Evolutionary and biogeographical history of an ancient …boyer/Site/Publications_files/Giribet et...

Evolutionary and biogeographical history of an ancientand global group of arachnids (Arachnida: Opiliones:Cyphophthalmi) with a new taxonomic arrangement

GONZALO GIRIBET1*, PRASHANT P. SHARMA1, LIGIA R. BENAVIDES2,SARAH L. BOYER3, RONALD M. CLOUSE1,4, BENJAMIN L. DE BIVORT1,5,DIMITAR DIMITROV6, GISELE Y. KAWAUCHI1, JEROME MURIENNE1,7 andPETER J. SCHWENDINGER8

1Museum of Comparative Zoology, Department of Organismic and Evolutionary Biology, HarvardUniversity, 26 Oxford Street, Cambridge, MA 02138, USA2Department of Biological Sciences, The George Washington University, 2023 G Street NW,Washington, DC 20052, USA3Biology Department, Macalester College, 1600 Grand Avenue, Saint Paul, MN 55105, USA4Division of Invertebrate Zoology, American Museum of Natural History, Central Park West at 79thStreet, New York, NY 10024, USA5Rowland Institute at Harvard, Harvard University, 100 Edwin Land Boulevard, Cambridge, MA02142, USA6Center for Macroecology, Evolution and Climate, Zoological Museum, University of Copenhagen,Department of Entomology, Universitetsparken 15, DK-2100, Copenhagen, Denmark7UMR 7205 CNRS, Département Systématique et Évolution, CP 50, Muséum national d’Histoirenaturelle, 45, rue Buffon, 75005 Paris, France8Muséum d’histoire naturelle, Département des Arthropodes et d’Entomologie I, 1, Route deMalagnou, Case Postale 6434, CH-1211 Genève 6, Switzerland

Received 27 May 2011; revised 8 July 2011; accepted for publication 8 July 2011bij_1774 1..39

We investigate the phylogeny, biogeography, time of origin and diversification, ancestral area reconstruction andlarge-scale distributional patterns of an ancient group of arachnids, the harvestman suborder Cyphophthalmi.Analysis of molecular and morphological data allow us to propose a new classification system for the group;Pettalidae constitutes the infraorder Scopulophthalmi new clade, sister group to all other families, which aredivided into the infraorders Sternophthalmi new clade and Boreophthalmi new clade. Sternophthalmi includesthe families Troglosironidae, Ogoveidae, and Neogoveidae; Boreophthalmi includes Stylocellidae and Sironidae,the latter family of questionable monophyly. The internal resolution of each family is discussed and traced backto its geological time origin, as well as to its original landmass, using methods for estimating divergence timesand ancestral area reconstruction. The origin of Cyphophthalmi can be traced back to the Carboniferous, whereasthe diversification time of most families ranges between the Carboniferous and the Jurassic, with the exceptionof Troglosironidae, whose current diversity originates in the Cretaceous/Tertiary. Ancestral area reconstruction isambiguous in most cases. Sternophthalmi is traced back to an ancestral land mass that contained New Caledoniaand West Africa in the Permian, whereas the ancestral landmass for Neogoveidae included the south-easternUSA and West Africa, dating back to the Triassic. For Pettalidae, most results include South Africa, or acombination of South Africa with the Australian plate of New Zealand or Sri Lanka, as the most likely ancestrallandmass, back in the Jurassic. Stylocellidae is reconstructed to the Thai-Malay Penisula during the Jurassic.Combination of the molecular and morphological data results in a hypothesis for all the cyphophthalmid genera,although the limited data available for some taxa represented only in the morphological partition negativelyaffects the phylogenetic reconstruction by decreasing nodal support in most clades. However, it resolves

*Corresponding author. E-mail: [email protected]

Biological Journal of the Linnean Society, 2011, ••, ••–••. With 10 figures

© 2011 The Linnean Society of London, Biological Journal of the Linnean Society, 2011, ••, ••–•• 1

the position of many monotypic genera not available for molecular analysis, such as Iberosiro, Odontosiro,Speleosiro, Managotria or Marwe, although it does not place Shearogovea or Ankaratra within any existingfamily. The biogeographical data show a strong correlation between relatedness and formerly adjacentlandmasses, and oceanic dispersal does not need to be postulated to explain disjunct distributions, especiallywhen considering the time of divergence. The data also allow testing of the hypotheses of the supposed totalsubmersion of New Zealand and New Caledonia, clearly falsifying submersion of the former, although the datacannot reject the latter. © 2011 The Linnean Society of London, Biological Journal of the Linnean Society,2011, ••, ••–••.

ADDITIONAL KEYWORDS: biogeography – distribution modelling – Gondwana – Laurasia – MAXENT –New Caledonia – New Zealand – Pangea.

INTRODUCTION

The harvestman suborder Cyphophthalmi (Fig. 1)constitutes an ancient lineage of arachnids and wasprobably one of the earliest inhabitants of terrestrialecosystems. Currently distributed on all continentallandmasses (with the exception of Antarctica) and onmost large islands of continental origin, the group isconsidered to have been in close association to theselandmasses since its origins (Juberthie & Massoud,1976; Boyer et al., 2007b). The fact that deep geneticdivergences in cytochrome c oxidase subunit I (COI)have been reported within one species (Boyer, Baker& Giribet, 2007a), and are suspected for many others(R. Clouse & P. Sharma, unpubl. data), corroboratesthe observations that individuals may live a long time(Juberthie, 1960b) and do not disperse far during thecourse of life history. These, together with the oldhistory of the group [a Burmese amber specimenprobably belonging to Stylocellidae is known from theEarly Cretaceous (Poinar, 2008) and the origins of thegroup has been estimated to have taken place duringthe Devonian or Carboniferous using moleculardating techniques (Giribet et al., 2010)] have resultedin a broad use of Cyphophthalmi for biogeographicalinferences and zoogeographical discussions (Rambla,1974; Juberthie & Massoud, 1976; Boyer, Karaman &

Giribet, 2005; Boyer & Giribet, 2007; Clouse &Giribet, 2007; Giribet & Kury, 2007; Boyer et al.,2007b; Boyer & Giribet, 2009; Clouse, de Bivort &Giribet, 2009; Karaman, 2009; Murienne & Giribet,2009; Sharma & Giribet, 2009a; Clouse & Giribet,2010; de Bivort & Giribet, 2010; Murienne, Karaman& Giribet, 2010b; Clouse et al., 2011). These includesome recent and more general debates on the totalsubmersion of large fragment islands such as NewCaledonia and New Zealand (Sharma & Giribet,2009a; Giribet & Boyer, 2010; Giribet et al., 2010).

However, to use a system for biogeographical infer-ences, a sound systematic hypothesis of the group isrequired. The taxonomy of Cyphophthalmi has ben-efitted from the contributions of many studies, espe-cially the synthetic work of Hansen & Sørensen(1904), who produced the first and still best mono-graph on the group, and established the first classi-fication system of the suborder Cyphophthalmi withone family, Sironidae, and two subfamilies, Stylocel-lini (including the genera Stylocellus Westwood, 1874,Ogovia Hansen & Sørensen, 1904, which was pre-occupied and became Ogovea Roewer, 1923, and Miop-salis Thorell, 1890) and Sironini (including PettalusThorell, 1876, Purcellia Hansen & Sørensen, 1904,Siro Latreille, 1796, and Parasiro Hansen &Sørensen, 1904). Another major contributor was

�Figure 1. Habitus. A, Karripurcellia harveyi (Pettalidae) from Warren National Park, Western Australia, July 2004. B,Pettalus thwaitesi (Pettalidae) from Peradeniya Botanical Gardens, Central Province, Sri Lanka, October 2007. C, Rakaiapauli (Pettalidae) from Kelcey’s bush, near Waimate, North Island, New Zealand, February 2008. D, Aoraki longitarsa(Pettalidae) from Governor’s bush, Mt Cook, South Island, New Zealand, January 2006. E, male Pettalus thwaitesi(Pettalidae) from Peradeniya Botanical Gardens, Central Province, Sri Lanka, June 2004. F, Ogovea cameroonensis(Ogoveidae) from Ototomo Forest, Central province, Cameroon, June 2009. G, Parogovia sp. (Neogoveidae) from Mt.Koupé, South-West Province, Cameroon, June 2009. H, two species of Parogovia from Campo Reserve, Littoral Province,Cameroon, June 2009; upper left, adult specimen of Parogovia n. sp.; lower right, juvenile specimen of Parogovia cf.sironoides. I, juvenile specimen of Paramiopsalis ramulosus (Sironidae) from P.N. Peneda Gerés, Portugal, May 2008. J,Paramiopsalis ramulosus (Sironidae) from P.N. Peneda Gerés, Portugal, May 2008. K, Parasiro minor (Sironidae) fromMonte Rasu, Sardinia, Italy, March 2008. L, Suzukielus sauteri (Sironidae) from Mt. Takao, Tokyo Prefecture, Honshu,Japan, April 2005. M, juvenile specimen of Leptopsalis sp. (Stylocellidae) from Bantimurung-Bulusaraung N.P., SulawesiSelatan, Indonesia, June 2006. N, female Leptopsalis sp. (Stylocellidae) from Bantimurung-Bulusaraung N.P., SulawesiSelatan, Indonesia, June 2006.

2 G. GIRIBET ET AL.

© 2011 The Linnean Society of London, Biological Journal of the Linnean Society, 2011, ••, ••–••

AA

C D E

F G

I J K

L M N

H

B

EVOLUTIONARY AND BIOGEOGRAPHICAL HISTORY OF ARACHNIDS 3


Juberthie, who described and monographed manygenera (e.g. Juberthie, 1956, 1958, 1960a, 1961, 1962,1969, 1970a, b; Juberthie & Muñoz-Cuevas, 1970;Juberthie, 1979) in addition to his contributions tothe biology of the group; the regional work of Forster(1948, 1952) in New Zealand, that of Lawrence (1931,1933, 1939, 1963) in South Africa, that of Rambla(Rambla & Fontarnau, 1984, 1986; Rambla, 1991,1994) in the Iberian Peninsula and southeast Asia, tomention just a few of them. More recently, Shear hascontributed with descriptions of numerous species inalmost all cyphophthalmid families (Shear, 1977,1979a, b, 1985, 1993a, b, c; Shear & Gruber, 1996). Healso proposed the bases of modern cyphophthalmidsystematics in a seminal first cladistic analysis of thegroup (Shear, 1980), with five families, three of whichwere new, and two infraorders, equivalent to Hansenand Sørensen’s subfamilies (Table 1). A sixth family,Troglosironidae, was also proposed a few years later(Shear, 1993b).

Two decades after Shear’s classification systemappeared, Giribet (2000) compiled all the cyphoph-thalmid literature to date, recognizing 113 species in26 genera. A subsequent analysis including represen-tatives of most genera and based on a numericalcladistic analysis of 32 morphological characters(Giribet & Boyer, 2002) recognized most of the fami-lies erected by Shear (1980) but also challenged someof his systematic propositions because the root ofthe tree, based on a limited molecular data set alsopublished in the same study, was placed betweenstylocellids and the rest (rendering Tropicophthalmiparaphyletic) or between pettalids and the rest (ren-dering Temperophthalmi paraphyletic). Shear (1993b)had also proposed the new family Troglosironidaeas sister to (Pettalidae + Sironidae) and this resultwas refuted by Giribet & Boyer (2002), who found itnested within an unresolved Neogoveidae. After someminor familial reassignments – Huitaca Shear, 1979was removed from Ogoveidae (Giribet & Prieto, 2003)

and subsequently included in Neogoveidae (Giribet,2007b); Fangensis Rambla, 1994 was transferred fromSironidae to Stylocellidae (Schwendinger & Giribet,2005); Metasiro was transferred from Sironidae toNeogoveidae (Giribet, 2007b); Meghalaya Giribet,Sharma & Bastawade, 2007 was included in Stylocel-lidae (Clouse et al., 2009); and Shearogovea mexasca(Shear, 1977) was excluded from Neogoveidae(Benavides & Giribet, 2007; Giribet, 2011) – the fami-lies are currently considered to be stable.

Recent phylogenetic analyses based on nucleotidesequence data have resolved the relationship amongsome of these families, providing strong support fora relationship of Troglosironidae and Neogoveidae(Boyer et al., 2007b; Boyer & Giribet, 2009; Sharma& Giribet, 2009a; Giribet et al., 2010), a result alsoobtained in a recent analysis of morphometric char-acters (de Bivort, Clouse & Giribet, 2010). Monophylyof Pettalidae is well supported both by discrete andcontinuous morphological characters (Giribet &Boyer, 2002; Giribet, 2003a; Boyer & Giribet, 2007; deBivort et al., 2010; de Bivort & Giribet, 2010), as wellas a diversity of molecular analyses (Boyer & Giribet,2007, 2009; Boyer et al., 2007b; Giribet et al., 2010).Stylocellidae is also well supported based on morphol-ogy (Giribet & Boyer, 2002; Clouse et al., 2009) andmolecules (Schwendinger & Giribet, 2005; Clouse &Giribet, 2007; Boyer et al., 2007b; Clouse et al., 2009;Clouse & Giribet, 2010; Giribet et al., 2010). However,monophyly of Sironidae, especially the membership ofthe Mediterranean genus Parasiro and the JapaneseSuzukielus Juberthie, 1970, remains controversial,both based on morphology (Giribet & Boyer, 2002; deBivort & Giribet, 2004; de Bivort et al., 2010), as wellas on molecular analyses (Boyer et al., 2005; Boyer &Giribet, 2007; Giribet et al., 2010).

In addition to the uncertainty about the monophylyof Sironidae, which we approach here by includingan expanded taxon sampling in problematic generapreviously represented by a single species (Parasiro),we include a much larger diversity of Neogoveidae,both from the Neotropics (29 species versus six usedin Boyer et al., 2007b; including data on the newBrazilian genus Canga DaSilva, Pinto-da-Rocha &Giribet, 2010) and from the Afrotropics (12 terminalsversus seven used in Boyer et al., 2007b). Most impor-tantly, we include the first molecular data on thefamily Ogoveidae, from specimens collected in Cam-eroon in 2009. In total, we provide novel sequencedata for 34 species (of a total of 162 molecular ter-minals), include 27 genera and a family previouslyunsampled, and include new landmasses (Mindanao,the eastern Neotropics, the westernmost distributionof the Afrotropics) not considered in previous phylo-genetic analyses. The present study also providesthe first total evidence analysis of molecules and

Table 1. Classification system of Shear (1980, 1993)

Suborder CyphophthalmiInfraorder Tropicophthalmi Shear, 1980

Superfamily Stylocelloidea Hansen & Sørensen, 1904Family Stylocellidae Hansen & Sørensen, 1904

Superfamily Ogoveoidea Shear, 1980Family Ogoveidae Shear, 1980Family Neogoveidae Shear, 1980

Infraorder Temperophthalmi Shear, 1980Superfamily Sironoidea Simon, 1879

Family Sironidae Simon, 1879Family Pettalidae Shear, 1980Family Troglosironidae Shear, 1993

4 G. GIRIBET ET AL.


morphology for the whole suborder Cyphophthalmiand new data on the timing of diversification andcladogenesis of the group, aiming to revisit interest-ing biogeographical topics. Finally, we provide anestimate of the ancestral area for each lineage andpresent the first habitat suitability and distributionalpatterns analysis for this dispersal-limited, yetglobally-distributed group of arthropods. Studyingmacroecological patterns in Cyphophthalmi is compli-cated as a result of the scarce occurrence data formost species. Thus, species-level assessment of large-scale distributional patterns and their primary eco-logical and evolutionary drivers is difficult. Recently,theoretical and practical arguments for the utility ofmodelling distributional patterns or even ecologicalniche characteristics above the species level havebeen proposed (Heino & Soininen, 2007; Hadly,Spaeth & Li, 2009; Diniz, De Marco & Hawkins,2010). Despite some obvious limitations (Diniz et al.,2010), this approach may be very useful to evaluatepatterns in groups with limited distributional datasuch as insects and other arthropods includingCyphophthalmi.

MATERIAL AND METHODSSPECIMENS

Most specimens used in the molecular part ofthis study (Fig. 2) were collected by one or moreof the authors through direct sifting of leaf litterand transferred to approximately 95% ethanol for

molecular and morphological study. Museum speci-mens have also been used for the morphologicalstudies. A detailed discussion of the specimen col-lecting effort is provided in the Supporting informa-tion (Appendix S1).

The study includes the first molecular data for thefamily Ogoveidae (Ogovea cameroonensis Giribet &Prieto, 2003) and includes additional sampling withinall other families, building upon previous studies onthe phylogenies of Pettalidae (Boyer & Giribet, 2007;Boyer et al., 2007b). Stylocellidae (Clouse et al., 2009;Clouse & Giribet, 2010; Clouse et al., 2011), Troglo-sironidae (Sharma & Giribet, 2009a), and Sironidae(Boyer et al., 2005; Giribet & Shear, 2010; Murienneet al., 2010b). However, data for Neogoveidae wererestricted to a single previous study (Boyer et al.,2007b), and the family was poorly sampled. For theAfrican diversity, we are now able to add data onanother described species, Parogovia gabonica (Juber-thie, 1969), from near its type locality (IpassaReserve, Makokou, Gabon). We also add new data ontwo new species from Mount Koupé and the Camporeserve, in Cameroon, and an additional specimen ofParogovia cf. sironoides also from the Campo reservein Cameroon. The third African species, Parogoviapabsgarnoni Legg, 1990, is known only from its typelocality in Sierra Leone, resulting in a large biogeo-graphical gap from the known distribution of speciesin the Gulf of Guinea, and differs morphologicallyfrom the other species in the genus in many keycharacters, showing very different spermatopositor

Figure 2. Distribution map of the sampled specimens for the molecular study (Table 2). Pettalidae are represented inred, Troglosironidae in purple, Ogoveidae in cyan, Neogoveidae in green, Sironidae in orange, and Stylocellidae in navyblue.



from the species in the Gulf of Guinea (Legg & Pabs-Garnon, 1989; Legg, 1990). A possible close relative tothis species is included here, based on data from afemale collected in Ivory Coast. The South Americansampling has also been enriched considerably. Themonotypic genus Huitaca is now represented by datafrom seven species all from Colombia, including thenominal Huitaca ventralis Shear, 1979. Metagovea isnow represented by 13 specimens in 12 putative newspecies, including one from Guyana. These includethe Colombian species identified as Neogovea in ourprevious studies, although they appear to be relatedto Metagovea. A large biogeographical gap in ourprevious studies was the easternmost distribution ofthe genus Neogovea (type species Neogovea immsiHinton, 1938, from Amapá, Brazil). We now includeputative representatives of this clade based on speci-mens collected in French Guiana and Guyana, includ-ing specimens from the recently described N. virginieJocqué & Jocqué, 2011. With the exception of onespecimen from Guyana clustering with Metagovea,these specimens group with the unidentified juvenilefrom Venezuela used in Boyer et al. (2007b) and witha species from the ‘Tepuis’ in Colombia, which wereassign here to the genus Brasilogovea Martens,1969, previously considered a synonym of Neogovea(Shear, 1980; Giribet, 2000). Finally, we were able toinclude data on the monotypic genus Canga based ona specimen from its type locality (DaSilva et al.,2010). However, large areas of the Neotropics withknown specimens of Neogoveidae remain unsampledin our molecular phylogeny, and they should beincluded in future studies (Benavides & Giribet, 2007:fig. 1).

MOLECULAR DATA

DNA extraction, amplification, and sequencing wereperformed as described in several of our previousstudies on molecular systematics of Cyphophthalmiusing the same markers (Schwendinger & Giribet,2005; Boyer et al., 2007b; Boyer & Giribet, 2009;Sharma & Giribet, 2009a; Clouse & Giribet, 2010).We used the five markers as in these previous studies,including the nuclear ribosomal 18S and 28S rRNA,the nuclear protein-encoding histone H3, the mito-chondrial ribosomal 16S rRNA, and the mitochondrialprotein-encoding COI genes. For outgroups, we usedseven noncyphophthalmid Opiliones from the subor-ders Eupnoi, Dyspnoi, and Laniatores (Table 2). Pub-lished sequence data from these studies and the noveldata presented here are deposited in GenBank andare shown in Table 2.

All sequence files for each gene were prepared withMacGDE (Linton, 2005). 18S rRNA sequence datawere divided into six fragments and it was available

for 170 terminals. The 28S rRNA fragment wasdivided into ten regions and was available for 169terminals. 16S rRNA was divided into eight frag-ments and was available for 127 terminals. All theribosomal genes were submitted to direct optimiza-tion or to multiple sequence alignment for homologyassignment. The 143 COI sequences, unlike in manyother organisms, show clade-specific considerablesequence length variation, and hence the gene wasdivided into seven fragments and analyzed underdynamic homology (Wheeler, 2005) or submitted tomultiple sequence alignment. The histone H3 datawere available for 108 terminals and were treated asprealigned in all analyses because they show nolength variation.

MORPHOLOGICAL DATA MATRIX

A morphological matrix of 62 characters was compiledfor 161 taxa based in part on our previous studies(Giribet & Boyer, 2002; Giribet, 2003a; de Bivort &Giribet, 2004; Boyer & Giribet, 2007; de Bivort &Giribet, 2010), direct observation of specimens,mostly through scanning electron microscopy, andcomplemented by some new literature sources(Karaman, 2009). All 19 multistate characters wereunordered. Spermatogenesis in Cyphophthalmi is apromising source of phylogenetic characters, asrecently outlined by Alberti, Giribet & Gutjahr (2009;see also Juberthie & Manier, 1976; Juberthie, Manier& Boissin, 1976; Juberthie & Manier, 1978; Alberti,1995, 2005), although taxon sampling is still sparseand these characters were not considered in this dataset (G. Alberti & G. Giribet, unpubl. data). We did nothave access to specimens of a few species that wereincluded in the data matrix based entirely on litera-ture sources. These have missing data for severalcharacters, especially those observed through scan-ning electron microscopy, such as the prosomalsternal characters. These species include Ankaratrafranzi Shear & Gruber, 1996, Manangotria taolanaroShear & Gruber, 1996, and Odontosiro lusitanicusJuberthie, 1961. Similarly, several CyphophthalmusJoseph, 1868 species, included in our molecularmatrix, were not scored for several morphologicalcharacters because males were never available forexamination, and their published descriptions do notinclude scanning electron micrographs of the relevantcharacters and their descriptions and illustrations areinadequate for scoring those features. Finally, a fewspecies scored in the matrix are not known for onegender, and therefore the corresponding scorings aremissing. The total number of missing cells was thus1798 (17% of cells). In the present study, we were notable to use morphometrics, as we have done in pre-vious studies (Clouse, 2010; de Bivort et al., 2010; de

6 G. GIRIBET ET AL.


Tab

le2.

Spe

cim

enan

dco

llec

tion

data

wit

hG

enB

ank

acce

ssio

nn

um

bers

Vou

cher

Loc

alit

yC

oord

inat

es18

SrR

NA

28S

rRN

A16

SrR

NA

CO

IH

3

FA

MIL

YP

ET

TA

LID

AE

Aor

aki

calc

arob

tusa

wes

tlan

dic

aD

NA

1011

29N

ewZ

eala

nd

-41.

7838

7,17

2.36

903

EU

6736

26D

Q51

8038

DQ

5180

70E

U67

3667

EU

6737

03A

orak

icr

ypta

DN

A10

1289

New

Zea

lan

d-3

7.53

404,

175.

7414

0D

Q51

8000

DQ

5180

43D

Q51

8068

DQ

5181

20D

Q51

8156

Aor

aki

den

ticu

lata

den

ticu

lata

DN

A10

0955

New

Zea

lan

d-4

2.09

283,

171.

3409

6E

U67

3618

EU

6736

54E

U67

3584

DQ

9923

09E

U67

3698

Aor

aki

den

ticu

lata

maj

orD

NA

1009

59N

ewZ

eala

nd

-43.

0340

8,17

1.76

464

EU

6736

20E

U67

3656

EU

6735

85D

Q99

2203

EU

6737

00A

orak

igr

anu

losa

DN

A10

1841

New

Zea

lan

d-3

9.93

478,

175.

6404

6D

Q51

7999

DQ

5180

39D

Q51

8071

––

Aor

aki

hea

lyi

DN

A10

0940

New

Zea

lan

d-4

1.08

673,

174.

1382

6D

Q51

8002

DQ

5180

42D

Q51

8067

DQ

5181

22D

Q51

8160

Aor

aki

iner

ma

DN

A10

0966

New

Zea

lan

d-3

8.79

686,

177.

1247

2E

U67

3622

EU

6736

58–

–E

U67

3702

Aor

aki

lon

gita

rsa

DN

A10

1806

New

Zea

lan

d-4

3.73

666,

170.

0922

2E

U67

3613

EU

6736

52–

DQ

9923

13E

U67

3695

Aor

aki

cf.

tum

idat

aD

OC

094

New

Zea

lan

d-3

9.66

7,17

5.63

7*E

U67

3614

––

DQ

9923

18–

Aor

aki

sp.

DN

A10

1126

New

Zea

lan

d-4

1.08

708,

174.

1370

9E

U67

3624

EU

6736

59–

DQ

9923

19–

Au

stro

purc

elli

aar

ctic

osa

DN

A10

0951

Qu

een

slan

d,A

ust

rali

a-1

6.16

610,

145.

4156

0D

Q51

7984

DQ

5180

23–

DQ

5181

11D

Q51

8147

Au

stro

purc

elli

ad

avie

sae

DN

A10

0947

Qu

een

slan

d,A

ust

rali

a-1

7.24

560,

145.

6420

7D

Q51

7985

DQ

5180

24–

DQ

5181

12D

Q51

8148

Au

stro

purc

elli

afo

rste

riD

NA

1009

45Q

uee

nsl

and,

Au

stra

lia

-16.

0615

1,14

5.46

217

DQ

5179

83D

Q51

8022

DQ

5180

64D

Q51

8110

DQ

5181

46A

ust

ropu

rcel

lia

scop

aria

DN

A10

0946

Qu

een

slan

d,A

ust

rali

a-1

6.59

458,

145.

2792

7D

Q51

7982

DQ

5180

21D

Q51

8065

DQ

5181

08E

U67

3678

Ch

ileo

gove

aoe

dip

us

DN

A10

0413

Ch

ile

-41.

5083

3,-7

2.61

666

DQ

1337

21D

Q13

3733

DQ

5180

55D

Q13

3745

–C

hil

eogo

vea

sp.

DN

A10

0490

Ch

ile

-39.

6666

6,-7

3.28

333*

DQ

1337

22D

Q13

3734

DQ

5180

54D

Q13

3746

EU

6736

72K

arri

purc

elli

ah

arve

yiD

NA

1013

03W

este

rnA

ust

rali

a-3

4.49

500,

115.

9752

7D

Q51

7980

DQ

5180

19D

Q51

8062

DQ

5181

06D

Q51

8143

Neo

purc

elli

asa

lmon

iD

NA

1009

39N

ewZ

eala

nd

-44.

1078

0,16

9.35

527

DQ

5179

98E

U67

3650

DQ

5180

66D

Q82

5638

EU

6736

94P

arap

urc

elli

am

onti

cola

DN

A10

0386

Sou

thA

fric

a-2

9.05

347,

29.3

8516

DQ

5189

73D

Q51

8009

–D

Q51

8098

DQ

5181

35P

arap

urc

elli

asi

lvic

ola

DN

A10

0385

Sou

thA

fric

a-2

8.74

421,

31.1

3763

AY

6394

94D

Q51

8008

DQ

5180

53A

Y63

9582

DQ

5181

36P

etta

lus

thw

aite

siD

NA

1012

23S

riL

anka

7.27

251,

80.5

9333

EU

6735

92E

U67

3633

EU

6735

69E

U67

3666

EU

6736

77P

etta

lus

sp.

DN

A10

1282

Sri

Lan

ka7.

3848

3,80

.816

96D

Q82

5537

EU

6736

32–

DQ

8256

36E

U67

3676

Pet

talu

ssp

.D

NA

1012

83S

riL

anka

7.38

483,

80.8

1696

DQ

5179

74D

Q51

8016

DQ

5180

56D

Q51

8100

DQ

5181

37P

etta

lus

sp.

DN

A10

1285

Sri

Lan

ka6.

9251

7,80

.819

49D

Q51

7976

DQ

5180

17D

Q51

8058

DQ

5181

02D

Q51

8139

Pet

talu

ssp

.D

NA

1012

86S

riL

anka

6.92

517,

80.8

1949

DQ

5179

77D

Q51

8013

DQ

5180

59D

Q51

8103

DQ

5181

40P

etta

lus

sp.

DN

A10

1287

Sri

Lan

ka6.

8235

9,80

.849

96D

Q51

7978

DQ

5180

14D

Q51

8060

DQ

5181

04D

Q51

8141

Pet

talu

ssp

.D

NA

1012

88S

riL

anka

6.55

551,

80.3

7030

DQ

5179

79D

Q51

8015

DQ

5180

61D

Q51

8105

DQ

5181

42P

urc

elli

ail

lust

ran

sD

NA

1003

87S

outh

Afr

ica

-33.

9829

4,18

.424

64E

U67

3589

EU

6736

29D

Q51

8052

EU

6736

65E

U67

3673

Rak

aia

anti

pod

ian

aD

NA

1009

57N

ewZ

eala

nd

-43.

2514

5,17

1.36

761

DQ

5179

88D

Q51

8031

DQ

5180

72D

Q51

8115

DQ

5181

51R

akai

ad

orot

hea

DN

A10

0943

New

Zea

lan

d-4

1.28

163,

174.

9096

1D

Q51

7990

DQ

5180

33D

Q51

8077

DQ

9923

31–

Rak

aia

flor

ensi

sD

NA

1012

95N

ewZ

eala

nd

-40.

8325

8,17

2.96

896

DQ

5179

86D

Q51

8025

DQ

5180

83D

Q51

8113

DQ

5181

49R

akai

ali

nd

sayi

DN

A10

1128

New

Zea

lan

d-4

6.89

327,

168.

1039

8D

Q51

7995

DQ

5180

27D

Q51

8081

DQ

5181

18D

Q51

8154

Rak

aia

mac

raD

NA

1018

08N

ewZ

eala

nd

-45.

9200

0,17

0.02

805

EU

6735

96E

U67

3636

EU

6735

71E

U67

3668

–R

akai

am

agn

aau

stra

lis

DN

A10

0962

New

Zea

lan

d-4

2.33

225,

172.

1712

6E

U67

3601

EU

6736

40E

U67

3575

DQ

9923

33E

U67

3684

Rak

aia

med

iaD

NA

1012

92N

ewZ

eala

nd

-39.

9347

8,17

5.64

046

DQ

5179

96D

Q51

8030

DQ

5180

74D

Q51

8125

DQ

5181

57R

akai

am

inu

tiss

ima

DN

A10

1291

New

Zea

lan

d-3

9.41

641,

175.

2185

8D

Q51

7987

DQ

5180

26D

Q51

8082

DQ

5181

14D

Q51

8150

Rak

aia

pau

liD

NA

1009

68N

ewZ

eala

nd

-44.

7007

3,17

0.96

557

DQ

5179

92D

Q51

8032

DQ

5180

73E

U67

3670

DQ

5181

61R

akai

aso

lita

ria

DN

A10

1294

New

Zea

lan

d-4

1.46

852,

175.

4488

5D

Q51

7997

DQ

5180

29D

Q51

8075

DQ

5181

19D

Q51

8155

Rak

aia

sore

nse

ni

sore

nse

ni

DN

A10

0969

New

Zea

lan

d-4

6.10

967,

167.

6903

4D

Q51

7993

DQ

5180

36D

Q51

8079

DQ

5181

16D

Q51

8153

Rak

aia

sore

nse

ni

dig

itat

aD

NA

1009

70N

ewZ

eala

nd

-46.

5817

7,16

9.20

901

DQ

5179

89D

Q51

8035

DQ

5180

78D

Q51

8123

DQ

5181

62R

akai

ast

ewar

tien

sis

DN

A10

0944

New

Zea

lan

d-4

6.89

327,

168.

1039

8D

Q51

7994

DQ

5180

28D

Q51

8080

DQ

5181

17–

Rak

aia

un

iloc

aD

NA

1018

12N

ewZ

eala

nd

-41.

2208

3,17

3.43

944

EU

6735

99E

U67

3638

–E

U67

3671

–R

akai

asp

.D

NA

1012

97N

ewZ

eala

nd

-40.

9759

5,17

5.11

747

EU

6736

08E

U67

3647

EU

6735

79D

Q99

2344

EU

6736

91R

akai

asp

.D

NA

1018

07N

ewZ

eala

nd

-45.

9016

6,16

9.46

277

EU

6736

06E

U67

3645

––

EU

6736

89R

akai

asp

.D

NA

1009

58N

ewZ

eala

nd

-43.

8086

9,17

3.02

144

EU

6735

97E

U67

3637

EU

6735

73D

Q99

2349

EU

6736

81R

akai

asp

.D

NA

1012

93N

ewZ

eala

nd

-40.

8518

1,17

4.93

233

EU

6736

10E

U67

3649

EU

6735

81D

Q99

2322

EU

6736

93R

akai

asp

.D

NA

1009

54N

ewZ

eala

nd

-41.

1575

7,17

5.02

168

EU

6736

03E

U67

3642

EU

6735

76D

Q99

2348

EU

6736

86



Tab

le2.

Con

tin

ued

Vou

cher

Loc

alit

yC

oord

inat

es18

SrR

NA

28S

rRN

A16

SrR

NA

CO

IH

3

FA

MIL

YS

IRO

NID

AE

Cyp

hop

hth

alm

us

corf

uan

us

DN

A10

2111

Gre

ece

39.6

1056

,20

.339

44F

J946

390

FJ9

4641

5F

J946

364

FJ9

4643

8–

Cyp

hop

hth

alm

us

du

rico

riu

sD

NA

1004

87S

love

nia

46.0

1667

,14

.666

67A

Y63

9461

DQ

5131

20A

Y63

9526

AY

6395

56–

Cyp

hop

hth

alm

us

ere

DN

A10

0499

Ser

bia

43.8

3333

,20

.05

AY

6394

62D

Q82

5593

AY

6395

27A

Y63

9557

AY

6394

44C

yph

oph

thal

mu

scf

.gj

orgj

evic

iD

NA

1004

98M

aced

onia

41.9

6667

,21

.35

AY

6394

64D

Q82

5587

AY

6395

29A

Y63

9559

–C

yph

oph

thal

mu

sgo

rdan

iD

NA

1004

95M

onte

neg

ro42

.45,

19.2

6667

AY

6394

67D

Q82

5592

AY

6395

32–

AY

6394

46C

yph

oph

thal

mu

sh

lava

ciD

NA

1020

99C

roat

ia43

.413

44,

16.9

1244

FJ9

4638

4F

J946

409

FJ9

4635

8F

J946

433

–C

yph

oph

thal

mu

sm

arko

iD

NA

1004

97M

aced

onia

41.4

1667

,22

.266

67A

Y63

9469

AY

6395

04A

Y63

9534

AY

6395

61A

Y63

9447

Cyp

hop

hth

alm

us

mar

ten

siD

NA

1004

94M

onte

neg

ro42

.4,

18.7

6667

AY

6394

71D

Q82

5589

AY

6395

36A

Y63

9563

AY

6394

49C

yph

oph

thal

mu

sm

inu

tus

DN

A10

0493

Mon

ten

egro

42.6

5,18

.666

67A

Y63

9473

DQ

8255

91A

Y63

9537

AY

6395

65A

Y63

9450

Cyp

hop

hth

alm

us

ogn

jen

ovic

iD

NA

1010

39B

osn

ia&

Her

zego

vin

a43

.016

67,

18.5

1667

AY

6394

75D

Q82

5594

–A

Y63

9567

AY

6394

51

Cyp

hop

hth

alm

us

rum

ijae

DN

A10

0492

Mon

ten

egro

42.1

6667

,19

.333

33A

Y63

9477

DQ

8255

88A

Y63

9539

AY

6395

69A

Y63

9453

Cyp

hop

hth

alm

us

serb

icu

sD

NA

1020

98S

erbi

a43

.279

17,

22.0

6389

FJ9

4638

3F

J946

408

FJ9

4635

7F

J946

432

–C

yph

oph

thal

mu

ste

yrov

skyi

DN

A10

0910

Mon

ten

egro

42.2

3333

,19

.066

67A

Y63

9482

DQ

5131

18A

Y63

9544

AY

6395

71A

Y63

9454

Cyp

hop

hth

alm

us

treb

inja

nu

sD

NA

1010

38B

osn

ia&

Her

zego

vin

a42

.233

33,

19.1

6667

AY

6394

83D

Q51

3119

–A

Y63

9572

–

Cyp

hop

hth

alm

us

zeta

eD

NA

1009

07M

onte

neg

ro42

.933

33,

18.5

AY

6394

85A

Y63

9515

AY

6395

46A

Y63

9574

AY

6394

56P

aram

iops

alis

edu

ard

oiD

NA

1018

78S

pain

43.4

1718

,-8

.063

56E

U63

8284

EU

6382

87E

U63

8281

EU

6382

88JF

7864

15P

aram

iops

alis

ram

ulo

sus

DN

A10

0459

Spa

in42

.315

80,

-8.4

8697

AY

6394

89D

Q51

3121

AY

6395

50D

Q82

5641

–P

aram

iops

alis

ram

ulo

sus

DN

A10

3538

Por

tuga

l41

.569

44,

-8.1

4027

JF93

4956

JF93

4990

JF93

5023

JF78

6389

–P

aram

iops

alis

sp.

DN

A10

4624

Spa

in43

.319

68,

-6.8

7282

JF93

4957

JF93

4991

JF93

5024

JF78

6390

JF78

6416

Par

asir

oco

iffa

iti

DN

A10

1383

Spa

in42

.152

51,

1.93

039

AY

9188

72D

Q51

3122

AY

9188

77D

Q82

5642

AY

9188

82P

aras

iro

min

orD

NA

1035

35S

ardi

nia

,It

aly

40.4

3047

,9.

0094

8JF

9349

58JF

9349

92JF

9350

25JF

7863

91–

Sir

oac

aroi

des

DN

A10

0488

Ore

gon

,U

SA

44.5

833,

-123

.516

6*A

Y63

9490

DQ

5131

28A

Y63

9551

DQ

8256

43–

Sir

obo

yera

eD

NA

1016

14W

ash

ingt

on,

US

A46

.992

21,

-121

.846

41D

Q51

3139

DQ

5131

25–

DQ

5131

12–

Sir

oca

lave

ras

DN

A10

1623

Cal

ifor

nia

,U

SA

38.2

7744

,-1

20.3

0543

DQ

5131

46D

Q51

3133

*–

––

Sir

oex

ilis

DN

A10

0489

Mar

ylan

d,U

SA

39.4

833,

-79.

4333

AY

6394

91D

Q82

5585

–A

Y63

9579

–S

iro

cf.

kam

iake

nsi

sD

NA

1016

11Id

aho,

US

A47

.746

46,

-116

.702

07D

Q51

3147

DQ

5131

34*

–D

Q51

3115

–S

iro

kam

iake

nsi

sD

NA

1016

13Id

aho,

US

A46

.867

77,

-117

.157

77JF

9349

59JF

9349

93–

––

Sir

oru

ben

sD

NA

1004

57F

ran

ce44

.083

38,

3.58

140

AY

4288

18D

Q82

5584

–D

Q51

3111

–S

iro

shas

taD

NA

1016

22C

alif

orn

ia,

US

A41

.063

67,

-122

.360

45D

Q51

3149

DQ

5131

36*

––

–S

iro

vall

eoru

mD

NA

1004

61It

aly

45.9

833,

9.86

66*

AY

6394

92D

Q51

3123

AY

6395

52A

Y63

9580

AY

6394

57.1

Su

zuki

elu

ssa

ute

riD

NA

1015

43Ja

pan

35.6

3440

,13

9.24

122

DQ

5131

38D

Q51

3116

DQ

5180

86D

Q51

3108

DQ

5181

66S

uzu

kiel

us

sau

teri

DN

A10

1550

Japa

n34

.833

33,

138.

9316

6D

Q82

5541

DQ

8255

83D

Q82

5615

DQ

8256

40D

Q82

5520

FA

MIL

YO

GO

VE

IDA

EO

gove

aca

mer

oon

ensi

sD

NA

1046

17C

amer

oon

3.64

621,

11.2

9079

JF93

4960

JF93

4994

JF93

5026

JF78

6392

JF78

6417

FA

MIL

YT

RO

GL

OS

IRO

NID

AE

Trog

losi

roae

llen

iD

NA

1003

45N

ewC

aled

onia

-21.

1833

,16

5.30

59A

Y63

9497

DQ

8255

80A

Y63

9555

AY

6395

84D

Q51

8164

Trog

losi

roju

bert

hie

iD

NA

1003

44N

ewC

aled

onia

-22.

0500

,16

6.46

67D

Q82

5540

EU

8871

21E

U88

7077

EU

8870

47–

Trog

losi

rolo

ngi

foss

aD

NA

1008

67N

ewC

aled

onia

-22.

3527

,16

6.97

36D

Q51

8089

DQ

8255

82D

Q51

8084

DQ

8256

39D

Q51

8165

Trog

losi

rom

onte

ith

iD

NA

1015

80N

ewC

aled

onia

-21.

6000

,16

5.71

67E

U88

7101

EU

8871

16E

U88

7074

EU

8870

43–

Trog

losi

ron

inqu

aD

NA

1005

77N

ewC

aled

onia

-21.

7500

,16

6.15

00D

Q51

8088

DQ

8255

81D

Q51

8055

DQ

5181

28–

Trog

losi

rou

rban

us

DN

A10

1710

New

Cal

edon

ia-2

2.19

45,

166.

5017

EU

8871

02E

U88

7119

EU

8870

73E

U88

7040

JF78

6340

Trog

losi

row

ilso

ni

DN

A10

2324

New

Cal

edon

ia-2

2.17

70,

166.

5106

EU

8871

07E

U88

7125

EU

8870

75E

U88

7061

–

8 G. GIRIBET ET AL.


Tab

le2.

Con

tin

ued

FA

MIL

YS

TY

LO

CE

LL

IDA

EF

ange

nsi

sin

sula

nu

sD

NA

1003

88,

DN

A10

1063

Th

aila

nd

7.88

5,98

.436

94G

Q48

8337

DQ

8255

51–

GQ

4881

81–

Fan

gen

sis

spel

aeu

sD

NA

1006

69T

hai

lan

d14

.299

72,

98.9

8306

DQ

1337

12D

Q82

5554

GQ

4881

95A

Y63

9583

AY

6394

60L

epto

psal

isly

dek

keri

DN

A10

1064

New

Gu

inea

,In

don

esia

-2.7

1667

,13

4.5*

DQ

1337

17G

Q48

8439

–G

Q48

8153

–

Lep

tops

alis

nov

agu

inea

DN

A10

1510

New

Gu

inea

,In

don

esia

-0.8

333,

134.

0333

*G

Q48

8322

GQ

4884

51G

Q48

8230

––

Lep

tops

alis

sp.

DN

A10

1514

Bor

neo

,M

alay

sia

1.76

667,

110.

3166

7G

Q48

8317

GQ

4884

35D

Q82

5611

GQ

4881

78G

Q48

8114

Lep

tops

alis

sp.

DN

A10

1932

Java

,In

don

esia

-6.7

9833

,10

7.01

583

GQ

4882

64G

Q48

8382

GQ

4882

06G

Q48

8144

GQ

4881

21L

epto

psal

issp

.D

NA

1019

44,

DN

A10

1945

Java

,In

don

esia

-6.7

5694

,10

6.52

333

GQ

4882

67G

Q48

8385

GQ

4882

26G

Q48

8146

GQ

4881

23

Lep

tops

alis

sp.

DN

A10

1093

,D

NA

1004

96T

hai

lan

d6.

67,

101.

15*

GQ

4882

83G

Q48

8402

GQ

4882

21G

Q48

8137

–

Lep

tops

alis

sp.

DN

A10

1483

Mal

aysi

a4.

3969

4,10

2.43

056

DQ

8255

32G

Q48

8454

DQ

8256

10G

Q48

8182

GQ

4881

28L

epto

psal

issp

.D

NA

1014

89M

alay

sia

3.71

639,

101.

7386

1D

Q51

8095

GQ

4884

56D

Q51

8087

GQ

4881

84–

Lep

tops

alis

sp.

DN

A10

1930

Java

,In

don

esia

-6.7

4,10

7.01

278

GQ

4882

62G

Q48

8380

GQ

4882

04–

GQ

4881

19L

epto

psal

issp

.D

NA

1019

37S

ula

wes

i,In

don

esia

1.49

028,

125.

1527

8G

Q48

8298

GQ

4884

22G

Q48

8211

GQ

4881

67G

Q48

8132

Lep

tops

alis

sp.

DN

A10

1938

Su

law

esi,

Indo

nes

ia-5

.042

5,11

9.73

556

GQ

4882

99G

Q48

8359

GQ

4882

12G

Q48

8168

GQ

4881

33L

epto

psal

issp

.D

NA

1020

32S

um

atra

,In

don

esia

-0.1

0583

,10

0.66

389

GQ

4883

08G

Q48

8361

GQ

4881

88G

Q48

8171

GQ

4881

34L

epto

psal

issp

.D

NA

1020

33,

DN

A10

2048

Su

mat

ra,

Indo

nes

ia0.

3461

1,10

0.06

917

GQ

4883

07G

Q48

8428

–G

Q48

8170

–

Lep

tops

alis

sp.

DN

A10

2039

Su

mat

ra,

Indo

nes

ia-0

.945

83,

100.

5436

1G

Q48

8250

GQ

4884

33G

Q48

8190

GQ

4881

75–

Lep

tops

alis

sp.

DN

A10

2042

Su

mat

ra,

Indo

nes

ia3.

2211

1,98

.497

22G

Q48

8314

GQ

4884

34G

Q48

8213

GQ

4881

76G

Q48

8136

Lep

tops

alis

sp.

DN

A10

2061

Su

mat

ra,

Indo

nes

ia-0

.477

22,

100.

3538

9G

Q48

8312

GQ

4884

32–

GQ

4881

74G

Q48

8135

Lep

tops

alis

sp.

DN

A10

3250

Th

aila

nd

9.76

667,

98.4

1389

GQ

4882

78G

Q48

8394

GQ

4881

92G

Q48

8155

–M

egh

alay

asp

.D

NA

1010

94,

DN

A10

1500

Th

aila

nd

7.88

528,

98.4

3722

DQ

8255

34G

Q48

8352

–G

Q48

8158

GQ

4881

27

Meg

hal

aya

sp.

DN

A10

1494

,D

NA

1015

06,

DN

A10

1765

Th

aila

nd

9.91

806,

98.9

4278

DQ

8255

30G

Q48

8398

–D

Q82

5632

–

Meg

hal

aya

sp.

DN

A10

1767

Th

aila

nd

6.97

,10

0.10

*G

Q48

8276

GQ

4883

90–

GQ

4881

54–

Meg

hal

aya

sp.

DN

A10

2051

Indi

a25

.507

78,

90.2

3167

GQ

4882

61G

Q48

8379

––

GQ

4881

18M

egh

alay

asp

.D

NA

1032

42,

DN

A10

3243

,D

NA

1032

44

Ch

ina

27.6

8833

,98

.277

78G

Q48

8233

GQ

4883

77G

Q48

8219

–G

Q48

8117

Meg

hal

aya

sp.

DN

A10

3251

Th

aila

nd

9.76

667,

98.4

1389

GQ

4882

80G

Q48

8396

GQ

4881

96–

–M

egh

alay

asp

.D

NA

1032

65T

hai

lan

d14

.25,

101.

98*

GQ

4882

39G

Q48

8392

––

–M

iops

alis

sp.

DN

A10

1519

Bor

neo

,In

don

esia

0.64

,11

7.09

*D

Q82

5526

GQ

4884

36G

Q48

8228

GQ

4881

80G

Q48

8115

Mio

psal

issp

.D

NA

1032

59B

orn

eo,

Mal

aysi

a5.

81,

116.

24*

GQ

4882

60G

Q48

8375

GQ

4881

93G

Q48

8142

GQ

4881

16M

iops

alis

sp.

DN

A10

1468

,D

NA

1019

50B

orn

eo,

Indo

nes

ia0.

64,

117.

09*

GQ

4883

28G

Q48

8444

––

–

Mio

psal

issp

.D

NA

1015

17B

orn

eo,

Indo

nes

ia1.

0666

7,11

7.83

333*

DQ

8255

27.1

DQ

8255

64–

GQ

4881

79D

Q82

5508

.1M

iops

alis

sp.

DN

A10

2032

,D

NA

1020

53,

DN

A10

2058

Su

mat

ra,

Indo

nes

ia-0

.105

83,

100.

6638

9G

Q48

8305

GQ

4884

25–

––

Mio

psal

issp

.D

NA

1032

49B

orn

eo,

Mal

aysi

a6.

01,

116.

53*

GQ

4882

52G

Q48

8367

GQ

4881

94G

Q48

8139

–M

iops

alis

sp.

DN

A10

3254

Bor

neo

,M

alay

sia

1.75

833,

110.

3297

2G

Q48

8257

GQ

4883

71–

––

Mio

psal

issp

.D

NA

1049

81M

inda

nao

,P

hil

ippi

nes

6.48

,12

5.09

*H

Q59

3868

HQ

5938

69–

HQ

5938

70H

Q59

3871

FA

MIL

YN

EO

GO

VE

IDA

EB

rasi

logo

vea

sp.

DN

A10

1665

Col

ombi

a0.

1797

2,-7

2.62

333

JF93

4963

JF93

5011

JF93

5028

JF78

6414

JF78

6430

‘Bra

silo

gove

a’sp

.To

boga

nD

NA

1008

69V

enez

uel

a5.

6500

0,-6

7.63

333

DQ

8255

45D

Q82

5600

DQ

8256

17–

–



Tab

le2.

Con

tin

ued

Vou

cher

Loc

alit

yC

oord

inat

es18

SrR

NA

28S

rRN

A16

SrR

NA

CO

IH

3

Can

gare

nat

aeD

NA

1056

80B

razi

l-6

.410

55,

-50.

3231

9JF

9349

64JF

9349

97JF

9350

29JF

7863

95JF

7864

20H

uit

aca

ven

tral

isD

NA

1016

74C

olom

bia

7.41

667,

-72.

4333

3JF

9349

80JF

9350

14–

JF78

6399

–H

uit

aca

sp.

DN

A10

1683

Col

ombi

a3.

5583

3,-7

6.58

278

JF93

4979

JF93

5016

–JF

7863

96JF

7864

21H

uit

aca

sp.

DN

A10

1407

Col

ombi

a5.

7795

6,-7

3.45

377

DQ

5180

90D

Q82

5596

DQ

5180

50D

Q51

8129

DQ

5181

67H

uit

aca

sp.

DN

A10

1681

Col

ombi

a5.

0995

6,-7

5.40

594

JF93

4977

JF93

5015

JF93

5032

JF78

6397

JF78

6422

Hu

itac

asp

.D

NA

1021

50C

olom

bia

3.55

833,

-76.

5827

8JF

9349

81JF

9350

12JF

9350

33–

JF78

6423

Hu

itac

asp

.D

NA

1046

46C

olom

bia

3.55

833,

-76.

5827

8JF

9349

82JF

9350

17JF

9350

30–

–H

uit

aca

sp.

DN

A10

1671

Col

ombi

a7.

4166

7,-7

2.43

333

JF93

4978

JF93

5013

JF93

5031

JF78

6398

JF78

6424

Met

agov

easp

.D

NA

1016

80C

olom

bia

5.09

542,

-75.

3907

5JF

9349

72JF

9349

99JF

9350

36JF

7864

00–

Met

agov

easp

.D

NA

1046

47C

olom

bia

3.55

833,

-76.

5827

8JF

9349

88JF

9350

03JF

9350

37–

–M

etag

ovea

sp.

DN

A10

4648

Col

ombi

a3.

5583

3,-7

6.58

278

JF93

4989

JF93

5004

JF93

5038

––

Met

agov

easp

.D

NA

1014

08C

olom

bia

-4.0

4495

,-6

9.98

979

DQ

8255

43D

Q82

5598

DQ

8256

18–

DQ

8255

14M

etag

ovea

sp.

DN

A10

1410

Col

ombi

a1.

2850

0,-7

8.07

367

DQ

5180

91D

Q82

5597

JF93

5034

GQ

9128

60D

Q51

8168

Met

agov

easp

.D

NA

1016

54C

olom

bia

1.25

,-7

8.25

*JF

9349

70JF

9350

00JF

9350

35JF

7864

01JF

7864

25M

etag

ovea

sp.

DN

A10

2151

Col

ombi

a5.

4858

3,-7

6.01

667

JF93

4971

JF93

5001

–JF

7864

02–

Met

agov

easp

.D

NA

1016

85C

olom

bia

3.56

889,

-76.

5886

1JF

9349

73JF

9349

98–

JF78

6403

–M

etag

ovea

sp.

DN

A10

1670

Col

ombi

a1.

6163

9,-7

6.10

417

JF93

4984

–5JF

9350

02–

––

Met

agov

easp

.D

NA

1014

09C

olom

bia

1.28

500,

-78.

0736

7D

Q82

5544

DQ

8255

99D

Q82

5619

DQ

8256

46JF

7864

26M

etag

ovea

sp.

DN

A10

1642

Col

ombi

a3.

5583

3,-7

6.58

278

JF93

4986

JF93

5006

JF93

5042

––

Met

agov

easp

.D

NA

1016

86C

olom

bia

3.56

889,

-76.

5886

1JF

9349

87JF

9350

05–

JF78

6404

–M

etag

ovea

sp.

DN

A10

5826

Gu

yan

a1.

3365

5,-5

8.96

510

JF93

4983

JF93

5010

JF93

5041

JF78

6408

–M

etas

iro

amer

ican

us

DN

A10

1532

Flo

rida

,U

SA

30.5

6489

,-8

4.95

163

DQ

8255

42D

Q82

5595

DQ

8256

16D

Q82

5645

DQ

8255

13M

etas

iro

amer

ican

us

DN

A10

5644

Sou

thC

arol

ina,

US

A35

.062

31,

-82.

795

JF93

4961

JF93

4995

–JF

7863

93JF

7864

18M

etas

iro

amer

ican

us

DN

A10

5645

Sou

thC

arol

ina,

US

A32

.189

23,

-81.

08JF

9349

62JF

9349

96JF

9350

27JF

7863

94JF

7864

19N

eogo

vea

virg

inie

DN

A10

4823

Fre

nch

Gu

ian

a4.

1951

1,-5

2.14

936

JF93

4974

JF93

5007

–JF

7864

05–

Neo

gove

avi

rgin

ieD

NA

1058

24F

ren

chG

uia

na

4.08

813,

-52.

6752

0JF

9349

75JF

9350

08JF

9350

39JF

7864

06–

Neo

gove

asp

.D

NA

1058

25G

uya

na

1.38

803,

-58.

9463

2JF

9349

76JF

9350

09JF

9350

40JF

7864

07–

Par

ogov

iaga

bon

ica

DN

A10

4620

Gab

on0.

5044

8,12

.795

25JF

9349

69JF

9350

19JF

9350

47JF

7864

11–

Par

ogov

iasi

ron

oid

esD

NA

1010

59B

ioko

,E

quat

oria

lG

uin

ea3.

7257

0,8.

8382

8D

Q51

8092

DQ

8256

06D

Q51

8051

DQ

5181

31D

Q51

8169

Par

ogov

iasi

ron

oid

esD

NA

1010

61B

ioko

,E

quat

oria

lG

uin

ea3.

7028

4,8.

8752

0D

Q82

5550

DQ

8256

07JF

9350

43D

Q82

5650

DQ

8255

19

Par

ogov

iacf

.si

ron

oid

esD

NA

1004

62E

quat

oria

lG

uin

ea2.

1830

5,9.

8030

5A

Y63

9493

DQ

8256

03–

–A

Y63

9459

Par

ogov

iacf

.si

ron

oid

esD

NA

1010

53E

quat

oria

lG

uin

ea1.

6581

5,10

.311

43D

Q82

5548

DQ

8256

04D

Q82

5623

–D

Q82

5517

Par

ogov

iacf

.si

ron

oid

esD

NA

1010

56E

quat

oria

lG

uin

ea1.

4485

8,9.

7808

6D

Q82

5549

DQ

8256

05D

Q82

5624

DQ

8256

50D

Q82

5518

Par

ogov

iacf

.si

ron

oid

esD

NA

1046

19C

amer

oon

2.74

108,

9.88

181

JF93

4967

JF93

5022

JF93

5045

JF78

6409

JF78

6428

Par

ogov

iasp

.D

NA

1010

52E

quat

oria

lG

uin

ea1.

6581

5,10

.311

43D

Q82

5546

DQ

8256

01D

Q82

5620

DQ

8256

48D

Q82

5515

Par

ogov

iasp

.D

NA

1010

57E

quat

oria

lG

uin

ea2.

1311

9,9.

8718

7D

Q82

5547

DQ

8256

02D

Q82

5621

–D

Q82

5516

Par

ogov

iasp

.D

NA

1046

15C

amer

oon

4.80

084,

9.66

326

JF93

4966

JF93

5021

JF93

5044

JF78

6410

JF78

6427

Par

ogov

iasp

.D

NA

1046

18C

amer

oon

2.74

108,

9.88

181

JF93

4968

JF93

5020

JF93

5046

JF78

6412

JF78

6429

Par

ogov

iasp

.D

NA

1056

71Iv

ory

Coa

st5.

8333

3,-7

.350

00*

JF93

4965

JF93

5018

JF93

5048

JF78

6413

–

OU

TG

RO

UP

SP

roto

loph

us

sin

gula

ris

DN

A10

1033

Cal

ifor

nia

,U

SA

EF

0280

95E

F02

8096

EF

1085

81E

F10

8586

EF

1085

92M

egal

opsa

lis

sp.

DN

A10

0783

SI,

New

Zea

lan

dE

F10

8573

EF

1085

76E

F10

8582

EF

1085

87E

F10

8593

Hes

pero

nem

asto

ma

mod

estu

mD

NA

1003

12O

rego

n,

US

AA

F12

4942

EF

1085

77E

F10

8583

EF

1085

88E

F10

8594

Den

dro

lasm

apa

rvu

lum

DN

A10

0318

Japa

nE

F10

8574

EF

1085

78E

F10

8584

EF

1085

89–

Equ

itiu

sd

oria

eD

NA

1006

07A

ust

rali

aU

3700

3E

F10

8579

–E

F10

8590

EF

1085

95S

and

okan

mal

ayan

us

DN

A10

0321

Mal

aysi

aE

F10

8575

EF

1085

80E

F10

8585

EF

1085

91E

F10

8596

Ast

eris

ksin

dica

teap

prox

imat

eco

ordi

nat

es.

10 G. GIRIBET ET AL.


Bivort & Giribet, 2010), as a result of the largernumber of specimens based on literature sources andthe lack of scanning electron micrographs of severalspecies.

When selecting the morphological terminals, weattempted to maximize overlapping with the molecu-lar matrix and also attempted to include the typespecies of each genus, with a few exceptions. Allmonotypic genera were also included, irrespective ofwhether molecular data were available or not. Mono-typic genera not represented by molecular data areAnkaratra Shear & Gruber, 1996, Iberosiro de Bivort& Giribet, 2004, Manangotria Shear & Gruber, 1996,Marwe Shear, 1985, Odontosiro Juberthie, 1961, Spe-leosiro Lawrence, 1931, and Stylocellus. Similarly,Shearogovea mexasca, now not considered as amember of Neogoveidae or Neogovea (Benavides &Giribet, 2007; Giribet, 2011), is not represented bymolecular data but was included in the combinedanalysis.

When a species was represented by multiplemolecular terminals, the morphological data matrixwas replicated so all molecular terminals were repre-sented by the same morphological codings. Thisapplies to the three populations of Metasiro america-nus (Davis, 1933), the two specimens of Parogoviasironoides Hansen, 1921 and four specimens of P. cf.sironoides, the two specimens of Metagovea sp.(DNA104648 and DNA104647), two specimens ofNeogovea virginie, and two specimens of Suzukielussauteri (Roewer, 1916).

The annotated morphological data matrix has beendeposited in Morphobank (morphobank.org) withaccession number P199 (http://morphobank.org/permalink/?P199).

PHYLOGENETIC ANALYSIS: DYNAMIC HOMOLOGY

UNDER PARSIMONY

Parsimony analysis under direct optimization(Wheeler, 1996) used the software POY, version 4.1.2(Varón, Sy Vinh & Wheeler, 2010) on six processors on

a Quad-Core Intel Xeon 3 GHz Mac Pro or on 40processors in the Odyssey cluster at Harvard Univer-sity FAS Research computing facility. Timed searches(multiple Wagner trees followed by SPR + TBR +ratchet and tree fusing) of 6–12 h each were run forthe combined analyses of all molecules under sixanalytical parameter sets (see below). Two additionalrounds of sensitivity analysis tree fusing (SATF)(Giribet, 2007a), taking all input trees from the pre-vious round of analyses, were conducted for the com-bined analysis of molecules under the multipleparameter sets evaluated. These were also 6-h timedsearches, and the results of these were plotted tocheck for stability in the results. Once a parameterset stabilized and the optimal result was found mul-tiple times, we stopped that inquiry but continuedwith additional rounds of searches for those param-eter sets that continued improving or that found theoptimal solution only once. The results of these analy-ses are shown in Table 3.

Because a broad parameter space has already beenexplored in detail in earlier studies (Boyer et al.,2007b), we restricted the dynamic homology analysesto six parameter sets, named 111, 121, 211, 221, 3221,and 3211. Parameter set 3221 (indel opening cost = 3;indel extension cost = 1; transversions = transitions= 2) has been favoured in many analyses and hasbeen justified philosophically as the best way of ana-lyzing data under direct optimization (De Laet, 2010).In addition, we explored a parameter set, named3211, where transversions and transitions receive dif-ferent costs (indel opening cost = 3; indel extensioncost = 1; transversion cost = 2; transition cost = 1),extending the idea of mixed-parameter sets ofSharma et al. (2011). Four other parameter sets 111,121, 211, and 221, optimal in the analyses of Boyeret al. (2007b) and aiming to limit the differencebetween indel costs and transformation costs (Spagna& Álvarez-Padilla, 2008), were explored. To calculatethe WILD (Wheeler, 1995; Sharma et al., 2011) eachindividual partition, or the combination of the twonuclear ribosomal RNA partitions, were run with a

Table 3. Search strategy and tree length stabilization after subsequent rounds of sensitivity analysis tree fusing (TFN)for each parameter set

TF4 TF5 TF6 TF7 TF8 TF9

111 27101 27074 27074 – – –121 41849 41773 41773 – – –211 28975 28944 28940 28940 – –221 45211 45179 45131 45118 45115 451153221 55982 55744 55729 55729 – –3211 43121 42874 42854 42854 – –

111 and 121 stabilized after five rounds of tree fusing; 221 stabilized after eight rounds of tree fusing



similar search strategy as described above with a 2-htimed search. The resulting WILD values are shown inTable 4.

A jackknife resampling analysis (Farris et al., 1996)with 1000 replicates and a probability of deletion ofeach character of 0.36 was applied to assess nodalsupport. Because resampling techniques may be mean-ingless under dynamic homology, different strategiescan be applied. Dynamic characters can be convertedto a static set, although this tends to inflate supportvalues because it is based on the implied alignmentthat favours the topology. Instead, we resample char-acters that were static a priori (morphology and pre-aligned protein-encoding genes), as well as fragmentsof the dynamic characters by both using the number offragments (eight fragments for 16S rRNA, six frag-ments for 18S rRNA, and ten fragments for 28S rRNA),as well as the command auto_sequence_partition,which evaluates each predetermined fragment. If along region appears to have no indels, then the frag-ment is broken inside that region.

PHYLOGENETIC ANALYSIS:PROBABILISTIC APPROACHES

Maximum likelihood (ML) analyses were conductedon static alignments, which were inferred as follows.Sequences of ribosomal genes were aligned usingMUSCLE, version 3.6 (Edgar, 2004) with defaultparameters, and subsequently treated withGBLOCKS, version 0.91b (Castresana, 2000) to cullpositions of ambiguous homology. For these genes,indels were permitted within blocks. Sequences of theprotein-encoding genes COI and histone H3 werealigned using MUSCLE, version 3.6 with defaultparameters as well, although alignments were con-firmed using protein sequence translations beforetreatment with GBLOCKS, and no gaps were permit-ted within blocks (COI has length variation, so theseregions were excluded in GBLOCKS). The size of

data matrices for each gene before and subsequentto treatment with GBLOCKS is provided in theAppendix (Table A1).

ML analysis was conducted using RaxML, version7.2.7 (Stamatakis, 2006) on 40 CPUs of a cluster atHarvard University, FAS Research Computing (http://rc.fas.harvard.edu/faq/odyssey). For the maximumlikelihood searches, a unique GTR model of sequenceevolution with corrections for a discrete gamma dis-tribution (GTR + G) was specified for each data par-tition, and 500 independent searches were conducted.Nodal support was estimated via the rapid bootstrapalgorithm (1000 replicates) using the GTR-CAT model(Stamatakis, Hoover & Rougemont, 2008), throughthe CIPRES, version 3, gateway, using the Abe DellIntel 64 Linux teragrid cluster housed at the NationalCenter for Supercomuting Applications (Universityof Illinois). Bootstrap resampling frequencies werethereafter mapped onto the optimal tree from theindependent searches.

ESTIMATION OF DIVERGENCE TIMES

Ages of clades were inferred using BEAST, version1.6.1 (Drummond et al., 2006; Drummond &Rambaut, 2007). We assigned the best fitting models(a GTR model of sequence evolution with correctionsfor a discrete gamma distribution and a proportion ofinvariant sites, GTR + G+ I) selected by MODELTEST,version 3.7 (Posada & Crandall, 1998; Posada, 2005)to each partition. Protein-encoding genes were parti-tioned into two sets by codon positions, separatingthird codon positions from the set of first and secondpositions. An uncorrelated lognormal clock model wasinferred for each partition, and a Yule speciationprocess was assumed for the tree prior. We selectedthe uncorrelated lognormal model because its accu-racy is comparable to an uncorrelated exponen-tial model, although it has narrower 95% highestposterior density (HPD) intervals. Additionally, the

Table 4. Tree lengths for different data partitions (rib, nuclear ribosomal genes; coi, cytochrome c oxidase subunit I; 16s,16S rRNA; h3, histone H3; mol, all molecular partitions) analyzed and incongruence length differences (ILD) between thedata sets

rib coi 16s h3 Mol wILD

111 5852 12315 6857 1449 27074 0.02220121 8846 18664 11350 1974 41773 0.02248211 6659 12502 7692 1449 28940 0.02205221 10312 18890 12897 1974 45115 0.02310

3211 9268 18810 11861 1967 42851 0.022053221 12212 24996 14397 2898 55713 0.02172

Parameter set 3221 (in italics) minimizes the ILD value.



variance of the uncorrelated lognormal model canbetter accommodate data that are already clock-like(Drummond et al., 2006). Priors were sequentiallyoptimized in a series of iterative test runs (data notshown). Markov chains were run for 50 000 000 gen-erations, sampling every 1000 generations. Conver-gence diagnostics were assessed using TRACER,version 1.5 (Rambaut & Drummond, 2007).

Fossil taxa were used to calibrate divergence times.We constrained the age of Eupnoi to 410 Mya usingthe Devonian harvestman Eophalangium sheari[Dunlop et al. 2004 [Dunlop et al., 2003; 2004 (for2003)]; a normal distribution with a standard devia-tion of 5 Myr was applied to this node to account foruncertainty in estimation of fossil age. Dyspnoi wereconstrained using a normal distribution with a meanof 300 Mya and a standard deviation of 10 Myr, on thebasis of the Carboniferous fossils Eotrogulus fayoliThevenin, 1901 and Nemastomoides elaveris Thev-enin, 1901 (Dunlop, 2007).

We explored constraining the family Stylocellidaeusing the Early Cretaceous Burmese amber fossilPalaeosiro burmanicum Poinar, 2008 (Poinar, 2008)1.We used a gamma distribution with shape param-eters (a, b) = (8, 14), and an offset of 105 Myr for thediversification of Stylocellidae; such a prior distribu-tion establishes a floor in the age of stylocellids(105 Mya), at the same time enabling estimates ofdiversification as early as the Late Permian, in accor-dance with previous estimates (Boyer et al., 2007b;Clouse & Giribet, 2010). However, because the inclu-sion of this last constraint did not affect the ageestimate of Stylocellidae, we ultimately did notinclude it in the analysis.

ANCESTRAL AREA RECONSTRUCTION

Likelihood analysis of ancestral area reconstructionwas conducted using the software LAGRANGE (Reeet al., 2005; Ree & Smith, 2008). We divided the datedtree from BEAST analysis into three parts for ana-lytical tractability: (1) the Pettalidae subtree; (2) the(Troglosironidae + Ogoveidae + Neogoveidae) subtree;and (3) the (Sironidae + Stylocellidae) subtree. Foreach subtree, we implemented stratified dispersalconstraint matrices for multiple spans of time for therelevant areas inhabited by the constituent taxa ofeach subtree. Geological events used to delimit the

time spans are sensu Sanmartín & Ronquist (2004)and Hall (2002). The maximum number of areas inancestral ranges was held at two (this conventionreflects empirical observations of Cyphophthalmispecies, the majority of which are narrowly distrib-uted endemics), and dispersal constraints were set to1.0 (if landmasses were connected), 0.1 (if landmasseswere disjunct) or 0 (if landmasses did not exist). Areasand geological intervals for each subtree are indicatedin the Appendix Table A2 (the Python scripts specify-ing dispersal constraint matrices are available uponrequest from the authors).

HABITAT SUITABILITY AND DISTRIBUTION MODELLING

To generate predictions of habitat suitability andpotential lineage distributions, habitat suita-bility models (HSMs) of the major lineages ofCyphophthalmi were reconstructed using the19 bioclimatic variables of Hijmans et al. (2005:http://www.worldclim.org/). These variables provide asummary of the monthly temperature and precipita-tion worldwide. These variables are well documentedand are widely used in studies relaying on niche anddistribution modelling (Evans et al., 2009; Smith &Donoghue, 2010). By contrast to the raw temperatureand precipitation data, they do provide biologicallyrelevant information. We have used all 19 variablesat 10 arc minutes resolution. In addition, analyseswith a subset of the environmental variables repre-senting only the most important variables were per-formed (thus reducing the dimensionality of theanalyses and the risk of over fitting) and the resultsobtained were compared. To evaluate the variablessignificance, we used jackknife (as implemented inMAXENT).

HSMs were built using the maximum entropyalgorithm implemented in MAXENT, version 3.3.3a(Phillips, Anderson & Schapire, 2006; Phillips &Dudik, 2008). Maximum entropy has shown a highperformance score in comparison with other methods(Araujo & Rahbek, 2006) and also allows workingwith fewer data points (Pearson et al., 2007). Thetotal number of unique localities with occurrenceobservations used for the modelling of habitat suit-ability was: Pettalidae, N = 107; Sironidae, N = 60;Stylocellidae, N = 127; and Sternophthalmi, N = 90.To evaluate the performance of the model, cross-validation as implemented in MAXENT (ten repli-cates) was used in all runs.

To test the sensitivity of the results to the model-ling algorithm, we have run the same set of analysisusing the simpler BIOCLIM (Nix, 1986) profilemethod as implemented in openModeller, version1.1.0 (de Souza Muñoz et al., 2011). Climatic enve-lopes’ extent and distribution were modelled world-

1Palaeosiro burmanicum Poinar, 2008 was placed withinSironidae in the original description based on the lack of asternal apophysis with gland pores and dentition on the tarsalclaw of leg II, although these only rule out placement withinTroglosironidae and Neogoveidae. Moreover, the shape andposition of the ozophores, the presence of eyes, the carina ofthe anal plate (similar to some Fangensis Schwendinger &Giribet, 2005), and the collecting locality of the fossil, are allconsistent with an early diverging lineage of Stylocellidae.



wide to compare the actual linage distributions withthe distribution of potentially suitable climates.

The software package ENMTools, version 1.3(Warren, Glor & Turelli, 2010) was used to accessclimatic envelopes’ differentiation. ENMTools imple-ments the I, Schoener’s D and relative rank metrics tocompare models predictions (Schoener, 1968; Warren,Glor & Turelli, 2008). The three indices measuresimilarity of predicted habitat suitability distribu-tions and range from 0, indicating no overlap, to 1,indicating complete overlap. In addition, habitat suit-ability score differences were evaluated by comparingthe similarity indices (models overlap) for the modelsbuilt from the actual occurrences of the two species toa null distribution generated by nonparametric resa-mpling. Comparisons were performed using the nicheidentity test (Warren et al., 2008) implemented inENMTools.

RESULTS

Analysis of the combined molecular data matrixunder selected parameter sets for direct optimizationresulted in topologies that agree on several basicaspects of cyphophthalmid phylogeny, includingmonophyly of the suborder, a sister group relationshipof Pettalidae to all other families, and a clade con-taining all members of the families Troglosironidae,Ogoveidae, and Neogoveidae. All parameter setsalso resulted in very similar WILD numbers, with3221 being slightly favoured above all others(WILD = 0.02172; the worst parameter set being 121,with WILD = 0.02248). Stabilization of parameter set3221 occurred after nine rounds of tree fusing. Theoptimal tree, along with the Navajo rugs (Giribet,2003b) for the familial monophyly and relationships,is presented in Figure 3. A clade containing the fami-lies Sironidae and Stylocellidae is also found undermost analytical conditions (Fig. 3).

Monophyly of Pettalidae, Troglosironidae, Stylocel-lidae, and Ogoveoidea (= Ogoveidae + Neogoveidae) issupported under every analyzed parameter set, asare many internal clades within the familiesStylocellidae, Sironidae, and Neogoveidae. However,Sironidae is not monophyletic under any parameterset when combining all data (Sironidae is monophyl-etic when nuclear ribosomal genes are analyzedalone). In this case, a clade containing the generaSiro, Paramiopsalis, and Cyphophthalmus is stableto parameter variation, although Parasiro and

Suzukielus often appear at the base of Stylocellidae,or as sister to a clade including the familiesStylocellidae, Troglosironidae, Ogoveidae, andNeogoveidae (parameter set 3211). The North Ameri-can Siro and the European Siro form reciprocallymonophyletic groups and this clade is sister to Para-miopsalis + Cyphophthalmus. In the case ofNeogoveidae, most parameter sets find Metasiro to bethe sister genus to all other neogoveids but, underparameter sets 111 and 211, Ogovea appears as sisterto the African genus Parogovia, both forming thesister clade to Metasiro. These are the only param-eter sets that find monophyly of the South Americanneogoveids, with Canga as sister genus to all otherSouth American genera. All other parameter setsinstead support monophyly of Neogoveidae, Metasiroas the sister genus to all other species, the Braziliangenus Canga as sister to the African genusParogovia, and the stable relationship of ((Brasil-ogovea, Neogovea) (Huitaca, Metagovea)). Relation-ships of Stylocellidae are well resolved, as:(Fangensis (Meghalaya (Miopsalis, Leptopsalis))).Although all pettalid genera are supported, theirrelationships remain unstable to parameter setvariation, and stable are only the sister group rela-tionships of Purcellia to Chileogovea Roewer, 1961and of Karripurcellia Giribet, 2003 to Pettalus. Twogenera appear as candidate sister groups to all otherpettalids, the South African genus ParapurcelliaRosas Costa, 1950 or the north-eastern Australianendemic Austropurcellia Juberthie, 1988. Jackknifesupport values for the pettalid generic relationshipsare, for the most part, below 50%.

The maximum likelihood analysis resulted in a treetopology with lnL = -103 563.078879. The likelihoodtree topology (Fig. 4) is largely comparable to resultsfrom parsimony analyses but notably recovers amonophyletic Sironidae (i.e. including the generaParasiro and Suzukielus), albeit with low nodalsupport (BS = 44%). As in the direct optimizationoptimal tree, Parapurcellia is sister to all other pet-talid genera, and Purcellia + Chileogovea form a sup-ported clade (66% bootstrap support; BS), whereasKarripurcellia and Pettalus form a clade without sig-nificant nodal support. No other generic relationshipsfind high support. Troglosironidae is sister to Ogoveo-idea (84% BS), and the structure of Neogoveidae isalmost identical to that of the optimal direct optimi-zation tree, with the exception that Brasilogovea ishere monophyletic (the sequences for one terminal

Figure 3. Phylogenetic tree based on the parsimony direct optimization analysis of molecular data under parameter set3221 (55 713 weighted steps). Clade colours correspond to those in Fig. 2. Navajo rugs indicate monophyly (black) ornon-monophyly (white) of a given node under the parameter set specified in the legend. Numbers above nodes indicatejackknife support values.

�



DNA101383

Metagovea sp. DNA101680

Rakaia sp. DNA101807

Rakaia macra DNA101808

Pettalus sp. DNA101286



Cyphophthalmus ognjenovici DNA101039

Leptopsalis sp. DNA101514

Cyphophthalmus duricorius DNA100487

Cyphophthalmus markoi DNA100497

Austropurcellia scoparia DNA100946

Aoraki sp. DNA101126

Parogovia sp. DNA101057


Rakaia stewartiensis DNA100944


Miopsalis sp. DNA101519


DNA101295

Rakaia dorothea DNA100943

Cyphophthalmus trebinjanus DNA101038

Meghalaya sp. DNA101767




Aoraki tumidata DOC094

Siro shasta DNA101622

Parapurcellia monticola DNA100386


Conomma oedipus

Marthana sp.


Austropurcellia forsteri DNA100945

Aoraki denticulata major DNA100959


Cyphophthalmus zetae DNA100907

Troglosiro longifossa DNA100867


Huitaca ventralis DNA101674

Rakaia minutissima DNA101291

Parogovia cf. sironoides DNA104619

Miopsalis sp. DNA102032, DNA102053, DNA102058

Parogovia gabonica DNA104620

Siro rubens DNA100457

Hesperonemastoma modestum

Huitaca sp. DNA101683

Siro calaveras DNA101623


Rakaia uniloca DNA101812

Suzukielus sauteri DNA101550

Meghalaya sp. DNA103242, DNA103243, DNA103244


Aoraki crypta DNA101289


Metasiro americanus DNA101532

Neogovea virginie DNA104823


Cyphophthalmus teyrovskyi DNA100910


Brasilogovea sp. DNA101665


Troglosiro ninqua DNA100577

“Brasilogovea” sp. Tobogan DNA100869

Fangensis insulanus DNA100388, DNA101063

Parasiro minor DNA103535




Neogovea sp. DNA105825




Austropurcellia arcticosa DNA100951

Cyphophthalmus gordani DNA100495

Austropurcellia daviesae DNA100947

Siro boyerae DNA101614

Karripurcellia harveyi DNA101303

Rakaia sorenseni digitata DNA100970



Cyphophthalmus rumijae DNA100492

Rakaia media DNA101292

Siro valleorum DNA100461

Troglosiro juberthiei DNA100344

Pettalus thwaitesi DNA101223

Troglosiro aelleni DNA100345

Sandokan malayanus

Miopsalis sp. DNA101468, DNA100950


Troglosiro wilsoni DNA102324

Cyphophthalmus gjorgjevici DNA100498

Aoraki calcarobtusa westlandica DNA101129

Neogovea virginie DNA105824

Leptopsalis sp. DNA101093, DNA1000496

Siro kamiakensis DNA101613


Metagovea sp. DNA101642Metagovea sp. DNA101409

Siro cf. kamiakensis DNA101611

Rakaia antipodiana DNA100957

Paramiopsalis ramulosus DNA103538


Rakaia pauli DNA100968

Parogovia sironoides DNA101059

Paramiopsalis sp. DNA104624


Rakaia lindsayi DNA101128

Purcellia illustrans DNA100387


Cyphophthalmus minutus DNA100493

Leptopsalis lydekkeri DNA101064

Cyphophthalmus serbicus DNA102098



Siro acaroides DNA100488





Aoraki longitarsa DNA101806


Paramiopsalis eduardoi DNA101878

Rhampsinitus sp.

Meghalaya sp. DNA101094, DNA101500



Aoraki granulosa DNA101841

Chileogovea sp. DNA100490


Neopurcellia salmoni DNA100939

Canga renatae DNA105680

Limulus polyphemus

Paramiopsalis ramulosus DNA100459

Aoraki healyi DNA100940

Siro exilis DNA100489




Cyphophthalmus ere DNA100499


Fangensis spelaeus DNA100669



Rakaia sorenseni sorenseni DNA100969

Rakaia magna australis DNA100962

Suzukielus sauteri DNA101543

Chileogovea oedipus DNA100413

Troglosiro monteithi DNA101580

Ogovea cameroonensis DNA104617



Rakaia solitaria DNA101294


Cyphophthalmus hlavaci DNA102099

Cyphophthalmus martensi DNA100494

Leptopsalis novaguinea DNA101510

Megalopsalis sp.


Aoraki denticulata denticulata DNA100955



Equitius doriae

Troglosiro urbanus DNA101710

Meghalaya sp. DNA101494, DNA101506, DNA101765


Cyphophthalmus corfuanus DNA102111


Aoraki inerma DNA100966

Parapurcellia silvicola DNA100385

93

98

92

97

87

98

82

100

100

100

55

52

83

57

60

80

97

98

100

94

92

76

83

64

97

99

65

81

100

73

82

52

99

98

99

73

58

88

100

88

96

59

51

100

100

92

80

64

52

56

94

70

83

56

99

100

81

72

96

81

56

69

57

87

88

72

72

64

52

100

69

99

52

58

75

99

67

83

99

100

73

85

54

100

98

68

87

59

100

69

91

55

98

83

100

91

70

99

97

32

100

100

100

100

99

99

82

Troglosironidae

Ogoveidae

Pettalidae100

Neogoveidae

Sironidae

Stylocellidae

100

100

100

100

99

99

100

100

100

111 121 211

221 3211 3221




Pettalus sp. DNA101287Pettalus sp. DNA101286

Pettalus sp. DNA101283Pettalus sp. DNA101288

Pettalus sp. DNA101285Pettalus thwaitesi DNA101223

Aoraki sp. DNA101126Aoraki healyi DNA100940

Aoraki inerma DNA100966Aoraki crypta DNA101289

Aoraki calcarobtusa westlandica DNA101129 Aoraki granulosa DNA101841

Aoraki tumidata DOC094Aoraki denticulata major DNA100959

Aoraki denticulata denticulata DNA100955Aoraki longitarsa DNA101806

Austropurcellia daviesae DNA100947Austropurcellia scoparia DNA100946

Austropurcellia arcticosa DNA100951Austropurcellia forsteri DNA100945

Parapurcellia monticola DNA100386Parapurcellia silvicola DNA100385

Purcellia illustrans DNA100387

Chileogovea sp. DNA100490Chileogovea oedipus DNA100413

Karripurcellia harveyi DNA101303

Neopurcellia salmoni DNA100939


Rakaia media DNA101292Rakaia solitaria DNA101294


Rakaia sp. DNA100954Rakaia sp. DNA101297Rakaia dorothea DNA100943


Rakaia pauli DNA100968Rakaia antipodiana DNA100957

Rakaia macra DNA101808Rakaia sp. DNA100958

Rakaia sp. DNA101807Rakaia sorenseni sorenseni DNA100969

Rakaia sorenseni digitata DNA100970

Rakaia stewartiensis DNA100944Rakaia lindsayi DNA101128

Rakaia minutissima DNA101291DNA101295


Troglosiro aelleni DNA100345Troglosiro ninqua DNA100577

Troglosiro monteithi DNA101580

Troglosiro juberthiei DNA100344

Troglosiro urbanus DNA101710Troglosiro longifossa DNA100867

Ogovea cameroonensis DNA104617



Metasiro americanus DNA101532Metasiro americanus DNA105644

Canga renatae DNA105680

Parogovia sp. DNA101057Parogovia sp. DNA101052

Parogovia sp. DNA104618Parogovia gabonica DNA104620

Parogovia sp. DNA104615Parogovia sironoides DNA101059



Parogovia cf. sironoides DNA100462Parogovia cf. sironoides DNA101056


Neogovea virginie DNA104823Neogovea virginie DNA105824


“Brasilogovea” sp. Tobogan DNA100869Brasilogovea sp. DNA101665






Metagovea sp. Valle DNA101685Metagovea sp. DNA102151m




Huitaca ventralis DNA101674Huitaca sp. DNA101407


Huitaca sp. DNA101683Huitaca sp. DNA104646

Huitaca sp. DNA102150Huitaca sp. DNA101681

Suzukielus sauteri DNA101550Suzukielus sauteri DNA101543

Parasiro minor DNA103535 DNA101383

Cyphophthalmus ognjenovici DNA101039Cyphophthalmus minutus DNA100493

Cyphophthalmus trebinjanus DNA101038Cyphophthalmus teyrovskyi DNA100910

Cyphophthalmus gordani DNA100495Cyphophthalmus martensi DNA100494


Cyphophthalmus corfuanus DNA102111Cyphophthalmus hlavaci DNA102099




Cyphophthalmus rumijae DNA100492Cyphophthalmus duricorius DNA100487


Paramiopsalis ramulosus DNA103538Paramiopsalis ramulosus DNA100459

Paramiopsalis sp. DNA104624Paramiopsalis eduardoi DNA101878

Siro shasta DNA101622Siro kamiakensis DNA101613

Siro cf. kamiakensis DNA101611

Siro exilis DNA100489Siro boyerae DNA101614

Siro calaveras DNA101623Siro acaroides DNA100488

Siro rubens DNA100457Siro valleorum DNA100461

Leptopsalis sp. DNA101932Leptopsalis sp. DNA101930


Leptopsalis lydekkeri DNA101064

Leptopsalis novaguinea DNA101510Leptopsalis sp. DNA101937






Leptopsalis sp. DNA101093, DNA101496Leptopsalis sp. DNA101938




Miopsalis sp. DNA101468, DNA101950Miopsalis sp. DNA102032, DNA102053, DNA102058


Miopsalis sp. DNA103249 Miopsalis sp. DNA103259

Miopsalis sp. DNA103254Stylocellidae gen. sp. DNA104981

Meghalaya sp. DNA101094, DNA101500Meghalaya sp. DNA101494, DNA101506, DNA101765



Meghalaya sp. DNA102051Meghalaya sp. DNA103242, DNA103243, DNA103244

Meghalaya sp. DNA103251Fangensis insulanus DNA100388

Fangensis spelaeus DNA100669

Equitius doriae DNA100607

Megalopsalis sp. DNA100783Marthana sp. DNA100613

Sandokan malayanus DNA100321Conomma oedipus DNA101051

Hesperonemastoma modestum DNA100312

Rhampsinitus sp. DNA100710

* ** *

85

*

91

72

0.1

**

66

97

**

*

*9985

60

85

98

68

50 67

90

55 94

*

97

* **

*

9870

*69

65 *86

*

*83

53** 92

84

81

8395

*99

*99 58

*98

88 92

76

** ***

*

**66

*90

75

99

** 93

* 94

* 6277

**99

97 71

69

* 9774

99

*

*99

8098

*87

*

*

7161

58

9550

*

**

* *

9896

53

92

92* *

*80 89

97

**

99

87

70

Pettalidae

Trogosironidae

Sironidae

Neogoveidae

Stylocellidae

Ogoveidae

I.O. Sternophthalmi

I.O. Boreophthalmi

I.O. Scopulophthalmi

Figure 4. Single most likely tree (lnL = -103 563.078879) for the combined molecular data set aligned using MUSCLEand subsequently trimmed with GBLOCKS and analyzed in RAxML under GTR + G. Clade colours correspond to thosein Fig. 2. Bootstrap support values are represented above each node; asterisks indicate 100% bootstrap value.



are based on a juvenile specimen, so the assignmentto this genus is tentative). Within Sironidae, Parasirois sister to all other genera, followed by Suzukielus,although bootstrap support for these basal nodes islow, as is the clade including the remaining sironidgenera. In this case, there is also reciprocal mono-phyly of the European and North American Siro, andthese form the sister group of Paramiopsalis +Cyphophthalmus. Structure of the genera within Sty-locellidae matches that of the direct optimizationanalyses.

The run of BEAST reached stationarity after10 000 000 generations; 20 000 000 generations werediscarded as burn-in. The tree topology recovered byBEAST (Fig. 5) is almost identical to that of theparsimony analysis under the parameter set 3221,with Parasiro and Suzukielus forming a paraphyleticgrade at the base of Stylocellidae, although posteriorprobabilities for the corresponding nodes are low(0.788 and 0.880, respectively). In all other aspects, italso resembles the topology of the maximum likeli-hood analysis, especially in the internal relationshipsamong the pettalid genera.

The diversification of Cyphophthalmi is estimatedat approximately 332 Mya (95% HPD: 297–362 Mya).Diversification times for the described familiesof Cyphophthalmi are estimated as: Neogoveidae,236 Mya (95% HPD: 208–266 Mya); Pettalidae,183 Mya (95% HPD: 148–218 Mya); Sironidae(excluding Parasiro and Suzukielus), 278 Mya (95%HPD: 243–311 Mya); Stylocellidae, 167 Mya (95%HPD: 140–195 Mya); and Troglosironidae, 57 Mya(95% HPD: 40–73 Mya). Ogoveidae, represented by asingle exemplar, diverged from Neogoveidae 261 Mya(95% HPD: 231–292 Mya), and Troglosironidaediverged from Ogoveoidea 279 Mya (95% HPD: 248–311 Mya). These results largely corroborate previousestimates of divergence times (Boyer et al., 2007b;Giribet et al., 2010), with the exception of Stylocel-lidae, the diversification of which is recovered asyounger than previously reported (Clouse & Giribet,2010). Although some species represented by multipleterminals are young (e.g. Suzukielus sauteri,Neogovea virginie, Parogovia sironoides), Metasiroamericanus is an old species, perhaps reflecting theexistence of cryptic species along its range.

All probabilistic approaches recognize a clade oftrans-Tasman Cyphophthalmi (the Australian andNew Zealand genera), although none of these land-masses or their constituent terranes appears mono-phyletic (Fig. 6). The ancestral area reconstruction ofthis clade is ambiguous, with the highest probabilityfor an origin in the Australian plate of New Zealand.The ancestral area of the family shows more ambigu-ity, the three most likely scenarios being a mixedSouth Africa/New Zealand Australian plate

(P = 0.291), South African (P = 0.207) or mixed SouthAfrica/Sri Lankan (P = 0.194). The ancestral areareconstruction of the clade including the three fami-lies with sternal opisthosomal gland openings (Fig. 7)is mostly West African/New Caledonian (P = 0.787),with the ogoveoid families being most likely WestAfrican (P = 0.662) or mixed between West Africa andthe south-eastern USA (P = 0.206), the latter beingonce connected to West Africa. A South American(Amazonian) origin of the family Neogoveidaereceives little support.

COMBINED ANALYSIS OF MOLECULES

AND MORPHOLOGY

The position of morphology-only taxa was unstable inthe first rounds of analyses, which (for example) didnot group the two Ogovea species, one represented bymorphology only, whereas the other one was repre-sented by molecules and morphology, despite beingalmost identical for the morphological data matrix.This appears to be a problem of the Wagner addition,as designed in most phylogenetic software, and wasresolved by fusing a jackknife tree and a first treeobtained during a normal search, as described above.The resulting trees of each subsequent analysis werethen fused to the previous pool of trees until results(topology and tree length) stabilized. The combinedanalysis of molecules and morphology in POYrequired eight rounds of tree fusing until stabilizingin a tree length of 56 984 weighted steps and findingthree trees differing only in the position of some of themorphology-only taxa (Fig. 9).

The overall topology is very similar to those of theanalyses with molecular data only, with a few excep-tions, and lowered jackknife support values. Pettal-idae is monophyletic (63%), and includes bothSpeleosiro and Manangotria from the morphology-only taxa. Speleosiro appears as sister to Purcelliaand Managotria is sister to Karripurcellia, althoughthese relationships, as with most other intergenericpettalid relationships, receive low support. Ankaratradoes not appear within Pettalidae and, instead,is basal to the clade containing Sironidae andStylocellidae.

Troglosironidae appears as sister to Ogoveoidea,although this tree differs from all previous trees inthat Neogoveidae is paraphyletic with respect toOgovea, which is sister to Parogovia, constituting anAfrican clade, sister to all the American species, withCanga as the sister group to Metasiro, and thisclade being sister to the remaining neogoveids [56%jackknife frequency (JF)]. The type species andmorphology-only species of Neogovea and Brasil-ogovea appear in a clade, although there is littlecorrespondence between the complete taxa and those



Pettalidae

Troglosironidae

Sironidae

Neogoveidae

Stylocellidae

Ogoveidae


Pettalus sp. DNA101287Pettalus sp. DNA101286Pettalus sp. DNA101283Pettalus sp. DNA101288 Pettalus sp. DNA101285Pettalus thwaitesi DNA101223

Aoraki sp. DNA101126Aoraki healyi DNA100940Aoraki inerma DNA100966Aoraki crypta DNA101289Aoraki calcarobtusa westlandica DNA101129 Aoraki granulosa DNA101841 Aoraki tumidata DOC094

Aoraki denticulata major DNA100959 Aoraki denticulata denticulata DNA100955




Parapurcellia monticola DNA100386Parapurcellia silvicola DNA100385Purcellia illustrans DNA100387


Karripurcellia harveyi DNA101303Neopurcellia salmoni DNA100939

Rakaia sp. DNA101293Rakaia media DNA101292Rakaia solitaria DNA101294


Rakaia sp. DNA100954Rakaia sp. DNA101297







Rakaia sorenseni digitata DNA100970Rakaia stewartiensis DNA100944Rakaia lindsayi DNA101128

Rakaia minutissima DNA101291Rakaia florensis DNA101295



Troglosiro monteithi DNA101580 Troglosiro juberthiei DNA100344 Troglosiro urbanus DNA101710Troglosiro longifossa DNA100867 Ogovea cameroonensis DNA104617


Metasiro americanus DNA105645Metasiro americanus DNA101532Metasiro americanus DNA105644Canga renatae DNA105680

Parogovia prietoi DNA101057Parogovia sp. DNA101052

Parogovia sp. Campo2 DNA104618Parogovia gabonica DNA104620Parogovia sp. Koupe DNA104615

Parogovia sironoides DNA101059Parogovia sironoides DNA101061



Metagovea sp. Valle DNA101686

Metagovea sp. DNA104648Metagovea sp. DNA104647Metagovea sp. Nambi DNA101409

Metagovea sp. Pasto DNA101642Metagovea sp. Huila DNA101670

Metagovea sp. Guyana DNA105826

Metagovea sp. Valle DNA101685Metagovea sp. Tatama DNA102151m

Metagovea sp. Caldas DNA101680

Metagovea sp. Planada DNA101654Metagovea sp. Nambi DNA101410

Metagovea sp. Leticia DNA101408

Huitaca ventralis DNA101674Huitaca boyacaensis DNA101407

Huitaca tama DNA101671

Huitaca bitaco DNA101683Huitaca sharkeyi DNA104646Huitaca depressa DNA102150Huitaca caldas DNA101681


Neogovea sp. Guyana DNA105825Brasilogovea sp. Tobogan DNA100869Brasilogovea chiribiqueta DNA101665


Parasiro minor DNA103535Parasiro coiffaiti DNA101383

Cyphophthalmus ognjenovici DNA101039Cyphophthalmus minutus DNA100493Cyphophthalmus trebinjanus DNA101038Cyphophthalmus teyrovskyi DNA100910Cyphophthalmus gordani DNA100495Cyphophthalmus martensi DNA100494








Paramiopsalis ramulosus DNA103538Paramiopsalis ramulosus DNA100459Paramiopsalis sp. DNA104624Paramiopsalis eduardoi DNA101878

Siro shasta DNA101622Siro kamiakensis DNA101613Siro cf. kamiakensis DNA101611Siro exilis DNA100489Siro boyerae DNA101614




Leptopsalis sp. DNA101944, DNA101945Leptopsalis lydekkeri DNA101064


Leptopsalis sp. DNA102042Leptopsalis sp. DNA102039Leptopsalis sp. DNA102032








Miopsalis sp. DNA102032, DNA102053, DNA102058Miopsalis sp. DNA101519


Miopsalis sp. DNA103254Miopsalis sp. DNA104981

Meghalaya sp. DNA101094, DNA101500Meghalaya sp. DNA101494, DNA101506, DNA101765Meghalaya sp. DNA101767

Meghalaya sp. DNA103265Meghalaya sp. DNA102051Meghalaya sp. DNA103242, DNA103243, DNA103244


Fangensis insulanus DNA100388, DNA101063Fangensis spelaeus DNA100669

65

CARBONIFEROUS PERMIAN TRIASSIC JURASSIC CRETACEOUS TERTIARY

144206248290354

*

*

*

*

**

*

0.983

0.929

0.792

0.871 *

Figure 5. Evolutionary time-tree of Cyphophthalmi inferred from BEAST analysis of all molecular data. Clade colourscorrespond to those in Fig. 2. Coloured bars indicate 95% highest posterior density (HPD) intervals for nodes of interest.Number on nodes indicate posterior probabilities; asterisks indicate posterior probability of 1.00.



represented by morphology only (i.e. Neogovea andBrasilogovea are not reciprocally monophyletic).

Ankaratra and Shearogovea form a grade at thebase of the Sironidae – Stylocellidae clade, withSironidae paraphyletic, as in the prior POY andBEAST analyses. Marwe and Iberosiro form a cladewith Paramiopsalis, and Odontosiro forms a cladewith Parasiro. Stylocellidae is monophyletic (63% JF),including the morphology-only species Stylocellussumatranus Westwood, 1874, Meghalaya annandaleiGiribet, Sharma & Bastawade, 2007, Miopsalis puli-caria Thorell, 1890, and Leptopsalis beccarii Thorell,1882–1883. Stylocellus sumatranus, the type speciesof Stylocellus, appears nested within the molecularMeghalaya clade; Meghalaya annandalei, the typespecies of Meghalaya, appears unresolved at the baseof the molecular Leptopsalis clade; Miopsalis puli-caria and Leptopsalis beccarii, the type species oftheir respective genera, appear nested deep within

the clade Leptopsalis. Although the stylocellid resultsmake little sense, this may be a result of the lack ofdiscrete characters useful for resolving their phyloge-netic relationships (see below).

A NEW CLASSIFICATION SYSTEM FOR

CYPHOPHTHALMI

Based on the results reported above, we providea new classification system for Cyphophthalmi,introducing three new infraorders: Scopulophthalminew clade, Sternophthalmi new clade, and Boreo-phthalmi new clade (Table 6). Scopulophthalmiis diagnosed as Pettalidae, and the name refers tothe presence of a scopula in the anal region ofthe male in many pettalid species. Sternophthalmiincludes the families Troglosironidae, Ogoveidae, andNeogoveidae, with its etymology referring to the pres-ence of an exocrine gland opening in the opisthosomal


Pettalus sp. DNA101287Pettalus sp. DNA101286Pettalus sp. DNA101283Pettalus sp. DNA101288 Pettalus sp. DNA101285Pettalus thwaitesi DNA101223

Aoraki sp. DNA101126Aoraki healyi DNA100940Aoraki inerma DNA100966Aoraki crypta DNA101289Aoraki calcarobtusa westlandica DNA101129 Aoraki granulosa DNA101841 Aoraki tumidata DOC094

Aoraki denticulata major DNA100959 Aoraki denticulata denticulata DNA100955




Parapurcellia monticola DNA100386Parapurcellia silvicola DNA100385Purcellia illustrans DNA100387


Karripurcellia harveyi DNA101303Neopurcellia salmoni DNA100939

Rakaia sp. DNA101293Rakaia media DNA101292Rakaia solitaria DNA101294


Rakaia sp. DNA100954Rakaia sp. DNA101297







Rakaia sorenseni digitata DNA100970Rakaia stewartiensis DNA100944Rakaia lindsayi DNA101128

Rakaia minutissima DNA101291Rakaia florensis DNA101295

65

JURASSIC CRETACEOUS TERTIARY

144206

pr = 0.938

pr = 0.898

pr = 0.050

pr = 0.609

pr = 0.154pr = 0.054

pr = 0.871

pr = 0.053

pr = 0.909

pr = 0.056

pr = 0.679pr = 0.140pr = 0.106

pr = 0.601pr = 0.151pr = 0.078

pr = 0.565

pr = 0.310pr = 0.108

pr = 0.720

pr = 0.152pr = 0.119

South Africa

Chile

Eastern Australia

Western Australia

New Zealand, Australian plate

New Zealand, Pacific plate

Sri Lanka

pr = 0.536pr = 0.168pr = 0.143

pr = 0.637pr = 0.187

pr = 0.443pr = 0.167pr = 0.137pr = 0.051

pr = 0.355

pr = 0.236pr = 0.112pr = 0.061

pr = 0.291

pr = 0.194pr = 0.207

pr = 0.098pr = 0.088

Figure 6. Ancestral range reconstructions for Pettalidae inferred by Lagrange analysis, using stratified models. Colouredsquares at terminals indicate ranges occupied by sampled species. Coloured squares on nodes indicate ranges recon-structed for hypothetical ancestors. Numbers on nodes indicate relative probability of ranges reconstructed.



sternal region of males in all troglosironids, allogoveids, and most neogoveids, as opposed to theother three families where the opisthosomal exocrineglands, when present, open in the posterior tergites.We maintain Shear’s superfamily Ogoveoidea, andrestrict Sironoidea to the family Sironidae and Stylo-celloidea to the family Stylocellidae, although we donot introduce other superfamilies because they wouldeach contain a single family. Boreophthalmi includesthe families Stylocellidae and Sironidae, which sub-sequent to Hansen & Sørensen’s (1904), had beenconsidered the representatives of the two maincyphophthalmid clades (Shear, 1980). The term refersto the mostly northern hemisphere distribution ofthese two families, although the origin of Stylocel-lidae can be probably traced to northern Gondwana(Clouse & Giribet, 2010). Sternophthalmi is sistergroup to Boreophthalmi.

The following taxa are thus abandoned as a resultof being non-monophyletic according to our phyloge-netic results: Infraorder Tropicophthalmi Shear, 1980and Infraorder Temperophthalmi Shear, 1980.Shear’s infraorders do not reflect the phylogeneticrelationships obtained here, as suggested in previousstudies (Giribet & Boyer, 2002; Boyer et al., 2007b;Giribet et al., 2010).

DISTRIBUTION MODELLING AND HABITAT

SUITABILITY OVERLAP

Habitat suitability models predicted by both theMAXENT and BIOCLIM methods were highly con-gruent and therefore we present only results fromMAXENT (Fig. 10) because it was found to outper-form other modelling algorithms (Elith et al., 2006).Model predictions were significantly distinct fromrandom and area under the curve (AUC) values werehigh or moderately high (in the range 0.84–0.99) inall runs independently of the modelling algorithm andthe set of variables used to build the model. For theanalyses with a reduced number of variables, we keptall variables that had jackknife regularized traininggain greater than one. As expected, when correlationamong variables is present, using a lower number ofvariables does not change significantly the AUCvalues but reduces over-fitting; hence, the resultingmodels find broader areas with suitable conditions.These are, however, congruent with results frommodels built with all BIOCLIM variables and differ-ences are generally associated with areas wherehabitat suitability is low (Fig. 10).

The variables with highest average relative contri-bution to the MAXENT habitat suitability model for



Troglosiro monteithi DNA101580 Troglosiro juberthiei DNA100344 Troglosiro urbanus DNA101710Troglosiro longifossa DNA100867 Ogovea cameroonensis DNA104617


Metasiro americanus DNA105645Metasiro americanus DNA101532Metasiro americanus DNA105644Canga renatae DNA105680

Parogovia sp. DNA101057Parogovia sp. DNA101052

Parogovia sp. DNA104618Parogovia gabonica DNA104620Parogovia sp. DNA104615

Parogovia sironoides DNA101059Parogovia sironoides DNA101061




Metagovea sp. DNA104648Metagovea sp. DNA104647Metagovea sp. DNA101409







Huitaca ventralis DNA101674Huitaca sp. DNA101407


Huitaca sp. DNA101683Huitaca sp. DNA104646Huitaca sp. DNA102150Huitaca sp. DNA101681


Neogovea sp. DNA105825“Brasilogovea” sp. Tobogan DNA100869Brasilogovea sp. DNA101665

65

PERMIAN TRIASSIC JURASSIC CRETACEOUS TERTIARY

144206248290

New Caledonia

West Africa

SE USA

Amazonia

pr = 0.993pr = 0.969

pr = 1.000

pr = 1.000

pr = 1.000

pr = 1.000

pr = 1.000

pr = 0.832pr = 0.152

pr = 0.986

pr = 0.696pr = 0.148pr = 0.088

pr = 0.206pr = 0.662

pr = 0.077

pr = 0.996

pr = 0.787pr = 0.109pr = 0.058

Figure 7. Ancestral range reconstructions for Sternophthalmi (Troglosironidae, Ogoveidae, Neogoveidae) inferred byLagrange analysis, using stratified models. Coloured squares at terminals indicate ranges occupied by sampled species.Coloured squares on nodes indicate ranges reconstructed for hypothetical ancestors. Numbers on nodes indicate relativeprobability of ranges reconstructed.



Pettalidae were isothermality (33.8%), precipitationof the driest month (22.8%) and annual mean tem-perature (9.2%). Jackknife tests of variable impor-tance indicate that temperature seasonality had thehighest gain in isolation. Mean diurnal temperaturerange decreased the gain the most when omitted,suggesting that it contained the most information not

present in the other variables. For Sironidae, theprecipitation of the coldest quarter (40.0%), meantemperature of the coldest quarter (16.3%), andannual mean temperature were the variables withhighest contribution. Mean temperature of thecoldest quarter had the highest gain in isolation, andannual precipitation decreased the gain the most


Parasiro minor DNA103535Parasiro coiffaiti DNA101383

Cyphophthalmus ognjenovici DNA101039Cyphophthalmus minutus DNA100493Cyphophthalmus trebinjanus DNA101038Cyphophthalmus teyrovskyi DNA100910Cyphophthalmus gordani DNA100495Cyphophthalmus martensi DNA100494








Paramiopsalis ramulosus DNA103538Paramiopsalis ramulosus DNA100459Paramiopsalis sp. DNA104624Paramiopsalis eduardoi DNA101878

Siro shasta DNA101622Siro kamiakensis DNA101613Siro cf. kamiakensis DNA101611Siro exilis DNA100489Siro boyerae DNA101614




Leptopsalis sp. DNA101944, DNA101945Leptopsalis lydekkeri DNA101064


Leptopsalis sp. DNA102042Leptopsalis sp. DNA102039Leptopsalis sp. DNA102032








Miopsalis sp. DNA102032, DNA102053, DNA102058Miopsalis sp. DNA101519


Miopsalis sp. DNA103254Miopsalis sp. DNA104981

Meghalaya sp. DNA101094 500Meghalaya sp. DNA101494, DNA101506, DNA101765Meghalaya sp. DNA101767

Meghalaya sp. DNA103265Meghalaya sp. DNA102051Meghalaya sp. DNA103242, DNA103243, DNA103244


Fangensis insulanus DNA100388Fangensis spelaeus DNA100669

65

CARBONIFEROUS PERMIAN TRIASSIC JURASSIC CRETACEOUS TERTIARY

144206248290354

Thai-Malay Peninsula

Eastern Himalayas

Borneo/Philippine plate

Indo-Malay Archipelago

North America

Western Europe

Mediterranean plate

Balkan Peninsula

Iberian Peninsula

Japan

pr = 0.940

pr = 0.971

pr = 0.943

pr = 0.154

pr = 0.767

pr = 0.619

pr = 0.193pr = 0.069

pr = 0.736

pr = 0.081

pr = 0.330pr = 0.222pr = 0.153

pr = 0.736

pr = 0.609 pr = 0.980

pr = 0.974

pr = 0.963

pr = 0.736

pr = 0.962

pr = 0.996

pr = 0.981

pr = 0.975

pr = 0.941

pr = 0.139pr = 0.722

pr = 0.727pr = 0.074

pr = 0.354pr = 0.178pr = 0.125

Figure 8. Ancestral range reconstructions for Boreophthalmi (Sironidae, Stylocellidae) inferred by Lagrange analysis,using stratified models. Coloured squares at terminals indicate ranges occupied by sampled species. Coloured squares onnodes indicate ranges reconstructed for hypothetical ancestors. Numbers on nodes indicate relative probability of rangesreconstructed.



Leptopsalis beccariiLeptopsalis sp. DNA101483Miopsalis pulicariaLeptopsalis lydekkeri DNA101064Leptopsalis sp. DNA101930Leptopsalis sp. DNA101932Leptopsalis sp. DNA101944, DNA101945Leptopsalis novaguinea DNA101510Leptopsalis sp. DNA101937Leptopsalis sp. DNA101489Leptopsalis sp. DNA102042Leptopsalis sp. DNA101093, DNA101496Leptopsalis sp. DNA101938Leptopsalis sp. DNA102032Leptopsalis sp. DNA102039Leptopsalis sp. DNA102061Leptopsalis sp. DNA103250Leptopsalis sp. DNA102033, DNA102048Meghalaya annandaleiLeptopsalis sp. DNA101514Miopsalis sp. DNA103259Miopsalis sp. DNA103249Miopsalis sp. DNA103254Miopsalis sp. DNA104981Miopsalis sp. DNA101517Stylocellus sumatranusMiopsalis sp. DNA101468, DNA101950Miopsalis sp. DNA102032, DNA102053, DNA102058Miopsalis sp. DNA101519Meghalaya sp. DNA101094, DNA101500Meghalaya sp. DNA101494, DNA101506, DNA101765Meghalaya sp. DNA101767Meghalaya sp. DNA102051Meghalaya sp. DNA103242, DNA103243, DNA103244Meghalaya sp. DNA103265Meghalaya sp. DNA103251Fangensis insulanus DNA100388Fangensis spelaeus DNA100669Suzukielus sauteri DNA101550Suzukielus sauteri DNA101543

DNA101383Parasiro minor DNA103535Odontosiro lusitanicusCyphophthalmus minutus DNA100493Cyphophthalmus ognjenovici DNA101039Cyphophthalmus trebinjanus DNA101038Cyphophthalmus teyrovskyi DNA100910Cyphophthalmus gordani DNA100495Cyphophthalmus martensi DNA100494Cyphophthalmus corfuanus DNA102111Cyphophthalmus zetae DNA100907Cyphophthalmus hlavaci DNA102099Cyphophthalmus markoi DNA100497Cyphophthalmus serbicus DNA102098Cyphophthalmus ere DNA100499Cyphophthalmus duricorius DNA100487Cyphophthalmus rumijae DNA100492Cyphophthalmus gjorgjevici DNA100498Paramiopsalis ramulosus DNA103538Paramiopsalis ramulosus DNA100459Paramiopsalis eduardoi DNA101878Paramiopsalis sp. DNA104624Iberosiro distylosMarwe coarctataSiro boyerae DNA101614Siro exilis DNA100489Siro cf. kamiakensis DNA101611Siro kamiakensis DNA101613Siro calaveras DNA101623Siro shasta DNA101622Siro acaroides DNA100488Siro rubens DNA100457Siro valleorum DNA100461Shearogovea mexascaAnkaratra franziMetagovea sp. DNA104648Metagovea sp. DNA104647Metagovea sp. DNA101686Metagovea sp. DNA101670Metagovea sp. DNA101409Metagovea sp. DNA101642Metagovea sp. DNA105826Metagovea sp. DNA101680Metagovea sp. DNA102151Metagovea sp. DNA101685Metagovea sp. DNA101410Metagovea sp. DNA101654Metagovea sp. DNA101408Huitaca sp. DNA101681Huitaca sp. DNA102150Huitaca sp. DNA101683Huitaca sp. DNA104646Huitaca ventralis DNA101674Huitaca sp. DNA101671Huitaca sp. DNA101407Neogovea sp. DNA105825Brasilogovea microphagaNeogovea kamakusaNeogovea virginie DNA104823Neogovea virginie DNA105824“Brasilogovea” sp. Tobogan DNA100869Brasilogovea sp. DNA101665Neogovea immsiNeogovea kartaboMetasiro americanus DNA105644Metasiro americanus DNA101532Metasiro americanus DNA105645Canga renatae DNA105680Parogovia cf. sironoides DNA104619Parogovia cf. sironoides DNA100462Parogovia cf. sironoides DNA101056Parogovia cf. sironoides DNA101053Parogovia sironoides DNA101061Parogovia sironoides DNA101059Parogovia sp. DNA104615Parogovia sp. DNA104618Parogovia gabonica DNA104620Parogovia sp. DNA101052Parogovia sp. DNA101057Parogovia pabsgarnoniParogovia sp. DNA105671Ogovea cameroonensis DNA104617Ogovea nasutaTroglosiro longifossa DNA100867Troglosiro urbanus DNA101710Troglosiro juberthiei DNA100344Troglosiro monteithi DNA101580Troglosiro wilsoni DNA102324Troglosiro ninqua DNA100577Troglosiro aelleni DNA100345Rakaia sp. DNA101293Rakaia solitaria DNA101294Rakaia media DNA101292Rakaia magna australis DNA100962Rakaia sp. DNA101297Rakaia sp. DNA100954Rakaia dorothea DNA100943Rakaia uniloca DNA101812Rakaia antipodiana DNA100957Rakaia pauli DNA100968Rakaia macra DNA101808Rakaia sp. DNA100958Rakaia sorenseni digitata DNA100970Rakaia sp. DNA101807Rakaia sorenseni sorenseni DNA100969

DNA101295Rakaia minutissima DNA101291Rakaia lindsayi DNA101128Rakaia stewartiensis DNA100944Aoraki calcarobtusa westlandica DNA101129Aoraki granulosa DNA101841Aoraki tumidata DOC094Aoraki sp. DNA101126Aoraki denticulata denticulata DNA100955Aoraki longitarsa DNA101806Aoraki denticulata major DNA100959Aoraki crypta DNA101289Aoraki inerma DNA100966Aoraki healyi DNA100940Austropurcellia daviesae DNA100947Austropurcellia scoparia DNA100946Austropurcellia arcticosa DNA100951Austropurcellia forsteri DNA100945Parapurcellia monticola DNA100386Parapurcellia silvicola DNA100385Neopurcellia salmoni DNA100939Chileogovea oedipus DNA100413Chileogovea sp. DNA100490Karripurcellia harveyi DNA101303Manangotria taolanaroPurcellia illustrans DNA100387Speleosiro argasiformisPettalus sp. DNA101285Pettalus sp. DNA101282Pettalus thwaitesi DNA101223Pettalus sp. DNA101288Pettalus sp. DNA101286Pettalus sp. DNA101287Pettalus sp. DNA101283

61

67

98

96

64

10098

6293

8383

87

7487

97

8571

91

5467

83

8396

88

83 99

5094

9568

67

5093

6453

99

56

86

80

85 67

6788

93

56

90 98

87

77

74

63

72

50

95

95

7887

8198

6867

5796

9764

6364

100

100

100

8954

10093

83

87

9051

8367

85 9285

100

10098

9852

6799

10074

Troglosironidae

Ogoveidae

Pettalidae

Neogoveidae

Sironidae

Stylocellidae

imla

hthp

olup

ocS .

O.Ii

mlaht

hpon

retS

.O.I

iml a

h thp

oero

B .O.I

Figure 9. Combined analysis of morphology and molecules. Strict consensus of three optimal trees based on theparsimony direct optimization analysis under parameter set 3221 (56 984 weighted steps). Clade colours correspond tothose in Fig. 2; taxa in red are represented by morphology only. Numbers above nodes indicate jackknife support values.



when omitted. The variables with highest contribu-tion for Stylocellidae were temperature seasonality(42.9%), annual precipitation (26.3%), and precipita-tion of the warmest quarter (11.8%). Annual pre-cipitation had the highest gain in isolation, andprecipitation of the warmest quarter decreased thegain the most when omitted. For the clade Sternoph-thalmi, the variables with the highest contributionwere annual precipitation (33.8%), precipitation ofthe driest quarter (24.1%), and isothermality (13.9%).Temperature annual range had the highest gainin isolation and also reduced gain the most whenomitted.

Results of the identity test for the MAXENT modelsbased on all variables are shown in Table 5. Resultsfrom the analysis using the Bioclim algorithm arecongruent (not shown). The identity test shows that,when considering relative ranks, the calculatedhabitat suitability scores for most of the groups aresignificantly distinct except for Stylocellidae versusSternophthalmi, the two tropical clades. Habitat suit-ability identity cannot be rejected either for Pettal-idae versus Stylocellidae at the 0.01% significancelevel. The higher values for I and D in those two casesalso show that there is significant overlap of thesuitable habitat of these clades. Pettalidae versusSternophthalmi shows also high values of I and D butidentity tests reject the null hypothesis of habitatsuitability identity.

DISCUSSION

The present data, of worldwide scope, and spanningthe geological scale from the Palaeozoic to the

present, allow us to study a group of soil arthropodsto a level of detail rarely seen in biogeographicaland phylogenetic studies. Taxonomic representationin the molecular data includes species from all non-monotypic genera and several monotypic genera; allgenera are represented in the morphological dataset. Geographical coverage includes all known worldregions where Cyphophthalmi have been reported,with the exception of Mexico (a few specimens fromtwo caves; Shear, 1977, 1980), Madagascar (fourspecimens known in total for two species; Shear &Gruber, 1996), Kenya (five specimens known from asingle cave; Shear, 1985), and the Philippine islandof Palawan (a single adult specimen known; Shear,1993c).

Our phylogenetic results provide the basis for anew classification of the suborder Cyphophthalmi.The results also set the geological time framework forthe origin and diversification of each family and theevolution of the niche preference in selected familiesor suprafamilial clades. This allows testing specificbiogeographical hypotheses, such as the supposedtotal submersion of New Caledonia (Murienne et al.,2005) or New Zealand (Goldberg, Trewick & Paterson,2008), or the reconstruction of the ancestral areas ofeach cyphophthalmid lineage.

DIRECT OPTIMIZATION ANALYSIS

To a lesser degree than for static homology, dynamichomology searches are difficult to evaluate in terms ofoptimality. In the present study, we used a strategy ofSATF with multiple rounds of analyses to decidewhen to stop the searches, and saw that searches of6–12 h run on a desktop computer stabilized after fiveto ten rounds, depending on the parameter set. Thestability of the results is used here as a criterion for

Table 5. Habitat suitability overlap statistics based onthe MAXENT analysis with all bioclim variables

Clade Relative rank I D

Pettalidae versusSironidae

0.727** 0.392 0.187

Pettalidae versusStylocellidae

0.826* (P = 0.015) 0.525 0.234

Pettalidae versusSternophthalmi

0.832** 0.638 0.329

Sironidae versusStylocellidae

0.666** 0.126 0.038

Sironidae versusSternophthalmi

0.768** 0.292 0.114

Stylocellidae versusSternophthalmi

0.836 (P = 0.133) 0.819 0.554

Relative rank significance calculated using ENMtoolsidentity test results. *0.01 < P < 0.05, **P < 0.01. I, the Istatistic (Warren et al., 2008); D, Schoener’s D (Schoener,1968).

Table 6. Classification system for Cyphophthalmi, usingestablished family and superfamily names

Suborder CyphophthalmiInfraorder Scopulophthalmi new clade

Family Pettalidae Shear, 1980Infraorder Sternophthalmi new clade

Family Troglosironidae Shear, 1993Superfamily Ogoveoidea Shear, 1980

Family Ogoveidae Shear, 1980Family Neogoveidae Shear, 1980

Infraorder Boreophthalmi new cladeSuperfamily Stylocelloidea Hansen & Sørensen,

1904 new compositionFamily Stylocellidae Hansen & Sørensen, 1904

Superfamily Sironoidea Simon, 1879 newcomposition

Family Sironidae Simon, 1879



reporting results, in the same fashion that drivensearches and similar techniques have been applied tothe computational problem of tree searching (Giribet,2007a; Goloboff, Farris & Nixon, 2008).

In previous studies of cyphophthalmid and harvest-men data, analyses based on direct optimization haveyielded results sometimes differing from those ofanalyses based on static homology (Boyer et al.,2007b; Giribet et al., 2010). However, this is not thecase in the present study, where taxon sampling andgeographical representation have been thoroughlyoptimized. One major difference remains, the mono-phyly of Sironidae (see below), although some ofthe static homology analyses (Fig. 5) are congruentwith the direct optimization tree (Fig. 3), whereas themaximum likelihood tree (Fig. 4) differs from theBayesian estimate (Fig. 5). Another major difference(but, again, among analyses, and not necessarily theresult of differences between dynamic and staticnotions of homology) is the internal resolution of thepettalid genera (see below).

SYSTEMATICS

The monophyly of Cyphophthalmi has been well sup-ported in all morphological analyses (Giribet & Boyer,2002), as well as from the earliest molecular analysesusing just a few sequences in the families Sironidaeand Stylocellidae (Giribet et al., 1999; Shultz &Regier, 2001; Giribet et al., 2002), a few representa-tives of the suborder (Giribet & Boyer, 2002) or, morerecently, in several much denser analyses (Boyeret al., 2007b; Giribet et al., 2010). Our new data addcorroboration to this well-delimited taxon, with thefamilial inter-relationships and their internal struc-ture being the real focus of the study, although, in thecombined analysis including taxa with morphologicaldata only, support for the monophyly of Cyphoph-thalmi decreases to 61%, probably as a result of someeffects of the missing data (see below). Among these,Pettalidae, Stylocellidae, Troglosironidae, andNeogoveidae are monophyletic in most of our analyses(but see discussion on Neogoveidae), Ogoveidae isrepresented by a single specimen in the molecularanalyses, and Sironidae remains contentious, espe-cially with respect to the placement of the two generaSuzukielus and Parasiro.

One of the outstanding issues in cyphophthalmidphylogenetics has been the placement of the root,which was suggested to occur: (1) between Stylocel-lidae and the remaining families; (2) between Pettal-idae and the remaining families; or (3) between aclade containing Suzukielus and Pettalidae and theremaining families (Giribet & Boyer, 2002), based onthe molecular rooting of a morphological tree, becausemost cyphophthalmid morphological characters are

inapplicable or meaningless outside the suborder.Subsequent analyses found alternative resolutionsplacing the root between Pettalidae and the rest orbetween Stylocellidae and the rest (Boyer et al.,2007b), depending on the analysis and optimalitycriterion employed. Different studies have assumedeither one of these alternative rootings until a recentbroader Opiliones study found the root between Pet-talidae and the rest (Sternophthalmi + Boreoph-thalmi), this time without distinction amongoptimality criteria or method of analysis (Giribetet al., 2010). This latter result is further corroboratedin the present study. This position of Pettalidae assister group to all other cyphophthalmid families fal-sifies the two infraorders introduced by Shear (1980),which should be abandoned, and allows for a muchclearer reconstruction of the cyphophthalmid ances-tor, which must have had laterally positioned simpleocelli (Alberti, Lipke & Giribet, 2008), a lamelliformadenostyle in the male fourth tarsus, and opisthoso-mal exocrine glands opening in the anal region in themale.

Internal resolution of Pettalidae does not differconsiderably from the source studies of this pettaliddata set (Boyer & Giribet, 2007, 2009) and, as inthese studies, South Africa, New Zealand and Aus-tralia are not monophyletic. Diversification of thefamily started 183 Mya, and therefore paralogy ofsome of its landmasses is easily explained by clado-genesis prior to the split of Gondwana into itscurrent continents. Nonetheless, relationships withinPettalidae remain unstable or poorly supported andimportant African diversity is missing from themolecular sampling, both from South Africa (deBivort & Giribet, 2010) and Madagascar (Shear &Gruber, 1996), although the combined analyses withmorphology place Speleosiro as sister group to Pur-cellia (64% JF) (Giribet, 2003a; de Bivort et al., 2010;de Bivort & Giribet, 2010), and Manangotria assister group to Karripurcellia, although with lownodal support.

Results within Pettalidae are congruent amongmethods of analysis in the monophyly of each genus,although their relationships remain contentious. Atrans-Tasman clade is found, albeit with low support,in the probabilistic analyses but not in the directoptimization analysis. Similarly, the most-basal posi-tion of Parapurcellia is not universally accepted.Other relationships discussed above are poorly sup-ported, with the exception of a Chileogovea + Purcel-lia clade. Whether the deficient sampling in SouthAfrica (whose genera appear to have influence at thebase of the tree) or perhaps lineage extinction duringthe cooling of Antarctica are responsible for the lackof resolution in the pettalid relationships, remainsuntested.



The present study introduces the first geneticdata for the monogeneric family Ogoveidae, whichclearly forms part of the previously establishedTroglosironidae–Neogoveidae clade (Boyer et al.,2007b; Sharma & Giribet, 2009a), now namedSternophthalmi. Ogoveidae forms a clade withNeogoveidae in all analyses, corroborating Shear’ssuperfamily Ogoveoidea, although not his infraorderTropicophthalmi, because Stylocellidae are unrelatedto Ogoveoidea. Ogoveoidea is thus a Pantropical cladeof probable African origin, although its original diver-sification dates back to 261 Mya. Some analyses (twosuboptimal parameter sets under direct optimization)place Ogoveidae as ingroup Neogoveidae, althoughmost analyses support monophyly of Neogoveidae.This is also found in the combined analysis withmorphology, where Ogovea and Parogovia form aclade of African Ogoveoidea, although support for thisclade is low. The latter clade is sister to a clade ofAmerican neogoveids. However, because of the uniquemorphology of ogoveids (Juberthie, 1969; Giribet &Prieto, 2003), and monophyly of Neogoveidae in mostanalyses, the family Ogoveidae is maintained as valid(after rediagnosis from Giribet & Prieto, 2003). Shear(1980) included the genus Huitaca in this family,although, earlier, he had postulated a sister grouprelationship of Huitaca and Metagovea (Shear,1979a), as shown in our analyses.

Neogoveidae began its own diversification soonafter (236 Mya), long before the opening of the Atlan-tic Ocean, as illustrated by the amphi-Atlantic claderelating the Eastern Brazilian genus Canga with theAfrican Parogovia (specimens from Cameroon, Gabon,and Equatorial Guinea), or the sister group relation-ships of the North American genus Metasiro to theAmazonian/West African clade. The specimen fromIvory Coast, probably related to P. pabsgarnoni, con-stitutes a new genus that will be described elsewhere.Other than Canga, the South American species forma well supported clade that we currently assign tofour genera: Brasilogovea, which we resurrect here,includes species from Amazonia and the ‘Tepuis’region of Colombia; Neogovea, represented by twospecies from Guyana and French Guiana; Huitaca,still endemic to Colombia, including a large number ofundescribed species; and Metagovea, including notonly most specimens from the Andean region, but alsosome Amazon specimens from Leticia and a specimenfrom Guyana, with the latter being sister to all otherMetagovea and possibly constituting another newgenus (L. Benavides & G. Giribet, unpubl. data). Thisspecies is clearly unrelated to the genus Neogovea,occurring in this part of the Neotropics, and it ischaracterized by a conspicuous opisthosomal mid-dorsal longitudinal sulcus; an adenostyle ending in abrush of setae and located at the base or towards the

centre of the dorsal side of tarsus IV; absence ofopisthosomal exocrine glands; and a spermatopositorcomplex with a crown-shaped structure at the tip,with additional perpendicular projections (L. Bena-vides & G. Giribet, unpubl. data). Our Metagoveaclade includes specimens that we previously placed inthe genera Metagovea and Neogovea (Boyer et al.,2007b; Giribet et al., 2010) because they differ consid-erably in their anatomy. Further subdivision ofMetagovea may be warranted, although not untilspecimens from the Manaus area (Brazil) are avail-able for molecular study. Nonetheless, relationshipsamong the four genera are well established, withBrasilogovea + Neogovea being the sister group to aclade including Huitaca and Metagovea, and thelatter genus generally divided among small species,or ‘typical’ Metagovea and larger species more similarto Neogovea. The large sampling within the super-family, including all the currently recognized genera,and new data for many mostly undescribed speciesand genera not represented in previous studies,allows us to provide a more comprehensive under-standing of this Pantropical group. The addition ofmorphological data of the types of the generaNeogovea and Brasilogovea did not fully resolve thisclade, although this is considered to be a result of thepoor preservation of these specimens (missing theventral opisthosomal region) that does not allowexamination of key characters such as the sternum orthe opisthosomal exocrine glands.

The sister group of Ogoveoidea is without doubt theNew Caledonia endemic genus Troglosiro, and bothseparated approximately 279 Mya at a time whenNew Caledonia was geographically located at theeastern margin of Gondwana. Diversification of Tro-glosironidae is, however, much more recent (57 Mya)and the error associated with this date does not allowfor an unambiguous interpretation of the postulatedtotal submersion of the island (Grandcolas et al.,2008; Murienne et al., 2008; Murienne, 2009). Theanalyses recognize a group with sternal opisthosomaldepressions associated with the sternal exocrineglands of the males, sensu Sharma & Giribet (2009a).

Stylocellidae have gone from being the most poorly-known group to arguably the most stable and bestunderstood phylogenetically. The results of thepresent study corroborate those from the recentstudies mostly by R. Clouse (Clouse & Giribet, 2007;Clouse et al., 2009; Clouse, 2010; Clouse & Giribet,2010; Clouse et al., 2011), from which all the dataincluded here were derived; see also Schwendinger &Giribet (2005). Fangensis, a clade with its origins inthe terrane that today constitutes the Thai-Malaypeninsula, is sister to all other stylocellids, whichdiversified towards the north in the genus Meghalayaand towards the south in the genera Miopsalis and



Leptopsalis, the former mostly found on Borneo,although giving rise to species in the Philippines andSumatra, and the latter radiating rapidly asSumatra, Java, and Sulawesi became accessible, andalso extending to New Guinea. The family diversified167 Mya and includes the only reported cases ofpossible transoceanic dispersal in Cyphophthalmi(Clouse & Giribet, 2007; see also Clouse et al., 2011).Morphological analysis of discrete character data alsosupport earlier studies excluding the nominal genusStylocellus from any of the four genera adopted here,with Stylocellus remaining monotypic (Clouse et al.,2009). However, the addition of the type species of thegenera Leptopsalis, Meghalaya, Miopsalis, and Stylo-cellus, not available for molecular analysis, does notresult in a well-resolved taxonomy. The problem,however, lies in the nature of the data because thetype of Stylocellus, an old, pinned, deformed speci-men, is difficult to position based on discrete-characters only (Clouse et al., 2009), whereas the typeof Miopsalis is a female and thus misses most discretecharacters coded for other specimens, and was notavailable to be included in the morphometric analysesof Clouse et al. (2009). Leptopsalis is, however,well placed within its supposed molecular clade,although the type of Meghalaya does not find goodsupport.

Sironidae monophyly has been disputed in previousanalyses based on morphology and molecules, wheretwo genera, Parasiro and Suzukielus, often do notcluster with the remaining sironids (in the generaSiro, Cyphophthalmus, Paramiopsalis, and Iberosiro)(Giribet & Boyer, 2002; de Bivort & Giribet, 2004;Boyer et al., 2007b; Giribet et al., 2010). Odontosiro,never included in a molecular analysis, is sister toParasiro outside of the typical sironids. The KenyanMarwe has been placed within sironids in some analy-ses based on morphological data (de Bivort & Giribet,2004), as also shown here, and appears related toParamiopsalis and Iberosiro. Sironids found theirmonophyly, however, in a recent morphological analy-sis of continuous characters (de Bivort et al., 2010)and in the maximum likelihood analysis of Giribetet al. (2010), as does the maximum likelihood analysisof the present data set, albeit with bootstrap supportbelow 50%. Monophyly of Sironidae is also foundin the analysis of the nuclear ribosomal data underdirect optimization, suggesting that the non-monophyly of the family may be an artefact intro-duced most probably by their unusual COI evolution(Boyer et al., 2005). However, the membership inSironidae of the genera Iberosiro, Odontosiro, or evenMarwe remains untested with molecular data andfuture sampling effort in the north-western IberianPeninsula and in Kenya should focus on these highlycontroversial genera.

MORPHOLOGICAL DATA

It is well known that combined analyses of moleculesand morphology are fundamental for understandingthe systematics of groups that include many taxa forwhich molecular data cannot be obtained (Eernisse &Kluge, 1993; Nixon & Carpenter, 1996). A typicalexample of the latter case is provided by fossil taxa(Giribet, 2010; Murienne, Edgecombe & Giribet,2010a; Pyron, 2011). When dealing with such taxa,missing data can become a concern, although it hasbeen shown, both with simulation and with empiricalresults, that missing data in and of themselves arenot always problematic; instead, it is informationcontent in the data at hand what really matters(Wiens, 2003; Goloboff et al., 2009; Hejnol et al., 2009;Wiens, 2009).

The fossil record of Cyphophthalmi is scarce(Dunlop & Giribet, 2003; Poinar, 2008; Dunlop &Mitov, 2011), and it does not add important diversitythat can be coded into an explicit data matrix.However, the problem of missing data is of greatimportance in Cyphophthalmi as a result of thegroup’s almost global but highly localized distribu-tion, making collecting an arduous task. For thesereasons, many species are known only from oldmuseum material, from a single male, or even fromonly females or juveniles, making it impossible toinclude them in the molecular matrix or presentingconsiderable amounts of missing data in our morpho-logical matrix. To mention just a few examples, thetype species of the genera Miopsalis and Ogovea areknown only from female individuals (Thorell, 1890–1891; Hansen & Sørensen, 1904); the second speciesin the genus Pettalus, Pettalus brevicauda Pocock1897, was based on a juvenile specimen (Giribet,2008); Stylocellus sumatranus, currently the onlyspecies in the genus, is based on a deformed specimenin very poor condition (Clouse et al., 2009); andseveral monotypic genera have never been examinedunder a scanning electron microscope: Ankaratra,Manangotria, Marwe, and Odontosiro.

When trying to maximize the diversity representedin the present study, we included all currently recog-nized genera in our morphological matrix, althoughsome of the species representing these genera aremissing important characters. Despite this problem,we followed earlier recommendations into a combinedanalysis in POY under the optimal parameter set andsubmitted it to a jackknife analysis. The resultingtree was not too different from that of an early analy-sis of all cyphophthalmid genera (Giribet & Boyer,2002) with respect to the lack of resolution for manyclades, which is otherwise well supported by themolecular data sets. Notable results are the place-ment of Managotria taolanaro within Pettalidae (63%



jackknife support), the monophyly of Stylocellidae(including the types of the genera Leptopsalis, Megha-laya, Miopsalis, and Stylocellus, despite the lack ofmolecular data; 63% jackknife support) or the mono-phyly of the Neotropical Neogoveidae (exceptingCanga), including the types and morphology-onlyspecies of Neogovea [N. immsi, Neogovea kartabo(Davis, 1937)], Neogovea kamakusa Shear, 1977), andBrasilogovea microphaga Martens, 1969 (56% jack-knife support). The inclusion of Parogovia pabsgar-noni affects the monophyly of the African neogoveids,as Parogovia sp. DNA105671 does not form a cladewith the other Parogovia (54% jackknife support).However, the instability of species such asShearogovea mexasca and Ankaratra franzi affectsthe monophyly of groups that are otherwise robustto molecular analysis such as Sternophthalmi +Boreophthalmi, Sternophthalmi, Ogoveoidea, Neogo-veidae or Boreophthalmi.

The current results combining morphology withmolecules lowered overall support for the tree. This isan unfortunate result because simulations haveshown that the accuracy in the phylogenetic place-ment of fossils often improves or stays the same whenusing molecular data and that only in a few casesaccuracy was significantly decreased (Wiens, 2009).The problem here may be related to the low numbersof discrete morphological characters available forthese Opiliones, often limited to variation amonggroups of species; hence, the recent use of continuouscharacters in some analyses of the group (Clouseet al., 2009; de Bivort et al., 2010; de Bivort & Giribet,2010). Some characters show low levels of homoplasy,greatly structuring the data (some of these characterswere the basis for older classification systems) butmany of our wild taxa show ‘unexpected’ states inthese characters. Character 7 is notable in thisrespect. The coxae of the walking legs of Cyphoph-thalmi show different degrees of fusion, with coxae IIIand IV of each side always fused and coxae I remain-ing moveable. Coxa II can be free (state 0) or fused tocoxae III and IV; among the former are most membersof the families Pettalidae, Troglosironidae, andSironidae (except for the genera Paramiopsalis andIberosiro); among the latter are the members of thefamilies Ogoveidae, Stylocellidae, and Neogoveidae(except for Canga and Metasiro). It is therefore notunexpected that this character defined the majorgroups Sironoidea and Stylocelloidea sensu Hansen &Sørensen (1904), and that a genus such as Metasirowas considered a member of Sironidae in previousstudies, nor that these genera are among the mostunstable ones when morphology is used. Thepresence/absence of eyes (character 1), ozophore type(character 2), and spiracle shape (character 49) arealso characters with relatively low levels of

homoplasy, which have played an important role incyphophthalmid systematics, and, again, it is notunexpeced that the taxa that present odd characterstates become unstable.

BIOGEOGRAPHICAL PATTERNS IN CONTINENTS

Cyphophthalmi have been shown to present a highcorrelation of their systematic position and landmassaffinity (Juberthie & Massoud, 1976; Shear, 1980;Giribet, 2000), to show strong genetic structure acrossshort distances (Boyer et al., 2007a), and have beenused as models to study vicariance biogeography(Giribet, 2003a; Boyer et al., 2005; Boyer & Giribet,2007; Giribet & Kury, 2007; Boyer et al., 2007b; Boyer& Giribet, 2009; Clouse et al., 2009; Sharma &Giribet, 2009a; Clouse, 2010; Clouse & Giribet, 2010;de Bivort & Giribet, 2010; Murienne et al., 2010b;Clouse et al., 2011). This study corroborates earlierfindings that suggest a temperate Gondwanan clade(Pettalidae; Fig. 6), a Pantropical clade (Sternoph-thalmi; Fig. 7), one clade originating in the Thai-Malay Peninsula (Stylocellidae; Fig. 8), and two ormore Laurasian clades whose ancestral area is diffi-cult to reconstruct with high probability but thatincludes the Iberian Peninsula, North America, andWestern Europe (Sironidae; Fig. 8). The origin of allthese clades is ancient, preceding the fragmentationof Pangea and therefore suggesting that many clado-genetic events were older than the vicariant eventsthat followed. This has important biogeographicalimplications with respect to using vicariant events ascalibration points because the mismatch between thetwo events could be very large (Kodandaramaiah,2011).

Most biogeographical patterns observed, in conjunc-tion with a well-dated phylogenetic hypothesis and areconstruction of the ancestral landmasses for eachclade, allow a thorough explanation of each clade. Theancestral area reconstruction of the family Pettalidae(Fig. 6) involves several cladogenetic events at thegenus-level because each genus is currently recog-nized to be restricted to a single landmass or toadjacent terranes (Boyer & Giribet, 2007). Althoughresolution among the genera finds low support, mostanalyses suggest the South African genus Parapur-cellia to be the sister group to all other genera, andplace the other South African genus, Purcellia, in thatclade, lending support to South Africa as one of thepossible centres of origin of the family. A relationshipbetween South Africa (Purcellia) and South America(Chileogovea) is found in most analyses, as is alsofound in the members of the peripatopsid Ony-chophora, with similar distribution and habitatrequirements as pettalids (Allwood et al., 2010). It isalso notable that the two Australian genera Austro-



purcellia (Queensland) and Karripurcellia (WesternAustralia) never form a clade, supporting earlierviews about using microareas in biogeographicalstudies of small soil organisms (Giribet & Edgecombe,2006). A relationship of Sri Lanka–Australia–NewZealand is found in several analyses.

The biogeographical patterns of Sternophthalmi(Fig. 7) are easily reconstructed, with two ancestrallineages occurring in the Neotropics (Canga is sisterto the African Parogovia clade), two ancestral lin-eages in Africa (Ogovea and Parogovia), and onelineage in North America (Metasiro), which separatedfrom the remaining neogoveids during the Triassic.Although older analyses suggested a relationship ofMetasiro to Parogovia, this was based on analyses

without several neogoveid lineages and withoutogoveids, and the current results are very stable. Thesister group relationship of Ogoveoidea to the NewCaledonian endemic genus Troglosiro has been foundin previous studies and it is discussed in more detailbelow. No paralogy is needed in this tree when con-sidering the timing of the diversification events,as the separation of the Neotropics from the Afrotro-pics is dated at 95 Mya (Raven & Axelrod, 1972;Sanmartín, 2002).

Stylocellid biogeography and their ancestral areashave been discussed recently (Clouse & Giribet, 2010)and our results corroborate this earlier analysis.The Thai-Malay Peninsula is reconstructed as theancestral terrane for the family with subsequent

A

B

C

D

Figure 10. MAXENT models of habitat suitability. Left column with all bioclim variables, right column with variableswith jackknife regularized training gain greater than one. Warmer (red-yellow) colours represent more suitable habitats.Maps in miniature represent actual presence observations. A, Pettalidae; B, Sironidae; C, Stylocellidae; D, Sternophthalmi(Troglosironidae + Ogoveidae + Neogoveidae).



expansions to the Eastern Himalayas during theCretaceous/Tertiary boundary, and radiations into theBorneo/Philippine plate and into the Indo-MalayArchipelago during the Cretaceous. Several lineagesmay have returned to the Thai-Malay Peninsula ormoved between islands around the Cretaceous/Tertiary boundary during a period in which south-east Asia was subjected to drastic changes and theIndo-Malay Archipelago variously connected (Hall,2002; Ali & Aitchison, 2008).

Reconstruction of the biogeographical history ofSironidae remains hindered by the instability of therelationships of Suzukielus and Parasiro; the formerendemic to Japan and the latter found in the Iberianand Mediterranean plates. Parasiro has its origins inthe Jurassic/Cretaceous of the Iberian Peninsula,where cyphophthalmids are so far restricted to areaswith Paleozoic rocks (Murienne & Giribet, 2009). Themain sironid clade includes three genera found inNorth America (Siro), Western Europe (Siro), theIberian Peninsula (Paramiopsalis), and the Balkanregion and Eastern Europe (Cyphophthalmus). Siroshows reciprocal monophyly of the two landmasses,the lineages separating during the Triassic, a resultnot supported in a recent analysis of the North Ameri-can diversity (Giribet & Shear, 2010). The sister-group relationship of the Iberian/Balkan clade hasbeen discussed thoroughly in recent studies (Boyeret al., 2005; Murienne et al., 2010b), which have alsoillustrated a correlation between an evolutionaryexplosion and the coming into contact of ancestrallandmasses in the Mediterranean region (Murienneet al., 2010b). From a geological point of view, theBalkan Peninsula, supporting the explosive evolutionof Cyphophthalmus, includes the margin of bothEurasia (the Moesian microplate) and Gondwana (theAdria microplate), as well as remnants of the Tethysand related marginal seas (made up of oceanic crust)(Karamata, 2006). The Adria microplate is the largestlithospheric fragment in the Central Mediterraneanregion. It was connected to Iberia in the west and tonorth-west Africa in the south (Wortmann et al.,2001) until the Middle–Late Triassic episodes ofrifting and breakup (Channell, D’Argenio & Horváth,1979; Pamic, Gusic & Jelaska, 1998), around thecladogenesis time for the split between Paramiopsalisand Cyphophthalmus. For most of the time, the Adriamicroplate was in a shallow-water environment(Scheibner & Speijer, 2008) in which the SouthernTethyan Megaplatform formed before disintegratinginto several carbonate platforms in the Early Jurassic(Vlahovic et al., 2005), before the diversification ofCyphophthalmus during the Jurassic/Cretaceous,when cycles of land submergence and emergence havebeen recorded in some carbonate platforms (Vlahovicet al., 2005; Márton et al., 2008). The ancestral area of

the family is however difficult to infer, perhaps,amongst other factors, as a result of the large numberof terranes that existed around the Tethys.

BIOGEOGRAPHY IN CONTINENTAL ISLANDS: THE

CASES OF NEW CALEDONIA AND NEW ZEALAND

Cyphophthalmi are present in most islands of conti-nental origin (fragment islands sensu Gillespie &Roderick, 2002), including Sri Lanka (Pocock, 1897;Sharma & Giribet, 2006; Giribet, 2008; Sharma,Karunarathna & Giribet, 2009), Chiloé (Roewer,1961; Juberthie & Muñoz-Cuevas, 1970; Shear,1993a), Corsica and Sardinia (Simon, 1872; Juber-thie, 1958), Honshu (Roewer, 1916; Juberthie, 1970b;Suzuki & Ohrui, 1972; Giribet, Tsurusaki & Boyer,2006), and the Indo-Malay archipelago (Westwood,1874; Thorell, 1882–1883; Pocock, 1897; Hansen &Sørensen, 1904; Shear, 1979b; Rambla, 1991; Shear,1993c; Giribet, 2002; Schwendinger & Giribet, 2005;Clouse & Giribet, 2007; Clouse et al., 2009; Clouse &Giribet, 2010), and, in all these cases, their presencein these islands is best explained as a result of vicari-ance. Similarly, New Caledonia and New Zealandhost a considerable diversity of Cyphophthalmi,although their presence in these islands as a resultof one or more vicariant evens has been recentlydisputed.

New Caledonia currently has 13 described speciesin the genus Troglosiro, the only genus in the familyTroglosironidae (Juberthie, 1979; Shear, 1993b;Sharma & Giribet, 2005, 2009a; Sharma & Giribet,2009b) considered to be endemic to the Grande Terreand unambiguously recovered as the sister group tothe Equatorial Ogoveoidea from Equatorial WestAfrica and the Equatorial Neotropical belt. Geologicaldata on the origins of the New Caledonian biodiver-sity argue in favour of a series of submersions duringthe Palaeocene and Eocene (Paris, Andreieff &Coudray, 1979; Aitchison et al., 1998; Pelletier, 2006),which has been used to support a total submersion ofthe island, re-emerging 37 Mya (Murienne et al.,2005; Grandcolas et al., 2008; Murienne et al., 2008).Indeed, molecular dating analyses of several NewCaledonian clades has supported diversification pro-cesses post-dating the critical date of 37 Mya (Muri-enne et al., 2005; Page et al., 2005; Murienne et al.,2008; Espeland & Johanson, 2010; Murienne, Edge-combe & Giribet, 2011), which has led some studies tosuggest that the entirety of the New Caledonian ter-restrial biota must have arrived to the islands viadispersal and that no trace of ancient vicariance isleft. One notable exception may be the family Troglo-sironidae, whose diversification has been dated at28–49 Mya by Boyer et al. (2007b) and 52–102 Mya byGiribet et al. (2010), although these studies used few



troglosironid samples. Refined analyses here suggesta Late Cretaceous–Early Tertiary diversificationof the family (57 Mya), predating the supposedre-emergence of New Caledonia, although the errorassociated with this date does not allow unambiguousdistinction of the hypothesis owing to the temporalproximity of the re-emergence of the island (37 Mya)and the floor of the diversification age estimate (95%HPD: 40–73 Mya). The ancestral area reconstructionfor the split between Troglosironidae and Ogoveoideais supported as a contiguous landmass containingWest Africa and New Caledonia, deep in the Permian,indicating that the range of the clade was muchbroader than it currently is (Figs 5, 7), and thatmassive extinctions may have occurred during theperiod comprised between 279–57 Mya. However,relict taxa (and Troglosironidae certainly is such anexample) and the problem of extinction, especially inthe absence of a fossil record, are mysteries that aredifficult to address in biogeography (Crisp, Trewick &Cook, 2011). This is indeed a unique case, whereTroglosironidae constitute a special lineage in thisrespect.

Another possibility is a trans-Pacific dispersal,again during the 279–57 Mya period, a phenomenonalso observed in at least two other opilionid lineages(e.g. the families Zalmoxidae and Samoidae; Sharma& Giribet, 2011). However, dispersals in the Cenozoicare possible to reconstruct unambiguously in Zal-moxidae and Samoidae insofar as lineages in one partof the Pacific form a grade with respect to a clade inanother part of the Pacific. In both these cases, Neo-tropical lineages form the paraphyletic grade withrespect to Pacific island lineages, rendering the ances-tral area reconstruction for the origin of these radia-tions as Neotropical. By contrast, Troglosironidae andthe clade (Ogoveidae + Neogoveidae) form reciprocallymonophyletic groups that diverged 279 Mya, which isinconsistent with recent dispersal. Moreover, thePermian origin of Troglosironidae also suggests thatany putative dispersal event had to have occurredsometime between 279–57 Mya, a hypothesis that isdifficult to test. As stated previously, we submit thatthe biogeographical history of Troglosironidae isinherently difficult to reconstruct as a result of therelictual nature of this lineage (Sharma & Giribet,2009a).

A similar case has been proposed for New Zealand,which includes 29 species in three pettalid genera(Aoraki, Neopurcellia, and Rakaia) (Forster, 1948,1952; Boyer & Giribet, 2003, 2007) found in twogeological terranes (the Australian plate and thePacific plate) (Boyer & Giribet, 2009). New Zealand’sgeology and biota reflect a dynamic history of ancientGondwanan origin, long-term isolation from othercontinental landmasses, marine inundating during

the Oligocene, glaciation during the Pleistocene, andevolutionary radiations that have produced a spec-tacular proportion of endemic species (Gibbs, 2006).Studies have focused on New Zealand’s biogeographywith particular vigour over the past two decadesbecause molecular systematics has provided newtools with which to approach evolutionary questions.Molecular systematists have addressed topics suchas the number and location of Pleistocene refugia(Marske et al., 2009; Buckley, Marske & Attanayake,2010), the Alpine Fault Hypothesis (Heads & Craw,2004), and, most contentiously, a vicariance versusdispersal-based origin of New Zealand’s terrestrialbiota (Trewick, Paterson & Campbell, 2007; Phillipset al., 2010). Although studies have long recognizedthat land area was drastically reduced (i.e. to lessthan 15% of its current size) during the marineincursions of the Oligocene (Cooper & Cooper, 1995),more recently Waters & Craw (2006) have suggestedthat there is no strong evidence for continuouslyemergent land throughout the period (Landis et al.,2008). Trewick et al. (2007) and Wallis & Trewick(2009) asserted that the preponderance of biogeo-graphical evidence favours a scenario of completesubmergence during the Oligocene, and some studieshave gone further, suggesting that the entire terres-trial biota arrived via dispersal during the last 22Myr (Landis et al., 2008), and that it is thereforemore like that of an oceanic archipelago than a con-tinent (Goldberg et al., 2008). Few have questionedthis new trend in New Zealand biogeography (Knappet al., 2007; Edgecombe & Giribet, 2008; Boyer &Giribet, 2009; Allwood et al., 2010; Giribet & Boyer,2010).

The evolutionary history of the New Zealandcyphophthalmid genera (Aoraki, Rakaia, and Neopur-cellia) has long been of interest as part of the evalu-ation of a hypothesis proposing total submersion ofNew Zealand in the Oligocene (Waters & Craw, 2006;Trewick et al., 2007; Goldberg et al., 2008; Landiset al., 2008; Wallis & Trewick, 2009; Giribet & Boyer,2010; Phillips et al., 2010). The persistence of theselineages through the Oligocene bottleneck was con-sidered to represent evidence of incomplete submer-sion of this landmass (Boyer & Giribet, 2007, 2009),although this hypothesis was not previously accom-panied by molecular dating. Consequently, the agesof diversification of these lineages have been opento interpretation as very young (e.g. approximately5 Mya; Goldberg et al., 2008). Moreover, Crisp et al.(2011) suggested that an important criterion for evi-dence of vicariance events is diversification time coin-cident with the timing of the geological event thatprecipitated the vicariance; in this case, the rifting ofZealandia from the Australian plate approximately85 Mya, although cladogenesis could be expected to be



much older than the vicariant event in taxa with lowvagility and small distribution ranges.

The present study, utilizing a robust methodologyfor simultaneous estimation of tree topology and cladedivergence times (sensu Crisp et al., 2011), and cali-brated using fossil taxa exclusively (Kodandara-maiah, 2011), obtains the following diversificationtimes for the New Zealand endemic genera Rakaiaand Aoraki: 91 Mya (95% HPD: 72–108 Mya) and90 Mya (95% HPD: 75–108 Mya), respectively. Thesediversification age estimates coincide with the riftingof Zealandia in the Late Cretaceous. We present evi-dence, therefore, based upon tree topology and cladedivergence times, of the persistence of multiple lin-eages through the Oligocene. In addition, the presentstudy reconstructs the origin of the genus Aoraki tothe Australian plate during the Cretaceous and thatof the genus Rakaia to a composite terrane in thePacific and Australian plate also in the middle of theCretaceous, although, later on, our study clearlyassigns a clade to each terrane (Figs 5, 6). Unfortu-nately, the generic relationships are highly unstableacross methods and parameter sets and furtherspeculation about the relationships of the Australianand New Zealand genera awaits further data. Wesubmit that these data falsify the hypothesis of com-plete submersion of New Zealand during the Oli-gocene Drowning. Recent and forthcoming studies ofother invertebrate lineages (Allwood et al., 2010;Giribet & Boyer, 2010; Murienne et al., 2010a;Marshall, 2011) are anticipated to corroborate thisconclusion.

HABITAT SUITABILITY MODELS

One common characteristic of all models that wepresent here is the larger suitable habitat than thearea actually occupied by the four clades of interest.This constitutes further evidence for the old cladogen-esis and low dispersal abilities of Cyphophthalmibecause many areas of suitable habitat have neverbeen in contact with a landmass occupied by the cladeof interest. This pattern also corroborates the hypoth-esis that tectonic movements and vicariance eventshave defined distributions and driven diversificationin this group of soil arthropods. Mysteries remainbecause certain temperate clades have migrated towarmer climates (e.g. Pettalus in Sri Lanka orAustropurcellia in Queensland, Australia), whereasothers may not have been able to adapt to changingclimates. This suggests that, in several lineages, pro-cesses of niche evolution might have taken place.However, the lack of detailed occurrence observationsfor many species does not currently allow studyingthe niche evolution in Cyphophthalmi in greaterdetail.

CONCLUDING REMARKS

Cyphophthalmi constitute an ancient lineage of Opil-iones distributed in temperate to tropical rainforestsworldwide but restricted for the most part to conti-nents and islands of continental origin, representingan ideal group of organisms for studying vicariancebiogeography. Both phylogenetic patterns derivedfrom molecular and morphological data and moleculardating using Opiliones fossils as calibration pointscorroborate the old age of the group and of its con-stituent clades. Ancestral area reconstruction furthercorroborates our biogeographical predictions byrequiring only minimal switches between land-masses, most of them through contiguous land, there-fore showing that the actual distribution is muchmore restricted than the potential distributiondefined by the modelled habitat suitability for thedifferent familial/suprafamilial clades. The data alsopermit tests of more general biogeographical hypoth-eses, such as the total submersion of New Caledoniaand New Zealand and, at least in the former case,contradict a scenario of complete inundation. Thepresent study provides refinement not only ofthe phylogenetic relationships and taxonomy of thegroup, but also its evolutionary and biogeographicalhistory.

ACKNOWLEDGEMENTS

This work is the result of more than a decade ofintense research on Cyphophthalmi and many indi-viduals and institutions have contributed to it. Firstand foremost, this work has been possible in largepart as a result of grants from the National ScienceFoundation (DEB-0236871 to G.G., DEB-0508789 toS.B. and G.G., and DEB-0205982 to M. Sharkey) andto numerous Putnam Expeditions Grants from theMuseum of Comparative Zoology to collect in Austra-lia, Cameroon, Gabon, Indonesia, New Zealand,South Africa, and Sri Lanka. J.M. was supported by aMarie Curie International Outgoing Fellowship(grant 221099) within the 7th European CommunityFramework Program. Many colleagues have assistedus in the field, and they are acknowledged in ourprevious papers. We want to emphasize, however, thehelp of Carlos Prieto for work in Equatorial Guinea,Indika Karunarathna for work in Sri Lanka, CahyoRahmadi for work in Indonesia, Nobuo Tsurusaki forwork in Japan, Phil Sirvid and Ricardo Palma forwork in New Zealand, Hervé Jourdan for work inNew Caledonia, Nono Legrand for work in Cameroon,and many other colleagues who provided samples,especially Louis Deharverg for south-east Asiansamples, Mike Sharkey for his Colombian samples,Rudy Jocqué for samples from Ivory Coast and



Guyana, Eduardo Mateos for Spanish samples, andRicardo Pinto-da-Rocha for Brazilian samples. GregEdgecombe, Salvador Carranza, and Michele Nishigu-chi accompanied us on many collecting trips. We arealso indebted to many museums and curators fortheir long-term loans that made this research pos-sible, especially Lorenzo Prendini and Norman Plat-nick from the American Museum of Natural History,New York, NY; Charles Griswold from the CaliforniaAcademy of Sciences, San Francisco, CA; Petra Sier-wald from the Field Museum, Chicago, IL; Jan Bec-caloni from The Natural History Museum, London;Arturo Muñoz Cuevas from the Muséum nationald’Histoire naturelle, Paris; Phil Sirvid from Te PapaTongarewa, Wellington; Jonathan Coddington fromthe US National Museum (Smithsonian Institution),Washington, DC; Nikolaj Scharff from the ZoologicalMuseum, Natural History Museum of Denmark,Copenhagen; and Jason Dunlop from the Zoologis-chen Museums, Berlin. Finally, Laura Leibenspergerhas assisted with specimen loans during all theseyears. The editor John A. Allen, an anonymousreviewer, and Gustavo Hormiga provided commentsthat helped to improve upon an earlier version of thismanuscript.

REFERENCES

Aitchison JC, Ireland TR, Clarke GL, Cluzel D, DavisAM, Meffre S. 1998. Regional implifications of U/PbSHRIMP age constraints on the tectonic evolution of NewCaledonia. Tectonophysics 299: 333–343.

Alberti G. 1995. Comparative spermatology of Chelicerata:review and perspective. In: Jamieson BGM, Ausió J, JustineJ-L, eds. Advances in spermatozoal phylogeny and tax-onomy. Mémoires du Muséum National d’Histoire Naturelle166: 203–230.

Alberti G. 2005. Double spermatogenesis in Chelicerata.Journal of Morphology 266: 281–297.

Alberti G, Giribet G, Gutjahr M. 2009. Ultrastructure ofspermatozoa of different species of Neogoveidae, Sironidae,and Stylocellidae (Cyphophthalmi: Opiliones). Contributionsto Natural History – Scientific papers from the NaturalHistory Museum Bern 12: 53–69.

Alberti G, Lipke E, Giribet G. 2008. On the ultrastructureand identity of the eyes of Cyphophthalmi based on a studyof Stylocellus sp. (Opiliones, Stylocellidae). Journal ofArachnology 36: 379–387.

Ali JR, Aitchison JC. 2008. Gondwana to Asia: plate tec-tonics, paleogeography and the biological connectivity of theIndian sub-continent from the Middle Jurassic throughlatest Eocene (166–35 Ma). Earth-Science Reviews 88: 145–166.

Allwood J, Gleeson D, Mayer G, Daniels S, Beggs JR,Buckley TR. 2010. Support for vicariant origins of the NewZealand Onychophora. Journal of Biogeography 37: 669–681.

Araujo MB, Rahbek C. 2006. How does climate changeaffect biodiversity? Science 313: 1396–1397.

Benavides LR, Giribet G. 2007. An illustrated catalogue ofthe South American species of the cyphophthalmid familyNeogoveidae (Arthropoda, Opiliones, Cyphophthalmi) witha report on 37 undescribed species. Zootaxa 1509: 1–15.

de Bivort B, Clouse RM, Giribet G. 2010. Amorphometrics-based phylogeny of the temperate Gond-wanan mite harvestmen (Opiliones, Cyphophthalmi, Pettal-idae). Journal of Zoological Systematics and EvolutionaryResearch 48: 294–309.

de Bivort B, Giribet G. 2010. A systematic revision of theSouth African Pettalidae (Arachnida : Opiliones : Cyphoph-thalmi) based on a combined analysis of discrete and con-tinuous morphological characters with the description ofseven new species. Invertebrate Systematics 24: 371–406.

de Bivort BL, Giribet G. 2004. A new genus of cyphoph-thalmid from the Iberian Peninsula with a phylogeneticanalysis of the Sironidae (Arachnida : Opiliones : Cyphoph-thalmi) and a SEM database of external morphology. Inver-tebrate Systematics 18: 7–52.

Boyer SL, Baker JM, Giribet G. 2007a. Deep geneticdivergences in Aoraki denticulata (Arachnida, Opiliones,Cyphophthalmi): a widespread ‘mite harvestman’ defiesDNA taxonomy. Molecular Ecology 16: 4999–5016.

Boyer SL, Clouse RM, Benavides LR, Sharma P, Sch-wendinger PJ, Karunarathna I, Giribet G. 2007b. Bio-geography of the world: a case study from cyphophthalmidOpiliones, a globally distributed group of arachnids. Journalof Biogeography 34: 2070–2085.

Boyer SL, Giribet G. 2003. A new Rakaia species (Opiliones,Cyphophthalmi, Pettalidae) from Otago, New Zealand.Zootaxa 133: 1–14.

Boyer SL, Giribet G. 2007. A new model Gondwanan taxon:systematics and biogeography of the harvestman familyPettalidae (Arachnida, Opiliones, Cyphophthalmi), with ataxonomic revision of genera from Australia and NewZealand. Cladistics: The International Journal of the WilliHennig Society 23: 337–361.

Boyer SL, Giribet G. 2009. Welcome back New Zealand:regional biogeography and Gondwanan origin of threeendemic genera of mite harvestmen (Arachnida, Opiliones,Cyphophthalmi). Journal of Biogeography 36: 1084–1099.

Boyer SL, Karaman I, Giribet G. 2005. The genusCyphophthalmus (Arachnida, Opiliones, Cyphophthalmi) inEurope: a phylogenetic approach to Balkan Peninsulabiogeography. Molecular Phylogenetics and Evolution 36:554–567.

Buckley TR, Marske K, Attanayake D. 2010. Phylogeog-raphy and ecological niche modelling of the New Zealandstick insect Clitarchus hookeri (White) support survival inmultiple coastal refugia. Journal of Biogeography 37: 682–695.

Castresana J. 2000. Selection of conserved blocks from mul-tiple alignments for their use in phylogenetic analysis.Molecular Biology and Evolution 17: 540–552.



Channell JET, D’Argenio B, Horváth F. 1979. Adria, theAfrican Promontory, in Mesozoic Mediterranean Paleogeog-raphy. Earth Science Reviews 15: 213–292.

Clouse RM. 2010. Molecular and morphometric phylogenet-ics and biogeography of a Southeast Asian arachnid family(Opiliones, Cyphophthalmi, Stylocellidae). PhD Thesis,Department of Organismic and Evolutionary Biology,Harvard University, Cambridge.

Clouse RM, de Bivort BL, Giribet G. 2009. A phylogeneticanalysis for the South-east Asian mite harvestman familyStylocellidae (Opiliones : Cyphophthalmi) – a combinedanalysis using morphometric and molecular data. Inverte-brate Systematics 23: 515–529.

Clouse RM, General DM, Diesmos AC, Giribet G.2011. An old lineage of Cyphophthalmi (Opiliones) discov-ered on Mindanao highlights the need for biogeographicalresearch in the Philippines. Journal of Arachnology 39:147–153.

Clouse RM, Giribet G. 2007. Across Lydekker’s Line – firstreport of mite harvestmen (Opiliones : Cyphophthalmi :Stylocellidae) from New Guinea. Invertebrate Systematics21: 207–227.

Clouse RM, Giribet G. 2010. When Thailand was an island– the phylogeny and biogeography of mite harvestmen(Opiliones, Cyphophthalmi, Stylocellidae) in Southeast Asia.Journal of Biogeography 37: 1114–1130.

Cooper A, Cooper RA. 1995. The Oligocene bottleneck andNew Zealand biota: genetic record of a past environmentalcrisis. Proceedings of the Royal Society of London Series B,Biological Sciences 261: 293–302.

Crisp MD, Trewick SA, Cook LG. 2011. Hypothesistesting in biogeography. Trends in Ecology & Evolution 26:66–72.

DaSilva MB, Pinto-da-Rocha R, Giribet G. 2010. Cangarenatae, a new genus and species of Cyphophthalmi fromBrazilian Amazon caves (Opiliones: Neogoveidae). Zootaxa2508: 45–55.

De Laet J. 2010. A problem in POY tree searches (and itswork-around) when some sequences are observed to beabsent in some terminals. Cladistics: The InternationalJournal of the Willi Hennig Society 26: 453–455.

Diniz JAF, De Marco P, Hawkins BA. 2010. Defying thecurse of ignorance: perspectives in insect macroecology andconservation biogeography. Insect Conservation and Diver-sity 3: 172–179.

Drummond AJ, Ho SY, Phillips MJ, Rambaut A. 2006.Relaxed phylogenetics and dating with confidence. PLoSBiology 4: e88.

Drummond AJ, Rambaut A. 2007. BEAST: Bayesian evo-lutionary analysis by sampling trees. BMC EvolutionaryBiology 7: 214.

Dunlop JA. 2007. Paleontology. In: Pinto-da-Rocha R,Machado G, Giribet G, eds. Harvestmen: the biology ofOpiliones. Cambridge, MA: Harvard University Press, 247–265.

Dunlop JA, Anderson LI, Kerp H, Hass H. 2003.Preserved organs of Devonian harvestmen. Nature 425:916.

Dunlop JA, Anderson LI, Kerp H, Hass H. 2004. for 2003).A harvestman (Arachnida: Opiliones) from the Early Devo-nian Rhynie cherts, Aberdeenshire, Scotland. Transactionsof the Royal Society of Edinburgh: Earth Sciences 94: 341–354.

Dunlop JA, Giribet G. 2003. The first fossil cyphophthalmid(Arachnida: Opiliones), from Bitterfeld amber, Germany.Journal of Arachnology 31: 371–378.

Dunlop JA, Mitov PG. 2011. The first fossil cyphophthalmidharvestman from Baltic amber. Arachnologische Mitteilun-gen 40: 47–54.

Edgar RC. 2004. MUSCLE: multiple sequence alignmentwith high accuracy and high throughput. Nucleic AcidsResearch 32: 1792–1797.

Edgecombe GD, Giribet G. 2008. A New Zealand species ofthe trans-Tasman centipede order Craterostigmomorpha(Arthropoda : Chilopoda) corroborated by molecular evi-dence. Invertebrate Systematics 22: 1–15.

Eernisse DJ, Kluge AG. 1993. Taxonomic congruence versustotal evidence, and amniote phylogeny inferred from fossils,molecules, and morphology. Molecular Biology and Evolu-tion 10: 1170–1195.

Elith J, Graham CH, Anderson RP, Dudík M, Ferrier S,Guisan A, Hijmans RJ, Huettmann F, Leathwick JR,Lehmann A, Li J, Lohmann LG, Loiselle BA, ManionG, Moritz C, Nakamura M, Nakazawa Y, Overton JM,Peterson AT, Phillips SJ, Richardson K, Scachetti-Pereira R, Schapire RE, Soberón J, Williams S, WiszMS, Zimmermannm NE. 2006. Novel methods improveprediction of species’ distributions from occurrence data.Ecography 29: 129–151.

Espeland M, Johanson KA. 2010. The effect of environmen-tal diversification on species diversification in New Cale-donian caddisflies (Insecta: Trichoptera: Hydropsychidae).Journal of Biogeography 37: 879–890.

Evans ME, Smith SA, Flynn RS, Donoghue MJ. 2009.Climate, niche evolution, and diversification of the ‘bird-cage’ evening primroses (Oenothera, sections Anogra andKleinia). American Naturalist 173: 225–240.

Farris JS, Albert VA, Källersjö M, Lipscomb D, KlugeAG. 1996. Parsimony jackknifing outperforms neighbor-joining. Cladistics: The International Journal of the WilliHennig Society 12: 99–124.

Forster RR. 1948. The sub-order Cyphophthalmi Simon inNew Zealand. Dominion Museum Records in Entomology 1:79–119.

Forster RR. 1952. Supplement to the sub-order Cyphoph-thalmi. Dominion Museum Records in Entomology 1: 179–211.

Gibbs G. 2006. Ghosts of Gondwana. The history of life inNew Zealand. Nelson: Craig Potton Publishing.

Gillespie RG, Roderick GK. 2002. Arthropods on islands:colonization, speciation, and conservation. Annual Review ofEntomology 47: 595–632.

Giribet G. 2000. Catalogue of the Cyphophthalmi of theWorld (Arachnida, Opiliones). Revista Ibérica de Arac-nología 2: 49–76.

Giribet G. 2002. Stylocellus ramblae, a new stylocellid



(Opiliones, Cyphophthalmi) from Singapore, with a discus-sion of the family Stylocellidae. Journal of Arachnology 30:1–9.

Giribet G. 2003a. Karripurcellia, a new pettalid genus(Arachnida : Opiliones : Cyphophthalmi) from Western Aus-tralia, with a cladistic analysis of the family Pettalidae.Invertebrate Systematics 17: 387–406.

Giribet G. 2003b. Stability in phylogenetic formulations andits relationship to nodal support. Systematic Biology 52:554–564.

Giribet G. 2007a. Efficient tree searches with available algo-rithms. Evolutionary Bioinformatics 3: 1–16.

Giribet G. 2007b. Neogoveidae Shear, 1980. In: Pinto-da-Rocha R, Machado G, Giribet G, eds. Harvestmen: thebiology of Opiliones. Cambridge, MA: Harvard UniversityPress, 95–97.

Giribet G. 2008. On the identity of Pettalus cimiciformis andP. brevicauda (Opiliones, Pettalidae) from Sri Lanka.Journal of Arachnology 36: 199–201.

Giribet G. 2010. A new dimension in combining data? Theuse of morphology and phylogenomic data in metazoansystematics. Acta Zoologica (Stockholm) 91: 11–19.

Giribet G. 2011. Shearogovea, a new genus of Cyphophthalmi(Arachnida, Opiliones) of uncertain position from Oaxacancaves, Mexico. Breviora 528: 1–7.

Giribet G, Boyer SL. 2002. A cladistic analysis of thecyphophthalmid genera (Opiliones, Cyphophthalmi).Journal of Arachnology 30: 110–128.

Giribet G, Boyer SL. 2010. ‘Moa’s Ark’ or ‘GoodbyeGondwana’: is the origin of New Zealand’s terrestrial inver-tebrate fauna ancient, recent, or both? Invertebrate System-atics 24: 1–8.

Giribet G, Edgecombe GD. 2006. The importance of lookingat small-scale patterns when inferring Gondwanan bio-geography: a case study of the centipede Paralamyctes(Chilopoda, Lithobiomorpha, Henicopidae). BiologicalJournal of the Linnean Society 89: 65–78.

Giribet G, Edgecombe GD, Wheeler WC, Babbitt C. 2002.Phylogeny and systematic position of Opiliones: a combinedanalysis of chelicerate relationships using morphologicaland molecular data. Cladistics: The International Journal ofthe Willi Hennig Society 18: 5–70.

Giribet G, Kury AB. 2007. Phylogeny and biogeography. In:Pinto-da-Rocha R, Machado G, Giribet G, eds. Harvestmen:the biology of Opiliones. Cambridge, MA: Harvard Univer-sity Press, 62–87.

Giribet G, Prieto CE. 2003. A new Afrotropical Ogovea(Opiliones, Cyphophthalmi) from Cameroon, with a discus-sion on the taxonomic characters in the family Ogoveidae.Zootaxa 329: 1–18.

Giribet G, Rambla M, Carranza S, Baguñà J, Riutort M,Ribera C. 1999. Phylogeny of the arachnid order Opiliones(Arthropoda) inferred from a combined approach of complete18S and partial 28S ribosomal DNA sequences and mor-phology. Molecular Phylogenetics and Evolution 11: 296–307.

Giribet G, Shear WA. 2010. The genus Siro Latreille, 1796(Opiliones, Cyphophthalmi, Sironidae), in North America

with a phylogenetic analysis based on molecular data andthe description of four new species. Bulletin of the Museumof Comparative Zoology 160: 1–33.

Giribet G, Tsurusaki N, Boyer SL. 2006. Confirmation ofthe type locality and the distributional range of Suzukielussauteri (Opiliones, Cyphophthalmi) in Japan. Acta Arach-nologica 55: 87–90.

Giribet G, Vogt L, Pérez González A, Sharma P,Kury AB. 2010. A multilocus approach to harvestman(Arachnida: Opiliones) phylogeny with emphasis on bioge-ography and the systematics of Laniatores. Cladistics: TheInternational Journal of the Willi Hennig Society 26: 408–437.

Goldberg J, Trewick SA, Paterson AM. 2008. Evolutionof New Zealand’s terrestrial fauna: a review of mole-cular evidence. Philosophical Transactions of the RoyalSociety of London Series B, Biological Sciences 363: 3319–3334.

Goloboff PA, Catalano SA, Mirande JM, Szumik CA,Arias JS, Källersjö M, Farris JS. 2009. Phylogeneticanalysis of 73 060 taxa corroborates major eukaryoticgroups. Cladistics: The International Journal of the WilliHennig Society 25: 211–230.

Goloboff PA, Farris JS, Nixon KC. 2008. TNT, a freeprogram for phylogenetic analysis. Cladistics: The Interna-tional Journal of the Willi Hennig Society 24: 774–786.

Grandcolas P, Murienne J, Robillard T, DeSutter-Grandcolas L, Jourdan H, Guilbert E. 2008. NewCaledonia: a very old Darwinian island? PhilosophicalTransactions of the Royal Society of London Series B, Bio-logical Sciences 363: 3309–3317.

Hadly EA, Spaeth PA, Li C. 2009. Niche conservatism abovethe species level. Proceedings of the National Academy ofSciences of the United States of America 106: 19707–19714.

Hall R. 2002. Cenozoic geological and plate tectonic evolutionof SE Asia and the SW Pacific: computer-based reconstruc-tions, model and animations. Journal of Asian EarthSciences 20: 353–431.

Hansen HJ, Sørensen W. 1904. On two orders of Arachnida:Opiliones, especially the suborder Cyphophthalmi, andRicinulei, namely the family Cryptostemmatoidae. Cam-bridge: Cambridge University Press.

Heads M, Craw R. 2004. The Alpine Fault biogeographichypothesis revisited. Cladistics: The International Journalof the Willi Hennig Society 20: 184–190.

Heino J, Soininen J. 2007. Are higher taxa adequate sur-rogates for species-level assemblage patterns and speciesrichness in stream organisms? Biological Conservation 137:78–89.

Hejnol A, Obst M, Stamatakis A, Ott M, Rouse GW,Edgecombe GD, Martinez P, Baguñà J, Bailly X, Jon-delius U, Wiens M, Müller WEG, Seaver E, WheelerWC, Martindale MQ, Giribet G, Dunn CW. 2009. Assess-ing the root of bilaterian animals with scalable phyloge-nomic methods. Proceedings of the Royal Society of LondonSeries B, Biological Sciences 276: 4261–4270.

Hijmans RJ, Cameron SE, Parra JL, Jones PG, Jarvis A.2005. Very high resolution interpolated climate surfaces for



global land areas. International Journal of Climatology 25:1965–1978.

Juberthie C. 1956. Une nouvelle espèce d’Opilions Sironidaede France et d’Espagne: Parasiro coiffaiti n. sp. Bulletin duMuséum National d’Histoire Naturelle, 2e série 28: 394–400.

Juberthie C. 1958. Révision du genre Parasiro (Opilions,Sironidae) et descriptions de Parasiro minor n. sp. Bulletindu Muséum National d’Histoire Naturelle, 2e série 30: 159–166.

Juberthie C. 1960a. Contribution a l’étude des opilionscyphophthalmes: description de Metasiro gen. n. Bulletin duMuséum National d’Histoire Naturelle, 2e série 32: 235–241.

Juberthie C. 1960b. Sur la biologie d’un opilion endogé, Sirorubens Latr. (Cyphophthalmes). Comptes rendus des séancesde L’Académie des Sciences 251: 1674–1676.

Juberthie C. 1961. Étude des opilions cyphophthalmes(arachnides) du Portugal: description d’Odontosiro lusitani-cus g. n., sp. n. Bulletin du Muséum National d’HistoireNaturelle, 2e série 33: 512–519.

Juberthie C. 1962. Étude des opilions cyphophthalmes Sty-locellinae du Portugal. Description de Paramiopsalis ramu-losus gen. n., sp. n. Bulletin du Muséum National d’HistoireNaturelle, 2e série 34: 267–275.

Juberthie C. 1969. Sur les opilions cyphophthalmes Stylo-cellinae du Gabon. Biologia Gabonica 5: 79–92.

Juberthie C. 1970a. Les genres d’opilions Sironinae(Cyphophthalmes). Bulletin du Muséum National d’HistoireNaturelle, 2e série 41: 1371–1390.

Juberthie C. 1970b. Sur Suzukielus sauteri (Roewer, 1916)opilion cyphophthalme du Japon. Revue d’Écologie et deBiologie du Sol 7: 563–569.

Juberthie C. 1979. Un cyphophthalme nouveau d’une grottede Nouvelle-Caledonie: Troglosiro aelleni n. gen., n. sp.(Opilion, Sironinae). Revue suisse de Zoologie 86: 221–231.

Juberthie C, Manier JF. 1976. Les grands traits de laspermiogenese chez les opilions Comptes Rendus de laTroisième Réunion des Arachnologistes d’ExpressionFrançaise. Station biologique des Eyzies. Les Eyzies:Académie de Paris, 74–82.

Juberthie C, Manier JF. 1978. Étude ultratructurale com-parée de la spermiogenèse des opilions et son intérêt phylé-tique. Symposium of the Zoological Society of London 42:407–416.

Juberthie C, Manier JF, Boissin L. 1976. Étude ultra-structurale de la double espermiogenèse chez l’opilioncyphophthalme Siro rubens Latreille. Journal deMicroscopie et de Biologie Cellulaire 25: 137–148.

Juberthie C, Massoud Z. 1976. Biogéographie, taxonomie etmorphologie ultrastructurale des opilions cyphophthalmes.Revue d’Écologie et de Biologie du Sol 13: 219–231.

Juberthie C, Muñoz-Cuevas A. 1970. Revision [sic] deChileogovea oedipus Roewer (Opiliones: Cyphophthalmi:Sironinae). Senckenbergiana biologica 51: 109–118.

Karaman IM. 2009. The taxonomical status and diversity ofBalkan sironids (Opiliones, Cyphophthalmi) with descrip-tions of twelve new species. Zoological Journal of theLinnean Society 156: 260–318.

Karamata S. 2006. The geological development of the BalkanPeninsula related to the approach, collision and compres-sion of Gondwanan and Eurasian units. In: Robertson AHF,Mountrakis D, eds. Tectonic development of the EasternMediterranean region. London: Geological Society, 155–178.

Knapp M, Mudaliar R, Havell D, Wagstaff SJ, LockhartPJ. 2007. The drowning of New Zealand and the problem ofAgathis. Systematic Biology 56: 862–870.

Kodandaramaiah U. 2011. Tectonic calibrations in molecu-lar dating. Current Zoology 57: 116–124.

Landis CA, Campbell HJ, Begg JG, Mildenhall DC,Paterson AM, Trewick SA. 2008. The WaipounamuErosion Surface: questioning the antiquity of the NewZealand land surface and terrestrial fauna and flora. Geo-logical Magazine 145: 173–197.

Lawrence RF. 1931. The harvest-spiders (Opiliones) ofSouth Africa. Annals of the South African Museum 29:341–508.

Lawrence RF. 1933. The harvest-spiders (Opiliones) ofNatal. Annals of the Natal Museum 7: 211–241.

Lawrence RF. 1939. A contribution to the opilionid fauna ofNatal and Zululand. Annals of the Natal Museum 9: 225–243.

Lawrence RF. 1963. The Opiliones of the Transvaal. Annalsof the Transvaal Museum 24: 275–304.

Legg G. 1990. Parogovia pabsgarnoni, sp. n. (Arachnida,Opiliones, Cyphophthalmi) from Sierra Leone, with notes onother African species of Parogovia. Bulletin of the BritishArachnological Society 8: 113–121.

Legg G, Pabs-Garnon EB. 1989. The life history of a tropi-cal forest cyphophthalmid from Sierra Leone (Arachnida,Opiliones). In: Haupt J, ed. Xi europäisches arachnologis-ches colloquium. Berlin: Technische Universität Berlin,222–230.

Linton EW. 2005. MacGDE: Genetic Data Environment forMacOS X, Version 2.0. Available at: http://www.msu.edu/~lintone/macgde/

Marshall BA. 2011. A new species of Latia Gray, 1850 (Gas-tropoda: Pulmonata: Hygrophila: Chilinoidea: Latiidae)from Miocene Palaeo-lake Manuherikia, southern NewZealand, and biogeographic implications. MolluscanResearch 31: 47–52.

Marske KA, Leschen RAB, Barker GM, Buckley TR.2009. Phylogeography and ecological niche modelling impli-cate coastal refugia and trans-alpine dispersal of a NewZealand fungus beetle. Molecular Ecology 18: 5126–5142.

Márton E, Cosovic V, Moro A, Zvocak S. 2008. The motionof Adria during the Late Jurassic and Cretaceous: newpaleomagnetic results from stable Istria. Tectonophysics454: 44–53.

Murienne J. 2009. Testing biodiversity hypotheses in NewCaledonia using phylogenetics. Journal of Biogeography 36:1433–1434.

Murienne J, Edgecombe GD, Giribet G. 2010a. Includingsecondary structure, fossils and molecular dating in thecentipede tree of life. Molecular Phylogenetics and Evolution57: 301–313.



Murienne J, Edgecombe GD, Giribet G. 2011. Compara-tive phylogeography of the centipedes Cryptops pictus andC. niuensis in New Caledonia, Fiji and Vanuatu. OrganismsDiversity & Evolution 11: 61–74.

Murienne J, Giribet G. 2009. The Iberian Peninsula:ancient history of a hot spot of mite harvestmen (Arachnida:Opiliones: Cyphophthalmi: Sironidae) diversity. ZoologicalJournal of the Linnean Society 156: 785–800.

Murienne J, Grandcolas P, Piulachs MD, Bellés X,D’Haese C, Legendre F, Pellens R, Guilbert E. 2005.Evolution of a shaky piece of Gondwana: is local endemismrecent in New Caledonia? Cladistics: The InternationalJournal of the Willi Hennig Society 21: 2–7.

Murienne J, Karaman I, Giribet G. 2010b. Explosive evo-lution of an ancient group of Cyphophthalmi (Arachnida:Opiliones) in the Balkan Peninsula. Journal of Biogeogra-phy 37: 90–102.

Murienne J, Pellens R, Budinoff R, Wheeler WC, Grand-colas P. 2008. Phylogenetic analysis of the endemic NewCaledonian cockroach Lauraesilpha. Testing competinghypotheses of diversification. Cladistics: The InternationalJournal of the Willi Hennig Society 24: 802–812.

Nix HA. 1986. A biogeographic analysis of Australian elapidsnakes. In: Longmore R, ed. Australian flora and faunaseries number 7. Atlas of elapid snakes of Australia. Can-berra: Australian Government Publishing Service, 4–15.

Nixon KC, Carpenter JM. 1996. On simultaneous analysis.Cladistics: The International Journal of the Willi HennigSociety 12: 221–241.

Page TJ, Baker AM, Cook BD, Hughes JM. 2005. Histori-cal transoceanic dispersal of a freshwater shrimp: the colo-nization of the South Pacific by the genus Paratya (Atyidae).Journal of Biogeography 32: 581–593.

Pamic J, Gusic I, Jelaska V. 1998. Geodynamic evolution ofthe Central Dinarides. Tectonophysics 297: 251–268.

Paris JP, Andreieff P, Coudray J. 1979. Sur l’âge Eocènesupérieur de la mise en place de la nappe ophiolitique deNouvelle-Calédonie. Comptes rendus de l’Académie des sci-ences, série B 288: 1659–1661.

Pearson RG, Raxworthy CJ, Nakamura M, Peterson AT.2007. Predicting species distributions from small numbersof occurrence records: a test case using cryptic geckos inMadagascar. Journal of Biogeography 34: 102–117.

Pelletier B. 2006. Geology of the New Caledonia region andits implications for the study of the New Caledonian biodi-versity. In: Payri C, Richer de Forges B, eds. Compendiumof marine species from New Caledonia. Doc. Sci. Tech. IRD.IRD, II 7, 17–30.

Phillips MJ, Gibb GC, Crimp EA, Penny D. 2010.Tinamous and moa flock together: mitochondrial genomesequence analysis reveals independent losses of flightamong ratites. Systematic Biology 59: 90–107.

Phillips SJ, Anderson RP, Schapire RE. 2006. Maximumentropy modeling of species geographic distributions. Eco-logical Modelling 190: 231–259.

Phillips SJ, Dudik M. 2008. Modeling of species distribu-tions with Maxent: new extensions and a comprehensiveevaluation. Ecography 31: 161–175.

Pocock RI. 1897. Descriptions of some new Oriental Opil-iones recently received by the British Museum. Annals andMagazine of Natural History, series 6 19: 283–292.

Poinar G. 2008. Palaeosiro burmanicum n. gen., n. sp., afossil Cyphophthalmi (Arachnida: Opiliones: Sironidae) inEarly Cretaceous Burmese amber. In: Makarov SE, Dimi-trijevic RN, eds. Advances in arachnology and developmen-tal biology. Papers dedicated to Prof. Dr. Bozidar Curcic.Vienna, Belgrade, Sofia: Faculty of Life Sciences, Universityof Vienna, and Serbian Academy of Sciences and Arts,267–274.

Posada D. 2005. Modeltest, Version 3.7. Available at: http://darwin.uvigo.es

Posada D, Crandall KA. 1998. MODELTEST: testing themodel of DNA substitution. Bioinformatics (Oxford,England) 14: 817–818.

Pyron RA. 2011. Divergence time estimation using fossils asterminal taxa and the origins of Lissamphibia. SystematicBiology 60: 466–481.

Rambaut A, Drummond AJ. 2007. Tracer, Version 1.5.Available at: http://beast.bio.ed.ac.uk/Tracer

Rambla M. 1974. Consideraciones sobre la biogeografía de losOpiliones de la Península Ibérica. Miscellanea AlcobéSpecial volume: 45–56.

Rambla M. 1991. A new Stylocellus from some caves ofBorneo, Malaysia (Opiliones, Cyphophthalmi, Stylocellidae).Mémoires de Biospéologie 18: 227–232.

Rambla M. 1994. Un nouveau Cyphophthalme du sud-estasiatique, Fangensis leclerci n. gen. n. sp. (Opiliones,Sironidae). Mémoires de Biospéologie 21: 109–114.

Rambla M, Fontarnau R. 1984. Les Opilions Cyphoph-thalmes (Arachnida) de la faune ibérique: I. Sur Paramiop-salis ramulosus Juberthie, 1962. Revue Arachnologique 5:145–152.

Rambla M, Fontarnau R. 1986. Les Opilions Cyphoph-thalmes (Arachnida) de la faune ibérique: III. Sur Odonto-siro lusitanicus Juberthie, 1961. Mémoirs de la Sociétéroyale belge d’Entomologie 33: 171–178.

Raven PH, Axelrod DI. 1972. Plate tectonics and Australa-sian paleobiogeography. Science 176: 1379–1386.

Ree RH, Moore BR, Webb CO, Donoghue MJ. 2005. Alikelihood framework for inferring the evolution of geo-graphic range on phylogenetic trees. Evolution 59: 2299–2311.

Ree RH, Smith SA. 2008. Maximum likelihood inference ofgeographic range evolution by dispersal, local extinction,and cladogenesis. Systematic Biology 57: 4–14.

Roewer CF. 1916. 7 neue Opilioniden des Zoolog. Museumsin Berlin. Archiv für Naturgeschichte 81: 6–13.

Roewer CF. 1961. Opiliones aus Süd-Chile. Senckenbergianabiologica 42: 99–105.

Sanmartín I. 2002. A paleogeographic history of the SouthernHemisphere. Uppsala: Uppsala University.

Sanmartín I, Ronquist F. 2004. Southern hemisphere bio-geography inferred by event-based models: plant versusanimal patterns. Systematic Biology 53: 216–243.

Scheibner C, Speijer RP. 2008. Late Paleocene-earlyEocene Tethyan carbonate platform evolution – a response



to long- and short-term paleoclimatic change. Earth ScienceReviews 90: 71–102.

Schoener TW. 1968. The Anolis lizards of Bimini: resourcepartitioning in a complex fauna. Ecology 49: 704–726.

Schwendinger PJ, Giribet G. 2005. The systematics of thesouth-east Asian genus Fangensis Rambla (Opiliones :Cyphophthalmi : Stylocellidae). Invertebrate Systematics 19:297–323.

Sharma P, Giribet G. 2005. A new Troglosiro species(Opiliones, Cyphophthalmi, Troglosironidae) from NewCaledonia. Zootaxa 1053: 47–60.

Sharma P, Giribet G. 2006. A new Pettalus species (Opil-iones, Cyphophthalmi, Pettalidae) from Sri Lanka with adiscussion on the evolution of eyes in Cyphophthalmi.Journal of Arachnology 34: 331–341.

Sharma P, Giribet G. 2009a. A relict in New Caledonia:phylogenetic relationships of the family Troglosironidae(Opiliones: Cyphophthalmi). Cladistics: The InternationalJournal of the Willi Hennig Society 25: 279–294.

Sharma P, Karunarathna I, Giribet G. 2009. On theendemic Sri Lankan genus Pettalus (Opiliones, Cyphoph-thalmi, Pettalidae) with the description of a new species anda discussion on the magnitude of its diversity. Journal ofArachnology 37: 60–67.

Sharma PP, Giribet G. 2009b. The family Troglosironidae(Opiliones: Cyphophthalmi) of New Caledonia Zoologia Neo-caledonica 7. Biodiversity Studies in New Caledonia. Paris:196, 83–123.

Sharma PP, Giribet G. 2011. The evolutionary and biogeo-graphic history of the armoured harvestmen – Laniatoresphylogeny based on ten molecular markers, with thedescription of two new families of Opiliones (Arachnida).Invertebrate Systematics 25: 106–142.

Sharma PP, Vahtera V, Kawauchi GY, Giribet G. 2011.Running WILD: the case for exploring mixed parameter setsin sensitivity analysis. Cladistics: The InternationalJournal of the Willi Hennig Society 27: 538–549.

Shear WA. 1977. The opilionid genus Neogovea Hinton, witha description of the first troglobitic cyphophthalmid fromthe western hemisphere (Opiliones, Cyphophthalmi).Journal of Arachnology 3: 165–175.

Shear WA. 1979a. Huitaca ventralis, n. gen., n. sp., with adescription of a gland complex new to cyphophthalmids(Opiliones, Cyphophthalmi). Journal of Arachnology 7: 237–243.

Shear WA. 1979b. Stylocellus sedgwicki n.sp., from PenangIsland, Malaysia (Opiliones, Cyphophthalmida, Stylocel-lidae). Bulletin of the British Arachnological Society 4: 356–360.

Shear WA. 1980. A review of the Cyphophthalmi of theUnited States and Mexico, with a proposed reclassificationof the suborder (Arachnida, Opiliones). American MuseumNovitates 2705: 1–34.

Shear WA. 1985. Marwe coarctata, a remarkable newcyphophthalmid from a limestone cave in Kenya (Arach-nida, Opiliones). American Museum Novitates 2830: 1–6.

Shear WA. 1993a. The genus Chileogovea (Opiliones,

Cyphophthalmi, Petallidae [sic]). Journal of Arachnology21: 73–78.

Shear WA. 1993b. The genus Troglosiro and the new familyTroglosironidae (Opiliones, Cyphophthalmi). Journal ofArachnology 21: 81–90.

Shear WA. 1993c. New species in the opilionid genus Stylo-cellus from Malaysia, Indonesia and the Philippines (Opil-iones, Cyphophthalmi, Stylocellidae). Bulletin of the BritishArachnological Society 9: 174–188.

Shear WA, Gruber J. 1996. Cyphophthalmid opilionids newto Madagascar: two new genera (Opiliones, Cyphophthalmi,?Pettalidae). Bulletin of the British Arachnological Society10: 181–186.

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: 189–200.

Simon E. 1872. Notice sur les Arachnides cavernicoles ethypogés. Annales de la Société entomologique de France, 5esérie 2: 215–244.

Smith SA, Donoghue MJ. 2010. Combining historical bio-geography with niche modeling in the Caprifolium clade ofLonicera (Caprifoliaceae, Dipsacales). Systematic Biology59: 322–341.

de Souza Muñoz ME, De Giovanni R, Ferreira deSiqueira M, Sutton T, Brewer P, Scachetti Pereira R,Llange Canhos DA, Perez Canhos V. 2011. openMod-eller: a generic approach to species’ potential distributionmodelling. Geoinformatica 15: 111–135.

Spagna JC, Álvarez-Padilla F. 2008. Finding an upperlimit for gap costs in direct optimization parsimony. Cladis-tics: The International Journal of the Willi Hennig Society24: 787–801.

Stamatakis A. 2006. RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa andmixed models. Bioinformatics (Oxford, England) 22: 2688–2690.

Stamatakis A, Hoover P, Rougemont J. 2008. A rapidbootstrap algorithm for the RAxML Web servers. SystematicBiology 57: 758–771.

Suzuki S, Ohrui M. 1972. Opiliones of the Izu Peninsula,Central Japan. Acta Arachnologica 24: 41–50. [In Japanese.

Thorell T. 1882–1883. Descrizione di alcuni Aracnidi inferioridell’ Arcipelago Malese. Annali del Museo civico di Storianaturale di Genova 18: 21–69.

Thorell T. 1890–1891. Aracnidi di Pinang raccolti nel 1889dai Sig.ri L. Loria e L. Fea. Annali del Museo civico diStoria naturale di Genova (Ser. 2a) 10: 269–383.

Trewick SA, Paterson AM, Campbell HJ. 2007. Hello NewZealand. Journal of Biogeography 34: 1–6.

Varón A, Sy Vinh L, Wheeler WC. 2010. POY version 4:phylogenetic analysis using dynamic homologies. Cladistics:The International Journal of the Willi Hennig Society 26:72–85.

Vlahovic I, Tisljar J, Velic I, Maticec D. 2005. Evolution ofthe Adriatic carbonate platform: palaeogeography, mainevents and depositional dynamics. Palaeogeography Palaeo-climatology Palaeoecology 220: 333–360.



Wallis GP, Trewick SA. 2009. New Zealand phylogeography:evolution on a small continent. Molecular Ecology 18: 3548–3580.

Warren DL, Glor RE, Turelli M. 2008. Environmental nicheequivalency versus conservatism: quantitative approachesto niche evolution. Evolution 62: 2868–2883.

Warren DL, Glor RE, Turelli M. 2010. ENMTools: a toolboxfor comparative studies of environmental niche models.Ecography 33: 607–611.

Waters JM, Craw D. 2006. Goodbye Gondwana? NewZealand biogeography, geology, and the problem of circular-ity. Systematic Biology 55: 351–356.

Westwood JO. 1874. Thesaurus entomologicus oxoniensis; or,illustrations of new, rare, and interesting insects, for themost part contained in the collections presented to the Uni-versity of Oxford by the Rev. F.W. Hope . . . with forty platesfrom drawings by the author. Oxford: Clarendon Press.

Wheeler WC. 1995. Sequence alignment, parameter sen-sitivity, and the phylogenetic analysis of molecular data.Systematic Biology 44: 321–331.

Wheeler WC. 1996. Optimization alignment: the end of mul-tiple sequence alignment in phylogenetics? Cladistics: TheInternational Journal of the Willi Hennig Society 12: 1–9.

Wheeler WC. 2005. Alignment, dynamic homology, and opti-mization. In: Albert VA, ed. Parsimony, phylogeny, andgenomics. Oxford: Oxford University Press, 71–80.

Wiens JJ. 2003. Missing data, incomplete taxa, and phylo-genetic accuracy. Systematic Biology 52: 528–538.

Wiens JJ. 2009. Paleontology, genomics, and combined-dataphylogenetics: can molecular data improve phylogeny esti-mation for fossil taxa? Systematic Biology 58: 87–99.

Wortmann UG, Weissert H, Funk H, Hauck J. 2001.Alpine plate kinematics revisited: the Adria problem.Tectonics 20: 134–147.

SUPPORTING INFORMATION

Additional Supporting information may be found in the online version of this article:

Appendix S1. Specimen sampling. Detailed accounts of specimens and collecting data.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materialssupplied by the authors. Any queries (other than missing material) should be directed to the correspondingauthor for the article.

APPENDIXTable A1. Size of data matrices for each gene before and subsequent to treatment with GBLOCKS

Data partition

Number ofpositions aftertreatment withMUSCLE

Number ofpositions aftertreatment withGBLOCKS

16S rRNA 598 40318S rRNA 1769 176928S rRNA 2267 2016COI 820 657Histone H3 327 327



Table A2. Lagrange analyses.

Subtree: PettalidaeAreas:(a) South Africa(b) Chile(c) Eastern Australia(d) Western Australia(e) New Zealand, Australian plate(f) New Zealand, Pacific plate(g) Sri Lanka

Geological intervals:(1) 0–35 Ma (disconnection of all landmasses)(2) 35–60 Ma (fragmentation of transantarctic connections between Australian plate and temperate South America)(3) 60–75 Ma (disconnection of Australia and Zealandia)(4) 75–110 Ma (disconnection of South America and West Africa)(5) 110–120 Ma (Sri Lanka + Madagascar + India separated from Africa)(6) 120–167 Ma (East Gondwana separated from West Gondwana)(7) 167–184 Ma (connection of all landmasses)

Subtree: SternophthalmiAreas:(a) Southeast USA(b) Amazonia(c) Tropical West Africa(d) New Caledonia

Geological intervals:(1) 0–35 Ma (disconnection of all three landmasses)(2) 35–45 Ma (submersion of New Caledonia)(3) 45–60 Ma (New Caledonia emergent and disconnected)(4) 60–75 Ma (submersion of New Caledonia)(5) 75–110 Ma (transantarctic connections between the Australian plate and temperate South America; disconnection of South

America and West Africa)(6) 110–206 Ma (connection of all landmasses)

Subtree: BoreophthalmiAreas:(a) Thai-Malay Peninsula(b) Eastern Himalayas(c) Borneo(d) Indo-Malay Archipelago(e) North America(f) Western Europe(g) Mediterranean(h) Balkans(i) Iberia(j) Japan

Geological intervals0–35 Ma (separation of Mediterranean plate from Western Europe; separation of Japan from Eurasia; connection of Iberia to Eurasia)35–45 Ma (separation of Borneo and Indo-Malay Archipelago from Eurasia)45–60 Ma (Balkans connected to Western Europe; Iberia connected to Mediterranean plate, Balkans and Japan)60–75 Ma (Iberia separated from Mediterranean plate, Balkans and Japan; North America separated from Western Europe;

emergence of Indo-Malay Archipelago)75–110 Ma (Mediterranean plate separated from North America; Iberia connected to western Laurasia; Balkans separated from North

America and Western Europe)110–120 Ma (Iberia disconnected from other landmasses; Western Europe, Mediterranean plate and North America separated from

Eastern Laurasia; emergence of Borneo; Indo-Malay Archipelago nonexistent)120–180 Ma (Iberia disconnected from other landmasses; Western Europe, Mediterranean plate and North America separated from

Eastern Laurasia; Borneo and Indo-Malay Archipelago nonexistent)180–250 Ma (Thai-Malay Peninsula disconnected from other landmasses; Eastern Himalayas disconnected from North America,

Western Europe and Iberia; Borneo and Indo-Malay Archipelago nonexistent)250–296 Ma (Borneo and Indo-Malay Archipelago nonexistent; other landmasses connected)



Evolutionary and biogeographical history of an ancient …boyer/Site/Publications_files/Giribet et...

Documents

Transcript of Evolutionary and biogeographical history of an ancient …boyer/Site/Publications_files/Giribet et...