arachnida

Apulmonata

Holotracheata

Cryptoperculata

Acaromorpha(ldquoAcarinomorphardquo)

Tetrapulmonata(rdquoMegoperculatardquo)

LabellataHaplo-cnemata

Weygoldt amp Paulus (1979)

Labellata

Tetrapulmonata(rdquoArachnidardquo)

Haplocnemata(rdquoApatellatardquo)

Van der Hammen (1989)

Xip

hosu

raP

alpi

gra

diA

rane

aeA

mbl

ypyg

iU

ropy

giR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Pedipalpi Haplocnemata

Novogenuata

Dromopoda

Shultz (1990)

Acaro-morpha

Micrura

Xip

hosu

raP

alpi

gra

diU

ropy

giA

mbl

ypyg

iA

rane

aeR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Labellata Haplocnemata

Novogenuata

Dromopoda

Wheeler amp Hayashi (1998)

Acaro-morpha

Micrura

Xip

hosu

raA

cari

Pal

pig

radi

Ric

inul

eiA

rane

aeA

mbl

ypyg

iU

ropy

giO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Tetrapulmonata

PedipalpiHaplocnemata

Novogenuata

Dromopoda

Giribet et al (2002)

ROOT

Micrura

Crypto-gnomae

Sternocoxata

Epimerata

RostrosomataMyliosomata

Neosternata

232 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 3 Consensus trees produced by parsimony analysis of neontological data constrained to produce relationships pro-posed in the recent literature The constrained (target) node is indicated by a black dot and the taxa encompassed by theconstraint are enclosed in a box Numbers below each tree represent the number of minimal-length constrained treeslength of minimal-length trees difference in the length of the unconstrained minimal-length tree and the constrained min-imal-length tree and percentage unconstrained bootstrap trees in which the target node was recovered respectively Thesetrees indicate the effect on branch length imposed by specific hypotheses and impact of constraining target nodes on non-target nodes

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Lipoctena4 385 2 lt5

Xi

Sc

Op

Ps

So

Pa

Ur

Am

Ar

Ri

An

Ac

Labellata6 397 14 lt5

Xi

Sc

Ar

Am

Ur

Op

Pa

Ps

So

Ri

An

Ac

Apulmonata2 390 7 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

So

Ri

An

Ac

Holotracheata5 390 7 lt5

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptoperculata2 388 5 lt5

Xi

Sc

Op

Pa

So

Ps

Ar

Am

Ur

Ri

An

Ac

Rostrosomata6 384 1 25

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Neosternata3 388 5 lt5

Xi

Sc

Op

Ar

Am

Ur

Ri

An

Ps

So

Pa

Ac

Coxisternata2 388 5 lt5

Epimerata7 387 4 11

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptognomae6 384 1 24

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Micrura3 385 2 7

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Megoperculata3 384 1 28

Xi

Sc

Ps

So

Op

Pa

Ar

Am

Ur

Ri

An

Ac

Novogenuata10 388 5 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

Ri

So

An

Ac

Solifugae + Acari4 388 5 lt5

Shultz (1990) Alberti amp Peretti (2002)

Van der Hammen (1989)

Weygoldt amp Paulus (1979)

Xi

Ar

Am

Ur

So

Ps

Sc

Op

Ri

Pa

An

Ac

PHYLOGENY OF THE ARACHNID ORDERS 233

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Results from analyses of node-specific hypothesesare summarized in Figures 3 4 In most cases pro-posed relationships were recovered in fewer than 5of bootstrap pseudoreplicates The notable exceptionswere Rostrosomata (25) Cryptognomae (24) andMegoperculata (28) and the strict-consensus con-strained minimal-length tree for each was only onestep longer than the unconstrained optimal tree Inmost cases constrained topologies did not add manysteps most were less than five steps longer and nonewas greater than 14 steps longer than the un-constrained minimal-length tree It is noteworthyhowever that even constrained nodes that imposerelatively few extra steps sometimes supportedimprobable relationships or eliminated clades thatwere well supported or stable in the unconstrainedanalysis Specific examples and their implications arediscussed below

DISCUSSION

RECENT ISSUES IN ARACHNID PHYLOGENY

Available morphological evidence consistently re-solves some interordinal relationships and fails to

resolve others (Fig 1) Continued progress depends onthe ability of arachnologists to discover new charac-ters and to assess the evidence critically Whilereviewing the recent arachnological literature severalaspects of phylogenetic practice emerged that seemedcounterproductive to both the perception and theactual rate of progress toward resolving arachnid phy-logeny Some of these are summarized here withspecific examples given in the remainder of theDiscussion and in the Appendix

There is a tendency to portray arachnid ordinal phy-logeny as more poorly resolved and contentious than isactually the case (Coddington et al 2004) Phyloge-netic hypotheses generated during different historicalperiods and using differing standards of evidence areoften cited as examples of current disagreement (egSelden 1993 Dunlop 1996 Selden amp Dunlop 1998Wheeler amp Hayashi 1998) In fact recent parsimony-based analyses of morphology have tended to convergeon topologies with internal structures congruent withthose found here (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) (Fig 2) Arachnidphylogeny is not fully resolved especially at its deep-est levels (Fig 1) but this does not mean that allaspects of arachnid phylogeny are controversial orpoorly supported by the available evidence

Matrices are sometimes constructed by uncriticallsquorecyclingrsquo of erroneous or problematic charactersbased on diverse secondary sources (Jenner 2001)Conclusions derived from mixtures of valid invalidspeculative and redundant characters are sometimesportrayed as the phylogenetic signal provided by mor-phology Data recycling can perpetuate errors (see 1320 39 52 136 140 144 152 169 171ndash174) legiti-mize speculations (see 13 32 63 95) or create dupli-cate or non-independent characters (see 13 30 70 77172) The lsquolateral organrsquo is a notable exampleYoshikura (1975) equated the embryonicearly post-embryonic lsquolateral organsrsquo of Amblypygi Thelyphonidaand Solifugae with the dissimilar lsquolateral organrsquo ofXiphosura but failed to note the very similar Cla-paregravede organ of Acariformes (see 173 174) This cod-ing was recycled by Wheeler amp Hayashi (1998) andthen by Giribet et al (2002) Error is probably inevi-table when assembling a large matrix from morphol-ogy including the one presented here but this can beminimized by making original observations and byconsulting primary sources

Some workers advocate weighting characters a pri-ori on the basis of structural or functional complexity(eg Kraus 1998 Dunlop amp Braddy 2001) and dismissphylogenetic conclusions derived from equal-weightsparsimony In short these workers criticize parsimonyfor emphasizing data quantity over data quality Thiscriticism ignores the intense debate in systematic biol-ogy that eventually led to widespread adoption of

Figure 4 Consensus trees produced by parsimony analy-sis of the full data matrix constrained to produce relation-ships proposed in the recent literature See legend toFigure 3 for details

Xi

Eu

Sc

Op

Pa

Ac

Ri

An

So

Ps

Tr

Ar

Ha

Am

UrEurypterida slat + Scorpiones12 435 9 lt5

Xi

Eu

Sc

Op

Pa

Ac

An

So

Ps

Ri

Tr

Ar

Ha

Am

UrTrigonotarbida+ Ricinulei28 434 8 lt5

Dunlop amp Braddy (2002)et al

Dunlop (1996)

234 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 5 Results from analysis of neontological data A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology AFor B and C relationships within terminal clades are the same as those shown in A

LimulusTachypleusProkoeneniaEukoeneniaCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAllothyrusAustralothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesLiphistiusAphonopelmaHypochilusCharinusPhrynusStenochrusProtoschizomusMastigoproctusCyphophthalmusChileogoveaCaddoLeiobunumSclerobunusGonyleptesCentruroidesHadrurusHeterometrusChthoniusFeaellaNeobisiumCheliferEremocostaGaleodes

10031

10018451

501561

733

723

944

633

10010

832

901

943

903

522

341

1007

1008

Acaromorpha

Haplocnemata

10012

Stomothecata

Opiliones

Scorpiones

Pseudo-scorpiones

Solifugae

Acariformes

Parasitiformes

Opilioacariformes

Ricinulei

Xiphosura

Palpigradi

Thelyphonida

Schizomida

Araneae

Amblypygi

10011

936853

100141005

1009

Tetrapulmonata Pedipalpi

Uropygi

692

522572

968

331

603

81

511

161

101

Acari

Dromopoda

Arachnida

A

Xiphosura

Stomothecata

Haplocnemata

Acaromorpha

Palpigradi

TetrapulmonataMicrura

Megoperculata

BXiphosura

Stomothecata

Acaromorpha

Haplocnemata

Palpigradi

TetrapulmonataMegoperculata

C

PHYLOGENY OF THE ARACHNID ORDERS 235

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 6 Results from analysis of the full data matrix A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology CFor B and C relationships within terminal clades are the same as those shown in A

WeinberginaLimuloidesEuproopsTachypleusLimulusStylonurusBaltoeurypterusChasmataspisDiploaspisOctoberaspisCyphophthamusChileogoveaCaddoLeiobunumSclerobunusGonyleptesProscorpiusPalaeoscorpiusStoermeroscorpioPrearcturusCentruroidesHadrurusHeterometrusProkoeneniaEukoeneniaTerpsicrotonPoliocheraCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAustralothyrusAllothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesEremocostaGaleodesChthoniusFeaellaNeobisiumCheliferGilboarachnePalaeocharinusLiphistiusAphonopelmaHypochilusPlesiosiroCharinusPhrynusStenochrusProtoschizomusProschizomusMastigoproctus

631

893

923

984

984

671

341

671

986

10011

423

974

Xiphosura

Eurypterida sstr

Chasmataspidida

Opiliones

Scorpiones

Palpigradi

Ricinulei

Opilioacariformes

Parasitiformes

Acariformes

Solifugae

Pseudoscorpiones

Trigonotarbida

Araneae

Haptopoda

Amblypygi

Schizomida

Thelyphonida

842

251331

361311

552

832

882

925

913

573

561291

443

976

541

954

743644

732521

571

331

833

663591

966

573

71

221

905

832

351

Uropygi

Pedipalpi

Haplocnemata

Tetrapulmonata

Anactinotrichida

Cryptognomae

Acaromorpha

Stomothecata

Pantetrapulmonata

Arachnida

Eurypterida slat

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatProscorpiusStoermeroscorpioPalaeoscorpiusPrearcturusCentruroidesHadrurusHeterometrusOpilionesHaplocnemataRicinuleiAnactinotrichidaAcariformesPalpigradiPantetrapulmonata

Megoperculata

Micrura

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatPalpigradiStomothecataHaplocnemataPantetrapulmonataAcaromorpha

A

B

C

Schizotarsata

236 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

parsimony and instead advocates a return to thespeculative and subjective approaches of the late 19thand early 20th century that once threatened thescientific legitimacy of the discipline (Bowler 1996)Furthermore it misrepresents the properties of parsi-mony-based analyses Specifically characters withfunctional significance are readily encompassed byparsimony analyses and in fact characters derivedfrom locomotor systems (eg 46ndash94) have played animportant role in developing current ideas aboutarachnid phylogeny (eg Shultz 1989 1991) In addi-tion lsquocomplex charactersrsquo can be viewed as compositesof several characters such that morphological com-plexity is effectively weighted by the number of inde-pendent lsquosubcharactersrsquo it contains For example thelsquosucking stomachrsquo once considered a synapomorphy ofLabellata (= Araneae + Amblypygi Fig 3) is codedhere as a composite of three characters (199ndash201) TheLabellata hypothesis was not corroborated in thepresent analysis (Figs 1 3ndash5) and in fact was highlydisfavoured (Fig 3) but this result cannot be dis-missed as a failure to acknowledge the complexity ofthe character

Some workers support specific (target) clades withone or more similarities without exploring the impacton overall homoplasy or relationships among non-target clades Each character offers its own phyloge-netic hypothesis which may or may not be consistentwith relationships implied by other characters It isexceedingly rare for all characters to be perfectly com-patible in the phylogenetic hypotheses they supportand criteria such as parsimony have been developed todiscover those hypotheses that minimize the conflict-ing phylogenetic signals of different characters Eventhough character conflict (homoplasy) is a virtuallyinescapable phenomenon in comparative biology it isnot uncommon for workers to discover one or morecharacters and to promote their phylogeneticimplications over alternative hypotheses even thosethat otherwise appear to be well supported Thisapproach may have value in highlighting new dataand perspectives but accomplishes this by promotingthe erroneous impression that all aspects of arachnidphylogeny are so tenuous that a single character canfalsify even well-supported hypotheses Several exam-ples of this approach have appeared in the recentarachnological literature

For example Alberti amp Peretti (2002) argued thataflagellate spermatozoa (163) are a compelling syna-pomorphy for a Solifugae + Acari clade and dismissedsome of the characters that support Solifugae + Pseu-doscorpiones as having lsquodebatable valuersquo Yet theirproposal rejects two hypotheses that are consistentlyrecovered in recent phylogenetic analyses that isHaplocnemata (= Solifugae + Pseudoscorpiones) andAcaromorpha (= Acari + Ricinulei) (Figs 1 2) Rejec-

tion of Haplocnemata would require its presumed syn-apomorphies to be reinterpreted as homoplasiesincluding features of the chelicerae (18 19 20) pre-oral chamber (13 32) legs (12 48) and respiratorysystem (126) The same reason applies to Acaromor-pha and its synapomorphies It is noteworthy that aSolifugae + Acari clade was recovered in the presentanalysis in fewer than 5 of bootstrap pseudorepli-cates and that analyses constrained to recover thisclade were five steps longer than the minimal-lengthtree and favoured a problematic clade unitingRicinulei and Pseudoscorpiones (Fig 3)

In another example Dunlop (1996) proposed a closerelationship between Trigonotarbida and Ricinuleibased on two-segmented chelicerae (18) prosomandashopisthosoma coupling mechanism (96) diplotergites(100 101) and longitudinally divided opisthosomaltergites (115) However phylogenetic analyses con-strained to recover this relationship required eightadditional steps eliminated support for Acari and nec-essarily rejected Acaromorpha and forced its synapo-morphies to be reinterpreted as homoplasies Alberti(2005) proposed an interesting hypothesis for the evo-lution of male gonads in tetrapulmonates (153) butchose to accept Labellata (= Araneae + Amblypygi) indeveloping his argument over the much more well-supported Pedipalpi (= Amblypygi + Uropygi) (Shultz1999) a phylogenetic reconfiguration that was recov-ered here in fewer than 5 of bootstrap pseudorepli-cates increased tree length by a minimum of 14 stepsand resulted in the collapse of Stomothecata and Hap-locnemata (Fig 3) Many other examples can be cited

Promoting or defending a specific phylogenetichypothesis via lists of compatible synapomorphies is acommon but problematic approach By restrictingattention to the states of specific characters at one ortwo target nodes one can easily overlook the unin-tended impact of the hypothesis on phylogenetic sig-nal elsewhere and its effect on non-target clades Anode supported by a long list of synapomorphies mayseem convincing taken in isolation but may becomeless acceptable when its full phylogenetic implicationsare explored

SUMMARY OF INTERORDINAL ARACHNID CLADES RECOVERED IN THIS ANALYSIS

Arachnida Lamarck 1801Analyses consistently recovered Arachnida as a mono-phyletic group with high bootstrap support (Figs 1 45) Possible synapomorphies include the loss of thecarapacal pleural doublure (9) cardiac lobe (10) pedalgnathobases (52) and moveable endites (53) and thegain of aerial respiration (120) and an anteriorly oranteroventrally directed mouth (185) Some tradi-tional synapomorphies such as slit sensilla (142) and

PHYLOGENY OF THE ARACHNID ORDERS 237

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

fluid feeding (184) may have appeared later in arach-nid evolution but this can only be decided once theinternal phylogeny of Arachnida has been determined

Some workers regard many proposed arachnid syn-apomorphies as adaptations to terrestrial life andthereby link the hypothesis of arachnid monophyly tothe hypothesis of a single ancestral aquatic-to-terres-trial transition and arachnid polyphyly to multipletransitions (eg Selden amp Jeram 1989 Dunlop 1997Dunlop amp Webster 1999 Dunlop amp Braddy 2001)Given this line of reasoning the existence of appar-ently aquatic scorpions in the fossil record (Kjellesvig-Waering 1986 Jeram 1998) and the inference thatterrestrialization occurred late in scorpion evolutionthese workers conclude that character states support-ing arachnid monophyly are actually convergencesand do not necessarily support arachnid monophyly

However this approach to assessing phylogenetichypotheses is founded on overly simplistic assump-tions such as the ability of the investigator to discrim-inate unerringly between characters that existexclusively in aquatic organisms (including fossils)from those that occur exclusively in terrestrial organ-isms There also appears to be an assumption thathomoplasy can be generated through parallelism (iemultiple aquatic-to-terrestrial events) but not throughterrestrial-to-aquatic reversals Furthermore anassumed dichotomy between exclusively aquatic andterrestrial life histories in ancestral arachnids is sim-plistic as illustrated by the amphibious life cycles ofbasal vertebrates and pterygote hexapods In factthese examples demonstrate that there is no neces-sary inconsistency in basing a hypothesis of arachnidmonophyly on the derived terrestrial features of anamphibious ancestor whose descendants thencompleted terrestrialization once or several timesindependently or even returned to a fully aquaticexistence Workers who link the frequency and direc-tion of aquaticndashterrestrial transitions to the assess-ment of arachnid phylogeny do so by endowingthemselves with substantially greater insight thanseems prudent by ignoring the huge gaps in ourunderstanding of early arachnid evolution and bydenying to arachnids the evolutionary complexityknown to exist in other groups

Several palaeontologists have been particularlyactive during the past decade in proposing new char-acters with the stated goal of removing Scorpionesfrom Arachnida and erecting a Scorpiones + Eurypter-ida clade (Braddy amp Dunlop 1997 Dunlop amp Braddy1997 Dunlop 1998 Braddy et al 1999 Dunlop ampWebster 1999) Dunlop amp Braddy (2001) recentlysummarized this evidence and conducted aparsimony-based analysis of Xiphosura EurypteridaScorpiones Opiliones and Tetrapulmonata (but notHaplocnemata Acaromorpha or Palpigradi) using 33

morphological characters Their analysis produced atopology congruent with those generated here (Figs 15 6) including recovery of Stomothecata (= Scorpiones+ Opiliones) However they rejected this result as aproduct of lsquoempirical cladisticsrsquo because it gives thesame weight to prosomal characters that supportarachnid monophyly and to selected opisthosomalcharacters that support their favoured Eurypterida +Scorpiones clade namely a five-segmented metasoma(116) suppression of opisthosomal tergite 1 (95) lossof lamellate respiratory organs on the postgenitalsomite (122) Kiemenplatten (125) loss of respiratorylamellae on the genital segment (121) and a lsquonon-stainingrsquo exocuticle (but see Grainge amp Pearson 1966for evidence of this in Opiliones see Appendix for com-ments on the other characters)

A Eurypterida + Scorpiones clade was not favouredin the present analysis (Fig 6) the strict consensus ofminimal-length trees constrained to recover this clade(Fig 4) is nine steps longer than the tree recovered byanalysis without this constraint It is also noteworthythat Opiliones was consistently reconstructed as thesister group to Eurypterida + Scorpiones in the mini-mal-length constrained trees a provocative result thatwas probably unintended and unanticipated by sup-porters of the Eurypterida + Scorpiones Given thatDunlop amp Braddy (2001) reject equal weights parsi-mony as an arbiter of phylogenetic hypotheses theywould presumably dismiss these results as irrelevantto their argument just as they dismissed their ownparsimony-based results However if the Eurypterida+ Scorpiones hypothesis is to be credible it must beopen to evaluation and potential falsification usingobjective criteria and the subjective or intuitive a pri-ori weighting of characters advocated by Dunlop ampBraddy clearly does not qualify At present it isunclear how one would objectively evaluate Dunlop ampBraddyrsquos proposal with criteria compatible with thoseused in its original formulation For now the Euryp-terida + Scorpiones concept advocated by Dunlop ampBraddy may persist outside the mainstream of mod-ern systematic practice but it is increasingly problem-atic within it

Stomothecata nomnovOpiliones and Scorpiones were consistently recoveredas a monophyletic group The proposed name acknowl-edges a unique preoral chamber the stomothecaformed by coxapophyses of the palp and leg 1 (50)often with an auxiliary role played by the coxapophy-sis of leg 2 (51) In addition the epistome appears tohave been modified for adduction of the palpal coxaeThe lateral walls of the epistome are fused to themedial surfaces of the palpal coxae and the epistomallumen is spanned by a transverse muscle (188) which

238 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

apparently adducts the palpal coxae thereby constrict-ing the stomothecal chamber Scorpions and opilionsare also unique in having a pair of large epistomalarms projecting rearward into the prosoma andattaching to the endosternite (189) The epistomalarms provide attachment sites for pharyngeal dilatormuscles (196) and extrinsic muscles of anterior proso-mal appendages (eg 37) The chelicera is equippedwith a muscle that arises on the carapace and insertson the ventral margin of the second cheliceral article(23) There is an anteriorly placed genital opening(155)

Several notable similarities were found whilereviewing the literature but information from otherarachnid groups was considered too incomplete toallow them to be included in the matrix For exampleboth orders have apparent haemocytopoeitic organsassociated with major nerves of the anterior opistho-soma These are termed supraneural organs in scorpi-ons (Farley 1999) and perineural organs in opilions(Kaumlstner 1935) Haemocytes develop in the cardiacwall in spiders (Seitz 1972) and perhaps amblypygids(Weygoldt 2000) but haemocytopoeitic organs areunknown in most other arachnid groups Germ cellsdifferentiate very early during embryogenesis in bothscorpions and opilions relative to spiders (Moritz1957 Anderson 1973) Additional research is requiredto determine the phylogenetic utility of thesecharacters

Haplocnemata Boumlrner 1904Several workers recognize a close phylogenetic rela-tionship between Pseudoscorpiones and Solifugae(Weygoldt amp Paulus 1979 Van der Hammen 1989Shultz 1990 Wheeler amp Hayashi 1998 Giribetet al 2002) and this was consistently recovered inthe present analysis Haplocnemata is supported byseveral synapomorphies including (i) a feeding com-plex formed by two-segmented chelicerae (18) with aventrolateral intrinsic articulation (19) and perhapsa dorsolateral articulation with the carapace (20)and (ii) a rostrosoma (32) a preoral apparatus formedby an anteriorly projecting epistome affixed dorsallyto enlarged palpal coxae and bordered by lateral pal-pal projections The epistome forms the dorsal compo-nent of a lsquobeakrsquo with the ventral part formed by amidventral sternapophysis (13) fused between thepalpal coxae The locomotor apparatus consists ofcoxae that meet along the ventral midline (12) elon-gate femur-like patellae (48) and an apotele in theadult modified as an eversible empodium or pulvillus(92) The respiratory system is formed entirely (Pseu-doscorpiones) or in part (Solifugae) by paired trachealstigmata opening on opisthosomal somites 3 and 4(126)

Acaromorpha Dubinin 1957A clade comprising Ricinulei and Acari is consistentlyrecovered in the present analysis It is united here bytwo unique and seven homoplasious synapomorphiesThese include a gnathosoma (31) defined in part bymedial fusion of the palpal coxae (30) although pres-ence of a gnathosoma in Ricinulei is debatable as dis-cussed in the Appendix Acaromorphs also have aunique post-embryonic development consisting of ahexapodal larva and up to three octopodal nymphalinstars (176) The pedal patellandashtibia joints are formedby a bicondylar dorsal hinge rather than a single mid-dorsal condyle (69) all postcheliceral segmental gan-glia are unified in the subesophageal ganglion (130)and a postcerebral pharynx is absent (192) The groupis also tentatively united by presence of differentiatedpedal basi- and telofemora (63 64) and the absence ofa ventral (sternal) pharyngeal dilator muscle (199)but these may be symplesiomorphies erroneouslyreconstructed as a synapomorphies

The internal phylogenetic structure of Acaromorphais controversial with many recent workers favouringa monophyletic Acari (Weygoldt amp Paulus 1979Shultz 1990 argued most thoroughly by Lindquist1984) and others advocating a diphyletic Acari withAcariformes being the sister group to Cryptognomae(= Ricinulei + Anactinotrichida) (especially Van derHammen 1979 1989) (Fig 3) Acari was recovered asmonophyletic when fossils were excluded (Fig 2) butnot when they were included (Fig 3) A lsquomite-centredrsquosurvey of arachnid characters may be needed if mor-phology is to offer a compelling solution to the internalphylogeny of Acari and its placement within Arach-nida These issues are not resolved by the presentanalysis

Pantetrapulmonata nom novPantetrapulmonata includes the extinct orders Trigo-notarbida Haptopoda and the extant orders AraneaeAmblypygi Schizomida and Thelyphonida The cladeis united by cheliceral structure (18 19) a megoper-culum (106) booklungs on the genital and first post-genital somites (121 122) and enlargement of theepipharyngeal sclerite (192) It is important to notehowever that most of these characters were coded asuncertain in Plesiosiro (Haptopoda) Aside from theplacement of Haptopoda the monophyly of Trigono-tarbida and the extant orders was anticipated byShear et al (1987) and is generally regarded as amonophyletic group

Tetrapulmonata Shultz 1990Tetrapulmonata was originally proposed on the basisof neontological analyses and encompassed Araneae

PHYLOGENY OF THE ARACHNID ORDERS 239

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Amblypygi Schizomida and Thelyphonida (Shultz1990) and was also recovered here in analysis of neon-tological data (Fig 5) However results from analysisof all taxa required that the Tetrapulmonata conceptbe expanded to include the fossil order HaptopodaFeatures uniting this group are problematic howeveras many states in Haptopoda were coded as uncertain

Schizotarsata nomnovHaptopoda and Pedipalpi (= Amblypygi + Schizomida+ Thelyphonida) are united here in a group named forpossession of divided pedal telotarsi (84) Synapomor-phies include a pointed anterior carapacal margin (3)and elongation of leg 1 (46) As already noted theplacement of Haptopoda should probably be regardedas tentative because the state of many characters inthis group was coded as unknown The placement ofHaptopoda as the sister group to Pedipalpi was antic-ipated by Dunlop (1999 2002c)

Pedipalpi Boumlrner 1904Pedipalpi encompasses Amblypygi and Uropygi(= Thelyphonida + Schizomida) Although widely rec-ognized in the past Pedipalpi was set aside in the mid-20th century in favour of Labellata (= Amblypygi +Araneae) based largely on the presumed synapomor-phies of a narrow prosomandashopisthosoma juncture (97)and sucking stomach (199ndash201) The situation beganto reverse when Shear et al (1987) again highlightedsimilarities of Amblypygi and Uropygi such as rapto-rial palps (35) and antenniform leg 1 (46) Shultz(1989 1990 1999) subsequently described numerousderived similarities including asymmetrical flexormuscles at the pedal femurndashpatella joint (70) threetelotarsomeres on legs 2ndash4 (85) and modification ofthe palpal coxae to provide support for extrinsic pha-ryngeal muscles (197) Pedipalpi has emerged as oneof the most well-supported interordinal relationshipsin Arachnida

Uropygi Thorell 1882Uropygi has long been accepted as a monophyleticunion of Thelyphonida and Schizomida Synapomor-phies include a unique mating behaviour (159) fusedpalpal coxae (30) 2-1-1-1 arrangement of tibial tricho-bothria (144) posterior defensive glands (102) elon-gated patella of leg 1 (68) and many others

ACKNOWLEDGEMENTS

I thank Nikolaj Scharff and three anonymous review-ers for useful comments This work was supported bythe Maryland Agricultural Experiment Station the

General Research Board of the University of Mary-land and National Science Foundation Grant 0416628to Susan Masta

REFERENCES

Abd el-Wahab A 1952 Notes on the morphology of the scor-pion Buthus quinquestriatus (HE) Publications de lrsquoInsti-tut Fouad 1er du Desert 3 1ndash129

Alberti G 1979 Licht- und elektronenmikroskopische Unter-suchungen an Coxaldruumlsen von Walzenspinnen (ArachnidaSolifugae) Zoologischer Anzeiger 203 48ndash64

Alberti G 1995 Comparative spermatology of Cheliceratareview and perspective Memoires du Museum NationaldrsquoHistoire Naturelle 166 203ndash230

Alberti G 2005 Double spermatogenesis in ChelicerataJournal of Morphology 266 281ndash297

Alberti G Coons LB 1999 Acari mites In Harrison FWFoelix RF eds Microscopic anatomy of invertebratesVol 8C Chelicerate Arthropoda New York Wiley Liss515ndash1215

Alberti G Peretti AV 2002 Fine structure of male genitalsystem and sperm in Solifugae does not support a sister-group relationship with Pseudoscorpiones (Arachnida)Journal of Arachnology 30 268ndash274

Alexander AJ 1967 Problems of limb extension in the scor-pion Opisthophthalmus latimanus Koch Transactions of theRoyal Society of South Africa 37 165ndash181

Anderson DT 1973 Embryology and phylogeny in annelidsand arthropods Oxford Pergamon Press

Anderson LI Selden PA 1997 Opisthosomal fusion andphylogeny of Palaeozoic Xiphosura Lethaia 30 19ndash31

Beall BS Labandeira CC 1990 Macroevolutionary patternsof the Chelicerata and Tracheata In Culver SJ ed Arthro-pod paleobiology Short Courses in Paleontology No 3 Lon-don The Paleontological Society 257ndash284

Bernard HM 1896 The comparative morphology of the Gale-odidae Transactions of the Linnean Society of London Sec-ond Series 6 305ndash417

Boumlrner C 1902a Arachnologische Studien (II und III) Zool-ogischer Anzeiger 25 433ndash466

Boumlrner C 1902b Arachnologische Studien IV Die Genitalor-gane der Pedipalpen Zoologischer Anzeiger 26 81ndash92

Boumlrner C 1904 Beitraumlge zur Morphologie der Arthropoden IEin Beitrag zur Kenntnis der Pedipalpen Zoologica (Stut-tgart) 42 1ndash174

Bowler PJ 1996 Lifersquos splendid drama Chicago Universityof Chicago Press

Braddy SJ Alridge RJ Gabbott SE Theron JN 1999Lamellate book-gills in a late Ordovician eurypterid from theSoom Shale South Africa support for a eurypterid-scorpionclade Lethaia 32 72ndash74

Braddy SJ Dunlop JA 1997 The functional morphology ofmating in the Silurian eurypterid Baltoeurypterus tet-ragonophthalmus (Fischer 1839) Zoological Journal of theLinnean Society 121 435ndash461

Brauer A 1895 Beitrage zur Kenntnis der Entwicklungsge-schichte des Skorpiones II Zeitschrift fuumlr WissenschaftlicheZoologie 59 351ndash433

240 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bremer K 1994 Branch support and tree stability Cladistics10 295ndash304

Bristowe WS 1932 The liphistiid spiders Proceedings of theZoological Society of London Part 4 1016ndash1057

Butt AG Taylor HH 1991 The function of spider coxalorgans effects of feeding and salt-loading on Porrhotheleantipodiana (Mygalomorpha Dipluridae) Journal of Exper-imental Biology 158 439ndash461

Buxton BH 1913 Coxal glands of arachnids ZoologischeJahrbucher (Suppl XIV) 16ndash50

Buxton BH 1917 Notes on the anatomy of arachnids Jour-nal of Morphology 29 1ndash31

Camin JH Clark GM Bourdeau FG 1956 The palpallsquotined setarsquo in the Mesostigmata a homologue of the palpalclaw in the Onychopalpida (Acarina) In Becker EC ed Pro-ceedings of the Tenth International Congress of EntomologyOttawa Mortimer 903ndash908

Chamberlin JC 1931 The arachnid order ChelonethidaStanford University Stanford University Press

Chapman RF 1998 The insects structure and function 4thedn Cambridge Cambridge University Press

Clarke KU 1979 Visceral anatomy and arthropod phylogenyIn Gupta AP ed Arthropod phylogeny New York Van Nos-trand Reinhold 467ndash549

Clarke J 1984 On the relationship between structure andfunction in the leg joints of Heteropoda venatoria (L) (Ara-neae Eusparassidae) Bulletin of the British ArachnologicalSociety 6 181ndash192

Clarke J 1986 The comparative functional morphology of theleg joints and muscles of five spiders Bulletin of the BritishArachnological Society 7 37ndash47

Clarke JM Ruedemann R 1912 The Eurypterida of NewYork New York State Museum Memoir 14 (1ndash2) 1ndash628

Coddington JA 2005 Phylogeny and classification of spi-ders In Ubick D Paquin P Cushing PE Roth V eds Spi-ders of North America an identification manual AmericanArachnological Society 18ndash24

Coddington JA Giribet G Harvey MS Prendini LWalter DE 2004 Arachnida In Cracraft J Donoghue MJeds Assembling the tree of life Oxford Oxford UniversityPress 269ndash318

Coddington JA Levi HW 1991 Systematics and evolutionof spiders (Araneae) Annual Review of Ecology and System-atics 22 565ndash592

Coineau Y Van der Hammen L 1979 The postembryonicdevelopment of Opilioacarida with notes on new taxa and ona general model for the evolution In Piffle E ed Proceed-ings of the 4th International Congress of Acarology Budap-est Akadeacutemiai Kiadoacute 437ndash441

Cokendolpher JC Reddell JR 1992 Revision of the Pro-toschizomidae (Arachnida Schizomida) with notes on thephylogeny of the order Texas Memorial Museum Speleolog-ical Monographs 3 31ndash74

Condeacute B 1991a Le dimorphisme sexuel des Palpigrades Bul-letin de la Socieacuteteacute Neuchacircteloise des Sciences Naturelles 11667ndash73

Condeacute B 1991b Lrsquoaire genitale male des Palpigrades Bulle-tin de la Socieacuteteacute Europeenne drsquoArachnologie 1990 64ndash69

Dearden PK Donly C Grbicccc M 2002 Expression of pair-rule gene homologues in a chelicerate early patterning ofthe two-spotted spider mite Tetranychus urticae Develop-ment 129 5461ndash5472

Delle Cave L 1975 The ldquocleaning brushrdquo in Damon diademaSimon 1876 and Phrynichus jayakari Pocock 1894 (Ambly-pygi Arachnida) Monitore Zoologico Italiano (NS) 9 175ndash184

Dunlop JA 1994 Filtration mechanisms in the mouthparts oftetrapulmonate arachnids (Trigonotarbida Araneae Ambly-pygi Uropygi Schizomida) Bulletin of the British Arachno-logical Society 9 267ndash273

Dunlop JA 1996 Evidence for a sister group relationshipbetween Ricinulei and Trigonotarbida Bulletin of the BritishArachnological Society 10 193ndash204

Dunlop JA 1997 Palaeozoic arachnids and their significancefor arachnid phylogeny In Zabka M ed Proceedings of the16th European colloquium of arachnology Siedlce PolandWyzsza Szkola Rolnicko-Pedagogiczna 65ndash82

Dunlop JA 1998 The origins of tetrapulmonate book lungsand their significance for chelicerate phylogeny In SeldenPA ed Proceedings of the 17th European Colloquium ofArachnology Edinburgh 1997 Burnham Beeches BritishArachnological Society 9ndash16

Dunlop JA 1999 A redescription of the Carboniferous arach-nid Plesiosiro madeleyi Pocock 1911 (Arachnida Hapto-poda) Transactions of the Royal Society of Edinburgh EarthSciences 90 29ndash47

Dunlop JA 2000 The epistomo-labral plate and lateral lips insolifuges pseudoscorpions and mites Ekoloacutegia (Bratislava)19 67ndash78

Dunlop JA 2002a Arthropods from the Lower Devonian Sev-ernaya Zemlya Formation of October Revolution Island(Russia) Geodiversitas 24 349ndash379

Dunlop JA 2002b Character states and evolution of the che-licerate claws In Toft S Scharff N eds European Arach-nology 2000 Aarhus Aarhus University Press 345ndash354

Dunlop JA 2002c Phylogeny of Chelicerata In LlorenteBousquets J Morrone JJ eds Biodiversidad Taxonomiacutea yBiogeografiacutea de Arthroacutepodos de Meacutexico Hacia Una Siacutentesisde Su Conocimiento Vol III Prensas de Ciencias Univer-sidad Nacional Autonoma de Meacutexico 117ndash141

Dunlop JA Anderson LI Braddy SJ 2004 A redescriptionof Chasmataspis laurencii Caster amp Brooks 1956 (Chelicer-ata Chasmataspidida) from the Middle Ordovician of Ten-nessee USA with remarks on chasmataspid phylogenyTransactions of the Royal Society of Edinburgh Earth Sci-ences 94 207ndash225

Dunlop JA Braddy SJ 1997 Slit-like structures on theprosomal appendages of the eurypterid BaltoeurypterusNeues Jahrbuch fuumlr Geologie und Palaumlontologie-Monatshefte1997 31ndash38

Dunlop JA Braddy SJ 2001 Scorpions and their sister-group relationships In Fet V Selden PA eds Scorpions2001 in Memoriam Gary A Polis London British Arachno-logical Society 1ndash24

Dunlop JA Horrocks CA 19951996 A new Upper Carbon-iferous whip scorpion (Arachnida Uropygi Thelyphonida)

PHYLOGENY OF THE ARACHNID ORDERS 241

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

with a revision of the British Carboniferous Uropygi Zoolo-gischer Anzeiger 234 293ndash306

Dunlop JA Poschmann M Anderson L 2001 On theEmsian (early Devonian) arthropods of the Rhenish SlateMountains 3 The chasmataspidid Diploaspis Palaumlon-tolgische Zeitschrift 75 253ndash269

Dunlop JA Webster M 1999 Fossil evidence terrestrializa-tion and arachnid phylogeny Journal of Arachnology 27 86ndash93

Edgecombe GD Wilson GDF Colgan DJ Gray MR Cas-sis G 2000 Arthropod cladistics combined analysis of his-tone H3 and U2 snRNA sequences and morphologyCladistics 16 155ndash203

Evans GO 1992 Principles of acarology Wallingford UKCABI

Fahrenbach WH 1999 Merostomata In Harrison FW Foe-lix RF eds Microscopic anatomy of invertebrates Vol 8AChelicerate arthropods New York Wiley-Liss 21ndash115

Farley RD 1999 Scorpiones In Harrison FW Foelix RF edsMicroscopic anatomy of invertebrates Vol 8A Cheliceratearthropods New York Wiley-Liss 117ndash222

Farley RD 2005 Developmental changes in the embryo pro-nymph and first molt of the scorpion Centruroides vittatus(Scorpiones Buthidae) Journal of Morphology 265 1ndash27

Fayers SR Dunlop JA Trewin NH 2004 A new earlyDevonian trigonotarbid arachnid from the Windyfield ChertRhynie Scotland Journal of Systematic Palaeontology 2269ndash284

Felsenstein J 1985 Confidence limits on phylogenies anapproach using the bootstrap Evolution 39 783ndash791

Fet V Sissom WD Lowe G Braunwalder ME 2000 Cat-alog of the scorpions of the world (1758ndash1998) New YorkNew York Entomological Society

Firstman B 1954 The central nervous system musculatureand segmentation of the cephalothorax of a tarantula (Eu-rypelma californicum Ausserer) Microentomology 87 1ndash40

Firstman B 1973 The relationship of the chelicerate arterialsystem to the evolution of the endosternite Journal ofArachnology 1 1ndash54

Foelix RF 1985 Mechano- and chemoreceptive sensilla InBarth FG ed Neurobiology of arachnids Berlin Spinger-Verlag 118ndash137

Foelix RF 1996 Biology of spiders 2nd edn Oxford OxfordUniversity Press

Giribet G Edgecombe GD Wheeler WC Babbit C 2002Phylogeny and systematic position of Opiliones a combinedanalysis of chelicerate relationships using morphologicaland molecular data Cladistics 18 5ndash70

Goloboff P 1993 Estimating character weights during treesearch Cladistics 9 83ndash91

Goloboff PA Farris JS Nixon K 2000 TNT Tree AnalysisUsing New Technologies Version 1 Program and documen-tation httpwwwzmucdkpublicphylogenytnt

Grainge CA Pearson RG 1966 Cuticular structure in thePhalangida Nature 211 866

Grandjean F 1938 Observations sur les Bdelles (Acariens)Annales de la Socieacuteteacute Entomologique de France Paris 107 1ndash24

Hansen HJ Soslashrensen W 1904 On two orders of ArachnidaCambridge Cambridge University Press

Hansen HJ Soslashrensen W 1905 The Tartarides a tribe of theorder Pedipalpi Arkiv fuumlr Zoologi 2 1ndash78

Hara MR Gnaspini P 2003 Comparative study of the defen-sive behavior and morphology of the gland opening areaamong harvestmen (Arachnida Opiliones Gonyleptidae)under a phylogenetic perspective Arthropod Structure andDevelopment 32 257ndash275

Harvey MS 1992 The phylogeny and classification of thePseudoscorpionida (Chelicerata Arachnida) InvertebrateTaxonomy 6 1373ndash1435

Harvey MS 2002 The neglected cousins what do we knowabout the smaller arachnid orders Journal of Arachnology30 357ndash372

Haupt J Song D 1996 Revision of East Asian whip scorpions(Arachnida Uropygi Theyphonida) I China and JapanArthropoda Selecta 5 43ndash52

Hjelle JT 1990 Anatomy and morphology In Polis GA edThe biology of scorpions Stanford University Stanford Uni-versity Press 9ndash63

Janczyk FSW 1956 Anatomie von Siro duricorius Joseph imVergleich mit anderen Opilioniden Sitzungsberichte Oster-reichischen Akademie der Wissenschaften Mathematisch-Natruwissenschaftliche Klasse 165 474ndash522

Jenner RA 2001 Bilaterian phylogeny and uncritical recy-cling of morphological data sets Systematic Biology 50 730ndash742

Jeram AJ 1998 Phylogeny classification and evolution of Sil-urian and Devonian scorpions In Selden PA ed Proceed-ings of the 17th European Colloquium of ArachnologyEdinburgh 1997 Burnham Beeches UK British Arachno-logical Society 15ndash31

Karaman IM 2005 Evidence of spermatophores in Cyphoph-thalmi (Arachnida Opiliones) Revue Suisse de Zoologie 1123ndash11

Kaumlstner A 1935 Die Funktion der sogenannten sympathis-chen Ganglien und die Exkretion bei den PhalangiidenZoologischer Anzeiger 109 273ndash288

Kierans JE Robbins RG 1999 A world checklist of generasubgenera and species of ticks (Acari Ixodida) publishedfrom 1973 to 1997 Journal of Vector Ecology 24 115ndash129

Kjellesvig-Waering EN 1986 A restudy of the fossil Scorpi-onida of the world Palaeontographica Americana 55 1ndash287

Klompen JSH 2000 Prelarva and larva of Opilioacarus (Neo-carus) texanus (Chamberlin and Mulaik) (Acari Opilioacar-ida) with notes on the patterns of setation and lyrifissuresJournal of Natural History 34 1977ndash1992

Kraus O 1998 Elucidating the historical process of phylog-eny phylogenetic systematics versus cladistic techniquesIn Selden PA ed Proceedings of the 17th European Collo-quium of Arachnology Edinburgh 1997 Burnham BeechesUK British Arachnological Society 1ndash7

Kraus O Kraus M 1993 Divergent transformation of cheli-cerae and original arrangement of eyes in spiders (Arach-nida Araneae) Memoirs of the Queensland Museum 33579ndash584

Lankester ER Benham WBS Beck EJ 1885 On the mus-

242 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

cular and endoskeletal systems of Limulus and Scorpio withsome notes on the anatomy and generic characters of scor-pions Transactions of the Zoological Society of London 11311ndash384

Leclerc P 1989 Considerations paleobiogeographiques a pro-pos la decouverte en Thailande drsquoopilioacariens nouveaux(Acari ndash Notostigmata) Compte Rendu des Seances de laSociete de Biogeographie 65 162ndash174

Legg G 1976 The external morphology of a new species ofricinuleid (Arachnida) from Sierra Leone Zoological Journalof the Linnean Society 59 1ndash58

Lindquist EE 1984 Current theories on the evolution ofmajor groups of Acari and on their relationships with othergroups of Arachnida with consequent implications for theirclassificaton In Griffiths DA Bowman CE eds AcarologyVI Vol 1 New York John Wiley amp Sons 28ndash62

Manning PL Dunlop JA 1995 The respiratory organs ofeurypterids Palaeontology 38 287ndash297

Manton SM 1958 Hydrostatic pressure and leg exten-sion in arthropods with special reference to arachnidsAnnals and Magazine of Natural History Series 13 1161ndash182

Manton SM 1964 Mandibular mechanisms and the evolutionof arthropods Philosophical Transactions of the Royal Soci-ety of London B 7 1ndash183

Maraun M Heethoff M Schneider K Scheu S WeigmannG Cianciolo J Thomas RH Norton RA 2004 Molecularphylogeny of oribatid mites (Oribatida Acari) evidence formultiple radiations of parthenogenetic lineages Experimen-tal and Applied Acarology 33 183ndash201

Marples BJ 1967 The spinnerets and epiandrous glands ofspiders Journal of the Linnean Society (Zoology) 46 209ndash222

Marples BJ 1983 Observations on the structure of the fore-gut of spiders Bulletin of the British Arachnological Society8 46ndash52

Millot J 1933 Notes compleacutementaires sur lrsquoanatomie desliphistiides et des hypchilides a propos drsquoun travail reacutecent deA Petrunkevitch Bulletin de la Socieacuteteacute Zoologique de France68 217ndash235

Millot J 1942 Sur lrsquoanatomie et lrsquohistophysiologie deKoenenia mirabilis Grassi (Arachnida Palpigradi) RevueFranccedilaise drsquoEntomologie 9 33ndash51

Millot J 1943 Notes complementaires sur lrsquoanatomielrsquohistologie et la repartition geographique en France deKoenenia mirabilis Grassi Revue Franccedilaise drsquoEntomologie 9127ndash135

Millot J 1945 Lrsquoanatomie interne des Ricinulei (Arachnides)Annales des Sciences Naturelles (Zoologie) Series 11 7 1ndash29

Millot J 1949a Morphologie generale et anatomie interne InGrasseacute P-P ed Traite de zoologie VI Paris Masson et Cie263ndash319

Millot J 1949b Ordre des Amblypyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 563ndash588

Millot J 1949c Ordre des Araneacuteides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 589ndash743

Millot J 1949d Ordre des Palpigrades In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 520ndash532

Millot J 1949e Ordre des Uropyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 533ndash562

Millot J 1949f Ordre des Ricinuleides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 744ndash760

Millot J Vachon M 1949 Ordre Des Solifuges In Grasseacute P-P ed Traite de zoologie VI Paris Masson et Cie 482ndash519

Moore RA Briggs DEG Bartels C 2005 A new specimen ofWeinbergina opitzi (Chelicerata Xiphosura) from the LowerDevonian Huumlnsruck Slate Germany PalaumlontologischeZeitschrift 79 399ndash408

Moritz M 1957 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) unter besonderer Beruumlcksichtigungder ausseren Morphologie der Bildung des Mitteldarmesund der Genitalanlage Zoologische Jahrbuumlcher Abteilungfuumlr Anatomie und Ontogenie der Tiere 76 331ndash370

Moritz M 1959 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) II Die Anlage und Entwicklung derCoxaldruumlse bei Phalangium opilio L Zoologische Jahr-buumlcher Abteilung fuumlr Anatomie und Ontogenie der Tiere 77229ndash240

Nixon K 1999 The parsimony ratchet a new method forrapid parsimony analysis Cladistics 15 407ndash414

Norton RA 1998 Morphological evidence for the evolutionaryorigin of Astigmata (Acari Acariformes) Experimental andApplied Acarology 22 559ndash594

Palmgren P 1978 On the muscular anatomy of spiders ActaZoologica Fennica 155 1ndash41

Patten W 1912 The evolution of the vertebrates and their kinPhiladelphia P Blakistonrsquos Son

Patten W Hazen AP 1900 The development of the coxalgland branchial cartilages and genital ducts of Limuluspolyphemus Journal of Morphology 16 459ndash490

Paulus HF 1979 Eye structure and the monophyly of theArthropoda In Gupta AP ed Arthropod phylogeny NewYork Van Nostrand Reinhold 299ndash383

Petrunkevitch A 1949 A study of Palaeozoic ArachnidaTransactions of the Connecticut Academy of Arts and Sci-ences 37 69ndash315

Pittard K Mitchell RW 1972 Comparative morphology ofthe life stages of Cryptocellus pelaezi (Arachnida Ricinulei)Graduate Studies Texas Tech University 1 1ndash77

Platnick NI 2005 The world spider catalog Version55 American Museum of Natural History httpresearchamnhorgentomologyspiderscatalogindexhtm

Platnick NI Gertsch WJ 1976 The suborders of spiders acladistic analysis (Arachnida Araneae) American MuseumNovitates 2607 1ndash15

Pocock RI 1893 On some points in the morphology of theArachnida (ss) with notes on the classification of the groupAnnals and Magazine of Natural History Series 6 11 1ndash19

Pocock RI 1902 Some points in the morphology and classi-fication of the Opiliones Annals and Magazine of NaturalHistory Series 7 10 504ndash516

Popadicccc A Panganiban G Rusch D Shear WA KaufmanTC 1998 Molecular evidence for the gnathobasic derivationof arthropod mandibles and for the appendicular origin ofthe labrum and other structures Development Genes andEvolution 208 142ndash150

PHYLOGENY OF THE ARACHNID ORDERS 243

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Raw F 1957 Origin of chelicerates Journal of Paleontology31 139ndash192

Reddell JR Cokendolpher JC 1995 Catalogue bibliogra-phy and generic revision of the order Schizomida (Arach-nida) Texas Memorial Museum Speleological Monographs4 1ndash170

Reissland A Goumlrner P 1985 Trichobothria In Barth FGed Neurobiology of arachnids Berlin Springer-Verlag 138ndash161

Roewer C-F 1934 Solifugae Palpigradi Bronns Klassen undOrdnungen ses Tierreich 5 IV 4 1ndash713

Rowland JM Cooke JAL 1973 Systematics of the arachnidorder Uropygida (=Thelyphonida) Journal of Arachnology 155ndash71

Rowland JM Sissom WD 1980 Report on a fossil palpi-grade from the Tertiary of Arizona and a review of themorphology and systematics of the order (ArachnidaPalpigradida) Journal of Arachnology 8 69ndash86

Rucker A 1901 The Texan Koenenia American Naturalist35 615ndash630

Ruhland M Rathmayer W 1978 Die Beinmuskulatur undihre Innervation bei der Vogelspinne Dugesiella henzti (Ch)(Araneae Aviculariidae) Zoomorphologie 89 33ndash46

Saboori A Kamali K 2000 Description of Allothrombiumtriticium adult (Acari Trombidiidae) from Iran Systematicand Applied Acarology 5 207ndash208

Schliwa M 1979 The retina of the phalangid Opilio raven-nae with particular reference to arhabdomeric cells Celland Tissue Research 204 473ndash495

Scholl G 1977 Beitrage zur Embryonalentwicklung von Lim-ulus polyphemus L (Chelicerata Xiphosura) Zoomorpholo-gie 86 99ndash154

Scholtz G Kamenz C 2006 The book lungs of Scorpionesand Tetrapulmonata (Chelicerata Arachnida) evidence forhomology and a single terrestrialization event of a commonarachnid ancestor Zoology 109 2ndash13

Seitz KA 1972 Zur Histologie und Feinstruktur des Herzensund der Haumlmocyten von Cupiennius salei Keys (AraneaeCtenidae) I Herzwandung Bildung und Differenzierungder Haumlmocyten Zoologische Jahrbuumlcher Abteilung fuumlr Anat-omie und Ontogenie der Tiere 89 351ndash384

Sekiguchi K ed 1988 Biology of horseshoe crabs Tokyo Sci-ence House

Selden PA 1981 Functional morphology of the prosoma ofBaltoeurypterus tetragonophthalmus (Fischer) (ChelicerataEurypterida) Transactions of the Royal Society of Edin-burgh Earth Sciences 72 9ndash48

Selden P 1992 Revision of the fossil ricinuleids Transactionsof the Royal Society of Edinburgh Earth Sciences 83 595ndash634

Selden PA 1993 Fossil arachnids ndash recent advances andfuture prospects Memoirs of the Queensland Museum 33389ndash400

Selden PA Dunlop JA 1998 Fossil taxa and relationshipsof chelicerates In Edgecombe GD ed Arthropod fossilsand phylogeny New York Columbia University Press 303ndash331

Selden PA Jeram AJ 1989 Palaeophysiology of terrestrial-

isation in the Chelicerata Transactions of the Royal Societyof Edinburgh Earth Sciences 80 303ndash310

Selden PA Shear WA Bonamo PM 1991 A spider and otherarachnids from the Devonian of New York and reinterpre-tations of Devonian Araneae Palaeontology 34 241ndash281

Sensenig A Shultz JW 2003 Mechanics of cuticular elasticenergy storage in leg joints lacking extensor muscles inarachnids Journal of Experimental Biology 206 771ndash784

Shear WA 1993 New species in the opilionid genus Stylocel-lus from Malaysia Indonesia and the Philippines (OpilionesCyphophthalmi Stylocellidae) Bulletin of the British Arach-nological Society 9 174ndash188

Shear WA Selden PA Rolfe WDI Bonamo PM GriersonJD 1987 New terrestrial arachnids from the Devonian ofGilboa New York (Arachnida Trigonotarbida) AmericanMuseum Novitates 2901 1ndash74

Shultz JW 1987 The origin of the spinning apparatus in spi-ders Biological Reviews of the Cambridge PhilosophicalSociety 62 89ndash113

Shultz JW 1989 Morphology of locomotor appendages inArachnida evolutionary trends and phylogenetic implica-tions Zoological Journal of the Linnean Society 97 1ndash56

Shultz JW 1990 Evolutionary morphology and phylogeny ofArachnida Cladistics 6 1ndash31

Shultz JW 1991 Evolution of locomotion in Arachnida thehydraulic pressure pump of the giant whipscorpion Masti-goproctus giganteus (Uropygi) Journal of Morphology 21013ndash31

Shultz JW 1993 Muscular anatomy of the giant whipscor-pion Mastigoproctus giganteus (Arachnida Uropygi) and itsevolutionary significance Zoological Journal of the LinneanSociety 108 335ndash365

Shultz JW 1998 Phylogeny of Opiliones (Arachnida) anassessment of the lsquoCyphopalpatoresrsquo concept Journal ofArachnology 26 257ndash272

Shultz JW 1999 Muscular anatomy of a whipspider Phrynuslongipes (Amblypygi) and its evolutionary significance Zoo-logical Journal of the Linnean Society 126 81ndash116

Shultz JW 2000 Skeletomuscular anatomy of the harvest-man Leiobunum aldrichi (Weed 1893) (Arachnida Opil-iones Palpatores) and its evolutionary significanceZoological Journal of the Linnean Society 128 401ndash438

Shultz JW 2001 Gross muscular anatomy of Limuluspolyphemus (Chelicerata Xiphosura) and its bearing on evo-lution in the Arachnida Journal of Arachnology 29 283ndash303

Shultz JW 2007 Morphology of the prosomal exoskeleton ofScorpiones (Arachnida) and a new hypothesis for the evolu-tion of cuticular cephalic endoskeletons in arthropodsArthropod Structure amp Development 36 77ndash102

Shultz JW Pinto da Rocha R 2007 Morphology and func-tional anatomy In Pinto da Rocha R Machado G Giribet Geds The harvestmen the biology of Opiliones CambridgeMA Harvard University Press 14ndash61

Shultz JW Regier JC 2001 Phylogenetic analysis of Pha-langida (Arachnida Opiliones) using two nuclear protein-encoding genes supports monophyly of Palpatores Journalof Arachnology 29 189ndash200

Sitnikova LG 1978 The main evolutionary trends of the

244 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Acari and the problem of their monophyletism Entomologis-cheskoe Obozrenie 57 431ndash457

Snodgrass RE 1948 The feeding organs of Arachnidaincluding mites and ticks Smithsonian Miscellaneous Col-lections 110 (10) 1ndash94

Snodgrass RE 1952 A textbook of arthropod anatomy Ith-aca NY Comstock Publishing

Soslashrensen W 1879 Om Bygningen af Gonyleptiderne en Typeaf Arachnidernes Classe Naturhistorisk Tidsskrifter Series3 8 97ndash222

Stockwell SA 1989 Revision of the phylogeny and higherclassification of scorpions (Chelicerata) Unpublished doc-toral dissertation University of California at Berkeley

Stoslashrmer L 1944 On the relationships and phylogeny of fossiland Recent Arachnomorpha Skrifter utgitt av Det NorskeVidenskaps-Akademi i Oslo I 5 1ndash158

Stoslashrmer L 1955 Merostomata In Moore RC ed Treatise oninvertebrate paleontology Part P Arthropoda 2 LawrenceKS University of Kansas Press and Geological Society ofAmerica 4ndash41

Swofford DL 2002 PAUP Phylogenetic analysis using par-simony (and other methods) Version Beta 10 SunderlandMA Sinauer Associates

Talarico G Palacios-Vargas JG Fuentes Silva M AlbertiG 2005 First ultrastructural observations of the tarsal poreorgan of Pseudocellus pearsei and P boneti (ArachnidaRicinulei) Journal of Arachnology 33 604ndash612

Templeton AR 1983 Phylogenetic inference from restrictionendonuclease cleavage site maps with articular reference tothe evolution of humans and the apes Evolution 37 221ndash244

Tetlie OE Braddy SJ 2004 The first Silurian chasmataspidLoganamaraspis dunlopi gen et sp nov (Chelicerata Chas-mataspidida) from Lesmahagow Scotland and its implica-tions for eurypterid phylogeny Transactions of the RoyalSociety of Edinburgh Earth Sciences 94 227ndash234

Thomas RH Telford MJ 1999 Appendage development inembryos of the oribatid mite Archegozetes longisetosus(Acari Oribatei Trhypochthoniidae) Acta Zoologica (Stock-holm) 80 193ndash200

Tollerton VP Jr 1989 Morphology taxonomy and classifica-tion of the order Eurypterida Burmeister 1843 Journal ofPaleontology 63 642ndash657

Vachon M 1949 Ordre des Pseudoscorpions In Grasseacute P-Ped Traite de zoologie VI Paris Masson et Cie 436ndash481

Van der Hammen L 1979 Comparative studies in Chelicer-ata I The Cryptognomae (Ricinulei Architarbi and Anacti-notrichida) Zoologische Verhandelingen 174 1ndash62

Van der Hammen L 1982 Comparative studies in Chelicer-ata II Epimerata (Palpigradi and Actinotrichida) Zoolo-gische Verhandelingen 196 1ndash70

Van der Hammen L 1985 Functional morphology and affin-ities of extant Chelicerata in evolutionary perspectiveTransactions of the Royal Society of Edinburgh 76 137ndash146

Van der Hammen L 1989 An introduction to comparativearachnology Leiden SPB Academic Publishing

Vyas AB 1970 Studies in myology of the scorpion Heter-ometrus fulvipes Koch Unpublished PhD DissertationDepartment of Zoology Gujarat University India

Vyas AB 1974 The cheliceral muscles of the scorpion Heter-ometrus fulvipes Bulletin Southern California Academy ofSciences 73 9ndash14

Walter DE 1988 Predation and mycophagy by endeostigma-tid mites (Acariformes Prostigmata) Experimental andApplied Acarology 4 159ndash166

Walter DE Proctor HC 1998 Feeding behaviour and phy-logeny observations on early derivative Acari Experimentaland Applied Acarology 22 39ndash50

Walter DE Proctor HC 1999 Mites ecology evolution andbehaviour Sydney University of New South Wales

Waterston CD 1979 Problems of functional morphology andclassification in stylonuroid eurypterids (Chelierata Meros-tomata) with observations on the Scottish Silurian Sty-lonuroidea Transactions of the Royal Society of EdinburghEarth Science 70 251ndash322

Weygoldt P 1969 Biology of pseudoscorpions CambridgeMA Harvard University Press

Weygoldt P 1984 Lrsquoautotomie chez les Amblypyges RevueArachnologique 5 321ndash327

Weygoldt P 1996 Evolutionary morphology of whip spiderstowards a phylogenetic system (Chelicerata ArachnidaAmblypygi) Journal of Zoological Systematics and Olution-ary Research 34 185ndash202

Weygoldt P 1998 Evolution and systematics of the Chelicer-ata Experimental and Applied Acarology 22 63ndash79

Weygoldt P 2000 Whip spiders (Chelicerata Amblypygi)their biology morphology and systematics StenstrupDenmark Apollo Books

Weygoldt P Paulus HF 1979 Untersuchungen zur Morphol-ogie Taxonomie und Phylogenie der Chelicerata Zeitschriftfuumlr Zoologische Systematik und Evolutionsforschung 17 85ndash116 177ndash200

Wheeler WC Hayashi CY 1998 The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192

Winkler D 1957 Die Entwicklung der aumlusseren Koumlrperge-stalt bei den Phalangiidae (Opiliones) Mitteilungen aus demZoologischen Museum in Berlin 33 355ndash389

With CJ 1904 The Notostigmata a new suborder of AcariVidenskabelige Meddelelser fra Dansk Naturhistorisk Foren-ingi Kjobenhaus 1904 137ndash192

Yamasaki T Makioka T Saito J 1988 Morphology InSekiguchi K ed Biology of horseshoe crabs Tokyo ScienceHouse 69ndash132

Yoshikura M 1975 Comparative embryology and phylogeny ofArachnida Kumamoto Journal of Science Biology 12 71ndash142

APPENDIX

Each character used in this analysis is numbereddefined and cross-referenced with characters used infour previous studies that are abbreviated in bracketsspecifically WP = Weygoldt amp Paulus (1979)S = Shultz (1990) WH = Wheeler amp Hayashi (1998)GEWB = Giribet et al (2002) Descriptions includecorrections of errors discussions of controversies orambiguities and justifications for state assignmentsto lsquoproblemrsquo taxa

PHYLOGENY OF THE ARACHNID ORDERS 245

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

PROSOMA

1 Dorsal sclerite formed by fusion of the prosomal car-apace the dorsal portion of the first opisthosomalsomite and the dorsomedial (axial) portion of the sec-ond opisthosomal somite 0 absent 1 presentState 1 is present in extant xiphosurids and close fos-sil relatives (Scholl 1977 Anderson amp Selden 1997Shultz 2001) The fossil record suggests that thisstructure originated with the disappearance ofthe already reduced first opisthosomal tergite(= microtergite) and axial portion of the secondopisthosomal tergite (Anderson amp Selden 1997)2 Single dorsal sclerite covering entire dorsal surfaceof body no lines indicating original segmentation 0absent 1 presentState 1 occurs throughout Holothyrida (Acari) (Vander Hammen 1989) and in many Mesostigmata(Alberti amp Coons 1999) including Glyptholaspis (Vander Hammen 1989)3 Anterior end of dorsal prosoma with medianmarginal or submarginal pointed process 0 absent1 presentState 1 occurs throughout Schizomida (Protoschizo-mus Stenochrus Reddell amp Cokendolpher 1995) The-lyphonida (Rowland amp Cooke 1973 Shultz unpublobs Proschizomus Dunlop amp Horrocks 19951996)Amblypygi (Charinus Millot 1949b Phrynus Shultz1999) and Plesiosiro (Dunlop 1999) Citing a figure inRowland amp Cooke (1973) Dunlop amp Horrocks assumedthat the anterior process was limited to lsquohypoctonidrsquothelyphonids and united them with Proschizomus andSchizomida If valid this would make Thelyphonidaparaphyletic However the doubtful monophyly oflsquohypoctonidsrsquo (Haupt amp Song 1996) and widespreadpresence of a submarginal anterior process in Pedi-palpi weaken this proposal (Harvey 2002)4 Ophthalmic ridges pair of longitudinal crests inter-secting or passing near the region of the lateral eyes orcomparable region where lateral eyes are absent 0absent 1 presentState 1 is present throughout Xiphosura (Anderson ampSelden 1997) Similar structures are present in Plesi-osiro (Dunlop 1999) and lsquonon-hypoctonidrsquo Thely-phonida (Mastigoproctus Rowland amp Cooke 1973)Dunlop (2002a) described apparent ophthalmic ridgesin the chasmataspidid Octoberaspis but these werenot observed in Diploaspis (Dunlop et al 2001) eitherdue to absence or inadequate preservation of speci-mens nor in Chasmataspis (Dunlop et al 2004)5 Carapace with demarcations (eg grooves scleritesphragmata) between pro- meso- or metapeltidia 0absent 1 present [S 1 WH 48 GEWB 29]State 1 is present in extant Scorpiones Cyphoph-thalmi (Hansen amp Soslashrensen 1904) and many Palpa-tores (Opiliones) including Caddo and Leiobunum

and most Pseudoscorpiones (Chamberlin 1931) Giri-bet et al (2002) coded this character as absent inCyphophthalmi (Opiliones) but a transverseprocurved groove is present throughout the group(Hansen amp Soslashrensen 1904) and represents the meso-metapeltidial border as indicated by attachment ofthe dorsal endosternal suspensor of postoral somite VIin Chileogovea oedipus and Siro exilis (Shultz unpublobserv) Kjellesvig-Waering (1986) did not explicitlydescribe such grooves in fossil scorpions although heillustrated corresponding structures in PrearcturusArchaeophonus Archaeoctonus Palaeophonus andother Palaeozoic scorpions (Kjellesvig-Waering 1986figs 14A 21E 27D 59) (coded as uncertain here) butnot in Proscorpius (contra Giribet et al 2002) Dis-tinct carapacal sclerites are present throughout Palpi-gradi Schizomida and Solifugae (6)

The condition in Acari is controversial but is codedhere as State 1 Van der Hammen (1989) consideredthe tergal region of the leg-bearing somites (postoralsomites IIIndashVI) or podosoma to be reduced andreplaced by posterior migration of the aspidosoma (ietergal region assumed to be associated with append-ages of the gnathosoma) and anterior migration of theopisthosomal tergal region to form the hysterosomaThe dorsal proterosomal and hysterosomal elementsmeet at a transverse sejugal furrow that continues lat-erally and passes ventrally between the coxae of legs 2and 3 According to this scheme the prosomandashopisthosoma border is expressed as the disjugal fur-row which passes from the ventral posterior margin ofthe podosoma anterodorsally to join the dorsal part ofthe sejugal furrow Based on the arrangement of setaeand slit sensilla (= lyrifissures) Van der Hammen(1989) interpreted the region above coxae 3 and 4 (ieregion C) as a fusion of the dorsal parts of the first twoopisthosomal somites (= postoral somites VII and VIII)in early divergent Anactinotrichida (ie Opilioacari-formes) and Acariformes (eg Alycus) These interpre-tations have been followed by many acarologistsalthough its speculative aspects are widely acknowl-edged (Evans 1992 Alberti amp Coons 1999)

From the standpoint of a general arachnologist theVan der Hammen system seems unnecessarily compli-cated it appears to have been formulated to explainbroad morphological themes in Oribatida but was thenextrapolated to other mites A more conservativelsquoarachnologicalrsquo scheme adopted here equates the pro-podosoma with the pars cephalica (= propeltidium) ofthe arachnid prosoma and region C with the pars tho-racica (= mesopeltidium + metapetidium) In fact theearliest and most recent treatments of Opilioacari-formes (With 1904 Klompen 2000) reached similarconclusions This interpretation is also consistent withevidence from Acariformes For example the regiondorsal to the coxa of legs 3 and 4 in Alycus is as readily

246 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

explained by the persistence of primitive prosomal ter-gal elements as by the supposition that these disap-peared and were replaced by two opisthosomalelements Furthermore several diverse lineages ofbasally divergent Acariformes display features that donot fit readily within the GrandjeanndashVan der Ham-men system but seem to correspond to the cephalicathoracica division of the prosoma They have a sejugalfurrow that may have demarcated the cephalicatho-racica border and a postpodosomal furrow thatappears to represent the prosomandashopisthosoma junc-tion there is no disjugal furrow Examples includepalaeosomate (especially aphelacarid) and pediculoch-elid Oribatida alicorhagiid and mircopsammidlsquoEndeostigmatarsquo and paratydeid Prostigmata (Albertiamp Coons 1999 fig 42) This organization may be sym-plesiomorphic for Acariformes but could be a conver-gence brought about by selection for enhancedflexibility6 Carapace with distinct pro- meso- or metapeltidialsclerites 0 absent 1 present - inapplicable due toabsence of pro- meso- or metapeltidial demarcations(5)State 1 occurs throughout Palpigradi Schizomida andSolifugae It also appears sporadically within Opil-iones (eg Leiobunum flavum Shultz unpubl observ)and Pseudoscorpiones (eg Pseudochiridium) (Cham-berlin 1931)7 Sejugal furrow circumferential zone of body flexi-bility that passes between the coxae of legs 2 and 3 0absent 1 presentState 1 is probably synapomorphic for Acariformes asit occurs in most lsquoendeostigmatidsrsquo including Alycus(Van der Hammen 1989) and is variously developedthroughout Sarcoptiformes (Alberti amp Coons 1999) Itis probably the primitive condition for Prostigmatabut is either absent or weakly expressed in represen-tatives included here See 5 for alternative morpholog-ical interpretations of this body region in Acari8 Prosomal ozopores 0 absent 1 present [WH 46GEWB 12]State 1 occurs throughout Opiliones and Holothyrida(Van der Hammen 1989)9 Carapacal pleural doublure 0 absent 1 present[WP 12 S 2 WH 49 GEWB 27]A carapacal pleural doublure occurs throughoutXiphosura (Stoslashrmer 1944) Eurypterida (Clarke ampRuedemann 1912 Selden 1981) and Chasmataspid-ida (Dunlop et al 2001 2004 Dunlop 2002a) and isabsent in all known Arachnida10 Cardiac lobe a longitudinal axial elevation of thecarapace 0 absent 1 presentState 1 occurs throughout Xiphosura (Anderson ampSelden 1997) and Eurypterida (Clarke amp Ruedemann1912 Selden 1981) Carapaces are not preserved suf-ficiently to determine this condition in the chasmatas-

pidids Diploaspis (Dunlop et al 2001) andOctoberaspis (Dunlop 2002a) but an apparent cardiaclobe occurs in the Chasmataspis (Dunlop et al 2004)A cardiac lobe is absent in Arachnida11 Moveable cucullus 0 absent 1 present [WP 54GEWB 22]State 1 is a synapomorphy of Ricinulei (Selden 1992)12 Medial intercoxal lsquosternalrsquo region 0 all pedalcoxae separated medially 1 anterior pedal coxae abut-ting medially posterior coxae separated 2 anteriorpedal coxae separated medially posterior coxae abut-ting 3 all pedal coxae abutting medially 4 epimeracoxae undifferentiated medially from ventral body wall[S sim3 WH sim50 GEWB sim17]Prosomal sternites appear to be those portions of theventral body wall not occupied by coxae Despite earlyattempts to assess metamerism of the ventralprosoma (protosternum deutosternum tritosternumetc) (Boumlrner 1902a Millot 1949a) no apparent mor-phological features such as borders of sclerites ormuscle attachments reliably demarcate the ventralbody wall of one somite from that of an adjacentsomite Thus coding schemes that focus on describingmetameric components of the sternal region are prob-lematic Shultz (1990) attempted to code the shape ofthe entire intercoxal region and subsequent workershave adopted this approach All pedal coxae are sepa-rated medially in Plesiosiro (Dunlop 1999) Palpigradi(Roewer 1934) Araneae (Millot 1949c) Amblypygi(Weygoldt 2000) and many Acari including Opilioac-ariformes Holothryrida Mesostigmata Ixodida Aly-cus and many Prostigmata (Van der Hammen 1989Evans 1992 Alberti amp Coons 1999) The anteriorcoxae are separated in extant Thelyphonida Schizo-mida (Hansen amp Soslashrensen 1905 Millot 1949e) andRicinulei (Pittard amp Mitchell 1972 Legg 1976) Thecoxae are fused to the ventral body wall in Sarcopti-formes (Van der Hammen 1989 Evans 1992)

Interpreting this character is complicated for theanterior coxae of Scorpiones and Opiliones and allcoxae in Xiphosura and Eurypterida due to the pres-ence of coxapophyses endites andor gnathobaseswhich form components of the preoral chambers inthese groups Thus coxae may abut medially whenthese structures are considered part of the coxae butare separate medially if these structures are ignoredAs coxapophyses etc are here coded as separate char-acters (50ndash53) I have chosen to code this character asif these structures did not exist Consequently pha-langid Opiliones are coded as having all coxae sepa-rated (Pocock 1902 Hansen amp Soslashrensen 1904) Fossilscorpions also show substantial variation in this char-acter All pairs of pedal coxae in Palaeoscorpius abutmedially but also have gnathobases (Jeram 1998)and this condition is coded here as uncertain A ster-num separates all pairs of pedal coxae medially in

PHYLOGENY OF THE ARACHNID ORDERS

247

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Proscorpius

and

Stoermeroscorpio

(State 0) and

Pre-arcturus

approximates modern scorpions (State 1)(Kjellesvig-Waering 1986 Jeram 1998) The coxalbases of Eurypterida and Xiphosura are separated bya small sternite the endostoma (Selden 1981 Shultz2001)

13 Postoral sternapophysis (

=

tritosternum labium)a cuticular evagination of the ventral body wall poste-riorly adjacent to the palpal coxae forming the poste-rior border of the preoral chamber in some taxa ordisplaced posteriorly by fusion of the palpal coxae inothers 0 absent 1 present

[S

sim

10 WH

sim

56 GEWB19

=

33]State 1 is present throughout Palpigradi AraneaeAmblypygi (Snodgrass 1948) Schizomida (Hansen ampSoslashrensen 1905 Van der Hammen 1989) Thely-phonida (Van der Hammen 1989) extant Ricinulei(Pittard amp Mitchell 1972 Legg 1976) Opilioacari-formes and Mesostigmata (Van der Hammen 1989Alberti amp Coons 1999)

Allothyrus constrictus

is theonly holothyrid mite known to have sternapophysesAn apparent sternapophysis has been described in apalaeocharinid trigonotarbid (Dunlop 1994) Giribet

et al

(2002) miscoded Ixodida and Acariformes as hav-ing a lsquolabium (

=

tritosternum)rsquo The lophognath ofpseudoscorpions (Chamberlin 1931) and lsquolabiumrsquo ofSolifugae (Roewer 1934) also appear to be sternapo-physes that have been incorporated into the ros-trosoma (

32

) The labium of phalangid Opiliones(Pocock 1902 Hansen amp Soslashrensen 1904) is a ster-napophysis but is associated with leg 1 rather thepedipalp (Winkler 1957) A small anterior sclerite inthe fossil scorpions

Proscorpius

Waeringoscorpio

and

Labriscorpio

has been interpreted as a lsquolabiumrsquo homol-ogous with the sternapophyses of other arachnids(Weygoldt 1998) despite absence of any indication ofthe relationship of this sclerite to the mouth and pos-sible alternative interpretations (eg

16

) Giribet

et al

coded this sclerite as a sternapophysis but the char-acter is coded here as uncertain in fossil scorpions

14 Channels on the body surface linking openings ofcoxal organs to preoral chamber 0 absent 1 present

Van der Hammen (1989) described bilaterally pairedcuticular tracts (lsquotaenidiarsquo) connecting orifices of coxalorgans to the preoral chamber in some arachnids andor a ventromedian groove (

=

intercoxal or subcapitulargutter) that leads in turn to the preoral chamber inother taxa Taenidia and intercoxal gutters arepresent in Opilioacariformes (Van der Hammen1989) Holothyrida (Van der Hammen 1989) Mesos-tigmata (

Glyptholapis

Van der Hammen 1989)Ricinulei (Pittard amp Mitchell 1972 Van der Hammen1989) Araneae (

Heptathela

Aphonopelma

Hypochi-lus

Shultz unpubl observ also

Porrhothele

(Mygalo-morphae Dipluridae) Butt amp Taylor 1991

Segestria

(Araneomorphae Segestriidae) Van der Hammen

1989) Amblypygi (Van der Hammen 1989) Schizo-mida (

Stenochrus

Shultz unpubl observ unidenti-fied Hubbardiinae Van der Hammen 1989) andThelyphonida (

Mastigoproctus

Shultz unpublobserv also

Tetrabalius

Van der Hammen 1989) Theconduction of large volumes of fluid from orifices of thecoxal organs to the preoral chamber has been observedin the mygalomorph spider

Porrhothele

(Dipluridae)(Butt amp Taylor 1991) and

Mastigoproctus

(Thely-phonida) (Shultz unpubl observ) The podocephaliccanal (

15

) is present in Acariformes (Acari) (Lindquist1984) and appears to have evolved as an invaginatedsupracoxal channel (Grandjean 1938 G Alberti inEvans 1992) Van der Hammen (1982) could not dis-cern with certainty whether a channel exists in

Eukoenenia

(Palpigradi) but none has been reportedby previous authors and they have not been observedby the present author Palpigradi is coded as havingState 0

15 Podocephalic canal cuticular channel andorduct draining multiple glands and opening nearmouthparts 0 absent 1 present - inapplicable due toabsence of channel (14)

State 1 occurs throughout Acariformes (Acari)(Lindquist 1984)

16 Heavily sclerotized suboral sclerite serving inpart as basal pivot point for coxae of appendages ofpostoral somites IIndashIV (

=

arachnid palp and legs 1 and2) 0 absent 1 present

The sclerite forms the posterior wall of the true mouthand is heavily sclerotized in extant scorpions It is notvisible externally due to the tightly fitting coxae of theanterior prosomal appendages (Shultz 2007) It maycorrespond to the so-called labium (Weygoldt 1998) orlabrum (Kjellesvig-Waering 1986) of certain fossilscorpions (see

13

)

17 Genal angles 0 rounded 1 pointed

Anderson amp Selden (1997) originally coded a pointedgenal angle and genal spine as separate charactersbut they are combined here State 1 is present in manyXiphosura (Anderson amp Selden 1997) and Chas-mataspidida (Dunlop

et al

2001 2004 Dunlop2002a)

A

PPENDAGES

OF

POSTORAL

SOMITE

I

CHELICERAE

18 Cheliceral segmentation 0 three articles 1 twoarticles

[WP

sim

25

+

42 WH

sim

15 GEWB

sim

44]State 0 is present in extant Xiphosura Eurypterida(Clarke amp Ruedemann 1912 Selden 1981) Palpi-gradi (Roewer 1934) Opiliones (Hansen amp Soslashrensen1904) Scorpiones (Hjelle 1990) and throughout non-acariform Acari (Evans 1992) State 1 is presentthroughout known Trigonotarbida Araneae Ambly-pygi Thelyphonida Schizomida Pseudoscorpionesand Solifugae Acariform mites have two unambigu-

248 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ous cheliceral articles but may have a reduced thirdarticle (Van der Hammen 1989 Evans 1992) acari-form mites are thus coded as uncertain Chelicerae arenot known or their condition is unclear in several fos-sil taxa including Xiphosura Chasmataspidida (Dun-lop et al 2001 2004 Dunlop 2002a) Riciniulei(Selden 1992) Plesiosiro (Dunlop 1999) Proschizo-mus (Dunlop amp Horrocks 19951996) and the fossilscorpions Praearcturus and Palaeoscorpius (Kjelles-vig-Waering 1986) Kjellesvig-Waering (1986) consis-tently interpreted chelicerae of fossil scorpions ashaving four articles but this is probably a misinter-pretation of the basal article (see also Stockwell1989) The basal article in extant scorpions consists ofa distal collar of cuticle and a large proximal processand this arrangement would probably appear as twoarticles in compressed specimens In any event thereare at least three cheliceral articles in Stoermeroscor-pio and Proscorpius (Kjellesvig-Waering 1986)19 Terminal cheliceral joint 0 laterally placedbicondylar hinge 1 dorsally placed bicondylar hinge2 ventrally placed bicondylar hinge [WP 25 + 42 Ssim11 + 12 WH sim15 GEWB sim44]State 0 is present in Xiphosura (Shultz 2001)extant Scorpiones (Millot amp Vachon 1949) Opil-iones (Hansen amp Soslashrensen 1904) Palpigradi(Boumlrner 1904 Millot 1949d) Ricinulei (Pittard ampMitchell 1972 Legg 1976) and Eurypterida includ-ing Baltoeurypterus (Selden 1981) and Stylonurus(Clarke amp Ruedemann 1912) State 0 is present inthe fossil scorpions Proscorpius and Stoermeroscor-pio but the state of the chelicerae is unclear in theother fossil scorpions represented here (Kjellesvig-Waering 1986) State 1 is present in Trigonotarbida(Shear et al 1987) mygalomorph Araneae Ambly-pygi Schizomida and Theylphonida (Weygoldt ampPaulus 1979) State 2 is present in Pseudoscorpi-ones (Chamberlin 1931) and Solifugae (Roewer1934) and appears to be a groundplan feature for allmajor groups of mites (eg Van der Hammen 1989Alberti amp Coons 1999) The difference between theorthognathy of mygalomorph Araneae and lsquoplagiog-nathyrsquo mesothele (Liphistius Heptathela) and palae-ocribellate Araneae (Hypochilus) (Kraus amp Kraus1993) is small compared with the variation thatoccurs throughout arachnids and plagiognathy iscoded as State 1 here20 Chelicera articulating with carapace at anterolat-eral pivot 0 absent 1 present [S 13 WH 57 GEWB45]State 1 occurs in Solifugae (Roewer 1934) and Pseu-doscorpiones except Chthonioidea and Feaelloidea(Chamberlin 1931) Giribet et al (2002) coded thecharacter as present in Pseudoscorpiones21 Chelicera pivoting on dorsal protuberance of epis-tome 0 absent 1 present

State 1 appears to be a unique synapomorphy of Opil-iones (Shultz 2000)22 Extrinsic cheliceral muscle arising on carapaceand inserting on dorsal margin of nonbasal cheliceralarticle 0 absent 1 present (Note Characters 22 and23 refer to two different muscles)State 1 is known only in extant Scorpiones (LankesterBenham amp Beck 1885 Vyas 1970 1974 Shultz2007) but may occur in acariform mites if their cheli-cerae are composed of three rather than two articles(Evans 1992) (see 18) Acariformes are coded asuncertain23 Extrinsic cheliceral muscle arising on carapaceand inserting on ventral margin of nonbasal cheliceralarticle 0 absent 1 present [GEWB 252] (Note Char-acters 22 and 23 refer to two different muscles)State 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) somemesostigmatid Acari (Evans 1992) and representa-tive Opiliones particularly Leiobunum (Shultz 2000)Chileogovea and Gonyleptes (Shultz unpubl observ)The status of the character in other opilions is not yetknown The character was originally defined as anextrinsic muscle inserting on the second segment ofthree-segmented chelicerae (Shultz 2000 Giribetet al 2002) but it has been redefined here to make itapplicable to arachnids with two-segmented cheli-cerae24 Extrinsic cheliceral muscles attaching to epistome0 absent 1 presentState 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) but mayoccur in Prostigmata if the sigmoid piece to which thecheliceral protractors attach (Evans 1992) is a modi-fication of the epistome25 Lateral tergocheliceral muscle with three heads 0absent 1 present - coded only for extant tetrapulmo-nates and the palpigrade Eukoenenia homology isunclear in other taxa [GEWB 230]State 1 is present in Pedipalpi (Shultz 1999)26 Cheliceral silk glands 0 absent 1 present [GEWB47]State 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931) The silk gland of spider mites(eg Tetranychus) are associated with the pedipalps(Evans 1992)27 Cheliceral venom glands 0 absent 1 present [WP34 WH 40 GEWB 46]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)28 Cheliceral serrula interior and exterior 0 absent1 presentState 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931 Harvey 1992)29 Cheliceral lsquoflagellumrsquo in male 0 absent 1 present[GEWB 48]

PHYLOGENY OF THE ARACHNID ORDERS 249

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITE II LEG 1 IN NONARACHNIDS AND PALP IN ARACHNIDS

30 Palpal coxae fused ventromedially and formingposterior wall of preoral chamber 0 absent 1 present[WP 27 S 18 WH 17 = 62 GEWB 33 + 63]State 1 occurs in Schizomida Thelyphonida Acari andRicinulei31 Gnathosoma 0 absent 1 present - inapplicablepalpal coxae not fused (30) [WP 56 S 19 WH sim63GEWB sim36 + 37]The gnathosoma is a functional complex comprisingan epistome attached to the dorsal surface of mediallyfused palpal coxae (= infra- or subcapitulum) (see 30)the chelicerae and in some taxa a supracheliceral tec-tum These components are tightly integrated andmove in unison relative to the body The gnathosomahas traditionally been considered a feature of AcariHowever Van der Hammen (1989) argued that a gna-thosoma is present in Ricinulei and this interpreta-tion has received wide acceptance (eg Lindquist1984) However it is not clear that the relationship ofthe chelicerae and subcapitulum is as intimate inRicinulei as in Acari and the ricinuleid subcapitulumdoes not appear to be as moveable In fact the lsquosubca-pitulumrsquo in Ricinulei is similar in many respects to thecondition in Schizomida and Thelyphonida (Hansen ampSoslashrensen 1904) which has never been regarded as agnathosoma Consequently I have coded the conditionof Ricinulei as uncertain32 Rostrosoma long narrow subcylindrical epistomeprojecting anteriorly with base fixed to dorsal surfaceof palpal coxae bordered laterally by lobes projectingfrom palpal coxae ventral wall of preoral chamberformed by anterior element of prosoma (sternapophy-sis) 0 absent 1 present [S 15 WH 59 GEWB 20 + 36]State 1 is present in Pseudoscorpiones and Solifugae(Chamberlin 1931 Roewer 1934 Van der Hammen1989) Dunlop (2000) attempted to homologize com-ponents of the rostrosoma and gnathosoma (31) anargument accepted by Giribet et al (2002) Howeverfusion of the palpal coxae (30) apparently evolvedindependently in the two structures Coxal fusion inthe gnathosoma is complete ventrally and excludesthe suboral sternapophysis (13) while fusion isincomplete ventrally in the rostrosoma and incorpo-rates the sternapophysis Unlike the gnathosomathe rostrosoma is largely immobile and cheliceralmovement is not coupled with that of the palpalcoxae33 Rutellacorniculi hypertrophied setae modified asmouthparts located on the anterior processes of the pal-pal coxae 0 absent 1 present [GEWB 39]

State 1 occurs in the anactinotrichid groups Opilioac-ariformes Holothyrida and Mesostigmata but not Ixo-dida (Van der Hammen 1989) It also occurs inSarcoptiformes (Evans 1992) and certain endeostig-matids (eg Alycus Van der Hammen 1989) but not inProstigmata Comparable structures are unknown inother chelicerate groups34 Terminal segments specially modified in adultmale as a clasper used to engage the female 0 absent1 presentState 1 is known only in extant Xiphosura (YamasakiMakioka amp Saito 1988)35 Robust raptorial 0 absent 1 present [S 17 WH61 GEWB sim59 97]State 1 is present in Amblypygi Thelyphonida andSchizomida (Shear et al 1987) and in Laniatores(Opiliones)36 Orientation of robust raptorial appendage 0operating in subtransverse plane 1 operating in sub-vertical plane - inapplicable coded only for PedipalpiState 0 is present in extant Thelyphonida and mostAmblypygi and State 1 is present in Schizomida (Cok-endolpher amp Reddell 1992) and in the basally diver-gent amblypygid Paracharon caecus (Weygoldt 2000)The orientation of the pedipalps in the fossil Proschi-zomus is uncertain (Dunlop amp Horrocks 19951996)37 Extrinsic muscle attaching to epistome 0 absent1 presentState 1 occurs in Chileogovea and Siro (OpilionesCyphophthalmi) and extant Scorpiones (Vyas 1970Shultz 2007)38 Muscle originating and inserting within coxa 0absent 1 present [GEWB sim232]State 1 occurs is in Amblypygi (Phrynus Shultz 1999Phrynichus Boumlrner 1904) Thelyphonida (Mastigo-proctus Shultz 1993 Thelyphonus Boumlrner 1904) andSchizomida (Stenochrus Shultz unpubl observ)There are two intracoxal muscles in amblypygids andthelyphonids the one coded here and another that isserially homologous with a muscle in the pedal coxae(54)39 Tarsus andor tibia with venom glands 0 absent1 present [GEWB 64]State 1 is known only from extant Pseudoscorpi-ones except the superfamilies Chthonioidea andFeaelloidea (Chamberlin 1931 Harvey 1992) Giri-bet et al (2002) coded this character as present forPseudoscorpiones40 Tarsal grooming organ 0 absent 1 present[GEWB 98]State 1 is a unique synapomorphy of Amblypygi (DelleCave 1975 Weygoldt 2000)41 Scorpionoid chela a large well-developed chelaformed by tibia (manus + fixed finger) and tarsus(moveable finger) 0 absent 1 present [S 16 WH 60GEWB 62]

250 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs throughout Scorpiones andPseudoscorpiones42 Modified in male as copulatory organ 0 absent 1present [WP 35 WH 39 GEWB 67]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)43 Apotele 0 apparently absent not differentiatedexternally from penultimate article 1 present [GEWB69]The apotele exists as a structure (ie dactyl clawempodium) differentiated from the tibiotarsus tarsusor telotarsus and is typically associated with a pair ofantagonistic muscles The apotele occurs as a distinctstructure in Xiphosura (Snodgrass 1948) Eurypter-ida (Clarke amp Ruedemann 1912) Schizomida (Coken-dolpher amp Reddell 1992) Araneae (Foelix 1996)early divergent Amblypygi (Charinus Weygoldt 2000not Phrynus Shultz 1999) Anactinotrichida (CaminClark amp Bourdeau 1956 Lindquist 1984) Ricinulei(Pittard amp Mitchell 1972 Legg 1976) and Solifugae(Roewer 1934) Giribet et al (2002) coded Solifugae asnot having a palpal apotele but it is present as anadhesive organ (45) Dunlop (1999) reconstructedHaptopoda as having a palpal claw but notes in hisdescription that the distal ends were not preserved inthe known specimens44 Apotele (claw) position 0 terminal 1 subtermi-nal - inapplicable coded only for taxa with a distinctapotele (43)State 1 occurs in Parasitiformes (Acari) (Evans 1992)45 Terminal adhesive organ 0 absent 1 present -inapplicable due to lack of an apotele (43) [GEWB 66]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITES IIIndashVI LEGS 2ndash5 IN NON-ARACHNIDS AND LEGS 1ndash4 IN ARACHNIDS

46 Appendage III (= arachnid leg 1) extremely elon-gate antenniform 0 absent 1 present [WP sim24 + 32S 20 WH 14 GEWB sim84]State 1 occurs in Amblypygi Thelyphonida Schizo-mida (Shear et al 1987) and Opilioacariformes (Vander Hammen 1989) State 1 is approximated in Solif-ugae Palpigradi (Roewer 1934) and Haptopoda(Dunlop 1999) and it is coded as uncertain in thesetaxa47 Appendage V (= arachnid leg 3) of male specializedfor sperm transfer 0 absent 1 present [WP 55 GEWB94]State 1 is known only from extant Ricinulei48 Appendages V and VI (= arachnid legs 3 and 4)with femur shorter than patella and with principal siteof flexionextension at patella-tibia joint (lsquoapatellatersquocondition sensu Van der Hammen 1989) 0 absent 1present [S 25 GEWB 86]

State 1 occurs in Pseudoscorpiones and Solifugae(Shultz 1989)49 Appendage VI (= arachnid leg 4) with terminusmodified as flattened blade 0 absent 1 presentState 1 occurs in some Chasmataspidida (DiploapisDunlop et al 2001 Octoberaspis Dunlop 2002a)uncertain in Chasmataspis Dunlop et al 2004 State0 in Loganamaraspis Tetlie amp Braddy 2004) andEurypterida except stylonuroids (Clarke amp Ruede-mann 1912 Waterston 1979 Tollerton 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI COXA AND BODYndashCOXA JOINT

50 Appendage III (= arachnid leg 1) with coxapophy-sis forming floor or wall of preoral chamber 0 absent1 present [S sim14 WH sim58]State 1 occurs in Opiliones (Hansen amp Soslashrensen 1904)and higher scorpions (Jeram 1998) A lobe is presenton the coxa of leg 1 in the fossil scorpion Prearcturusand perhaps Stoermeroscorpio and is coded here as acoxapophysis Coxapophysis of leg 1 together with theepistome and coxapophysis of the palp form a uniquepreoral chamber the stomotheca Weygoldt (1998) andDunlop amp Braddy (2001) reject the homology of cox-apophyses and stomothecae in Opiliones and Scorpi-ones because these structures are not apparent infossils of those scorpions thought to have been aquaticHowever the coxapophyses in Opiliones are formedlargely from soft cuticular lsquolipsrsquo and similar structuresin fossil scorpions would probably not have beenpreserved51 Appendage IV (= arachnid leg 2) with coxapophy-sis 0 absent 1 present [S sim14 WH sim58 GEWB 75]State 1 occurs in extant Scorpiones and most Opil-iones except Dyspnoi (Shultz 1998)52 At least one pair of coxal gnathobases on append-ages III-VI (= arachnid legs 1ndash4) 0 absent 1 present[S sim23 WH sim66 GEWB 78]State 1 occurs in all appropriately preserved Euryp-terida including Baltoeurypterus (Selden 1981) andStylonurus (Clarke amp Ruedemann 1912) extantXiphosura on appendages IIndashVI and apparentlyProscorpius on appendage III (= arachnid leg 1)(Kjellesvig-Waering 1986 Jeram 1998 contra Giribetet al 2002) Fossilized appendages with gnathobaseshave been found in association with Chasmataspis(Chasmataspidida) suggesting that they may havehad State 1 (Dunlop et al 2004)53 Coxae of appendages IIIndashV (= arachnid legs 1ndash3)with jointed moveable endites 0 absent 1 present [Ssim23]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) and in adequately preserved eurypterids(Clarke amp Ruedemann 1912) (BaltoeurypterusSelden 1981)

PHYLOGENY OF THE ARACHNID ORDERS 251

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

54 Intracoxal muscle a muscle arising on anteriorwall of coxa and inserting on posterior wall 0 absent1 present [GEWB 237]State 1 occurs in Thelyphonida (MastigoproctusShultz 1993) and Amblypygi (Phryus Shultz 1999)It is unclear whether the muscle operating the move-able endite (53) in extant Xiphosura (Manton 1964Shultz 2001) is homologous with this muscle and iscoded here as uncertain It is not present in Steno-chrus (Schizomida) (Shultz unpubl observ)55 Coxae of appendages IIndashVI (= arachnid palp andlegs) with dorsal articulation with carapace 0 absent1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)56 Flabellum (exite) on coxa of appendage VI(= arachnid leg 4) 0 absent 1 present [GEWB sim110]State 1 occurs in extant Xiphosura (Yamasaki et al1988) An apparent exite occurs on appendages asso-ciated with Chasmataspis (Chasmataspidida) (Dunlopet al 2004)57 Insertion process of anteromedial tergo-coxal mus-cle 0 weakly developed 1 well to extremely well devel-oped [GEWB 238]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Boumlrner 1904 Shultz 1999)58 Musculi laterales enlarged lateral tergocoxalmuscle with attachment shifted from coxa to adjacentpleural membrane 0 absent 1 present [S 22 WH 65GEWB 242]State 1 occurs in Araneae and Thelyphonida (Shultz1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI TROCHANTER AND COXAndashTROCHANTER JOINT

59 Coxandashtrochanter joint with complex posteriorarticulation composed of two articulating sclerites 0absent 1 present [S 24 WH 67 GEWB 85]State 1 occurs in Araneae Amblypygi Thelyphonidaand Schizomida (Shultz 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI FEMUR AND TROCHANTERndashFEMUR JOINT

60 Depressor muscle (or homologue) of trochanter-femur joint 0 absent 1 present [WH 42 GEWB 220]State 0 appears to be a unique synapomorphy of Ara-neae (Liphistius Shultz 1989 Aphonopelma Ruh-land amp Rathmayer 1978 Hypochilus present studyother examples Clarke 1984 1986)61 Trochanterndashfemur joint with dorsal hinge or pivotoperated by flexor muscles only 0 absent 1 present[WH sim44 GEWB sim91]State 1 occurs in Palpigradi Other chelicerates have abicondylar articulation and are typically equipped

with antagonistic muscles (Van der Hammen 1985Shultz 1989) but see 60 for an exception62 Superior trochanterndashfemur muscle (or homologue)originating broadly in femur inserting on distal mar-gin of trochanter 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 19892001)63 Basifemurndashtelofemur joint of appendages III andIV (= arachnid legs 1 and 2) in adult 0 absent 1present [S 25 WH 68 GEWB 86]The basifemur and telofemur apparently develop aslsquosisterrsquo articles through division of a parental articlethe femur (eg opilioacariform endeostigmatidpalaeosomatid mites) (Coineau amp Van der Hammen1979 Evans 1992) A single animal can have all legswith divided or undivided femora or have some com-bination of the two typically with the more posteriorlegs having the divided femora Thus anterior legsoften tend to be paedomorphic with respect toposterior legs Most chelicerates have undivided fem-ora on some or all legs but muscles homologous withthose of the basifemurndashtelofemur joint may stilldevelop (66) (Shultz 1989) suggesting that the devel-opment of cuticular and muscular components of ajoint have a degree of developmental and evolutionaryindependence

Undivided femora occur in most extant euchelicer-ates including Xiphosura Palpigradi AraneaeAmblypygi Thelyphonida Schizomida ScorpionesOpiliones Pseudoscorpiones higher Sarcoptiformes(Archegozetes Van der Hammen 1989) many Prostig-mata (Tetranychus Evans 1992) and throughout Par-asitiformes Divided femora occur in all legs in manyearly divergent Acariformes including many Endeo-stigmata basally divergent Sarcoptiformes (egPalaeacarus) and Prostigmata (eg AllothrombiumMicrocaeculus) (Van der Hammen 1989 Evans1992) Only leg 4 has basi- and telofemora in Alycus(Van der Hammen 1989) Basi- and telofemora occuron legs 3 and 4 in Solifugae Ricinulei Opilioacari-formes and Eurypterida (Clarke amp Ruedemann 1912Selden 1981 Shultz 1989)

Giribet et al (2002) coded the lsquoannulusrsquo of certainTrigonotarbida (Shear et al 1987) as a basifemur buta similar non-muscularized annulus is present in theposterior legs of extant Thelyphonida In both casesthe condyles of the trochanterndashannulus and annulusndashfemur joints are aligned in such away that they per-form essentially identical movements and appear notto enhance either the range or degree of leg move-ment [This contrasts with the condition in the lasttwo pairs of legs in Baltoeurypterus where condyles ofthe apparent trochanter-basifemur and basifemurndashtelofemur joints have different arrangements (Selden1981)] The annulus in Thelyphonida appears to func-tion in maintaining constant volume of the tro-

252 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

chanterndashfemur joint by preventing ballooning of thearthrodial membrane thereby minimizing the effect ofinternal fluid pressure on movement64 Basifemurndashtelofemur joint of appendage V(= arachnid leg 3) in adult 0 absent 1 present See 6365 Basifemurndashtelofemur joint of appendage VI(= arachnid leg 4) in adult 0 absent 1 present See6366 Cuticular differentiation of basifemur-telofemurjoint absent but muscles present 0 absent 1 present- inapplicable due to presence of joint (63ndash65)State 1 occurs in appendages of postoral somites IIIndashIV (= arachnid legs 1ndash4) in Xiphosura AraneaeAmblypygi Thelyphonida Schizomida Scorpionesand Pseudoscorpiones (Shultz 1989)67 Circumfemoral ring 0 absent 1 presentThe femora of one or more pairs of legs in anactinot-richid Acari have a basal groove or ring associatedwith slit sensilla Acarologists term that portionproximal to the ring the lsquobasifemurrsquo and that portiondistal to the ring the lsquotelofemurrsquo and they refer tothe basifemur as the lsquosecond trochanterrsquo and thetelofemur as the lsquofemurrsquo Despite the confusingterminology the basifemur discussed in 63ndash65 is nothomologous with the lsquobasifemurrsquo of opilioacariformand parasitiform Acari as both a true basifemurndashtelofemur joint and a circumfemoral ring are presentin legs 3 and 4 of Opilioacariformes (Van der Ham-men 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI PATELLA AND FEMURndashPATELLA JOINT

68 Patella of appendage of postoral somite III(= arachnid leg 1) proportionally much longer thanthose of more posterior appendages 0 absent 1presentState 1 occurs in Thelyphonida Schizomida (Hansenamp Soslashrensen 1905 Shultz 1989) and Opilioacari-formes (Van der Hammen 1989)69 Femurndashpatella joint 0 monocondylar several axesof movement and multifunctional muscles 1 bicondy-lar hinge one axis of movement and antagonistic mus-cles 2 hinge one axis of movement flexor muscleswithout muscular antagonists [S sim26 WH sim69 GEWBsim87]State 0 occurs in Solifugae State 1 occurs in Phalang-ida (Opiliones) Scorpiones and PseudoscorpionesState 2 occurs in the remaining taxa (Shultz 1989)70 Patellar plagula 0 absent or with simple medianattachment 1 symmetrical Y-shaped with long prox-imal stem 2 symmetrical U-shaped (= arcuate scler-ite) 3 asymmetrical attaching to patella only atanterior margin [S 27 WH sim70 GEWB 234 = 243]State 1 occurs in extant Xiphosura (Shultz 1989)State 2 occurs in Araneae (Manton 1958 Ruhland amp

Rathmayer 1978 Clarke 1986 Shultz 1989) State 3occurs in Amblypygi Thelyphonida and Schizomida(Shultz 1989) Giribet et al (2002) included this char-acter twice

APPENDAGES OF POSTORAL SOMITES IIIndashVI TIBIA AND PATELLAndashTIBIA JOINT

71 Tibiae divided by one or more joints 0 absent 1presentState 1 occurs in Amblypygi (Weygoldt 2000)72 Patellandashtibia joint 0 monocondylar with or with-out CZY (73) 1 bicondylar hinge one axis of move-ment antagonistic muscles 2 hinge one axis ofmovement muscles without muscular antagonists [Ssim31 WH 74 GEWB sim92] (Shultz 1989 Selden Shearamp Bonamo 1991)73 Patellandashtibia joint with posterior compression zone(lsquoCZYrsquo) 0 absent 1 present - inapplicable due toabsence of monocondylar articulation [GEWB 99]State 1 is a unique synapomorphy of Araneae (Seldenet al 1991)74 Patellandashtibia joint of appendages IIIndashVI(= arachnid legs 1ndash4) with deep-set monocondylar pivotbordered by a pair of tibial processes to which extensormuscles attach 0 absent 1 present - inapplicabledue to absence of monocondylar joint (72)State 1 occurs in extant Xiphosura (Snodgrass 1952Shultz 1989) although a similar arrangement ispresent in the appendage III (= arachnid leg 1) in The-lyphonida (Shultz 1993)75 Patellandashtibia joint largely immobile specializedfor autotomy 0 absent 1 presentState 1 occurs throughout Amblypygi (Weygoldt 19842000)76 Anterior femurndashtibia or femoropatellandashtibia(transpatellar) muscle 0 absent 1 present [S sim30GEWB 222] (Shultz 1989)77 Proximal attachment of posterior femurndashtibia orfemoropatellarndashtibia (transpatellar) muscle 0 muscleabsent 1 dorsoposterior surface of femur andor pos-terior surface of patella 2 distal process of femur mus-cle acting as extensor of femur-patella joint [S sim28 WHsim71 + 72 GEWB 235 = 244 245] (Shultz 1989)78 Distal attachment of posterior femurndashtibia (trans-patellar) muscle 0 posterior 1 dorsal acting as exten-sor of patellandashtibia joint - inapplicable due to absenceof muscle (77) [S 29 GEWB 221] (Shultz 1989)79 Anterior patellandashtibia muscle 0 absent 1 present[S sim32 WH sim73 GEWB 246] (Shultz 1989)80 Posterior patellandashtibia muscle 0 absent 1 present[S 33 WH 76 GEWB 248] (Shultz 1989)81 Patellandashtibia joint spanned by elastic (lsquospringlikersquo)sclerite 0 absent 1 present [GEWB 107]State 1 is a unique synapomorphy of Solifugae (Sens-enig amp Shultz 2003)

PHYLOGENY OF THE ARACHNID ORDERS 253

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

APPENDAGES OF POSTORAL SOMITES IIIndashVI TARSUS AND TIBIAndashTARSUS JOINT

82 Tarsus divided into proximal basitarsus(= metatarsus) and distal telotarsus (= distitarsus orlsquotarsusrsquo) 0 absent 1 presentState 0 occurs in extant Xiphosura (Shultz 1989) andthroughout Acariformes (Lindquist 1984 Evans1992) in the anterior two pairs of legs in chthonioidpseudoscorpions and all legs in lsquomonosphyronidrsquo pseu-doscorpions including Feaelloidea and Cheliferoidea(Chamberlin 1931) A circumtarsal ring (83) associ-ated with slit sensilla in Parasitiformes (Acari) mayrepresent a joint between a telotarsus and basitarsus(Evans 1992)83 Circumtarsal ring 0 absent 1 presentState 1 is an apparent synapomorphy of Parasiti-formes (Evans 1992)84 Telotarsus in adult with two or more tarsomeres 0absent 1 present - inapplicable due to absence of telo-tarsus (82) [GEWB 71]State 1 occurs in Schizomida Thelyphonida Ambly-pygi Palpigradi Ricinulei Solifugae (Shultz 1989)Opilioacariformes (Van der Hammen 1989) and Ple-siosiro (Dunlop 1999)85 Three telotarsomeres on appendages of postoralsomites IV-VI (= arachnid legs 2ndash4) 0 absent 1present - inapplicable due to absence of telotarsus (82)or absence of telotarsomeres (84) [S 34 WH 77 GEWB102]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Shultz 1989)86 Tibiandashtarsus joint spanned by well-developed elas-tic (lsquospringlikersquo) sclerite 0 absent 1 present [GEWB107]State 1 occurs in Scorpiones phalangid Opiliones andSolifugae (Alexander 1967 Shultz 2000 Sensenig ampShultz 2003) Extant Xiphosura is coded for the lastprosomal appendage only (see 88)87 Appendage of postoral somite VI (= arachnid leg 4)with ring of large basally articulated spatulate pro-cesses at tibia-tarsus joint 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)88 Appendages postoral somites IIIndashV (= arachnidlegs 1ndash3) with tibiotarsus (ie tibia and tarsus not dif-ferentiated) 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 1989) butnot apparently in Weinbergina (Moore et al 2005) theonly synxiphosurid in which appendages are suffi-ciently preserved to determine the number ofpodomeres89 Ambulacrum peduncle-like extension of the tarsuswith internal condylophores and terminating distallywith apotele (eg claws) andor pulvillus 0 absent 1present

State 1 is present throughout Acari (Alberti amp Coons1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI APOTELE AND TARSUSndashAPOTELE JOINT

90 Apotele of appendage III (= arachnid leg 1) 0absent or not apparent 1 present [GEWB 101]State 0 occurs in Amblypygi Thelyphonida and Schi-zomida (Shultz 1999) However the apotele is proba-bly present but undifferentiated because muscles thatwould normally insert on the apotele terminate on theend of the lsquotarsusrsquo Dunlop (1999) reconstructed Hap-topoda as lacking an apotelic claw on leg 1 but noted inthe text that this was speculative Dunlop (2002b) hasreviewed apotelic diversity in Chelicerata91 Appendages of postoral somites IIIndashVI (= arachnidlegs) chelate with chela formed from tibiotarsus andapotele or tarsus and apotele 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988) The condition appears to occur in Chasmataspisbased on associated appendages (Dunlop et al 2004)but not in Diploaspis and Octoberaspis (Dunlop et al2001 Dunlop 2002a)92 Apotele with eversible or padlike empodium(= pulvillus) in adult 0 absent although empodialclaw may be present 1 present [S sim38 WH 81 GEWB95]State 1 occurs throughout Pseudoscorpiones (Cham-berlin 1931) Solifugae (Roewer 1934) Opilioacari-formes Holothyrida Ixodida and free-livingMesostigmata (Evans 1992) An eversible lsquopulvillusrsquo isprobably a primitive feature of Amblypygi (eg Char-inus Weygoldt 2000) but is absent in higher groups(eg Phrynus Weygoldt 2000)93 Inferior apotele muscle (= claw depressor) with tib-ial attachment 0 absent 1 present [S 35 WH 78GEWB 249]The character is coded for extant Xiphosura usingappendage of postoral somite VI (= arachnid leg 4) asother legs lack a differentiated tibia (88) State 1appears to occur throughout Arachnida (Shultz 1989)94 Inferior apotele muscle (= claw depressor) withpatellar attachment 0 absent 1 present [S 36 WH 79GEWB 250]State 0 occurs in extant Xiphosura Acari AmblypygiPalpigradi and Ricinulei State 1 occurs in all otherarachnids examined here (Shultz 1989)

OPISTHOSOMA

95 Number of opisthosomal somites in adult 0 five1 eight 2 nine 3 10 4 11 5 12 6 13 [WH sim33GEWB sim190]Chasmataspidids appear to have 13 opisthosomalsomites (Dunlop et al 2004) The synziphosurans

254 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Weinbergina (Moore et al 2005) and Limuloides(Anderson amp Selden 1997) appear to have ten opistho-somal somites Moore et al (2005) found no evidence ofan anterior microtergite in Weinbergina as proposedby Anderson amp Selden (1997) Xiphosurids appear tohave nine opisthosomal somites (Scholl 1977 Ander-son amp Selden 1997 Shultz 2001) Members of the fol-lowing orders have 12 opisthosomal somites Araneae(Millot 1949ac) Amblypygi (Weygoldt 2000) Thely-phonida and Schizomida (Hansen amp Soslashrensen 1905)and Pseudoscorpiones (Chamberlin 1931) Elevensomites are present in Palpigradi Solifugae (Roewer1934) and Trigonotarbida (Gilboarachne Shear et al1987 Palaeocharinus Fayers et al 2004) Plesiosiromay have 11 or 12 somites (Dunlop 1999) Justifica-tion for coding of more controversial taxa is providedbelow

Scorpiones Two principal hypotheses regarding thenumber of opisthosomal somites in scorpions havebeen advocated a 13-somite hypothesis derived fromembryological studies (Brauer 1895 Patten 1912Farley 1999 2005) and a 12-somite hypothesis basedon comparative anatomy of adults (Weygoldt amp Pau-lus 1979) The embryological interpretation is basedon the observation of pregenital genital and pectinalsomites (each with segmental ganglia and paired limbbuds) in early scorpion embryos followed by extremereduction or loss of the pregenital somite in laterembryos According to this view a missing pregenitalsomite should be added to the 12 apparent opisthoso-mal somites of post-embryonic scorpions to achieve afinal number of 13 The anatomy-based hypothesiswas introduced by Weygoldt amp Paulus (1979) whoadvocated a literal interpretation of post-embryonicsegmentation based on opisthosomal tergites Specifi-cally these authors argued that the last pair of dorsalendosternal suspensor muscles of non-scorpionarachnids especially Pedipalpi attach to the first(= pregenital) somite that this condition also occurs inscorpions and that there is no reason to invoke a miss-ing pregenital tergite They proposed that the pectinesbelong to the genital somite and that functional spe-cializations of the nervous system for pectinal functiongive the apparence of an extra neuromere duringembryonic development

I recently dissected the prosoma and anterioropisthosoma of the scorpions Centruroides Hadrurusand Heterometrus and focused on the composition ofthe muscular diaphragm (103) that separates thehaemocoelic compartments of the prosoma andopisthosoma (Shultz 2007) The diaphragm is com-posed of a metameric series of axial muscles fromthree somites the anterior somite corresponds to thelast prosomal somite and the posterior somite corre-sponds to the genital somite The middle elements

insert dorsally along a tranverse tendon attached tothe anterior margin of the first tergite These obser-vations are consistent with the embryological inter-pretation that a pregenital somite is present but itstergite is not expressed It appears likely that the thepregenital somite was compressed longitudinally dur-ing the evolution of the diaphragm I code scorpions ashaving 13 opisthosomal somites

Eurypterida The eurypterid opisthosoma is widelyassumed to have 12 somites (Clarke amp Ruedemann1912 Stoslashrmer 1944) However in a speculative paperon the evolutionary morphology of trilobites and che-licerates Raw (1957) proposed that both scorpions andeurypterids have 15 opisthosomal somites Rawassumed that scorpions have 13 apparent opisthoso-mal somites based on the transient pregenital somiteof scorpion embryology that scorpions and eurypteridsare close relatives and should have the same numberof somites and that olenellid trilobites and chelicer-ates always have somites in multiples of three Rawachieved 15 somites in scorpions by assuming the lastopisthosomal somite to be a diplosomite and that thetelson is a postanal somite even though there is noevidence for either of these proposals He attributedthese features to eurypterids as well He also notedthat the connection between the prosomal carapaceand first opisthosomal tergite in Eurypterida differedstructurally from the connection between adjacentopisthosomal tergites and regarded this as evidencefor a reduced pregenital tergite in the prosomandashopisthosoma junction Rawrsquos speculations were largelyforgotten until Dunlop amp Webster (1999) resurrectedthe proposal that eurypterids have a reduced opistho-somal tergite and therefore share a unique similaritywith scorpions Unfortunately Dunlop amp Bullocktreated Rawrsquos conjecture as if it were based on empir-ical evidence rather than an attempt to force euryp-terid morphology into a peculiar numerical system Inthe absence of convincing evidence to the contrary Ihave coded eurypterids as having 12 opisthosomalsomites

Ricinulei The opisthosoma consists of a membranouspedicel (97) bearing the gonopore ventrally Thefemale gonopore is bordered by an anterior plate and aposterior plate The remainder of the opisthosoma iscomposed of thick sclerites separated by less heavilysclerotized cuticle The first dorsal sclerite is short andfunctions as part of a prosomandashopisthosoma couplingmechanism (96) that may or may not be a specializedcomponent of the following tergite This is followed byfour tergites and sternites and a three-segmentedmetasoma (= lsquopygidiumrsquo) Millot (1945 1949f) rea-soned that the opisthosoma contains ten somites withsomites VIIndashIX incorporated into the pedicel XndashXIIIexpressed as tergites and sternites and the metasoma

PHYLOGENY OF THE ARACHNID ORDERS 255

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

comprising four somites He did not regard the dorsalcoupling sclerite as a separate tergite and his inter-pretation of four rather than three metasomal somiteshas been rejected Pittard amp Mitchell (1972) also pro-posed ten opisthosomal somites but achieved thisnumber by regarding the dorsal coupling sclerite as atergite of somite IX and by recognizing three metaso-mal somites Legg (1976) adopted the system proposedby Pittard amp Mitchell but did not regard the dorsalcoupling sclerite as separate from the following tergite(X) Van der Hammen (1979 1989) reconstructed 13somites The pedicel was regarded as having twosomites (VII + VIII) the following four sets of tergitesand sternites were considered diplosomites (IX + XXI + XII XIII + XIV XV + XVI) and the metasoma hadthree somites (XVIIndashXIX) This interpretation mayhave been biased by Van der Hammenrsquos attempt tounite Ricinulei with anactinotrichid Acari which healso regarded as having a primitive number of 13somites Selden (1992) described a fossil ricinuleidTerpsicroton that shows two pairs of depressions onthe three large premetasomal tergites which con-trasts with the single pair seen in extant species Thisobservation appears to corroborate the diplosomitehypothesis but the evidence does not indicate that thetergite anterior to these is a diplosomite

Dunlop (1996) attempted to homologize theprosomandashopisthosoma coupling mechanisms ofRicinulei and Trigonotarbida a goal that required anovel and rather forced interpretation of the dorsalsclerites He regarded the coupling sclerite as homol-ogous with the first opisthosomal tergite (VII) of trig-onotarbids and then followed Van der Hammenrsquosdiplosomite hypothesis to achieve 12 somites in totalDunloprsquos scheme differs from previous systems in sug-gesting that the pedicel does not contain the dorsalelements of the first and second somites and is in-consistent with Millotrsquos (1945) observation that thepre- and postgenital plates each have dorsoventralmuscles Here I code Ricinulei as having 12 somitesThere are three metasomal somites (XVIndashXVIII)three diplosomites (= six somites) (XndashXV) one cou-pling somite (IX) and two somites in the pedicel (iethe genital and pregenital somites) (VII VIII)

Opiliones Harvestmen have nine opisthosomalsomites and an anal operculum that is traditionallyregarded as the tergite of a tenth somite a tenth ster-nite is lacking (Hansen amp Soslashrensen 1904 Winkler1957) However the anal operculum appears to repre-sent a persistent embryonic telson (Moritz 1957) andis therefore likely to be a postsegmental structurecomparable with the stinger of scorpions flagella ofthelyphonids etc (Shultz 2000)

Acari The number of opisthosomal somites is probe-matic for most mite taxa due to uncertainty about the

location of the prosomandashopisthosoma boundry (5)extensive simplification or loss of metamericallyarranged sclerites and muscle attachments a paucityof developmental studies of engrailed expression andin Acariformes opisthosomal anamorphosis and het-erochronic modification of somite number (Evans1992) Ixodids appear to be the exception developmen-tal studies indicate five somites in the opisthosoma ofticks (Evans 1992)

Some mites retain external evidence of segmenta-tion ndash metameric patterns of furrows muscle attach-ments and slit sensilla ndash and with certainassumptions the number of opisthosomal somites canbe estimated Here it is assumed that the dorsal sur-faces of the last two prosomal somites are present inmites and retain their primitive association with legs3 and 4 (contra Van der Hammen 1989) (see 5 for jus-tification) Given this there appear to be 11 somites inthe opisthosoma of Opilioacariformes a conclusionalso reached by other workers (eg With 1904 Sitni-kova 1978 Klompen 2000) Similar reasoning sug-gests that Alycus has seven opisthosomal somites notnine as advocated by Van der Hammen (1989) Unfor-tunately external evidence is ambiguous for determin-ing the number of somites in other mite lineages andthese are coded as uncertain96 Prosoma-opisthosoma coupling mechanism 0absent 1 present [GEWB sim24]The posterior margin of the carapace and the anteriormargin of the first apparent opisthosomal tergite arespecialized as a coupling mechanism in Ricinulei andTrigonotarbida and Dunlop amp Horrocks (1996) pro-posed that these are synapomorphic for the twogroups However there is uncertainty about thehomology of the anterior opisthosomal somites in Trig-onotarbida and Ricinulei (95)97 Pedicel 0 absent 1 aranean type 2 ricinuleidtype [WP 30 S sim40 WH sim20 GEWB sim126]The body narrows at or near the prosomandashopisthosoma juncture in several arachnid lineages (ieSolifugae Palpigradi Amblypygi Araneae Ricinulei)and this lsquowaistrsquo has often been used as character at theinterordinal level (eg Pocock 1893) However thisinterpretation is rejected here for being subjective andfor uniting non-homologous conditions For exampleAraneae and Amblypygi are often grouped on the basisof a lsquopedicelrsquo yet it is a highly specialized structure inAraneae and its parts are not readily homologizedwith those of Amblypygi In contrast the condition inAmblypygi is a slightly narrower version of the highlymoveable prosomandashopisthosoma juncture in Uropygiwhich is not generally considered a lsquopedicelrsquo Thepedicel in Ricinulei is also unique a weakly sclero-tized stalk containing the genital opening (Cryptocel-lus Pittard amp Mitchell 1972 Ricinoides Legg 1976)See 95

256 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

98 Opisthosoma with three- segmented lsquobucklerrsquo 0absent 1 presentState 1 occurs throughout Chasmatapsidida (Dunlop2002a)99 Thoracetron consolidation of tergites of postgeni-tal somites 0 absent 1 present - inapplicable codedonly for Xiphosura (Anderson amp Selden 1997)100 Fusion of tergites of postoral somites VIII and IX(= opisthosomal somites 2 and 3) only 0 absent 1present [GEWB 146]State 1 appears to be a synapomorphy Trigonotarbida(Shear et al 1987) Ricinuleids also have diplotergites(101) but these appear to encompass a different com-bination of somites (see 95)101 Three diplotergites 0 absent presentState 1 is a unique synapomorphy of Ricinulei (Selden1992) See 95102 Paired opisthosomal defensive glands openingvia ducts on either side of the anus 0 absent 1present [S 46 WH 86 GEWB 122]State 1 occurs in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905)103 Muscular diaphragm separating prosomal andopisthosomal compartments formed by dorsoventralmuscles of postoral somites VIndashVIII and extrinsic mus-cles of leg 4 0 absent 1 presentState 1 is known only in extant scorpions (Lankesteret al 1885 Firstman 1973) The diaphragm is oftentreated as a single structure but recent anatomicalwork (Shultz 2007) has shown that it is a composite ofdorsal endosternal suspensors and extrinsic leg mus-cles (see 95) A diaphragm is also present in Solifugaebut it is located more posteriorly and does not appearto be homologous with that of scorpions (Roewer1934)104 Opisthosomal appendicular chondrites 0absent 1 presentThese cartilage-like columns of mesodermally derivedtissue are associated with each opisthosomal append-age including the chilaria in extant Xiphosura(Patten amp Hazen 1900 Yamasaki et al 1988 Fahr-enbach 1999 Shultz 2001)105 Paired appendages on ventral surface of postoralsomite VII (= opisthosomal somite 1) in adult 0absent 1 present [WP sim13 S 39 WH 82 GEWB 143]State 1 is known in Weinbergina (Moore et al 2005)and extant Xiphosura The bilobed structure of themetastoma in certain eurypterids as well as the cor-responding placement of chilaria in Xiphosura haveled some workers to regard the metastoma as fusedappendages of postoral somite VII (eg Stoslashrmer1955) This hypothesis is problematic given that (i) nospecial explanation is needed to account for bilaterallysymmetrical structures in bilaterally symmetricalorganisms (ii) incorporation of the ventral part of thefirst opisthosomal sternite as a functional element of

the prosoma is typical of arachnids and perhapsEuchelicerata generally and (iii) some workers regardthe metastoma as a sternite (eg Jeram 1998) Thusthe metastoma is coded here as uncertain Dunlop hasproposed that the sternum of scorpions is derivedfrom appendages citing the presence of transient limbbuds in scorpion development (Brauer 1895 Patten1912) but this is not compelling evidence for a persis-tent appendicular contribution to the sternite of theadult106 Megoperculum 0 absent 1 present [S sim41 + 42WH sim83 GEWB 162 + 163]State 1 occurs in Thelyphonida Schizomida Ambly-pygi Araneae (Shultz 1993 1999) Trigonotarbida(Shear et al 1987) and probably Haptopoda (Dunlop1999) and Palpigradi The megoperculum consists ofappendages of opisthosomal somite 2 (= genital somiteor postoral somite VIII) (Shultz 1993 1999 Popadicet al 1998) that have fused medially and displacedthe ventral body wall of the somite anteriorly (oftenrepresented by a small sternite to which dorsoventralmuscles of postoral somite VIII attach) The megoper-culum projects posteriorly to form the ventral surfaceof a pregenital chamber The dorsal surface of thechamber is formed by the ventral body wall of opistho-somal somite 3 (= postoral somite IX) which serves asthe ventral attachment of the paired dorsoventralmuscles of that somite The megoperculum bears book-lungs (where present) and sometimes paired gonopodsthat are probably derived from telopodites A similarbut less developed operculum is present on theopisthosomal somite 3 which may also bear booklungsand eversible vesicles (eg Amblypygi) (see 111) corre-sponding to the booklungs and gonopods of the mego-perculum (Shultz 1999)

I contend here that a megoperculum is present inPalpigradi and that this is particularly evident infemales Specifically a large unpaired lobe projectsposteriorly from the genital somite over the ventroan-terior surface of the first postgenital somite therebyforming a pregenital chamber The dorsoventral mus-cles of the genital somite attach ventrally near theanterior border of postoral somite VIII and those ofpostoral somite IX attach to the upper surface of thepregenital chamber (Rucker 1901 Boumlrner 1902bRoewer 1934 Millot 1949d Van der Hammen 1989)Weygoldt (1998) has questioned this interpretation ofPalpigradi107 Postgenital operculum or lsquosternitersquo 0 sclerotized1 not sclerotized - inapplicable coded only for Pan-tetrapulmonata (Platnick amp Gertsch 1976)State 1 is present in Trigonotarbida Amblypygi The-lyphonida Schizomida and Mesothelae (Araneae) butnot Opisthothelae (Araneae)108 Genital opening of female guarded by four plates(one pregenital one postgenital two laterogenitals)

PHYLOGENY OF THE ARACHNID ORDERS 257

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

genital opening of male guarded by two plates 0absent 1 presentThis an apparent synapomorphy of Holothyrida(Acari) (Van der Hammen 1989)109 Anterior margin of genital opening in male withglands secreting via fusules 0 absent 1 presentState 1 occurs in Araneae as epigastric or lsquoepiandrousrsquoglands (Marples 1967) and in Palpigradi (Condeacute1991b)110 Paired valve-like plates apparently formed fromcomponents of three somites covering triradiate genitalopening 0 absent 1 presentThis is an apparent synapomorphy of Acariformes(Evans 1992 Alberti amp Coons 1999)111 Eversible lsquoappendagesrsquo on the ventral surface ofpostgenital somites 0 absent 1 present [S sim43 WHsim84 GEWB sim127]Paired ventral lsquoappendagesrsquo operated in part byhaemolymph pressure occur in the form of ventralsacs in certain Amblypygi (postoral somite IX) (Wey-goldt 2000) (Charinus Millot 1949b not PhrynusShultz 1999) prokoenenian Palpigradi (postoralsomites XndashXII) (Rucker 1901 Condeacute 1991a) as spin-nerets in Araneae (XndashXI) (Shultz 1987) and as geni-tal papillae in many Acariformes (Alberti amp Coons1999) including Oribatida many Endeostigmata (Aly-cus Van der Hammen 1989) and Prostigmata (Allo-thrombium Saboori amp Kamali 2000 MicrocaeculusEvans 1992) Eversible vesicles similar to those ofamblypygids occur in fossils of the trigonotarbidPalaeocharinus (postoral somite IX) (Fayers et al2004) Two cuticular structures (lsquogenital verrucaersquo)occur anterior to the genital opening in Opilioacari-formes and each covers the opening to a thin-walledinvaginated sac (lsquogenital papillarsquo) of unknown function(Van der Hammen 1989) Van der Hammen proposedthat it is homologous to a genital papilla of Acari-formes but Alberti amp Coons (1999) have questionedthis interpretation on several grounds and refer tothese structures as pregenital capsules112 Opisthosomal silk glands and spinnerets derivedfrom appendages on postoral somites X and XI(= opisthosomal somites 4 and 5) 0 absent 1 present[WP 33 WH 41 GEWB 123 + 142]State 1 is a unique synapomorphy of Araneae (Plat-nick amp Gertsch 1976) although potentially homolo-gous glands and spigot-like setae occur on the ventralopisthosomal surface in Palpigradi (Millot 1943Condeacute 1991a)113 Opisthsosomal spinnerets location 0 near mid-dle of opisthosoma 1 near posterior end of opisthso-som ndash inapplicable coded only for Araneae (Platnickamp Gertsch 1976)114 Anterior medial lsquospinneretsrsquo 0 absent 1 present- inapplicable coded only for Araneae (Platnick ampGertsch 1976)

115 Opisthosomal tergites divided longitudinally intoone median and two lateral plates 0 absent 1 present[GEWB 145]State 1 occurs in non-curculioid Ricinulei (Selden1992) and many Trigonotarbida (Dunlop 1996) Thisfeature or something very like it (eg distinct opistho-somal trilobation) occurs sporadically in severaleuchelicerate groups including Chasmataspis (Chas-mataspidida) (Dunlop et al 2004) most fossil xipho-surans (Anderson amp Selden 1997) and a few derivedEurypterida (especially Mixopteroidea) (Tollerton1989)116 Number of metasomal somites 0 zero 1 two 2three 3 five 4 nine [S sim44 WH sim85 GEWB 128 144]The metasoma is a preanal region comprising multiplesomites that are substantially narrower than the pre-ceding somites or mesosoma Metasomal somitesoften lack pleural membranes and take the form ofsclerotized rings State 1 occurs in TrigonotarbidaState 2 is present in synziphosurid Xiphosura (Ander-son amp Selden 1997) Amblypyi Thelyphonida Schizo-mida and Ricinulei State 3 occurs in Scorpiones andEurypterida and State 4 occurs in Chasmatispidida

Some workers appear to consider a five-segmentedmetasoma as conclusive evidence for the monophylyof a Eurypterida + Scorpiones clade (eg Dunlop ampBraddy 2001) presumably because the character rep-resents a kind of tagmosis and should therefore begiven substantial phylogenetic weight However Wey-goldt (1998) has pointed out that a metasoma contain-ing three somites is probably part of the ground planof Euchelicerata The known diversity of the eucheli-cerate metasoma indicates that this feature can inincrease or decrease its segmental composition in evo-lution or be eliminated entirely without necessarilychanging the total number of opisthosomal somitesConsequently it is problematic to assume that thefive-segmented metasoma has any special immunityto homoplasy117 Postanal structure (telson) 0 absent or not obvi-ously developed 1 presentPostanal structures are considered non-segmentalbecause they are modifications of the embryonicregion posterior to the site of somite addition (embry-onic growth zone) A well-developed postanal struc-ture persists in adults in Xiphosura EurypteridaChasmataspidida Scorpiones Schizomida Thely-phonida and Palpigradi The anal operculum of Opil-iones is generally considered by morphologists to bethe tergite of the tenth opisthosomal or anal somite(eg Hansen amp Soslashrensen 1904) but embryological evi-dence (Moritz 1957) suggests that it is derived fromthe postproliferative zone and thus corresponds to thetelson of other chelicerates (see 95)118 Postanal structure shape 0 caudal spine 1aculeus 2 flagellum 3 anal operculum - inapplica-

258 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ble due to absence of postanal structure (117) [WP sim41S sim45 WH sim24 GEWB sim121 129 + 147]State 0 is present in Xiphosura Eurypterida andChasmataspidida State 1 occurs in Scorpiones State2 occurs in Palpigradi Thelyphonida and SchizomidaState 3 occurs in Opiliones (see 95 117)119 Specialized postanal flagellum in male (see 157ndash159) 0 absent 1 present - inapplicable coded onlyfor taxa with postanal flagellum (118) [GEWB 131]State 1 is a unique synapomorphy of Schizomida(Hansen amp Soslashrensen 1905 Cokendolpher amp Reddell1992 Reddell amp Cokendolpher 1995)

RESPIRATORY SYSTEM

120 Respiratory medium 0 water 1 airAmong the terminal taxa included only fossil scorpi-ons are problematic for this character They are codedhere as uncertain Based on a study of book lungmicrosculpture Scholtz amp Kamenz (2006) have arguedthat arachnids are primitively terrestrial and pulmo-nate (see also Firstman 1973) and have questionedwhether any fossil scorpions were aquatic No positionon this proposal is taken here121 Respiratory lamellae on opisthosomal somite 2(= genital somite postoral somite VIII) 0 absent 1present [WP sim37 S sim51 WH sim22 GWEB sim133ndash137]This character encompasses book gills (= lamellae thatfunction in water) and book lungs (= lamellae thatfunction in air) The distinction between book gills andbook lungs is accommodated here by 120 in combina-tion with 121ndash124 Respiratory lamellae are presenton the genital somite in Trigonotarbida AraneaeAmblypygi Thelyphonida and Schizomida Petrunk-evitch (1949) reconstructed Plesiosiro as having booklungs but Dunlop (1999) could not corroborate thisThis character was used by Dunlop amp Webster (1999)to propose that Xiphosura and Scorpiones are closelyrelated because they both lack respiratory lamellae onthe genital somite Dunlop amp Braddy (2001) alsoargued for the placement of Eurypterida with Xipho-sura and Scorpiones based in part on this characterAt least some eurypterids may have had respiratorylamellae (eg Manning amp Dunlop 1995) but the onlyevidence of their segmental distribution is derivedfrom one specimen (Braddy et al 1999)122 Respiratory lamellae on opisthosomal somite 3(= postoral somite IX) 0 absent 1 present [GEWB136]State 1 occurs in Amblypygi Thelyphonida and Xipho-sura It is a groundplan feature of Araneae (Platnick ampGertsch 1976) and is present in all representativetaxa included here Braddy et al (1999) proposed theexistence of respiratory lamellae on postoral somitesIXndashXII in Eurypterida based on evidence from onespecimen This character is coded as uncertain for the

Eurypterida included here Extant scorpions are codedhere as lacking respiratory lamellae on this somite(= pectinal somite) based on the interpretation of theopisthosomal segmentation discussed in 95123 Respiratory lamellae on opisthosomal somites 4ndash6 (= postoral somites XndashXII) 0 absent 1 presentState 1 is definitely present in extant Xiphosura andScorpiones and may have occurred in Eurypterida124 Respiratory lamellae on opisthosomal somite 7(= postoral somite XIII)State 1 occurs in extant Xiphosura and Scorpionesonly125 Kiemenplatten 0 absent 1 presentThese structures are located on the roof of opercularchambers in Eurypterida (Clarke amp Ruedemann1912) where they take the form of ventrally projectingcones with a distinct cuticular microsculputure (Man-ning amp Dunlop 1995) Dunlop amp Braddy (2001)inferred the existence of Kiemenplatten in all Palaeo-zoic scorpions based on a description and photos of onespecimen of Paraisobuthus duobicarinatus by Kjelles-vig-Waering (1986 pls 16ndash18) The plates depict darkcone-like rearward-pointing denticles distributedwithin a white amorphous material The denticlesappear to lack microsculpture even though the mag-nifications at which the photos were taken (times90ndashtimes330)are comparable with those illustrating the cones ofKiemenplatten (times170ndashtimes350 in Manning amp Dunlop1995 figs 1 2) Given the diversity of cuticular struc-tures present in the atria of booklungs book gills andtracheae the denticles appear to bear no special sim-ilarity to the cones of Kiemenplatten126 Tracheal system 0 absent 1 paired ventral stig-mata on postoral somite VIII (= opisthosomal somite2) 2 paired ventral stigmata on postoral somites IXand X 3 one pair of stigmata opening near legs 3 or 44 paired stigmata associated with chelicerae 5 fourpairs of stigmata on dorsal surface of opisthosoma -inapplicable aquatic (120) [WP 40 sim43 S sim52 + 5354 WH sim23 45 88 GEWB sim138 139]Firstman (1973) and Weygoldt amp Paulus (1979)hypothesized that tracheae are homologous in all tra-cheate arachnids except spiders and subsequentworkers have entertained this hypothesis by includinga character for the presenceabsence of tracheal sys-tems that ignores the diverse arrangement of stig-mata in arachnids (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) However thisapproach assumes that internal tracheal systems areconserved but that tracheal openings (stigmata)appear and disappear on different parts of the bodywith higher evolutionary frequency Here I assumethat stigmata are conserved in evolution and that dif-ferences in their anatomical placement reflect the evo-lution of new tracheae State 1 occurs throughoutOpiliones State 2 occurs throughout Solifugae

PHYLOGENY OF THE ARACHNID ORDERS 259

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

(Roewer 1934) and Pseudoscorpiones (Chamberlin1931) State 3 recognizes the possible homology of stig-mata in Ricinulei Opilioacariformes and Parasiti-formes (Acari) (Lindquist 1984 Van der Hammen1989) Tracheal systems are apparently absent in thegroundplan of Acariformes but Prostigmata typicallyhave tracheae associated with the chelicerae Tra-cheae derived from the posterior pair of book lungs arewidespread in araneomorph Araneae but the booklungs are retained and tracheae are absent in themost basally divergent groups (eg Hypochilus)

BOXndashTRUSS AXIAL MUSCLE SYSTEM (BTAMS)

127 Posterior oblique muscles of BTAMS of postoralsomites IndashVI 0 absent 1 present in one or moresomitesState 0 occurs in Xiphosura and state 1 occurs inPalpigradi Araneae Amblypygi Thelyphonida andSchizomida (Shultz 2001) The condition in AcariRicinulei Opiliones and Solifugae is not known128 Anterior oblique muscles of BTAMS posterior topostoral somite VI 0 absent 1 presentState 1 occurs in Xiphosura which is probably theprimitive condition based on comparison with otherarthropods (Shultz 2001) State 0 occurs in all arach-nids examined thus far129 Ventral attachments of posterior oblique musclesof opisthosomal BTAMS located in prosoma 0 absent1 presentState 1 occurs in extant Xiphosura (Shultz 2001)130 Endosternite fenestrate 0 absent 1 present [S 8WH 54 GEWB 35]State 1 occurs in Thelyphonida and hubbardiid Schi-zomida (Firstman 1973) The condition in protoschi-zomid Schizomida is not known131 Suboral suspensor a tendon that arises from theBTAMS and inserts on the ventral surface of the oralcavity via muscle 0 absent 1 presentState 1 occurs in Palpigradi (Eukoenenia Millot1943) Amblypygi (Shultz 1999) and Thelyphonida(Shultz 1993)132 Perineural vascular membrane in adult 0absent 1 present [WH 28 GEWB 155]Wheeler amp Hayashi (1998) coded extant Xiphosura asunknown although the primary source for this char-acter (Firstman 1973) stated that adult Limulus havea perineural vascular membrane Within Arachnidapresence of a perineural vascular membrane is appar-ently correlated with the presence of tracheae (126)133 Ventral endosternal suspensor attaching on coxaof anteriorly adjacent somite 0 absent 1 present[GEWB 239 253]State 1 occurs in Amblypygi (Shultz 1999) Thely-phonida (Shultz 1993) and Schizomida (Shultzunpubl observ)

134 Posteriormost postoral somite with a pair of dor-soventral muscles 0 VI 1 VII 2 VIII 3 XII 4 XIII5 XIV 6 XV 7 XVIDorsoventral muscles tend to run in a continuousmetameric series beginning in the prosoma Deter-mining the last tergite on which the muscle seriesends is substantially easier than counting especiallygiven frequent anatomical complexities near theprosomandashopisthosoma juncture The dorsoventralmuscle series ends on postoral somite VII inCyphophthalmi (eg Chileogovea) and Laniatores(Opiliones) (Shultz unpubl observ) on postoralsomite VII or VIII in Eupnoi (Leiobunum Shultz2000 unpubl observ) on postoral somite XII in Pal-pigradi (Roewer 1934) Solifugae (Bernard 1896Roewer 1934 Millot amp Vachon 1949) and Araneae(Liphistius Hypochilus Millot 1933 but segmenta-tion reinterpreted here following Shultz 1993 1999)on postoral somite XIII in hubbardiid Schizomida(Cokendolpher amp Reddell 1992) and Scorpiones(Lankester et al 1885) and on postoral somite XIVin extant Xiphosura (Shultz 2001) Thelyphonida(Shultz 1993) protoschizomid Schizomida (Cokendol-pher amp Reddell 1992) and Amblypygi (Shultz 1999)Dunlop (1999) reconstructed Plesiosiro as havingpaired tergal apodemes ending on postoral somite XVbut his figures indicate that the series ends on posto-ral somite XIV The interpretation of opisthosomalsegmentation in Ricinulei coded here (95) indicatesthat the series ends on postoral somite XV in Terpsi-croton and this is treated as the groundplan for theorder The posteriormost dorsoventral muscles inPseudoscorpiones occur on postoral somite XVI(Vachon 1949)

NERVOUS SYSTEM

135 Segmental ganglia 0 consolidated in prosoma1 one or more present in opisthosoma [WH 30 GEWB210 + 216]Adult neuromeres are exclusively prosomal in Ambly-pygi liphistiomorph and araneomorph Araneae Opil-iones Pseudoscorpiones Ricinulei and Acari (Millot1949a)136 Dorsal median eyes 0 absent 1 present [WP14 + 47 S 50 WH sim7 GEWB 1]Median eyes are a groundplan feature of Euchelicer-ata and are retained in extant Xiphosura Eurypteridaand occur in all arachnid orders except PalpigradiSchizomida Ricinulei and Pseudoscorpiones (Paulus1979) Median eyes are also absent in opilioacariformand parasitiform Acari (Lindquist 1984) and perhapsthe fossil whipscorpion Proschizomus (Dunlop amp Hor-rocks 19951996) They are unknown in synziphosu-ran Xiphosura Median eyes are present in a variety ofbasal acariform mites including Endeostigmata (but

260 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

not Alycus) Prostigmata (Microcaeculus) and certainoribatids (Palaeacarus) (Evans 1992 Alberti amp Coons1999) Giribet et al (2002) coded all representativeAcari as lacking median eyes It is not known whetherthe eyes of some cyphophthalmid opilions are medianor lateral evidence from tracheal branching inCyphophthalmus (Janczyk 1956) suggests that theyare median eyes (Shultz amp Pinto da Rocha 2007) andpresence of a tapetum in Stylocellus is consistent withlateral eyes (Shear 1993)137 Retinula cells of dorsal median eyes 0 organizedinto closed rhabdoms 1 organized into network ofrhabdomeres 2 disorganized - inapplicable due toabsence of median eyes (136) [GEWB sim3]State 0 is present in Scorpiones Thelyphonida andAmblypygi State 1 is present in Solifugae and Ara-neae state 2 is present in Xiphosura (Paulus 1979)The retinula cells of median eyes have been studied inseveral prostigmatid Acariformes and are organized ina network in some taxa and in an irregular pattern inothers (Alberti amp Coons 1999) Retinulae in phalangidOpiliones have state 0 proximally and state 2 distally(Schliwa 1979)138 Ventral median eyes 0 absent 1 presentState 1 occurs in early instars of extant Xiphosura(Paulus 1979)139 Lateral eyes 0 absent 1 presentLateral eyes are primitively present in Chelicerataand are absent in Palpigradi and Opiliones (Paulus1979) It is unclear whether the eyes of cyphoph-thalmid opilions are median or lateral (see 136)140 Arrangement and number of lateral eyes 0 com-pound many 1 five or more pairs (includes micro-lenses) 2 three primary pairs (excludes microlenses)3 two pairs 4 one pair - inapplicable due to absenceof lateral eyes (139) [WP sim13 + 18 + 38 + 44 + 52 Ssim49 WH sim10 GEWB sim4 + 5]True compound eyes are present in XiphosuraEurypterida Chasmataspidida and many PalaeozoicScorpiones Among the anactinotrichid Acari Opilioa-cariformes have two (Neocarus) or three (Siamacarus)pairs of lateral eyes (although at least one species ofSiamacarus lacks eyes) the allothryrid Holothyridaand many Ixodida have a single pair of lateral lensesand the remainder apparently lack eyes (Evans1992) Thelyphonida was coded as lsquo12rsquo to reflect fivepairs of lenses comprising three pairs of primarylenses and two pairs of small accessory lenses (notsimply three pairs as coded by Giribet et al 2002)Trigonotarbida is also coded as lsquo12rsquo to reflect threepairs of primary lenses and multiple small accessorylenses Dunlop (1999) illustrated Haptopoda as havingpaired triads of lateral eyes but noted that there wasactually no evidence of this in the fossils The lateraleyes of fossil ricinuleids have two pairs of lenses(Selden 1992) One pair of eyes or eyespots are

present in extant Ricinulei and many Schizomidaalthough five genera of hubbardiid Schizomida have apair of lenses (Reddell amp Cokendolpher 1995) State 1is the groundplan for extant Scorpiones and State 3occurs in the pseudoscorpions Chthonius FeaellaNeobisium (but State 4 in Chelifer) and Solifugae(Paulus 1979)141 Lateral eyes with closed rhabdoms 0 absent 1present - inapplicable due to absence of lateral eyes(139) [WP 21 WH 12 GEWB 6]Presence of closed rhabdoms is probably primitive forChelicerata and is retained in extant Xiphosura andScorpiones Retinula cells form a network of rhab-domeres in other extant chelicerates (Paulus 1979)but these networks can differ substantially in detailWeygoldt (1998) and other workers have given sub-stantial weight to the network character in unitingnon-scorpion arachnids but these authors seem not togrant comparable phylogenetic significance to theanalogous character of the median eyes shared bySolifugae and Araneae (137)142 Slit sensilla 0 absent 1 present [WP 19 S 47WH 11 GEWB 209]State 1 occurs in all arachnid orders except Palpigradi(Shultz 1990) The proposal that lsquoprimitiversquo slit sen-silla were present in Eurypterida (Dunlop amp Braddy1997) appears to be based on the over-interpretation ofa comparatively large notch that occurs in Baltoeu-rypterus at the terminus of the tibia (= podomere 7)(see also Edgecombe et al 2000 Giribet et al 2002)The neural construction of slit sensilla like function-ally similar campaniform sensilla of hexapods is sim-ilar to that of trichoid sensilla (Chapman 1998Klompen 2000) Thus these cuticular stress receptorsmay represent modified bases of sensory setae a viewsupported by the replacement of setae by slit sensilladuring post-embryonic development in some Opilioac-ariformes (Klompen 2000)143 Trichobothria 0 absent 1 present [GEWB 213]Trichobothria are present in extant Scorpiones(Jeram 1998) Pseudoscorpiones (Chamberlin 1931)some endeostigmatid (Alycus Van der Hammen 1989)and prostigmatid Acariformes and most Oribatida(Lindquist 1984) They occur on the ventral surface ofthe pedal femora in the opilioacariform Siamacarus(Leclerc 1989) They are also present in Araneae (Foe-lix 1996) Amblypygi (Weygoldt 2000) Schizomidaand Thelyphonida (Hansen amp Soslashrensen 1905) Theyare apparently absent in all non-arachnid chelicer-ates Solifugae Ricinulei Opiliones (Reissland ampGoumlrner 1985) and Parasitiformes (Acari) (Lindquist1984)144 Tibial trichobothria with 2-1-1-1 pattern onappendages IIIndashVI (= arachnid legs 1ndash4) 0 absent 1present - inapplicable due to absence of trichobothria(143) [S 48 WH 87 GEWB 88]

PHYLOGENY OF THE ARACHNID ORDERS 261

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs only in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905) Note that Shultz (1990)erroneously described this character as a 2-2-1-1 pat-tern and that this error was repeated by Giribet et al(2002) and described as 2-1-1 by Wheeler amp Hayashi(1998)145 Paired trichobothria on dorsal surface ofprosoma 0 absent 1 present - inapplicable due toabsence of trichobothria (143)State 1 is an apparent synapomorphy of Acariformesand is present in all acariforms included in this study(Alberti amp Coons 1999)146 Malleoli 0 absent 1 present [WP 45 GEWB 96]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)147 Pectines 0 absent 1 present [WP 20 WH 38GEWB 120]State 1 is a unique synapomorphy of Scorpiones148 Intercheliceral median organ 0 absent 1present [GEWB 212]State 1 is a unique synapomorphy of Palpigradi(Roewer 1934)149 Tarsal organ on appendage of postoral somite III(= arachnid leg 1) (= Hallerrsquos organ) sensilla containedwithin a cuticular depression on the superior surface ofthe tarsus of appendage III (= arachnid leg 1) 0absent 1 present [GEWB sim100]Klompen (2000) has noted that State 1 occurs in Opil-ioacariformes Parasitiformes (except Mesostigmata)and Ricinulei where it also occurs on leg 2 (Talaricoet al 2005) State 1 occurs on all legs in Araneae andperhaps Scorpiones (Foelix 1985)150 Tarsal organ on appendage of postoral somite IV(= arachnid leg 2) 0 absent 1 present (See 149)

REPRODUCTION

151 Gonads 0 primarily prosomal 1 primarilyopisthosomalState 0 is limited to Xiphosura (Sekiguchi 1988)State 1 occurs throughout Arachnida (Millot1949a)152 Ladder-like opisthosomal gonadsaccessoryglands (see 153) 0 absent 1 present [WH sim37 GEWBsim158]Giribet et al (2002) followed Wheeler amp Hayashi(1998) who followed Clarke (1979) in coding gonads asreticulate (Xiphosura) ladder-like (Scorpiones Thely-phonida Schizomida) or lsquosaccularrsquo (all remainingArachnida) However comparisons between Xipho-sura and Arachnida are problematic given that thexiphosuran gonads are primarily prosomal and thoseof arachnids are primarily opisthosomal (151) Fur-ther the reticulate pattern in Limulus (Xiphosura)but not other extant xiphosurans contains a distinctlyladder-like component The lsquosaccularrsquo state is probably

artificial as it encompasses a wide variety of pairedand unpaired structures153 Male gonads in two distinct parts one producingsperm and another (tubular gland) producing a holo-crine secretion similar to degenerate sperm 0 absent1 presentState 1 occurs in Thelyphonida Schizomida andAmblypygi although the holocrine material is pro-duced by ventral organs in Amblypygi and dorsalorgans in Thelyphonida and Schizomida (Alberti2005)154 Number of gonopores 0 two 1 oneExtant xiphosurans have two small genital openingson the base of the genital telopodite and all extantarachnids have a single opening The condition inEurypterida is not known Clarke amp Ruedemann(1912) located a pair of openings near the base of themedian organ (161) that are the outlets of the lsquohornorgansrsquo but it is unclear whether these are genitalducts or accessory structures Braddy amp Dunlop (1997)have developed numerous speculations about thesestructures and extended their arguments far beyondthe available evidence155 Genital opening (ie gonopore or gonostome)appearing to open in prosomal region (ie between legcoxae or anterior to posterior carapacal margin) 0absent 1 present [WP sim50 WH 26 GEWB 166]The genital opening in Euchelicerata is located onpostoral somite VIII (= opisthosomal somite 2) but ithas shifted anterior to the posterior margin of thecarapace or between the coxae of the last pair of legsin most Scorpiones (but not in PalaeoscorpiusKjellesvig-Waering 1986) and Opiliones The geni-tal opening occurs near or anterior to the last coxaein the opilioacariform and parasitiform Acari repre-sented here It is variable in Acariformes but islocated posterior to the coxae in all representativetaxa156 Ovipositor 0 absent 1 present [WP sim51 S sim60WH sim91 GEWB sim172]An ovipositor with a trilobed terminus is an apparentgroundplan character of Oribatida (Lindquist 1984Alberti amp Coons 1999) An ovipositor is also present inOpilioacariformes and Opiliones (Van der Hammen1989)157 Stalked spermatophore attached to substratum0 absent 1 present [S 57]State 1 occurs in Scorpiones PseudoscorpionesAmblypygi Thelyphonida and Schizomida The mech-anism of sperm transfer is unknown in PalpigradiOpilioacariformes and Holothryrida158 Male turns posterior end toward female duringmating behaviour 0 absent 1 presentState 1 occurs in Amblypygi (Weygoldt 2000) The-lyphonida and Schizomida (Weygoldt amp Paulus1979)

262 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

159 Female grasps male opisthosoma during matingbehaviour 0 absent 1 present [WP 29 S 58 WH 19GEWB 188]State 1 occurs in Thelyphonida and Schizomida (Wey-goldt amp Paulus 1979) Evolution of this behaviour wasprobably facilitated by 158160 Penis 0 absent 1 present [WP sim51 GEWB 167]State 1 is a unique synapomorphy of Opiliones A truepenis occurs in Phalangida A clearly homologousstructure is present in Cyphophthalmi (Opiliones) andapparently functions in depositing a spermatophore inthe femalersquos genital chamber (Karaman 2005) A truepenis may occur in some mites but it does not appearto be a groundplan feature of any major group (Evans1992) The lsquopenisrsquo in Oribatida is really a spermatop-ositor it functions in construction of a spermatophore(Alberti amp Coons 1999)161 Median organ 0 absent 1 presentState 1 occurs throughout Eurypterida (Clarke amp Rue-demann 1912) and has been observed in diploaspididChasmataspidida (Dunlop 2002a also Loganamaras-pis Tetlie amp Braddy 2004)

SPERM MORPHOLOGY

162 Nucleus with manchette of microtubules 0absent 1 present [S 54 WH 89 GEWB 193]State 1 occurs in Araneae Amblypygi Thely-phonida Schizomida Ricinulei and gonyleptid Lani-atores (Opiliones) (Alberti 1995 Giribet et al2002)163 Axoneme 0 absent 1 present [WP sim49 S sim55WH sim25 GEWB 195]State 1 occurs in Xiphosura Scorpiones Cyphoph-thalmi (Opiliones) Pseudoscorpiones RicinuleiAraneae Amblypygi Thelyphonida and Schizomida(Alberti 1995)164 Coiled axoneme 0 absent 1 present - inappli-cable due to absence of axoneme (163) [WP 22 S sim55WH 13 GEWB 196]State 1 occurs in Pseudoscorpiones Ricinulei Ara-neae Amblypygi Thelyphonida and Schizomida(Alberti 1995)165 Microtubule arrangement in axoneme 0 9 + 0 19 + 1 2 9 + 2 3 9 + 3 - inapplicable due to absence ofaxoneme (163) [WP sim26 S sim56 WH sim16 GEWB 198](Alberti 1995)166 Helical or corkscrew shaped nucleus 0 absent 1present [GEWB sim204]State 1 occurs in Araneae Amblypygi ThelyphonidaSchizomida and certain Scorpiones (eg Hadrurus)(Alberti 1995)167 Vacuolated-type sperm 0 absent 1 present[GEWB 205]State 1 is unique to Opilioacariformes and Parasiti-formes (Acari) (Alberti 1995)

168 Sperm aggregates 0 absent 1 present [GEWB206] (Alberti 1995)

DEVELOPMENT

169 Yolk in early embryo 0 concentrated (centroleci-thal or telolecithal) 1 evenly distributed (isolecithal)[WH sim34 GEWB sim191]Wheeler amp Hayashi (1998) and Giribet et al (2002)coded this character as lsquo0 isolecithal or telolecithal 1centrolecithalrsquo based on information presented byYoshikura (1975) Their coding appears to highlightseparate conditions found in Scorpiones (ie isoleci-thal and telolecithal) rather than a property intrinsicto the character itself170 Embryonic nutrition other than yolk 0 absent 1present [WP sim49]State 1 occurs in extant Scorpiones (except buthidsand chaerilids) and Pseudoscorpiones (Weygoldt1969)171 Embryological growth zone 0 initiating segmentaddition within prosoma 1 initiating segment addi-tion posterior to prosoma [GEWB sim192]Giribet et al (2002) followed Dunlop amp Webster (1999)in stating that only Xiphosura and Scorpiones havestate 0 among Chelicerata However the last prosomalsomite develops from the growth zone in Ixodida(Anderson 1973 Evans 1992) A similar processapparently occurs in the acariform mites Tyroglyphus(Sarcoptiformes Astigmata) (Yoshikura 1975) andArchegozetes (Sarcoptiformes Oribatida) (Thomas ampTelford 1999) but not apparently in Tetranychus(Prostigmata) (Dearden Donly amp Grbic 2002)172 Eggsembryos maintained in external attachedbrood sac secreted by genital glands 0 absent 1present [WP sim46 S 59 WH 90 GEWB 208 = 219]State 1 occurs throughout Amblypygi SchizomidaThelyphonida and Pseudoscorpions Giribet et al(2002) followed Shear et al (1987) in homologizingspinneret-derived silken egg sacs of spiders with gen-ital-gland-derived brood sacs of Pedipalpi However asthese structures are not homologous in secretory ori-gin construction or composition (Shultz 1987) theyare not regarded here as homologous Giribet et al(2002) miscoded the character as absent in Pseudo-scorpiones and without justification in PalpigradiFemale palpigrades have well-developed glands asso-ciated with postgenital somites and an array of setaefusules and other structures (112) (Boumlrner 1902b Mil-lot 1942 Condeacute 1991a) which are consistent withsome sort of brood care and Palpigradi should there-fore be coded as unknown for this character173 Embryonic and early postembryonic lsquolateralrsquo orClaparegravede organs associated with coxa of postoralsomite IV (= arachnid leg 2) 0 absent 1 present [WHsim31 GEWB sim211]

PHYLOGENY OF THE ARACHNID ORDERS 263

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bilaterally paired rounded protuberances locatedbetween coxae of appendages of postoral somites IIIand IV (arachnid legs 1 and 2) in prelarval and larvalinstars appear to be a primitive feature of Acariformes(Lindquist 1984 Evans 1992 Alberti amp Coons 1999)It is known to be derived embryologically from thecoxa of leg 2 in the oribatid Archegozetes (Thomas ampTelford 1999) Structures apparently homologouswith the acariform Claparegravede organ are present inembryonic Amblypygi (Weygoldt 2000) and Thely-phonida (Yoshikura 1975) and in embryonic and earlypost-embryonic stages of Solifugae (Roewer 1934)where they are termed lateral organs The so-calledlsquolateral organrsquo of embryonic Xiphosura (174) occurs onthe lateral surface of the carapace and does not appearto be homologous with the coxa-associated structuressharing their name In fact the xiphosuran lateralorgan develops even when the coxa of postoral somiteIV has been excised (Sekiguchi 1988) In contrastWheeler amp Hayashi (1998) and Giribet et al (2002)followed Yoshikura (1975) in regarding all lsquolateralorgansrsquo as homologous and in miscoding this characteras absent in Acariformes Van der Hammen (1989)suggested that the sternal verrucae of Opilioacari-formes are homologous with the Claparegravede organ butit differs in gross structure and ontogenetic timing174 Embryonic lateral organ associated with cara-pace 0 absent 1 present [WH sim31 GEWB sim211]State 1 is known only from embryonic stages of extantXiphosura (Yoshikura 1975) In contrast to previousinterpretations this feature does not appear to behomologous with the lsquolateralrsquo or Clarapegravede organ ofcertain arachnids (173)175 Live birth 0 absent 1 presentState 1 occurs in extant Scorpiones and has evolvedseveral times in a few mites (Evans 1992)176 Hexapodal larva and 1ndash3 nymphal stages 0absent 1 present [WP 53 S sim61 WH 27 GEWB 186]State 1 occurs in Acari and Ricinulei (Lindquist1984)177 Hexapodal prelarva 0 absent 1 present [S sim61WH 47]State 1 occurs in Opilioacariformes (NeocarusKlompen 2000) and many Acariformes (Evans 1992)

EXCRETORYOSMOREGULATORY SYSTEM

178 Malpighian tubules 0 absent 1 present [WP 17S 62 WH 9 GEWB 153]State 1 occurs in all major arachnid groups except Pal-pigradi Pseudoscorpiones Opiliones Oribatida (Sar-coptiformes) and Prostigmata Within Acari State 1occurs in all major anactinotrichid lineages (Opilioac-ariformes Holothyrida Mesostigmata Ixodida) andsimilar structures are present in certain Astigmata(Evans 1992 Alberti amp Coons 1999)

179 Dorsomedian excretory organ 0 absent 1presentThis is a specialized excretoryosmoregulatory organformed by the postventriculus and proctodeum inProstigmata (Alberti amp Coons 1999)180 Adult coxal organ opening on or near coxa ofappendage III (= arachnid leg 1) 0 absent 1 present[S 64 WH 93 GEWB 90]State 1 occurs throughout Acari (Evans 1992 Albertiamp Coons 1999) Ricinulei (Pittard amp Mitchell 1972Legg 1976) Palpigradi (Millot 1942) AraneaeAmblypygi Thelyphonida and Schizomida (Buxton1913 1917) There is evidence that the second lsquoozo-porersquo of gonyleptids (Opiliones Laniatores) (Hara ampGnaspini 2003) is a persistent opening to the coxalorgan associated with leg 1 (Soslashrensen 1879) whichalso occurs in embryonic Phalangium (Opiliones Eup-noi) (Moritz 1959)181 Adult coxal organ opening on or near coxa ofappendage V (= arachnid leg 3) 0 absent 1 present [S63 WH 92 GEWB 89]State 1 occurs in extant Xiphosura (Yamasaki et al1988) Scorpiones mygalomorph (Buxton 1913 1917)and mesothele Araneae (Liphistius J Millot in Bris-towe 1932) State 1 occurs in basal Amblypygi(Charon Charinus Buxton 1913 1917) but degener-ates prior to the adult stage in higher groups(Phrynus Weygoldt 2000) Adequately preservedeurypterids show a small submarginal opening on thecoxa of appendage V (Baltoeurypterus Selden 1981also Eurypterus Hughmilleria Clarke amp Ruedemann1912) and this is interpreted here as the opening to acoxal organ Giribet et al (2002) coded this characteras uncertain for Eurypterida182 Adult coxal organ opening on coxa of appendageII (= arachnid palp) 0 absent 1 present [GEWB 65]Buxton (1913 1917) noted that the coxal organ ofSolifugae opens on the prolateral surface of the palpalcoxa The proximal end of the duct leading to the ori-fice is associated with glands and apparently theglandular secretion andor coxal fluid serve as saliva(Alberti 1979) (see also 14 15)183 Genital papillae 0 absent 1 present - inappli-cable coded only for those taxa with eversible lsquoappend-agesrsquo (111)The genital papillae are associated with the genitalopening in postlarval Acariformes (Evans 1992Alberti amp Coons 1999) These structures are probablyserially homologous with the prosomal Claparegravedeorgan (173) and both structures probably function inwater and ion regulation

DIGESTIVE SYSTEM

184 Ingestion 0 solid food 1 primarily liquid foodwith or without preoral digestion [WP sim16 WH sim8GEWB sim218]

264 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Extant xiphosurans ingest solid food and members ofmost extant arachnid orders ingest fluids Opilionesand Acari are the principal exceptions The phalangidOpiliones ingest solids but Shultz (2000 unpublobserv) has shown that the precerebral pharyngealapparatus of Cyphophthalmi is very similar to that ofthe fluid-feeding Scorpiones and differs substantiallyfrom phalangid Opiliones Examination of gut con-tents of the cyphophthalmids Siro acaroides S exilisand Chileogovea oedipus have failed to reveal solidparticles typically found in the guts of phalangids(Shultz unpubl observ) the character is coded asunknown for Cyphophthalmi Within Acari consump-tion of particles has been documented in Opilioacari-formes (Neocarus Van der Hammen 1989unidentified Walter amp Proctor 1998) Holothyridanon-parasitic Sarcoptiformes (Evans 1992 Walter ampProctor 1998 Alberti amp Coons 1999) and many endeo-stigmatids including the nematophagous Alycusroseus (Walter 1988)185 Mouth 0 directed posteroventrally 1 directedanteroventrally [S 9 WH 55 GEWB 28]State 1 occurs throughout Arachnida (Shultz 1990)186 Oral cavity dilated by muscles arising from coxaeand constricted by large circular sphincter 0 absent1 presentState 1 occurs in extant Xiphosura (Manton 1964Scholl 1977 Shultz 2001)187 Palate plate 0 absent 1 present [GEWB 159]State 1 is a unique synapomorphy of Araneae (Dunlop1994)188 Lateral walls of epistome broadly fused to medialwalls of palpal coxae opposite sides connected by well-developed transverse epistomal muscle 0 absent 1presentState 1 occurs throughout Opiliones and Scorpiones(Shultz 2000 unpubl observ)189 Epistome with a pair of lateral arms that projectsposteriorly into the prosoma on either side of the phar-ynx 0 absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)190 Epistome with four pairs of suspensor musclesattaching to the carapace 0 absent 1 presentState 1 occurs in extant Scorpiones (Lankester et al1885 Vyas 1970 Shultz unpubl observ) and appar-ently in Solifugae (Roewer 1934)191 Intercheliceral epipharyngeal sclerite 0 absent1 present [GEWB 31]State 1 occurs in Palpigradi (Boumlrner 1904) Trigo-notarbida (Palaeocharinus Dunlop 1994) Araneae(Firstman 1954 Marples 1983) Amblypygi (Mil-lot 1949b Shultz 1999) Schizomida (Millot1949e) Thelyphonida (Millot 1949e Shultz 1993)and Phalangida (Opiliones) (Shultz 2000 unpublobserv)

192 Epipharyngeal sclerite large projecting posteri-orly 0 absent 1 present - inapplicable due to absenceof sclerite (191) [GEWB 32]State 1 occurs in Amblypygi (Millot 1949b Shultz1999) Schizomida (Millot 1949e) Thelyphonida (Mil-lot 1949e Shultz 1993) some Araneae (Heptathela ndashcoded as Liphistius ndash and Hypochilus Marples 1983not Aphonopelma Firstman 1954) and perhaps Trig-onotarbida (Palaeocharinus Dunlop 1994)193 Dorsal dilator muscle of precerebral pharynxattaching to intercheliceral septum or associatedepipharyngeal sclerite (191) 0 absent 1 presentState 1 occurs in Palpigradi (Millot 1943) Araneae(Marples 1983) Amblypygi Schizomida and Thely-phonida (Millot 1949e Shultz 1993 1999)Extremely fine muscles have been documented insome Scorpiones (Shultz 2007) and in Leiobunum(Opiliones) (Shultz 2000) Lankester et al (1885)reported a muscle in Limulus that would appear tocorrespond to this muscle but the muscle does notexist (Manton 1964 Shultz 2001)194 Dorsal dilator muscle of precerebral pharynxattaching to dorsal surface of prosoma 0 absent 1present [GEWB 228]State 1 occurs in Palpigradi (Roewer 1934) and mostAraneae (Palmgren 1978 Aphonopelma Firstman1973) but not in Heptathela (coded as Liphistius) orHypochilus (Marples 1983)195 Lateral dilator muscle of precerebral pharynxattaching to endosternite 0 absent 1 present [GEWBsim226]State 1 is a unique feature of Araneae (Marples 1983)196 Lateral dilator muscle of precerebral pharynxattaching to lateral surface of epistomal processes 0absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)197 Lateral dilator muscle of precerebral pharynxattaching to medial process of coxa of appendage II(= arachnid palp) 0 absent 1 present [GEWB sim226]State 1 occurs in Amblypygi Thelyphonida and Schi-zomida (Boumlrner 1904 Shultz 1993 1999)198 Dilator muscle of precerebral pharynx andorpreoral cavity attaching to ventral surface of prosoma0 absent 1 present [GEWB 227]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) Araneae (Marples 1983) Palpigradi(Millot 1943) and Solifugae (Roewer 1934)199 Postcerebral pharynx 0 absent 1 present [WPsim31 S sim5 WH sim21 GEWB sim154]The cuticle-lined foregut passes through the centralnervous system in many arthropods That portion pos-terior to the CNS is here termed the postcerebralpharynx regardless of the details of its morphology orfunctional specialization A substantial postcerebralpharynx is known in extant Xiphosura (Lankester

PHYLOGENY OF THE ARACHNID ORDERS 265

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

et al 1885 Manton 1964 Yamasaki et al 1988)Scorpiones (Centruroides Hadrurus HeterometrusShultz 2007 also Androctonus Abd el-Wahab 1952)Solifugae (Millot amp Vachon 1949) Araneae Ambly-pygi Thelyphonida and Schizomida (Millot 1949andasheShultz 1993) Weygoldt amp Paulus (1979) applied theterm lsquopostcerebral pharynxrsquo only to the lsquosucking stom-achrsquo of Araneae and Amblypygi but the sucking stom-ach is here regarded as a complex of three characters199ndash201

200 Dilator muscle of postcerebral pharynx attach-ing to endosternite 0 absent 1 present - inapplicabledue to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Lankester et al1885 Manton 1964 Shultz 2001) Scorpiones (Cen-truroides Hadrurus Heterometrus Shultz 2007 alsoAndroctonus Abd el-Wahab 1952) Araneae (First-

man 1954 Palmgren 1978) Amblypygi (Millot1949b Shultz 1999) and apparently the palpigradeProkoenenia wheeleri (Rucker 1901) but not inEukoenenia miriabilis (Millot 1943 contra Boumlrner1904)201 Dilator muscle of postcerebral pharynx attachingto dorsal surface of prosoma 0 absent 1 present -inapplicable due to absence of postcerebral pharynx(199)State 1 occurs throughout Araneae (Millot 1949cFirstman 1954 Palmgren 1978) and Amblypygi(Charinus Millot 1949b Phrynus Shultz 1999)202 Crop and gizzard 0 absent 1 present - inap-plicable due to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Sekiguchi 1988)but not in extant Arachnida

230 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

COMPARISONS OF ALTERNATIVE HYPOTHESES

Phylogenetic hypotheses proposed in the recent liter-ature were also evaluated including those thatattempted to resolve arachnid phylogeny completely(Fig 2) their hypothesized subclades (Fig 3) andhypotheses that proposed only specific nodes (Figs 34) The fully resolved topologies of Weygoldt amp Paulus(1979) Van der Hammen (1989) Shultz (1990)Wheeler amp Hayashi (1998) and Giribet et al (2002)were compared with the optimal topology using theTempleton test (Templeton 1983) as implementedin PAUP Internal relationships of multisampledorders were constrained to match those of the optimaltopology unless the original authors explicitlyfavoured an alternative Node-specific hypotheseswere evaluated by determining the frequency withwhich the node was recovered in bootstrap analysisThey were also evaluated by constraining parsimonyanalysis in TNT to recover the shortest tree contain-ing each specific node and then assessing the effect onrelative tree length and overall resolution takingnote of effects on otherwise stable or well-supportedclades The entire matrix was used to assess the twocases where node-specific hypotheses involved fossiltaxa (Fig 4)

RESULTS

EXTANT TAXA

Unweighted analysis of extant taxa produced two min-imal-length trees (length 383 CI 057) (Fig 5A) withconflicts limited to relationships within Pseudoscorpi-ones Multiply sampled orders were recovered asmonophyletic with high nodal support as measured bybootstrap percentage (BP gt 80) except Acari (BP 30)There were few well-supported interordinal groupsexcept Uropygi (BP 100) Pedipalpi (BP 100) Tetra-pulmonata (BP 93) and ArachnidaXiphosura (BP100) All other interordinal relationships were recov-ered with BP less than 60

Results from implied weights analysis (IWA) indi-cated a significant effect of homoplasy on resolution ofthe deepest interordinal relationships although theinterordinal clades Stomotheca Haplocnemata Acar-omorpha Tetrapulmonata Pedipalpi and Uropygiwere recovered under all values of k IWA with k = 1 (1tree best score = 5124) recovered Megoperculata(= Palpigradi + Tetrapulmonata) and Micrura(= Megoperculata + Acaromorpha) with Micrura beingthe sister group to Haplocnemata (= Pseudoscorpiones+ Solifugae) and with Stomothecata (= Opiliones +Scorpiones) being the sister group to all other arach-nids (Fig 5B) With k = 2 (1 tree best score = 3807)and k = 3 (1 tree best score = 3054) Megoperculatawas recovered and Stomotheca remained the sister

group to other arachnids but the relative positions ofHaplocnemata and Acaromorpha changed (Fig 5C)IWA using k = 4 (1 tree best score = 2559) k = 5 (1tree best score = 2205) and k = 6 (1 tree bestscore = 1938) recovered topologies consistent with theminimal-length topology derived from unweightedcharacters (Fig 5A)

EXTANT AND FOSSIL TAXA

Unweighted analysis of extant and fossil taxa pro-duced 18 minimal-length trees (length 426 CI 056)(Fig 6A) with conflicts limited to relationships withinPseudoscorpiones Scorpiones and Araneae Multiplysampled orders were again recovered as monophyleticwith high nodal support (BP gt 80) except Acari whichwas reconstructed as diphyletic with Anactinotrichidabeing the sister group to Ricinulei (BP 33) Therewere no well-supported interordinal groupings (ieBP gt 80) except Uropygi (BP 89) Pedipalpi (BP 92)and Arachnida (BP 90)

Results from IWA indicated a significant effect ofhomoplasy on resolution of the deepest interordinalrelationships in Arachnida although StomothecataHaplocnemata Acaromorpha Cryptognomae Pantet-rapulmonata Tetrapulmonata Schizotarsata Pedi-palpi and Uropygi were recovered under all values ofk IWA with k = 1 (7 trees best score = 5675) recov-ered Megoperculata and Micrura with Micrura beingthe sister group to Haplocnemata Stomothecata wasnot recovered in the strict consensus (Fig 6B) Withk = 2 (3 trees best score = 4243) IWA recovered thetopology shown in Fig 6C with Palpigradi being thesister group to all other arachnids The strict consen-sus topologies reconstructed using k = 3ndash6 were iden-tical to those obtained with k = 2 (k = 3 3 trees bestscore = 3419 k = 4 3 trees best score = 2874 k = 53 trees best score = 2483 k = 6 3 trees bestscore = 2188)

COMPARISONS OF ALTERNATIVE HYPOTHESES

The hypothesis of Weygoldt amp Paulus (1979) (Fig 2)was 398 steps (15 steps longer than the unconstrainedminimal-length tree) Van der Hammen (1989) was400 steps (17 steps longer) Shultz (1990) was 399steps (six steps longer) Wheeler amp Hayashi (1998)was 391 steps (eight steps longer) and Giribet et al(2002) was 394 steps (11 steps longer) Templetontests conducted at the 005 significance level rejectedthe hypothesis of no difference between the optimaltopology and those of Weygoldt amp Paulus (P = 00002)Van der Hammen (P = 00011) Wheeler amp Hayashi(P = 00018) and Giribet et al (P = 00343) The Shultztopology was not significantly different (P = 01444)from the optimal topology

PHYLOGENY OF THE ARACHNID ORDERS 231

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 2 Fully resolved phylogenetic hypotheses of extant euchelicerate groups proposed in the recent literature Note thesimilarity in the topologies of the parsimony-based analyses by Shultz (1990) Wheeler amp Hayashi (1998) and Giribet et al(2002) The Giribet et al topology is based on neontological data (morphology and molecules) and the original lsquoROOTrsquo maybe an artefact from use of the highly divergent pycnogonids as an outgroup

Xip

hosu

raS

corp

ione

sO

pilio

nes

Ara

neae

Am

blyp

ygi

Uro

pygi

Ric

inul

eiA

nact

inot

richi

daS

olifu

gae

Pse

udos

corp

ione

sP

alpi

gra

diA

carif

orm

es

Xip

hosu

raS

corp

ione

sA

rane

aeA

mbl

ypyg

iU

ropy

giP

alpi

gra

diS

olifu

gae

Pse

udos

corp

ione

sO

pilio

nes

Ric

inul

eiA

cari

Lipoctena

Arachnida

Apulmonata

Holotracheata

Cryptoperculata

Acaromorpha(ldquoAcarinomorphardquo)

Tetrapulmonata(rdquoMegoperculatardquo)

LabellataHaplo-cnemata

Weygoldt amp Paulus (1979)

Labellata

Tetrapulmonata(rdquoArachnidardquo)

Haplocnemata(rdquoApatellatardquo)

Van der Hammen (1989)

Xip

hosu

raP

alpi

gra

diA

rane

aeA

mbl

ypyg

iU

ropy

giR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Pedipalpi Haplocnemata

Novogenuata

Dromopoda

Shultz (1990)

Acaro-morpha

Micrura

Xip

hosu

raP

alpi

gra

diU

ropy

giA

mbl

ypyg

iA

rane

aeR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Labellata Haplocnemata

Novogenuata

Dromopoda

Wheeler amp Hayashi (1998)

Acaro-morpha

Micrura

Xip

hosu

raA

cari

Pal

pig

radi

Ric

inul

eiA

rane

aeA

mbl

ypyg

iU

ropy

giO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Tetrapulmonata

PedipalpiHaplocnemata

Novogenuata

Dromopoda

Giribet et al (2002)

ROOT

Micrura

Crypto-gnomae

Sternocoxata

Epimerata

RostrosomataMyliosomata

Neosternata

232 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 3 Consensus trees produced by parsimony analysis of neontological data constrained to produce relationships pro-posed in the recent literature The constrained (target) node is indicated by a black dot and the taxa encompassed by theconstraint are enclosed in a box Numbers below each tree represent the number of minimal-length constrained treeslength of minimal-length trees difference in the length of the unconstrained minimal-length tree and the constrained min-imal-length tree and percentage unconstrained bootstrap trees in which the target node was recovered respectively Thesetrees indicate the effect on branch length imposed by specific hypotheses and impact of constraining target nodes on non-target nodes

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Lipoctena4 385 2 lt5

Xi

Sc

Op

Ps

So

Pa

Ur

Am

Ar

Ri

An

Ac

Labellata6 397 14 lt5

Xi

Sc

Ar

Am

Ur

Op

Pa

Ps

So

Ri

An

Ac

Apulmonata2 390 7 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

So

Ri

An

Ac

Holotracheata5 390 7 lt5

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptoperculata2 388 5 lt5

Xi

Sc

Op

Pa

So

Ps

Ar

Am

Ur

Ri

An

Ac

Rostrosomata6 384 1 25

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Neosternata3 388 5 lt5

Xi

Sc

Op

Ar

Am

Ur

Ri

An

Ps

So

Pa

Ac

Coxisternata2 388 5 lt5

Epimerata7 387 4 11

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptognomae6 384 1 24

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Micrura3 385 2 7

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Megoperculata3 384 1 28

Xi

Sc

Ps

So

Op

Pa

Ar

Am

Ur

Ri

An

Ac

Novogenuata10 388 5 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

Ri

So

An

Ac

Solifugae + Acari4 388 5 lt5

Shultz (1990) Alberti amp Peretti (2002)

Van der Hammen (1989)

Weygoldt amp Paulus (1979)

Xi

Ar

Am

Ur

So

Ps

Sc

Op

Ri

Pa

An

Ac

PHYLOGENY OF THE ARACHNID ORDERS 233

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Results from analyses of node-specific hypothesesare summarized in Figures 3 4 In most cases pro-posed relationships were recovered in fewer than 5of bootstrap pseudoreplicates The notable exceptionswere Rostrosomata (25) Cryptognomae (24) andMegoperculata (28) and the strict-consensus con-strained minimal-length tree for each was only onestep longer than the unconstrained optimal tree Inmost cases constrained topologies did not add manysteps most were less than five steps longer and nonewas greater than 14 steps longer than the un-constrained minimal-length tree It is noteworthyhowever that even constrained nodes that imposerelatively few extra steps sometimes supportedimprobable relationships or eliminated clades thatwere well supported or stable in the unconstrainedanalysis Specific examples and their implications arediscussed below

DISCUSSION

RECENT ISSUES IN ARACHNID PHYLOGENY

Available morphological evidence consistently re-solves some interordinal relationships and fails to

resolve others (Fig 1) Continued progress depends onthe ability of arachnologists to discover new charac-ters and to assess the evidence critically Whilereviewing the recent arachnological literature severalaspects of phylogenetic practice emerged that seemedcounterproductive to both the perception and theactual rate of progress toward resolving arachnid phy-logeny Some of these are summarized here withspecific examples given in the remainder of theDiscussion and in the Appendix

There is a tendency to portray arachnid ordinal phy-logeny as more poorly resolved and contentious than isactually the case (Coddington et al 2004) Phyloge-netic hypotheses generated during different historicalperiods and using differing standards of evidence areoften cited as examples of current disagreement (egSelden 1993 Dunlop 1996 Selden amp Dunlop 1998Wheeler amp Hayashi 1998) In fact recent parsimony-based analyses of morphology have tended to convergeon topologies with internal structures congruent withthose found here (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) (Fig 2) Arachnidphylogeny is not fully resolved especially at its deep-est levels (Fig 1) but this does not mean that allaspects of arachnid phylogeny are controversial orpoorly supported by the available evidence

Matrices are sometimes constructed by uncriticallsquorecyclingrsquo of erroneous or problematic charactersbased on diverse secondary sources (Jenner 2001)Conclusions derived from mixtures of valid invalidspeculative and redundant characters are sometimesportrayed as the phylogenetic signal provided by mor-phology Data recycling can perpetuate errors (see 1320 39 52 136 140 144 152 169 171ndash174) legiti-mize speculations (see 13 32 63 95) or create dupli-cate or non-independent characters (see 13 30 70 77172) The lsquolateral organrsquo is a notable exampleYoshikura (1975) equated the embryonicearly post-embryonic lsquolateral organsrsquo of Amblypygi Thelyphonidaand Solifugae with the dissimilar lsquolateral organrsquo ofXiphosura but failed to note the very similar Cla-paregravede organ of Acariformes (see 173 174) This cod-ing was recycled by Wheeler amp Hayashi (1998) andthen by Giribet et al (2002) Error is probably inevi-table when assembling a large matrix from morphol-ogy including the one presented here but this can beminimized by making original observations and byconsulting primary sources

Some workers advocate weighting characters a pri-ori on the basis of structural or functional complexity(eg Kraus 1998 Dunlop amp Braddy 2001) and dismissphylogenetic conclusions derived from equal-weightsparsimony In short these workers criticize parsimonyfor emphasizing data quantity over data quality Thiscriticism ignores the intense debate in systematic biol-ogy that eventually led to widespread adoption of

Figure 4 Consensus trees produced by parsimony analy-sis of the full data matrix constrained to produce relation-ships proposed in the recent literature See legend toFigure 3 for details

Xi

Eu

Sc

Op

Pa

Ac

Ri

An

So

Ps

Tr

Ar

Ha

Am

UrEurypterida slat + Scorpiones12 435 9 lt5

Xi

Eu

Sc

Op

Pa

Ac

An

So

Ps

Ri

Tr

Ar

Ha

Am

UrTrigonotarbida+ Ricinulei28 434 8 lt5

Dunlop amp Braddy (2002)et al

Dunlop (1996)

234 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 5 Results from analysis of neontological data A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology AFor B and C relationships within terminal clades are the same as those shown in A

LimulusTachypleusProkoeneniaEukoeneniaCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAllothyrusAustralothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesLiphistiusAphonopelmaHypochilusCharinusPhrynusStenochrusProtoschizomusMastigoproctusCyphophthalmusChileogoveaCaddoLeiobunumSclerobunusGonyleptesCentruroidesHadrurusHeterometrusChthoniusFeaellaNeobisiumCheliferEremocostaGaleodes

10031

10018451

501561

733

723

944

633

10010

832

901

943

903

522

341

1007

1008

Acaromorpha

Haplocnemata

10012

Stomothecata

Opiliones

Scorpiones

Pseudo-scorpiones

Solifugae

Acariformes

Parasitiformes

Opilioacariformes

Ricinulei

Xiphosura

Palpigradi

Thelyphonida

Schizomida

Araneae

Amblypygi

10011

936853

100141005

1009

Tetrapulmonata Pedipalpi

Uropygi

692

522572

968

331

603

81

511

161

101

Acari

Dromopoda

Arachnida

A

Xiphosura

Stomothecata

Haplocnemata

Acaromorpha

Palpigradi

TetrapulmonataMicrura

Megoperculata

BXiphosura

Stomothecata

Acaromorpha

Haplocnemata

Palpigradi

TetrapulmonataMegoperculata

C

PHYLOGENY OF THE ARACHNID ORDERS 235

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 6 Results from analysis of the full data matrix A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology CFor B and C relationships within terminal clades are the same as those shown in A

WeinberginaLimuloidesEuproopsTachypleusLimulusStylonurusBaltoeurypterusChasmataspisDiploaspisOctoberaspisCyphophthamusChileogoveaCaddoLeiobunumSclerobunusGonyleptesProscorpiusPalaeoscorpiusStoermeroscorpioPrearcturusCentruroidesHadrurusHeterometrusProkoeneniaEukoeneniaTerpsicrotonPoliocheraCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAustralothyrusAllothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesEremocostaGaleodesChthoniusFeaellaNeobisiumCheliferGilboarachnePalaeocharinusLiphistiusAphonopelmaHypochilusPlesiosiroCharinusPhrynusStenochrusProtoschizomusProschizomusMastigoproctus

631

893

923

984

984

671

341

671

986

10011

423

974

Xiphosura

Eurypterida sstr

Chasmataspidida

Opiliones

Scorpiones

Palpigradi

Ricinulei

Opilioacariformes

Parasitiformes

Acariformes

Solifugae

Pseudoscorpiones

Trigonotarbida

Araneae

Haptopoda

Amblypygi

Schizomida

Thelyphonida

842

251331

361311

552

832

882

925

913

573

561291

443

976

541

954

743644

732521

571

331

833

663591

966

573

71

221

905

832

351

Uropygi

Pedipalpi

Haplocnemata

Tetrapulmonata

Anactinotrichida

Cryptognomae

Acaromorpha

Stomothecata

Pantetrapulmonata

Arachnida

Eurypterida slat

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatProscorpiusStoermeroscorpioPalaeoscorpiusPrearcturusCentruroidesHadrurusHeterometrusOpilionesHaplocnemataRicinuleiAnactinotrichidaAcariformesPalpigradiPantetrapulmonata

Megoperculata

Micrura

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatPalpigradiStomothecataHaplocnemataPantetrapulmonataAcaromorpha

A

B

C

Schizotarsata

236 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

parsimony and instead advocates a return to thespeculative and subjective approaches of the late 19thand early 20th century that once threatened thescientific legitimacy of the discipline (Bowler 1996)Furthermore it misrepresents the properties of parsi-mony-based analyses Specifically characters withfunctional significance are readily encompassed byparsimony analyses and in fact characters derivedfrom locomotor systems (eg 46ndash94) have played animportant role in developing current ideas aboutarachnid phylogeny (eg Shultz 1989 1991) In addi-tion lsquocomplex charactersrsquo can be viewed as compositesof several characters such that morphological com-plexity is effectively weighted by the number of inde-pendent lsquosubcharactersrsquo it contains For example thelsquosucking stomachrsquo once considered a synapomorphy ofLabellata (= Araneae + Amblypygi Fig 3) is codedhere as a composite of three characters (199ndash201) TheLabellata hypothesis was not corroborated in thepresent analysis (Figs 1 3ndash5) and in fact was highlydisfavoured (Fig 3) but this result cannot be dis-missed as a failure to acknowledge the complexity ofthe character

Some workers support specific (target) clades withone or more similarities without exploring the impacton overall homoplasy or relationships among non-target clades Each character offers its own phyloge-netic hypothesis which may or may not be consistentwith relationships implied by other characters It isexceedingly rare for all characters to be perfectly com-patible in the phylogenetic hypotheses they supportand criteria such as parsimony have been developed todiscover those hypotheses that minimize the conflict-ing phylogenetic signals of different characters Eventhough character conflict (homoplasy) is a virtuallyinescapable phenomenon in comparative biology it isnot uncommon for workers to discover one or morecharacters and to promote their phylogeneticimplications over alternative hypotheses even thosethat otherwise appear to be well supported Thisapproach may have value in highlighting new dataand perspectives but accomplishes this by promotingthe erroneous impression that all aspects of arachnidphylogeny are so tenuous that a single character canfalsify even well-supported hypotheses Several exam-ples of this approach have appeared in the recentarachnological literature

For example Alberti amp Peretti (2002) argued thataflagellate spermatozoa (163) are a compelling syna-pomorphy for a Solifugae + Acari clade and dismissedsome of the characters that support Solifugae + Pseu-doscorpiones as having lsquodebatable valuersquo Yet theirproposal rejects two hypotheses that are consistentlyrecovered in recent phylogenetic analyses that isHaplocnemata (= Solifugae + Pseudoscorpiones) andAcaromorpha (= Acari + Ricinulei) (Figs 1 2) Rejec-

tion of Haplocnemata would require its presumed syn-apomorphies to be reinterpreted as homoplasiesincluding features of the chelicerae (18 19 20) pre-oral chamber (13 32) legs (12 48) and respiratorysystem (126) The same reason applies to Acaromor-pha and its synapomorphies It is noteworthy that aSolifugae + Acari clade was recovered in the presentanalysis in fewer than 5 of bootstrap pseudorepli-cates and that analyses constrained to recover thisclade were five steps longer than the minimal-lengthtree and favoured a problematic clade unitingRicinulei and Pseudoscorpiones (Fig 3)

In another example Dunlop (1996) proposed a closerelationship between Trigonotarbida and Ricinuleibased on two-segmented chelicerae (18) prosomandashopisthosoma coupling mechanism (96) diplotergites(100 101) and longitudinally divided opisthosomaltergites (115) However phylogenetic analyses con-strained to recover this relationship required eightadditional steps eliminated support for Acari and nec-essarily rejected Acaromorpha and forced its synapo-morphies to be reinterpreted as homoplasies Alberti(2005) proposed an interesting hypothesis for the evo-lution of male gonads in tetrapulmonates (153) butchose to accept Labellata (= Araneae + Amblypygi) indeveloping his argument over the much more well-supported Pedipalpi (= Amblypygi + Uropygi) (Shultz1999) a phylogenetic reconfiguration that was recov-ered here in fewer than 5 of bootstrap pseudorepli-cates increased tree length by a minimum of 14 stepsand resulted in the collapse of Stomothecata and Hap-locnemata (Fig 3) Many other examples can be cited

Promoting or defending a specific phylogenetichypothesis via lists of compatible synapomorphies is acommon but problematic approach By restrictingattention to the states of specific characters at one ortwo target nodes one can easily overlook the unin-tended impact of the hypothesis on phylogenetic sig-nal elsewhere and its effect on non-target clades Anode supported by a long list of synapomorphies mayseem convincing taken in isolation but may becomeless acceptable when its full phylogenetic implicationsare explored

SUMMARY OF INTERORDINAL ARACHNID CLADES RECOVERED IN THIS ANALYSIS

Arachnida Lamarck 1801Analyses consistently recovered Arachnida as a mono-phyletic group with high bootstrap support (Figs 1 45) Possible synapomorphies include the loss of thecarapacal pleural doublure (9) cardiac lobe (10) pedalgnathobases (52) and moveable endites (53) and thegain of aerial respiration (120) and an anteriorly oranteroventrally directed mouth (185) Some tradi-tional synapomorphies such as slit sensilla (142) and

PHYLOGENY OF THE ARACHNID ORDERS 237

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

fluid feeding (184) may have appeared later in arach-nid evolution but this can only be decided once theinternal phylogeny of Arachnida has been determined

Some workers regard many proposed arachnid syn-apomorphies as adaptations to terrestrial life andthereby link the hypothesis of arachnid monophyly tothe hypothesis of a single ancestral aquatic-to-terres-trial transition and arachnid polyphyly to multipletransitions (eg Selden amp Jeram 1989 Dunlop 1997Dunlop amp Webster 1999 Dunlop amp Braddy 2001)Given this line of reasoning the existence of appar-ently aquatic scorpions in the fossil record (Kjellesvig-Waering 1986 Jeram 1998) and the inference thatterrestrialization occurred late in scorpion evolutionthese workers conclude that character states support-ing arachnid monophyly are actually convergencesand do not necessarily support arachnid monophyly

However this approach to assessing phylogenetichypotheses is founded on overly simplistic assump-tions such as the ability of the investigator to discrim-inate unerringly between characters that existexclusively in aquatic organisms (including fossils)from those that occur exclusively in terrestrial organ-isms There also appears to be an assumption thathomoplasy can be generated through parallelism (iemultiple aquatic-to-terrestrial events) but not throughterrestrial-to-aquatic reversals Furthermore anassumed dichotomy between exclusively aquatic andterrestrial life histories in ancestral arachnids is sim-plistic as illustrated by the amphibious life cycles ofbasal vertebrates and pterygote hexapods In factthese examples demonstrate that there is no neces-sary inconsistency in basing a hypothesis of arachnidmonophyly on the derived terrestrial features of anamphibious ancestor whose descendants thencompleted terrestrialization once or several timesindependently or even returned to a fully aquaticexistence Workers who link the frequency and direc-tion of aquaticndashterrestrial transitions to the assess-ment of arachnid phylogeny do so by endowingthemselves with substantially greater insight thanseems prudent by ignoring the huge gaps in ourunderstanding of early arachnid evolution and bydenying to arachnids the evolutionary complexityknown to exist in other groups

Several palaeontologists have been particularlyactive during the past decade in proposing new char-acters with the stated goal of removing Scorpionesfrom Arachnida and erecting a Scorpiones + Eurypter-ida clade (Braddy amp Dunlop 1997 Dunlop amp Braddy1997 Dunlop 1998 Braddy et al 1999 Dunlop ampWebster 1999) Dunlop amp Braddy (2001) recentlysummarized this evidence and conducted aparsimony-based analysis of Xiphosura EurypteridaScorpiones Opiliones and Tetrapulmonata (but notHaplocnemata Acaromorpha or Palpigradi) using 33

morphological characters Their analysis produced atopology congruent with those generated here (Figs 15 6) including recovery of Stomothecata (= Scorpiones+ Opiliones) However they rejected this result as aproduct of lsquoempirical cladisticsrsquo because it gives thesame weight to prosomal characters that supportarachnid monophyly and to selected opisthosomalcharacters that support their favoured Eurypterida +Scorpiones clade namely a five-segmented metasoma(116) suppression of opisthosomal tergite 1 (95) lossof lamellate respiratory organs on the postgenitalsomite (122) Kiemenplatten (125) loss of respiratorylamellae on the genital segment (121) and a lsquonon-stainingrsquo exocuticle (but see Grainge amp Pearson 1966for evidence of this in Opiliones see Appendix for com-ments on the other characters)

A Eurypterida + Scorpiones clade was not favouredin the present analysis (Fig 6) the strict consensus ofminimal-length trees constrained to recover this clade(Fig 4) is nine steps longer than the tree recovered byanalysis without this constraint It is also noteworthythat Opiliones was consistently reconstructed as thesister group to Eurypterida + Scorpiones in the mini-mal-length constrained trees a provocative result thatwas probably unintended and unanticipated by sup-porters of the Eurypterida + Scorpiones Given thatDunlop amp Braddy (2001) reject equal weights parsi-mony as an arbiter of phylogenetic hypotheses theywould presumably dismiss these results as irrelevantto their argument just as they dismissed their ownparsimony-based results However if the Eurypterida+ Scorpiones hypothesis is to be credible it must beopen to evaluation and potential falsification usingobjective criteria and the subjective or intuitive a pri-ori weighting of characters advocated by Dunlop ampBraddy clearly does not qualify At present it isunclear how one would objectively evaluate Dunlop ampBraddyrsquos proposal with criteria compatible with thoseused in its original formulation For now the Euryp-terida + Scorpiones concept advocated by Dunlop ampBraddy may persist outside the mainstream of mod-ern systematic practice but it is increasingly problem-atic within it

Stomothecata nomnovOpiliones and Scorpiones were consistently recoveredas a monophyletic group The proposed name acknowl-edges a unique preoral chamber the stomothecaformed by coxapophyses of the palp and leg 1 (50)often with an auxiliary role played by the coxapophy-sis of leg 2 (51) In addition the epistome appears tohave been modified for adduction of the palpal coxaeThe lateral walls of the epistome are fused to themedial surfaces of the palpal coxae and the epistomallumen is spanned by a transverse muscle (188) which

238 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

apparently adducts the palpal coxae thereby constrict-ing the stomothecal chamber Scorpions and opilionsare also unique in having a pair of large epistomalarms projecting rearward into the prosoma andattaching to the endosternite (189) The epistomalarms provide attachment sites for pharyngeal dilatormuscles (196) and extrinsic muscles of anterior proso-mal appendages (eg 37) The chelicera is equippedwith a muscle that arises on the carapace and insertson the ventral margin of the second cheliceral article(23) There is an anteriorly placed genital opening(155)

Several notable similarities were found whilereviewing the literature but information from otherarachnid groups was considered too incomplete toallow them to be included in the matrix For exampleboth orders have apparent haemocytopoeitic organsassociated with major nerves of the anterior opistho-soma These are termed supraneural organs in scorpi-ons (Farley 1999) and perineural organs in opilions(Kaumlstner 1935) Haemocytes develop in the cardiacwall in spiders (Seitz 1972) and perhaps amblypygids(Weygoldt 2000) but haemocytopoeitic organs areunknown in most other arachnid groups Germ cellsdifferentiate very early during embryogenesis in bothscorpions and opilions relative to spiders (Moritz1957 Anderson 1973) Additional research is requiredto determine the phylogenetic utility of thesecharacters

Haplocnemata Boumlrner 1904Several workers recognize a close phylogenetic rela-tionship between Pseudoscorpiones and Solifugae(Weygoldt amp Paulus 1979 Van der Hammen 1989Shultz 1990 Wheeler amp Hayashi 1998 Giribetet al 2002) and this was consistently recovered inthe present analysis Haplocnemata is supported byseveral synapomorphies including (i) a feeding com-plex formed by two-segmented chelicerae (18) with aventrolateral intrinsic articulation (19) and perhapsa dorsolateral articulation with the carapace (20)and (ii) a rostrosoma (32) a preoral apparatus formedby an anteriorly projecting epistome affixed dorsallyto enlarged palpal coxae and bordered by lateral pal-pal projections The epistome forms the dorsal compo-nent of a lsquobeakrsquo with the ventral part formed by amidventral sternapophysis (13) fused between thepalpal coxae The locomotor apparatus consists ofcoxae that meet along the ventral midline (12) elon-gate femur-like patellae (48) and an apotele in theadult modified as an eversible empodium or pulvillus(92) The respiratory system is formed entirely (Pseu-doscorpiones) or in part (Solifugae) by paired trachealstigmata opening on opisthosomal somites 3 and 4(126)

Acaromorpha Dubinin 1957A clade comprising Ricinulei and Acari is consistentlyrecovered in the present analysis It is united here bytwo unique and seven homoplasious synapomorphiesThese include a gnathosoma (31) defined in part bymedial fusion of the palpal coxae (30) although pres-ence of a gnathosoma in Ricinulei is debatable as dis-cussed in the Appendix Acaromorphs also have aunique post-embryonic development consisting of ahexapodal larva and up to three octopodal nymphalinstars (176) The pedal patellandashtibia joints are formedby a bicondylar dorsal hinge rather than a single mid-dorsal condyle (69) all postcheliceral segmental gan-glia are unified in the subesophageal ganglion (130)and a postcerebral pharynx is absent (192) The groupis also tentatively united by presence of differentiatedpedal basi- and telofemora (63 64) and the absence ofa ventral (sternal) pharyngeal dilator muscle (199)but these may be symplesiomorphies erroneouslyreconstructed as a synapomorphies

The internal phylogenetic structure of Acaromorphais controversial with many recent workers favouringa monophyletic Acari (Weygoldt amp Paulus 1979Shultz 1990 argued most thoroughly by Lindquist1984) and others advocating a diphyletic Acari withAcariformes being the sister group to Cryptognomae(= Ricinulei + Anactinotrichida) (especially Van derHammen 1979 1989) (Fig 3) Acari was recovered asmonophyletic when fossils were excluded (Fig 2) butnot when they were included (Fig 3) A lsquomite-centredrsquosurvey of arachnid characters may be needed if mor-phology is to offer a compelling solution to the internalphylogeny of Acari and its placement within Arach-nida These issues are not resolved by the presentanalysis

Pantetrapulmonata nom novPantetrapulmonata includes the extinct orders Trigo-notarbida Haptopoda and the extant orders AraneaeAmblypygi Schizomida and Thelyphonida The cladeis united by cheliceral structure (18 19) a megoper-culum (106) booklungs on the genital and first post-genital somites (121 122) and enlargement of theepipharyngeal sclerite (192) It is important to notehowever that most of these characters were coded asuncertain in Plesiosiro (Haptopoda) Aside from theplacement of Haptopoda the monophyly of Trigono-tarbida and the extant orders was anticipated byShear et al (1987) and is generally regarded as amonophyletic group

Tetrapulmonata Shultz 1990Tetrapulmonata was originally proposed on the basisof neontological analyses and encompassed Araneae

PHYLOGENY OF THE ARACHNID ORDERS 239

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Amblypygi Schizomida and Thelyphonida (Shultz1990) and was also recovered here in analysis of neon-tological data (Fig 5) However results from analysisof all taxa required that the Tetrapulmonata conceptbe expanded to include the fossil order HaptopodaFeatures uniting this group are problematic howeveras many states in Haptopoda were coded as uncertain

Schizotarsata nomnovHaptopoda and Pedipalpi (= Amblypygi + Schizomida+ Thelyphonida) are united here in a group named forpossession of divided pedal telotarsi (84) Synapomor-phies include a pointed anterior carapacal margin (3)and elongation of leg 1 (46) As already noted theplacement of Haptopoda should probably be regardedas tentative because the state of many characters inthis group was coded as unknown The placement ofHaptopoda as the sister group to Pedipalpi was antic-ipated by Dunlop (1999 2002c)

Pedipalpi Boumlrner 1904Pedipalpi encompasses Amblypygi and Uropygi(= Thelyphonida + Schizomida) Although widely rec-ognized in the past Pedipalpi was set aside in the mid-20th century in favour of Labellata (= Amblypygi +Araneae) based largely on the presumed synapomor-phies of a narrow prosomandashopisthosoma juncture (97)and sucking stomach (199ndash201) The situation beganto reverse when Shear et al (1987) again highlightedsimilarities of Amblypygi and Uropygi such as rapto-rial palps (35) and antenniform leg 1 (46) Shultz(1989 1990 1999) subsequently described numerousderived similarities including asymmetrical flexormuscles at the pedal femurndashpatella joint (70) threetelotarsomeres on legs 2ndash4 (85) and modification ofthe palpal coxae to provide support for extrinsic pha-ryngeal muscles (197) Pedipalpi has emerged as oneof the most well-supported interordinal relationshipsin Arachnida

Uropygi Thorell 1882Uropygi has long been accepted as a monophyleticunion of Thelyphonida and Schizomida Synapomor-phies include a unique mating behaviour (159) fusedpalpal coxae (30) 2-1-1-1 arrangement of tibial tricho-bothria (144) posterior defensive glands (102) elon-gated patella of leg 1 (68) and many others

ACKNOWLEDGEMENTS

I thank Nikolaj Scharff and three anonymous review-ers for useful comments This work was supported bythe Maryland Agricultural Experiment Station the

General Research Board of the University of Mary-land and National Science Foundation Grant 0416628to Susan Masta

REFERENCES

Abd el-Wahab A 1952 Notes on the morphology of the scor-pion Buthus quinquestriatus (HE) Publications de lrsquoInsti-tut Fouad 1er du Desert 3 1ndash129

Alberti G 1979 Licht- und elektronenmikroskopische Unter-suchungen an Coxaldruumlsen von Walzenspinnen (ArachnidaSolifugae) Zoologischer Anzeiger 203 48ndash64

Alberti G 1995 Comparative spermatology of Cheliceratareview and perspective Memoires du Museum NationaldrsquoHistoire Naturelle 166 203ndash230

Alberti G 2005 Double spermatogenesis in ChelicerataJournal of Morphology 266 281ndash297

Alberti G Coons LB 1999 Acari mites In Harrison FWFoelix RF eds Microscopic anatomy of invertebratesVol 8C Chelicerate Arthropoda New York Wiley Liss515ndash1215

Alberti G Peretti AV 2002 Fine structure of male genitalsystem and sperm in Solifugae does not support a sister-group relationship with Pseudoscorpiones (Arachnida)Journal of Arachnology 30 268ndash274

Alexander AJ 1967 Problems of limb extension in the scor-pion Opisthophthalmus latimanus Koch Transactions of theRoyal Society of South Africa 37 165ndash181

Anderson DT 1973 Embryology and phylogeny in annelidsand arthropods Oxford Pergamon Press

Anderson LI Selden PA 1997 Opisthosomal fusion andphylogeny of Palaeozoic Xiphosura Lethaia 30 19ndash31

Beall BS Labandeira CC 1990 Macroevolutionary patternsof the Chelicerata and Tracheata In Culver SJ ed Arthro-pod paleobiology Short Courses in Paleontology No 3 Lon-don The Paleontological Society 257ndash284

Bernard HM 1896 The comparative morphology of the Gale-odidae Transactions of the Linnean Society of London Sec-ond Series 6 305ndash417

Boumlrner C 1902a Arachnologische Studien (II und III) Zool-ogischer Anzeiger 25 433ndash466

Boumlrner C 1902b Arachnologische Studien IV Die Genitalor-gane der Pedipalpen Zoologischer Anzeiger 26 81ndash92

Boumlrner C 1904 Beitraumlge zur Morphologie der Arthropoden IEin Beitrag zur Kenntnis der Pedipalpen Zoologica (Stut-tgart) 42 1ndash174

Bowler PJ 1996 Lifersquos splendid drama Chicago Universityof Chicago Press

Braddy SJ Alridge RJ Gabbott SE Theron JN 1999Lamellate book-gills in a late Ordovician eurypterid from theSoom Shale South Africa support for a eurypterid-scorpionclade Lethaia 32 72ndash74

Braddy SJ Dunlop JA 1997 The functional morphology ofmating in the Silurian eurypterid Baltoeurypterus tet-ragonophthalmus (Fischer 1839) Zoological Journal of theLinnean Society 121 435ndash461

Brauer A 1895 Beitrage zur Kenntnis der Entwicklungsge-schichte des Skorpiones II Zeitschrift fuumlr WissenschaftlicheZoologie 59 351ndash433

240 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bremer K 1994 Branch support and tree stability Cladistics10 295ndash304

Bristowe WS 1932 The liphistiid spiders Proceedings of theZoological Society of London Part 4 1016ndash1057

Butt AG Taylor HH 1991 The function of spider coxalorgans effects of feeding and salt-loading on Porrhotheleantipodiana (Mygalomorpha Dipluridae) Journal of Exper-imental Biology 158 439ndash461

Buxton BH 1913 Coxal glands of arachnids ZoologischeJahrbucher (Suppl XIV) 16ndash50

Buxton BH 1917 Notes on the anatomy of arachnids Jour-nal of Morphology 29 1ndash31

Camin JH Clark GM Bourdeau FG 1956 The palpallsquotined setarsquo in the Mesostigmata a homologue of the palpalclaw in the Onychopalpida (Acarina) In Becker EC ed Pro-ceedings of the Tenth International Congress of EntomologyOttawa Mortimer 903ndash908

Chamberlin JC 1931 The arachnid order ChelonethidaStanford University Stanford University Press

Chapman RF 1998 The insects structure and function 4thedn Cambridge Cambridge University Press

Clarke KU 1979 Visceral anatomy and arthropod phylogenyIn Gupta AP ed Arthropod phylogeny New York Van Nos-trand Reinhold 467ndash549

Clarke J 1984 On the relationship between structure andfunction in the leg joints of Heteropoda venatoria (L) (Ara-neae Eusparassidae) Bulletin of the British ArachnologicalSociety 6 181ndash192

Clarke J 1986 The comparative functional morphology of theleg joints and muscles of five spiders Bulletin of the BritishArachnological Society 7 37ndash47

Clarke JM Ruedemann R 1912 The Eurypterida of NewYork New York State Museum Memoir 14 (1ndash2) 1ndash628

Coddington JA 2005 Phylogeny and classification of spi-ders In Ubick D Paquin P Cushing PE Roth V eds Spi-ders of North America an identification manual AmericanArachnological Society 18ndash24

Coddington JA Giribet G Harvey MS Prendini LWalter DE 2004 Arachnida In Cracraft J Donoghue MJeds Assembling the tree of life Oxford Oxford UniversityPress 269ndash318

Coddington JA Levi HW 1991 Systematics and evolutionof spiders (Araneae) Annual Review of Ecology and System-atics 22 565ndash592

Coineau Y Van der Hammen L 1979 The postembryonicdevelopment of Opilioacarida with notes on new taxa and ona general model for the evolution In Piffle E ed Proceed-ings of the 4th International Congress of Acarology Budap-est Akadeacutemiai Kiadoacute 437ndash441

Cokendolpher JC Reddell JR 1992 Revision of the Pro-toschizomidae (Arachnida Schizomida) with notes on thephylogeny of the order Texas Memorial Museum Speleolog-ical Monographs 3 31ndash74

Condeacute B 1991a Le dimorphisme sexuel des Palpigrades Bul-letin de la Socieacuteteacute Neuchacircteloise des Sciences Naturelles 11667ndash73

Condeacute B 1991b Lrsquoaire genitale male des Palpigrades Bulle-tin de la Socieacuteteacute Europeenne drsquoArachnologie 1990 64ndash69

Dearden PK Donly C Grbicccc M 2002 Expression of pair-rule gene homologues in a chelicerate early patterning ofthe two-spotted spider mite Tetranychus urticae Develop-ment 129 5461ndash5472

Delle Cave L 1975 The ldquocleaning brushrdquo in Damon diademaSimon 1876 and Phrynichus jayakari Pocock 1894 (Ambly-pygi Arachnida) Monitore Zoologico Italiano (NS) 9 175ndash184

Dunlop JA 1994 Filtration mechanisms in the mouthparts oftetrapulmonate arachnids (Trigonotarbida Araneae Ambly-pygi Uropygi Schizomida) Bulletin of the British Arachno-logical Society 9 267ndash273

Dunlop JA 1996 Evidence for a sister group relationshipbetween Ricinulei and Trigonotarbida Bulletin of the BritishArachnological Society 10 193ndash204

Dunlop JA 1997 Palaeozoic arachnids and their significancefor arachnid phylogeny In Zabka M ed Proceedings of the16th European colloquium of arachnology Siedlce PolandWyzsza Szkola Rolnicko-Pedagogiczna 65ndash82

Dunlop JA 1998 The origins of tetrapulmonate book lungsand their significance for chelicerate phylogeny In SeldenPA ed Proceedings of the 17th European Colloquium ofArachnology Edinburgh 1997 Burnham Beeches BritishArachnological Society 9ndash16

Dunlop JA 1999 A redescription of the Carboniferous arach-nid Plesiosiro madeleyi Pocock 1911 (Arachnida Hapto-poda) Transactions of the Royal Society of Edinburgh EarthSciences 90 29ndash47

Dunlop JA 2000 The epistomo-labral plate and lateral lips insolifuges pseudoscorpions and mites Ekoloacutegia (Bratislava)19 67ndash78

Dunlop JA 2002a Arthropods from the Lower Devonian Sev-ernaya Zemlya Formation of October Revolution Island(Russia) Geodiversitas 24 349ndash379

Dunlop JA 2002b Character states and evolution of the che-licerate claws In Toft S Scharff N eds European Arach-nology 2000 Aarhus Aarhus University Press 345ndash354

Dunlop JA 2002c Phylogeny of Chelicerata In LlorenteBousquets J Morrone JJ eds Biodiversidad Taxonomiacutea yBiogeografiacutea de Arthroacutepodos de Meacutexico Hacia Una Siacutentesisde Su Conocimiento Vol III Prensas de Ciencias Univer-sidad Nacional Autonoma de Meacutexico 117ndash141

Dunlop JA Anderson LI Braddy SJ 2004 A redescriptionof Chasmataspis laurencii Caster amp Brooks 1956 (Chelicer-ata Chasmataspidida) from the Middle Ordovician of Ten-nessee USA with remarks on chasmataspid phylogenyTransactions of the Royal Society of Edinburgh Earth Sci-ences 94 207ndash225

Dunlop JA Braddy SJ 1997 Slit-like structures on theprosomal appendages of the eurypterid BaltoeurypterusNeues Jahrbuch fuumlr Geologie und Palaumlontologie-Monatshefte1997 31ndash38

Dunlop JA Braddy SJ 2001 Scorpions and their sister-group relationships In Fet V Selden PA eds Scorpions2001 in Memoriam Gary A Polis London British Arachno-logical Society 1ndash24

Dunlop JA Horrocks CA 19951996 A new Upper Carbon-iferous whip scorpion (Arachnida Uropygi Thelyphonida)

PHYLOGENY OF THE ARACHNID ORDERS 241

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

with a revision of the British Carboniferous Uropygi Zoolo-gischer Anzeiger 234 293ndash306

Dunlop JA Poschmann M Anderson L 2001 On theEmsian (early Devonian) arthropods of the Rhenish SlateMountains 3 The chasmataspidid Diploaspis Palaumlon-tolgische Zeitschrift 75 253ndash269

Dunlop JA Webster M 1999 Fossil evidence terrestrializa-tion and arachnid phylogeny Journal of Arachnology 27 86ndash93

Edgecombe GD Wilson GDF Colgan DJ Gray MR Cas-sis G 2000 Arthropod cladistics combined analysis of his-tone H3 and U2 snRNA sequences and morphologyCladistics 16 155ndash203

Evans GO 1992 Principles of acarology Wallingford UKCABI

Fahrenbach WH 1999 Merostomata In Harrison FW Foe-lix RF eds Microscopic anatomy of invertebrates Vol 8AChelicerate arthropods New York Wiley-Liss 21ndash115

Farley RD 1999 Scorpiones In Harrison FW Foelix RF edsMicroscopic anatomy of invertebrates Vol 8A Cheliceratearthropods New York Wiley-Liss 117ndash222

Farley RD 2005 Developmental changes in the embryo pro-nymph and first molt of the scorpion Centruroides vittatus(Scorpiones Buthidae) Journal of Morphology 265 1ndash27

Fayers SR Dunlop JA Trewin NH 2004 A new earlyDevonian trigonotarbid arachnid from the Windyfield ChertRhynie Scotland Journal of Systematic Palaeontology 2269ndash284

Felsenstein J 1985 Confidence limits on phylogenies anapproach using the bootstrap Evolution 39 783ndash791

Fet V Sissom WD Lowe G Braunwalder ME 2000 Cat-alog of the scorpions of the world (1758ndash1998) New YorkNew York Entomological Society

Firstman B 1954 The central nervous system musculatureand segmentation of the cephalothorax of a tarantula (Eu-rypelma californicum Ausserer) Microentomology 87 1ndash40

Firstman B 1973 The relationship of the chelicerate arterialsystem to the evolution of the endosternite Journal ofArachnology 1 1ndash54

Foelix RF 1985 Mechano- and chemoreceptive sensilla InBarth FG ed Neurobiology of arachnids Berlin Spinger-Verlag 118ndash137

Foelix RF 1996 Biology of spiders 2nd edn Oxford OxfordUniversity Press

Giribet G Edgecombe GD Wheeler WC Babbit C 2002Phylogeny and systematic position of Opiliones a combinedanalysis of chelicerate relationships using morphologicaland molecular data Cladistics 18 5ndash70

Goloboff P 1993 Estimating character weights during treesearch Cladistics 9 83ndash91

Goloboff PA Farris JS Nixon K 2000 TNT Tree AnalysisUsing New Technologies Version 1 Program and documen-tation httpwwwzmucdkpublicphylogenytnt

Grainge CA Pearson RG 1966 Cuticular structure in thePhalangida Nature 211 866

Grandjean F 1938 Observations sur les Bdelles (Acariens)Annales de la Socieacuteteacute Entomologique de France Paris 107 1ndash24

Hansen HJ Soslashrensen W 1904 On two orders of ArachnidaCambridge Cambridge University Press

Hansen HJ Soslashrensen W 1905 The Tartarides a tribe of theorder Pedipalpi Arkiv fuumlr Zoologi 2 1ndash78

Hara MR Gnaspini P 2003 Comparative study of the defen-sive behavior and morphology of the gland opening areaamong harvestmen (Arachnida Opiliones Gonyleptidae)under a phylogenetic perspective Arthropod Structure andDevelopment 32 257ndash275

Harvey MS 1992 The phylogeny and classification of thePseudoscorpionida (Chelicerata Arachnida) InvertebrateTaxonomy 6 1373ndash1435

Harvey MS 2002 The neglected cousins what do we knowabout the smaller arachnid orders Journal of Arachnology30 357ndash372

Haupt J Song D 1996 Revision of East Asian whip scorpions(Arachnida Uropygi Theyphonida) I China and JapanArthropoda Selecta 5 43ndash52

Hjelle JT 1990 Anatomy and morphology In Polis GA edThe biology of scorpions Stanford University Stanford Uni-versity Press 9ndash63

Janczyk FSW 1956 Anatomie von Siro duricorius Joseph imVergleich mit anderen Opilioniden Sitzungsberichte Oster-reichischen Akademie der Wissenschaften Mathematisch-Natruwissenschaftliche Klasse 165 474ndash522

Jenner RA 2001 Bilaterian phylogeny and uncritical recy-cling of morphological data sets Systematic Biology 50 730ndash742

Jeram AJ 1998 Phylogeny classification and evolution of Sil-urian and Devonian scorpions In Selden PA ed Proceed-ings of the 17th European Colloquium of ArachnologyEdinburgh 1997 Burnham Beeches UK British Arachno-logical Society 15ndash31

Karaman IM 2005 Evidence of spermatophores in Cyphoph-thalmi (Arachnida Opiliones) Revue Suisse de Zoologie 1123ndash11

Kaumlstner A 1935 Die Funktion der sogenannten sympathis-chen Ganglien und die Exkretion bei den PhalangiidenZoologischer Anzeiger 109 273ndash288

Kierans JE Robbins RG 1999 A world checklist of generasubgenera and species of ticks (Acari Ixodida) publishedfrom 1973 to 1997 Journal of Vector Ecology 24 115ndash129

Kjellesvig-Waering EN 1986 A restudy of the fossil Scorpi-onida of the world Palaeontographica Americana 55 1ndash287

Klompen JSH 2000 Prelarva and larva of Opilioacarus (Neo-carus) texanus (Chamberlin and Mulaik) (Acari Opilioacar-ida) with notes on the patterns of setation and lyrifissuresJournal of Natural History 34 1977ndash1992

Kraus O 1998 Elucidating the historical process of phylog-eny phylogenetic systematics versus cladistic techniquesIn Selden PA ed Proceedings of the 17th European Collo-quium of Arachnology Edinburgh 1997 Burnham BeechesUK British Arachnological Society 1ndash7

Kraus O Kraus M 1993 Divergent transformation of cheli-cerae and original arrangement of eyes in spiders (Arach-nida Araneae) Memoirs of the Queensland Museum 33579ndash584

Lankester ER Benham WBS Beck EJ 1885 On the mus-

242 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

cular and endoskeletal systems of Limulus and Scorpio withsome notes on the anatomy and generic characters of scor-pions Transactions of the Zoological Society of London 11311ndash384

Leclerc P 1989 Considerations paleobiogeographiques a pro-pos la decouverte en Thailande drsquoopilioacariens nouveaux(Acari ndash Notostigmata) Compte Rendu des Seances de laSociete de Biogeographie 65 162ndash174

Legg G 1976 The external morphology of a new species ofricinuleid (Arachnida) from Sierra Leone Zoological Journalof the Linnean Society 59 1ndash58

Lindquist EE 1984 Current theories on the evolution ofmajor groups of Acari and on their relationships with othergroups of Arachnida with consequent implications for theirclassificaton In Griffiths DA Bowman CE eds AcarologyVI Vol 1 New York John Wiley amp Sons 28ndash62

Manning PL Dunlop JA 1995 The respiratory organs ofeurypterids Palaeontology 38 287ndash297

Manton SM 1958 Hydrostatic pressure and leg exten-sion in arthropods with special reference to arachnidsAnnals and Magazine of Natural History Series 13 1161ndash182

Manton SM 1964 Mandibular mechanisms and the evolutionof arthropods Philosophical Transactions of the Royal Soci-ety of London B 7 1ndash183

Maraun M Heethoff M Schneider K Scheu S WeigmannG Cianciolo J Thomas RH Norton RA 2004 Molecularphylogeny of oribatid mites (Oribatida Acari) evidence formultiple radiations of parthenogenetic lineages Experimen-tal and Applied Acarology 33 183ndash201

Marples BJ 1967 The spinnerets and epiandrous glands ofspiders Journal of the Linnean Society (Zoology) 46 209ndash222

Marples BJ 1983 Observations on the structure of the fore-gut of spiders Bulletin of the British Arachnological Society8 46ndash52

Millot J 1933 Notes compleacutementaires sur lrsquoanatomie desliphistiides et des hypchilides a propos drsquoun travail reacutecent deA Petrunkevitch Bulletin de la Socieacuteteacute Zoologique de France68 217ndash235

Millot J 1942 Sur lrsquoanatomie et lrsquohistophysiologie deKoenenia mirabilis Grassi (Arachnida Palpigradi) RevueFranccedilaise drsquoEntomologie 9 33ndash51

Millot J 1943 Notes complementaires sur lrsquoanatomielrsquohistologie et la repartition geographique en France deKoenenia mirabilis Grassi Revue Franccedilaise drsquoEntomologie 9127ndash135

Millot J 1945 Lrsquoanatomie interne des Ricinulei (Arachnides)Annales des Sciences Naturelles (Zoologie) Series 11 7 1ndash29

Millot J 1949a Morphologie generale et anatomie interne InGrasseacute P-P ed Traite de zoologie VI Paris Masson et Cie263ndash319

Millot J 1949b Ordre des Amblypyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 563ndash588

Millot J 1949c Ordre des Araneacuteides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 589ndash743

Millot J 1949d Ordre des Palpigrades In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 520ndash532

Millot J 1949e Ordre des Uropyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 533ndash562

Millot J 1949f Ordre des Ricinuleides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 744ndash760

Millot J Vachon M 1949 Ordre Des Solifuges In Grasseacute P-P ed Traite de zoologie VI Paris Masson et Cie 482ndash519

Moore RA Briggs DEG Bartels C 2005 A new specimen ofWeinbergina opitzi (Chelicerata Xiphosura) from the LowerDevonian Huumlnsruck Slate Germany PalaumlontologischeZeitschrift 79 399ndash408

Moritz M 1957 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) unter besonderer Beruumlcksichtigungder ausseren Morphologie der Bildung des Mitteldarmesund der Genitalanlage Zoologische Jahrbuumlcher Abteilungfuumlr Anatomie und Ontogenie der Tiere 76 331ndash370

Moritz M 1959 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) II Die Anlage und Entwicklung derCoxaldruumlse bei Phalangium opilio L Zoologische Jahr-buumlcher Abteilung fuumlr Anatomie und Ontogenie der Tiere 77229ndash240

Nixon K 1999 The parsimony ratchet a new method forrapid parsimony analysis Cladistics 15 407ndash414

Norton RA 1998 Morphological evidence for the evolutionaryorigin of Astigmata (Acari Acariformes) Experimental andApplied Acarology 22 559ndash594

Palmgren P 1978 On the muscular anatomy of spiders ActaZoologica Fennica 155 1ndash41

Patten W 1912 The evolution of the vertebrates and their kinPhiladelphia P Blakistonrsquos Son

Patten W Hazen AP 1900 The development of the coxalgland branchial cartilages and genital ducts of Limuluspolyphemus Journal of Morphology 16 459ndash490

Paulus HF 1979 Eye structure and the monophyly of theArthropoda In Gupta AP ed Arthropod phylogeny NewYork Van Nostrand Reinhold 299ndash383

Petrunkevitch A 1949 A study of Palaeozoic ArachnidaTransactions of the Connecticut Academy of Arts and Sci-ences 37 69ndash315

Pittard K Mitchell RW 1972 Comparative morphology ofthe life stages of Cryptocellus pelaezi (Arachnida Ricinulei)Graduate Studies Texas Tech University 1 1ndash77

Platnick NI 2005 The world spider catalog Version55 American Museum of Natural History httpresearchamnhorgentomologyspiderscatalogindexhtm

Platnick NI Gertsch WJ 1976 The suborders of spiders acladistic analysis (Arachnida Araneae) American MuseumNovitates 2607 1ndash15

Pocock RI 1893 On some points in the morphology of theArachnida (ss) with notes on the classification of the groupAnnals and Magazine of Natural History Series 6 11 1ndash19

Pocock RI 1902 Some points in the morphology and classi-fication of the Opiliones Annals and Magazine of NaturalHistory Series 7 10 504ndash516

Popadicccc A Panganiban G Rusch D Shear WA KaufmanTC 1998 Molecular evidence for the gnathobasic derivationof arthropod mandibles and for the appendicular origin ofthe labrum and other structures Development Genes andEvolution 208 142ndash150

PHYLOGENY OF THE ARACHNID ORDERS 243

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Raw F 1957 Origin of chelicerates Journal of Paleontology31 139ndash192

Reddell JR Cokendolpher JC 1995 Catalogue bibliogra-phy and generic revision of the order Schizomida (Arach-nida) Texas Memorial Museum Speleological Monographs4 1ndash170

Reissland A Goumlrner P 1985 Trichobothria In Barth FGed Neurobiology of arachnids Berlin Springer-Verlag 138ndash161

Roewer C-F 1934 Solifugae Palpigradi Bronns Klassen undOrdnungen ses Tierreich 5 IV 4 1ndash713

Rowland JM Cooke JAL 1973 Systematics of the arachnidorder Uropygida (=Thelyphonida) Journal of Arachnology 155ndash71

Rowland JM Sissom WD 1980 Report on a fossil palpi-grade from the Tertiary of Arizona and a review of themorphology and systematics of the order (ArachnidaPalpigradida) Journal of Arachnology 8 69ndash86

Rucker A 1901 The Texan Koenenia American Naturalist35 615ndash630

Ruhland M Rathmayer W 1978 Die Beinmuskulatur undihre Innervation bei der Vogelspinne Dugesiella henzti (Ch)(Araneae Aviculariidae) Zoomorphologie 89 33ndash46

Saboori A Kamali K 2000 Description of Allothrombiumtriticium adult (Acari Trombidiidae) from Iran Systematicand Applied Acarology 5 207ndash208

Schliwa M 1979 The retina of the phalangid Opilio raven-nae with particular reference to arhabdomeric cells Celland Tissue Research 204 473ndash495

Scholl G 1977 Beitrage zur Embryonalentwicklung von Lim-ulus polyphemus L (Chelicerata Xiphosura) Zoomorpholo-gie 86 99ndash154

Scholtz G Kamenz C 2006 The book lungs of Scorpionesand Tetrapulmonata (Chelicerata Arachnida) evidence forhomology and a single terrestrialization event of a commonarachnid ancestor Zoology 109 2ndash13

Seitz KA 1972 Zur Histologie und Feinstruktur des Herzensund der Haumlmocyten von Cupiennius salei Keys (AraneaeCtenidae) I Herzwandung Bildung und Differenzierungder Haumlmocyten Zoologische Jahrbuumlcher Abteilung fuumlr Anat-omie und Ontogenie der Tiere 89 351ndash384

Sekiguchi K ed 1988 Biology of horseshoe crabs Tokyo Sci-ence House

Selden PA 1981 Functional morphology of the prosoma ofBaltoeurypterus tetragonophthalmus (Fischer) (ChelicerataEurypterida) Transactions of the Royal Society of Edin-burgh Earth Sciences 72 9ndash48

Selden P 1992 Revision of the fossil ricinuleids Transactionsof the Royal Society of Edinburgh Earth Sciences 83 595ndash634

Selden PA 1993 Fossil arachnids ndash recent advances andfuture prospects Memoirs of the Queensland Museum 33389ndash400

Selden PA Dunlop JA 1998 Fossil taxa and relationshipsof chelicerates In Edgecombe GD ed Arthropod fossilsand phylogeny New York Columbia University Press 303ndash331

Selden PA Jeram AJ 1989 Palaeophysiology of terrestrial-

isation in the Chelicerata Transactions of the Royal Societyof Edinburgh Earth Sciences 80 303ndash310

Selden PA Shear WA Bonamo PM 1991 A spider and otherarachnids from the Devonian of New York and reinterpre-tations of Devonian Araneae Palaeontology 34 241ndash281

Sensenig A Shultz JW 2003 Mechanics of cuticular elasticenergy storage in leg joints lacking extensor muscles inarachnids Journal of Experimental Biology 206 771ndash784

Shear WA 1993 New species in the opilionid genus Stylocel-lus from Malaysia Indonesia and the Philippines (OpilionesCyphophthalmi Stylocellidae) Bulletin of the British Arach-nological Society 9 174ndash188

Shear WA Selden PA Rolfe WDI Bonamo PM GriersonJD 1987 New terrestrial arachnids from the Devonian ofGilboa New York (Arachnida Trigonotarbida) AmericanMuseum Novitates 2901 1ndash74

Shultz JW 1987 The origin of the spinning apparatus in spi-ders Biological Reviews of the Cambridge PhilosophicalSociety 62 89ndash113

Shultz JW 1989 Morphology of locomotor appendages inArachnida evolutionary trends and phylogenetic implica-tions Zoological Journal of the Linnean Society 97 1ndash56

Shultz JW 1990 Evolutionary morphology and phylogeny ofArachnida Cladistics 6 1ndash31

Shultz JW 1991 Evolution of locomotion in Arachnida thehydraulic pressure pump of the giant whipscorpion Masti-goproctus giganteus (Uropygi) Journal of Morphology 21013ndash31

Shultz JW 1993 Muscular anatomy of the giant whipscor-pion Mastigoproctus giganteus (Arachnida Uropygi) and itsevolutionary significance Zoological Journal of the LinneanSociety 108 335ndash365

Shultz JW 1998 Phylogeny of Opiliones (Arachnida) anassessment of the lsquoCyphopalpatoresrsquo concept Journal ofArachnology 26 257ndash272

Shultz JW 1999 Muscular anatomy of a whipspider Phrynuslongipes (Amblypygi) and its evolutionary significance Zoo-logical Journal of the Linnean Society 126 81ndash116

Shultz JW 2000 Skeletomuscular anatomy of the harvest-man Leiobunum aldrichi (Weed 1893) (Arachnida Opil-iones Palpatores) and its evolutionary significanceZoological Journal of the Linnean Society 128 401ndash438

Shultz JW 2001 Gross muscular anatomy of Limuluspolyphemus (Chelicerata Xiphosura) and its bearing on evo-lution in the Arachnida Journal of Arachnology 29 283ndash303

Shultz JW 2007 Morphology of the prosomal exoskeleton ofScorpiones (Arachnida) and a new hypothesis for the evolu-tion of cuticular cephalic endoskeletons in arthropodsArthropod Structure amp Development 36 77ndash102

Shultz JW Pinto da Rocha R 2007 Morphology and func-tional anatomy In Pinto da Rocha R Machado G Giribet Geds The harvestmen the biology of Opiliones CambridgeMA Harvard University Press 14ndash61

Shultz JW Regier JC 2001 Phylogenetic analysis of Pha-langida (Arachnida Opiliones) using two nuclear protein-encoding genes supports monophyly of Palpatores Journalof Arachnology 29 189ndash200

Sitnikova LG 1978 The main evolutionary trends of the

244 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Acari and the problem of their monophyletism Entomologis-cheskoe Obozrenie 57 431ndash457

Snodgrass RE 1948 The feeding organs of Arachnidaincluding mites and ticks Smithsonian Miscellaneous Col-lections 110 (10) 1ndash94

Snodgrass RE 1952 A textbook of arthropod anatomy Ith-aca NY Comstock Publishing

Soslashrensen W 1879 Om Bygningen af Gonyleptiderne en Typeaf Arachnidernes Classe Naturhistorisk Tidsskrifter Series3 8 97ndash222

Stockwell SA 1989 Revision of the phylogeny and higherclassification of scorpions (Chelicerata) Unpublished doc-toral dissertation University of California at Berkeley

Stoslashrmer L 1944 On the relationships and phylogeny of fossiland Recent Arachnomorpha Skrifter utgitt av Det NorskeVidenskaps-Akademi i Oslo I 5 1ndash158

Stoslashrmer L 1955 Merostomata In Moore RC ed Treatise oninvertebrate paleontology Part P Arthropoda 2 LawrenceKS University of Kansas Press and Geological Society ofAmerica 4ndash41

Swofford DL 2002 PAUP Phylogenetic analysis using par-simony (and other methods) Version Beta 10 SunderlandMA Sinauer Associates

Talarico G Palacios-Vargas JG Fuentes Silva M AlbertiG 2005 First ultrastructural observations of the tarsal poreorgan of Pseudocellus pearsei and P boneti (ArachnidaRicinulei) Journal of Arachnology 33 604ndash612

Templeton AR 1983 Phylogenetic inference from restrictionendonuclease cleavage site maps with articular reference tothe evolution of humans and the apes Evolution 37 221ndash244

Tetlie OE Braddy SJ 2004 The first Silurian chasmataspidLoganamaraspis dunlopi gen et sp nov (Chelicerata Chas-mataspidida) from Lesmahagow Scotland and its implica-tions for eurypterid phylogeny Transactions of the RoyalSociety of Edinburgh Earth Sciences 94 227ndash234

Thomas RH Telford MJ 1999 Appendage development inembryos of the oribatid mite Archegozetes longisetosus(Acari Oribatei Trhypochthoniidae) Acta Zoologica (Stock-holm) 80 193ndash200

Tollerton VP Jr 1989 Morphology taxonomy and classifica-tion of the order Eurypterida Burmeister 1843 Journal ofPaleontology 63 642ndash657

Vachon M 1949 Ordre des Pseudoscorpions In Grasseacute P-Ped Traite de zoologie VI Paris Masson et Cie 436ndash481

Van der Hammen L 1979 Comparative studies in Chelicer-ata I The Cryptognomae (Ricinulei Architarbi and Anacti-notrichida) Zoologische Verhandelingen 174 1ndash62

Van der Hammen L 1982 Comparative studies in Chelicer-ata II Epimerata (Palpigradi and Actinotrichida) Zoolo-gische Verhandelingen 196 1ndash70

Van der Hammen L 1985 Functional morphology and affin-ities of extant Chelicerata in evolutionary perspectiveTransactions of the Royal Society of Edinburgh 76 137ndash146

Van der Hammen L 1989 An introduction to comparativearachnology Leiden SPB Academic Publishing

Vyas AB 1970 Studies in myology of the scorpion Heter-ometrus fulvipes Koch Unpublished PhD DissertationDepartment of Zoology Gujarat University India

Vyas AB 1974 The cheliceral muscles of the scorpion Heter-ometrus fulvipes Bulletin Southern California Academy ofSciences 73 9ndash14

Walter DE 1988 Predation and mycophagy by endeostigma-tid mites (Acariformes Prostigmata) Experimental andApplied Acarology 4 159ndash166

Walter DE Proctor HC 1998 Feeding behaviour and phy-logeny observations on early derivative Acari Experimentaland Applied Acarology 22 39ndash50

Walter DE Proctor HC 1999 Mites ecology evolution andbehaviour Sydney University of New South Wales

Waterston CD 1979 Problems of functional morphology andclassification in stylonuroid eurypterids (Chelierata Meros-tomata) with observations on the Scottish Silurian Sty-lonuroidea Transactions of the Royal Society of EdinburghEarth Science 70 251ndash322

Weygoldt P 1969 Biology of pseudoscorpions CambridgeMA Harvard University Press

Weygoldt P 1984 Lrsquoautotomie chez les Amblypyges RevueArachnologique 5 321ndash327

Weygoldt P 1996 Evolutionary morphology of whip spiderstowards a phylogenetic system (Chelicerata ArachnidaAmblypygi) Journal of Zoological Systematics and Olution-ary Research 34 185ndash202

Weygoldt P 1998 Evolution and systematics of the Chelicer-ata Experimental and Applied Acarology 22 63ndash79

Weygoldt P 2000 Whip spiders (Chelicerata Amblypygi)their biology morphology and systematics StenstrupDenmark Apollo Books

Weygoldt P Paulus HF 1979 Untersuchungen zur Morphol-ogie Taxonomie und Phylogenie der Chelicerata Zeitschriftfuumlr Zoologische Systematik und Evolutionsforschung 17 85ndash116 177ndash200

Wheeler WC Hayashi CY 1998 The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192

Winkler D 1957 Die Entwicklung der aumlusseren Koumlrperge-stalt bei den Phalangiidae (Opiliones) Mitteilungen aus demZoologischen Museum in Berlin 33 355ndash389

With CJ 1904 The Notostigmata a new suborder of AcariVidenskabelige Meddelelser fra Dansk Naturhistorisk Foren-ingi Kjobenhaus 1904 137ndash192

Yamasaki T Makioka T Saito J 1988 Morphology InSekiguchi K ed Biology of horseshoe crabs Tokyo ScienceHouse 69ndash132

Yoshikura M 1975 Comparative embryology and phylogeny ofArachnida Kumamoto Journal of Science Biology 12 71ndash142

APPENDIX

Each character used in this analysis is numbereddefined and cross-referenced with characters used infour previous studies that are abbreviated in bracketsspecifically WP = Weygoldt amp Paulus (1979)S = Shultz (1990) WH = Wheeler amp Hayashi (1998)GEWB = Giribet et al (2002) Descriptions includecorrections of errors discussions of controversies orambiguities and justifications for state assignmentsto lsquoproblemrsquo taxa

PHYLOGENY OF THE ARACHNID ORDERS 245

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

PROSOMA

1 Dorsal sclerite formed by fusion of the prosomal car-apace the dorsal portion of the first opisthosomalsomite and the dorsomedial (axial) portion of the sec-ond opisthosomal somite 0 absent 1 presentState 1 is present in extant xiphosurids and close fos-sil relatives (Scholl 1977 Anderson amp Selden 1997Shultz 2001) The fossil record suggests that thisstructure originated with the disappearance ofthe already reduced first opisthosomal tergite(= microtergite) and axial portion of the secondopisthosomal tergite (Anderson amp Selden 1997)2 Single dorsal sclerite covering entire dorsal surfaceof body no lines indicating original segmentation 0absent 1 presentState 1 occurs throughout Holothyrida (Acari) (Vander Hammen 1989) and in many Mesostigmata(Alberti amp Coons 1999) including Glyptholaspis (Vander Hammen 1989)3 Anterior end of dorsal prosoma with medianmarginal or submarginal pointed process 0 absent1 presentState 1 occurs throughout Schizomida (Protoschizo-mus Stenochrus Reddell amp Cokendolpher 1995) The-lyphonida (Rowland amp Cooke 1973 Shultz unpublobs Proschizomus Dunlop amp Horrocks 19951996)Amblypygi (Charinus Millot 1949b Phrynus Shultz1999) and Plesiosiro (Dunlop 1999) Citing a figure inRowland amp Cooke (1973) Dunlop amp Horrocks assumedthat the anterior process was limited to lsquohypoctonidrsquothelyphonids and united them with Proschizomus andSchizomida If valid this would make Thelyphonidaparaphyletic However the doubtful monophyly oflsquohypoctonidsrsquo (Haupt amp Song 1996) and widespreadpresence of a submarginal anterior process in Pedi-palpi weaken this proposal (Harvey 2002)4 Ophthalmic ridges pair of longitudinal crests inter-secting or passing near the region of the lateral eyes orcomparable region where lateral eyes are absent 0absent 1 presentState 1 is present throughout Xiphosura (Anderson ampSelden 1997) Similar structures are present in Plesi-osiro (Dunlop 1999) and lsquonon-hypoctonidrsquo Thely-phonida (Mastigoproctus Rowland amp Cooke 1973)Dunlop (2002a) described apparent ophthalmic ridgesin the chasmataspidid Octoberaspis but these werenot observed in Diploaspis (Dunlop et al 2001) eitherdue to absence or inadequate preservation of speci-mens nor in Chasmataspis (Dunlop et al 2004)5 Carapace with demarcations (eg grooves scleritesphragmata) between pro- meso- or metapeltidia 0absent 1 present [S 1 WH 48 GEWB 29]State 1 is present in extant Scorpiones Cyphoph-thalmi (Hansen amp Soslashrensen 1904) and many Palpa-tores (Opiliones) including Caddo and Leiobunum

and most Pseudoscorpiones (Chamberlin 1931) Giri-bet et al (2002) coded this character as absent inCyphophthalmi (Opiliones) but a transverseprocurved groove is present throughout the group(Hansen amp Soslashrensen 1904) and represents the meso-metapeltidial border as indicated by attachment ofthe dorsal endosternal suspensor of postoral somite VIin Chileogovea oedipus and Siro exilis (Shultz unpublobserv) Kjellesvig-Waering (1986) did not explicitlydescribe such grooves in fossil scorpions although heillustrated corresponding structures in PrearcturusArchaeophonus Archaeoctonus Palaeophonus andother Palaeozoic scorpions (Kjellesvig-Waering 1986figs 14A 21E 27D 59) (coded as uncertain here) butnot in Proscorpius (contra Giribet et al 2002) Dis-tinct carapacal sclerites are present throughout Palpi-gradi Schizomida and Solifugae (6)

The condition in Acari is controversial but is codedhere as State 1 Van der Hammen (1989) consideredthe tergal region of the leg-bearing somites (postoralsomites IIIndashVI) or podosoma to be reduced andreplaced by posterior migration of the aspidosoma (ietergal region assumed to be associated with append-ages of the gnathosoma) and anterior migration of theopisthosomal tergal region to form the hysterosomaThe dorsal proterosomal and hysterosomal elementsmeet at a transverse sejugal furrow that continues lat-erally and passes ventrally between the coxae of legs 2and 3 According to this scheme the prosomandashopisthosoma border is expressed as the disjugal fur-row which passes from the ventral posterior margin ofthe podosoma anterodorsally to join the dorsal part ofthe sejugal furrow Based on the arrangement of setaeand slit sensilla (= lyrifissures) Van der Hammen(1989) interpreted the region above coxae 3 and 4 (ieregion C) as a fusion of the dorsal parts of the first twoopisthosomal somites (= postoral somites VII and VIII)in early divergent Anactinotrichida (ie Opilioacari-formes) and Acariformes (eg Alycus) These interpre-tations have been followed by many acarologistsalthough its speculative aspects are widely acknowl-edged (Evans 1992 Alberti amp Coons 1999)

From the standpoint of a general arachnologist theVan der Hammen system seems unnecessarily compli-cated it appears to have been formulated to explainbroad morphological themes in Oribatida but was thenextrapolated to other mites A more conservativelsquoarachnologicalrsquo scheme adopted here equates the pro-podosoma with the pars cephalica (= propeltidium) ofthe arachnid prosoma and region C with the pars tho-racica (= mesopeltidium + metapetidium) In fact theearliest and most recent treatments of Opilioacari-formes (With 1904 Klompen 2000) reached similarconclusions This interpretation is also consistent withevidence from Acariformes For example the regiondorsal to the coxa of legs 3 and 4 in Alycus is as readily

246 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

explained by the persistence of primitive prosomal ter-gal elements as by the supposition that these disap-peared and were replaced by two opisthosomalelements Furthermore several diverse lineages ofbasally divergent Acariformes display features that donot fit readily within the GrandjeanndashVan der Ham-men system but seem to correspond to the cephalicathoracica division of the prosoma They have a sejugalfurrow that may have demarcated the cephalicatho-racica border and a postpodosomal furrow thatappears to represent the prosomandashopisthosoma junc-tion there is no disjugal furrow Examples includepalaeosomate (especially aphelacarid) and pediculoch-elid Oribatida alicorhagiid and mircopsammidlsquoEndeostigmatarsquo and paratydeid Prostigmata (Albertiamp Coons 1999 fig 42) This organization may be sym-plesiomorphic for Acariformes but could be a conver-gence brought about by selection for enhancedflexibility6 Carapace with distinct pro- meso- or metapeltidialsclerites 0 absent 1 present - inapplicable due toabsence of pro- meso- or metapeltidial demarcations(5)State 1 occurs throughout Palpigradi Schizomida andSolifugae It also appears sporadically within Opil-iones (eg Leiobunum flavum Shultz unpubl observ)and Pseudoscorpiones (eg Pseudochiridium) (Cham-berlin 1931)7 Sejugal furrow circumferential zone of body flexi-bility that passes between the coxae of legs 2 and 3 0absent 1 presentState 1 is probably synapomorphic for Acariformes asit occurs in most lsquoendeostigmatidsrsquo including Alycus(Van der Hammen 1989) and is variously developedthroughout Sarcoptiformes (Alberti amp Coons 1999) Itis probably the primitive condition for Prostigmatabut is either absent or weakly expressed in represen-tatives included here See 5 for alternative morpholog-ical interpretations of this body region in Acari8 Prosomal ozopores 0 absent 1 present [WH 46GEWB 12]State 1 occurs throughout Opiliones and Holothyrida(Van der Hammen 1989)9 Carapacal pleural doublure 0 absent 1 present[WP 12 S 2 WH 49 GEWB 27]A carapacal pleural doublure occurs throughoutXiphosura (Stoslashrmer 1944) Eurypterida (Clarke ampRuedemann 1912 Selden 1981) and Chasmataspid-ida (Dunlop et al 2001 2004 Dunlop 2002a) and isabsent in all known Arachnida10 Cardiac lobe a longitudinal axial elevation of thecarapace 0 absent 1 presentState 1 occurs throughout Xiphosura (Anderson ampSelden 1997) and Eurypterida (Clarke amp Ruedemann1912 Selden 1981) Carapaces are not preserved suf-ficiently to determine this condition in the chasmatas-

pidids Diploaspis (Dunlop et al 2001) andOctoberaspis (Dunlop 2002a) but an apparent cardiaclobe occurs in the Chasmataspis (Dunlop et al 2004)A cardiac lobe is absent in Arachnida11 Moveable cucullus 0 absent 1 present [WP 54GEWB 22]State 1 is a synapomorphy of Ricinulei (Selden 1992)12 Medial intercoxal lsquosternalrsquo region 0 all pedalcoxae separated medially 1 anterior pedal coxae abut-ting medially posterior coxae separated 2 anteriorpedal coxae separated medially posterior coxae abut-ting 3 all pedal coxae abutting medially 4 epimeracoxae undifferentiated medially from ventral body wall[S sim3 WH sim50 GEWB sim17]Prosomal sternites appear to be those portions of theventral body wall not occupied by coxae Despite earlyattempts to assess metamerism of the ventralprosoma (protosternum deutosternum tritosternumetc) (Boumlrner 1902a Millot 1949a) no apparent mor-phological features such as borders of sclerites ormuscle attachments reliably demarcate the ventralbody wall of one somite from that of an adjacentsomite Thus coding schemes that focus on describingmetameric components of the sternal region are prob-lematic Shultz (1990) attempted to code the shape ofthe entire intercoxal region and subsequent workershave adopted this approach All pedal coxae are sepa-rated medially in Plesiosiro (Dunlop 1999) Palpigradi(Roewer 1934) Araneae (Millot 1949c) Amblypygi(Weygoldt 2000) and many Acari including Opilioac-ariformes Holothryrida Mesostigmata Ixodida Aly-cus and many Prostigmata (Van der Hammen 1989Evans 1992 Alberti amp Coons 1999) The anteriorcoxae are separated in extant Thelyphonida Schizo-mida (Hansen amp Soslashrensen 1905 Millot 1949e) andRicinulei (Pittard amp Mitchell 1972 Legg 1976) Thecoxae are fused to the ventral body wall in Sarcopti-formes (Van der Hammen 1989 Evans 1992)

Interpreting this character is complicated for theanterior coxae of Scorpiones and Opiliones and allcoxae in Xiphosura and Eurypterida due to the pres-ence of coxapophyses endites andor gnathobaseswhich form components of the preoral chambers inthese groups Thus coxae may abut medially whenthese structures are considered part of the coxae butare separate medially if these structures are ignoredAs coxapophyses etc are here coded as separate char-acters (50ndash53) I have chosen to code this character asif these structures did not exist Consequently pha-langid Opiliones are coded as having all coxae sepa-rated (Pocock 1902 Hansen amp Soslashrensen 1904) Fossilscorpions also show substantial variation in this char-acter All pairs of pedal coxae in Palaeoscorpius abutmedially but also have gnathobases (Jeram 1998)and this condition is coded here as uncertain A ster-num separates all pairs of pedal coxae medially in

PHYLOGENY OF THE ARACHNID ORDERS

247

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Proscorpius

and

Stoermeroscorpio

(State 0) and

Pre-arcturus

approximates modern scorpions (State 1)(Kjellesvig-Waering 1986 Jeram 1998) The coxalbases of Eurypterida and Xiphosura are separated bya small sternite the endostoma (Selden 1981 Shultz2001)

13 Postoral sternapophysis (

=

tritosternum labium)a cuticular evagination of the ventral body wall poste-riorly adjacent to the palpal coxae forming the poste-rior border of the preoral chamber in some taxa ordisplaced posteriorly by fusion of the palpal coxae inothers 0 absent 1 present

[S

sim

10 WH

sim

56 GEWB19

=

33]State 1 is present throughout Palpigradi AraneaeAmblypygi (Snodgrass 1948) Schizomida (Hansen ampSoslashrensen 1905 Van der Hammen 1989) Thely-phonida (Van der Hammen 1989) extant Ricinulei(Pittard amp Mitchell 1972 Legg 1976) Opilioacari-formes and Mesostigmata (Van der Hammen 1989Alberti amp Coons 1999)

Allothyrus constrictus

is theonly holothyrid mite known to have sternapophysesAn apparent sternapophysis has been described in apalaeocharinid trigonotarbid (Dunlop 1994) Giribet

et al

(2002) miscoded Ixodida and Acariformes as hav-ing a lsquolabium (

=

tritosternum)rsquo The lophognath ofpseudoscorpions (Chamberlin 1931) and lsquolabiumrsquo ofSolifugae (Roewer 1934) also appear to be sternapo-physes that have been incorporated into the ros-trosoma (

32

) The labium of phalangid Opiliones(Pocock 1902 Hansen amp Soslashrensen 1904) is a ster-napophysis but is associated with leg 1 rather thepedipalp (Winkler 1957) A small anterior sclerite inthe fossil scorpions

Proscorpius

Waeringoscorpio

and

Labriscorpio

has been interpreted as a lsquolabiumrsquo homol-ogous with the sternapophyses of other arachnids(Weygoldt 1998) despite absence of any indication ofthe relationship of this sclerite to the mouth and pos-sible alternative interpretations (eg

16

) Giribet

et al

coded this sclerite as a sternapophysis but the char-acter is coded here as uncertain in fossil scorpions

14 Channels on the body surface linking openings ofcoxal organs to preoral chamber 0 absent 1 present

Van der Hammen (1989) described bilaterally pairedcuticular tracts (lsquotaenidiarsquo) connecting orifices of coxalorgans to the preoral chamber in some arachnids andor a ventromedian groove (

=

intercoxal or subcapitulargutter) that leads in turn to the preoral chamber inother taxa Taenidia and intercoxal gutters arepresent in Opilioacariformes (Van der Hammen1989) Holothyrida (Van der Hammen 1989) Mesos-tigmata (

Glyptholapis

Van der Hammen 1989)Ricinulei (Pittard amp Mitchell 1972 Van der Hammen1989) Araneae (

Heptathela

Aphonopelma

Hypochi-lus

Shultz unpubl observ also

Porrhothele

(Mygalo-morphae Dipluridae) Butt amp Taylor 1991

Segestria

(Araneomorphae Segestriidae) Van der Hammen

1989) Amblypygi (Van der Hammen 1989) Schizo-mida (

Stenochrus

Shultz unpubl observ unidenti-fied Hubbardiinae Van der Hammen 1989) andThelyphonida (

Mastigoproctus

Shultz unpublobserv also

Tetrabalius

Van der Hammen 1989) Theconduction of large volumes of fluid from orifices of thecoxal organs to the preoral chamber has been observedin the mygalomorph spider

Porrhothele

(Dipluridae)(Butt amp Taylor 1991) and

Mastigoproctus

(Thely-phonida) (Shultz unpubl observ) The podocephaliccanal (

15

) is present in Acariformes (Acari) (Lindquist1984) and appears to have evolved as an invaginatedsupracoxal channel (Grandjean 1938 G Alberti inEvans 1992) Van der Hammen (1982) could not dis-cern with certainty whether a channel exists in

Eukoenenia

(Palpigradi) but none has been reportedby previous authors and they have not been observedby the present author Palpigradi is coded as havingState 0

15 Podocephalic canal cuticular channel andorduct draining multiple glands and opening nearmouthparts 0 absent 1 present - inapplicable due toabsence of channel (14)

State 1 occurs throughout Acariformes (Acari)(Lindquist 1984)

16 Heavily sclerotized suboral sclerite serving inpart as basal pivot point for coxae of appendages ofpostoral somites IIndashIV (

=

arachnid palp and legs 1 and2) 0 absent 1 present

The sclerite forms the posterior wall of the true mouthand is heavily sclerotized in extant scorpions It is notvisible externally due to the tightly fitting coxae of theanterior prosomal appendages (Shultz 2007) It maycorrespond to the so-called labium (Weygoldt 1998) orlabrum (Kjellesvig-Waering 1986) of certain fossilscorpions (see

13

)

17 Genal angles 0 rounded 1 pointed

Anderson amp Selden (1997) originally coded a pointedgenal angle and genal spine as separate charactersbut they are combined here State 1 is present in manyXiphosura (Anderson amp Selden 1997) and Chas-mataspidida (Dunlop

et al

2001 2004 Dunlop2002a)

A

PPENDAGES

OF

POSTORAL

SOMITE

I

CHELICERAE

18 Cheliceral segmentation 0 three articles 1 twoarticles

[WP

sim

25

+

42 WH

sim

15 GEWB

sim

44]State 0 is present in extant Xiphosura Eurypterida(Clarke amp Ruedemann 1912 Selden 1981) Palpi-gradi (Roewer 1934) Opiliones (Hansen amp Soslashrensen1904) Scorpiones (Hjelle 1990) and throughout non-acariform Acari (Evans 1992) State 1 is presentthroughout known Trigonotarbida Araneae Ambly-pygi Thelyphonida Schizomida Pseudoscorpionesand Solifugae Acariform mites have two unambigu-

248 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ous cheliceral articles but may have a reduced thirdarticle (Van der Hammen 1989 Evans 1992) acari-form mites are thus coded as uncertain Chelicerae arenot known or their condition is unclear in several fos-sil taxa including Xiphosura Chasmataspidida (Dun-lop et al 2001 2004 Dunlop 2002a) Riciniulei(Selden 1992) Plesiosiro (Dunlop 1999) Proschizo-mus (Dunlop amp Horrocks 19951996) and the fossilscorpions Praearcturus and Palaeoscorpius (Kjelles-vig-Waering 1986) Kjellesvig-Waering (1986) consis-tently interpreted chelicerae of fossil scorpions ashaving four articles but this is probably a misinter-pretation of the basal article (see also Stockwell1989) The basal article in extant scorpions consists ofa distal collar of cuticle and a large proximal processand this arrangement would probably appear as twoarticles in compressed specimens In any event thereare at least three cheliceral articles in Stoermeroscor-pio and Proscorpius (Kjellesvig-Waering 1986)19 Terminal cheliceral joint 0 laterally placedbicondylar hinge 1 dorsally placed bicondylar hinge2 ventrally placed bicondylar hinge [WP 25 + 42 Ssim11 + 12 WH sim15 GEWB sim44]State 0 is present in Xiphosura (Shultz 2001)extant Scorpiones (Millot amp Vachon 1949) Opil-iones (Hansen amp Soslashrensen 1904) Palpigradi(Boumlrner 1904 Millot 1949d) Ricinulei (Pittard ampMitchell 1972 Legg 1976) and Eurypterida includ-ing Baltoeurypterus (Selden 1981) and Stylonurus(Clarke amp Ruedemann 1912) State 0 is present inthe fossil scorpions Proscorpius and Stoermeroscor-pio but the state of the chelicerae is unclear in theother fossil scorpions represented here (Kjellesvig-Waering 1986) State 1 is present in Trigonotarbida(Shear et al 1987) mygalomorph Araneae Ambly-pygi Schizomida and Theylphonida (Weygoldt ampPaulus 1979) State 2 is present in Pseudoscorpi-ones (Chamberlin 1931) and Solifugae (Roewer1934) and appears to be a groundplan feature for allmajor groups of mites (eg Van der Hammen 1989Alberti amp Coons 1999) The difference between theorthognathy of mygalomorph Araneae and lsquoplagiog-nathyrsquo mesothele (Liphistius Heptathela) and palae-ocribellate Araneae (Hypochilus) (Kraus amp Kraus1993) is small compared with the variation thatoccurs throughout arachnids and plagiognathy iscoded as State 1 here20 Chelicera articulating with carapace at anterolat-eral pivot 0 absent 1 present [S 13 WH 57 GEWB45]State 1 occurs in Solifugae (Roewer 1934) and Pseu-doscorpiones except Chthonioidea and Feaelloidea(Chamberlin 1931) Giribet et al (2002) coded thecharacter as present in Pseudoscorpiones21 Chelicera pivoting on dorsal protuberance of epis-tome 0 absent 1 present

State 1 appears to be a unique synapomorphy of Opil-iones (Shultz 2000)22 Extrinsic cheliceral muscle arising on carapaceand inserting on dorsal margin of nonbasal cheliceralarticle 0 absent 1 present (Note Characters 22 and23 refer to two different muscles)State 1 is known only in extant Scorpiones (LankesterBenham amp Beck 1885 Vyas 1970 1974 Shultz2007) but may occur in acariform mites if their cheli-cerae are composed of three rather than two articles(Evans 1992) (see 18) Acariformes are coded asuncertain23 Extrinsic cheliceral muscle arising on carapaceand inserting on ventral margin of nonbasal cheliceralarticle 0 absent 1 present [GEWB 252] (Note Char-acters 22 and 23 refer to two different muscles)State 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) somemesostigmatid Acari (Evans 1992) and representa-tive Opiliones particularly Leiobunum (Shultz 2000)Chileogovea and Gonyleptes (Shultz unpubl observ)The status of the character in other opilions is not yetknown The character was originally defined as anextrinsic muscle inserting on the second segment ofthree-segmented chelicerae (Shultz 2000 Giribetet al 2002) but it has been redefined here to make itapplicable to arachnids with two-segmented cheli-cerae24 Extrinsic cheliceral muscles attaching to epistome0 absent 1 presentState 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) but mayoccur in Prostigmata if the sigmoid piece to which thecheliceral protractors attach (Evans 1992) is a modi-fication of the epistome25 Lateral tergocheliceral muscle with three heads 0absent 1 present - coded only for extant tetrapulmo-nates and the palpigrade Eukoenenia homology isunclear in other taxa [GEWB 230]State 1 is present in Pedipalpi (Shultz 1999)26 Cheliceral silk glands 0 absent 1 present [GEWB47]State 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931) The silk gland of spider mites(eg Tetranychus) are associated with the pedipalps(Evans 1992)27 Cheliceral venom glands 0 absent 1 present [WP34 WH 40 GEWB 46]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)28 Cheliceral serrula interior and exterior 0 absent1 presentState 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931 Harvey 1992)29 Cheliceral lsquoflagellumrsquo in male 0 absent 1 present[GEWB 48]

PHYLOGENY OF THE ARACHNID ORDERS 249

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITE II LEG 1 IN NONARACHNIDS AND PALP IN ARACHNIDS

30 Palpal coxae fused ventromedially and formingposterior wall of preoral chamber 0 absent 1 present[WP 27 S 18 WH 17 = 62 GEWB 33 + 63]State 1 occurs in Schizomida Thelyphonida Acari andRicinulei31 Gnathosoma 0 absent 1 present - inapplicablepalpal coxae not fused (30) [WP 56 S 19 WH sim63GEWB sim36 + 37]The gnathosoma is a functional complex comprisingan epistome attached to the dorsal surface of mediallyfused palpal coxae (= infra- or subcapitulum) (see 30)the chelicerae and in some taxa a supracheliceral tec-tum These components are tightly integrated andmove in unison relative to the body The gnathosomahas traditionally been considered a feature of AcariHowever Van der Hammen (1989) argued that a gna-thosoma is present in Ricinulei and this interpreta-tion has received wide acceptance (eg Lindquist1984) However it is not clear that the relationship ofthe chelicerae and subcapitulum is as intimate inRicinulei as in Acari and the ricinuleid subcapitulumdoes not appear to be as moveable In fact the lsquosubca-pitulumrsquo in Ricinulei is similar in many respects to thecondition in Schizomida and Thelyphonida (Hansen ampSoslashrensen 1904) which has never been regarded as agnathosoma Consequently I have coded the conditionof Ricinulei as uncertain32 Rostrosoma long narrow subcylindrical epistomeprojecting anteriorly with base fixed to dorsal surfaceof palpal coxae bordered laterally by lobes projectingfrom palpal coxae ventral wall of preoral chamberformed by anterior element of prosoma (sternapophy-sis) 0 absent 1 present [S 15 WH 59 GEWB 20 + 36]State 1 is present in Pseudoscorpiones and Solifugae(Chamberlin 1931 Roewer 1934 Van der Hammen1989) Dunlop (2000) attempted to homologize com-ponents of the rostrosoma and gnathosoma (31) anargument accepted by Giribet et al (2002) Howeverfusion of the palpal coxae (30) apparently evolvedindependently in the two structures Coxal fusion inthe gnathosoma is complete ventrally and excludesthe suboral sternapophysis (13) while fusion isincomplete ventrally in the rostrosoma and incorpo-rates the sternapophysis Unlike the gnathosomathe rostrosoma is largely immobile and cheliceralmovement is not coupled with that of the palpalcoxae33 Rutellacorniculi hypertrophied setae modified asmouthparts located on the anterior processes of the pal-pal coxae 0 absent 1 present [GEWB 39]

State 1 occurs in the anactinotrichid groups Opilioac-ariformes Holothyrida and Mesostigmata but not Ixo-dida (Van der Hammen 1989) It also occurs inSarcoptiformes (Evans 1992) and certain endeostig-matids (eg Alycus Van der Hammen 1989) but not inProstigmata Comparable structures are unknown inother chelicerate groups34 Terminal segments specially modified in adultmale as a clasper used to engage the female 0 absent1 presentState 1 is known only in extant Xiphosura (YamasakiMakioka amp Saito 1988)35 Robust raptorial 0 absent 1 present [S 17 WH61 GEWB sim59 97]State 1 is present in Amblypygi Thelyphonida andSchizomida (Shear et al 1987) and in Laniatores(Opiliones)36 Orientation of robust raptorial appendage 0operating in subtransverse plane 1 operating in sub-vertical plane - inapplicable coded only for PedipalpiState 0 is present in extant Thelyphonida and mostAmblypygi and State 1 is present in Schizomida (Cok-endolpher amp Reddell 1992) and in the basally diver-gent amblypygid Paracharon caecus (Weygoldt 2000)The orientation of the pedipalps in the fossil Proschi-zomus is uncertain (Dunlop amp Horrocks 19951996)37 Extrinsic muscle attaching to epistome 0 absent1 presentState 1 occurs in Chileogovea and Siro (OpilionesCyphophthalmi) and extant Scorpiones (Vyas 1970Shultz 2007)38 Muscle originating and inserting within coxa 0absent 1 present [GEWB sim232]State 1 occurs is in Amblypygi (Phrynus Shultz 1999Phrynichus Boumlrner 1904) Thelyphonida (Mastigo-proctus Shultz 1993 Thelyphonus Boumlrner 1904) andSchizomida (Stenochrus Shultz unpubl observ)There are two intracoxal muscles in amblypygids andthelyphonids the one coded here and another that isserially homologous with a muscle in the pedal coxae(54)39 Tarsus andor tibia with venom glands 0 absent1 present [GEWB 64]State 1 is known only from extant Pseudoscorpi-ones except the superfamilies Chthonioidea andFeaelloidea (Chamberlin 1931 Harvey 1992) Giri-bet et al (2002) coded this character as present forPseudoscorpiones40 Tarsal grooming organ 0 absent 1 present[GEWB 98]State 1 is a unique synapomorphy of Amblypygi (DelleCave 1975 Weygoldt 2000)41 Scorpionoid chela a large well-developed chelaformed by tibia (manus + fixed finger) and tarsus(moveable finger) 0 absent 1 present [S 16 WH 60GEWB 62]

250 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs throughout Scorpiones andPseudoscorpiones42 Modified in male as copulatory organ 0 absent 1present [WP 35 WH 39 GEWB 67]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)43 Apotele 0 apparently absent not differentiatedexternally from penultimate article 1 present [GEWB69]The apotele exists as a structure (ie dactyl clawempodium) differentiated from the tibiotarsus tarsusor telotarsus and is typically associated with a pair ofantagonistic muscles The apotele occurs as a distinctstructure in Xiphosura (Snodgrass 1948) Eurypter-ida (Clarke amp Ruedemann 1912) Schizomida (Coken-dolpher amp Reddell 1992) Araneae (Foelix 1996)early divergent Amblypygi (Charinus Weygoldt 2000not Phrynus Shultz 1999) Anactinotrichida (CaminClark amp Bourdeau 1956 Lindquist 1984) Ricinulei(Pittard amp Mitchell 1972 Legg 1976) and Solifugae(Roewer 1934) Giribet et al (2002) coded Solifugae asnot having a palpal apotele but it is present as anadhesive organ (45) Dunlop (1999) reconstructedHaptopoda as having a palpal claw but notes in hisdescription that the distal ends were not preserved inthe known specimens44 Apotele (claw) position 0 terminal 1 subtermi-nal - inapplicable coded only for taxa with a distinctapotele (43)State 1 occurs in Parasitiformes (Acari) (Evans 1992)45 Terminal adhesive organ 0 absent 1 present -inapplicable due to lack of an apotele (43) [GEWB 66]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITES IIIndashVI LEGS 2ndash5 IN NON-ARACHNIDS AND LEGS 1ndash4 IN ARACHNIDS

46 Appendage III (= arachnid leg 1) extremely elon-gate antenniform 0 absent 1 present [WP sim24 + 32S 20 WH 14 GEWB sim84]State 1 occurs in Amblypygi Thelyphonida Schizo-mida (Shear et al 1987) and Opilioacariformes (Vander Hammen 1989) State 1 is approximated in Solif-ugae Palpigradi (Roewer 1934) and Haptopoda(Dunlop 1999) and it is coded as uncertain in thesetaxa47 Appendage V (= arachnid leg 3) of male specializedfor sperm transfer 0 absent 1 present [WP 55 GEWB94]State 1 is known only from extant Ricinulei48 Appendages V and VI (= arachnid legs 3 and 4)with femur shorter than patella and with principal siteof flexionextension at patella-tibia joint (lsquoapatellatersquocondition sensu Van der Hammen 1989) 0 absent 1present [S 25 GEWB 86]

State 1 occurs in Pseudoscorpiones and Solifugae(Shultz 1989)49 Appendage VI (= arachnid leg 4) with terminusmodified as flattened blade 0 absent 1 presentState 1 occurs in some Chasmataspidida (DiploapisDunlop et al 2001 Octoberaspis Dunlop 2002a)uncertain in Chasmataspis Dunlop et al 2004 State0 in Loganamaraspis Tetlie amp Braddy 2004) andEurypterida except stylonuroids (Clarke amp Ruede-mann 1912 Waterston 1979 Tollerton 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI COXA AND BODYndashCOXA JOINT

50 Appendage III (= arachnid leg 1) with coxapophy-sis forming floor or wall of preoral chamber 0 absent1 present [S sim14 WH sim58]State 1 occurs in Opiliones (Hansen amp Soslashrensen 1904)and higher scorpions (Jeram 1998) A lobe is presenton the coxa of leg 1 in the fossil scorpion Prearcturusand perhaps Stoermeroscorpio and is coded here as acoxapophysis Coxapophysis of leg 1 together with theepistome and coxapophysis of the palp form a uniquepreoral chamber the stomotheca Weygoldt (1998) andDunlop amp Braddy (2001) reject the homology of cox-apophyses and stomothecae in Opiliones and Scorpi-ones because these structures are not apparent infossils of those scorpions thought to have been aquaticHowever the coxapophyses in Opiliones are formedlargely from soft cuticular lsquolipsrsquo and similar structuresin fossil scorpions would probably not have beenpreserved51 Appendage IV (= arachnid leg 2) with coxapophy-sis 0 absent 1 present [S sim14 WH sim58 GEWB 75]State 1 occurs in extant Scorpiones and most Opil-iones except Dyspnoi (Shultz 1998)52 At least one pair of coxal gnathobases on append-ages III-VI (= arachnid legs 1ndash4) 0 absent 1 present[S sim23 WH sim66 GEWB 78]State 1 occurs in all appropriately preserved Euryp-terida including Baltoeurypterus (Selden 1981) andStylonurus (Clarke amp Ruedemann 1912) extantXiphosura on appendages IIndashVI and apparentlyProscorpius on appendage III (= arachnid leg 1)(Kjellesvig-Waering 1986 Jeram 1998 contra Giribetet al 2002) Fossilized appendages with gnathobaseshave been found in association with Chasmataspis(Chasmataspidida) suggesting that they may havehad State 1 (Dunlop et al 2004)53 Coxae of appendages IIIndashV (= arachnid legs 1ndash3)with jointed moveable endites 0 absent 1 present [Ssim23]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) and in adequately preserved eurypterids(Clarke amp Ruedemann 1912) (BaltoeurypterusSelden 1981)

PHYLOGENY OF THE ARACHNID ORDERS 251

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

54 Intracoxal muscle a muscle arising on anteriorwall of coxa and inserting on posterior wall 0 absent1 present [GEWB 237]State 1 occurs in Thelyphonida (MastigoproctusShultz 1993) and Amblypygi (Phryus Shultz 1999)It is unclear whether the muscle operating the move-able endite (53) in extant Xiphosura (Manton 1964Shultz 2001) is homologous with this muscle and iscoded here as uncertain It is not present in Steno-chrus (Schizomida) (Shultz unpubl observ)55 Coxae of appendages IIndashVI (= arachnid palp andlegs) with dorsal articulation with carapace 0 absent1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)56 Flabellum (exite) on coxa of appendage VI(= arachnid leg 4) 0 absent 1 present [GEWB sim110]State 1 occurs in extant Xiphosura (Yamasaki et al1988) An apparent exite occurs on appendages asso-ciated with Chasmataspis (Chasmataspidida) (Dunlopet al 2004)57 Insertion process of anteromedial tergo-coxal mus-cle 0 weakly developed 1 well to extremely well devel-oped [GEWB 238]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Boumlrner 1904 Shultz 1999)58 Musculi laterales enlarged lateral tergocoxalmuscle with attachment shifted from coxa to adjacentpleural membrane 0 absent 1 present [S 22 WH 65GEWB 242]State 1 occurs in Araneae and Thelyphonida (Shultz1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI TROCHANTER AND COXAndashTROCHANTER JOINT

59 Coxandashtrochanter joint with complex posteriorarticulation composed of two articulating sclerites 0absent 1 present [S 24 WH 67 GEWB 85]State 1 occurs in Araneae Amblypygi Thelyphonidaand Schizomida (Shultz 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI FEMUR AND TROCHANTERndashFEMUR JOINT

60 Depressor muscle (or homologue) of trochanter-femur joint 0 absent 1 present [WH 42 GEWB 220]State 0 appears to be a unique synapomorphy of Ara-neae (Liphistius Shultz 1989 Aphonopelma Ruh-land amp Rathmayer 1978 Hypochilus present studyother examples Clarke 1984 1986)61 Trochanterndashfemur joint with dorsal hinge or pivotoperated by flexor muscles only 0 absent 1 present[WH sim44 GEWB sim91]State 1 occurs in Palpigradi Other chelicerates have abicondylar articulation and are typically equipped

with antagonistic muscles (Van der Hammen 1985Shultz 1989) but see 60 for an exception62 Superior trochanterndashfemur muscle (or homologue)originating broadly in femur inserting on distal mar-gin of trochanter 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 19892001)63 Basifemurndashtelofemur joint of appendages III andIV (= arachnid legs 1 and 2) in adult 0 absent 1present [S 25 WH 68 GEWB 86]The basifemur and telofemur apparently develop aslsquosisterrsquo articles through division of a parental articlethe femur (eg opilioacariform endeostigmatidpalaeosomatid mites) (Coineau amp Van der Hammen1979 Evans 1992) A single animal can have all legswith divided or undivided femora or have some com-bination of the two typically with the more posteriorlegs having the divided femora Thus anterior legsoften tend to be paedomorphic with respect toposterior legs Most chelicerates have undivided fem-ora on some or all legs but muscles homologous withthose of the basifemurndashtelofemur joint may stilldevelop (66) (Shultz 1989) suggesting that the devel-opment of cuticular and muscular components of ajoint have a degree of developmental and evolutionaryindependence

Undivided femora occur in most extant euchelicer-ates including Xiphosura Palpigradi AraneaeAmblypygi Thelyphonida Schizomida ScorpionesOpiliones Pseudoscorpiones higher Sarcoptiformes(Archegozetes Van der Hammen 1989) many Prostig-mata (Tetranychus Evans 1992) and throughout Par-asitiformes Divided femora occur in all legs in manyearly divergent Acariformes including many Endeo-stigmata basally divergent Sarcoptiformes (egPalaeacarus) and Prostigmata (eg AllothrombiumMicrocaeculus) (Van der Hammen 1989 Evans1992) Only leg 4 has basi- and telofemora in Alycus(Van der Hammen 1989) Basi- and telofemora occuron legs 3 and 4 in Solifugae Ricinulei Opilioacari-formes and Eurypterida (Clarke amp Ruedemann 1912Selden 1981 Shultz 1989)

Giribet et al (2002) coded the lsquoannulusrsquo of certainTrigonotarbida (Shear et al 1987) as a basifemur buta similar non-muscularized annulus is present in theposterior legs of extant Thelyphonida In both casesthe condyles of the trochanterndashannulus and annulusndashfemur joints are aligned in such away that they per-form essentially identical movements and appear notto enhance either the range or degree of leg move-ment [This contrasts with the condition in the lasttwo pairs of legs in Baltoeurypterus where condyles ofthe apparent trochanter-basifemur and basifemurndashtelofemur joints have different arrangements (Selden1981)] The annulus in Thelyphonida appears to func-tion in maintaining constant volume of the tro-

252 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

chanterndashfemur joint by preventing ballooning of thearthrodial membrane thereby minimizing the effect ofinternal fluid pressure on movement64 Basifemurndashtelofemur joint of appendage V(= arachnid leg 3) in adult 0 absent 1 present See 6365 Basifemurndashtelofemur joint of appendage VI(= arachnid leg 4) in adult 0 absent 1 present See6366 Cuticular differentiation of basifemur-telofemurjoint absent but muscles present 0 absent 1 present- inapplicable due to presence of joint (63ndash65)State 1 occurs in appendages of postoral somites IIIndashIV (= arachnid legs 1ndash4) in Xiphosura AraneaeAmblypygi Thelyphonida Schizomida Scorpionesand Pseudoscorpiones (Shultz 1989)67 Circumfemoral ring 0 absent 1 presentThe femora of one or more pairs of legs in anactinot-richid Acari have a basal groove or ring associatedwith slit sensilla Acarologists term that portionproximal to the ring the lsquobasifemurrsquo and that portiondistal to the ring the lsquotelofemurrsquo and they refer tothe basifemur as the lsquosecond trochanterrsquo and thetelofemur as the lsquofemurrsquo Despite the confusingterminology the basifemur discussed in 63ndash65 is nothomologous with the lsquobasifemurrsquo of opilioacariformand parasitiform Acari as both a true basifemurndashtelofemur joint and a circumfemoral ring are presentin legs 3 and 4 of Opilioacariformes (Van der Ham-men 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI PATELLA AND FEMURndashPATELLA JOINT

68 Patella of appendage of postoral somite III(= arachnid leg 1) proportionally much longer thanthose of more posterior appendages 0 absent 1presentState 1 occurs in Thelyphonida Schizomida (Hansenamp Soslashrensen 1905 Shultz 1989) and Opilioacari-formes (Van der Hammen 1989)69 Femurndashpatella joint 0 monocondylar several axesof movement and multifunctional muscles 1 bicondy-lar hinge one axis of movement and antagonistic mus-cles 2 hinge one axis of movement flexor muscleswithout muscular antagonists [S sim26 WH sim69 GEWBsim87]State 0 occurs in Solifugae State 1 occurs in Phalang-ida (Opiliones) Scorpiones and PseudoscorpionesState 2 occurs in the remaining taxa (Shultz 1989)70 Patellar plagula 0 absent or with simple medianattachment 1 symmetrical Y-shaped with long prox-imal stem 2 symmetrical U-shaped (= arcuate scler-ite) 3 asymmetrical attaching to patella only atanterior margin [S 27 WH sim70 GEWB 234 = 243]State 1 occurs in extant Xiphosura (Shultz 1989)State 2 occurs in Araneae (Manton 1958 Ruhland amp

Rathmayer 1978 Clarke 1986 Shultz 1989) State 3occurs in Amblypygi Thelyphonida and Schizomida(Shultz 1989) Giribet et al (2002) included this char-acter twice

APPENDAGES OF POSTORAL SOMITES IIIndashVI TIBIA AND PATELLAndashTIBIA JOINT

71 Tibiae divided by one or more joints 0 absent 1presentState 1 occurs in Amblypygi (Weygoldt 2000)72 Patellandashtibia joint 0 monocondylar with or with-out CZY (73) 1 bicondylar hinge one axis of move-ment antagonistic muscles 2 hinge one axis ofmovement muscles without muscular antagonists [Ssim31 WH 74 GEWB sim92] (Shultz 1989 Selden Shearamp Bonamo 1991)73 Patellandashtibia joint with posterior compression zone(lsquoCZYrsquo) 0 absent 1 present - inapplicable due toabsence of monocondylar articulation [GEWB 99]State 1 is a unique synapomorphy of Araneae (Seldenet al 1991)74 Patellandashtibia joint of appendages IIIndashVI(= arachnid legs 1ndash4) with deep-set monocondylar pivotbordered by a pair of tibial processes to which extensormuscles attach 0 absent 1 present - inapplicabledue to absence of monocondylar joint (72)State 1 occurs in extant Xiphosura (Snodgrass 1952Shultz 1989) although a similar arrangement ispresent in the appendage III (= arachnid leg 1) in The-lyphonida (Shultz 1993)75 Patellandashtibia joint largely immobile specializedfor autotomy 0 absent 1 presentState 1 occurs throughout Amblypygi (Weygoldt 19842000)76 Anterior femurndashtibia or femoropatellandashtibia(transpatellar) muscle 0 absent 1 present [S sim30GEWB 222] (Shultz 1989)77 Proximal attachment of posterior femurndashtibia orfemoropatellarndashtibia (transpatellar) muscle 0 muscleabsent 1 dorsoposterior surface of femur andor pos-terior surface of patella 2 distal process of femur mus-cle acting as extensor of femur-patella joint [S sim28 WHsim71 + 72 GEWB 235 = 244 245] (Shultz 1989)78 Distal attachment of posterior femurndashtibia (trans-patellar) muscle 0 posterior 1 dorsal acting as exten-sor of patellandashtibia joint - inapplicable due to absenceof muscle (77) [S 29 GEWB 221] (Shultz 1989)79 Anterior patellandashtibia muscle 0 absent 1 present[S sim32 WH sim73 GEWB 246] (Shultz 1989)80 Posterior patellandashtibia muscle 0 absent 1 present[S 33 WH 76 GEWB 248] (Shultz 1989)81 Patellandashtibia joint spanned by elastic (lsquospringlikersquo)sclerite 0 absent 1 present [GEWB 107]State 1 is a unique synapomorphy of Solifugae (Sens-enig amp Shultz 2003)

PHYLOGENY OF THE ARACHNID ORDERS 253

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

APPENDAGES OF POSTORAL SOMITES IIIndashVI TARSUS AND TIBIAndashTARSUS JOINT

82 Tarsus divided into proximal basitarsus(= metatarsus) and distal telotarsus (= distitarsus orlsquotarsusrsquo) 0 absent 1 presentState 0 occurs in extant Xiphosura (Shultz 1989) andthroughout Acariformes (Lindquist 1984 Evans1992) in the anterior two pairs of legs in chthonioidpseudoscorpions and all legs in lsquomonosphyronidrsquo pseu-doscorpions including Feaelloidea and Cheliferoidea(Chamberlin 1931) A circumtarsal ring (83) associ-ated with slit sensilla in Parasitiformes (Acari) mayrepresent a joint between a telotarsus and basitarsus(Evans 1992)83 Circumtarsal ring 0 absent 1 presentState 1 is an apparent synapomorphy of Parasiti-formes (Evans 1992)84 Telotarsus in adult with two or more tarsomeres 0absent 1 present - inapplicable due to absence of telo-tarsus (82) [GEWB 71]State 1 occurs in Schizomida Thelyphonida Ambly-pygi Palpigradi Ricinulei Solifugae (Shultz 1989)Opilioacariformes (Van der Hammen 1989) and Ple-siosiro (Dunlop 1999)85 Three telotarsomeres on appendages of postoralsomites IV-VI (= arachnid legs 2ndash4) 0 absent 1present - inapplicable due to absence of telotarsus (82)or absence of telotarsomeres (84) [S 34 WH 77 GEWB102]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Shultz 1989)86 Tibiandashtarsus joint spanned by well-developed elas-tic (lsquospringlikersquo) sclerite 0 absent 1 present [GEWB107]State 1 occurs in Scorpiones phalangid Opiliones andSolifugae (Alexander 1967 Shultz 2000 Sensenig ampShultz 2003) Extant Xiphosura is coded for the lastprosomal appendage only (see 88)87 Appendage of postoral somite VI (= arachnid leg 4)with ring of large basally articulated spatulate pro-cesses at tibia-tarsus joint 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)88 Appendages postoral somites IIIndashV (= arachnidlegs 1ndash3) with tibiotarsus (ie tibia and tarsus not dif-ferentiated) 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 1989) butnot apparently in Weinbergina (Moore et al 2005) theonly synxiphosurid in which appendages are suffi-ciently preserved to determine the number ofpodomeres89 Ambulacrum peduncle-like extension of the tarsuswith internal condylophores and terminating distallywith apotele (eg claws) andor pulvillus 0 absent 1present

State 1 is present throughout Acari (Alberti amp Coons1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI APOTELE AND TARSUSndashAPOTELE JOINT

90 Apotele of appendage III (= arachnid leg 1) 0absent or not apparent 1 present [GEWB 101]State 0 occurs in Amblypygi Thelyphonida and Schi-zomida (Shultz 1999) However the apotele is proba-bly present but undifferentiated because muscles thatwould normally insert on the apotele terminate on theend of the lsquotarsusrsquo Dunlop (1999) reconstructed Hap-topoda as lacking an apotelic claw on leg 1 but noted inthe text that this was speculative Dunlop (2002b) hasreviewed apotelic diversity in Chelicerata91 Appendages of postoral somites IIIndashVI (= arachnidlegs) chelate with chela formed from tibiotarsus andapotele or tarsus and apotele 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988) The condition appears to occur in Chasmataspisbased on associated appendages (Dunlop et al 2004)but not in Diploaspis and Octoberaspis (Dunlop et al2001 Dunlop 2002a)92 Apotele with eversible or padlike empodium(= pulvillus) in adult 0 absent although empodialclaw may be present 1 present [S sim38 WH 81 GEWB95]State 1 occurs throughout Pseudoscorpiones (Cham-berlin 1931) Solifugae (Roewer 1934) Opilioacari-formes Holothyrida Ixodida and free-livingMesostigmata (Evans 1992) An eversible lsquopulvillusrsquo isprobably a primitive feature of Amblypygi (eg Char-inus Weygoldt 2000) but is absent in higher groups(eg Phrynus Weygoldt 2000)93 Inferior apotele muscle (= claw depressor) with tib-ial attachment 0 absent 1 present [S 35 WH 78GEWB 249]The character is coded for extant Xiphosura usingappendage of postoral somite VI (= arachnid leg 4) asother legs lack a differentiated tibia (88) State 1appears to occur throughout Arachnida (Shultz 1989)94 Inferior apotele muscle (= claw depressor) withpatellar attachment 0 absent 1 present [S 36 WH 79GEWB 250]State 0 occurs in extant Xiphosura Acari AmblypygiPalpigradi and Ricinulei State 1 occurs in all otherarachnids examined here (Shultz 1989)

OPISTHOSOMA

95 Number of opisthosomal somites in adult 0 five1 eight 2 nine 3 10 4 11 5 12 6 13 [WH sim33GEWB sim190]Chasmataspidids appear to have 13 opisthosomalsomites (Dunlop et al 2004) The synziphosurans

254 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Weinbergina (Moore et al 2005) and Limuloides(Anderson amp Selden 1997) appear to have ten opistho-somal somites Moore et al (2005) found no evidence ofan anterior microtergite in Weinbergina as proposedby Anderson amp Selden (1997) Xiphosurids appear tohave nine opisthosomal somites (Scholl 1977 Ander-son amp Selden 1997 Shultz 2001) Members of the fol-lowing orders have 12 opisthosomal somites Araneae(Millot 1949ac) Amblypygi (Weygoldt 2000) Thely-phonida and Schizomida (Hansen amp Soslashrensen 1905)and Pseudoscorpiones (Chamberlin 1931) Elevensomites are present in Palpigradi Solifugae (Roewer1934) and Trigonotarbida (Gilboarachne Shear et al1987 Palaeocharinus Fayers et al 2004) Plesiosiromay have 11 or 12 somites (Dunlop 1999) Justifica-tion for coding of more controversial taxa is providedbelow

Scorpiones Two principal hypotheses regarding thenumber of opisthosomal somites in scorpions havebeen advocated a 13-somite hypothesis derived fromembryological studies (Brauer 1895 Patten 1912Farley 1999 2005) and a 12-somite hypothesis basedon comparative anatomy of adults (Weygoldt amp Pau-lus 1979) The embryological interpretation is basedon the observation of pregenital genital and pectinalsomites (each with segmental ganglia and paired limbbuds) in early scorpion embryos followed by extremereduction or loss of the pregenital somite in laterembryos According to this view a missing pregenitalsomite should be added to the 12 apparent opisthoso-mal somites of post-embryonic scorpions to achieve afinal number of 13 The anatomy-based hypothesiswas introduced by Weygoldt amp Paulus (1979) whoadvocated a literal interpretation of post-embryonicsegmentation based on opisthosomal tergites Specifi-cally these authors argued that the last pair of dorsalendosternal suspensor muscles of non-scorpionarachnids especially Pedipalpi attach to the first(= pregenital) somite that this condition also occurs inscorpions and that there is no reason to invoke a miss-ing pregenital tergite They proposed that the pectinesbelong to the genital somite and that functional spe-cializations of the nervous system for pectinal functiongive the apparence of an extra neuromere duringembryonic development

I recently dissected the prosoma and anterioropisthosoma of the scorpions Centruroides Hadrurusand Heterometrus and focused on the composition ofthe muscular diaphragm (103) that separates thehaemocoelic compartments of the prosoma andopisthosoma (Shultz 2007) The diaphragm is com-posed of a metameric series of axial muscles fromthree somites the anterior somite corresponds to thelast prosomal somite and the posterior somite corre-sponds to the genital somite The middle elements

insert dorsally along a tranverse tendon attached tothe anterior margin of the first tergite These obser-vations are consistent with the embryological inter-pretation that a pregenital somite is present but itstergite is not expressed It appears likely that the thepregenital somite was compressed longitudinally dur-ing the evolution of the diaphragm I code scorpions ashaving 13 opisthosomal somites

Eurypterida The eurypterid opisthosoma is widelyassumed to have 12 somites (Clarke amp Ruedemann1912 Stoslashrmer 1944) However in a speculative paperon the evolutionary morphology of trilobites and che-licerates Raw (1957) proposed that both scorpions andeurypterids have 15 opisthosomal somites Rawassumed that scorpions have 13 apparent opisthoso-mal somites based on the transient pregenital somiteof scorpion embryology that scorpions and eurypteridsare close relatives and should have the same numberof somites and that olenellid trilobites and chelicer-ates always have somites in multiples of three Rawachieved 15 somites in scorpions by assuming the lastopisthosomal somite to be a diplosomite and that thetelson is a postanal somite even though there is noevidence for either of these proposals He attributedthese features to eurypterids as well He also notedthat the connection between the prosomal carapaceand first opisthosomal tergite in Eurypterida differedstructurally from the connection between adjacentopisthosomal tergites and regarded this as evidencefor a reduced pregenital tergite in the prosomandashopisthosoma junction Rawrsquos speculations were largelyforgotten until Dunlop amp Webster (1999) resurrectedthe proposal that eurypterids have a reduced opistho-somal tergite and therefore share a unique similaritywith scorpions Unfortunately Dunlop amp Bullocktreated Rawrsquos conjecture as if it were based on empir-ical evidence rather than an attempt to force euryp-terid morphology into a peculiar numerical system Inthe absence of convincing evidence to the contrary Ihave coded eurypterids as having 12 opisthosomalsomites

Ricinulei The opisthosoma consists of a membranouspedicel (97) bearing the gonopore ventrally Thefemale gonopore is bordered by an anterior plate and aposterior plate The remainder of the opisthosoma iscomposed of thick sclerites separated by less heavilysclerotized cuticle The first dorsal sclerite is short andfunctions as part of a prosomandashopisthosoma couplingmechanism (96) that may or may not be a specializedcomponent of the following tergite This is followed byfour tergites and sternites and a three-segmentedmetasoma (= lsquopygidiumrsquo) Millot (1945 1949f) rea-soned that the opisthosoma contains ten somites withsomites VIIndashIX incorporated into the pedicel XndashXIIIexpressed as tergites and sternites and the metasoma

PHYLOGENY OF THE ARACHNID ORDERS 255

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

comprising four somites He did not regard the dorsalcoupling sclerite as a separate tergite and his inter-pretation of four rather than three metasomal somiteshas been rejected Pittard amp Mitchell (1972) also pro-posed ten opisthosomal somites but achieved thisnumber by regarding the dorsal coupling sclerite as atergite of somite IX and by recognizing three metaso-mal somites Legg (1976) adopted the system proposedby Pittard amp Mitchell but did not regard the dorsalcoupling sclerite as separate from the following tergite(X) Van der Hammen (1979 1989) reconstructed 13somites The pedicel was regarded as having twosomites (VII + VIII) the following four sets of tergitesand sternites were considered diplosomites (IX + XXI + XII XIII + XIV XV + XVI) and the metasoma hadthree somites (XVIIndashXIX) This interpretation mayhave been biased by Van der Hammenrsquos attempt tounite Ricinulei with anactinotrichid Acari which healso regarded as having a primitive number of 13somites Selden (1992) described a fossil ricinuleidTerpsicroton that shows two pairs of depressions onthe three large premetasomal tergites which con-trasts with the single pair seen in extant species Thisobservation appears to corroborate the diplosomitehypothesis but the evidence does not indicate that thetergite anterior to these is a diplosomite

Dunlop (1996) attempted to homologize theprosomandashopisthosoma coupling mechanisms ofRicinulei and Trigonotarbida a goal that required anovel and rather forced interpretation of the dorsalsclerites He regarded the coupling sclerite as homol-ogous with the first opisthosomal tergite (VII) of trig-onotarbids and then followed Van der Hammenrsquosdiplosomite hypothesis to achieve 12 somites in totalDunloprsquos scheme differs from previous systems in sug-gesting that the pedicel does not contain the dorsalelements of the first and second somites and is in-consistent with Millotrsquos (1945) observation that thepre- and postgenital plates each have dorsoventralmuscles Here I code Ricinulei as having 12 somitesThere are three metasomal somites (XVIndashXVIII)three diplosomites (= six somites) (XndashXV) one cou-pling somite (IX) and two somites in the pedicel (iethe genital and pregenital somites) (VII VIII)

Opiliones Harvestmen have nine opisthosomalsomites and an anal operculum that is traditionallyregarded as the tergite of a tenth somite a tenth ster-nite is lacking (Hansen amp Soslashrensen 1904 Winkler1957) However the anal operculum appears to repre-sent a persistent embryonic telson (Moritz 1957) andis therefore likely to be a postsegmental structurecomparable with the stinger of scorpions flagella ofthelyphonids etc (Shultz 2000)

Acari The number of opisthosomal somites is probe-matic for most mite taxa due to uncertainty about the

location of the prosomandashopisthosoma boundry (5)extensive simplification or loss of metamericallyarranged sclerites and muscle attachments a paucityof developmental studies of engrailed expression andin Acariformes opisthosomal anamorphosis and het-erochronic modification of somite number (Evans1992) Ixodids appear to be the exception developmen-tal studies indicate five somites in the opisthosoma ofticks (Evans 1992)

Some mites retain external evidence of segmenta-tion ndash metameric patterns of furrows muscle attach-ments and slit sensilla ndash and with certainassumptions the number of opisthosomal somites canbe estimated Here it is assumed that the dorsal sur-faces of the last two prosomal somites are present inmites and retain their primitive association with legs3 and 4 (contra Van der Hammen 1989) (see 5 for jus-tification) Given this there appear to be 11 somites inthe opisthosoma of Opilioacariformes a conclusionalso reached by other workers (eg With 1904 Sitni-kova 1978 Klompen 2000) Similar reasoning sug-gests that Alycus has seven opisthosomal somites notnine as advocated by Van der Hammen (1989) Unfor-tunately external evidence is ambiguous for determin-ing the number of somites in other mite lineages andthese are coded as uncertain96 Prosoma-opisthosoma coupling mechanism 0absent 1 present [GEWB sim24]The posterior margin of the carapace and the anteriormargin of the first apparent opisthosomal tergite arespecialized as a coupling mechanism in Ricinulei andTrigonotarbida and Dunlop amp Horrocks (1996) pro-posed that these are synapomorphic for the twogroups However there is uncertainty about thehomology of the anterior opisthosomal somites in Trig-onotarbida and Ricinulei (95)97 Pedicel 0 absent 1 aranean type 2 ricinuleidtype [WP 30 S sim40 WH sim20 GEWB sim126]The body narrows at or near the prosomandashopisthosoma juncture in several arachnid lineages (ieSolifugae Palpigradi Amblypygi Araneae Ricinulei)and this lsquowaistrsquo has often been used as character at theinterordinal level (eg Pocock 1893) However thisinterpretation is rejected here for being subjective andfor uniting non-homologous conditions For exampleAraneae and Amblypygi are often grouped on the basisof a lsquopedicelrsquo yet it is a highly specialized structure inAraneae and its parts are not readily homologizedwith those of Amblypygi In contrast the condition inAmblypygi is a slightly narrower version of the highlymoveable prosomandashopisthosoma juncture in Uropygiwhich is not generally considered a lsquopedicelrsquo Thepedicel in Ricinulei is also unique a weakly sclero-tized stalk containing the genital opening (Cryptocel-lus Pittard amp Mitchell 1972 Ricinoides Legg 1976)See 95

256 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

98 Opisthosoma with three- segmented lsquobucklerrsquo 0absent 1 presentState 1 occurs throughout Chasmatapsidida (Dunlop2002a)99 Thoracetron consolidation of tergites of postgeni-tal somites 0 absent 1 present - inapplicable codedonly for Xiphosura (Anderson amp Selden 1997)100 Fusion of tergites of postoral somites VIII and IX(= opisthosomal somites 2 and 3) only 0 absent 1present [GEWB 146]State 1 appears to be a synapomorphy Trigonotarbida(Shear et al 1987) Ricinuleids also have diplotergites(101) but these appear to encompass a different com-bination of somites (see 95)101 Three diplotergites 0 absent presentState 1 is a unique synapomorphy of Ricinulei (Selden1992) See 95102 Paired opisthosomal defensive glands openingvia ducts on either side of the anus 0 absent 1present [S 46 WH 86 GEWB 122]State 1 occurs in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905)103 Muscular diaphragm separating prosomal andopisthosomal compartments formed by dorsoventralmuscles of postoral somites VIndashVIII and extrinsic mus-cles of leg 4 0 absent 1 presentState 1 is known only in extant scorpions (Lankesteret al 1885 Firstman 1973) The diaphragm is oftentreated as a single structure but recent anatomicalwork (Shultz 2007) has shown that it is a composite ofdorsal endosternal suspensors and extrinsic leg mus-cles (see 95) A diaphragm is also present in Solifugaebut it is located more posteriorly and does not appearto be homologous with that of scorpions (Roewer1934)104 Opisthosomal appendicular chondrites 0absent 1 presentThese cartilage-like columns of mesodermally derivedtissue are associated with each opisthosomal append-age including the chilaria in extant Xiphosura(Patten amp Hazen 1900 Yamasaki et al 1988 Fahr-enbach 1999 Shultz 2001)105 Paired appendages on ventral surface of postoralsomite VII (= opisthosomal somite 1) in adult 0absent 1 present [WP sim13 S 39 WH 82 GEWB 143]State 1 is known in Weinbergina (Moore et al 2005)and extant Xiphosura The bilobed structure of themetastoma in certain eurypterids as well as the cor-responding placement of chilaria in Xiphosura haveled some workers to regard the metastoma as fusedappendages of postoral somite VII (eg Stoslashrmer1955) This hypothesis is problematic given that (i) nospecial explanation is needed to account for bilaterallysymmetrical structures in bilaterally symmetricalorganisms (ii) incorporation of the ventral part of thefirst opisthosomal sternite as a functional element of

the prosoma is typical of arachnids and perhapsEuchelicerata generally and (iii) some workers regardthe metastoma as a sternite (eg Jeram 1998) Thusthe metastoma is coded here as uncertain Dunlop hasproposed that the sternum of scorpions is derivedfrom appendages citing the presence of transient limbbuds in scorpion development (Brauer 1895 Patten1912) but this is not compelling evidence for a persis-tent appendicular contribution to the sternite of theadult106 Megoperculum 0 absent 1 present [S sim41 + 42WH sim83 GEWB 162 + 163]State 1 occurs in Thelyphonida Schizomida Ambly-pygi Araneae (Shultz 1993 1999) Trigonotarbida(Shear et al 1987) and probably Haptopoda (Dunlop1999) and Palpigradi The megoperculum consists ofappendages of opisthosomal somite 2 (= genital somiteor postoral somite VIII) (Shultz 1993 1999 Popadicet al 1998) that have fused medially and displacedthe ventral body wall of the somite anteriorly (oftenrepresented by a small sternite to which dorsoventralmuscles of postoral somite VIII attach) The megoper-culum projects posteriorly to form the ventral surfaceof a pregenital chamber The dorsal surface of thechamber is formed by the ventral body wall of opistho-somal somite 3 (= postoral somite IX) which serves asthe ventral attachment of the paired dorsoventralmuscles of that somite The megoperculum bears book-lungs (where present) and sometimes paired gonopodsthat are probably derived from telopodites A similarbut less developed operculum is present on theopisthosomal somite 3 which may also bear booklungsand eversible vesicles (eg Amblypygi) (see 111) corre-sponding to the booklungs and gonopods of the mego-perculum (Shultz 1999)

I contend here that a megoperculum is present inPalpigradi and that this is particularly evident infemales Specifically a large unpaired lobe projectsposteriorly from the genital somite over the ventroan-terior surface of the first postgenital somite therebyforming a pregenital chamber The dorsoventral mus-cles of the genital somite attach ventrally near theanterior border of postoral somite VIII and those ofpostoral somite IX attach to the upper surface of thepregenital chamber (Rucker 1901 Boumlrner 1902bRoewer 1934 Millot 1949d Van der Hammen 1989)Weygoldt (1998) has questioned this interpretation ofPalpigradi107 Postgenital operculum or lsquosternitersquo 0 sclerotized1 not sclerotized - inapplicable coded only for Pan-tetrapulmonata (Platnick amp Gertsch 1976)State 1 is present in Trigonotarbida Amblypygi The-lyphonida Schizomida and Mesothelae (Araneae) butnot Opisthothelae (Araneae)108 Genital opening of female guarded by four plates(one pregenital one postgenital two laterogenitals)

PHYLOGENY OF THE ARACHNID ORDERS 257

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

genital opening of male guarded by two plates 0absent 1 presentThis an apparent synapomorphy of Holothyrida(Acari) (Van der Hammen 1989)109 Anterior margin of genital opening in male withglands secreting via fusules 0 absent 1 presentState 1 occurs in Araneae as epigastric or lsquoepiandrousrsquoglands (Marples 1967) and in Palpigradi (Condeacute1991b)110 Paired valve-like plates apparently formed fromcomponents of three somites covering triradiate genitalopening 0 absent 1 presentThis is an apparent synapomorphy of Acariformes(Evans 1992 Alberti amp Coons 1999)111 Eversible lsquoappendagesrsquo on the ventral surface ofpostgenital somites 0 absent 1 present [S sim43 WHsim84 GEWB sim127]Paired ventral lsquoappendagesrsquo operated in part byhaemolymph pressure occur in the form of ventralsacs in certain Amblypygi (postoral somite IX) (Wey-goldt 2000) (Charinus Millot 1949b not PhrynusShultz 1999) prokoenenian Palpigradi (postoralsomites XndashXII) (Rucker 1901 Condeacute 1991a) as spin-nerets in Araneae (XndashXI) (Shultz 1987) and as geni-tal papillae in many Acariformes (Alberti amp Coons1999) including Oribatida many Endeostigmata (Aly-cus Van der Hammen 1989) and Prostigmata (Allo-thrombium Saboori amp Kamali 2000 MicrocaeculusEvans 1992) Eversible vesicles similar to those ofamblypygids occur in fossils of the trigonotarbidPalaeocharinus (postoral somite IX) (Fayers et al2004) Two cuticular structures (lsquogenital verrucaersquo)occur anterior to the genital opening in Opilioacari-formes and each covers the opening to a thin-walledinvaginated sac (lsquogenital papillarsquo) of unknown function(Van der Hammen 1989) Van der Hammen proposedthat it is homologous to a genital papilla of Acari-formes but Alberti amp Coons (1999) have questionedthis interpretation on several grounds and refer tothese structures as pregenital capsules112 Opisthosomal silk glands and spinnerets derivedfrom appendages on postoral somites X and XI(= opisthosomal somites 4 and 5) 0 absent 1 present[WP 33 WH 41 GEWB 123 + 142]State 1 is a unique synapomorphy of Araneae (Plat-nick amp Gertsch 1976) although potentially homolo-gous glands and spigot-like setae occur on the ventralopisthosomal surface in Palpigradi (Millot 1943Condeacute 1991a)113 Opisthsosomal spinnerets location 0 near mid-dle of opisthosoma 1 near posterior end of opisthso-som ndash inapplicable coded only for Araneae (Platnickamp Gertsch 1976)114 Anterior medial lsquospinneretsrsquo 0 absent 1 present- inapplicable coded only for Araneae (Platnick ampGertsch 1976)

115 Opisthosomal tergites divided longitudinally intoone median and two lateral plates 0 absent 1 present[GEWB 145]State 1 occurs in non-curculioid Ricinulei (Selden1992) and many Trigonotarbida (Dunlop 1996) Thisfeature or something very like it (eg distinct opistho-somal trilobation) occurs sporadically in severaleuchelicerate groups including Chasmataspis (Chas-mataspidida) (Dunlop et al 2004) most fossil xipho-surans (Anderson amp Selden 1997) and a few derivedEurypterida (especially Mixopteroidea) (Tollerton1989)116 Number of metasomal somites 0 zero 1 two 2three 3 five 4 nine [S sim44 WH sim85 GEWB 128 144]The metasoma is a preanal region comprising multiplesomites that are substantially narrower than the pre-ceding somites or mesosoma Metasomal somitesoften lack pleural membranes and take the form ofsclerotized rings State 1 occurs in TrigonotarbidaState 2 is present in synziphosurid Xiphosura (Ander-son amp Selden 1997) Amblypyi Thelyphonida Schizo-mida and Ricinulei State 3 occurs in Scorpiones andEurypterida and State 4 occurs in Chasmatispidida

Some workers appear to consider a five-segmentedmetasoma as conclusive evidence for the monophylyof a Eurypterida + Scorpiones clade (eg Dunlop ampBraddy 2001) presumably because the character rep-resents a kind of tagmosis and should therefore begiven substantial phylogenetic weight However Wey-goldt (1998) has pointed out that a metasoma contain-ing three somites is probably part of the ground planof Euchelicerata The known diversity of the eucheli-cerate metasoma indicates that this feature can inincrease or decrease its segmental composition in evo-lution or be eliminated entirely without necessarilychanging the total number of opisthosomal somitesConsequently it is problematic to assume that thefive-segmented metasoma has any special immunityto homoplasy117 Postanal structure (telson) 0 absent or not obvi-ously developed 1 presentPostanal structures are considered non-segmentalbecause they are modifications of the embryonicregion posterior to the site of somite addition (embry-onic growth zone) A well-developed postanal struc-ture persists in adults in Xiphosura EurypteridaChasmataspidida Scorpiones Schizomida Thely-phonida and Palpigradi The anal operculum of Opil-iones is generally considered by morphologists to bethe tergite of the tenth opisthosomal or anal somite(eg Hansen amp Soslashrensen 1904) but embryological evi-dence (Moritz 1957) suggests that it is derived fromthe postproliferative zone and thus corresponds to thetelson of other chelicerates (see 95)118 Postanal structure shape 0 caudal spine 1aculeus 2 flagellum 3 anal operculum - inapplica-

258 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ble due to absence of postanal structure (117) [WP sim41S sim45 WH sim24 GEWB sim121 129 + 147]State 0 is present in Xiphosura Eurypterida andChasmataspidida State 1 occurs in Scorpiones State2 occurs in Palpigradi Thelyphonida and SchizomidaState 3 occurs in Opiliones (see 95 117)119 Specialized postanal flagellum in male (see 157ndash159) 0 absent 1 present - inapplicable coded onlyfor taxa with postanal flagellum (118) [GEWB 131]State 1 is a unique synapomorphy of Schizomida(Hansen amp Soslashrensen 1905 Cokendolpher amp Reddell1992 Reddell amp Cokendolpher 1995)

RESPIRATORY SYSTEM

120 Respiratory medium 0 water 1 airAmong the terminal taxa included only fossil scorpi-ons are problematic for this character They are codedhere as uncertain Based on a study of book lungmicrosculpture Scholtz amp Kamenz (2006) have arguedthat arachnids are primitively terrestrial and pulmo-nate (see also Firstman 1973) and have questionedwhether any fossil scorpions were aquatic No positionon this proposal is taken here121 Respiratory lamellae on opisthosomal somite 2(= genital somite postoral somite VIII) 0 absent 1present [WP sim37 S sim51 WH sim22 GWEB sim133ndash137]This character encompasses book gills (= lamellae thatfunction in water) and book lungs (= lamellae thatfunction in air) The distinction between book gills andbook lungs is accommodated here by 120 in combina-tion with 121ndash124 Respiratory lamellae are presenton the genital somite in Trigonotarbida AraneaeAmblypygi Thelyphonida and Schizomida Petrunk-evitch (1949) reconstructed Plesiosiro as having booklungs but Dunlop (1999) could not corroborate thisThis character was used by Dunlop amp Webster (1999)to propose that Xiphosura and Scorpiones are closelyrelated because they both lack respiratory lamellae onthe genital somite Dunlop amp Braddy (2001) alsoargued for the placement of Eurypterida with Xipho-sura and Scorpiones based in part on this characterAt least some eurypterids may have had respiratorylamellae (eg Manning amp Dunlop 1995) but the onlyevidence of their segmental distribution is derivedfrom one specimen (Braddy et al 1999)122 Respiratory lamellae on opisthosomal somite 3(= postoral somite IX) 0 absent 1 present [GEWB136]State 1 occurs in Amblypygi Thelyphonida and Xipho-sura It is a groundplan feature of Araneae (Platnick ampGertsch 1976) and is present in all representativetaxa included here Braddy et al (1999) proposed theexistence of respiratory lamellae on postoral somitesIXndashXII in Eurypterida based on evidence from onespecimen This character is coded as uncertain for the

Eurypterida included here Extant scorpions are codedhere as lacking respiratory lamellae on this somite(= pectinal somite) based on the interpretation of theopisthosomal segmentation discussed in 95123 Respiratory lamellae on opisthosomal somites 4ndash6 (= postoral somites XndashXII) 0 absent 1 presentState 1 is definitely present in extant Xiphosura andScorpiones and may have occurred in Eurypterida124 Respiratory lamellae on opisthosomal somite 7(= postoral somite XIII)State 1 occurs in extant Xiphosura and Scorpionesonly125 Kiemenplatten 0 absent 1 presentThese structures are located on the roof of opercularchambers in Eurypterida (Clarke amp Ruedemann1912) where they take the form of ventrally projectingcones with a distinct cuticular microsculputure (Man-ning amp Dunlop 1995) Dunlop amp Braddy (2001)inferred the existence of Kiemenplatten in all Palaeo-zoic scorpions based on a description and photos of onespecimen of Paraisobuthus duobicarinatus by Kjelles-vig-Waering (1986 pls 16ndash18) The plates depict darkcone-like rearward-pointing denticles distributedwithin a white amorphous material The denticlesappear to lack microsculpture even though the mag-nifications at which the photos were taken (times90ndashtimes330)are comparable with those illustrating the cones ofKiemenplatten (times170ndashtimes350 in Manning amp Dunlop1995 figs 1 2) Given the diversity of cuticular struc-tures present in the atria of booklungs book gills andtracheae the denticles appear to bear no special sim-ilarity to the cones of Kiemenplatten126 Tracheal system 0 absent 1 paired ventral stig-mata on postoral somite VIII (= opisthosomal somite2) 2 paired ventral stigmata on postoral somites IXand X 3 one pair of stigmata opening near legs 3 or 44 paired stigmata associated with chelicerae 5 fourpairs of stigmata on dorsal surface of opisthosoma -inapplicable aquatic (120) [WP 40 sim43 S sim52 + 5354 WH sim23 45 88 GEWB sim138 139]Firstman (1973) and Weygoldt amp Paulus (1979)hypothesized that tracheae are homologous in all tra-cheate arachnids except spiders and subsequentworkers have entertained this hypothesis by includinga character for the presenceabsence of tracheal sys-tems that ignores the diverse arrangement of stig-mata in arachnids (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) However thisapproach assumes that internal tracheal systems areconserved but that tracheal openings (stigmata)appear and disappear on different parts of the bodywith higher evolutionary frequency Here I assumethat stigmata are conserved in evolution and that dif-ferences in their anatomical placement reflect the evo-lution of new tracheae State 1 occurs throughoutOpiliones State 2 occurs throughout Solifugae

PHYLOGENY OF THE ARACHNID ORDERS 259

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

(Roewer 1934) and Pseudoscorpiones (Chamberlin1931) State 3 recognizes the possible homology of stig-mata in Ricinulei Opilioacariformes and Parasiti-formes (Acari) (Lindquist 1984 Van der Hammen1989) Tracheal systems are apparently absent in thegroundplan of Acariformes but Prostigmata typicallyhave tracheae associated with the chelicerae Tra-cheae derived from the posterior pair of book lungs arewidespread in araneomorph Araneae but the booklungs are retained and tracheae are absent in themost basally divergent groups (eg Hypochilus)

BOXndashTRUSS AXIAL MUSCLE SYSTEM (BTAMS)

127 Posterior oblique muscles of BTAMS of postoralsomites IndashVI 0 absent 1 present in one or moresomitesState 0 occurs in Xiphosura and state 1 occurs inPalpigradi Araneae Amblypygi Thelyphonida andSchizomida (Shultz 2001) The condition in AcariRicinulei Opiliones and Solifugae is not known128 Anterior oblique muscles of BTAMS posterior topostoral somite VI 0 absent 1 presentState 1 occurs in Xiphosura which is probably theprimitive condition based on comparison with otherarthropods (Shultz 2001) State 0 occurs in all arach-nids examined thus far129 Ventral attachments of posterior oblique musclesof opisthosomal BTAMS located in prosoma 0 absent1 presentState 1 occurs in extant Xiphosura (Shultz 2001)130 Endosternite fenestrate 0 absent 1 present [S 8WH 54 GEWB 35]State 1 occurs in Thelyphonida and hubbardiid Schi-zomida (Firstman 1973) The condition in protoschi-zomid Schizomida is not known131 Suboral suspensor a tendon that arises from theBTAMS and inserts on the ventral surface of the oralcavity via muscle 0 absent 1 presentState 1 occurs in Palpigradi (Eukoenenia Millot1943) Amblypygi (Shultz 1999) and Thelyphonida(Shultz 1993)132 Perineural vascular membrane in adult 0absent 1 present [WH 28 GEWB 155]Wheeler amp Hayashi (1998) coded extant Xiphosura asunknown although the primary source for this char-acter (Firstman 1973) stated that adult Limulus havea perineural vascular membrane Within Arachnidapresence of a perineural vascular membrane is appar-ently correlated with the presence of tracheae (126)133 Ventral endosternal suspensor attaching on coxaof anteriorly adjacent somite 0 absent 1 present[GEWB 239 253]State 1 occurs in Amblypygi (Shultz 1999) Thely-phonida (Shultz 1993) and Schizomida (Shultzunpubl observ)

134 Posteriormost postoral somite with a pair of dor-soventral muscles 0 VI 1 VII 2 VIII 3 XII 4 XIII5 XIV 6 XV 7 XVIDorsoventral muscles tend to run in a continuousmetameric series beginning in the prosoma Deter-mining the last tergite on which the muscle seriesends is substantially easier than counting especiallygiven frequent anatomical complexities near theprosomandashopisthosoma juncture The dorsoventralmuscle series ends on postoral somite VII inCyphophthalmi (eg Chileogovea) and Laniatores(Opiliones) (Shultz unpubl observ) on postoralsomite VII or VIII in Eupnoi (Leiobunum Shultz2000 unpubl observ) on postoral somite XII in Pal-pigradi (Roewer 1934) Solifugae (Bernard 1896Roewer 1934 Millot amp Vachon 1949) and Araneae(Liphistius Hypochilus Millot 1933 but segmenta-tion reinterpreted here following Shultz 1993 1999)on postoral somite XIII in hubbardiid Schizomida(Cokendolpher amp Reddell 1992) and Scorpiones(Lankester et al 1885) and on postoral somite XIVin extant Xiphosura (Shultz 2001) Thelyphonida(Shultz 1993) protoschizomid Schizomida (Cokendol-pher amp Reddell 1992) and Amblypygi (Shultz 1999)Dunlop (1999) reconstructed Plesiosiro as havingpaired tergal apodemes ending on postoral somite XVbut his figures indicate that the series ends on posto-ral somite XIV The interpretation of opisthosomalsegmentation in Ricinulei coded here (95) indicatesthat the series ends on postoral somite XV in Terpsi-croton and this is treated as the groundplan for theorder The posteriormost dorsoventral muscles inPseudoscorpiones occur on postoral somite XVI(Vachon 1949)

NERVOUS SYSTEM

135 Segmental ganglia 0 consolidated in prosoma1 one or more present in opisthosoma [WH 30 GEWB210 + 216]Adult neuromeres are exclusively prosomal in Ambly-pygi liphistiomorph and araneomorph Araneae Opil-iones Pseudoscorpiones Ricinulei and Acari (Millot1949a)136 Dorsal median eyes 0 absent 1 present [WP14 + 47 S 50 WH sim7 GEWB 1]Median eyes are a groundplan feature of Euchelicer-ata and are retained in extant Xiphosura Eurypteridaand occur in all arachnid orders except PalpigradiSchizomida Ricinulei and Pseudoscorpiones (Paulus1979) Median eyes are also absent in opilioacariformand parasitiform Acari (Lindquist 1984) and perhapsthe fossil whipscorpion Proschizomus (Dunlop amp Hor-rocks 19951996) They are unknown in synziphosu-ran Xiphosura Median eyes are present in a variety ofbasal acariform mites including Endeostigmata (but

260 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

not Alycus) Prostigmata (Microcaeculus) and certainoribatids (Palaeacarus) (Evans 1992 Alberti amp Coons1999) Giribet et al (2002) coded all representativeAcari as lacking median eyes It is not known whetherthe eyes of some cyphophthalmid opilions are medianor lateral evidence from tracheal branching inCyphophthalmus (Janczyk 1956) suggests that theyare median eyes (Shultz amp Pinto da Rocha 2007) andpresence of a tapetum in Stylocellus is consistent withlateral eyes (Shear 1993)137 Retinula cells of dorsal median eyes 0 organizedinto closed rhabdoms 1 organized into network ofrhabdomeres 2 disorganized - inapplicable due toabsence of median eyes (136) [GEWB sim3]State 0 is present in Scorpiones Thelyphonida andAmblypygi State 1 is present in Solifugae and Ara-neae state 2 is present in Xiphosura (Paulus 1979)The retinula cells of median eyes have been studied inseveral prostigmatid Acariformes and are organized ina network in some taxa and in an irregular pattern inothers (Alberti amp Coons 1999) Retinulae in phalangidOpiliones have state 0 proximally and state 2 distally(Schliwa 1979)138 Ventral median eyes 0 absent 1 presentState 1 occurs in early instars of extant Xiphosura(Paulus 1979)139 Lateral eyes 0 absent 1 presentLateral eyes are primitively present in Chelicerataand are absent in Palpigradi and Opiliones (Paulus1979) It is unclear whether the eyes of cyphoph-thalmid opilions are median or lateral (see 136)140 Arrangement and number of lateral eyes 0 com-pound many 1 five or more pairs (includes micro-lenses) 2 three primary pairs (excludes microlenses)3 two pairs 4 one pair - inapplicable due to absenceof lateral eyes (139) [WP sim13 + 18 + 38 + 44 + 52 Ssim49 WH sim10 GEWB sim4 + 5]True compound eyes are present in XiphosuraEurypterida Chasmataspidida and many PalaeozoicScorpiones Among the anactinotrichid Acari Opilioa-cariformes have two (Neocarus) or three (Siamacarus)pairs of lateral eyes (although at least one species ofSiamacarus lacks eyes) the allothryrid Holothyridaand many Ixodida have a single pair of lateral lensesand the remainder apparently lack eyes (Evans1992) Thelyphonida was coded as lsquo12rsquo to reflect fivepairs of lenses comprising three pairs of primarylenses and two pairs of small accessory lenses (notsimply three pairs as coded by Giribet et al 2002)Trigonotarbida is also coded as lsquo12rsquo to reflect threepairs of primary lenses and multiple small accessorylenses Dunlop (1999) illustrated Haptopoda as havingpaired triads of lateral eyes but noted that there wasactually no evidence of this in the fossils The lateraleyes of fossil ricinuleids have two pairs of lenses(Selden 1992) One pair of eyes or eyespots are

present in extant Ricinulei and many Schizomidaalthough five genera of hubbardiid Schizomida have apair of lenses (Reddell amp Cokendolpher 1995) State 1is the groundplan for extant Scorpiones and State 3occurs in the pseudoscorpions Chthonius FeaellaNeobisium (but State 4 in Chelifer) and Solifugae(Paulus 1979)141 Lateral eyes with closed rhabdoms 0 absent 1present - inapplicable due to absence of lateral eyes(139) [WP 21 WH 12 GEWB 6]Presence of closed rhabdoms is probably primitive forChelicerata and is retained in extant Xiphosura andScorpiones Retinula cells form a network of rhab-domeres in other extant chelicerates (Paulus 1979)but these networks can differ substantially in detailWeygoldt (1998) and other workers have given sub-stantial weight to the network character in unitingnon-scorpion arachnids but these authors seem not togrant comparable phylogenetic significance to theanalogous character of the median eyes shared bySolifugae and Araneae (137)142 Slit sensilla 0 absent 1 present [WP 19 S 47WH 11 GEWB 209]State 1 occurs in all arachnid orders except Palpigradi(Shultz 1990) The proposal that lsquoprimitiversquo slit sen-silla were present in Eurypterida (Dunlop amp Braddy1997) appears to be based on the over-interpretation ofa comparatively large notch that occurs in Baltoeu-rypterus at the terminus of the tibia (= podomere 7)(see also Edgecombe et al 2000 Giribet et al 2002)The neural construction of slit sensilla like function-ally similar campaniform sensilla of hexapods is sim-ilar to that of trichoid sensilla (Chapman 1998Klompen 2000) Thus these cuticular stress receptorsmay represent modified bases of sensory setae a viewsupported by the replacement of setae by slit sensilladuring post-embryonic development in some Opilioac-ariformes (Klompen 2000)143 Trichobothria 0 absent 1 present [GEWB 213]Trichobothria are present in extant Scorpiones(Jeram 1998) Pseudoscorpiones (Chamberlin 1931)some endeostigmatid (Alycus Van der Hammen 1989)and prostigmatid Acariformes and most Oribatida(Lindquist 1984) They occur on the ventral surface ofthe pedal femora in the opilioacariform Siamacarus(Leclerc 1989) They are also present in Araneae (Foe-lix 1996) Amblypygi (Weygoldt 2000) Schizomidaand Thelyphonida (Hansen amp Soslashrensen 1905) Theyare apparently absent in all non-arachnid chelicer-ates Solifugae Ricinulei Opiliones (Reissland ampGoumlrner 1985) and Parasitiformes (Acari) (Lindquist1984)144 Tibial trichobothria with 2-1-1-1 pattern onappendages IIIndashVI (= arachnid legs 1ndash4) 0 absent 1present - inapplicable due to absence of trichobothria(143) [S 48 WH 87 GEWB 88]

PHYLOGENY OF THE ARACHNID ORDERS 261

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs only in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905) Note that Shultz (1990)erroneously described this character as a 2-2-1-1 pat-tern and that this error was repeated by Giribet et al(2002) and described as 2-1-1 by Wheeler amp Hayashi(1998)145 Paired trichobothria on dorsal surface ofprosoma 0 absent 1 present - inapplicable due toabsence of trichobothria (143)State 1 is an apparent synapomorphy of Acariformesand is present in all acariforms included in this study(Alberti amp Coons 1999)146 Malleoli 0 absent 1 present [WP 45 GEWB 96]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)147 Pectines 0 absent 1 present [WP 20 WH 38GEWB 120]State 1 is a unique synapomorphy of Scorpiones148 Intercheliceral median organ 0 absent 1present [GEWB 212]State 1 is a unique synapomorphy of Palpigradi(Roewer 1934)149 Tarsal organ on appendage of postoral somite III(= arachnid leg 1) (= Hallerrsquos organ) sensilla containedwithin a cuticular depression on the superior surface ofthe tarsus of appendage III (= arachnid leg 1) 0absent 1 present [GEWB sim100]Klompen (2000) has noted that State 1 occurs in Opil-ioacariformes Parasitiformes (except Mesostigmata)and Ricinulei where it also occurs on leg 2 (Talaricoet al 2005) State 1 occurs on all legs in Araneae andperhaps Scorpiones (Foelix 1985)150 Tarsal organ on appendage of postoral somite IV(= arachnid leg 2) 0 absent 1 present (See 149)

REPRODUCTION

151 Gonads 0 primarily prosomal 1 primarilyopisthosomalState 0 is limited to Xiphosura (Sekiguchi 1988)State 1 occurs throughout Arachnida (Millot1949a)152 Ladder-like opisthosomal gonadsaccessoryglands (see 153) 0 absent 1 present [WH sim37 GEWBsim158]Giribet et al (2002) followed Wheeler amp Hayashi(1998) who followed Clarke (1979) in coding gonads asreticulate (Xiphosura) ladder-like (Scorpiones Thely-phonida Schizomida) or lsquosaccularrsquo (all remainingArachnida) However comparisons between Xipho-sura and Arachnida are problematic given that thexiphosuran gonads are primarily prosomal and thoseof arachnids are primarily opisthosomal (151) Fur-ther the reticulate pattern in Limulus (Xiphosura)but not other extant xiphosurans contains a distinctlyladder-like component The lsquosaccularrsquo state is probably

artificial as it encompasses a wide variety of pairedand unpaired structures153 Male gonads in two distinct parts one producingsperm and another (tubular gland) producing a holo-crine secretion similar to degenerate sperm 0 absent1 presentState 1 occurs in Thelyphonida Schizomida andAmblypygi although the holocrine material is pro-duced by ventral organs in Amblypygi and dorsalorgans in Thelyphonida and Schizomida (Alberti2005)154 Number of gonopores 0 two 1 oneExtant xiphosurans have two small genital openingson the base of the genital telopodite and all extantarachnids have a single opening The condition inEurypterida is not known Clarke amp Ruedemann(1912) located a pair of openings near the base of themedian organ (161) that are the outlets of the lsquohornorgansrsquo but it is unclear whether these are genitalducts or accessory structures Braddy amp Dunlop (1997)have developed numerous speculations about thesestructures and extended their arguments far beyondthe available evidence155 Genital opening (ie gonopore or gonostome)appearing to open in prosomal region (ie between legcoxae or anterior to posterior carapacal margin) 0absent 1 present [WP sim50 WH 26 GEWB 166]The genital opening in Euchelicerata is located onpostoral somite VIII (= opisthosomal somite 2) but ithas shifted anterior to the posterior margin of thecarapace or between the coxae of the last pair of legsin most Scorpiones (but not in PalaeoscorpiusKjellesvig-Waering 1986) and Opiliones The geni-tal opening occurs near or anterior to the last coxaein the opilioacariform and parasitiform Acari repre-sented here It is variable in Acariformes but islocated posterior to the coxae in all representativetaxa156 Ovipositor 0 absent 1 present [WP sim51 S sim60WH sim91 GEWB sim172]An ovipositor with a trilobed terminus is an apparentgroundplan character of Oribatida (Lindquist 1984Alberti amp Coons 1999) An ovipositor is also present inOpilioacariformes and Opiliones (Van der Hammen1989)157 Stalked spermatophore attached to substratum0 absent 1 present [S 57]State 1 occurs in Scorpiones PseudoscorpionesAmblypygi Thelyphonida and Schizomida The mech-anism of sperm transfer is unknown in PalpigradiOpilioacariformes and Holothryrida158 Male turns posterior end toward female duringmating behaviour 0 absent 1 presentState 1 occurs in Amblypygi (Weygoldt 2000) The-lyphonida and Schizomida (Weygoldt amp Paulus1979)

262 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

159 Female grasps male opisthosoma during matingbehaviour 0 absent 1 present [WP 29 S 58 WH 19GEWB 188]State 1 occurs in Thelyphonida and Schizomida (Wey-goldt amp Paulus 1979) Evolution of this behaviour wasprobably facilitated by 158160 Penis 0 absent 1 present [WP sim51 GEWB 167]State 1 is a unique synapomorphy of Opiliones A truepenis occurs in Phalangida A clearly homologousstructure is present in Cyphophthalmi (Opiliones) andapparently functions in depositing a spermatophore inthe femalersquos genital chamber (Karaman 2005) A truepenis may occur in some mites but it does not appearto be a groundplan feature of any major group (Evans1992) The lsquopenisrsquo in Oribatida is really a spermatop-ositor it functions in construction of a spermatophore(Alberti amp Coons 1999)161 Median organ 0 absent 1 presentState 1 occurs throughout Eurypterida (Clarke amp Rue-demann 1912) and has been observed in diploaspididChasmataspidida (Dunlop 2002a also Loganamaras-pis Tetlie amp Braddy 2004)

SPERM MORPHOLOGY

162 Nucleus with manchette of microtubules 0absent 1 present [S 54 WH 89 GEWB 193]State 1 occurs in Araneae Amblypygi Thely-phonida Schizomida Ricinulei and gonyleptid Lani-atores (Opiliones) (Alberti 1995 Giribet et al2002)163 Axoneme 0 absent 1 present [WP sim49 S sim55WH sim25 GEWB 195]State 1 occurs in Xiphosura Scorpiones Cyphoph-thalmi (Opiliones) Pseudoscorpiones RicinuleiAraneae Amblypygi Thelyphonida and Schizomida(Alberti 1995)164 Coiled axoneme 0 absent 1 present - inappli-cable due to absence of axoneme (163) [WP 22 S sim55WH 13 GEWB 196]State 1 occurs in Pseudoscorpiones Ricinulei Ara-neae Amblypygi Thelyphonida and Schizomida(Alberti 1995)165 Microtubule arrangement in axoneme 0 9 + 0 19 + 1 2 9 + 2 3 9 + 3 - inapplicable due to absence ofaxoneme (163) [WP sim26 S sim56 WH sim16 GEWB 198](Alberti 1995)166 Helical or corkscrew shaped nucleus 0 absent 1present [GEWB sim204]State 1 occurs in Araneae Amblypygi ThelyphonidaSchizomida and certain Scorpiones (eg Hadrurus)(Alberti 1995)167 Vacuolated-type sperm 0 absent 1 present[GEWB 205]State 1 is unique to Opilioacariformes and Parasiti-formes (Acari) (Alberti 1995)

168 Sperm aggregates 0 absent 1 present [GEWB206] (Alberti 1995)

DEVELOPMENT

169 Yolk in early embryo 0 concentrated (centroleci-thal or telolecithal) 1 evenly distributed (isolecithal)[WH sim34 GEWB sim191]Wheeler amp Hayashi (1998) and Giribet et al (2002)coded this character as lsquo0 isolecithal or telolecithal 1centrolecithalrsquo based on information presented byYoshikura (1975) Their coding appears to highlightseparate conditions found in Scorpiones (ie isoleci-thal and telolecithal) rather than a property intrinsicto the character itself170 Embryonic nutrition other than yolk 0 absent 1present [WP sim49]State 1 occurs in extant Scorpiones (except buthidsand chaerilids) and Pseudoscorpiones (Weygoldt1969)171 Embryological growth zone 0 initiating segmentaddition within prosoma 1 initiating segment addi-tion posterior to prosoma [GEWB sim192]Giribet et al (2002) followed Dunlop amp Webster (1999)in stating that only Xiphosura and Scorpiones havestate 0 among Chelicerata However the last prosomalsomite develops from the growth zone in Ixodida(Anderson 1973 Evans 1992) A similar processapparently occurs in the acariform mites Tyroglyphus(Sarcoptiformes Astigmata) (Yoshikura 1975) andArchegozetes (Sarcoptiformes Oribatida) (Thomas ampTelford 1999) but not apparently in Tetranychus(Prostigmata) (Dearden Donly amp Grbic 2002)172 Eggsembryos maintained in external attachedbrood sac secreted by genital glands 0 absent 1present [WP sim46 S 59 WH 90 GEWB 208 = 219]State 1 occurs throughout Amblypygi SchizomidaThelyphonida and Pseudoscorpions Giribet et al(2002) followed Shear et al (1987) in homologizingspinneret-derived silken egg sacs of spiders with gen-ital-gland-derived brood sacs of Pedipalpi However asthese structures are not homologous in secretory ori-gin construction or composition (Shultz 1987) theyare not regarded here as homologous Giribet et al(2002) miscoded the character as absent in Pseudo-scorpiones and without justification in PalpigradiFemale palpigrades have well-developed glands asso-ciated with postgenital somites and an array of setaefusules and other structures (112) (Boumlrner 1902b Mil-lot 1942 Condeacute 1991a) which are consistent withsome sort of brood care and Palpigradi should there-fore be coded as unknown for this character173 Embryonic and early postembryonic lsquolateralrsquo orClaparegravede organs associated with coxa of postoralsomite IV (= arachnid leg 2) 0 absent 1 present [WHsim31 GEWB sim211]

PHYLOGENY OF THE ARACHNID ORDERS 263

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bilaterally paired rounded protuberances locatedbetween coxae of appendages of postoral somites IIIand IV (arachnid legs 1 and 2) in prelarval and larvalinstars appear to be a primitive feature of Acariformes(Lindquist 1984 Evans 1992 Alberti amp Coons 1999)It is known to be derived embryologically from thecoxa of leg 2 in the oribatid Archegozetes (Thomas ampTelford 1999) Structures apparently homologouswith the acariform Claparegravede organ are present inembryonic Amblypygi (Weygoldt 2000) and Thely-phonida (Yoshikura 1975) and in embryonic and earlypost-embryonic stages of Solifugae (Roewer 1934)where they are termed lateral organs The so-calledlsquolateral organrsquo of embryonic Xiphosura (174) occurs onthe lateral surface of the carapace and does not appearto be homologous with the coxa-associated structuressharing their name In fact the xiphosuran lateralorgan develops even when the coxa of postoral somiteIV has been excised (Sekiguchi 1988) In contrastWheeler amp Hayashi (1998) and Giribet et al (2002)followed Yoshikura (1975) in regarding all lsquolateralorgansrsquo as homologous and in miscoding this characteras absent in Acariformes Van der Hammen (1989)suggested that the sternal verrucae of Opilioacari-formes are homologous with the Claparegravede organ butit differs in gross structure and ontogenetic timing174 Embryonic lateral organ associated with cara-pace 0 absent 1 present [WH sim31 GEWB sim211]State 1 is known only from embryonic stages of extantXiphosura (Yoshikura 1975) In contrast to previousinterpretations this feature does not appear to behomologous with the lsquolateralrsquo or Clarapegravede organ ofcertain arachnids (173)175 Live birth 0 absent 1 presentState 1 occurs in extant Scorpiones and has evolvedseveral times in a few mites (Evans 1992)176 Hexapodal larva and 1ndash3 nymphal stages 0absent 1 present [WP 53 S sim61 WH 27 GEWB 186]State 1 occurs in Acari and Ricinulei (Lindquist1984)177 Hexapodal prelarva 0 absent 1 present [S sim61WH 47]State 1 occurs in Opilioacariformes (NeocarusKlompen 2000) and many Acariformes (Evans 1992)

EXCRETORYOSMOREGULATORY SYSTEM

178 Malpighian tubules 0 absent 1 present [WP 17S 62 WH 9 GEWB 153]State 1 occurs in all major arachnid groups except Pal-pigradi Pseudoscorpiones Opiliones Oribatida (Sar-coptiformes) and Prostigmata Within Acari State 1occurs in all major anactinotrichid lineages (Opilioac-ariformes Holothyrida Mesostigmata Ixodida) andsimilar structures are present in certain Astigmata(Evans 1992 Alberti amp Coons 1999)

179 Dorsomedian excretory organ 0 absent 1presentThis is a specialized excretoryosmoregulatory organformed by the postventriculus and proctodeum inProstigmata (Alberti amp Coons 1999)180 Adult coxal organ opening on or near coxa ofappendage III (= arachnid leg 1) 0 absent 1 present[S 64 WH 93 GEWB 90]State 1 occurs throughout Acari (Evans 1992 Albertiamp Coons 1999) Ricinulei (Pittard amp Mitchell 1972Legg 1976) Palpigradi (Millot 1942) AraneaeAmblypygi Thelyphonida and Schizomida (Buxton1913 1917) There is evidence that the second lsquoozo-porersquo of gonyleptids (Opiliones Laniatores) (Hara ampGnaspini 2003) is a persistent opening to the coxalorgan associated with leg 1 (Soslashrensen 1879) whichalso occurs in embryonic Phalangium (Opiliones Eup-noi) (Moritz 1959)181 Adult coxal organ opening on or near coxa ofappendage V (= arachnid leg 3) 0 absent 1 present [S63 WH 92 GEWB 89]State 1 occurs in extant Xiphosura (Yamasaki et al1988) Scorpiones mygalomorph (Buxton 1913 1917)and mesothele Araneae (Liphistius J Millot in Bris-towe 1932) State 1 occurs in basal Amblypygi(Charon Charinus Buxton 1913 1917) but degener-ates prior to the adult stage in higher groups(Phrynus Weygoldt 2000) Adequately preservedeurypterids show a small submarginal opening on thecoxa of appendage V (Baltoeurypterus Selden 1981also Eurypterus Hughmilleria Clarke amp Ruedemann1912) and this is interpreted here as the opening to acoxal organ Giribet et al (2002) coded this characteras uncertain for Eurypterida182 Adult coxal organ opening on coxa of appendageII (= arachnid palp) 0 absent 1 present [GEWB 65]Buxton (1913 1917) noted that the coxal organ ofSolifugae opens on the prolateral surface of the palpalcoxa The proximal end of the duct leading to the ori-fice is associated with glands and apparently theglandular secretion andor coxal fluid serve as saliva(Alberti 1979) (see also 14 15)183 Genital papillae 0 absent 1 present - inappli-cable coded only for those taxa with eversible lsquoappend-agesrsquo (111)The genital papillae are associated with the genitalopening in postlarval Acariformes (Evans 1992Alberti amp Coons 1999) These structures are probablyserially homologous with the prosomal Claparegravedeorgan (173) and both structures probably function inwater and ion regulation

DIGESTIVE SYSTEM

184 Ingestion 0 solid food 1 primarily liquid foodwith or without preoral digestion [WP sim16 WH sim8GEWB sim218]

264 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Extant xiphosurans ingest solid food and members ofmost extant arachnid orders ingest fluids Opilionesand Acari are the principal exceptions The phalangidOpiliones ingest solids but Shultz (2000 unpublobserv) has shown that the precerebral pharyngealapparatus of Cyphophthalmi is very similar to that ofthe fluid-feeding Scorpiones and differs substantiallyfrom phalangid Opiliones Examination of gut con-tents of the cyphophthalmids Siro acaroides S exilisand Chileogovea oedipus have failed to reveal solidparticles typically found in the guts of phalangids(Shultz unpubl observ) the character is coded asunknown for Cyphophthalmi Within Acari consump-tion of particles has been documented in Opilioacari-formes (Neocarus Van der Hammen 1989unidentified Walter amp Proctor 1998) Holothyridanon-parasitic Sarcoptiformes (Evans 1992 Walter ampProctor 1998 Alberti amp Coons 1999) and many endeo-stigmatids including the nematophagous Alycusroseus (Walter 1988)185 Mouth 0 directed posteroventrally 1 directedanteroventrally [S 9 WH 55 GEWB 28]State 1 occurs throughout Arachnida (Shultz 1990)186 Oral cavity dilated by muscles arising from coxaeand constricted by large circular sphincter 0 absent1 presentState 1 occurs in extant Xiphosura (Manton 1964Scholl 1977 Shultz 2001)187 Palate plate 0 absent 1 present [GEWB 159]State 1 is a unique synapomorphy of Araneae (Dunlop1994)188 Lateral walls of epistome broadly fused to medialwalls of palpal coxae opposite sides connected by well-developed transverse epistomal muscle 0 absent 1presentState 1 occurs throughout Opiliones and Scorpiones(Shultz 2000 unpubl observ)189 Epistome with a pair of lateral arms that projectsposteriorly into the prosoma on either side of the phar-ynx 0 absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)190 Epistome with four pairs of suspensor musclesattaching to the carapace 0 absent 1 presentState 1 occurs in extant Scorpiones (Lankester et al1885 Vyas 1970 Shultz unpubl observ) and appar-ently in Solifugae (Roewer 1934)191 Intercheliceral epipharyngeal sclerite 0 absent1 present [GEWB 31]State 1 occurs in Palpigradi (Boumlrner 1904) Trigo-notarbida (Palaeocharinus Dunlop 1994) Araneae(Firstman 1954 Marples 1983) Amblypygi (Mil-lot 1949b Shultz 1999) Schizomida (Millot1949e) Thelyphonida (Millot 1949e Shultz 1993)and Phalangida (Opiliones) (Shultz 2000 unpublobserv)

192 Epipharyngeal sclerite large projecting posteri-orly 0 absent 1 present - inapplicable due to absenceof sclerite (191) [GEWB 32]State 1 occurs in Amblypygi (Millot 1949b Shultz1999) Schizomida (Millot 1949e) Thelyphonida (Mil-lot 1949e Shultz 1993) some Araneae (Heptathela ndashcoded as Liphistius ndash and Hypochilus Marples 1983not Aphonopelma Firstman 1954) and perhaps Trig-onotarbida (Palaeocharinus Dunlop 1994)193 Dorsal dilator muscle of precerebral pharynxattaching to intercheliceral septum or associatedepipharyngeal sclerite (191) 0 absent 1 presentState 1 occurs in Palpigradi (Millot 1943) Araneae(Marples 1983) Amblypygi Schizomida and Thely-phonida (Millot 1949e Shultz 1993 1999)Extremely fine muscles have been documented insome Scorpiones (Shultz 2007) and in Leiobunum(Opiliones) (Shultz 2000) Lankester et al (1885)reported a muscle in Limulus that would appear tocorrespond to this muscle but the muscle does notexist (Manton 1964 Shultz 2001)194 Dorsal dilator muscle of precerebral pharynxattaching to dorsal surface of prosoma 0 absent 1present [GEWB 228]State 1 occurs in Palpigradi (Roewer 1934) and mostAraneae (Palmgren 1978 Aphonopelma Firstman1973) but not in Heptathela (coded as Liphistius) orHypochilus (Marples 1983)195 Lateral dilator muscle of precerebral pharynxattaching to endosternite 0 absent 1 present [GEWBsim226]State 1 is a unique feature of Araneae (Marples 1983)196 Lateral dilator muscle of precerebral pharynxattaching to lateral surface of epistomal processes 0absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)197 Lateral dilator muscle of precerebral pharynxattaching to medial process of coxa of appendage II(= arachnid palp) 0 absent 1 present [GEWB sim226]State 1 occurs in Amblypygi Thelyphonida and Schi-zomida (Boumlrner 1904 Shultz 1993 1999)198 Dilator muscle of precerebral pharynx andorpreoral cavity attaching to ventral surface of prosoma0 absent 1 present [GEWB 227]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) Araneae (Marples 1983) Palpigradi(Millot 1943) and Solifugae (Roewer 1934)199 Postcerebral pharynx 0 absent 1 present [WPsim31 S sim5 WH sim21 GEWB sim154]The cuticle-lined foregut passes through the centralnervous system in many arthropods That portion pos-terior to the CNS is here termed the postcerebralpharynx regardless of the details of its morphology orfunctional specialization A substantial postcerebralpharynx is known in extant Xiphosura (Lankester

PHYLOGENY OF THE ARACHNID ORDERS 265

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

et al 1885 Manton 1964 Yamasaki et al 1988)Scorpiones (Centruroides Hadrurus HeterometrusShultz 2007 also Androctonus Abd el-Wahab 1952)Solifugae (Millot amp Vachon 1949) Araneae Ambly-pygi Thelyphonida and Schizomida (Millot 1949andasheShultz 1993) Weygoldt amp Paulus (1979) applied theterm lsquopostcerebral pharynxrsquo only to the lsquosucking stom-achrsquo of Araneae and Amblypygi but the sucking stom-ach is here regarded as a complex of three characters199ndash201

200 Dilator muscle of postcerebral pharynx attach-ing to endosternite 0 absent 1 present - inapplicabledue to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Lankester et al1885 Manton 1964 Shultz 2001) Scorpiones (Cen-truroides Hadrurus Heterometrus Shultz 2007 alsoAndroctonus Abd el-Wahab 1952) Araneae (First-

man 1954 Palmgren 1978) Amblypygi (Millot1949b Shultz 1999) and apparently the palpigradeProkoenenia wheeleri (Rucker 1901) but not inEukoenenia miriabilis (Millot 1943 contra Boumlrner1904)201 Dilator muscle of postcerebral pharynx attachingto dorsal surface of prosoma 0 absent 1 present -inapplicable due to absence of postcerebral pharynx(199)State 1 occurs throughout Araneae (Millot 1949cFirstman 1954 Palmgren 1978) and Amblypygi(Charinus Millot 1949b Phrynus Shultz 1999)202 Crop and gizzard 0 absent 1 present - inap-plicable due to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Sekiguchi 1988)but not in extant Arachnida

PHYLOGENY OF THE ARACHNID ORDERS 231

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 2 Fully resolved phylogenetic hypotheses of extant euchelicerate groups proposed in the recent literature Note thesimilarity in the topologies of the parsimony-based analyses by Shultz (1990) Wheeler amp Hayashi (1998) and Giribet et al(2002) The Giribet et al topology is based on neontological data (morphology and molecules) and the original lsquoROOTrsquo maybe an artefact from use of the highly divergent pycnogonids as an outgroup

Xip

hosu

raS

corp

ione

sO

pilio

nes

Ara

neae

Am

blyp

ygi

Uro

pygi

Ric

inul

eiA

nact

inot

richi

daS

olifu

gae

Pse

udos

corp

ione

sP

alpi

gra

diA

carif

orm

es

Xip

hosu

raS

corp

ione

sA

rane

aeA

mbl

ypyg

iU

ropy

giP

alpi

gra

diS

olifu

gae

Pse

udos

corp

ione

sO

pilio

nes

Ric

inul

eiA

cari

Lipoctena

Arachnida

Apulmonata

Holotracheata

Cryptoperculata

Acaromorpha(ldquoAcarinomorphardquo)

Tetrapulmonata(rdquoMegoperculatardquo)

LabellataHaplo-cnemata

Weygoldt amp Paulus (1979)

Labellata

Tetrapulmonata(rdquoArachnidardquo)

Haplocnemata(rdquoApatellatardquo)

Van der Hammen (1989)

Xip

hosu

raP

alpi

gra

diA

rane

aeA

mbl

ypyg

iU

ropy

giR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Pedipalpi Haplocnemata

Novogenuata

Dromopoda

Shultz (1990)

Acaro-morpha

Micrura

Xip

hosu

raP

alpi

gra

diU

ropy

giA

mbl

ypyg

iA

rane

aeR

icin

ulei

Aca

riO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Megoperculata

Tetrapulmonata

Labellata Haplocnemata

Novogenuata

Dromopoda

Wheeler amp Hayashi (1998)

Acaro-morpha

Micrura

Xip

hosu

raA

cari

Pal

pig

radi

Ric

inul

eiA

rane

aeA

mbl

ypyg

iU

ropy

giO

pilio

nes

Sco

rpio

nes

Pse

udos

corp

ione

sS

olifu

gae

Arachnida

Tetrapulmonata

PedipalpiHaplocnemata

Novogenuata

Dromopoda

Giribet et al (2002)

ROOT

Micrura

Crypto-gnomae

Sternocoxata

Epimerata

RostrosomataMyliosomata

Neosternata

232 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 3 Consensus trees produced by parsimony analysis of neontological data constrained to produce relationships pro-posed in the recent literature The constrained (target) node is indicated by a black dot and the taxa encompassed by theconstraint are enclosed in a box Numbers below each tree represent the number of minimal-length constrained treeslength of minimal-length trees difference in the length of the unconstrained minimal-length tree and the constrained min-imal-length tree and percentage unconstrained bootstrap trees in which the target node was recovered respectively Thesetrees indicate the effect on branch length imposed by specific hypotheses and impact of constraining target nodes on non-target nodes

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Lipoctena4 385 2 lt5

Xi

Sc

Op

Ps

So

Pa

Ur

Am

Ar

Ri

An

Ac

Labellata6 397 14 lt5

Xi

Sc

Ar

Am

Ur

Op

Pa

Ps

So

Ri

An

Ac

Apulmonata2 390 7 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

So

Ri

An

Ac

Holotracheata5 390 7 lt5

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptoperculata2 388 5 lt5

Xi

Sc

Op

Pa

So

Ps

Ar

Am

Ur

Ri

An

Ac

Rostrosomata6 384 1 25

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Neosternata3 388 5 lt5

Xi

Sc

Op

Ar

Am

Ur

Ri

An

Ps

So

Pa

Ac

Coxisternata2 388 5 lt5

Epimerata7 387 4 11

Xi

Pa

Ar

Am

Ur

So

Ps

Sc

Op

Ri

An

Ac

Cryptognomae6 384 1 24

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Micrura3 385 2 7

Xi

Sc

Op

Ps

So

Pa

Ar

Am

Ur

Ri

An

Ac

Megoperculata3 384 1 28

Xi

Sc

Ps

So

Op

Pa

Ar

Am

Ur

Ri

An

Ac

Novogenuata10 388 5 lt5

Xi

Sc

Pa

Ar

Am

Ur

Op

Ps

Ri

So

An

Ac

Solifugae + Acari4 388 5 lt5

Shultz (1990) Alberti amp Peretti (2002)

Van der Hammen (1989)

Weygoldt amp Paulus (1979)

Xi

Ar

Am

Ur

So

Ps

Sc

Op

Ri

Pa

An

Ac

PHYLOGENY OF THE ARACHNID ORDERS 233

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Results from analyses of node-specific hypothesesare summarized in Figures 3 4 In most cases pro-posed relationships were recovered in fewer than 5of bootstrap pseudoreplicates The notable exceptionswere Rostrosomata (25) Cryptognomae (24) andMegoperculata (28) and the strict-consensus con-strained minimal-length tree for each was only onestep longer than the unconstrained optimal tree Inmost cases constrained topologies did not add manysteps most were less than five steps longer and nonewas greater than 14 steps longer than the un-constrained minimal-length tree It is noteworthyhowever that even constrained nodes that imposerelatively few extra steps sometimes supportedimprobable relationships or eliminated clades thatwere well supported or stable in the unconstrainedanalysis Specific examples and their implications arediscussed below

DISCUSSION

RECENT ISSUES IN ARACHNID PHYLOGENY

Available morphological evidence consistently re-solves some interordinal relationships and fails to

resolve others (Fig 1) Continued progress depends onthe ability of arachnologists to discover new charac-ters and to assess the evidence critically Whilereviewing the recent arachnological literature severalaspects of phylogenetic practice emerged that seemedcounterproductive to both the perception and theactual rate of progress toward resolving arachnid phy-logeny Some of these are summarized here withspecific examples given in the remainder of theDiscussion and in the Appendix

There is a tendency to portray arachnid ordinal phy-logeny as more poorly resolved and contentious than isactually the case (Coddington et al 2004) Phyloge-netic hypotheses generated during different historicalperiods and using differing standards of evidence areoften cited as examples of current disagreement (egSelden 1993 Dunlop 1996 Selden amp Dunlop 1998Wheeler amp Hayashi 1998) In fact recent parsimony-based analyses of morphology have tended to convergeon topologies with internal structures congruent withthose found here (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) (Fig 2) Arachnidphylogeny is not fully resolved especially at its deep-est levels (Fig 1) but this does not mean that allaspects of arachnid phylogeny are controversial orpoorly supported by the available evidence

Matrices are sometimes constructed by uncriticallsquorecyclingrsquo of erroneous or problematic charactersbased on diverse secondary sources (Jenner 2001)Conclusions derived from mixtures of valid invalidspeculative and redundant characters are sometimesportrayed as the phylogenetic signal provided by mor-phology Data recycling can perpetuate errors (see 1320 39 52 136 140 144 152 169 171ndash174) legiti-mize speculations (see 13 32 63 95) or create dupli-cate or non-independent characters (see 13 30 70 77172) The lsquolateral organrsquo is a notable exampleYoshikura (1975) equated the embryonicearly post-embryonic lsquolateral organsrsquo of Amblypygi Thelyphonidaand Solifugae with the dissimilar lsquolateral organrsquo ofXiphosura but failed to note the very similar Cla-paregravede organ of Acariformes (see 173 174) This cod-ing was recycled by Wheeler amp Hayashi (1998) andthen by Giribet et al (2002) Error is probably inevi-table when assembling a large matrix from morphol-ogy including the one presented here but this can beminimized by making original observations and byconsulting primary sources

Some workers advocate weighting characters a pri-ori on the basis of structural or functional complexity(eg Kraus 1998 Dunlop amp Braddy 2001) and dismissphylogenetic conclusions derived from equal-weightsparsimony In short these workers criticize parsimonyfor emphasizing data quantity over data quality Thiscriticism ignores the intense debate in systematic biol-ogy that eventually led to widespread adoption of

Figure 4 Consensus trees produced by parsimony analy-sis of the full data matrix constrained to produce relation-ships proposed in the recent literature See legend toFigure 3 for details

Xi

Eu

Sc

Op

Pa

Ac

Ri

An

So

Ps

Tr

Ar

Ha

Am

UrEurypterida slat + Scorpiones12 435 9 lt5

Xi

Eu

Sc

Op

Pa

Ac

An

So

Ps

Ri

Tr

Ar

Ha

Am

UrTrigonotarbida+ Ricinulei28 434 8 lt5

Dunlop amp Braddy (2002)et al

Dunlop (1996)

234 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 5 Results from analysis of neontological data A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology AFor B and C relationships within terminal clades are the same as those shown in A

LimulusTachypleusProkoeneniaEukoeneniaCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAllothyrusAustralothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesLiphistiusAphonopelmaHypochilusCharinusPhrynusStenochrusProtoschizomusMastigoproctusCyphophthalmusChileogoveaCaddoLeiobunumSclerobunusGonyleptesCentruroidesHadrurusHeterometrusChthoniusFeaellaNeobisiumCheliferEremocostaGaleodes

10031

10018451

501561

733

723

944

633

10010

832

901

943

903

522

341

1007

1008

Acaromorpha

Haplocnemata

10012

Stomothecata

Opiliones

Scorpiones

Pseudo-scorpiones

Solifugae

Acariformes

Parasitiformes

Opilioacariformes

Ricinulei

Xiphosura

Palpigradi

Thelyphonida

Schizomida

Araneae

Amblypygi

10011

936853

100141005

1009

Tetrapulmonata Pedipalpi

Uropygi

692

522572

968

331

603

81

511

161

101

Acari

Dromopoda

Arachnida

A

Xiphosura

Stomothecata

Haplocnemata

Acaromorpha

Palpigradi

TetrapulmonataMicrura

Megoperculata

BXiphosura

Stomothecata

Acaromorpha

Haplocnemata

Palpigradi

TetrapulmonataMegoperculata

C

PHYLOGENY OF THE ARACHNID ORDERS 235

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Figure 6 Results from analysis of the full data matrix A minimal-length topology Numbers below internodes are boot-strap percentagesBremer support values B parsimony tree produced by implied weights with k = 1 C parsimony tree pro-duced by implied weights with k = 2 Parsimony trees produced by implied weights with k = 3ndash6 are identical to topology CFor B and C relationships within terminal clades are the same as those shown in A

WeinberginaLimuloidesEuproopsTachypleusLimulusStylonurusBaltoeurypterusChasmataspisDiploaspisOctoberaspisCyphophthamusChileogoveaCaddoLeiobunumSclerobunusGonyleptesProscorpiusPalaeoscorpiusStoermeroscorpioPrearcturusCentruroidesHadrurusHeterometrusProkoeneniaEukoeneniaTerpsicrotonPoliocheraCryptocellusRicinoidesNeocarusSiamacarusAmblyommaArgasAustralothyrusAllothyrusGlyptholaspisAllothrombiumMicrocaeculusAlycusPalaeacarusArchegozetesEremocostaGaleodesChthoniusFeaellaNeobisiumCheliferGilboarachnePalaeocharinusLiphistiusAphonopelmaHypochilusPlesiosiroCharinusPhrynusStenochrusProtoschizomusProschizomusMastigoproctus

631

893

923

984

984

671

341

671

986

10011

423

974

Xiphosura

Eurypterida sstr

Chasmataspidida

Opiliones

Scorpiones

Palpigradi

Ricinulei

Opilioacariformes

Parasitiformes

Acariformes

Solifugae

Pseudoscorpiones

Trigonotarbida

Araneae

Haptopoda

Amblypygi

Schizomida

Thelyphonida

842

251331

361311

552

832

882

925

913

573

561291

443

976

541

954

743644

732521

571

331

833

663591

966

573

71

221

905

832

351

Uropygi

Pedipalpi

Haplocnemata

Tetrapulmonata

Anactinotrichida

Cryptognomae

Acaromorpha

Stomothecata

Pantetrapulmonata

Arachnida

Eurypterida slat

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatProscorpiusStoermeroscorpioPalaeoscorpiusPrearcturusCentruroidesHadrurusHeterometrusOpilionesHaplocnemataRicinuleiAnactinotrichidaAcariformesPalpigradiPantetrapulmonata

Megoperculata

Micrura

WeinberginaLimuloidesEuproopsTachypleusLimulusEurypterida slatPalpigradiStomothecataHaplocnemataPantetrapulmonataAcaromorpha

A

B

C

Schizotarsata

236 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

parsimony and instead advocates a return to thespeculative and subjective approaches of the late 19thand early 20th century that once threatened thescientific legitimacy of the discipline (Bowler 1996)Furthermore it misrepresents the properties of parsi-mony-based analyses Specifically characters withfunctional significance are readily encompassed byparsimony analyses and in fact characters derivedfrom locomotor systems (eg 46ndash94) have played animportant role in developing current ideas aboutarachnid phylogeny (eg Shultz 1989 1991) In addi-tion lsquocomplex charactersrsquo can be viewed as compositesof several characters such that morphological com-plexity is effectively weighted by the number of inde-pendent lsquosubcharactersrsquo it contains For example thelsquosucking stomachrsquo once considered a synapomorphy ofLabellata (= Araneae + Amblypygi Fig 3) is codedhere as a composite of three characters (199ndash201) TheLabellata hypothesis was not corroborated in thepresent analysis (Figs 1 3ndash5) and in fact was highlydisfavoured (Fig 3) but this result cannot be dis-missed as a failure to acknowledge the complexity ofthe character

Some workers support specific (target) clades withone or more similarities without exploring the impacton overall homoplasy or relationships among non-target clades Each character offers its own phyloge-netic hypothesis which may or may not be consistentwith relationships implied by other characters It isexceedingly rare for all characters to be perfectly com-patible in the phylogenetic hypotheses they supportand criteria such as parsimony have been developed todiscover those hypotheses that minimize the conflict-ing phylogenetic signals of different characters Eventhough character conflict (homoplasy) is a virtuallyinescapable phenomenon in comparative biology it isnot uncommon for workers to discover one or morecharacters and to promote their phylogeneticimplications over alternative hypotheses even thosethat otherwise appear to be well supported Thisapproach may have value in highlighting new dataand perspectives but accomplishes this by promotingthe erroneous impression that all aspects of arachnidphylogeny are so tenuous that a single character canfalsify even well-supported hypotheses Several exam-ples of this approach have appeared in the recentarachnological literature

For example Alberti amp Peretti (2002) argued thataflagellate spermatozoa (163) are a compelling syna-pomorphy for a Solifugae + Acari clade and dismissedsome of the characters that support Solifugae + Pseu-doscorpiones as having lsquodebatable valuersquo Yet theirproposal rejects two hypotheses that are consistentlyrecovered in recent phylogenetic analyses that isHaplocnemata (= Solifugae + Pseudoscorpiones) andAcaromorpha (= Acari + Ricinulei) (Figs 1 2) Rejec-

tion of Haplocnemata would require its presumed syn-apomorphies to be reinterpreted as homoplasiesincluding features of the chelicerae (18 19 20) pre-oral chamber (13 32) legs (12 48) and respiratorysystem (126) The same reason applies to Acaromor-pha and its synapomorphies It is noteworthy that aSolifugae + Acari clade was recovered in the presentanalysis in fewer than 5 of bootstrap pseudorepli-cates and that analyses constrained to recover thisclade were five steps longer than the minimal-lengthtree and favoured a problematic clade unitingRicinulei and Pseudoscorpiones (Fig 3)

In another example Dunlop (1996) proposed a closerelationship between Trigonotarbida and Ricinuleibased on two-segmented chelicerae (18) prosomandashopisthosoma coupling mechanism (96) diplotergites(100 101) and longitudinally divided opisthosomaltergites (115) However phylogenetic analyses con-strained to recover this relationship required eightadditional steps eliminated support for Acari and nec-essarily rejected Acaromorpha and forced its synapo-morphies to be reinterpreted as homoplasies Alberti(2005) proposed an interesting hypothesis for the evo-lution of male gonads in tetrapulmonates (153) butchose to accept Labellata (= Araneae + Amblypygi) indeveloping his argument over the much more well-supported Pedipalpi (= Amblypygi + Uropygi) (Shultz1999) a phylogenetic reconfiguration that was recov-ered here in fewer than 5 of bootstrap pseudorepli-cates increased tree length by a minimum of 14 stepsand resulted in the collapse of Stomothecata and Hap-locnemata (Fig 3) Many other examples can be cited

Promoting or defending a specific phylogenetichypothesis via lists of compatible synapomorphies is acommon but problematic approach By restrictingattention to the states of specific characters at one ortwo target nodes one can easily overlook the unin-tended impact of the hypothesis on phylogenetic sig-nal elsewhere and its effect on non-target clades Anode supported by a long list of synapomorphies mayseem convincing taken in isolation but may becomeless acceptable when its full phylogenetic implicationsare explored

SUMMARY OF INTERORDINAL ARACHNID CLADES RECOVERED IN THIS ANALYSIS

Arachnida Lamarck 1801Analyses consistently recovered Arachnida as a mono-phyletic group with high bootstrap support (Figs 1 45) Possible synapomorphies include the loss of thecarapacal pleural doublure (9) cardiac lobe (10) pedalgnathobases (52) and moveable endites (53) and thegain of aerial respiration (120) and an anteriorly oranteroventrally directed mouth (185) Some tradi-tional synapomorphies such as slit sensilla (142) and

PHYLOGENY OF THE ARACHNID ORDERS 237

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

fluid feeding (184) may have appeared later in arach-nid evolution but this can only be decided once theinternal phylogeny of Arachnida has been determined

Some workers regard many proposed arachnid syn-apomorphies as adaptations to terrestrial life andthereby link the hypothesis of arachnid monophyly tothe hypothesis of a single ancestral aquatic-to-terres-trial transition and arachnid polyphyly to multipletransitions (eg Selden amp Jeram 1989 Dunlop 1997Dunlop amp Webster 1999 Dunlop amp Braddy 2001)Given this line of reasoning the existence of appar-ently aquatic scorpions in the fossil record (Kjellesvig-Waering 1986 Jeram 1998) and the inference thatterrestrialization occurred late in scorpion evolutionthese workers conclude that character states support-ing arachnid monophyly are actually convergencesand do not necessarily support arachnid monophyly

However this approach to assessing phylogenetichypotheses is founded on overly simplistic assump-tions such as the ability of the investigator to discrim-inate unerringly between characters that existexclusively in aquatic organisms (including fossils)from those that occur exclusively in terrestrial organ-isms There also appears to be an assumption thathomoplasy can be generated through parallelism (iemultiple aquatic-to-terrestrial events) but not throughterrestrial-to-aquatic reversals Furthermore anassumed dichotomy between exclusively aquatic andterrestrial life histories in ancestral arachnids is sim-plistic as illustrated by the amphibious life cycles ofbasal vertebrates and pterygote hexapods In factthese examples demonstrate that there is no neces-sary inconsistency in basing a hypothesis of arachnidmonophyly on the derived terrestrial features of anamphibious ancestor whose descendants thencompleted terrestrialization once or several timesindependently or even returned to a fully aquaticexistence Workers who link the frequency and direc-tion of aquaticndashterrestrial transitions to the assess-ment of arachnid phylogeny do so by endowingthemselves with substantially greater insight thanseems prudent by ignoring the huge gaps in ourunderstanding of early arachnid evolution and bydenying to arachnids the evolutionary complexityknown to exist in other groups

Several palaeontologists have been particularlyactive during the past decade in proposing new char-acters with the stated goal of removing Scorpionesfrom Arachnida and erecting a Scorpiones + Eurypter-ida clade (Braddy amp Dunlop 1997 Dunlop amp Braddy1997 Dunlop 1998 Braddy et al 1999 Dunlop ampWebster 1999) Dunlop amp Braddy (2001) recentlysummarized this evidence and conducted aparsimony-based analysis of Xiphosura EurypteridaScorpiones Opiliones and Tetrapulmonata (but notHaplocnemata Acaromorpha or Palpigradi) using 33

morphological characters Their analysis produced atopology congruent with those generated here (Figs 15 6) including recovery of Stomothecata (= Scorpiones+ Opiliones) However they rejected this result as aproduct of lsquoempirical cladisticsrsquo because it gives thesame weight to prosomal characters that supportarachnid monophyly and to selected opisthosomalcharacters that support their favoured Eurypterida +Scorpiones clade namely a five-segmented metasoma(116) suppression of opisthosomal tergite 1 (95) lossof lamellate respiratory organs on the postgenitalsomite (122) Kiemenplatten (125) loss of respiratorylamellae on the genital segment (121) and a lsquonon-stainingrsquo exocuticle (but see Grainge amp Pearson 1966for evidence of this in Opiliones see Appendix for com-ments on the other characters)

A Eurypterida + Scorpiones clade was not favouredin the present analysis (Fig 6) the strict consensus ofminimal-length trees constrained to recover this clade(Fig 4) is nine steps longer than the tree recovered byanalysis without this constraint It is also noteworthythat Opiliones was consistently reconstructed as thesister group to Eurypterida + Scorpiones in the mini-mal-length constrained trees a provocative result thatwas probably unintended and unanticipated by sup-porters of the Eurypterida + Scorpiones Given thatDunlop amp Braddy (2001) reject equal weights parsi-mony as an arbiter of phylogenetic hypotheses theywould presumably dismiss these results as irrelevantto their argument just as they dismissed their ownparsimony-based results However if the Eurypterida+ Scorpiones hypothesis is to be credible it must beopen to evaluation and potential falsification usingobjective criteria and the subjective or intuitive a pri-ori weighting of characters advocated by Dunlop ampBraddy clearly does not qualify At present it isunclear how one would objectively evaluate Dunlop ampBraddyrsquos proposal with criteria compatible with thoseused in its original formulation For now the Euryp-terida + Scorpiones concept advocated by Dunlop ampBraddy may persist outside the mainstream of mod-ern systematic practice but it is increasingly problem-atic within it

Stomothecata nomnovOpiliones and Scorpiones were consistently recoveredas a monophyletic group The proposed name acknowl-edges a unique preoral chamber the stomothecaformed by coxapophyses of the palp and leg 1 (50)often with an auxiliary role played by the coxapophy-sis of leg 2 (51) In addition the epistome appears tohave been modified for adduction of the palpal coxaeThe lateral walls of the epistome are fused to themedial surfaces of the palpal coxae and the epistomallumen is spanned by a transverse muscle (188) which

238 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

apparently adducts the palpal coxae thereby constrict-ing the stomothecal chamber Scorpions and opilionsare also unique in having a pair of large epistomalarms projecting rearward into the prosoma andattaching to the endosternite (189) The epistomalarms provide attachment sites for pharyngeal dilatormuscles (196) and extrinsic muscles of anterior proso-mal appendages (eg 37) The chelicera is equippedwith a muscle that arises on the carapace and insertson the ventral margin of the second cheliceral article(23) There is an anteriorly placed genital opening(155)

Several notable similarities were found whilereviewing the literature but information from otherarachnid groups was considered too incomplete toallow them to be included in the matrix For exampleboth orders have apparent haemocytopoeitic organsassociated with major nerves of the anterior opistho-soma These are termed supraneural organs in scorpi-ons (Farley 1999) and perineural organs in opilions(Kaumlstner 1935) Haemocytes develop in the cardiacwall in spiders (Seitz 1972) and perhaps amblypygids(Weygoldt 2000) but haemocytopoeitic organs areunknown in most other arachnid groups Germ cellsdifferentiate very early during embryogenesis in bothscorpions and opilions relative to spiders (Moritz1957 Anderson 1973) Additional research is requiredto determine the phylogenetic utility of thesecharacters

Haplocnemata Boumlrner 1904Several workers recognize a close phylogenetic rela-tionship between Pseudoscorpiones and Solifugae(Weygoldt amp Paulus 1979 Van der Hammen 1989Shultz 1990 Wheeler amp Hayashi 1998 Giribetet al 2002) and this was consistently recovered inthe present analysis Haplocnemata is supported byseveral synapomorphies including (i) a feeding com-plex formed by two-segmented chelicerae (18) with aventrolateral intrinsic articulation (19) and perhapsa dorsolateral articulation with the carapace (20)and (ii) a rostrosoma (32) a preoral apparatus formedby an anteriorly projecting epistome affixed dorsallyto enlarged palpal coxae and bordered by lateral pal-pal projections The epistome forms the dorsal compo-nent of a lsquobeakrsquo with the ventral part formed by amidventral sternapophysis (13) fused between thepalpal coxae The locomotor apparatus consists ofcoxae that meet along the ventral midline (12) elon-gate femur-like patellae (48) and an apotele in theadult modified as an eversible empodium or pulvillus(92) The respiratory system is formed entirely (Pseu-doscorpiones) or in part (Solifugae) by paired trachealstigmata opening on opisthosomal somites 3 and 4(126)

Acaromorpha Dubinin 1957A clade comprising Ricinulei and Acari is consistentlyrecovered in the present analysis It is united here bytwo unique and seven homoplasious synapomorphiesThese include a gnathosoma (31) defined in part bymedial fusion of the palpal coxae (30) although pres-ence of a gnathosoma in Ricinulei is debatable as dis-cussed in the Appendix Acaromorphs also have aunique post-embryonic development consisting of ahexapodal larva and up to three octopodal nymphalinstars (176) The pedal patellandashtibia joints are formedby a bicondylar dorsal hinge rather than a single mid-dorsal condyle (69) all postcheliceral segmental gan-glia are unified in the subesophageal ganglion (130)and a postcerebral pharynx is absent (192) The groupis also tentatively united by presence of differentiatedpedal basi- and telofemora (63 64) and the absence ofa ventral (sternal) pharyngeal dilator muscle (199)but these may be symplesiomorphies erroneouslyreconstructed as a synapomorphies

The internal phylogenetic structure of Acaromorphais controversial with many recent workers favouringa monophyletic Acari (Weygoldt amp Paulus 1979Shultz 1990 argued most thoroughly by Lindquist1984) and others advocating a diphyletic Acari withAcariformes being the sister group to Cryptognomae(= Ricinulei + Anactinotrichida) (especially Van derHammen 1979 1989) (Fig 3) Acari was recovered asmonophyletic when fossils were excluded (Fig 2) butnot when they were included (Fig 3) A lsquomite-centredrsquosurvey of arachnid characters may be needed if mor-phology is to offer a compelling solution to the internalphylogeny of Acari and its placement within Arach-nida These issues are not resolved by the presentanalysis

Pantetrapulmonata nom novPantetrapulmonata includes the extinct orders Trigo-notarbida Haptopoda and the extant orders AraneaeAmblypygi Schizomida and Thelyphonida The cladeis united by cheliceral structure (18 19) a megoper-culum (106) booklungs on the genital and first post-genital somites (121 122) and enlargement of theepipharyngeal sclerite (192) It is important to notehowever that most of these characters were coded asuncertain in Plesiosiro (Haptopoda) Aside from theplacement of Haptopoda the monophyly of Trigono-tarbida and the extant orders was anticipated byShear et al (1987) and is generally regarded as amonophyletic group

Tetrapulmonata Shultz 1990Tetrapulmonata was originally proposed on the basisof neontological analyses and encompassed Araneae

PHYLOGENY OF THE ARACHNID ORDERS 239

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Amblypygi Schizomida and Thelyphonida (Shultz1990) and was also recovered here in analysis of neon-tological data (Fig 5) However results from analysisof all taxa required that the Tetrapulmonata conceptbe expanded to include the fossil order HaptopodaFeatures uniting this group are problematic howeveras many states in Haptopoda were coded as uncertain

Schizotarsata nomnovHaptopoda and Pedipalpi (= Amblypygi + Schizomida+ Thelyphonida) are united here in a group named forpossession of divided pedal telotarsi (84) Synapomor-phies include a pointed anterior carapacal margin (3)and elongation of leg 1 (46) As already noted theplacement of Haptopoda should probably be regardedas tentative because the state of many characters inthis group was coded as unknown The placement ofHaptopoda as the sister group to Pedipalpi was antic-ipated by Dunlop (1999 2002c)

Pedipalpi Boumlrner 1904Pedipalpi encompasses Amblypygi and Uropygi(= Thelyphonida + Schizomida) Although widely rec-ognized in the past Pedipalpi was set aside in the mid-20th century in favour of Labellata (= Amblypygi +Araneae) based largely on the presumed synapomor-phies of a narrow prosomandashopisthosoma juncture (97)and sucking stomach (199ndash201) The situation beganto reverse when Shear et al (1987) again highlightedsimilarities of Amblypygi and Uropygi such as rapto-rial palps (35) and antenniform leg 1 (46) Shultz(1989 1990 1999) subsequently described numerousderived similarities including asymmetrical flexormuscles at the pedal femurndashpatella joint (70) threetelotarsomeres on legs 2ndash4 (85) and modification ofthe palpal coxae to provide support for extrinsic pha-ryngeal muscles (197) Pedipalpi has emerged as oneof the most well-supported interordinal relationshipsin Arachnida

Uropygi Thorell 1882Uropygi has long been accepted as a monophyleticunion of Thelyphonida and Schizomida Synapomor-phies include a unique mating behaviour (159) fusedpalpal coxae (30) 2-1-1-1 arrangement of tibial tricho-bothria (144) posterior defensive glands (102) elon-gated patella of leg 1 (68) and many others

ACKNOWLEDGEMENTS

I thank Nikolaj Scharff and three anonymous review-ers for useful comments This work was supported bythe Maryland Agricultural Experiment Station the

General Research Board of the University of Mary-land and National Science Foundation Grant 0416628to Susan Masta

REFERENCES

Abd el-Wahab A 1952 Notes on the morphology of the scor-pion Buthus quinquestriatus (HE) Publications de lrsquoInsti-tut Fouad 1er du Desert 3 1ndash129

Alberti G 1979 Licht- und elektronenmikroskopische Unter-suchungen an Coxaldruumlsen von Walzenspinnen (ArachnidaSolifugae) Zoologischer Anzeiger 203 48ndash64

Alberti G 1995 Comparative spermatology of Cheliceratareview and perspective Memoires du Museum NationaldrsquoHistoire Naturelle 166 203ndash230

Alberti G 2005 Double spermatogenesis in ChelicerataJournal of Morphology 266 281ndash297

Alberti G Coons LB 1999 Acari mites In Harrison FWFoelix RF eds Microscopic anatomy of invertebratesVol 8C Chelicerate Arthropoda New York Wiley Liss515ndash1215

Alberti G Peretti AV 2002 Fine structure of male genitalsystem and sperm in Solifugae does not support a sister-group relationship with Pseudoscorpiones (Arachnida)Journal of Arachnology 30 268ndash274

Alexander AJ 1967 Problems of limb extension in the scor-pion Opisthophthalmus latimanus Koch Transactions of theRoyal Society of South Africa 37 165ndash181

Anderson DT 1973 Embryology and phylogeny in annelidsand arthropods Oxford Pergamon Press

Anderson LI Selden PA 1997 Opisthosomal fusion andphylogeny of Palaeozoic Xiphosura Lethaia 30 19ndash31

Beall BS Labandeira CC 1990 Macroevolutionary patternsof the Chelicerata and Tracheata In Culver SJ ed Arthro-pod paleobiology Short Courses in Paleontology No 3 Lon-don The Paleontological Society 257ndash284

Bernard HM 1896 The comparative morphology of the Gale-odidae Transactions of the Linnean Society of London Sec-ond Series 6 305ndash417

Boumlrner C 1902a Arachnologische Studien (II und III) Zool-ogischer Anzeiger 25 433ndash466

Boumlrner C 1902b Arachnologische Studien IV Die Genitalor-gane der Pedipalpen Zoologischer Anzeiger 26 81ndash92

Boumlrner C 1904 Beitraumlge zur Morphologie der Arthropoden IEin Beitrag zur Kenntnis der Pedipalpen Zoologica (Stut-tgart) 42 1ndash174

Bowler PJ 1996 Lifersquos splendid drama Chicago Universityof Chicago Press

Braddy SJ Alridge RJ Gabbott SE Theron JN 1999Lamellate book-gills in a late Ordovician eurypterid from theSoom Shale South Africa support for a eurypterid-scorpionclade Lethaia 32 72ndash74

Braddy SJ Dunlop JA 1997 The functional morphology ofmating in the Silurian eurypterid Baltoeurypterus tet-ragonophthalmus (Fischer 1839) Zoological Journal of theLinnean Society 121 435ndash461

Brauer A 1895 Beitrage zur Kenntnis der Entwicklungsge-schichte des Skorpiones II Zeitschrift fuumlr WissenschaftlicheZoologie 59 351ndash433

240 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bremer K 1994 Branch support and tree stability Cladistics10 295ndash304

Bristowe WS 1932 The liphistiid spiders Proceedings of theZoological Society of London Part 4 1016ndash1057

Butt AG Taylor HH 1991 The function of spider coxalorgans effects of feeding and salt-loading on Porrhotheleantipodiana (Mygalomorpha Dipluridae) Journal of Exper-imental Biology 158 439ndash461

Buxton BH 1913 Coxal glands of arachnids ZoologischeJahrbucher (Suppl XIV) 16ndash50

Buxton BH 1917 Notes on the anatomy of arachnids Jour-nal of Morphology 29 1ndash31

Camin JH Clark GM Bourdeau FG 1956 The palpallsquotined setarsquo in the Mesostigmata a homologue of the palpalclaw in the Onychopalpida (Acarina) In Becker EC ed Pro-ceedings of the Tenth International Congress of EntomologyOttawa Mortimer 903ndash908

Chamberlin JC 1931 The arachnid order ChelonethidaStanford University Stanford University Press

Chapman RF 1998 The insects structure and function 4thedn Cambridge Cambridge University Press

Clarke KU 1979 Visceral anatomy and arthropod phylogenyIn Gupta AP ed Arthropod phylogeny New York Van Nos-trand Reinhold 467ndash549

Clarke J 1984 On the relationship between structure andfunction in the leg joints of Heteropoda venatoria (L) (Ara-neae Eusparassidae) Bulletin of the British ArachnologicalSociety 6 181ndash192

Clarke J 1986 The comparative functional morphology of theleg joints and muscles of five spiders Bulletin of the BritishArachnological Society 7 37ndash47

Clarke JM Ruedemann R 1912 The Eurypterida of NewYork New York State Museum Memoir 14 (1ndash2) 1ndash628

Coddington JA 2005 Phylogeny and classification of spi-ders In Ubick D Paquin P Cushing PE Roth V eds Spi-ders of North America an identification manual AmericanArachnological Society 18ndash24

Coddington JA Giribet G Harvey MS Prendini LWalter DE 2004 Arachnida In Cracraft J Donoghue MJeds Assembling the tree of life Oxford Oxford UniversityPress 269ndash318

Coddington JA Levi HW 1991 Systematics and evolutionof spiders (Araneae) Annual Review of Ecology and System-atics 22 565ndash592

Coineau Y Van der Hammen L 1979 The postembryonicdevelopment of Opilioacarida with notes on new taxa and ona general model for the evolution In Piffle E ed Proceed-ings of the 4th International Congress of Acarology Budap-est Akadeacutemiai Kiadoacute 437ndash441

Cokendolpher JC Reddell JR 1992 Revision of the Pro-toschizomidae (Arachnida Schizomida) with notes on thephylogeny of the order Texas Memorial Museum Speleolog-ical Monographs 3 31ndash74

Condeacute B 1991a Le dimorphisme sexuel des Palpigrades Bul-letin de la Socieacuteteacute Neuchacircteloise des Sciences Naturelles 11667ndash73

Condeacute B 1991b Lrsquoaire genitale male des Palpigrades Bulle-tin de la Socieacuteteacute Europeenne drsquoArachnologie 1990 64ndash69

Dearden PK Donly C Grbicccc M 2002 Expression of pair-rule gene homologues in a chelicerate early patterning ofthe two-spotted spider mite Tetranychus urticae Develop-ment 129 5461ndash5472

Delle Cave L 1975 The ldquocleaning brushrdquo in Damon diademaSimon 1876 and Phrynichus jayakari Pocock 1894 (Ambly-pygi Arachnida) Monitore Zoologico Italiano (NS) 9 175ndash184

Dunlop JA 1994 Filtration mechanisms in the mouthparts oftetrapulmonate arachnids (Trigonotarbida Araneae Ambly-pygi Uropygi Schizomida) Bulletin of the British Arachno-logical Society 9 267ndash273

Dunlop JA 1996 Evidence for a sister group relationshipbetween Ricinulei and Trigonotarbida Bulletin of the BritishArachnological Society 10 193ndash204

Dunlop JA 1997 Palaeozoic arachnids and their significancefor arachnid phylogeny In Zabka M ed Proceedings of the16th European colloquium of arachnology Siedlce PolandWyzsza Szkola Rolnicko-Pedagogiczna 65ndash82

Dunlop JA 1998 The origins of tetrapulmonate book lungsand their significance for chelicerate phylogeny In SeldenPA ed Proceedings of the 17th European Colloquium ofArachnology Edinburgh 1997 Burnham Beeches BritishArachnological Society 9ndash16

Dunlop JA 1999 A redescription of the Carboniferous arach-nid Plesiosiro madeleyi Pocock 1911 (Arachnida Hapto-poda) Transactions of the Royal Society of Edinburgh EarthSciences 90 29ndash47

Dunlop JA 2000 The epistomo-labral plate and lateral lips insolifuges pseudoscorpions and mites Ekoloacutegia (Bratislava)19 67ndash78

Dunlop JA 2002a Arthropods from the Lower Devonian Sev-ernaya Zemlya Formation of October Revolution Island(Russia) Geodiversitas 24 349ndash379

Dunlop JA 2002b Character states and evolution of the che-licerate claws In Toft S Scharff N eds European Arach-nology 2000 Aarhus Aarhus University Press 345ndash354

Dunlop JA 2002c Phylogeny of Chelicerata In LlorenteBousquets J Morrone JJ eds Biodiversidad Taxonomiacutea yBiogeografiacutea de Arthroacutepodos de Meacutexico Hacia Una Siacutentesisde Su Conocimiento Vol III Prensas de Ciencias Univer-sidad Nacional Autonoma de Meacutexico 117ndash141

Dunlop JA Anderson LI Braddy SJ 2004 A redescriptionof Chasmataspis laurencii Caster amp Brooks 1956 (Chelicer-ata Chasmataspidida) from the Middle Ordovician of Ten-nessee USA with remarks on chasmataspid phylogenyTransactions of the Royal Society of Edinburgh Earth Sci-ences 94 207ndash225

Dunlop JA Braddy SJ 1997 Slit-like structures on theprosomal appendages of the eurypterid BaltoeurypterusNeues Jahrbuch fuumlr Geologie und Palaumlontologie-Monatshefte1997 31ndash38

Dunlop JA Braddy SJ 2001 Scorpions and their sister-group relationships In Fet V Selden PA eds Scorpions2001 in Memoriam Gary A Polis London British Arachno-logical Society 1ndash24

Dunlop JA Horrocks CA 19951996 A new Upper Carbon-iferous whip scorpion (Arachnida Uropygi Thelyphonida)

PHYLOGENY OF THE ARACHNID ORDERS 241

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

with a revision of the British Carboniferous Uropygi Zoolo-gischer Anzeiger 234 293ndash306

Dunlop JA Poschmann M Anderson L 2001 On theEmsian (early Devonian) arthropods of the Rhenish SlateMountains 3 The chasmataspidid Diploaspis Palaumlon-tolgische Zeitschrift 75 253ndash269

Dunlop JA Webster M 1999 Fossil evidence terrestrializa-tion and arachnid phylogeny Journal of Arachnology 27 86ndash93

Edgecombe GD Wilson GDF Colgan DJ Gray MR Cas-sis G 2000 Arthropod cladistics combined analysis of his-tone H3 and U2 snRNA sequences and morphologyCladistics 16 155ndash203

Evans GO 1992 Principles of acarology Wallingford UKCABI

Fahrenbach WH 1999 Merostomata In Harrison FW Foe-lix RF eds Microscopic anatomy of invertebrates Vol 8AChelicerate arthropods New York Wiley-Liss 21ndash115

Farley RD 1999 Scorpiones In Harrison FW Foelix RF edsMicroscopic anatomy of invertebrates Vol 8A Cheliceratearthropods New York Wiley-Liss 117ndash222

Farley RD 2005 Developmental changes in the embryo pro-nymph and first molt of the scorpion Centruroides vittatus(Scorpiones Buthidae) Journal of Morphology 265 1ndash27

Fayers SR Dunlop JA Trewin NH 2004 A new earlyDevonian trigonotarbid arachnid from the Windyfield ChertRhynie Scotland Journal of Systematic Palaeontology 2269ndash284

Felsenstein J 1985 Confidence limits on phylogenies anapproach using the bootstrap Evolution 39 783ndash791

Fet V Sissom WD Lowe G Braunwalder ME 2000 Cat-alog of the scorpions of the world (1758ndash1998) New YorkNew York Entomological Society

Firstman B 1954 The central nervous system musculatureand segmentation of the cephalothorax of a tarantula (Eu-rypelma californicum Ausserer) Microentomology 87 1ndash40

Firstman B 1973 The relationship of the chelicerate arterialsystem to the evolution of the endosternite Journal ofArachnology 1 1ndash54

Foelix RF 1985 Mechano- and chemoreceptive sensilla InBarth FG ed Neurobiology of arachnids Berlin Spinger-Verlag 118ndash137

Foelix RF 1996 Biology of spiders 2nd edn Oxford OxfordUniversity Press

Giribet G Edgecombe GD Wheeler WC Babbit C 2002Phylogeny and systematic position of Opiliones a combinedanalysis of chelicerate relationships using morphologicaland molecular data Cladistics 18 5ndash70

Goloboff P 1993 Estimating character weights during treesearch Cladistics 9 83ndash91

Goloboff PA Farris JS Nixon K 2000 TNT Tree AnalysisUsing New Technologies Version 1 Program and documen-tation httpwwwzmucdkpublicphylogenytnt

Grainge CA Pearson RG 1966 Cuticular structure in thePhalangida Nature 211 866

Grandjean F 1938 Observations sur les Bdelles (Acariens)Annales de la Socieacuteteacute Entomologique de France Paris 107 1ndash24

Hansen HJ Soslashrensen W 1904 On two orders of ArachnidaCambridge Cambridge University Press

Hansen HJ Soslashrensen W 1905 The Tartarides a tribe of theorder Pedipalpi Arkiv fuumlr Zoologi 2 1ndash78

Hara MR Gnaspini P 2003 Comparative study of the defen-sive behavior and morphology of the gland opening areaamong harvestmen (Arachnida Opiliones Gonyleptidae)under a phylogenetic perspective Arthropod Structure andDevelopment 32 257ndash275

Harvey MS 1992 The phylogeny and classification of thePseudoscorpionida (Chelicerata Arachnida) InvertebrateTaxonomy 6 1373ndash1435

Harvey MS 2002 The neglected cousins what do we knowabout the smaller arachnid orders Journal of Arachnology30 357ndash372

Haupt J Song D 1996 Revision of East Asian whip scorpions(Arachnida Uropygi Theyphonida) I China and JapanArthropoda Selecta 5 43ndash52

Hjelle JT 1990 Anatomy and morphology In Polis GA edThe biology of scorpions Stanford University Stanford Uni-versity Press 9ndash63

Janczyk FSW 1956 Anatomie von Siro duricorius Joseph imVergleich mit anderen Opilioniden Sitzungsberichte Oster-reichischen Akademie der Wissenschaften Mathematisch-Natruwissenschaftliche Klasse 165 474ndash522

Jenner RA 2001 Bilaterian phylogeny and uncritical recy-cling of morphological data sets Systematic Biology 50 730ndash742

Jeram AJ 1998 Phylogeny classification and evolution of Sil-urian and Devonian scorpions In Selden PA ed Proceed-ings of the 17th European Colloquium of ArachnologyEdinburgh 1997 Burnham Beeches UK British Arachno-logical Society 15ndash31

Karaman IM 2005 Evidence of spermatophores in Cyphoph-thalmi (Arachnida Opiliones) Revue Suisse de Zoologie 1123ndash11

Kaumlstner A 1935 Die Funktion der sogenannten sympathis-chen Ganglien und die Exkretion bei den PhalangiidenZoologischer Anzeiger 109 273ndash288

Kierans JE Robbins RG 1999 A world checklist of generasubgenera and species of ticks (Acari Ixodida) publishedfrom 1973 to 1997 Journal of Vector Ecology 24 115ndash129

Kjellesvig-Waering EN 1986 A restudy of the fossil Scorpi-onida of the world Palaeontographica Americana 55 1ndash287

Klompen JSH 2000 Prelarva and larva of Opilioacarus (Neo-carus) texanus (Chamberlin and Mulaik) (Acari Opilioacar-ida) with notes on the patterns of setation and lyrifissuresJournal of Natural History 34 1977ndash1992

Kraus O 1998 Elucidating the historical process of phylog-eny phylogenetic systematics versus cladistic techniquesIn Selden PA ed Proceedings of the 17th European Collo-quium of Arachnology Edinburgh 1997 Burnham BeechesUK British Arachnological Society 1ndash7

Kraus O Kraus M 1993 Divergent transformation of cheli-cerae and original arrangement of eyes in spiders (Arach-nida Araneae) Memoirs of the Queensland Museum 33579ndash584

Lankester ER Benham WBS Beck EJ 1885 On the mus-

242 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

cular and endoskeletal systems of Limulus and Scorpio withsome notes on the anatomy and generic characters of scor-pions Transactions of the Zoological Society of London 11311ndash384

Leclerc P 1989 Considerations paleobiogeographiques a pro-pos la decouverte en Thailande drsquoopilioacariens nouveaux(Acari ndash Notostigmata) Compte Rendu des Seances de laSociete de Biogeographie 65 162ndash174

Legg G 1976 The external morphology of a new species ofricinuleid (Arachnida) from Sierra Leone Zoological Journalof the Linnean Society 59 1ndash58

Lindquist EE 1984 Current theories on the evolution ofmajor groups of Acari and on their relationships with othergroups of Arachnida with consequent implications for theirclassificaton In Griffiths DA Bowman CE eds AcarologyVI Vol 1 New York John Wiley amp Sons 28ndash62

Manning PL Dunlop JA 1995 The respiratory organs ofeurypterids Palaeontology 38 287ndash297

Manton SM 1958 Hydrostatic pressure and leg exten-sion in arthropods with special reference to arachnidsAnnals and Magazine of Natural History Series 13 1161ndash182

Manton SM 1964 Mandibular mechanisms and the evolutionof arthropods Philosophical Transactions of the Royal Soci-ety of London B 7 1ndash183

Maraun M Heethoff M Schneider K Scheu S WeigmannG Cianciolo J Thomas RH Norton RA 2004 Molecularphylogeny of oribatid mites (Oribatida Acari) evidence formultiple radiations of parthenogenetic lineages Experimen-tal and Applied Acarology 33 183ndash201

Marples BJ 1967 The spinnerets and epiandrous glands ofspiders Journal of the Linnean Society (Zoology) 46 209ndash222

Marples BJ 1983 Observations on the structure of the fore-gut of spiders Bulletin of the British Arachnological Society8 46ndash52

Millot J 1933 Notes compleacutementaires sur lrsquoanatomie desliphistiides et des hypchilides a propos drsquoun travail reacutecent deA Petrunkevitch Bulletin de la Socieacuteteacute Zoologique de France68 217ndash235

Millot J 1942 Sur lrsquoanatomie et lrsquohistophysiologie deKoenenia mirabilis Grassi (Arachnida Palpigradi) RevueFranccedilaise drsquoEntomologie 9 33ndash51

Millot J 1943 Notes complementaires sur lrsquoanatomielrsquohistologie et la repartition geographique en France deKoenenia mirabilis Grassi Revue Franccedilaise drsquoEntomologie 9127ndash135

Millot J 1945 Lrsquoanatomie interne des Ricinulei (Arachnides)Annales des Sciences Naturelles (Zoologie) Series 11 7 1ndash29

Millot J 1949a Morphologie generale et anatomie interne InGrasseacute P-P ed Traite de zoologie VI Paris Masson et Cie263ndash319

Millot J 1949b Ordre des Amblypyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 563ndash588

Millot J 1949c Ordre des Araneacuteides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 589ndash743

Millot J 1949d Ordre des Palpigrades In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 520ndash532

Millot J 1949e Ordre des Uropyges In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 533ndash562

Millot J 1949f Ordre des Ricinuleides In Grasseacute P-P edTraite de zoologie VI Paris Masson et Cie 744ndash760

Millot J Vachon M 1949 Ordre Des Solifuges In Grasseacute P-P ed Traite de zoologie VI Paris Masson et Cie 482ndash519

Moore RA Briggs DEG Bartels C 2005 A new specimen ofWeinbergina opitzi (Chelicerata Xiphosura) from the LowerDevonian Huumlnsruck Slate Germany PalaumlontologischeZeitschrift 79 399ndash408

Moritz M 1957 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) unter besonderer Beruumlcksichtigungder ausseren Morphologie der Bildung des Mitteldarmesund der Genitalanlage Zoologische Jahrbuumlcher Abteilungfuumlr Anatomie und Ontogenie der Tiere 76 331ndash370

Moritz M 1959 Zur Embryonalentwicklung der Phalangiiden(Opiliones Palpatores) II Die Anlage und Entwicklung derCoxaldruumlse bei Phalangium opilio L Zoologische Jahr-buumlcher Abteilung fuumlr Anatomie und Ontogenie der Tiere 77229ndash240

Nixon K 1999 The parsimony ratchet a new method forrapid parsimony analysis Cladistics 15 407ndash414

Norton RA 1998 Morphological evidence for the evolutionaryorigin of Astigmata (Acari Acariformes) Experimental andApplied Acarology 22 559ndash594

Palmgren P 1978 On the muscular anatomy of spiders ActaZoologica Fennica 155 1ndash41

Patten W 1912 The evolution of the vertebrates and their kinPhiladelphia P Blakistonrsquos Son

Patten W Hazen AP 1900 The development of the coxalgland branchial cartilages and genital ducts of Limuluspolyphemus Journal of Morphology 16 459ndash490

Paulus HF 1979 Eye structure and the monophyly of theArthropoda In Gupta AP ed Arthropod phylogeny NewYork Van Nostrand Reinhold 299ndash383

Petrunkevitch A 1949 A study of Palaeozoic ArachnidaTransactions of the Connecticut Academy of Arts and Sci-ences 37 69ndash315

Pittard K Mitchell RW 1972 Comparative morphology ofthe life stages of Cryptocellus pelaezi (Arachnida Ricinulei)Graduate Studies Texas Tech University 1 1ndash77

Platnick NI 2005 The world spider catalog Version55 American Museum of Natural History httpresearchamnhorgentomologyspiderscatalogindexhtm

Platnick NI Gertsch WJ 1976 The suborders of spiders acladistic analysis (Arachnida Araneae) American MuseumNovitates 2607 1ndash15

Pocock RI 1893 On some points in the morphology of theArachnida (ss) with notes on the classification of the groupAnnals and Magazine of Natural History Series 6 11 1ndash19

Pocock RI 1902 Some points in the morphology and classi-fication of the Opiliones Annals and Magazine of NaturalHistory Series 7 10 504ndash516

Popadicccc A Panganiban G Rusch D Shear WA KaufmanTC 1998 Molecular evidence for the gnathobasic derivationof arthropod mandibles and for the appendicular origin ofthe labrum and other structures Development Genes andEvolution 208 142ndash150

PHYLOGENY OF THE ARACHNID ORDERS 243

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Raw F 1957 Origin of chelicerates Journal of Paleontology31 139ndash192

Reddell JR Cokendolpher JC 1995 Catalogue bibliogra-phy and generic revision of the order Schizomida (Arach-nida) Texas Memorial Museum Speleological Monographs4 1ndash170

Reissland A Goumlrner P 1985 Trichobothria In Barth FGed Neurobiology of arachnids Berlin Springer-Verlag 138ndash161

Roewer C-F 1934 Solifugae Palpigradi Bronns Klassen undOrdnungen ses Tierreich 5 IV 4 1ndash713

Rowland JM Cooke JAL 1973 Systematics of the arachnidorder Uropygida (=Thelyphonida) Journal of Arachnology 155ndash71

Rowland JM Sissom WD 1980 Report on a fossil palpi-grade from the Tertiary of Arizona and a review of themorphology and systematics of the order (ArachnidaPalpigradida) Journal of Arachnology 8 69ndash86

Rucker A 1901 The Texan Koenenia American Naturalist35 615ndash630

Ruhland M Rathmayer W 1978 Die Beinmuskulatur undihre Innervation bei der Vogelspinne Dugesiella henzti (Ch)(Araneae Aviculariidae) Zoomorphologie 89 33ndash46

Saboori A Kamali K 2000 Description of Allothrombiumtriticium adult (Acari Trombidiidae) from Iran Systematicand Applied Acarology 5 207ndash208

Schliwa M 1979 The retina of the phalangid Opilio raven-nae with particular reference to arhabdomeric cells Celland Tissue Research 204 473ndash495

Scholl G 1977 Beitrage zur Embryonalentwicklung von Lim-ulus polyphemus L (Chelicerata Xiphosura) Zoomorpholo-gie 86 99ndash154

Scholtz G Kamenz C 2006 The book lungs of Scorpionesand Tetrapulmonata (Chelicerata Arachnida) evidence forhomology and a single terrestrialization event of a commonarachnid ancestor Zoology 109 2ndash13

Seitz KA 1972 Zur Histologie und Feinstruktur des Herzensund der Haumlmocyten von Cupiennius salei Keys (AraneaeCtenidae) I Herzwandung Bildung und Differenzierungder Haumlmocyten Zoologische Jahrbuumlcher Abteilung fuumlr Anat-omie und Ontogenie der Tiere 89 351ndash384

Sekiguchi K ed 1988 Biology of horseshoe crabs Tokyo Sci-ence House

Selden PA 1981 Functional morphology of the prosoma ofBaltoeurypterus tetragonophthalmus (Fischer) (ChelicerataEurypterida) Transactions of the Royal Society of Edin-burgh Earth Sciences 72 9ndash48

Selden P 1992 Revision of the fossil ricinuleids Transactionsof the Royal Society of Edinburgh Earth Sciences 83 595ndash634

Selden PA 1993 Fossil arachnids ndash recent advances andfuture prospects Memoirs of the Queensland Museum 33389ndash400

Selden PA Dunlop JA 1998 Fossil taxa and relationshipsof chelicerates In Edgecombe GD ed Arthropod fossilsand phylogeny New York Columbia University Press 303ndash331

Selden PA Jeram AJ 1989 Palaeophysiology of terrestrial-

isation in the Chelicerata Transactions of the Royal Societyof Edinburgh Earth Sciences 80 303ndash310

Selden PA Shear WA Bonamo PM 1991 A spider and otherarachnids from the Devonian of New York and reinterpre-tations of Devonian Araneae Palaeontology 34 241ndash281

Sensenig A Shultz JW 2003 Mechanics of cuticular elasticenergy storage in leg joints lacking extensor muscles inarachnids Journal of Experimental Biology 206 771ndash784

Shear WA 1993 New species in the opilionid genus Stylocel-lus from Malaysia Indonesia and the Philippines (OpilionesCyphophthalmi Stylocellidae) Bulletin of the British Arach-nological Society 9 174ndash188

Shear WA Selden PA Rolfe WDI Bonamo PM GriersonJD 1987 New terrestrial arachnids from the Devonian ofGilboa New York (Arachnida Trigonotarbida) AmericanMuseum Novitates 2901 1ndash74

Shultz JW 1987 The origin of the spinning apparatus in spi-ders Biological Reviews of the Cambridge PhilosophicalSociety 62 89ndash113

Shultz JW 1989 Morphology of locomotor appendages inArachnida evolutionary trends and phylogenetic implica-tions Zoological Journal of the Linnean Society 97 1ndash56

Shultz JW 1990 Evolutionary morphology and phylogeny ofArachnida Cladistics 6 1ndash31

Shultz JW 1991 Evolution of locomotion in Arachnida thehydraulic pressure pump of the giant whipscorpion Masti-goproctus giganteus (Uropygi) Journal of Morphology 21013ndash31

Shultz JW 1993 Muscular anatomy of the giant whipscor-pion Mastigoproctus giganteus (Arachnida Uropygi) and itsevolutionary significance Zoological Journal of the LinneanSociety 108 335ndash365

Shultz JW 1998 Phylogeny of Opiliones (Arachnida) anassessment of the lsquoCyphopalpatoresrsquo concept Journal ofArachnology 26 257ndash272

Shultz JW 1999 Muscular anatomy of a whipspider Phrynuslongipes (Amblypygi) and its evolutionary significance Zoo-logical Journal of the Linnean Society 126 81ndash116

Shultz JW 2000 Skeletomuscular anatomy of the harvest-man Leiobunum aldrichi (Weed 1893) (Arachnida Opil-iones Palpatores) and its evolutionary significanceZoological Journal of the Linnean Society 128 401ndash438

Shultz JW 2001 Gross muscular anatomy of Limuluspolyphemus (Chelicerata Xiphosura) and its bearing on evo-lution in the Arachnida Journal of Arachnology 29 283ndash303

Shultz JW 2007 Morphology of the prosomal exoskeleton ofScorpiones (Arachnida) and a new hypothesis for the evolu-tion of cuticular cephalic endoskeletons in arthropodsArthropod Structure amp Development 36 77ndash102

Shultz JW Pinto da Rocha R 2007 Morphology and func-tional anatomy In Pinto da Rocha R Machado G Giribet Geds The harvestmen the biology of Opiliones CambridgeMA Harvard University Press 14ndash61

Shultz JW Regier JC 2001 Phylogenetic analysis of Pha-langida (Arachnida Opiliones) using two nuclear protein-encoding genes supports monophyly of Palpatores Journalof Arachnology 29 189ndash200

Sitnikova LG 1978 The main evolutionary trends of the

244 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Acari and the problem of their monophyletism Entomologis-cheskoe Obozrenie 57 431ndash457

Snodgrass RE 1948 The feeding organs of Arachnidaincluding mites and ticks Smithsonian Miscellaneous Col-lections 110 (10) 1ndash94

Snodgrass RE 1952 A textbook of arthropod anatomy Ith-aca NY Comstock Publishing

Soslashrensen W 1879 Om Bygningen af Gonyleptiderne en Typeaf Arachnidernes Classe Naturhistorisk Tidsskrifter Series3 8 97ndash222

Stockwell SA 1989 Revision of the phylogeny and higherclassification of scorpions (Chelicerata) Unpublished doc-toral dissertation University of California at Berkeley

Stoslashrmer L 1944 On the relationships and phylogeny of fossiland Recent Arachnomorpha Skrifter utgitt av Det NorskeVidenskaps-Akademi i Oslo I 5 1ndash158

Stoslashrmer L 1955 Merostomata In Moore RC ed Treatise oninvertebrate paleontology Part P Arthropoda 2 LawrenceKS University of Kansas Press and Geological Society ofAmerica 4ndash41

Swofford DL 2002 PAUP Phylogenetic analysis using par-simony (and other methods) Version Beta 10 SunderlandMA Sinauer Associates

Talarico G Palacios-Vargas JG Fuentes Silva M AlbertiG 2005 First ultrastructural observations of the tarsal poreorgan of Pseudocellus pearsei and P boneti (ArachnidaRicinulei) Journal of Arachnology 33 604ndash612

Templeton AR 1983 Phylogenetic inference from restrictionendonuclease cleavage site maps with articular reference tothe evolution of humans and the apes Evolution 37 221ndash244

Tetlie OE Braddy SJ 2004 The first Silurian chasmataspidLoganamaraspis dunlopi gen et sp nov (Chelicerata Chas-mataspidida) from Lesmahagow Scotland and its implica-tions for eurypterid phylogeny Transactions of the RoyalSociety of Edinburgh Earth Sciences 94 227ndash234

Thomas RH Telford MJ 1999 Appendage development inembryos of the oribatid mite Archegozetes longisetosus(Acari Oribatei Trhypochthoniidae) Acta Zoologica (Stock-holm) 80 193ndash200

Tollerton VP Jr 1989 Morphology taxonomy and classifica-tion of the order Eurypterida Burmeister 1843 Journal ofPaleontology 63 642ndash657

Vachon M 1949 Ordre des Pseudoscorpions In Grasseacute P-Ped Traite de zoologie VI Paris Masson et Cie 436ndash481

Van der Hammen L 1979 Comparative studies in Chelicer-ata I The Cryptognomae (Ricinulei Architarbi and Anacti-notrichida) Zoologische Verhandelingen 174 1ndash62

Van der Hammen L 1982 Comparative studies in Chelicer-ata II Epimerata (Palpigradi and Actinotrichida) Zoolo-gische Verhandelingen 196 1ndash70

Van der Hammen L 1985 Functional morphology and affin-ities of extant Chelicerata in evolutionary perspectiveTransactions of the Royal Society of Edinburgh 76 137ndash146

Van der Hammen L 1989 An introduction to comparativearachnology Leiden SPB Academic Publishing

Vyas AB 1970 Studies in myology of the scorpion Heter-ometrus fulvipes Koch Unpublished PhD DissertationDepartment of Zoology Gujarat University India

Vyas AB 1974 The cheliceral muscles of the scorpion Heter-ometrus fulvipes Bulletin Southern California Academy ofSciences 73 9ndash14

Walter DE 1988 Predation and mycophagy by endeostigma-tid mites (Acariformes Prostigmata) Experimental andApplied Acarology 4 159ndash166

Walter DE Proctor HC 1998 Feeding behaviour and phy-logeny observations on early derivative Acari Experimentaland Applied Acarology 22 39ndash50

Walter DE Proctor HC 1999 Mites ecology evolution andbehaviour Sydney University of New South Wales

Waterston CD 1979 Problems of functional morphology andclassification in stylonuroid eurypterids (Chelierata Meros-tomata) with observations on the Scottish Silurian Sty-lonuroidea Transactions of the Royal Society of EdinburghEarth Science 70 251ndash322

Weygoldt P 1969 Biology of pseudoscorpions CambridgeMA Harvard University Press

Weygoldt P 1984 Lrsquoautotomie chez les Amblypyges RevueArachnologique 5 321ndash327

Weygoldt P 1996 Evolutionary morphology of whip spiderstowards a phylogenetic system (Chelicerata ArachnidaAmblypygi) Journal of Zoological Systematics and Olution-ary Research 34 185ndash202

Weygoldt P 1998 Evolution and systematics of the Chelicer-ata Experimental and Applied Acarology 22 63ndash79

Weygoldt P 2000 Whip spiders (Chelicerata Amblypygi)their biology morphology and systematics StenstrupDenmark Apollo Books

Weygoldt P Paulus HF 1979 Untersuchungen zur Morphol-ogie Taxonomie und Phylogenie der Chelicerata Zeitschriftfuumlr Zoologische Systematik und Evolutionsforschung 17 85ndash116 177ndash200

Wheeler WC Hayashi CY 1998 The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192

Winkler D 1957 Die Entwicklung der aumlusseren Koumlrperge-stalt bei den Phalangiidae (Opiliones) Mitteilungen aus demZoologischen Museum in Berlin 33 355ndash389

With CJ 1904 The Notostigmata a new suborder of AcariVidenskabelige Meddelelser fra Dansk Naturhistorisk Foren-ingi Kjobenhaus 1904 137ndash192

Yamasaki T Makioka T Saito J 1988 Morphology InSekiguchi K ed Biology of horseshoe crabs Tokyo ScienceHouse 69ndash132

Yoshikura M 1975 Comparative embryology and phylogeny ofArachnida Kumamoto Journal of Science Biology 12 71ndash142

APPENDIX

Each character used in this analysis is numbereddefined and cross-referenced with characters used infour previous studies that are abbreviated in bracketsspecifically WP = Weygoldt amp Paulus (1979)S = Shultz (1990) WH = Wheeler amp Hayashi (1998)GEWB = Giribet et al (2002) Descriptions includecorrections of errors discussions of controversies orambiguities and justifications for state assignmentsto lsquoproblemrsquo taxa

PHYLOGENY OF THE ARACHNID ORDERS 245

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

PROSOMA

1 Dorsal sclerite formed by fusion of the prosomal car-apace the dorsal portion of the first opisthosomalsomite and the dorsomedial (axial) portion of the sec-ond opisthosomal somite 0 absent 1 presentState 1 is present in extant xiphosurids and close fos-sil relatives (Scholl 1977 Anderson amp Selden 1997Shultz 2001) The fossil record suggests that thisstructure originated with the disappearance ofthe already reduced first opisthosomal tergite(= microtergite) and axial portion of the secondopisthosomal tergite (Anderson amp Selden 1997)2 Single dorsal sclerite covering entire dorsal surfaceof body no lines indicating original segmentation 0absent 1 presentState 1 occurs throughout Holothyrida (Acari) (Vander Hammen 1989) and in many Mesostigmata(Alberti amp Coons 1999) including Glyptholaspis (Vander Hammen 1989)3 Anterior end of dorsal prosoma with medianmarginal or submarginal pointed process 0 absent1 presentState 1 occurs throughout Schizomida (Protoschizo-mus Stenochrus Reddell amp Cokendolpher 1995) The-lyphonida (Rowland amp Cooke 1973 Shultz unpublobs Proschizomus Dunlop amp Horrocks 19951996)Amblypygi (Charinus Millot 1949b Phrynus Shultz1999) and Plesiosiro (Dunlop 1999) Citing a figure inRowland amp Cooke (1973) Dunlop amp Horrocks assumedthat the anterior process was limited to lsquohypoctonidrsquothelyphonids and united them with Proschizomus andSchizomida If valid this would make Thelyphonidaparaphyletic However the doubtful monophyly oflsquohypoctonidsrsquo (Haupt amp Song 1996) and widespreadpresence of a submarginal anterior process in Pedi-palpi weaken this proposal (Harvey 2002)4 Ophthalmic ridges pair of longitudinal crests inter-secting or passing near the region of the lateral eyes orcomparable region where lateral eyes are absent 0absent 1 presentState 1 is present throughout Xiphosura (Anderson ampSelden 1997) Similar structures are present in Plesi-osiro (Dunlop 1999) and lsquonon-hypoctonidrsquo Thely-phonida (Mastigoproctus Rowland amp Cooke 1973)Dunlop (2002a) described apparent ophthalmic ridgesin the chasmataspidid Octoberaspis but these werenot observed in Diploaspis (Dunlop et al 2001) eitherdue to absence or inadequate preservation of speci-mens nor in Chasmataspis (Dunlop et al 2004)5 Carapace with demarcations (eg grooves scleritesphragmata) between pro- meso- or metapeltidia 0absent 1 present [S 1 WH 48 GEWB 29]State 1 is present in extant Scorpiones Cyphoph-thalmi (Hansen amp Soslashrensen 1904) and many Palpa-tores (Opiliones) including Caddo and Leiobunum

and most Pseudoscorpiones (Chamberlin 1931) Giri-bet et al (2002) coded this character as absent inCyphophthalmi (Opiliones) but a transverseprocurved groove is present throughout the group(Hansen amp Soslashrensen 1904) and represents the meso-metapeltidial border as indicated by attachment ofthe dorsal endosternal suspensor of postoral somite VIin Chileogovea oedipus and Siro exilis (Shultz unpublobserv) Kjellesvig-Waering (1986) did not explicitlydescribe such grooves in fossil scorpions although heillustrated corresponding structures in PrearcturusArchaeophonus Archaeoctonus Palaeophonus andother Palaeozoic scorpions (Kjellesvig-Waering 1986figs 14A 21E 27D 59) (coded as uncertain here) butnot in Proscorpius (contra Giribet et al 2002) Dis-tinct carapacal sclerites are present throughout Palpi-gradi Schizomida and Solifugae (6)

The condition in Acari is controversial but is codedhere as State 1 Van der Hammen (1989) consideredthe tergal region of the leg-bearing somites (postoralsomites IIIndashVI) or podosoma to be reduced andreplaced by posterior migration of the aspidosoma (ietergal region assumed to be associated with append-ages of the gnathosoma) and anterior migration of theopisthosomal tergal region to form the hysterosomaThe dorsal proterosomal and hysterosomal elementsmeet at a transverse sejugal furrow that continues lat-erally and passes ventrally between the coxae of legs 2and 3 According to this scheme the prosomandashopisthosoma border is expressed as the disjugal fur-row which passes from the ventral posterior margin ofthe podosoma anterodorsally to join the dorsal part ofthe sejugal furrow Based on the arrangement of setaeand slit sensilla (= lyrifissures) Van der Hammen(1989) interpreted the region above coxae 3 and 4 (ieregion C) as a fusion of the dorsal parts of the first twoopisthosomal somites (= postoral somites VII and VIII)in early divergent Anactinotrichida (ie Opilioacari-formes) and Acariformes (eg Alycus) These interpre-tations have been followed by many acarologistsalthough its speculative aspects are widely acknowl-edged (Evans 1992 Alberti amp Coons 1999)

From the standpoint of a general arachnologist theVan der Hammen system seems unnecessarily compli-cated it appears to have been formulated to explainbroad morphological themes in Oribatida but was thenextrapolated to other mites A more conservativelsquoarachnologicalrsquo scheme adopted here equates the pro-podosoma with the pars cephalica (= propeltidium) ofthe arachnid prosoma and region C with the pars tho-racica (= mesopeltidium + metapetidium) In fact theearliest and most recent treatments of Opilioacari-formes (With 1904 Klompen 2000) reached similarconclusions This interpretation is also consistent withevidence from Acariformes For example the regiondorsal to the coxa of legs 3 and 4 in Alycus is as readily

246 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

explained by the persistence of primitive prosomal ter-gal elements as by the supposition that these disap-peared and were replaced by two opisthosomalelements Furthermore several diverse lineages ofbasally divergent Acariformes display features that donot fit readily within the GrandjeanndashVan der Ham-men system but seem to correspond to the cephalicathoracica division of the prosoma They have a sejugalfurrow that may have demarcated the cephalicatho-racica border and a postpodosomal furrow thatappears to represent the prosomandashopisthosoma junc-tion there is no disjugal furrow Examples includepalaeosomate (especially aphelacarid) and pediculoch-elid Oribatida alicorhagiid and mircopsammidlsquoEndeostigmatarsquo and paratydeid Prostigmata (Albertiamp Coons 1999 fig 42) This organization may be sym-plesiomorphic for Acariformes but could be a conver-gence brought about by selection for enhancedflexibility6 Carapace with distinct pro- meso- or metapeltidialsclerites 0 absent 1 present - inapplicable due toabsence of pro- meso- or metapeltidial demarcations(5)State 1 occurs throughout Palpigradi Schizomida andSolifugae It also appears sporadically within Opil-iones (eg Leiobunum flavum Shultz unpubl observ)and Pseudoscorpiones (eg Pseudochiridium) (Cham-berlin 1931)7 Sejugal furrow circumferential zone of body flexi-bility that passes between the coxae of legs 2 and 3 0absent 1 presentState 1 is probably synapomorphic for Acariformes asit occurs in most lsquoendeostigmatidsrsquo including Alycus(Van der Hammen 1989) and is variously developedthroughout Sarcoptiformes (Alberti amp Coons 1999) Itis probably the primitive condition for Prostigmatabut is either absent or weakly expressed in represen-tatives included here See 5 for alternative morpholog-ical interpretations of this body region in Acari8 Prosomal ozopores 0 absent 1 present [WH 46GEWB 12]State 1 occurs throughout Opiliones and Holothyrida(Van der Hammen 1989)9 Carapacal pleural doublure 0 absent 1 present[WP 12 S 2 WH 49 GEWB 27]A carapacal pleural doublure occurs throughoutXiphosura (Stoslashrmer 1944) Eurypterida (Clarke ampRuedemann 1912 Selden 1981) and Chasmataspid-ida (Dunlop et al 2001 2004 Dunlop 2002a) and isabsent in all known Arachnida10 Cardiac lobe a longitudinal axial elevation of thecarapace 0 absent 1 presentState 1 occurs throughout Xiphosura (Anderson ampSelden 1997) and Eurypterida (Clarke amp Ruedemann1912 Selden 1981) Carapaces are not preserved suf-ficiently to determine this condition in the chasmatas-

pidids Diploaspis (Dunlop et al 2001) andOctoberaspis (Dunlop 2002a) but an apparent cardiaclobe occurs in the Chasmataspis (Dunlop et al 2004)A cardiac lobe is absent in Arachnida11 Moveable cucullus 0 absent 1 present [WP 54GEWB 22]State 1 is a synapomorphy of Ricinulei (Selden 1992)12 Medial intercoxal lsquosternalrsquo region 0 all pedalcoxae separated medially 1 anterior pedal coxae abut-ting medially posterior coxae separated 2 anteriorpedal coxae separated medially posterior coxae abut-ting 3 all pedal coxae abutting medially 4 epimeracoxae undifferentiated medially from ventral body wall[S sim3 WH sim50 GEWB sim17]Prosomal sternites appear to be those portions of theventral body wall not occupied by coxae Despite earlyattempts to assess metamerism of the ventralprosoma (protosternum deutosternum tritosternumetc) (Boumlrner 1902a Millot 1949a) no apparent mor-phological features such as borders of sclerites ormuscle attachments reliably demarcate the ventralbody wall of one somite from that of an adjacentsomite Thus coding schemes that focus on describingmetameric components of the sternal region are prob-lematic Shultz (1990) attempted to code the shape ofthe entire intercoxal region and subsequent workershave adopted this approach All pedal coxae are sepa-rated medially in Plesiosiro (Dunlop 1999) Palpigradi(Roewer 1934) Araneae (Millot 1949c) Amblypygi(Weygoldt 2000) and many Acari including Opilioac-ariformes Holothryrida Mesostigmata Ixodida Aly-cus and many Prostigmata (Van der Hammen 1989Evans 1992 Alberti amp Coons 1999) The anteriorcoxae are separated in extant Thelyphonida Schizo-mida (Hansen amp Soslashrensen 1905 Millot 1949e) andRicinulei (Pittard amp Mitchell 1972 Legg 1976) Thecoxae are fused to the ventral body wall in Sarcopti-formes (Van der Hammen 1989 Evans 1992)

Interpreting this character is complicated for theanterior coxae of Scorpiones and Opiliones and allcoxae in Xiphosura and Eurypterida due to the pres-ence of coxapophyses endites andor gnathobaseswhich form components of the preoral chambers inthese groups Thus coxae may abut medially whenthese structures are considered part of the coxae butare separate medially if these structures are ignoredAs coxapophyses etc are here coded as separate char-acters (50ndash53) I have chosen to code this character asif these structures did not exist Consequently pha-langid Opiliones are coded as having all coxae sepa-rated (Pocock 1902 Hansen amp Soslashrensen 1904) Fossilscorpions also show substantial variation in this char-acter All pairs of pedal coxae in Palaeoscorpius abutmedially but also have gnathobases (Jeram 1998)and this condition is coded here as uncertain A ster-num separates all pairs of pedal coxae medially in

PHYLOGENY OF THE ARACHNID ORDERS

247

copy 2007 The Linnean Society of London

Zoological Journal of the Linnean Society

2007

150

221ndash265

Proscorpius

and

Stoermeroscorpio

(State 0) and

Pre-arcturus

approximates modern scorpions (State 1)(Kjellesvig-Waering 1986 Jeram 1998) The coxalbases of Eurypterida and Xiphosura are separated bya small sternite the endostoma (Selden 1981 Shultz2001)

13 Postoral sternapophysis (

=

tritosternum labium)a cuticular evagination of the ventral body wall poste-riorly adjacent to the palpal coxae forming the poste-rior border of the preoral chamber in some taxa ordisplaced posteriorly by fusion of the palpal coxae inothers 0 absent 1 present

[S

sim

10 WH

sim

56 GEWB19

=

33]State 1 is present throughout Palpigradi AraneaeAmblypygi (Snodgrass 1948) Schizomida (Hansen ampSoslashrensen 1905 Van der Hammen 1989) Thely-phonida (Van der Hammen 1989) extant Ricinulei(Pittard amp Mitchell 1972 Legg 1976) Opilioacari-formes and Mesostigmata (Van der Hammen 1989Alberti amp Coons 1999)

Allothyrus constrictus

is theonly holothyrid mite known to have sternapophysesAn apparent sternapophysis has been described in apalaeocharinid trigonotarbid (Dunlop 1994) Giribet

et al

(2002) miscoded Ixodida and Acariformes as hav-ing a lsquolabium (

=

tritosternum)rsquo The lophognath ofpseudoscorpions (Chamberlin 1931) and lsquolabiumrsquo ofSolifugae (Roewer 1934) also appear to be sternapo-physes that have been incorporated into the ros-trosoma (

32

) The labium of phalangid Opiliones(Pocock 1902 Hansen amp Soslashrensen 1904) is a ster-napophysis but is associated with leg 1 rather thepedipalp (Winkler 1957) A small anterior sclerite inthe fossil scorpions

Proscorpius

Waeringoscorpio

and

Labriscorpio

has been interpreted as a lsquolabiumrsquo homol-ogous with the sternapophyses of other arachnids(Weygoldt 1998) despite absence of any indication ofthe relationship of this sclerite to the mouth and pos-sible alternative interpretations (eg

16

) Giribet

et al

coded this sclerite as a sternapophysis but the char-acter is coded here as uncertain in fossil scorpions

14 Channels on the body surface linking openings ofcoxal organs to preoral chamber 0 absent 1 present

Van der Hammen (1989) described bilaterally pairedcuticular tracts (lsquotaenidiarsquo) connecting orifices of coxalorgans to the preoral chamber in some arachnids andor a ventromedian groove (

=

intercoxal or subcapitulargutter) that leads in turn to the preoral chamber inother taxa Taenidia and intercoxal gutters arepresent in Opilioacariformes (Van der Hammen1989) Holothyrida (Van der Hammen 1989) Mesos-tigmata (

Glyptholapis

Van der Hammen 1989)Ricinulei (Pittard amp Mitchell 1972 Van der Hammen1989) Araneae (

Heptathela

Aphonopelma

Hypochi-lus

Shultz unpubl observ also

Porrhothele

(Mygalo-morphae Dipluridae) Butt amp Taylor 1991

Segestria

(Araneomorphae Segestriidae) Van der Hammen

1989) Amblypygi (Van der Hammen 1989) Schizo-mida (

Stenochrus

Shultz unpubl observ unidenti-fied Hubbardiinae Van der Hammen 1989) andThelyphonida (

Mastigoproctus

Shultz unpublobserv also

Tetrabalius

Van der Hammen 1989) Theconduction of large volumes of fluid from orifices of thecoxal organs to the preoral chamber has been observedin the mygalomorph spider

Porrhothele

(Dipluridae)(Butt amp Taylor 1991) and

Mastigoproctus

(Thely-phonida) (Shultz unpubl observ) The podocephaliccanal (

15

) is present in Acariformes (Acari) (Lindquist1984) and appears to have evolved as an invaginatedsupracoxal channel (Grandjean 1938 G Alberti inEvans 1992) Van der Hammen (1982) could not dis-cern with certainty whether a channel exists in

Eukoenenia

(Palpigradi) but none has been reportedby previous authors and they have not been observedby the present author Palpigradi is coded as havingState 0

15 Podocephalic canal cuticular channel andorduct draining multiple glands and opening nearmouthparts 0 absent 1 present - inapplicable due toabsence of channel (14)

State 1 occurs throughout Acariformes (Acari)(Lindquist 1984)

16 Heavily sclerotized suboral sclerite serving inpart as basal pivot point for coxae of appendages ofpostoral somites IIndashIV (

=

arachnid palp and legs 1 and2) 0 absent 1 present

The sclerite forms the posterior wall of the true mouthand is heavily sclerotized in extant scorpions It is notvisible externally due to the tightly fitting coxae of theanterior prosomal appendages (Shultz 2007) It maycorrespond to the so-called labium (Weygoldt 1998) orlabrum (Kjellesvig-Waering 1986) of certain fossilscorpions (see

13

)

17 Genal angles 0 rounded 1 pointed

Anderson amp Selden (1997) originally coded a pointedgenal angle and genal spine as separate charactersbut they are combined here State 1 is present in manyXiphosura (Anderson amp Selden 1997) and Chas-mataspidida (Dunlop

et al

2001 2004 Dunlop2002a)

A

PPENDAGES

OF

POSTORAL

SOMITE

I

CHELICERAE

18 Cheliceral segmentation 0 three articles 1 twoarticles

[WP

sim

25

+

42 WH

sim

15 GEWB

sim

44]State 0 is present in extant Xiphosura Eurypterida(Clarke amp Ruedemann 1912 Selden 1981) Palpi-gradi (Roewer 1934) Opiliones (Hansen amp Soslashrensen1904) Scorpiones (Hjelle 1990) and throughout non-acariform Acari (Evans 1992) State 1 is presentthroughout known Trigonotarbida Araneae Ambly-pygi Thelyphonida Schizomida Pseudoscorpionesand Solifugae Acariform mites have two unambigu-

248 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ous cheliceral articles but may have a reduced thirdarticle (Van der Hammen 1989 Evans 1992) acari-form mites are thus coded as uncertain Chelicerae arenot known or their condition is unclear in several fos-sil taxa including Xiphosura Chasmataspidida (Dun-lop et al 2001 2004 Dunlop 2002a) Riciniulei(Selden 1992) Plesiosiro (Dunlop 1999) Proschizo-mus (Dunlop amp Horrocks 19951996) and the fossilscorpions Praearcturus and Palaeoscorpius (Kjelles-vig-Waering 1986) Kjellesvig-Waering (1986) consis-tently interpreted chelicerae of fossil scorpions ashaving four articles but this is probably a misinter-pretation of the basal article (see also Stockwell1989) The basal article in extant scorpions consists ofa distal collar of cuticle and a large proximal processand this arrangement would probably appear as twoarticles in compressed specimens In any event thereare at least three cheliceral articles in Stoermeroscor-pio and Proscorpius (Kjellesvig-Waering 1986)19 Terminal cheliceral joint 0 laterally placedbicondylar hinge 1 dorsally placed bicondylar hinge2 ventrally placed bicondylar hinge [WP 25 + 42 Ssim11 + 12 WH sim15 GEWB sim44]State 0 is present in Xiphosura (Shultz 2001)extant Scorpiones (Millot amp Vachon 1949) Opil-iones (Hansen amp Soslashrensen 1904) Palpigradi(Boumlrner 1904 Millot 1949d) Ricinulei (Pittard ampMitchell 1972 Legg 1976) and Eurypterida includ-ing Baltoeurypterus (Selden 1981) and Stylonurus(Clarke amp Ruedemann 1912) State 0 is present inthe fossil scorpions Proscorpius and Stoermeroscor-pio but the state of the chelicerae is unclear in theother fossil scorpions represented here (Kjellesvig-Waering 1986) State 1 is present in Trigonotarbida(Shear et al 1987) mygalomorph Araneae Ambly-pygi Schizomida and Theylphonida (Weygoldt ampPaulus 1979) State 2 is present in Pseudoscorpi-ones (Chamberlin 1931) and Solifugae (Roewer1934) and appears to be a groundplan feature for allmajor groups of mites (eg Van der Hammen 1989Alberti amp Coons 1999) The difference between theorthognathy of mygalomorph Araneae and lsquoplagiog-nathyrsquo mesothele (Liphistius Heptathela) and palae-ocribellate Araneae (Hypochilus) (Kraus amp Kraus1993) is small compared with the variation thatoccurs throughout arachnids and plagiognathy iscoded as State 1 here20 Chelicera articulating with carapace at anterolat-eral pivot 0 absent 1 present [S 13 WH 57 GEWB45]State 1 occurs in Solifugae (Roewer 1934) and Pseu-doscorpiones except Chthonioidea and Feaelloidea(Chamberlin 1931) Giribet et al (2002) coded thecharacter as present in Pseudoscorpiones21 Chelicera pivoting on dorsal protuberance of epis-tome 0 absent 1 present

State 1 appears to be a unique synapomorphy of Opil-iones (Shultz 2000)22 Extrinsic cheliceral muscle arising on carapaceand inserting on dorsal margin of nonbasal cheliceralarticle 0 absent 1 present (Note Characters 22 and23 refer to two different muscles)State 1 is known only in extant Scorpiones (LankesterBenham amp Beck 1885 Vyas 1970 1974 Shultz2007) but may occur in acariform mites if their cheli-cerae are composed of three rather than two articles(Evans 1992) (see 18) Acariformes are coded asuncertain23 Extrinsic cheliceral muscle arising on carapaceand inserting on ventral margin of nonbasal cheliceralarticle 0 absent 1 present [GEWB 252] (Note Char-acters 22 and 23 refer to two different muscles)State 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) somemesostigmatid Acari (Evans 1992) and representa-tive Opiliones particularly Leiobunum (Shultz 2000)Chileogovea and Gonyleptes (Shultz unpubl observ)The status of the character in other opilions is not yetknown The character was originally defined as anextrinsic muscle inserting on the second segment ofthree-segmented chelicerae (Shultz 2000 Giribetet al 2002) but it has been redefined here to make itapplicable to arachnids with two-segmented cheli-cerae24 Extrinsic cheliceral muscles attaching to epistome0 absent 1 presentState 1 is known only in extant Scorpiones (Lankesteret al 1885 Vyas 1970 1974 Shultz 2007) but mayoccur in Prostigmata if the sigmoid piece to which thecheliceral protractors attach (Evans 1992) is a modi-fication of the epistome25 Lateral tergocheliceral muscle with three heads 0absent 1 present - coded only for extant tetrapulmo-nates and the palpigrade Eukoenenia homology isunclear in other taxa [GEWB 230]State 1 is present in Pedipalpi (Shultz 1999)26 Cheliceral silk glands 0 absent 1 present [GEWB47]State 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931) The silk gland of spider mites(eg Tetranychus) are associated with the pedipalps(Evans 1992)27 Cheliceral venom glands 0 absent 1 present [WP34 WH 40 GEWB 46]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)28 Cheliceral serrula interior and exterior 0 absent1 presentState 1 is a unique synapomorphy of Pseudoscorpiones(Chamberlin 1931 Harvey 1992)29 Cheliceral lsquoflagellumrsquo in male 0 absent 1 present[GEWB 48]

PHYLOGENY OF THE ARACHNID ORDERS 249

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITE II LEG 1 IN NONARACHNIDS AND PALP IN ARACHNIDS

30 Palpal coxae fused ventromedially and formingposterior wall of preoral chamber 0 absent 1 present[WP 27 S 18 WH 17 = 62 GEWB 33 + 63]State 1 occurs in Schizomida Thelyphonida Acari andRicinulei31 Gnathosoma 0 absent 1 present - inapplicablepalpal coxae not fused (30) [WP 56 S 19 WH sim63GEWB sim36 + 37]The gnathosoma is a functional complex comprisingan epistome attached to the dorsal surface of mediallyfused palpal coxae (= infra- or subcapitulum) (see 30)the chelicerae and in some taxa a supracheliceral tec-tum These components are tightly integrated andmove in unison relative to the body The gnathosomahas traditionally been considered a feature of AcariHowever Van der Hammen (1989) argued that a gna-thosoma is present in Ricinulei and this interpreta-tion has received wide acceptance (eg Lindquist1984) However it is not clear that the relationship ofthe chelicerae and subcapitulum is as intimate inRicinulei as in Acari and the ricinuleid subcapitulumdoes not appear to be as moveable In fact the lsquosubca-pitulumrsquo in Ricinulei is similar in many respects to thecondition in Schizomida and Thelyphonida (Hansen ampSoslashrensen 1904) which has never been regarded as agnathosoma Consequently I have coded the conditionof Ricinulei as uncertain32 Rostrosoma long narrow subcylindrical epistomeprojecting anteriorly with base fixed to dorsal surfaceof palpal coxae bordered laterally by lobes projectingfrom palpal coxae ventral wall of preoral chamberformed by anterior element of prosoma (sternapophy-sis) 0 absent 1 present [S 15 WH 59 GEWB 20 + 36]State 1 is present in Pseudoscorpiones and Solifugae(Chamberlin 1931 Roewer 1934 Van der Hammen1989) Dunlop (2000) attempted to homologize com-ponents of the rostrosoma and gnathosoma (31) anargument accepted by Giribet et al (2002) Howeverfusion of the palpal coxae (30) apparently evolvedindependently in the two structures Coxal fusion inthe gnathosoma is complete ventrally and excludesthe suboral sternapophysis (13) while fusion isincomplete ventrally in the rostrosoma and incorpo-rates the sternapophysis Unlike the gnathosomathe rostrosoma is largely immobile and cheliceralmovement is not coupled with that of the palpalcoxae33 Rutellacorniculi hypertrophied setae modified asmouthparts located on the anterior processes of the pal-pal coxae 0 absent 1 present [GEWB 39]

State 1 occurs in the anactinotrichid groups Opilioac-ariformes Holothyrida and Mesostigmata but not Ixo-dida (Van der Hammen 1989) It also occurs inSarcoptiformes (Evans 1992) and certain endeostig-matids (eg Alycus Van der Hammen 1989) but not inProstigmata Comparable structures are unknown inother chelicerate groups34 Terminal segments specially modified in adultmale as a clasper used to engage the female 0 absent1 presentState 1 is known only in extant Xiphosura (YamasakiMakioka amp Saito 1988)35 Robust raptorial 0 absent 1 present [S 17 WH61 GEWB sim59 97]State 1 is present in Amblypygi Thelyphonida andSchizomida (Shear et al 1987) and in Laniatores(Opiliones)36 Orientation of robust raptorial appendage 0operating in subtransverse plane 1 operating in sub-vertical plane - inapplicable coded only for PedipalpiState 0 is present in extant Thelyphonida and mostAmblypygi and State 1 is present in Schizomida (Cok-endolpher amp Reddell 1992) and in the basally diver-gent amblypygid Paracharon caecus (Weygoldt 2000)The orientation of the pedipalps in the fossil Proschi-zomus is uncertain (Dunlop amp Horrocks 19951996)37 Extrinsic muscle attaching to epistome 0 absent1 presentState 1 occurs in Chileogovea and Siro (OpilionesCyphophthalmi) and extant Scorpiones (Vyas 1970Shultz 2007)38 Muscle originating and inserting within coxa 0absent 1 present [GEWB sim232]State 1 occurs is in Amblypygi (Phrynus Shultz 1999Phrynichus Boumlrner 1904) Thelyphonida (Mastigo-proctus Shultz 1993 Thelyphonus Boumlrner 1904) andSchizomida (Stenochrus Shultz unpubl observ)There are two intracoxal muscles in amblypygids andthelyphonids the one coded here and another that isserially homologous with a muscle in the pedal coxae(54)39 Tarsus andor tibia with venom glands 0 absent1 present [GEWB 64]State 1 is known only from extant Pseudoscorpi-ones except the superfamilies Chthonioidea andFeaelloidea (Chamberlin 1931 Harvey 1992) Giri-bet et al (2002) coded this character as present forPseudoscorpiones40 Tarsal grooming organ 0 absent 1 present[GEWB 98]State 1 is a unique synapomorphy of Amblypygi (DelleCave 1975 Weygoldt 2000)41 Scorpionoid chela a large well-developed chelaformed by tibia (manus + fixed finger) and tarsus(moveable finger) 0 absent 1 present [S 16 WH 60GEWB 62]

250 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs throughout Scorpiones andPseudoscorpiones42 Modified in male as copulatory organ 0 absent 1present [WP 35 WH 39 GEWB 67]State 1 appears to be a unique synapomorphy of Ara-neae (Platnick amp Gertsch 1976)43 Apotele 0 apparently absent not differentiatedexternally from penultimate article 1 present [GEWB69]The apotele exists as a structure (ie dactyl clawempodium) differentiated from the tibiotarsus tarsusor telotarsus and is typically associated with a pair ofantagonistic muscles The apotele occurs as a distinctstructure in Xiphosura (Snodgrass 1948) Eurypter-ida (Clarke amp Ruedemann 1912) Schizomida (Coken-dolpher amp Reddell 1992) Araneae (Foelix 1996)early divergent Amblypygi (Charinus Weygoldt 2000not Phrynus Shultz 1999) Anactinotrichida (CaminClark amp Bourdeau 1956 Lindquist 1984) Ricinulei(Pittard amp Mitchell 1972 Legg 1976) and Solifugae(Roewer 1934) Giribet et al (2002) coded Solifugae asnot having a palpal apotele but it is present as anadhesive organ (45) Dunlop (1999) reconstructedHaptopoda as having a palpal claw but notes in hisdescription that the distal ends were not preserved inthe known specimens44 Apotele (claw) position 0 terminal 1 subtermi-nal - inapplicable coded only for taxa with a distinctapotele (43)State 1 occurs in Parasitiformes (Acari) (Evans 1992)45 Terminal adhesive organ 0 absent 1 present -inapplicable due to lack of an apotele (43) [GEWB 66]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)

APPENDAGES OF POSTORAL SOMITES IIIndashVI LEGS 2ndash5 IN NON-ARACHNIDS AND LEGS 1ndash4 IN ARACHNIDS

46 Appendage III (= arachnid leg 1) extremely elon-gate antenniform 0 absent 1 present [WP sim24 + 32S 20 WH 14 GEWB sim84]State 1 occurs in Amblypygi Thelyphonida Schizo-mida (Shear et al 1987) and Opilioacariformes (Vander Hammen 1989) State 1 is approximated in Solif-ugae Palpigradi (Roewer 1934) and Haptopoda(Dunlop 1999) and it is coded as uncertain in thesetaxa47 Appendage V (= arachnid leg 3) of male specializedfor sperm transfer 0 absent 1 present [WP 55 GEWB94]State 1 is known only from extant Ricinulei48 Appendages V and VI (= arachnid legs 3 and 4)with femur shorter than patella and with principal siteof flexionextension at patella-tibia joint (lsquoapatellatersquocondition sensu Van der Hammen 1989) 0 absent 1present [S 25 GEWB 86]

State 1 occurs in Pseudoscorpiones and Solifugae(Shultz 1989)49 Appendage VI (= arachnid leg 4) with terminusmodified as flattened blade 0 absent 1 presentState 1 occurs in some Chasmataspidida (DiploapisDunlop et al 2001 Octoberaspis Dunlop 2002a)uncertain in Chasmataspis Dunlop et al 2004 State0 in Loganamaraspis Tetlie amp Braddy 2004) andEurypterida except stylonuroids (Clarke amp Ruede-mann 1912 Waterston 1979 Tollerton 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI COXA AND BODYndashCOXA JOINT

50 Appendage III (= arachnid leg 1) with coxapophy-sis forming floor or wall of preoral chamber 0 absent1 present [S sim14 WH sim58]State 1 occurs in Opiliones (Hansen amp Soslashrensen 1904)and higher scorpions (Jeram 1998) A lobe is presenton the coxa of leg 1 in the fossil scorpion Prearcturusand perhaps Stoermeroscorpio and is coded here as acoxapophysis Coxapophysis of leg 1 together with theepistome and coxapophysis of the palp form a uniquepreoral chamber the stomotheca Weygoldt (1998) andDunlop amp Braddy (2001) reject the homology of cox-apophyses and stomothecae in Opiliones and Scorpi-ones because these structures are not apparent infossils of those scorpions thought to have been aquaticHowever the coxapophyses in Opiliones are formedlargely from soft cuticular lsquolipsrsquo and similar structuresin fossil scorpions would probably not have beenpreserved51 Appendage IV (= arachnid leg 2) with coxapophy-sis 0 absent 1 present [S sim14 WH sim58 GEWB 75]State 1 occurs in extant Scorpiones and most Opil-iones except Dyspnoi (Shultz 1998)52 At least one pair of coxal gnathobases on append-ages III-VI (= arachnid legs 1ndash4) 0 absent 1 present[S sim23 WH sim66 GEWB 78]State 1 occurs in all appropriately preserved Euryp-terida including Baltoeurypterus (Selden 1981) andStylonurus (Clarke amp Ruedemann 1912) extantXiphosura on appendages IIndashVI and apparentlyProscorpius on appendage III (= arachnid leg 1)(Kjellesvig-Waering 1986 Jeram 1998 contra Giribetet al 2002) Fossilized appendages with gnathobaseshave been found in association with Chasmataspis(Chasmataspidida) suggesting that they may havehad State 1 (Dunlop et al 2004)53 Coxae of appendages IIIndashV (= arachnid legs 1ndash3)with jointed moveable endites 0 absent 1 present [Ssim23]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) and in adequately preserved eurypterids(Clarke amp Ruedemann 1912) (BaltoeurypterusSelden 1981)

PHYLOGENY OF THE ARACHNID ORDERS 251

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

54 Intracoxal muscle a muscle arising on anteriorwall of coxa and inserting on posterior wall 0 absent1 present [GEWB 237]State 1 occurs in Thelyphonida (MastigoproctusShultz 1993) and Amblypygi (Phryus Shultz 1999)It is unclear whether the muscle operating the move-able endite (53) in extant Xiphosura (Manton 1964Shultz 2001) is homologous with this muscle and iscoded here as uncertain It is not present in Steno-chrus (Schizomida) (Shultz unpubl observ)55 Coxae of appendages IIndashVI (= arachnid palp andlegs) with dorsal articulation with carapace 0 absent1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)56 Flabellum (exite) on coxa of appendage VI(= arachnid leg 4) 0 absent 1 present [GEWB sim110]State 1 occurs in extant Xiphosura (Yamasaki et al1988) An apparent exite occurs on appendages asso-ciated with Chasmataspis (Chasmataspidida) (Dunlopet al 2004)57 Insertion process of anteromedial tergo-coxal mus-cle 0 weakly developed 1 well to extremely well devel-oped [GEWB 238]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Boumlrner 1904 Shultz 1999)58 Musculi laterales enlarged lateral tergocoxalmuscle with attachment shifted from coxa to adjacentpleural membrane 0 absent 1 present [S 22 WH 65GEWB 242]State 1 occurs in Araneae and Thelyphonida (Shultz1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI TROCHANTER AND COXAndashTROCHANTER JOINT

59 Coxandashtrochanter joint with complex posteriorarticulation composed of two articulating sclerites 0absent 1 present [S 24 WH 67 GEWB 85]State 1 occurs in Araneae Amblypygi Thelyphonidaand Schizomida (Shultz 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI FEMUR AND TROCHANTERndashFEMUR JOINT

60 Depressor muscle (or homologue) of trochanter-femur joint 0 absent 1 present [WH 42 GEWB 220]State 0 appears to be a unique synapomorphy of Ara-neae (Liphistius Shultz 1989 Aphonopelma Ruh-land amp Rathmayer 1978 Hypochilus present studyother examples Clarke 1984 1986)61 Trochanterndashfemur joint with dorsal hinge or pivotoperated by flexor muscles only 0 absent 1 present[WH sim44 GEWB sim91]State 1 occurs in Palpigradi Other chelicerates have abicondylar articulation and are typically equipped

with antagonistic muscles (Van der Hammen 1985Shultz 1989) but see 60 for an exception62 Superior trochanterndashfemur muscle (or homologue)originating broadly in femur inserting on distal mar-gin of trochanter 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 19892001)63 Basifemurndashtelofemur joint of appendages III andIV (= arachnid legs 1 and 2) in adult 0 absent 1present [S 25 WH 68 GEWB 86]The basifemur and telofemur apparently develop aslsquosisterrsquo articles through division of a parental articlethe femur (eg opilioacariform endeostigmatidpalaeosomatid mites) (Coineau amp Van der Hammen1979 Evans 1992) A single animal can have all legswith divided or undivided femora or have some com-bination of the two typically with the more posteriorlegs having the divided femora Thus anterior legsoften tend to be paedomorphic with respect toposterior legs Most chelicerates have undivided fem-ora on some or all legs but muscles homologous withthose of the basifemurndashtelofemur joint may stilldevelop (66) (Shultz 1989) suggesting that the devel-opment of cuticular and muscular components of ajoint have a degree of developmental and evolutionaryindependence

Undivided femora occur in most extant euchelicer-ates including Xiphosura Palpigradi AraneaeAmblypygi Thelyphonida Schizomida ScorpionesOpiliones Pseudoscorpiones higher Sarcoptiformes(Archegozetes Van der Hammen 1989) many Prostig-mata (Tetranychus Evans 1992) and throughout Par-asitiformes Divided femora occur in all legs in manyearly divergent Acariformes including many Endeo-stigmata basally divergent Sarcoptiformes (egPalaeacarus) and Prostigmata (eg AllothrombiumMicrocaeculus) (Van der Hammen 1989 Evans1992) Only leg 4 has basi- and telofemora in Alycus(Van der Hammen 1989) Basi- and telofemora occuron legs 3 and 4 in Solifugae Ricinulei Opilioacari-formes and Eurypterida (Clarke amp Ruedemann 1912Selden 1981 Shultz 1989)

Giribet et al (2002) coded the lsquoannulusrsquo of certainTrigonotarbida (Shear et al 1987) as a basifemur buta similar non-muscularized annulus is present in theposterior legs of extant Thelyphonida In both casesthe condyles of the trochanterndashannulus and annulusndashfemur joints are aligned in such away that they per-form essentially identical movements and appear notto enhance either the range or degree of leg move-ment [This contrasts with the condition in the lasttwo pairs of legs in Baltoeurypterus where condyles ofthe apparent trochanter-basifemur and basifemurndashtelofemur joints have different arrangements (Selden1981)] The annulus in Thelyphonida appears to func-tion in maintaining constant volume of the tro-

252 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

chanterndashfemur joint by preventing ballooning of thearthrodial membrane thereby minimizing the effect ofinternal fluid pressure on movement64 Basifemurndashtelofemur joint of appendage V(= arachnid leg 3) in adult 0 absent 1 present See 6365 Basifemurndashtelofemur joint of appendage VI(= arachnid leg 4) in adult 0 absent 1 present See6366 Cuticular differentiation of basifemur-telofemurjoint absent but muscles present 0 absent 1 present- inapplicable due to presence of joint (63ndash65)State 1 occurs in appendages of postoral somites IIIndashIV (= arachnid legs 1ndash4) in Xiphosura AraneaeAmblypygi Thelyphonida Schizomida Scorpionesand Pseudoscorpiones (Shultz 1989)67 Circumfemoral ring 0 absent 1 presentThe femora of one or more pairs of legs in anactinot-richid Acari have a basal groove or ring associatedwith slit sensilla Acarologists term that portionproximal to the ring the lsquobasifemurrsquo and that portiondistal to the ring the lsquotelofemurrsquo and they refer tothe basifemur as the lsquosecond trochanterrsquo and thetelofemur as the lsquofemurrsquo Despite the confusingterminology the basifemur discussed in 63ndash65 is nothomologous with the lsquobasifemurrsquo of opilioacariformand parasitiform Acari as both a true basifemurndashtelofemur joint and a circumfemoral ring are presentin legs 3 and 4 of Opilioacariformes (Van der Ham-men 1989)

APPENDAGES OF POSTORAL SOMITES IIIndashVI PATELLA AND FEMURndashPATELLA JOINT

68 Patella of appendage of postoral somite III(= arachnid leg 1) proportionally much longer thanthose of more posterior appendages 0 absent 1presentState 1 occurs in Thelyphonida Schizomida (Hansenamp Soslashrensen 1905 Shultz 1989) and Opilioacari-formes (Van der Hammen 1989)69 Femurndashpatella joint 0 monocondylar several axesof movement and multifunctional muscles 1 bicondy-lar hinge one axis of movement and antagonistic mus-cles 2 hinge one axis of movement flexor muscleswithout muscular antagonists [S sim26 WH sim69 GEWBsim87]State 0 occurs in Solifugae State 1 occurs in Phalang-ida (Opiliones) Scorpiones and PseudoscorpionesState 2 occurs in the remaining taxa (Shultz 1989)70 Patellar plagula 0 absent or with simple medianattachment 1 symmetrical Y-shaped with long prox-imal stem 2 symmetrical U-shaped (= arcuate scler-ite) 3 asymmetrical attaching to patella only atanterior margin [S 27 WH sim70 GEWB 234 = 243]State 1 occurs in extant Xiphosura (Shultz 1989)State 2 occurs in Araneae (Manton 1958 Ruhland amp

Rathmayer 1978 Clarke 1986 Shultz 1989) State 3occurs in Amblypygi Thelyphonida and Schizomida(Shultz 1989) Giribet et al (2002) included this char-acter twice

APPENDAGES OF POSTORAL SOMITES IIIndashVI TIBIA AND PATELLAndashTIBIA JOINT

71 Tibiae divided by one or more joints 0 absent 1presentState 1 occurs in Amblypygi (Weygoldt 2000)72 Patellandashtibia joint 0 monocondylar with or with-out CZY (73) 1 bicondylar hinge one axis of move-ment antagonistic muscles 2 hinge one axis ofmovement muscles without muscular antagonists [Ssim31 WH 74 GEWB sim92] (Shultz 1989 Selden Shearamp Bonamo 1991)73 Patellandashtibia joint with posterior compression zone(lsquoCZYrsquo) 0 absent 1 present - inapplicable due toabsence of monocondylar articulation [GEWB 99]State 1 is a unique synapomorphy of Araneae (Seldenet al 1991)74 Patellandashtibia joint of appendages IIIndashVI(= arachnid legs 1ndash4) with deep-set monocondylar pivotbordered by a pair of tibial processes to which extensormuscles attach 0 absent 1 present - inapplicabledue to absence of monocondylar joint (72)State 1 occurs in extant Xiphosura (Snodgrass 1952Shultz 1989) although a similar arrangement ispresent in the appendage III (= arachnid leg 1) in The-lyphonida (Shultz 1993)75 Patellandashtibia joint largely immobile specializedfor autotomy 0 absent 1 presentState 1 occurs throughout Amblypygi (Weygoldt 19842000)76 Anterior femurndashtibia or femoropatellandashtibia(transpatellar) muscle 0 absent 1 present [S sim30GEWB 222] (Shultz 1989)77 Proximal attachment of posterior femurndashtibia orfemoropatellarndashtibia (transpatellar) muscle 0 muscleabsent 1 dorsoposterior surface of femur andor pos-terior surface of patella 2 distal process of femur mus-cle acting as extensor of femur-patella joint [S sim28 WHsim71 + 72 GEWB 235 = 244 245] (Shultz 1989)78 Distal attachment of posterior femurndashtibia (trans-patellar) muscle 0 posterior 1 dorsal acting as exten-sor of patellandashtibia joint - inapplicable due to absenceof muscle (77) [S 29 GEWB 221] (Shultz 1989)79 Anterior patellandashtibia muscle 0 absent 1 present[S sim32 WH sim73 GEWB 246] (Shultz 1989)80 Posterior patellandashtibia muscle 0 absent 1 present[S 33 WH 76 GEWB 248] (Shultz 1989)81 Patellandashtibia joint spanned by elastic (lsquospringlikersquo)sclerite 0 absent 1 present [GEWB 107]State 1 is a unique synapomorphy of Solifugae (Sens-enig amp Shultz 2003)

PHYLOGENY OF THE ARACHNID ORDERS 253

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

APPENDAGES OF POSTORAL SOMITES IIIndashVI TARSUS AND TIBIAndashTARSUS JOINT

82 Tarsus divided into proximal basitarsus(= metatarsus) and distal telotarsus (= distitarsus orlsquotarsusrsquo) 0 absent 1 presentState 0 occurs in extant Xiphosura (Shultz 1989) andthroughout Acariformes (Lindquist 1984 Evans1992) in the anterior two pairs of legs in chthonioidpseudoscorpions and all legs in lsquomonosphyronidrsquo pseu-doscorpions including Feaelloidea and Cheliferoidea(Chamberlin 1931) A circumtarsal ring (83) associ-ated with slit sensilla in Parasitiformes (Acari) mayrepresent a joint between a telotarsus and basitarsus(Evans 1992)83 Circumtarsal ring 0 absent 1 presentState 1 is an apparent synapomorphy of Parasiti-formes (Evans 1992)84 Telotarsus in adult with two or more tarsomeres 0absent 1 present - inapplicable due to absence of telo-tarsus (82) [GEWB 71]State 1 occurs in Schizomida Thelyphonida Ambly-pygi Palpigradi Ricinulei Solifugae (Shultz 1989)Opilioacariformes (Van der Hammen 1989) and Ple-siosiro (Dunlop 1999)85 Three telotarsomeres on appendages of postoralsomites IV-VI (= arachnid legs 2ndash4) 0 absent 1present - inapplicable due to absence of telotarsus (82)or absence of telotarsomeres (84) [S 34 WH 77 GEWB102]State 1 occurs in Thelyphonida Schizomida andAmblypygi (Shultz 1989)86 Tibiandashtarsus joint spanned by well-developed elas-tic (lsquospringlikersquo) sclerite 0 absent 1 present [GEWB107]State 1 occurs in Scorpiones phalangid Opiliones andSolifugae (Alexander 1967 Shultz 2000 Sensenig ampShultz 2003) Extant Xiphosura is coded for the lastprosomal appendage only (see 88)87 Appendage of postoral somite VI (= arachnid leg 4)with ring of large basally articulated spatulate pro-cesses at tibia-tarsus joint 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988)88 Appendages postoral somites IIIndashV (= arachnidlegs 1ndash3) with tibiotarsus (ie tibia and tarsus not dif-ferentiated) 0 absent 1 presentState 1 occurs in extant Xiphosura (Shultz 1989) butnot apparently in Weinbergina (Moore et al 2005) theonly synxiphosurid in which appendages are suffi-ciently preserved to determine the number ofpodomeres89 Ambulacrum peduncle-like extension of the tarsuswith internal condylophores and terminating distallywith apotele (eg claws) andor pulvillus 0 absent 1present

State 1 is present throughout Acari (Alberti amp Coons1999)

APPENDAGES OF POSTORAL SOMITES IIIndashVI APOTELE AND TARSUSndashAPOTELE JOINT

90 Apotele of appendage III (= arachnid leg 1) 0absent or not apparent 1 present [GEWB 101]State 0 occurs in Amblypygi Thelyphonida and Schi-zomida (Shultz 1999) However the apotele is proba-bly present but undifferentiated because muscles thatwould normally insert on the apotele terminate on theend of the lsquotarsusrsquo Dunlop (1999) reconstructed Hap-topoda as lacking an apotelic claw on leg 1 but noted inthe text that this was speculative Dunlop (2002b) hasreviewed apotelic diversity in Chelicerata91 Appendages of postoral somites IIIndashVI (= arachnidlegs) chelate with chela formed from tibiotarsus andapotele or tarsus and apotele 0 absent 1 presentState 1 occurs in extant Xiphosura (Yamasaki et al1988) The condition appears to occur in Chasmataspisbased on associated appendages (Dunlop et al 2004)but not in Diploaspis and Octoberaspis (Dunlop et al2001 Dunlop 2002a)92 Apotele with eversible or padlike empodium(= pulvillus) in adult 0 absent although empodialclaw may be present 1 present [S sim38 WH 81 GEWB95]State 1 occurs throughout Pseudoscorpiones (Cham-berlin 1931) Solifugae (Roewer 1934) Opilioacari-formes Holothyrida Ixodida and free-livingMesostigmata (Evans 1992) An eversible lsquopulvillusrsquo isprobably a primitive feature of Amblypygi (eg Char-inus Weygoldt 2000) but is absent in higher groups(eg Phrynus Weygoldt 2000)93 Inferior apotele muscle (= claw depressor) with tib-ial attachment 0 absent 1 present [S 35 WH 78GEWB 249]The character is coded for extant Xiphosura usingappendage of postoral somite VI (= arachnid leg 4) asother legs lack a differentiated tibia (88) State 1appears to occur throughout Arachnida (Shultz 1989)94 Inferior apotele muscle (= claw depressor) withpatellar attachment 0 absent 1 present [S 36 WH 79GEWB 250]State 0 occurs in extant Xiphosura Acari AmblypygiPalpigradi and Ricinulei State 1 occurs in all otherarachnids examined here (Shultz 1989)

OPISTHOSOMA

95 Number of opisthosomal somites in adult 0 five1 eight 2 nine 3 10 4 11 5 12 6 13 [WH sim33GEWB sim190]Chasmataspidids appear to have 13 opisthosomalsomites (Dunlop et al 2004) The synziphosurans

254 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Weinbergina (Moore et al 2005) and Limuloides(Anderson amp Selden 1997) appear to have ten opistho-somal somites Moore et al (2005) found no evidence ofan anterior microtergite in Weinbergina as proposedby Anderson amp Selden (1997) Xiphosurids appear tohave nine opisthosomal somites (Scholl 1977 Ander-son amp Selden 1997 Shultz 2001) Members of the fol-lowing orders have 12 opisthosomal somites Araneae(Millot 1949ac) Amblypygi (Weygoldt 2000) Thely-phonida and Schizomida (Hansen amp Soslashrensen 1905)and Pseudoscorpiones (Chamberlin 1931) Elevensomites are present in Palpigradi Solifugae (Roewer1934) and Trigonotarbida (Gilboarachne Shear et al1987 Palaeocharinus Fayers et al 2004) Plesiosiromay have 11 or 12 somites (Dunlop 1999) Justifica-tion for coding of more controversial taxa is providedbelow

Scorpiones Two principal hypotheses regarding thenumber of opisthosomal somites in scorpions havebeen advocated a 13-somite hypothesis derived fromembryological studies (Brauer 1895 Patten 1912Farley 1999 2005) and a 12-somite hypothesis basedon comparative anatomy of adults (Weygoldt amp Pau-lus 1979) The embryological interpretation is basedon the observation of pregenital genital and pectinalsomites (each with segmental ganglia and paired limbbuds) in early scorpion embryos followed by extremereduction or loss of the pregenital somite in laterembryos According to this view a missing pregenitalsomite should be added to the 12 apparent opisthoso-mal somites of post-embryonic scorpions to achieve afinal number of 13 The anatomy-based hypothesiswas introduced by Weygoldt amp Paulus (1979) whoadvocated a literal interpretation of post-embryonicsegmentation based on opisthosomal tergites Specifi-cally these authors argued that the last pair of dorsalendosternal suspensor muscles of non-scorpionarachnids especially Pedipalpi attach to the first(= pregenital) somite that this condition also occurs inscorpions and that there is no reason to invoke a miss-ing pregenital tergite They proposed that the pectinesbelong to the genital somite and that functional spe-cializations of the nervous system for pectinal functiongive the apparence of an extra neuromere duringembryonic development

I recently dissected the prosoma and anterioropisthosoma of the scorpions Centruroides Hadrurusand Heterometrus and focused on the composition ofthe muscular diaphragm (103) that separates thehaemocoelic compartments of the prosoma andopisthosoma (Shultz 2007) The diaphragm is com-posed of a metameric series of axial muscles fromthree somites the anterior somite corresponds to thelast prosomal somite and the posterior somite corre-sponds to the genital somite The middle elements

insert dorsally along a tranverse tendon attached tothe anterior margin of the first tergite These obser-vations are consistent with the embryological inter-pretation that a pregenital somite is present but itstergite is not expressed It appears likely that the thepregenital somite was compressed longitudinally dur-ing the evolution of the diaphragm I code scorpions ashaving 13 opisthosomal somites

Eurypterida The eurypterid opisthosoma is widelyassumed to have 12 somites (Clarke amp Ruedemann1912 Stoslashrmer 1944) However in a speculative paperon the evolutionary morphology of trilobites and che-licerates Raw (1957) proposed that both scorpions andeurypterids have 15 opisthosomal somites Rawassumed that scorpions have 13 apparent opisthoso-mal somites based on the transient pregenital somiteof scorpion embryology that scorpions and eurypteridsare close relatives and should have the same numberof somites and that olenellid trilobites and chelicer-ates always have somites in multiples of three Rawachieved 15 somites in scorpions by assuming the lastopisthosomal somite to be a diplosomite and that thetelson is a postanal somite even though there is noevidence for either of these proposals He attributedthese features to eurypterids as well He also notedthat the connection between the prosomal carapaceand first opisthosomal tergite in Eurypterida differedstructurally from the connection between adjacentopisthosomal tergites and regarded this as evidencefor a reduced pregenital tergite in the prosomandashopisthosoma junction Rawrsquos speculations were largelyforgotten until Dunlop amp Webster (1999) resurrectedthe proposal that eurypterids have a reduced opistho-somal tergite and therefore share a unique similaritywith scorpions Unfortunately Dunlop amp Bullocktreated Rawrsquos conjecture as if it were based on empir-ical evidence rather than an attempt to force euryp-terid morphology into a peculiar numerical system Inthe absence of convincing evidence to the contrary Ihave coded eurypterids as having 12 opisthosomalsomites

Ricinulei The opisthosoma consists of a membranouspedicel (97) bearing the gonopore ventrally Thefemale gonopore is bordered by an anterior plate and aposterior plate The remainder of the opisthosoma iscomposed of thick sclerites separated by less heavilysclerotized cuticle The first dorsal sclerite is short andfunctions as part of a prosomandashopisthosoma couplingmechanism (96) that may or may not be a specializedcomponent of the following tergite This is followed byfour tergites and sternites and a three-segmentedmetasoma (= lsquopygidiumrsquo) Millot (1945 1949f) rea-soned that the opisthosoma contains ten somites withsomites VIIndashIX incorporated into the pedicel XndashXIIIexpressed as tergites and sternites and the metasoma

PHYLOGENY OF THE ARACHNID ORDERS 255

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

comprising four somites He did not regard the dorsalcoupling sclerite as a separate tergite and his inter-pretation of four rather than three metasomal somiteshas been rejected Pittard amp Mitchell (1972) also pro-posed ten opisthosomal somites but achieved thisnumber by regarding the dorsal coupling sclerite as atergite of somite IX and by recognizing three metaso-mal somites Legg (1976) adopted the system proposedby Pittard amp Mitchell but did not regard the dorsalcoupling sclerite as separate from the following tergite(X) Van der Hammen (1979 1989) reconstructed 13somites The pedicel was regarded as having twosomites (VII + VIII) the following four sets of tergitesand sternites were considered diplosomites (IX + XXI + XII XIII + XIV XV + XVI) and the metasoma hadthree somites (XVIIndashXIX) This interpretation mayhave been biased by Van der Hammenrsquos attempt tounite Ricinulei with anactinotrichid Acari which healso regarded as having a primitive number of 13somites Selden (1992) described a fossil ricinuleidTerpsicroton that shows two pairs of depressions onthe three large premetasomal tergites which con-trasts with the single pair seen in extant species Thisobservation appears to corroborate the diplosomitehypothesis but the evidence does not indicate that thetergite anterior to these is a diplosomite

Dunlop (1996) attempted to homologize theprosomandashopisthosoma coupling mechanisms ofRicinulei and Trigonotarbida a goal that required anovel and rather forced interpretation of the dorsalsclerites He regarded the coupling sclerite as homol-ogous with the first opisthosomal tergite (VII) of trig-onotarbids and then followed Van der Hammenrsquosdiplosomite hypothesis to achieve 12 somites in totalDunloprsquos scheme differs from previous systems in sug-gesting that the pedicel does not contain the dorsalelements of the first and second somites and is in-consistent with Millotrsquos (1945) observation that thepre- and postgenital plates each have dorsoventralmuscles Here I code Ricinulei as having 12 somitesThere are three metasomal somites (XVIndashXVIII)three diplosomites (= six somites) (XndashXV) one cou-pling somite (IX) and two somites in the pedicel (iethe genital and pregenital somites) (VII VIII)

Opiliones Harvestmen have nine opisthosomalsomites and an anal operculum that is traditionallyregarded as the tergite of a tenth somite a tenth ster-nite is lacking (Hansen amp Soslashrensen 1904 Winkler1957) However the anal operculum appears to repre-sent a persistent embryonic telson (Moritz 1957) andis therefore likely to be a postsegmental structurecomparable with the stinger of scorpions flagella ofthelyphonids etc (Shultz 2000)

Acari The number of opisthosomal somites is probe-matic for most mite taxa due to uncertainty about the

location of the prosomandashopisthosoma boundry (5)extensive simplification or loss of metamericallyarranged sclerites and muscle attachments a paucityof developmental studies of engrailed expression andin Acariformes opisthosomal anamorphosis and het-erochronic modification of somite number (Evans1992) Ixodids appear to be the exception developmen-tal studies indicate five somites in the opisthosoma ofticks (Evans 1992)

Some mites retain external evidence of segmenta-tion ndash metameric patterns of furrows muscle attach-ments and slit sensilla ndash and with certainassumptions the number of opisthosomal somites canbe estimated Here it is assumed that the dorsal sur-faces of the last two prosomal somites are present inmites and retain their primitive association with legs3 and 4 (contra Van der Hammen 1989) (see 5 for jus-tification) Given this there appear to be 11 somites inthe opisthosoma of Opilioacariformes a conclusionalso reached by other workers (eg With 1904 Sitni-kova 1978 Klompen 2000) Similar reasoning sug-gests that Alycus has seven opisthosomal somites notnine as advocated by Van der Hammen (1989) Unfor-tunately external evidence is ambiguous for determin-ing the number of somites in other mite lineages andthese are coded as uncertain96 Prosoma-opisthosoma coupling mechanism 0absent 1 present [GEWB sim24]The posterior margin of the carapace and the anteriormargin of the first apparent opisthosomal tergite arespecialized as a coupling mechanism in Ricinulei andTrigonotarbida and Dunlop amp Horrocks (1996) pro-posed that these are synapomorphic for the twogroups However there is uncertainty about thehomology of the anterior opisthosomal somites in Trig-onotarbida and Ricinulei (95)97 Pedicel 0 absent 1 aranean type 2 ricinuleidtype [WP 30 S sim40 WH sim20 GEWB sim126]The body narrows at or near the prosomandashopisthosoma juncture in several arachnid lineages (ieSolifugae Palpigradi Amblypygi Araneae Ricinulei)and this lsquowaistrsquo has often been used as character at theinterordinal level (eg Pocock 1893) However thisinterpretation is rejected here for being subjective andfor uniting non-homologous conditions For exampleAraneae and Amblypygi are often grouped on the basisof a lsquopedicelrsquo yet it is a highly specialized structure inAraneae and its parts are not readily homologizedwith those of Amblypygi In contrast the condition inAmblypygi is a slightly narrower version of the highlymoveable prosomandashopisthosoma juncture in Uropygiwhich is not generally considered a lsquopedicelrsquo Thepedicel in Ricinulei is also unique a weakly sclero-tized stalk containing the genital opening (Cryptocel-lus Pittard amp Mitchell 1972 Ricinoides Legg 1976)See 95

256 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

98 Opisthosoma with three- segmented lsquobucklerrsquo 0absent 1 presentState 1 occurs throughout Chasmatapsidida (Dunlop2002a)99 Thoracetron consolidation of tergites of postgeni-tal somites 0 absent 1 present - inapplicable codedonly for Xiphosura (Anderson amp Selden 1997)100 Fusion of tergites of postoral somites VIII and IX(= opisthosomal somites 2 and 3) only 0 absent 1present [GEWB 146]State 1 appears to be a synapomorphy Trigonotarbida(Shear et al 1987) Ricinuleids also have diplotergites(101) but these appear to encompass a different com-bination of somites (see 95)101 Three diplotergites 0 absent presentState 1 is a unique synapomorphy of Ricinulei (Selden1992) See 95102 Paired opisthosomal defensive glands openingvia ducts on either side of the anus 0 absent 1present [S 46 WH 86 GEWB 122]State 1 occurs in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905)103 Muscular diaphragm separating prosomal andopisthosomal compartments formed by dorsoventralmuscles of postoral somites VIndashVIII and extrinsic mus-cles of leg 4 0 absent 1 presentState 1 is known only in extant scorpions (Lankesteret al 1885 Firstman 1973) The diaphragm is oftentreated as a single structure but recent anatomicalwork (Shultz 2007) has shown that it is a composite ofdorsal endosternal suspensors and extrinsic leg mus-cles (see 95) A diaphragm is also present in Solifugaebut it is located more posteriorly and does not appearto be homologous with that of scorpions (Roewer1934)104 Opisthosomal appendicular chondrites 0absent 1 presentThese cartilage-like columns of mesodermally derivedtissue are associated with each opisthosomal append-age including the chilaria in extant Xiphosura(Patten amp Hazen 1900 Yamasaki et al 1988 Fahr-enbach 1999 Shultz 2001)105 Paired appendages on ventral surface of postoralsomite VII (= opisthosomal somite 1) in adult 0absent 1 present [WP sim13 S 39 WH 82 GEWB 143]State 1 is known in Weinbergina (Moore et al 2005)and extant Xiphosura The bilobed structure of themetastoma in certain eurypterids as well as the cor-responding placement of chilaria in Xiphosura haveled some workers to regard the metastoma as fusedappendages of postoral somite VII (eg Stoslashrmer1955) This hypothesis is problematic given that (i) nospecial explanation is needed to account for bilaterallysymmetrical structures in bilaterally symmetricalorganisms (ii) incorporation of the ventral part of thefirst opisthosomal sternite as a functional element of

the prosoma is typical of arachnids and perhapsEuchelicerata generally and (iii) some workers regardthe metastoma as a sternite (eg Jeram 1998) Thusthe metastoma is coded here as uncertain Dunlop hasproposed that the sternum of scorpions is derivedfrom appendages citing the presence of transient limbbuds in scorpion development (Brauer 1895 Patten1912) but this is not compelling evidence for a persis-tent appendicular contribution to the sternite of theadult106 Megoperculum 0 absent 1 present [S sim41 + 42WH sim83 GEWB 162 + 163]State 1 occurs in Thelyphonida Schizomida Ambly-pygi Araneae (Shultz 1993 1999) Trigonotarbida(Shear et al 1987) and probably Haptopoda (Dunlop1999) and Palpigradi The megoperculum consists ofappendages of opisthosomal somite 2 (= genital somiteor postoral somite VIII) (Shultz 1993 1999 Popadicet al 1998) that have fused medially and displacedthe ventral body wall of the somite anteriorly (oftenrepresented by a small sternite to which dorsoventralmuscles of postoral somite VIII attach) The megoper-culum projects posteriorly to form the ventral surfaceof a pregenital chamber The dorsal surface of thechamber is formed by the ventral body wall of opistho-somal somite 3 (= postoral somite IX) which serves asthe ventral attachment of the paired dorsoventralmuscles of that somite The megoperculum bears book-lungs (where present) and sometimes paired gonopodsthat are probably derived from telopodites A similarbut less developed operculum is present on theopisthosomal somite 3 which may also bear booklungsand eversible vesicles (eg Amblypygi) (see 111) corre-sponding to the booklungs and gonopods of the mego-perculum (Shultz 1999)

I contend here that a megoperculum is present inPalpigradi and that this is particularly evident infemales Specifically a large unpaired lobe projectsposteriorly from the genital somite over the ventroan-terior surface of the first postgenital somite therebyforming a pregenital chamber The dorsoventral mus-cles of the genital somite attach ventrally near theanterior border of postoral somite VIII and those ofpostoral somite IX attach to the upper surface of thepregenital chamber (Rucker 1901 Boumlrner 1902bRoewer 1934 Millot 1949d Van der Hammen 1989)Weygoldt (1998) has questioned this interpretation ofPalpigradi107 Postgenital operculum or lsquosternitersquo 0 sclerotized1 not sclerotized - inapplicable coded only for Pan-tetrapulmonata (Platnick amp Gertsch 1976)State 1 is present in Trigonotarbida Amblypygi The-lyphonida Schizomida and Mesothelae (Araneae) butnot Opisthothelae (Araneae)108 Genital opening of female guarded by four plates(one pregenital one postgenital two laterogenitals)

PHYLOGENY OF THE ARACHNID ORDERS 257

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

genital opening of male guarded by two plates 0absent 1 presentThis an apparent synapomorphy of Holothyrida(Acari) (Van der Hammen 1989)109 Anterior margin of genital opening in male withglands secreting via fusules 0 absent 1 presentState 1 occurs in Araneae as epigastric or lsquoepiandrousrsquoglands (Marples 1967) and in Palpigradi (Condeacute1991b)110 Paired valve-like plates apparently formed fromcomponents of three somites covering triradiate genitalopening 0 absent 1 presentThis is an apparent synapomorphy of Acariformes(Evans 1992 Alberti amp Coons 1999)111 Eversible lsquoappendagesrsquo on the ventral surface ofpostgenital somites 0 absent 1 present [S sim43 WHsim84 GEWB sim127]Paired ventral lsquoappendagesrsquo operated in part byhaemolymph pressure occur in the form of ventralsacs in certain Amblypygi (postoral somite IX) (Wey-goldt 2000) (Charinus Millot 1949b not PhrynusShultz 1999) prokoenenian Palpigradi (postoralsomites XndashXII) (Rucker 1901 Condeacute 1991a) as spin-nerets in Araneae (XndashXI) (Shultz 1987) and as geni-tal papillae in many Acariformes (Alberti amp Coons1999) including Oribatida many Endeostigmata (Aly-cus Van der Hammen 1989) and Prostigmata (Allo-thrombium Saboori amp Kamali 2000 MicrocaeculusEvans 1992) Eversible vesicles similar to those ofamblypygids occur in fossils of the trigonotarbidPalaeocharinus (postoral somite IX) (Fayers et al2004) Two cuticular structures (lsquogenital verrucaersquo)occur anterior to the genital opening in Opilioacari-formes and each covers the opening to a thin-walledinvaginated sac (lsquogenital papillarsquo) of unknown function(Van der Hammen 1989) Van der Hammen proposedthat it is homologous to a genital papilla of Acari-formes but Alberti amp Coons (1999) have questionedthis interpretation on several grounds and refer tothese structures as pregenital capsules112 Opisthosomal silk glands and spinnerets derivedfrom appendages on postoral somites X and XI(= opisthosomal somites 4 and 5) 0 absent 1 present[WP 33 WH 41 GEWB 123 + 142]State 1 is a unique synapomorphy of Araneae (Plat-nick amp Gertsch 1976) although potentially homolo-gous glands and spigot-like setae occur on the ventralopisthosomal surface in Palpigradi (Millot 1943Condeacute 1991a)113 Opisthsosomal spinnerets location 0 near mid-dle of opisthosoma 1 near posterior end of opisthso-som ndash inapplicable coded only for Araneae (Platnickamp Gertsch 1976)114 Anterior medial lsquospinneretsrsquo 0 absent 1 present- inapplicable coded only for Araneae (Platnick ampGertsch 1976)

115 Opisthosomal tergites divided longitudinally intoone median and two lateral plates 0 absent 1 present[GEWB 145]State 1 occurs in non-curculioid Ricinulei (Selden1992) and many Trigonotarbida (Dunlop 1996) Thisfeature or something very like it (eg distinct opistho-somal trilobation) occurs sporadically in severaleuchelicerate groups including Chasmataspis (Chas-mataspidida) (Dunlop et al 2004) most fossil xipho-surans (Anderson amp Selden 1997) and a few derivedEurypterida (especially Mixopteroidea) (Tollerton1989)116 Number of metasomal somites 0 zero 1 two 2three 3 five 4 nine [S sim44 WH sim85 GEWB 128 144]The metasoma is a preanal region comprising multiplesomites that are substantially narrower than the pre-ceding somites or mesosoma Metasomal somitesoften lack pleural membranes and take the form ofsclerotized rings State 1 occurs in TrigonotarbidaState 2 is present in synziphosurid Xiphosura (Ander-son amp Selden 1997) Amblypyi Thelyphonida Schizo-mida and Ricinulei State 3 occurs in Scorpiones andEurypterida and State 4 occurs in Chasmatispidida

Some workers appear to consider a five-segmentedmetasoma as conclusive evidence for the monophylyof a Eurypterida + Scorpiones clade (eg Dunlop ampBraddy 2001) presumably because the character rep-resents a kind of tagmosis and should therefore begiven substantial phylogenetic weight However Wey-goldt (1998) has pointed out that a metasoma contain-ing three somites is probably part of the ground planof Euchelicerata The known diversity of the eucheli-cerate metasoma indicates that this feature can inincrease or decrease its segmental composition in evo-lution or be eliminated entirely without necessarilychanging the total number of opisthosomal somitesConsequently it is problematic to assume that thefive-segmented metasoma has any special immunityto homoplasy117 Postanal structure (telson) 0 absent or not obvi-ously developed 1 presentPostanal structures are considered non-segmentalbecause they are modifications of the embryonicregion posterior to the site of somite addition (embry-onic growth zone) A well-developed postanal struc-ture persists in adults in Xiphosura EurypteridaChasmataspidida Scorpiones Schizomida Thely-phonida and Palpigradi The anal operculum of Opil-iones is generally considered by morphologists to bethe tergite of the tenth opisthosomal or anal somite(eg Hansen amp Soslashrensen 1904) but embryological evi-dence (Moritz 1957) suggests that it is derived fromthe postproliferative zone and thus corresponds to thetelson of other chelicerates (see 95)118 Postanal structure shape 0 caudal spine 1aculeus 2 flagellum 3 anal operculum - inapplica-

258 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

ble due to absence of postanal structure (117) [WP sim41S sim45 WH sim24 GEWB sim121 129 + 147]State 0 is present in Xiphosura Eurypterida andChasmataspidida State 1 occurs in Scorpiones State2 occurs in Palpigradi Thelyphonida and SchizomidaState 3 occurs in Opiliones (see 95 117)119 Specialized postanal flagellum in male (see 157ndash159) 0 absent 1 present - inapplicable coded onlyfor taxa with postanal flagellum (118) [GEWB 131]State 1 is a unique synapomorphy of Schizomida(Hansen amp Soslashrensen 1905 Cokendolpher amp Reddell1992 Reddell amp Cokendolpher 1995)

RESPIRATORY SYSTEM

120 Respiratory medium 0 water 1 airAmong the terminal taxa included only fossil scorpi-ons are problematic for this character They are codedhere as uncertain Based on a study of book lungmicrosculpture Scholtz amp Kamenz (2006) have arguedthat arachnids are primitively terrestrial and pulmo-nate (see also Firstman 1973) and have questionedwhether any fossil scorpions were aquatic No positionon this proposal is taken here121 Respiratory lamellae on opisthosomal somite 2(= genital somite postoral somite VIII) 0 absent 1present [WP sim37 S sim51 WH sim22 GWEB sim133ndash137]This character encompasses book gills (= lamellae thatfunction in water) and book lungs (= lamellae thatfunction in air) The distinction between book gills andbook lungs is accommodated here by 120 in combina-tion with 121ndash124 Respiratory lamellae are presenton the genital somite in Trigonotarbida AraneaeAmblypygi Thelyphonida and Schizomida Petrunk-evitch (1949) reconstructed Plesiosiro as having booklungs but Dunlop (1999) could not corroborate thisThis character was used by Dunlop amp Webster (1999)to propose that Xiphosura and Scorpiones are closelyrelated because they both lack respiratory lamellae onthe genital somite Dunlop amp Braddy (2001) alsoargued for the placement of Eurypterida with Xipho-sura and Scorpiones based in part on this characterAt least some eurypterids may have had respiratorylamellae (eg Manning amp Dunlop 1995) but the onlyevidence of their segmental distribution is derivedfrom one specimen (Braddy et al 1999)122 Respiratory lamellae on opisthosomal somite 3(= postoral somite IX) 0 absent 1 present [GEWB136]State 1 occurs in Amblypygi Thelyphonida and Xipho-sura It is a groundplan feature of Araneae (Platnick ampGertsch 1976) and is present in all representativetaxa included here Braddy et al (1999) proposed theexistence of respiratory lamellae on postoral somitesIXndashXII in Eurypterida based on evidence from onespecimen This character is coded as uncertain for the

Eurypterida included here Extant scorpions are codedhere as lacking respiratory lamellae on this somite(= pectinal somite) based on the interpretation of theopisthosomal segmentation discussed in 95123 Respiratory lamellae on opisthosomal somites 4ndash6 (= postoral somites XndashXII) 0 absent 1 presentState 1 is definitely present in extant Xiphosura andScorpiones and may have occurred in Eurypterida124 Respiratory lamellae on opisthosomal somite 7(= postoral somite XIII)State 1 occurs in extant Xiphosura and Scorpionesonly125 Kiemenplatten 0 absent 1 presentThese structures are located on the roof of opercularchambers in Eurypterida (Clarke amp Ruedemann1912) where they take the form of ventrally projectingcones with a distinct cuticular microsculputure (Man-ning amp Dunlop 1995) Dunlop amp Braddy (2001)inferred the existence of Kiemenplatten in all Palaeo-zoic scorpions based on a description and photos of onespecimen of Paraisobuthus duobicarinatus by Kjelles-vig-Waering (1986 pls 16ndash18) The plates depict darkcone-like rearward-pointing denticles distributedwithin a white amorphous material The denticlesappear to lack microsculpture even though the mag-nifications at which the photos were taken (times90ndashtimes330)are comparable with those illustrating the cones ofKiemenplatten (times170ndashtimes350 in Manning amp Dunlop1995 figs 1 2) Given the diversity of cuticular struc-tures present in the atria of booklungs book gills andtracheae the denticles appear to bear no special sim-ilarity to the cones of Kiemenplatten126 Tracheal system 0 absent 1 paired ventral stig-mata on postoral somite VIII (= opisthosomal somite2) 2 paired ventral stigmata on postoral somites IXand X 3 one pair of stigmata opening near legs 3 or 44 paired stigmata associated with chelicerae 5 fourpairs of stigmata on dorsal surface of opisthosoma -inapplicable aquatic (120) [WP 40 sim43 S sim52 + 5354 WH sim23 45 88 GEWB sim138 139]Firstman (1973) and Weygoldt amp Paulus (1979)hypothesized that tracheae are homologous in all tra-cheate arachnids except spiders and subsequentworkers have entertained this hypothesis by includinga character for the presenceabsence of tracheal sys-tems that ignores the diverse arrangement of stig-mata in arachnids (eg Shultz 1990 Wheeler ampHayashi 1998 Giribet et al 2002) However thisapproach assumes that internal tracheal systems areconserved but that tracheal openings (stigmata)appear and disappear on different parts of the bodywith higher evolutionary frequency Here I assumethat stigmata are conserved in evolution and that dif-ferences in their anatomical placement reflect the evo-lution of new tracheae State 1 occurs throughoutOpiliones State 2 occurs throughout Solifugae

PHYLOGENY OF THE ARACHNID ORDERS 259

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

(Roewer 1934) and Pseudoscorpiones (Chamberlin1931) State 3 recognizes the possible homology of stig-mata in Ricinulei Opilioacariformes and Parasiti-formes (Acari) (Lindquist 1984 Van der Hammen1989) Tracheal systems are apparently absent in thegroundplan of Acariformes but Prostigmata typicallyhave tracheae associated with the chelicerae Tra-cheae derived from the posterior pair of book lungs arewidespread in araneomorph Araneae but the booklungs are retained and tracheae are absent in themost basally divergent groups (eg Hypochilus)

BOXndashTRUSS AXIAL MUSCLE SYSTEM (BTAMS)

127 Posterior oblique muscles of BTAMS of postoralsomites IndashVI 0 absent 1 present in one or moresomitesState 0 occurs in Xiphosura and state 1 occurs inPalpigradi Araneae Amblypygi Thelyphonida andSchizomida (Shultz 2001) The condition in AcariRicinulei Opiliones and Solifugae is not known128 Anterior oblique muscles of BTAMS posterior topostoral somite VI 0 absent 1 presentState 1 occurs in Xiphosura which is probably theprimitive condition based on comparison with otherarthropods (Shultz 2001) State 0 occurs in all arach-nids examined thus far129 Ventral attachments of posterior oblique musclesof opisthosomal BTAMS located in prosoma 0 absent1 presentState 1 occurs in extant Xiphosura (Shultz 2001)130 Endosternite fenestrate 0 absent 1 present [S 8WH 54 GEWB 35]State 1 occurs in Thelyphonida and hubbardiid Schi-zomida (Firstman 1973) The condition in protoschi-zomid Schizomida is not known131 Suboral suspensor a tendon that arises from theBTAMS and inserts on the ventral surface of the oralcavity via muscle 0 absent 1 presentState 1 occurs in Palpigradi (Eukoenenia Millot1943) Amblypygi (Shultz 1999) and Thelyphonida(Shultz 1993)132 Perineural vascular membrane in adult 0absent 1 present [WH 28 GEWB 155]Wheeler amp Hayashi (1998) coded extant Xiphosura asunknown although the primary source for this char-acter (Firstman 1973) stated that adult Limulus havea perineural vascular membrane Within Arachnidapresence of a perineural vascular membrane is appar-ently correlated with the presence of tracheae (126)133 Ventral endosternal suspensor attaching on coxaof anteriorly adjacent somite 0 absent 1 present[GEWB 239 253]State 1 occurs in Amblypygi (Shultz 1999) Thely-phonida (Shultz 1993) and Schizomida (Shultzunpubl observ)

134 Posteriormost postoral somite with a pair of dor-soventral muscles 0 VI 1 VII 2 VIII 3 XII 4 XIII5 XIV 6 XV 7 XVIDorsoventral muscles tend to run in a continuousmetameric series beginning in the prosoma Deter-mining the last tergite on which the muscle seriesends is substantially easier than counting especiallygiven frequent anatomical complexities near theprosomandashopisthosoma juncture The dorsoventralmuscle series ends on postoral somite VII inCyphophthalmi (eg Chileogovea) and Laniatores(Opiliones) (Shultz unpubl observ) on postoralsomite VII or VIII in Eupnoi (Leiobunum Shultz2000 unpubl observ) on postoral somite XII in Pal-pigradi (Roewer 1934) Solifugae (Bernard 1896Roewer 1934 Millot amp Vachon 1949) and Araneae(Liphistius Hypochilus Millot 1933 but segmenta-tion reinterpreted here following Shultz 1993 1999)on postoral somite XIII in hubbardiid Schizomida(Cokendolpher amp Reddell 1992) and Scorpiones(Lankester et al 1885) and on postoral somite XIVin extant Xiphosura (Shultz 2001) Thelyphonida(Shultz 1993) protoschizomid Schizomida (Cokendol-pher amp Reddell 1992) and Amblypygi (Shultz 1999)Dunlop (1999) reconstructed Plesiosiro as havingpaired tergal apodemes ending on postoral somite XVbut his figures indicate that the series ends on posto-ral somite XIV The interpretation of opisthosomalsegmentation in Ricinulei coded here (95) indicatesthat the series ends on postoral somite XV in Terpsi-croton and this is treated as the groundplan for theorder The posteriormost dorsoventral muscles inPseudoscorpiones occur on postoral somite XVI(Vachon 1949)

NERVOUS SYSTEM

135 Segmental ganglia 0 consolidated in prosoma1 one or more present in opisthosoma [WH 30 GEWB210 + 216]Adult neuromeres are exclusively prosomal in Ambly-pygi liphistiomorph and araneomorph Araneae Opil-iones Pseudoscorpiones Ricinulei and Acari (Millot1949a)136 Dorsal median eyes 0 absent 1 present [WP14 + 47 S 50 WH sim7 GEWB 1]Median eyes are a groundplan feature of Euchelicer-ata and are retained in extant Xiphosura Eurypteridaand occur in all arachnid orders except PalpigradiSchizomida Ricinulei and Pseudoscorpiones (Paulus1979) Median eyes are also absent in opilioacariformand parasitiform Acari (Lindquist 1984) and perhapsthe fossil whipscorpion Proschizomus (Dunlop amp Hor-rocks 19951996) They are unknown in synziphosu-ran Xiphosura Median eyes are present in a variety ofbasal acariform mites including Endeostigmata (but

260 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

not Alycus) Prostigmata (Microcaeculus) and certainoribatids (Palaeacarus) (Evans 1992 Alberti amp Coons1999) Giribet et al (2002) coded all representativeAcari as lacking median eyes It is not known whetherthe eyes of some cyphophthalmid opilions are medianor lateral evidence from tracheal branching inCyphophthalmus (Janczyk 1956) suggests that theyare median eyes (Shultz amp Pinto da Rocha 2007) andpresence of a tapetum in Stylocellus is consistent withlateral eyes (Shear 1993)137 Retinula cells of dorsal median eyes 0 organizedinto closed rhabdoms 1 organized into network ofrhabdomeres 2 disorganized - inapplicable due toabsence of median eyes (136) [GEWB sim3]State 0 is present in Scorpiones Thelyphonida andAmblypygi State 1 is present in Solifugae and Ara-neae state 2 is present in Xiphosura (Paulus 1979)The retinula cells of median eyes have been studied inseveral prostigmatid Acariformes and are organized ina network in some taxa and in an irregular pattern inothers (Alberti amp Coons 1999) Retinulae in phalangidOpiliones have state 0 proximally and state 2 distally(Schliwa 1979)138 Ventral median eyes 0 absent 1 presentState 1 occurs in early instars of extant Xiphosura(Paulus 1979)139 Lateral eyes 0 absent 1 presentLateral eyes are primitively present in Chelicerataand are absent in Palpigradi and Opiliones (Paulus1979) It is unclear whether the eyes of cyphoph-thalmid opilions are median or lateral (see 136)140 Arrangement and number of lateral eyes 0 com-pound many 1 five or more pairs (includes micro-lenses) 2 three primary pairs (excludes microlenses)3 two pairs 4 one pair - inapplicable due to absenceof lateral eyes (139) [WP sim13 + 18 + 38 + 44 + 52 Ssim49 WH sim10 GEWB sim4 + 5]True compound eyes are present in XiphosuraEurypterida Chasmataspidida and many PalaeozoicScorpiones Among the anactinotrichid Acari Opilioa-cariformes have two (Neocarus) or three (Siamacarus)pairs of lateral eyes (although at least one species ofSiamacarus lacks eyes) the allothryrid Holothyridaand many Ixodida have a single pair of lateral lensesand the remainder apparently lack eyes (Evans1992) Thelyphonida was coded as lsquo12rsquo to reflect fivepairs of lenses comprising three pairs of primarylenses and two pairs of small accessory lenses (notsimply three pairs as coded by Giribet et al 2002)Trigonotarbida is also coded as lsquo12rsquo to reflect threepairs of primary lenses and multiple small accessorylenses Dunlop (1999) illustrated Haptopoda as havingpaired triads of lateral eyes but noted that there wasactually no evidence of this in the fossils The lateraleyes of fossil ricinuleids have two pairs of lenses(Selden 1992) One pair of eyes or eyespots are

present in extant Ricinulei and many Schizomidaalthough five genera of hubbardiid Schizomida have apair of lenses (Reddell amp Cokendolpher 1995) State 1is the groundplan for extant Scorpiones and State 3occurs in the pseudoscorpions Chthonius FeaellaNeobisium (but State 4 in Chelifer) and Solifugae(Paulus 1979)141 Lateral eyes with closed rhabdoms 0 absent 1present - inapplicable due to absence of lateral eyes(139) [WP 21 WH 12 GEWB 6]Presence of closed rhabdoms is probably primitive forChelicerata and is retained in extant Xiphosura andScorpiones Retinula cells form a network of rhab-domeres in other extant chelicerates (Paulus 1979)but these networks can differ substantially in detailWeygoldt (1998) and other workers have given sub-stantial weight to the network character in unitingnon-scorpion arachnids but these authors seem not togrant comparable phylogenetic significance to theanalogous character of the median eyes shared bySolifugae and Araneae (137)142 Slit sensilla 0 absent 1 present [WP 19 S 47WH 11 GEWB 209]State 1 occurs in all arachnid orders except Palpigradi(Shultz 1990) The proposal that lsquoprimitiversquo slit sen-silla were present in Eurypterida (Dunlop amp Braddy1997) appears to be based on the over-interpretation ofa comparatively large notch that occurs in Baltoeu-rypterus at the terminus of the tibia (= podomere 7)(see also Edgecombe et al 2000 Giribet et al 2002)The neural construction of slit sensilla like function-ally similar campaniform sensilla of hexapods is sim-ilar to that of trichoid sensilla (Chapman 1998Klompen 2000) Thus these cuticular stress receptorsmay represent modified bases of sensory setae a viewsupported by the replacement of setae by slit sensilladuring post-embryonic development in some Opilioac-ariformes (Klompen 2000)143 Trichobothria 0 absent 1 present [GEWB 213]Trichobothria are present in extant Scorpiones(Jeram 1998) Pseudoscorpiones (Chamberlin 1931)some endeostigmatid (Alycus Van der Hammen 1989)and prostigmatid Acariformes and most Oribatida(Lindquist 1984) They occur on the ventral surface ofthe pedal femora in the opilioacariform Siamacarus(Leclerc 1989) They are also present in Araneae (Foe-lix 1996) Amblypygi (Weygoldt 2000) Schizomidaand Thelyphonida (Hansen amp Soslashrensen 1905) Theyare apparently absent in all non-arachnid chelicer-ates Solifugae Ricinulei Opiliones (Reissland ampGoumlrner 1985) and Parasitiformes (Acari) (Lindquist1984)144 Tibial trichobothria with 2-1-1-1 pattern onappendages IIIndashVI (= arachnid legs 1ndash4) 0 absent 1present - inapplicable due to absence of trichobothria(143) [S 48 WH 87 GEWB 88]

PHYLOGENY OF THE ARACHNID ORDERS 261

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

State 1 occurs only in Thelyphonida and Schizomida(Hansen amp Soslashrensen 1905) Note that Shultz (1990)erroneously described this character as a 2-2-1-1 pat-tern and that this error was repeated by Giribet et al(2002) and described as 2-1-1 by Wheeler amp Hayashi(1998)145 Paired trichobothria on dorsal surface ofprosoma 0 absent 1 present - inapplicable due toabsence of trichobothria (143)State 1 is an apparent synapomorphy of Acariformesand is present in all acariforms included in this study(Alberti amp Coons 1999)146 Malleoli 0 absent 1 present [WP 45 GEWB 96]State 1 is a unique synapomorphy of Solifugae(Roewer 1934)147 Pectines 0 absent 1 present [WP 20 WH 38GEWB 120]State 1 is a unique synapomorphy of Scorpiones148 Intercheliceral median organ 0 absent 1present [GEWB 212]State 1 is a unique synapomorphy of Palpigradi(Roewer 1934)149 Tarsal organ on appendage of postoral somite III(= arachnid leg 1) (= Hallerrsquos organ) sensilla containedwithin a cuticular depression on the superior surface ofthe tarsus of appendage III (= arachnid leg 1) 0absent 1 present [GEWB sim100]Klompen (2000) has noted that State 1 occurs in Opil-ioacariformes Parasitiformes (except Mesostigmata)and Ricinulei where it also occurs on leg 2 (Talaricoet al 2005) State 1 occurs on all legs in Araneae andperhaps Scorpiones (Foelix 1985)150 Tarsal organ on appendage of postoral somite IV(= arachnid leg 2) 0 absent 1 present (See 149)

REPRODUCTION

151 Gonads 0 primarily prosomal 1 primarilyopisthosomalState 0 is limited to Xiphosura (Sekiguchi 1988)State 1 occurs throughout Arachnida (Millot1949a)152 Ladder-like opisthosomal gonadsaccessoryglands (see 153) 0 absent 1 present [WH sim37 GEWBsim158]Giribet et al (2002) followed Wheeler amp Hayashi(1998) who followed Clarke (1979) in coding gonads asreticulate (Xiphosura) ladder-like (Scorpiones Thely-phonida Schizomida) or lsquosaccularrsquo (all remainingArachnida) However comparisons between Xipho-sura and Arachnida are problematic given that thexiphosuran gonads are primarily prosomal and thoseof arachnids are primarily opisthosomal (151) Fur-ther the reticulate pattern in Limulus (Xiphosura)but not other extant xiphosurans contains a distinctlyladder-like component The lsquosaccularrsquo state is probably

artificial as it encompasses a wide variety of pairedand unpaired structures153 Male gonads in two distinct parts one producingsperm and another (tubular gland) producing a holo-crine secretion similar to degenerate sperm 0 absent1 presentState 1 occurs in Thelyphonida Schizomida andAmblypygi although the holocrine material is pro-duced by ventral organs in Amblypygi and dorsalorgans in Thelyphonida and Schizomida (Alberti2005)154 Number of gonopores 0 two 1 oneExtant xiphosurans have two small genital openingson the base of the genital telopodite and all extantarachnids have a single opening The condition inEurypterida is not known Clarke amp Ruedemann(1912) located a pair of openings near the base of themedian organ (161) that are the outlets of the lsquohornorgansrsquo but it is unclear whether these are genitalducts or accessory structures Braddy amp Dunlop (1997)have developed numerous speculations about thesestructures and extended their arguments far beyondthe available evidence155 Genital opening (ie gonopore or gonostome)appearing to open in prosomal region (ie between legcoxae or anterior to posterior carapacal margin) 0absent 1 present [WP sim50 WH 26 GEWB 166]The genital opening in Euchelicerata is located onpostoral somite VIII (= opisthosomal somite 2) but ithas shifted anterior to the posterior margin of thecarapace or between the coxae of the last pair of legsin most Scorpiones (but not in PalaeoscorpiusKjellesvig-Waering 1986) and Opiliones The geni-tal opening occurs near or anterior to the last coxaein the opilioacariform and parasitiform Acari repre-sented here It is variable in Acariformes but islocated posterior to the coxae in all representativetaxa156 Ovipositor 0 absent 1 present [WP sim51 S sim60WH sim91 GEWB sim172]An ovipositor with a trilobed terminus is an apparentgroundplan character of Oribatida (Lindquist 1984Alberti amp Coons 1999) An ovipositor is also present inOpilioacariformes and Opiliones (Van der Hammen1989)157 Stalked spermatophore attached to substratum0 absent 1 present [S 57]State 1 occurs in Scorpiones PseudoscorpionesAmblypygi Thelyphonida and Schizomida The mech-anism of sperm transfer is unknown in PalpigradiOpilioacariformes and Holothryrida158 Male turns posterior end toward female duringmating behaviour 0 absent 1 presentState 1 occurs in Amblypygi (Weygoldt 2000) The-lyphonida and Schizomida (Weygoldt amp Paulus1979)

262 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

159 Female grasps male opisthosoma during matingbehaviour 0 absent 1 present [WP 29 S 58 WH 19GEWB 188]State 1 occurs in Thelyphonida and Schizomida (Wey-goldt amp Paulus 1979) Evolution of this behaviour wasprobably facilitated by 158160 Penis 0 absent 1 present [WP sim51 GEWB 167]State 1 is a unique synapomorphy of Opiliones A truepenis occurs in Phalangida A clearly homologousstructure is present in Cyphophthalmi (Opiliones) andapparently functions in depositing a spermatophore inthe femalersquos genital chamber (Karaman 2005) A truepenis may occur in some mites but it does not appearto be a groundplan feature of any major group (Evans1992) The lsquopenisrsquo in Oribatida is really a spermatop-ositor it functions in construction of a spermatophore(Alberti amp Coons 1999)161 Median organ 0 absent 1 presentState 1 occurs throughout Eurypterida (Clarke amp Rue-demann 1912) and has been observed in diploaspididChasmataspidida (Dunlop 2002a also Loganamaras-pis Tetlie amp Braddy 2004)

SPERM MORPHOLOGY

162 Nucleus with manchette of microtubules 0absent 1 present [S 54 WH 89 GEWB 193]State 1 occurs in Araneae Amblypygi Thely-phonida Schizomida Ricinulei and gonyleptid Lani-atores (Opiliones) (Alberti 1995 Giribet et al2002)163 Axoneme 0 absent 1 present [WP sim49 S sim55WH sim25 GEWB 195]State 1 occurs in Xiphosura Scorpiones Cyphoph-thalmi (Opiliones) Pseudoscorpiones RicinuleiAraneae Amblypygi Thelyphonida and Schizomida(Alberti 1995)164 Coiled axoneme 0 absent 1 present - inappli-cable due to absence of axoneme (163) [WP 22 S sim55WH 13 GEWB 196]State 1 occurs in Pseudoscorpiones Ricinulei Ara-neae Amblypygi Thelyphonida and Schizomida(Alberti 1995)165 Microtubule arrangement in axoneme 0 9 + 0 19 + 1 2 9 + 2 3 9 + 3 - inapplicable due to absence ofaxoneme (163) [WP sim26 S sim56 WH sim16 GEWB 198](Alberti 1995)166 Helical or corkscrew shaped nucleus 0 absent 1present [GEWB sim204]State 1 occurs in Araneae Amblypygi ThelyphonidaSchizomida and certain Scorpiones (eg Hadrurus)(Alberti 1995)167 Vacuolated-type sperm 0 absent 1 present[GEWB 205]State 1 is unique to Opilioacariformes and Parasiti-formes (Acari) (Alberti 1995)

168 Sperm aggregates 0 absent 1 present [GEWB206] (Alberti 1995)

DEVELOPMENT

169 Yolk in early embryo 0 concentrated (centroleci-thal or telolecithal) 1 evenly distributed (isolecithal)[WH sim34 GEWB sim191]Wheeler amp Hayashi (1998) and Giribet et al (2002)coded this character as lsquo0 isolecithal or telolecithal 1centrolecithalrsquo based on information presented byYoshikura (1975) Their coding appears to highlightseparate conditions found in Scorpiones (ie isoleci-thal and telolecithal) rather than a property intrinsicto the character itself170 Embryonic nutrition other than yolk 0 absent 1present [WP sim49]State 1 occurs in extant Scorpiones (except buthidsand chaerilids) and Pseudoscorpiones (Weygoldt1969)171 Embryological growth zone 0 initiating segmentaddition within prosoma 1 initiating segment addi-tion posterior to prosoma [GEWB sim192]Giribet et al (2002) followed Dunlop amp Webster (1999)in stating that only Xiphosura and Scorpiones havestate 0 among Chelicerata However the last prosomalsomite develops from the growth zone in Ixodida(Anderson 1973 Evans 1992) A similar processapparently occurs in the acariform mites Tyroglyphus(Sarcoptiformes Astigmata) (Yoshikura 1975) andArchegozetes (Sarcoptiformes Oribatida) (Thomas ampTelford 1999) but not apparently in Tetranychus(Prostigmata) (Dearden Donly amp Grbic 2002)172 Eggsembryos maintained in external attachedbrood sac secreted by genital glands 0 absent 1present [WP sim46 S 59 WH 90 GEWB 208 = 219]State 1 occurs throughout Amblypygi SchizomidaThelyphonida and Pseudoscorpions Giribet et al(2002) followed Shear et al (1987) in homologizingspinneret-derived silken egg sacs of spiders with gen-ital-gland-derived brood sacs of Pedipalpi However asthese structures are not homologous in secretory ori-gin construction or composition (Shultz 1987) theyare not regarded here as homologous Giribet et al(2002) miscoded the character as absent in Pseudo-scorpiones and without justification in PalpigradiFemale palpigrades have well-developed glands asso-ciated with postgenital somites and an array of setaefusules and other structures (112) (Boumlrner 1902b Mil-lot 1942 Condeacute 1991a) which are consistent withsome sort of brood care and Palpigradi should there-fore be coded as unknown for this character173 Embryonic and early postembryonic lsquolateralrsquo orClaparegravede organs associated with coxa of postoralsomite IV (= arachnid leg 2) 0 absent 1 present [WHsim31 GEWB sim211]

PHYLOGENY OF THE ARACHNID ORDERS 263

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Bilaterally paired rounded protuberances locatedbetween coxae of appendages of postoral somites IIIand IV (arachnid legs 1 and 2) in prelarval and larvalinstars appear to be a primitive feature of Acariformes(Lindquist 1984 Evans 1992 Alberti amp Coons 1999)It is known to be derived embryologically from thecoxa of leg 2 in the oribatid Archegozetes (Thomas ampTelford 1999) Structures apparently homologouswith the acariform Claparegravede organ are present inembryonic Amblypygi (Weygoldt 2000) and Thely-phonida (Yoshikura 1975) and in embryonic and earlypost-embryonic stages of Solifugae (Roewer 1934)where they are termed lateral organs The so-calledlsquolateral organrsquo of embryonic Xiphosura (174) occurs onthe lateral surface of the carapace and does not appearto be homologous with the coxa-associated structuressharing their name In fact the xiphosuran lateralorgan develops even when the coxa of postoral somiteIV has been excised (Sekiguchi 1988) In contrastWheeler amp Hayashi (1998) and Giribet et al (2002)followed Yoshikura (1975) in regarding all lsquolateralorgansrsquo as homologous and in miscoding this characteras absent in Acariformes Van der Hammen (1989)suggested that the sternal verrucae of Opilioacari-formes are homologous with the Claparegravede organ butit differs in gross structure and ontogenetic timing174 Embryonic lateral organ associated with cara-pace 0 absent 1 present [WH sim31 GEWB sim211]State 1 is known only from embryonic stages of extantXiphosura (Yoshikura 1975) In contrast to previousinterpretations this feature does not appear to behomologous with the lsquolateralrsquo or Clarapegravede organ ofcertain arachnids (173)175 Live birth 0 absent 1 presentState 1 occurs in extant Scorpiones and has evolvedseveral times in a few mites (Evans 1992)176 Hexapodal larva and 1ndash3 nymphal stages 0absent 1 present [WP 53 S sim61 WH 27 GEWB 186]State 1 occurs in Acari and Ricinulei (Lindquist1984)177 Hexapodal prelarva 0 absent 1 present [S sim61WH 47]State 1 occurs in Opilioacariformes (NeocarusKlompen 2000) and many Acariformes (Evans 1992)

EXCRETORYOSMOREGULATORY SYSTEM

178 Malpighian tubules 0 absent 1 present [WP 17S 62 WH 9 GEWB 153]State 1 occurs in all major arachnid groups except Pal-pigradi Pseudoscorpiones Opiliones Oribatida (Sar-coptiformes) and Prostigmata Within Acari State 1occurs in all major anactinotrichid lineages (Opilioac-ariformes Holothyrida Mesostigmata Ixodida) andsimilar structures are present in certain Astigmata(Evans 1992 Alberti amp Coons 1999)

179 Dorsomedian excretory organ 0 absent 1presentThis is a specialized excretoryosmoregulatory organformed by the postventriculus and proctodeum inProstigmata (Alberti amp Coons 1999)180 Adult coxal organ opening on or near coxa ofappendage III (= arachnid leg 1) 0 absent 1 present[S 64 WH 93 GEWB 90]State 1 occurs throughout Acari (Evans 1992 Albertiamp Coons 1999) Ricinulei (Pittard amp Mitchell 1972Legg 1976) Palpigradi (Millot 1942) AraneaeAmblypygi Thelyphonida and Schizomida (Buxton1913 1917) There is evidence that the second lsquoozo-porersquo of gonyleptids (Opiliones Laniatores) (Hara ampGnaspini 2003) is a persistent opening to the coxalorgan associated with leg 1 (Soslashrensen 1879) whichalso occurs in embryonic Phalangium (Opiliones Eup-noi) (Moritz 1959)181 Adult coxal organ opening on or near coxa ofappendage V (= arachnid leg 3) 0 absent 1 present [S63 WH 92 GEWB 89]State 1 occurs in extant Xiphosura (Yamasaki et al1988) Scorpiones mygalomorph (Buxton 1913 1917)and mesothele Araneae (Liphistius J Millot in Bris-towe 1932) State 1 occurs in basal Amblypygi(Charon Charinus Buxton 1913 1917) but degener-ates prior to the adult stage in higher groups(Phrynus Weygoldt 2000) Adequately preservedeurypterids show a small submarginal opening on thecoxa of appendage V (Baltoeurypterus Selden 1981also Eurypterus Hughmilleria Clarke amp Ruedemann1912) and this is interpreted here as the opening to acoxal organ Giribet et al (2002) coded this characteras uncertain for Eurypterida182 Adult coxal organ opening on coxa of appendageII (= arachnid palp) 0 absent 1 present [GEWB 65]Buxton (1913 1917) noted that the coxal organ ofSolifugae opens on the prolateral surface of the palpalcoxa The proximal end of the duct leading to the ori-fice is associated with glands and apparently theglandular secretion andor coxal fluid serve as saliva(Alberti 1979) (see also 14 15)183 Genital papillae 0 absent 1 present - inappli-cable coded only for those taxa with eversible lsquoappend-agesrsquo (111)The genital papillae are associated with the genitalopening in postlarval Acariformes (Evans 1992Alberti amp Coons 1999) These structures are probablyserially homologous with the prosomal Claparegravedeorgan (173) and both structures probably function inwater and ion regulation

DIGESTIVE SYSTEM

184 Ingestion 0 solid food 1 primarily liquid foodwith or without preoral digestion [WP sim16 WH sim8GEWB sim218]

264 J W SHULTZ

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

Extant xiphosurans ingest solid food and members ofmost extant arachnid orders ingest fluids Opilionesand Acari are the principal exceptions The phalangidOpiliones ingest solids but Shultz (2000 unpublobserv) has shown that the precerebral pharyngealapparatus of Cyphophthalmi is very similar to that ofthe fluid-feeding Scorpiones and differs substantiallyfrom phalangid Opiliones Examination of gut con-tents of the cyphophthalmids Siro acaroides S exilisand Chileogovea oedipus have failed to reveal solidparticles typically found in the guts of phalangids(Shultz unpubl observ) the character is coded asunknown for Cyphophthalmi Within Acari consump-tion of particles has been documented in Opilioacari-formes (Neocarus Van der Hammen 1989unidentified Walter amp Proctor 1998) Holothyridanon-parasitic Sarcoptiformes (Evans 1992 Walter ampProctor 1998 Alberti amp Coons 1999) and many endeo-stigmatids including the nematophagous Alycusroseus (Walter 1988)185 Mouth 0 directed posteroventrally 1 directedanteroventrally [S 9 WH 55 GEWB 28]State 1 occurs throughout Arachnida (Shultz 1990)186 Oral cavity dilated by muscles arising from coxaeand constricted by large circular sphincter 0 absent1 presentState 1 occurs in extant Xiphosura (Manton 1964Scholl 1977 Shultz 2001)187 Palate plate 0 absent 1 present [GEWB 159]State 1 is a unique synapomorphy of Araneae (Dunlop1994)188 Lateral walls of epistome broadly fused to medialwalls of palpal coxae opposite sides connected by well-developed transverse epistomal muscle 0 absent 1presentState 1 occurs throughout Opiliones and Scorpiones(Shultz 2000 unpubl observ)189 Epistome with a pair of lateral arms that projectsposteriorly into the prosoma on either side of the phar-ynx 0 absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)190 Epistome with four pairs of suspensor musclesattaching to the carapace 0 absent 1 presentState 1 occurs in extant Scorpiones (Lankester et al1885 Vyas 1970 Shultz unpubl observ) and appar-ently in Solifugae (Roewer 1934)191 Intercheliceral epipharyngeal sclerite 0 absent1 present [GEWB 31]State 1 occurs in Palpigradi (Boumlrner 1904) Trigo-notarbida (Palaeocharinus Dunlop 1994) Araneae(Firstman 1954 Marples 1983) Amblypygi (Mil-lot 1949b Shultz 1999) Schizomida (Millot1949e) Thelyphonida (Millot 1949e Shultz 1993)and Phalangida (Opiliones) (Shultz 2000 unpublobserv)

192 Epipharyngeal sclerite large projecting posteri-orly 0 absent 1 present - inapplicable due to absenceof sclerite (191) [GEWB 32]State 1 occurs in Amblypygi (Millot 1949b Shultz1999) Schizomida (Millot 1949e) Thelyphonida (Mil-lot 1949e Shultz 1993) some Araneae (Heptathela ndashcoded as Liphistius ndash and Hypochilus Marples 1983not Aphonopelma Firstman 1954) and perhaps Trig-onotarbida (Palaeocharinus Dunlop 1994)193 Dorsal dilator muscle of precerebral pharynxattaching to intercheliceral septum or associatedepipharyngeal sclerite (191) 0 absent 1 presentState 1 occurs in Palpigradi (Millot 1943) Araneae(Marples 1983) Amblypygi Schizomida and Thely-phonida (Millot 1949e Shultz 1993 1999)Extremely fine muscles have been documented insome Scorpiones (Shultz 2007) and in Leiobunum(Opiliones) (Shultz 2000) Lankester et al (1885)reported a muscle in Limulus that would appear tocorrespond to this muscle but the muscle does notexist (Manton 1964 Shultz 2001)194 Dorsal dilator muscle of precerebral pharynxattaching to dorsal surface of prosoma 0 absent 1present [GEWB 228]State 1 occurs in Palpigradi (Roewer 1934) and mostAraneae (Palmgren 1978 Aphonopelma Firstman1973) but not in Heptathela (coded as Liphistius) orHypochilus (Marples 1983)195 Lateral dilator muscle of precerebral pharynxattaching to endosternite 0 absent 1 present [GEWBsim226]State 1 is a unique feature of Araneae (Marples 1983)196 Lateral dilator muscle of precerebral pharynxattaching to lateral surface of epistomal processes 0absent 1 present [GEWB sim38]State 1 occurs in Scorpiones and Opiliones (Shultz2000)197 Lateral dilator muscle of precerebral pharynxattaching to medial process of coxa of appendage II(= arachnid palp) 0 absent 1 present [GEWB sim226]State 1 occurs in Amblypygi Thelyphonida and Schi-zomida (Boumlrner 1904 Shultz 1993 1999)198 Dilator muscle of precerebral pharynx andorpreoral cavity attaching to ventral surface of prosoma0 absent 1 present [GEWB 227]State 1 occurs in extant Xiphosura (Manton 1964Shultz 2001) Araneae (Marples 1983) Palpigradi(Millot 1943) and Solifugae (Roewer 1934)199 Postcerebral pharynx 0 absent 1 present [WPsim31 S sim5 WH sim21 GEWB sim154]The cuticle-lined foregut passes through the centralnervous system in many arthropods That portion pos-terior to the CNS is here termed the postcerebralpharynx regardless of the details of its morphology orfunctional specialization A substantial postcerebralpharynx is known in extant Xiphosura (Lankester

PHYLOGENY OF THE ARACHNID ORDERS 265

copy 2007 The Linnean Society of London Zoological Journal of the Linnean Society 2007 150 221ndash265

et al 1885 Manton 1964 Yamasaki et al 1988)Scorpiones (Centruroides Hadrurus HeterometrusShultz 2007 also Androctonus Abd el-Wahab 1952)Solifugae (Millot amp Vachon 1949) Araneae Ambly-pygi Thelyphonida and Schizomida (Millot 1949andasheShultz 1993) Weygoldt amp Paulus (1979) applied theterm lsquopostcerebral pharynxrsquo only to the lsquosucking stom-achrsquo of Araneae and Amblypygi but the sucking stom-ach is here regarded as a complex of three characters199ndash201

200 Dilator muscle of postcerebral pharynx attach-ing to endosternite 0 absent 1 present - inapplicabledue to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Lankester et al1885 Manton 1964 Shultz 2001) Scorpiones (Cen-truroides Hadrurus Heterometrus Shultz 2007 alsoAndroctonus Abd el-Wahab 1952) Araneae (First-

man 1954 Palmgren 1978) Amblypygi (Millot1949b Shultz 1999) and apparently the palpigradeProkoenenia wheeleri (Rucker 1901) but not inEukoenenia miriabilis (Millot 1943 contra Boumlrner1904)201 Dilator muscle of postcerebral pharynx attachingto dorsal surface of prosoma 0 absent 1 present -inapplicable due to absence of postcerebral pharynx(199)State 1 occurs throughout Araneae (Millot 1949cFirstman 1954 Palmgren 1978) and Amblypygi(Charinus Millot 1949b Phrynus Shultz 1999)202 Crop and gizzard 0 absent 1 present - inap-plicable due to absence of postcerebral pharynx (199)State 1 occurs in extant Xiphosura (Sekiguchi 1988)but not in extant Arachnida