PhylogenyofFrogsofthe PhysalaemusPustulosus SpeciesGroup ... · 1998...

Syst Biol 47(2) 311 plusmn 335 1998

Phylogeny of Frogs of the Physalae m us Pu stu losus Species G roup

W ith an Examinat ion of D ata Incongruence

DAVID C CANNATELLA 1 2 DAVID M H ILLIS 1 PAUL T CHIPPINDALE 1 3 LEE W EIG T4

A S TANLEY RAND 4 5 AND M ICHAEL J RYAN1 4

1 Departm ent o f Zoology University o f Texa s Austin Texa s 78712 USA

E -ma il (D M H ) (M J R )h illis bull zo utexa s edu mryan ma il utexas edu2Texas M emoria l M useum University o f Texa s Austin Texas 78705 USA

E -m ail ca treg sh ma il utexas edu

Abstract ETH Characters derive d from advertisem en t calls m orph ology allozym es and the

se que nce s of the sm all subunit of the m itochondrial ribosom al gen e (12S ) and the cytochrom e

oxidase I (CO I) m itochondrial gen e were use d to e stimate the phylogeny of frogs of the Physa -

la emus pustulosus group (Le ptodactylidae ) The com binability of these data partitions was

assessed in several ways m easures of ph ylogen etic signal ch aracter support for tre es con -

gruence of tre e topologies com patib ility of da ta partition s with suboptimal tree s and hom o-

gene ity of da ta partitions Com bine d parsim ony analysis of a ll data eq ually weigh ted yielded

the sam e tre e as the 12S partition analyzed un der parsimony and m axim um likelih ood The

C O I allozym e and m orph ology partitions were gen erally congruent and com patib le with the

tree de rived from com bine d data The call data were sign ireg can tly diƒerent from all oth er

partition s whethe r considered in terms of tree topology alone partition hom oge neity or com -

patib ility of data with tre es derive d from other partition s Th e lack of eƒect of the call data on

the topology of the com bined tree is prob ab ly due to the sm all num be r of ca ll characters The

gene ral incongrue nce of the call data with othe r data partition s is con sistent with the ide a that

the adve rtisem en t calls of this group of frogs are under strong sexual se lection

[ Advertisem e nt calls behavior com bine d-da ta analysis data partitions frogs Leptodac-

tylidae Physa laem us sen sory e xploita tion hypoth esis]

W hether or not to combine data setshas been discussed widely in the recentliterature (Bull et al 1993 Eern isse andKluge 1993 Chippindale and W iens1994 de Q ueiroz et al 1996) Less dis-cussed is the identireg cation and localiza-tion of incongruence am ong datapartitions (but see Huelsenbeck and Bull1996 Poe 1996 Mason-G am er andKellogg 1996 Lutzoni 1997) It has beenargued that if diƒerent data partitions areno more diƒerent than expected by sam -pling error then the data can be com -bined in to a single analysis (Bull et al

ETH ETH ETH ETH ETH3 Present address D epartm en t of Biology U ni-

versity of Texas Arlington Texas 76019 U SA

E-m ail pau lc albertutaedu4 Sm ith sonian Tropica l Research In stitu te Un it

0948 APO AA 34002 Presen t ddress (L W ) Fie ld

M useum of Natural History Roosevelt Road at Lake

S hore D r C hicago Illin ois 60605 US A E-m ail

weigt fm pprfm nhorg5 E-m ail ran d gam boasiedu

1993) Although there are many reasonsto favor a com bined analysis (Eernisseand Kluge 1993 Chippindale and W iens1994) it can be enlightening to exam ineincongruence am ong data partitions

Behavioral data are rece iving increas-ing attention in phylogenetic analysis (deQ ueiroz and W imberger 1993 Foster etal 1996 G ittleman et al 1996 Irwin 1996 Kennedy et al 1996 W im bergerand de Queiroz 1996) In th is articlewe use a diverse original data setfrom advertisem ent calls m orphologyallozymes and the 12S and cytochrom eoxidase I (CO I) mitochondrial genes toestimate the phylogeny of frogs of thePhysalaemus pustulosus group (Cannatellaand Duellm an 1984) This clade hasserved as a m ode l for examin ing aspectsof behavioral evolution such as sexualse lection and signal-rece iver evolution(Ryan and Rand 1993 1995 Ryan 1996) Additionally we assess incongruenceam ong data partitions with several

311

312 S YSTEM ATIC BIO LO G Y VO L 47

methods and discuss the phylogeneticutility of the advertisem ent calls of thesefrogs

M ATERIALS AND M ETHO DS

S pecim ens were collected in the reg e ldtissues extracted and the voucher speci-mens preserved or prepared as ske le tons(Appendix 1) S pecim ens are deposited atthe United States National Museum andthe Texas M em orial M useum Universityof Texas S ome ske le tal m aterial wasborrowed from the American M useumof Natural History University ofKansas Museum of Natural History the Museum of Com parative ZoologyHarvard University and the LouisianaState University M useum of NaturalScience

Taxon Sampling

The species sam pled are listed inAppendix 1 All known valid species inthe ingroup were sampled we treated apopulation of P petersi that may be refer-ab le to the nom inal taxon P freibergi(Cannate lla and Duellm an 1984) as a dis-tinct taxon Monophyly of the ingroup issupported by four synapom orphies(Cannate lla and Duellman 1984) Out-group taxa were Physalaemus ephipp ifer Physa laemus sp A and Physa laemusenesefa e These species were chosenbecause our preliminary survey of m or-phology and calls among 75 of thespecies suggested that they are the mostsimilar to the pustulosus group in externalmorphology osteology and the generalcharacteristics of the call A more com -prehensive phylogenetic analysis ofrelationships in the genus is in progress

Data Pa rtitions

The following character se ts were desig-nated as data partitions morphologicalcharacters (n 5 12 M ORPHOLOG Y) ad-vertisem ent calls (n 5 12 CALLS ) allo-zym e electromorphs (n 5 27 ALLO -ZYM ES ) DNA seq uence of the cyto-chrom e oxidase I gene (n 5 543 COI)

and DNA seq uence of the sm all sub-unit of the m itochondrial ribosomal gene(n 5 1214 12S ) The combined dataset was designated as CO MBINED(n 5 1808)

M orphological characters (Appendix 2)were taken from dissections of wholespecim ens and alizarin -and-alcian plusmnstained skele tons (Dingerkus and Uhler1977) Although sample sizes of skeletonsfor most species were two or three asurvey of 30 skeletons of Physalaemuspustulosus (Cannate lla and Duellm an 1984) indicated no intraspecireg c polymor-phism in the characters examined andnone was noted in the present study

Advertisem ent calls were recorded inthe reg eld onto m etal tape with either aS ony TCD 5M M arantz PM D 420 orS ony Professional W alkm an using aM E-80 S ennheiser microphone with aK3-U power module and wind screen Tem peratures at the calling sites of eachfrog were recorded and usually were25 6 2degC S uch a small temperature dif-ferential has no substantial in macr uence oncall variation

The advertisement calls of the Physa -laemus pustulosus species group (exceptspecies C) and the three outgroup speciesare all sim ilar in that they are rather longfrequency sweeps W e refer to these callsas whines which describes the sound tothe hum an observer S ome species m ayadd to their call a suffix which isdescribed as a chuck TuAcirc ngara thecom m on nam e for P pustulosus is anonomatopoeia for the whine followed bytwo chucks Because the whine is thecom ponent required for species recogni-tion (Ryan 1985 Rand et al 1992 Ryanand Rand 1995) it is the only call com-ponent considered The whines diƒer intheir spectral properties (the onse t oƒse tand dom inant frequency) as well as inthe duration and shape of the frequencysweep All of the whines have upper har-m onics but in P pustulosus these harmo-nics have no inmacr uence on the calls rsquoattractiveness to fem ales (Rand et al1992 W ilczynski e t al 1995) These har-m onics are not considered here all

1998 CAN NATELLA ET AL ETH PHYSALAEM US PH YLO G ENY 313

TABLE 1 Allozym e loci exam ine d and buƒer system s and tissues used E C n um ber 5 Enzym e Com -

m ission num ber from In ternation al Un ion of Bioche m istry (1984) Buƒe r system s follow M urphy et a l

(1996) 1 5 Tris-citrate II pH 80 2 5 Tris-citra te-ED TA pH 70 3 5 Tris-b orate -EDTA II pH 86 4 5 Tris-

c itrate borate ge l pH 87

Locu s Abbrevia tion E C num ber Buƒer system

Acon itase h ydratase-1 Aco-1 4113 1 1 NAD P

Ade nylate kinase Ak 2743 1

Aspartate am inotran sferase Aat-M 2611 3

(m itoch ondrial form )

Aspartate am inotran sferase Aat-S 2611 1 3

(supernatan t form )

Creatine kinase C k 2732 1

Cytosol am inopeptidase C ap 34111 1

Este rase D Est-D 311-

Fructose -b iph osph atase Fbp 31311 1 1 NAD P

G lucose-6-phosphate dehydrogen ase G 6pdh 11149 4 1 NAD P

G lucose-6-phosphate isom erase G pi 5319 4

G lutam ate de hydrogenase G tdh 1414 1

G lutath ione reductase G r 1642 1

G lycerol-3-phosphate de hydrogenase G 3pdh 1118 2

Isocitrate de hydroge nase -1 Idh-2 11142 2


Lactate dehydrogen ase-A Ldh -A 11127 2

Lactate dehydrogen ase-B Ldh -B 11127 2

M alate de hydrogenase -1 M dh -1 11137 1


M alate de hydrogenase -1 (N ADP + ) M dh p-1 11140 2 1 NAD P


Peptidase A (glycyl-L- leucine ) Pep-A 34- - 1

Ph osph oglucom utase Pgm 5422 2 1 NAD

Ph osph ogluconate de hydrogenase Pgdh 11144 1 1 NAD P

Supe roxide dism utase S od-S 11511 2


Triose-ph osph ate isom e rase Tpi 5311 2

values re fer to the fundam ental fre -quency

S pectral properties of calls except fordom inant freq uency were analyzed on aUniscan sonograph Tem poral propertieswere analyzed on a DATA 6000 digitalwaveform analyzer Calls were digitizedat a rate of 20 kHz therefore the Nyq uistfreq uency is 10 kHz substantially abovethe highest frequencies in any of the callsanalyzed The dominan t frequency of thecall also was analyzed on the DATA 6000by taking a fast Fourier transform of theentire call The following call variableswere q uantireg ed Duration (TLDURmsec) freq uency at onset of call (INHZHz) maximum freq uency (MXHZ Hz)time to the maximum frequency (TM M Xmsec) tim e to mid-freq uency (TMHFHZmsec) frequency at oƒset of call (FNHZ

Hz) dominant frequency (DO MHZ Hz)duration of am plitude-modulated com-ponent (AM DUR msec) rise time (RSTM in msec) time to m id-rise (TMHFRS m sec) fall time (FLTM msec) and tim eto m id-fall (TM HFFL m sec)

Calls and tissues for DNA andallozyme analysis are from the sam e indi-viduals except for Physalaemus pustulosus in which they are from diƒerent individ-uals in the sam e population The COIand 12S seq uence data for P pustulosuswere obtained from diƒerent individualsbut these cam e from the sam e popu-lation Each species is represented by onepopulation intraspecireg c variation wasnot assessed Although there are signireg -cant diƒerences in call parameters with ina species (eg Ryan and W ilczynski1988 1991) from studies of Physalaemus


pustulosus we know that in traspecireg cvariation is far less than variation amongthe species (Ryan et al 1996)

Liver heart and thigh muscle weredissected from 10 individuals from eachpopulation in the reg e ld and imm ediate lyfrozen in liq uid nitrogen until transporta-tion to the University of Texas Austin atwhich time they were m aintained in anultracold freezer at less than plusmn 70(CMethods for allozym e electrophoresis fol-lowed the horizontal starch ge l protocolsdescribed by Murphy et al (1996) Gelswere made from 12 starch (S tarch Artlot W 561-2) Table 1 shows the enzym eloci scored and buƒer system used toscore each locus Appendix 1 lists thelocalities of the specim ens examined

M ethods for DNA isolation am plireg ca-tion cloning and seq uencing followedHillis e t al (1996) protocol numbers inthe following description re fer to thatpaper W hole genomic DNA was isolatedusing protocol 1

Data partition 12S consisted of thecomple te mitochondrial 12S rRNA genecomple te valine-tRNA gene and the

adjacent approxim ate ly 200 bp of the 16SrRNA gene These were am plireg ed by thepolym erase chain reaction (see Palum bi1996) using primers 12Sh and 16Sh (Table2) The am plireg ed product was clonedusing TA cloning (protocol 18 part B)Plasm id DNA was isolated according toprotocol 14 and sequenced (protocols 2122 and 25) using the primers shown inTable 2 The 12S sequences were alignedusing M ALIG N (W heeler and G ladstein 1992)

The same extracted DNA sampleswere used to sequence the cytochrom eoxidase I gene DNA from the followingspecies was am plireg ed using the poly-m erase chain reaction with CO If andCO Ia primers (Palumbi 1996) P eph ipp ifer P freibergi P sp B P sp A and P pustulosus The remain ingspecies were amplireg ed with COIf andCO Ia2 (designed for these species) P coloradorum P enesefa e P petersi P pustula tus P sp C The region of analysisincluded sites 55 plusmn 597

After amplireg cation the product wasseparated and excised from an agarose

TABLE 2 Prime rs used to se que nce 12S rRNA valine -tRNA and 16S rRNA genes (upper part of tab le )

an d C O I gene (lower part) Th e 12S prim er loca tions refer to the positions in the P pustulosus seq uen ce

The designations pp6 plusmn pp9 are inte rna l prim ers for CO I

12S prim er nam e Prim er sequence Position

12S a 5rsquo-AAAC TG G G ATTAG ATAC C CC ACTAT-3rsquo 413plusmn 437

12S ar 5rsquo-ATAGTG G G G TATCTAATC C CAG TTT-3rsquo 437plusmn 413

12S b 5rsquo-G AG G G TG AC G G G CG G TG TG T-3rsquo 835plusmn 816

12S c 5rsquo-AAG G CG G ATTTAG CAG TAAA-3rsquo 754plusmn 773

12S d 5rsquo-TC G TG C CAG CC RCC G CG G T-3rsquo 230plusmn 248

12S e 5rsquo-G G G AAG AAATG G G C TACATTTTC T-3rsquo 689plusmn 712

12S h 5rsquo-AAAG G TTTG G TCC TAG C CTT-3rsquo 1 plusmn 20

12S k 5rsquo-G G G AACTAC G AG C AAAG C TT-3rsquo 475plusmn 494

12S l 5rsquo-G G AC AG G C TCC TC TAG G TG G -3rsquo 545plusmn 526

16S h 5rsquo-G C TAG AC CATKATG C AAAAG G TA-3rsquo 1202plusmn 1180

M 13re v 5rsquo-C AG G AAACAG CTATG AC -3rsquo vector

T7 prom ote r 5rsquo-AATACG ACTCACTATAG -3rsquo vector

CO I prime r n am e Prim er sequence Position

C O If 5rsquo-C CT G C A G G A G G A G G A G AY C C-3rsquo 1 plusmn 20

C O Ia 5rsquo-AGT ATA AG C G TC TG G G TA G TC -3rsquo 660plusmn 681

C O Ia2 5rsquo-C CT G C Y AR Y C CT AR R AAR TG T TG A G G -3rsquo 616plusmn 641

pp6 5rsquo-TC T G C A AC A ATA ATY ATY G C A ATT CC A AC -3rsquo 256plusmn 284

pp7 5rsquo-G TT G G A ATT G CR ATR ATT ATT G TT G C A G A-3rsquo 284plusmn 256

pp8 5rsquo-TC T C TA G AY ATT G TA TTA C AT G A-3rsquo 421plusmn 443

pp9 5rsquo-TC A TG T AAT AC A ATR TC T AG A G A-3rsquo 443plusmn 421


gel and resuspended for a second roundof PCR am plireg cation The product waspurireg ed via G eneclean III (BIO 101 LaJolla California) Cycle sequencing wasdone with the ABI Prism m ix sequencingkit Sequences were run on an ABI 377automated DNA sequencer (Applied Bio-system s Perkin plusmn Elmer Foster C ityCalifornia) using the m anufacturerrsquo srecom mended protocols Seq uences wereread verireg ed and aligned with the ABIsoftware package SeqEd

G enbank accession num bers areAF058957-66 The NEXUS reg le (M addisonet al 1997) is availab le at http wwwutexasedu deptssystb iol

Phylogenetic Ana lysis

Coding of the call variab les followed aprocedure inspired by Maddison andSlatkin (1990) The min im um andmaxim um values of a variab le (datapooled over all species) were scaled to 0and 25 respective ly (Tab le 3) The speciesmean was then scaled m onotonically tothe nearest integer Each character wasdownweighted to unity and analyzed asordered In th is way the re lative distancebetween each pair of values was main -tained and calculation of homoplasyindices was possible

Phylogenetic analyses were done usingPAUP 311 (S woƒord 1993) and PAUPtest versions 400d26 plusmn 400d28 (providedby David S woƒord) The allozymic datawere coded using step m atrices so that areg xed change at a locus was weighted asone step in the parsimony analysis andany intermediate combination of alleleswas counted as a half-step Thus achange from a reg xed to a polym orphiccondition or vice versa (eg aa to ab orab to bb) was counted as a half stepwhereas a reg xed or mutually exclusive dif-ference (e g aa to bb or ab to cd) wascoded as a full step Parsim ony analysesof the DNA data included (1) all charac-ter transform ations weighted equallywith gaps treated as a reg fth character (2)all character transform ations weightedeq ually but gaps treated as m issing data

and (3) a weighted parsim ony analysis inwhich transversions were given weightsof two and reg ve tim es re lative to tran-sitions These values were based on thesubstitution m atrix estimated by averag-ing across all m ost parsim onious recon-structions of characters on an initialunweighted tree using M acClade(M addison and M addison 1992) M aximum-like lihood analyses included(1) a one-parameter analysis (all classesof substitutions eq ually like ly) assum ingeq ual base frequencies (2) a one-param eter analysis using em pirical(observed) base freq uencies (3) a two-param eter analysis (allowing diƒerentrates of transitions and transversions)with equal base frequencies and (4) atwo-parameter analysis with empiricallydetermined base freq uencies

Data were weighted as follows 12S CO I M O RPHO LO G Y and m ono-m orphic loci from ALLO ZYM ES wereweighted 1000 polymorphic loci fromALLO ZYM ES were weighted 500 andCALLS were scaled with a base weight of1000 In th is way the total variation ineach character was eq ually weightedEach data partition was analyzed separa-te ly and the data were pooled for a com-bined analysis

Nonparametric bootstrap analyseswere conducted with 5000 iterationsDecay values (Brem er support branchsupport) were calculated using theHypercard utility Autodecay 295(Eriksson 1996 h ttp wwwbotansuseS ystematikFolkTorstenhtm l) 10 ran-dom-addition seq uences were used todetermine the decay value for each nodeof each tree The resulting trees aredepicted with the outgroup arbitrarilyshown as monophyletic Bootstrapdecayvalues for the branch connecting theingroup and outgroup were arbitrarilyplaced at the base of the ingroupBecause no data on calls were availablefor Physalaemus sp C the results of theCO MBINED analysis were used to con-strain that species to be the sister speciesof Physalaemus sp B for com parisons oftree topologies


TA

BL

E3

S

um

ma

ryo

fca

llsta

tisti

cs

for

the

Ph

ysa

laem

us

pu

stu

losu

sg

rou

pa

nd

clo

se

rela

tiv

es

Th

ev

ari

ab

les

are

tota

ld

ura

tio

no

fca

ll(T

LD

UR

m

se

c)

fre

qu

en

cy

at

on

set

of

ca

ll(I

NH

Z

Hz)

ma

xim

um

fre

qu

en

cy

(MX

HZ

H

z)

tim

eto

the

ma

xim

um

fre

qu

en

cy

(TM

MX

m

se

c)

tim

eto

mid

-fre

qu

en

cy

(TM

HF

HZ

mse

c)

fre

qu

en

cy

at

oƒs

et

of

ca

ll(F

NH

Z

Hz)

do

min

an

tfr

eq

ue

ncy

(DO

MH

Z

Hz)

du

rati

on

of

am

pli

tud

e-m

od

ula

ted

co

mp

on

en

t(A

MD

UR

m

se

c)

rise

tim

e

(RS

TM

in

mse

c)

tim

eto

mid

-ris

e(T

MH

FR

S

mse

c)

fall

tim

e(F

LT

M

mse

c)

an

dti

me

tom

id-f

all

(TM

HF

FL

m

se

c)

Th

em

ea

nis

giv

en

wit

hth

era

ng

eb

elo

w

Th

ela

tte

rin

pa

ren

the

se

sfo

llo

win

gth

em

ea

nis

the

ch

ara

cte

r-sta

teco

de

se

eM

ate

ria

lsa

nd

Me

tho

ds

Th

ev

ari

ab

les

are

dis

cu

sse

din

mo

red

eta

ilin

Co

cro

ft

an

dR

ya

n(1

99

5)

Sp

ecie

sn

TL

DU

RIN

HZ

MX

HZ

TM

MX

TM

HF

HZ

FN

HZ

DO

MH

ZA

MD

UR

RS

TM

TM

HF

RS

FL

TM

TM

HF

FL

sp

A1

339

(j)

812

(d)

876

(a)

65

(k)

16

0(j

)46

0(j

)9

83

(v)

0(

)9

46

(h)

56

6(h

)25

16

(n)

711

(h)

02

34

plusmn4

47

800

plusmn84

08

00

plusmn9

20

43plusmn

10

015

0plusmn1

87

40

0plusmn

520

767

plusmn13

62

0plusmn

06

4plusmn

11

637

plusmn8

317

4plusmn

339

31plusmn

11

6

eph

ipp

ifer

12

66

(g)

900

(h)

944

(e)

62

(j)

14

0(i

)57

6(q

)9

44

(t)

0(

)8

35

(g)

39

4(e

)17

74

(i)

606

(g)

238

plusmn30

88

40

plusmn1

00

08

40

plusmn1

04

04

3plusmn

81

11

2plusmn1

62

52

0plusmn

60

08

45

plusmn10

25

0plusmn

04

8plusmn

97

4plusmn

56

12

9plusmn

236

27plusmn

10

7

enes

efae

17

47

(z)

944

(j)

976

(g)

16

2(z

)38

6(z

)69

2(z

)9

62

(u)

38

48

(z)

30

15

(z)

166

6(z

)44

57

(z)

20

37

(z)

06

31

plusmn9

03

880

plusmn1

04

09

20

plusmn1

04

08

1plusmn

28

730

0plusmn

45

664

0plusmn

720

844

plusmn13

84

23

8plusmn

76

023

0plusmn

40

71

25

plusmn24

237

2plusmn

546

99plusmn

33

8

pu

stu

losu

s1

370

(k)

884

(h)

884

(a)

0(a

)12

4(h

)48

4(k

)7

12

(j)

223

(b)

240

(a)

79

(a)

34

28

(s)

17

50

(v)

02

52

plusmn49

68

40

plusmn9

60

840

plusmn9

60

0plusmn

08

7plusmn1

75

44

0plusmn

60

06

05

plusmn8

83

119

plusmn3

27

93

plusmn60

62

2plusmn1

77

23

6plusmn

45

01

06plusmn

25

2

pet

ersi

12

46

(f)

12

20

(x)

12

20

(w)

0(a

)2

8(b

)38

4(d

)6

28

(g)

120

(a)

137

(a)

11

5(a

)23

03

(m)

471

(e)

02

06

plusmn35

010

40

plusmn1

40

010

40

plusmn1

40

00plusmn

01

8plusmn5

032

0plusmn

48

05

96

plusmn6

93

84

plusmn2

23

60

plusmn1

88

96

plusmn14

219

4plusmn

331

82

plusmn1

61

frei

ber

gi

11

04

(a)

12

53

(z)

12

53

(z)

0(a

)1

2(a

)33

0(a

)4

82

(a)

150

(a)

193

(a)

17

6(b

)3

00

(a)

126

(a)

04

82

plusmn1

408

10

00

plusmn1

42

410

00

plusmn1

42

40plusmn

06

plusmn2

527

2plusmn

368

361

plusmn58

545

plusmn2

31

42

plusmn29

913

3plusmn

24

13

41

plusmn12

47

12

plusmn3

79

colo

rad

oru

m9

221

(e)

10

31

(o)

10

71

(m)

25

(d)

83

(e)

55

6(p

)10

07

(w)

470

(d)

534

(d)

23

3(c

)16

17

(h)

470

(e)

152

plusmn3

58

960

plusmn1

08

010

00

plusmn1

16

00plusmn

62

50

plusmn1

00

48

0plusmn

64

08

89

plusmn11

33

33plusmn

72

25plusmn

72

12

plusmn52

64plusmn

285

8plusmn

12

8

pu

stu

latu

s1

206

(d)

964

(k)

964

(f)

0(a

)8

8(f

)67

6(x

)10

62

(z)

943

(g)

995

(h)

95

0(n

)10

43

(e)

529

(f)

01

86

plusmn23

08

80

plusmn1

08

08

80

plusmn1

08

00plusmn

05

6plusmn1

18

64

0plusmn

80

08

20

plusmn12

54

731

plusmn10

76

903

plusmn1

09

477

7plusmn1

08

27

69

plusmn1

29

01

95

plusmn12

0

sp

B1

395

(l)

740

(a)

888

(a)

11

2(r

)11

5(g

)44

4(h

)8

94

(r)

921

(f)

10

51

(h)

69

4(j

)29

37

(p)

933

(k)

03

22

plusmn60

86

80

plusmn88

08

40

plusmn9

60

43plusmn

15

08

1plusmn1

62

40

0plusmn

48

08

54

plusmn98

16

5plusmn

14

95

0plusmn

15

425

plusmn12

023

8plusmn

44

41

5plusmn

16

9


Assessments of Combinability

There are several issues re lated to theconcept of combinab ility (1) phylogen-etic signal or data structure (2) strengthof support for a resulting tree topology (3) congruence of trees from diƒerentdata partitions (4) homogeneity of datapartitions (5) compatib ility of a data par-tition with a suboptim al tree and (6)strength of support (assuming 5 is true)of a data partition for a suboptimal tree

Phylogenetic signa l ETH If a data set hasno structure that is signireg cantly diƒerentfrom random then little con reg dence canbe placed in the resulting estim ates oftree topology However lack of discern -ib le structure may be an artifact of sm allnumbers of characters W e assessed datastructure using the PTP test (Faith 1991)as im plem ented in PAUP using 5000random m atrices

Strength of support for a tree topology ETHCon reg dence in trees was q uantireg ed forbranches using character resampling(nonparam etric bootstrap Hillis and Bull1993) and Bremer support (decay index)value and for the entire tree using ``totalsupportrsquo rsquo test and the constrained treeT-PTP Clades with 70 bootstrapvalues are considered strongly sup-ported

The ``total supportrsquo rsquo test described byKaEgrave lle rs joEgrave e t al (1992) and recomm endedby Brem er (1994) consists of computingtotal support (the sum of all Bremersupport values also called decay indices)for the observed data and com paring th isto a distribution of total support valuesfrom random ly perm uted m atrices Onehundred m atrices were produced usingMacClade 305 and decay indices foreach m atrix were calculated using Auto-decay 295 (Eriksson 1996) 10 random -addition heuristic searches were used foreach decay value

The constrained-tree T-PTP test is anextension of Faithrsquo s monophyly test (seealso Faith and Cranston 1991) in whichan entire tree rather than a single nodeis used as a constraint It is im plem entedas the TPTP test in PAUP but an entire

tree is de reg ned as a constraint rather thanjust one node (see S woƒord et al 1996for a criticism of T-PTP tests) The lengthdiƒerence between the observed shortesttree and the shortest tree that is incon-gruent in any part of the tree is used asthe test statistic and com pared to a nulldistribution of length diƒerences gener-ated from perm uted data This testam ounts to a test of the m onophyly of thenode with the weakest decay indexRe jection of the null hypothesis is in ter-preted as signireg cant support for a speci-reg ed topology as opposed to generalcladistic structure in the case of the PTPtest The null distribution is essentiallyone of decay indices based on perm uteddata G enerally 1000 randomizedm atrices were used to generate the nulldistribution If the perm utation-tail prob-ab ility was 005 or less the test was rerunwith 5000 matrices to increase resolutionin the tail of the distribution Theconstrained-tree test diƒers in details ofexecution from the ``all-groupsrsquo rsquo test pro-posed by Faith and Ballard (1994)although the purpose (assessing overallsupport of a data set for a tree) is similar

Congruence of trees ETH A th ird issue is thecongruence of trees resulting from datapartitions W e assessed tree congruenceby strict consensus trees (S woƒord 1991)and tree sim ilarity by the symm etric-diƒerence distance or partition m etric(Robinson and Foulds 1981) which isde reg ned as the number of subclades thatappear on either of the two trees but notboth This m etric q uantireg es diƒerences intree topology (``taxonomic congruence rsquo rsquo)irrespective of the character supportPenny and Hendy (1985) discussedseveral attractive features of this metricwhich can be used with unrooted orrooted and binary or nonbinary treesValues range from 0 to 2n 2 6 where n isthe number of term inals (S tee l andPenny 1993) It should be noted that aterminal with diƒering position on twootherwise similar trees m ay yield a largevalue in the way that a strict consensustree would appear largely unresolved


under similar conditions The probab ilitythat two given trees are drawn at randomfrom all possib le trees was determ inedusing Tab le 3 in Hendy et al (1984) thusrejection of the null hypothesis indicatesthat two labeled topologies are moresimilar than one would expect by chance

Homogene ity of partitions ETH Bull et al(1993) argued that one should be cautiousin combin ing data partitions that are sig-nireg cantly heterogeneous W e do notargue for or against combining heter-ogeneous partitions rather we sim plywish to determ ine heterogeneity beforefurther analysis W e assessed partitionhom ogeneity using PAUP The partition-hom ogeneity test generally assumesthat if diƒerent data partitions arehom ogeneous then random ly allocatingcharacters am ong those partitions shouldyie ld trees that are not signireg cantlydiƒerent As proposed by Farris e t al(1994 1995) the test relies on theobserved incongruence length diƒerence

com pared to a null distribution gen-Dxy erated by pooling the m 1 n charactersfrom partitions (matrices) x and y andthen random ly allocating these in to twomatrices of original sizes m and n Theincongruence length diƒerence isDxy de reg ned

Dxy 5 L(x+ y )

2 (Lx 1 Ly)

where and are the lengths of theLx Lyshortest trees for m atrices x and y and

is the length of the shortest tree forL(x+ y )

the combined m atrix Farris e t al (1994)argued that did not need to be cal-L

(x+ y)culated because it was a com mon termThus the test becom es a com parison ofthe sum of observed tree lengths com -pared to the sum of tree lengths fromrandom character partitions If the datapartitions are congruent then the length-sum s of the random partitions will beless than or equal to that of the observedpartition If the partitions are highlyincongruent then the length-sums of therandom partitions will be greater thanthat of the observed partition becauserandom partitions will tend to produce

(longer) trees with more hom oplasyPAUP determines the signireg cance of thetest by P 5 1 2 (SW ) where S is thenum ber of replicates in which the length-sum is greater than the length-sum forthe observed partition and W is the totalnum ber of observed and random parti-tions Farris e t al (1994) noted that theexact lengths were not crucial andapproximate parsim ony calculations (ega ``one-passrsquo rsquo heuristic search) were suffi-cient but because of the small num ber oftaxa we used heuristic searches with TBRbranch-swapping Partition-homogeneitytests were done for all pairwise compari-sons of data partitions and a simulta-neous reg ve-partition test with 1000iterations for each test

Compatib ility of da ta partitions with sub -op tima l trees ETH Even though two data par-titions strongly support diƒerent trees itm ay be that one partition is com patible(does not conmacr ict) with the other(suboptim al) tree S uch com patibility wastested using Tem ple tonrsquos test and thecom pare -2 T-PTP

Tem pleton rsquos test (Tem ple ton 1983 Larson 1994) is a W ilcoxon signed rankstest (Zar 1974) of the diƒerence inlengths of characters when a data parti-tion is optim ized on one tree versusanother Its results can be in terpre ted asa statement about the compatib ility of adata partition with a suboptim al tree rather than a statem ent about two treetopologies The more conservative two-tailed test was used (Felsenstein 1985) although it can be argued that the one-tailed test is appropriate

The com pare-2 T-PTP was suggestedby Faith (1991) and is im plem ented inPAUP A data se t is optimized usingparsim ony on each of two constrainttrees and the diƒerence in length is usedas a statistic and compared to a null dis-tribution of length diƒerences from ran-domly perm uted data If one of theconstraint trees is the shortest tree thenthe test re macr ects the compatib ility of thedata partition with the second sub-optimal tree


Strength of support for suboptima l trees ETHIt is of interest whether a data partitiongives signireg cant support to a suboptim altopology in addition to being compatib lewith it This was assessed using aconstrained-tree T-PTP as describedearlie r

Other considerations ETH The T-PTP per-mutation tests are im plem ented inPAUP as a priori tests (Faith 1991) inwhich no particular hypothesis of mono-phyly is be ing tested In cases where aparticular hypothesis of monophyly istested the a posteriori test is moreappropriate Using the a priori test canincrease Type 1 error (wrongly re jectingthe null hypothesis) The constrained-tree test can be perform ed as an a prioritest because there was no expectationof particular monophyletic groupsHowever it is not clear that the compare-2 tests are properly executed as a prioritests In the case of the test for mono-phyly of a clade the a posteriori mono-phyly test is performed by subtractingthe m in imum length under a monophylyconstrain t from the length under non-monophyly the length diƒerences arecalculated for the observed and m anypermuted data matrices However for aparticular perm uted m atrix the lengthdiƒerence is calculated using the largestvalue found for all groupings of taxa thesam e size as the clade of interest (Faith1991) Thus the length diƒerence wouldbe evaluated for example for each of the35 com binations of three taxa from theseven ingroup taxa for each permutedmatrix

The T-PTP tests used here in (both theconstrained-tree and compare-2) diƒerfrom the m onophyly test in that theentire tree is constrained and Faithrsquo s(1991) procedure of evaluating clades ofeq ual size amounts to exam in ing alterna-tive trees as is done in the a priori testThus it would seem that if the entire treeis constrained there is no operational dif-ference between a priori and a posterioritests However we fee l that the issuedeserves further exam ination (e g S wof-

ford et al 1996) and because a solutionis not obvious we have perform ed allperm utation tests as a priori tests One ofthe purposes of th is paper is to exam inethe behavior of these tests and theresults of these tests are very consistentwith other tests (see Results)

W e have used the COM BINED data setas if it were any other data partition However th is in troduces a degree ofnonindependence in pairwise compari-sons Curiosity about the behavior of theCO MBINED partition in these tests out-weighs our concerns about nonin-dependence and the results can bereadily in terpre ted

A seq uential Bonferroni correction(Rice 1989) was applied to the tables ofprobability values resulting from thepairwise procedures

RES ULTS

The statistics for the call variables andthe coding for each are shown in Tab le 3The alle le frequencies for the presum p-tive loci are presented in Table 4

Phylogenetic Analysis

Phylogenetic signa l and phylogenyestimation ETH The PTP test indicated thateach data partition had signireg cantphylogenetic structure (Tab le 5) S tatisticsfrom the results of the separate andcom bined phylogenetic analyses areshown in Tab le 5 and Figure 1 Eitherone or two m ost parsim onious treeswere found for each partition TheCO MBINED data set and the 12Spartition produced the sam e tree

W eighting transversions twice as muchas transitions yie lded the same shortesttrees for the CO MBINED 12S and COIpartitions W eighting transversions reg vetim es as m uch as transitions yie lded thesame shortest trees for the CO MBINEDand 12S partitions and for the COI parti-tion yie lded one of the two trees found inthe unweighted analysis the one with the((P coloradorum pustula tus ) (sp B sp C))topology

For the 12S data partition allm axim um-like lihood analyses yielded


TA

BL

E4

A

llo

zy

me

ge

no

typ

es

ob

se

rve

din

10

sp

ecie

so

fP

hy

sala

emu

s

Se

eT

ab

le1

for

exp

lan

ati

on

of

ab

bre

via

tio

ns

of

loci

Ge

no

typ

es

are

foll

ow

ed

by

the

nu

mb

er

of

ind

ivid

ua

lso

bse

rve

dw

ith

the

resp

ecti

ve

ge

no

typ

e

Ifn

og

en

oty

pe

issh

ow

n

the

nn

on

eo

fth

esp

ecim

en

se

xh

ibit

ed

inte

rpre

tab

lea

cti

vit

y(t

he

se

loci

we

resco

red

as

mis

sin

gin

the

ph

ylo

ge

ne

tic

an

aly

sis

)

Sp

ecie

s

Lo

cu

sco

lora

doru

men

esef

ae

eph

ipp

ifer

frei

berg

ip

eter

sip

ust

ula

tus

pu

stu

losu

ssp

A

sp

B

sp

C

Aco

-1cc

10

ee

9d

d8

dd

8d

d1

0cc

8a

b2

dd

8cc

10

cc

4

cd

1b

b8

de

2

Aco

-2d

d8

bb

1a

a2

cc

7cc

10

bd

3b

b9

bb

10

dd

6b

d4

bf

2a

b2

dd

6e

e1

ƒ3

bb

4

Ak

a6

cc

10

cc

7cc

8cc

9a

a1

cc

10

cc

7cc

8a

a4

aa

2a

c3

cc

1cc

1

Ast-

Mb

b1

0cc

10

cc

8b

b9

cc

10

bb

9d

d1

0e

e1

0a

a9

bb

4

bb

1

Aa

t-S

bb

10

dd

10

bb

8cc

9cc

10

bb

9b

b8

bc

3a

a1

0b

b4

bc

2cc

3

Ck

cc

10

bb

10

aa

8a

a9

aa

10

bb

5b

b1

0a

a1

0b

b1

0b

b4

Ca

pa

a1

0b

b6

bb

8a

a9

bb

10

bb

9b

b1

0b

b1

0cc

8a

a4

Est-

Db

b1

0b

b1

0b

b8

bb

9b

b1

0a

b1

bb

10

bb

10

bb

10

bb

4

bb

8

Fb

pe

e8

dd

10

bb

8d

d9

dd

10

ƒ9

cc

10

bb

9b

b1

0b

b4

G6

pd

ha

a9

bb

10

bb

8a

a9

aa

5a

a9

aa

10

bb

10

aa

10

aa

4

Gp

ia

a8

cc

10

dd

6ii

9e

e1

0a

c4

gg

10

hh

10

cc

10

jj4

ac

2d

f2

cc

5

Gtd

ha

a1

0b

b1

0b

b8

aa

9a

a1

0a

a9

aa

10

bb

10

aa

10

aa

4

Gr

aa

8d

d1

0d

d7

dd

9d

d1

0cc

8b

b3

dd

10

dd

10

dd

3

bf

5

ƒ2

G3

pd

hb

b1

0b

b9

bb

8b

b9

bb

10

aa

9b

b1

0b

b1

0b

b1

0b

b4

Idh

-1d

d1

0d

d1

0e

e1

0d

d9

aa

4ƒ

9b

b1

0d

d1

0d

d1

0d

d4

ad

6

Idh

-2cc

10

aa

10

ab

1d

d9

dd

10

dd

8b

d4

bb

10

ee

10

ce

1

bb

7d

d6

ee

3

Ld

h-A

bb

10

bb

10

aa

8cc

9cc

9b

b9

cc

10

bb

10

bb

10

bb

4

cd

1

Ld

h-B

aa

7g

g9

ETHd

d9

dd

10

ƒ9

bb

10

ETHcc

10

hh

4

ae

1

ee

2

Md

h-1

bb

10

bb

10

cc

8d

d9

dd

10

bb

9a

a9

ee

10

ab

1ƒ

4

bb

9


TA

BL

E4

C

on

tin

ue

d

Sp

ecie

s

Lo

cu

sco

lora

doru

men

esef

ae

eph

ipp

ifer

freib

ergi

pet

ersi

pu

stu

latu

sp

ust

ulo

sus

sp

A

sp

B

sp

C

Md

h-2

aa

10

aa

10

ETHd

d9

dd

10

aa

3a

a1

0ETH

aa

10

ETH

ad

6

Md

hp

-1a

a1

0a

a1

0a

g1

bb

4b

b8

ee

9b

b1

0b

b1

0cc

9ƒ

4

bg

2b

c4

bc

2

gg

5cc

1

Md

ph

-2a

a4

ee

10

ee

8d

d9

dd

10

dd

8a

a6

ab

1a

a5

bb

4

ab

3a

b1

bb

8a

b3

bb

3b

b1

bb

1

ad

1

dd

1

Pe

p-A

dd

10

cc

9e

e7

bb

8cc

10

dd

9cc

10

aa

10

cc

10

cc

4

Pg

ma

a7

gg

10

cc

8h

h9

ee

10

bb

1ƒ

9g

g1

0cc

1jj

4

ab

3b

c4

reg1

cf

2

cc

4ƒ

7

Pg

dh

cc

10

aa

2a

b2

dd

9d

e1

aa

8e

g3

aa

1e

e1

0h

h4

ab

4b

b4

ee

7a

c1

gg

6a

b4

bb

4b

f2

bb

4

So

d-S

aa

10

dd

9d

d8

ƒ1

ee

2b

b5

cc

10

dd

10

bb

10

bb

4

fg4

eg

3b

i2

gg

4g

g1

ii1

gh

2

hh

1

Tp

ia

a1

0d

d1

0cc

10

gg

9g

g9

bb

9e

e3

cc

10

dd

9ii

4

gh

1e

g6

df

1

gg

1


TABLE 5 Ph ylogen y estim ation statistics for e ach data partition CI 5 consistency inde x RI 5 retention

inde x t 5 total support ti 5 total support index an d m pt 5 n um ber of m ost parsimonious trees The

constra ined- tree T-PTP is the probab ility that the da ta support the constraint ree The PTP is the probabil-

ity associate d with the test for sign ireg can t ph ylogen etic structure

Inform a- C onstrain-

Total tive ed- tree

Data partition characters ch aracters C I RI Len gth t ti T-PTP PTP m pt

C O M BIN ED 1808 442 068 060 12739 4 16098 0126 00002 00002 1

12S 1214 255 073 066 709 107 0151 00002 00002 1

C O I 543 138 060 045 425 26 0061 00002 00002 2

ALLO ZYM ES 27 25 080 053 102 95 0093 00002 00002 1

C ALLS 12 12 071 061 168 232 0138 00002 00004 2

M O RPHO L- 12 12 100 100 113 13 1000 00002 00002 1

O G Y

the sam e trees as did the unweightedparsimony analysis (Fig 1) For the CO Ipartition only one of the two best par-simony trees having the sam e topologyas the tree from the weighted parsim onyanalysis was found Under both one-and two-parameter m ode ls the 12S COM BINED topology (Fig 1) had ahigher likelihood using em pirical basefreq uencies than did the alternate CO Itree W hen eq ual base freq uencies wereassum ed the COI tree had a higher like l-ihood than the 12S CO MBINED treeBecause the results of the maxim um -like lihood analyses do not diƒer signireg -can tly from those of the parsim onyanalysis they are not discussed further

In all trees except CALLS the ingroupwas found to be m onophyletic W ithinthe ingroup the cis-Andean species (P pustulosus petersi and freibergi ) form ed aclade in the M ORPHOLO GY 12S CO M -BINED and ALLOZYM ES trees Thetrans-Andean species (P coloradorum pustula tus sp B and sp C) form ed aclade in the COI MO RPHO LO G Y 12S and CO M BINED tree Neither of thesegeographic groups was m onophyletic inthe CALLS tree In all trees P petersi andP freibergi were sister taxa

Strength of support for a tree topology ETHThe CO M BINED tree has the strongestsupport only one bootstrap value (63)was below 90 Bootstrap values for theALLO ZYM ES and CALLS trees were thelowest The statistical signireg cance of thedecay index values (Fig 1) is undeter-

m ined but they are strongly corre latedwith the bootstrap values (S pearm an rsquosrho 5 0879 P 5 00001) The total supporttest values for each data partition weresignireg cant (Fig 2) indicating departurefrom random m atrices However thebehavior of th is test has not beenexplored The null distribution from per-m uted matrices is highly asym metricwith m ost values being 0 A total supportvalue of 0 means that no branch in thetree calculated from a randomized m atrixhad a decay index greater than 0

The constrained-tree T-PTP tests(Table 5) indicate that each data partitionsignireg cantly supports the tree derivedfrom that partition

Congruence of trees ETH A strict consensustree of the reg ve topologies is unresolvedexcept for the P petersi plusmn freibergi clade(these species were considered conspe-cireg c by Cannate lla and Duellm an [ 1984] )In the CALLS tree the ingroup is notm onophyletic If the CALLS tree isexcluded from the consensus analysisthe only additional resolved node is theingroup

The signireg cance test of the symm etric-diƒerence metric (Table 6) indicated thatthe CALLS tree is not sim ilar to any othertree beyond random expectation as is thesim ilarity of the CO Iplusmn ALLO ZYM ES pairAny other pair of trees is too sim ilar tohave been drawn at random

Partition homogene ity ETH The null hy-pothesis that the reg ve data partitionswere hom ogeneous was not re jected


FIGURE 1 Phylogenies of the P pustulosus group base d on individual da ta partitions and the CO M -

BIN ED partition (se e Tab le 5) Bootstrap values are given above the bran ch and decay values below


FIGURE 2 Fre quency distributions of total support values from 100 ran dom ize d m atrices Arrows indi-

cate ob se rve d tota l support value s which lie signi reg can tly outside of the distribution of value s from random

m atrices


TABLE 6 Prob ab ilitie s (an d associate d sym m etric-diƒe rence distan ces in pare ntheses) that a pa ir of

tre es with 10 term inals are n o m ore similar than a pair of tree s drawn from a random distribution of

nonb inary tre es (H endy et al 1984) The Bonferroni-corrected critical value for a tab le-wide alph a of 005

was 0010 Aste risk indicates signi reg can t value

Topology Com bined12S C O I Allozym es Calls

C O I 00001 (3)

Allozym es 00047 (6) 00180 (7)

C alls 0127 (9) 00548 (8) 0259 (11)

M orphology 00002 (4) 00010 (5) 00047 (6) 0127 (9)

(P 5 0389) Results from pairwise tests(Tab le 7) indicate that the null hypothesiswas not re jected except for the CALLS plusmnMO RPHOLOG Y com parison These twopartitions have the fewest characters

TABLE 7 Prob ab ility values from pairwise

partition-h om ogene ity tests (1000 ran dom

partitions) for all data partitions The Bonferroni-

corrected critical vlaue for a table-wide a lpha of

005 was 0005 A signi reg cant value (aste risk) indi-

cate s hete rogen eity be tween paired data partition s

12S C O I Allozym e s Calls

CO I 0724

Allozym es 0570 0749

Calls 0293 0440 0502

M orph ology 0719 0202 0452 0002

Compatib ility of data with suboptima ltrees ETH Tem ple ton tests (Table 8) indicatethat all data partitions are incom patiblewith the CALLS tree Additionally thetwo largest data partitions 12S andCO MBINED are incompatib le with theALLO ZYM ES trees All other data parti-tions are com patible with the remain ingsuboptim al trees

Interestingly the sam e incompati-b ilities were obtained from thecom pare -2 tests (Table 9) In addition the four sm allest nonsignireg cant prob-ab ilities in Table 8 were found tobe signireg cant by the com pare-2test (CALLS plusmn COI CO Iplusmn ALLOZYMES CO MBINED plusmn M ORPHOLO GY and 12S plusmnM O RPHO LO G Y) By this test all data

TABLE 8 Results from Tem pleton tests under the null hypothe sis that a data partition is equally com -

patib le with a suboptim al tree The Bonfe rroni-corrected critical value for a tab le -wide a lpha of 005 was

00029 In e ach cell the sam ple size an d W ilcoxonrsquos T are separated by a com m a on the reg rst line an d the

probability (on e- taile d test) is given be low For n 100 the probab ility was taken from Tab le D 18 in Zar

(1974 ) in terpolation was performe d as n eeded for n $ 100 the normal approxim ation was used Aste risk

indicates sign ireg can t value

Alternative tree

Partition C om b 12S C O Ia Allozym es C allsa M orphologya

Com bine d ETH 74 13365 118 1913 245 30475 87 12915

025 00001 00001 00042

12S ETH 38 351 63 480 149 6825 47 312

025 00001 00001 00038

CO Ia 24 125 ETH 63 7135 52 220 56 644

025 001 00001 010

Allozym es 7 95 10 17 ETH 15 0 5 25

025 010 00001b 010

Callsa 11 22 12 8 11 25 ETH 10 185

010 005 025 010

M orph ologya 1 0 4 0 4 0 10 0 ETH

05 010 010 000098 b

a In cases whe re m ultiple eq ually parsimon ious tre es were com pared the largest prob ab ility value (least

likely to re je ct) is reported However in each case all values either uniform ly reject or fail to re ject the null

h ypothesisb Because sufficiently accurate table values were n ot availab le the sign test was pe rform ed


TABLE 9 Results from com pare-2 perm utation tests under the n ull hypothe sis that a da ta partition is

e qually com patib le with an alternative suboptim al tree The Bonfe rroni-corrected critical value at which a

table-wide alpha of 005 was obtained was 00038 1000 or 5000 replicate s were used as described in the

text Asterisk indicate s sign ireg can t value

Alternative tree

Partition C om b 12S CO I Allozym e s Calls M orphology

C om bined ETH 0210 00002 00002 00002

12S ETH 0246 00002 00002 00008

C O I 0201 ETH 00030 00002 0010

Allozym es 0206 0070 ETH 00002 00366

C alls 0014 00002 0093 ETH 0021

M orphology 0599 0045 0101 00002 ETH

partitions are extrem ely incompatib lewith the CALLS tree

Strength of support for suboptima l trees ETHThe results of the constrained-tree T-PTP(Tab le 10) were consistent with those ofthe compare-2 tests (Table 9) That is inall cases (11) in which the compare-2tests indicated signireg can t incom patib ilitythe constrained-tree test showed no sig-nireg cant support for the suboptimal treeConverse ly in all cases in which theconstrained-tree test indicated signireg can tdata support for an alternative tree thecompare-2 results showed compatib ilitywith the suboptimal tree

Certain data partitions providedsupport for suboptim al trees (Tab le 10)Am ong the larger data partitions CO M -BINED and 12S provide signireg cant signalfor the COI tree and vice-versa TheCOM BINED tree is strongly supportedby CO I ALLOZYMES and MO RPHOL-

O G Y but not CALLS (which supports nosuboptim al tree)

D ISCUSSION

Incongruence

Overall the tests indicate that eachdata partition is signireg can tly (non-randomly) structured (PTP tests) andeach strongly supports its own short-est tree (constrained-tree T-PTP totalsupport test) Do the phylogenies derivedfrom these partitions disagree Thisdepends on what one m eans by disagree-m ent It has been argued (Barrett e t al1991) that strict consensus trees are con-servative and mask estim ates of relation-sh ip and our results support th is claim the strict consensus tree is unresolvedexcept for the P petersi plusmn freibergi clade Incontrast the sym m etric-diƒerence testshows that most of the pairwise com-binations of topologies are too sim ilar to

TABLE 10 Results from constraine d- tre e perm utation tests under the null h ypothesis that a data parti-

tion provide s n o signi reg can t support for a suboptim al tree The Bonferroni-corrected critical value at wh ich

a tab le-wide alph a of 005 was ob taine d was 00025 1000 or 5000 replica tes were used as describe d in the

text Aste risk indica tes signi reg can t value

Alternative tre e

Partition Com b12S C O I Allozym es C alls M orphology

Com bined ETH 00004 0555 1000 0195

12S ETH 00002 0333 1000 0187

CO I 00004 ETH 0153 0914 0046

Allozym es 00004 00048 ETH 0945 00034

Calls 0047 0441 0016 ETH 0079

M orphology 00012 0072 0042 1000 ETH


have been chosen at random (except forall pairings of CALLS with other treesand ALLO ZYM ES plusmn CO I) This is sugges-tive of underlying signal in com mon to alldata partitions except for CALLS

The sim ultaneous and pairwisepartition-homogeneity tests are inter-pre ted as indicating that the partitionsare m ostly com binable with the excep-tion of the CALLS plusmn M O RPHO LO G Y pairIt is perhaps no coincidence that theCALLS partition is not com binable withthe m ost internally congruent data parti-tion These results considered toge therindicate that the data partitions are eachwell structured and generally agree intheir estim ates of re lationships

Examination of the com patib ility of adata partition with a suboptim al treegives perhaps a m ore accurate as well asmore com plex picture of the re lations ofdata partitions The CALLS tree has littlesimilarity to other trees All other datapartitions provide no signireg can t supportfor the CALLS tree and indeed all areincom patib le with it Nonetheless theCALLS data partition is com patible withmost of the other trees (except for theCOI tree in the com pare -2 test) eventhough its symm etric-diƒerence distanceto any other tree is large Additionallythe 12S (and COM BINED) partition isincom patib le with the ALLOZYM ES treeand incompatib le (com pare-2 test) ormarginally com patible (Templeton test)with the M ORPHOLO GY tree but theALLO ZYM ES and M ORPHOLOG Y parti-tions both with few characters are com -patib le with the 12S COM BINED treeunder both tests W e suggest that th is``com binabilityrsquo rsquo is due to the sm all sizeof the partitions That is a small well-structured partition might be expected tobe com patible with the tree derived froma large partition but the large partition isincom patib le with the tree derived fromthe sm all partition If the partitions arecombined the sm all one is eƒective lyswam ped out by the larger one

This possib le eƒect of sm all partitionswas examined using Fisherrsquo s exact test inwhich sm all (ALLO ZYM ES MO RPHOL-

O G Y CALLS ) versus large (12S CO I)partitions were scored as being compat-ible or incom patib le with a suboptimaltree according to Templeton rsquos test Thenull hypothesis of no association betweenpartition size and data com patib ilitywas marginally rejected at P 5 0046However when applied to the results ofthe com pare-2 tests the null hypothesiswas not re jected (P 5 0168) A com para-b le exam ination of other data sets m ightbe enlightening

Our principal goal has been to identifyand localize incongruence in these datapartitions rather than to argue for oragainst combin ing data Nonetheless it isclear that the tree from COM BINED dataset is the best supported of the trees Ifone assum es that a com bined analysisyie lds the best estim ate of the phylogenyit becom es particularly interesting thatom itting the 12S partition (the largest)from the com bined analysis still yie ldsthe CO MBINED tree This result issim ilar to that observed by Olm stead andS weere (1994) Additionally it may be anexam ple of consistency (Huelsenbeck1995) in which the accumulation of suffi-cient data (even in the absence of the 12Spartition) leads the analysis to convergeon the ``correctrsquo rsquo phylogeny

Relationsh ips and Ca ll Evolution

The Physalaemus pustulosus speciesgroup was the reg rst example used toargue for the role of sensory exploitationin sexual selection (Ryan et al 1990b)This hypothesis states sim ply that malesevolve traits to exploit preexisting fem alepreferences The data that test thishypothesis com e from exam ining sex-ually selected male traits and pre ferencesfor those traits in taxa with and withoutthe traits This behavioral in form ation toge ther with an estim ate of phylogeneticre lationships is then used to de term inethe m ost parsim onious in terpre tation ofpatterns of trait and pre ference evolution The sensory exploitation hypothesis pre-dicts that the pre ference existed prior tothe trait while other hypotheses such as


runaway sexual se lection and selectionfor good genes predict coevolution oftrait and preference (Ryan 1990 Kirk-patrick and Ryan 1991)

In P pustulosus fem ales prefer callswith chucks added to the whine andthey also pre fer lower frequency chucksto higher freq uency chucks (Ryan 1980 however the strength of the frequencypre ference is weaker than previouslysuggested [ W ilczynski e t al 1995] ) Phy-sa laemus coloradorum m ales (and all otherPhysa laemus except P freibergi ) do notproduce chucks but females prefer theconspecireg c call to which chucks havebeen arti reg cially added over their unal-tered conspecireg c calls (Ryan and Rand1993) Also P coloradorum females havethe same neural tuning which in P pus-tulosus is thought to guide females towardlower frequency chucks (Ryan et al1990b) G iven the phylogeny presentedherein the m ost parsim onious hypothe-sis is that the preference for the chuckand the neural b ias toward lower fre-quency chucks existed prior to the evolu-tion of the chuck (Fig 3a) This is truewhether the chuck evolved twice inde-pendently in the pustulosus -petersi - frei-bergi clade or once in the com monancestor of that clade with subsequentloss in P petersi (see also Ryan 1996)

The in itial phylogenetic hypothesis forrelationships with in the P pustulosusspecies group which provided the phylo-genetic fram ework for testing the sensoryexploitation hypothesis (Ryan et al1990b) was suggested by Cannatella andDuellm an (1984) This is the sam ehypothesis supported here although wenow recognize additional species Sub-seq uent to these earlier studies we rea-lized the necessity of verifying thephylogenetic hypothesis of Cannate llaand Duellm an (1984) given the criticalnature of this hypothesis in evaluatingthe sensory exploitation hypothesisW hen subsequent data for sensoryexploitation were presented howeverour prelim inary molecular analysis(based on a subset of the allozyme dataand about 400 bp of the 12S gene) sug-

FIGURE 3 S cen arios for evolution of the chuck

com ponen t of the call m apped onto alternative

tree s G ene ralized geographic distribution s of taxa

are presen ted (a) Tree supported by a prelim inary

analysis (Ryan 1996) (b ) Tree favored by present

analysis

gested a tree m atching the COI topology(Fig 3a) in which P pustulosus was thesister species to all other species in theingroup (Ryan and Rand 1993) This treeproduced an unexpected biogeographicpattern in which P petersi was m oreclose ly re lated to species on the otherside of the Andes than to its neighbor P pustulosus

Pom iankowski (1994) suggested thatth is pre lim inary phylogeny com plicatedsupport for the sensory exploitationhypothesis However he did notcom m ent upon additional exam ples ofsensory exploitation in the group thatwere not ``com plicatedrsquo rsquo by the pectinateand prelim inary phylogeny (e g fem aleP pustulosus prefer the ir own calls withthe am plitude-m odulated pre reg x of P pustula tus ) The present phylogeneticanalysis yields the sam e topology that


was originally and clearly used to arguefor sensory exploitation as an im portantforce in sexual selection in th is speciesgroup

Although the present analysis (see alsoRyan and Rand 1995 Ryan 1996) hasreturned to the schem e of relationships(Fig 3b ) in which P pustulosus and P petersi (and P freibergi ) form a clade theevolutionary scenario is m ore am biguousOne most parsim onious in terpretation isthat the chuck evolved twice but aneq ually parsim onious one is that thechuck evolved once and was lost in Phy-sa laemus petersi This latter in terpre tationconmacr icts with results reported from thepre lim inary data set (Fig 3a) The conmacr ictis exem plireg ed by the trees derived fromthe 12S partition and the CO I partitionand in each it is the re lationsh ip of the P petersi 1 freibergi cluster that diƒers Onecan also view this conmacr ict as a rootingissue if one excludes the outgroupsthe unrooted 12S and COI trees(corresponding to Figs 3a and 3b) are thesam e Relationsh ips am ong the out-groups becom e im portant and we areexpanding the sam ple of outgroup taxa

Behaviora l Characters in PhylogenyEstimation

Diƒerences of opinion exist aboutwhether behavioral characters m ight beexpected to be re liab le in phylogeneticanalysis (G ittleman et al 1996 M artins1996 Ryan 1996) De Q ueiroz and W im -berger (1993) and W im berger and deQueiroz (1996) have argued that there isno reason to expect that behavioral char-acters should in general be poor indica-tors of phylogenetic re lationships On theother hand certain classes of behavioralcharacters such as mate-recognitionsignals m ay evolve rapidly (Ryan et al1990a) Rapid evolution m ight increasehom oplasy obscure the ``true rsquo rsquo phylogen-etic signal and even suggest a m isleadingsignal Thus one might hypothesize thatrapidly evolving characters involved inbehavioral display are less re liab le inphylogeny estimation (but see Foster et

al 1996) For example in male cricketsthe call is often the reg rst phenotype todiverge am ong lineages (S haw 1996a)and in the cricke t genus Laupa la there isa lack of congruence between the mtDNAhaplotype phylogeny and taxonomicspecies as de reg ned by song type (S haw1996b) Likewise Ryan et al (1996)showed for 30 populations of Physalaemuspustulosus along a 5000-km transect thatcall sim ilarity and genetic (allozym e)sim ilarity covary only slightly signireg -cantly after the eƒects of geographicproximity are controlled also call sim i-larity and geographic proximity arestrongly corre lated when controlling forallozyme similarity

Although the evolutionary lab ility ofthe call characters is a possib le explana-tion for the incongruence of CALLS there are two other explanations One isthe sm all number of characters whichsuggests that the apparent incongruenceis due to sampling error M ORPHO L-O G Y is also small but is internally con-sistent and also com patib le with m ostother partitions Perhaps the re levantparam eter is not the number of charac-ters but the number of in form ative char-acter states The CALLS partition has alarger num ber of such character statesthan does MO RPHOLOG Y because ofthe way the continuous data were madediscre te There is som e indication thatcoding procedures that m axim ize thenum ber of inform ative charactersincrease the m easure of phyogeneticsignal in a data se t (W iens 1995) A m oregeneral consideration of these issuesusing m ultiple data se ts is desirable

In the P pustulosus group the incon-gruence exhibited between the CALLSdata partition and all others and thegeneral congruence am ong the other par-titions suggest that the call charactersif considered alone m islead the phylo-genetic analysis The only se t of relation-sh ips with which the CALLS partitionagrees with all other data partitions isthe P petersi plusmn freibergi clade a pair ofcryptic species that was considered onespecies based on external m orphology


(Cannate lla and Duellm an 1984) W eargue that these limited data indicatethat the hom ologous sim ilarity in calls ofrecently separated species is q uickly lostas the species diverge However PTPtests suggest that the call characterspossess signireg cant phylogenetic signal this m ight result from correlations amongthe characters that produce structure inthe data even though that structure doesnot re macr ect phylogeny This observationcoupled with the preceding conclusionsis consistent with observation of strongse lection on the call signal in Physa laemus(Ryan 1985) It m ay be that sexuallyse lected character complexes associatedwith evolving signal-rece iver system s willbe generally unsuitable for use in phy-logeny estim ation However additionalstudies are needed to determ ine the gen-erality of th is conclusion

ACKNOW LEDGM ENTS

W e thank Jim M cG uire S teve Poe M ary M cKi-

trick Allan Larson and espe cially Alan de Q ueiroz

for com m ents on the m anuscript Th e 12S and 16S

sequen ce data were collected by M arty Badgett

David S woƒord allowed us to use PAUP Speci-

m en s were len t by the following curators John

Cadle W illiam E Duellm an David G ood and

Ch arles W M yers W e also than k the n um e rous

pe ople who aided in the collection of tissue sam ples

an d logistica l aspe cts of the reg e ld work these

pe rson s have be en acknowledge d in previous

papers

REFERENCES

BARRETT M M J DONO GHUE AND E S OBER 1991

Against con sensus S yst Biol 40 486 plusmn 493

BREMER K 1994 Branch support and tre e stability

C ladistics 10 295 plusmn 304

BULL J J J P H UELSENBECK C W C UNNING HAM

D L S WOFFORD AND P J W ADDELL 1993 Par -

titioning and com bining data in phylogene tic

an alysis S yst Biol 42 384 plusmn 397

CANNATELLA D C AND W E DUELLM AN 1984

Le ptodactylid frogs of the Physa laem us pustulosus

group C opeia 1984 902 plusmn 921

CHIPPIND ALE P T AND J J W IENS 1994 W eigh -

ting partition ing and com bin ing characte rs in

ph ylogen etic an alysis S yst Biol 43 278 plusmn 87

COCROFT R B AND M J RYAN 1995 Patterns of

advertisem e nt ca ll evolution in toads an d ch orus

frogs Anim Beh av 49 283 plusmn 303

DE Q UEIROZ A M J DONOGHUE AND J K IM

1996 S eparate versus com bined analysis of

phylogene tic e vidence Ann u Rev Ecol S yst

26 657 plusmn 681

DE Q UEIROZ A AND P H W IMBERG ER 1993 Th e

usefu lness of beh avior for ph ylogen y e stimation

Leve ls of h om oplasy in beh avioral an d m orpho-

logical characters Evolution 47 46 plusmn 60

D ING ERKUS G AND L D UHLER 1977 Enzym e

clearing of alcian b lue stained whole sm all verte-

brates for de m onstration of cartilage S tain Tech

52 229 plusmn 232

EERNISS E D J AND A G KLUGE 1993 Taxon om ic

con gruence versus tota l evide nce an d am n iote

phylogeny in ferred from fossils m olecu les an d

m orphology M ol Biol Evol 10 1170 plusmn 1195

ERIKS SO N T 1996 AutoDecay version 295 Bot-

an iska Institution en S tockholm U niversity S tock-

holm

FAITH D P 1991 C ladistic perm utation tests for

m onoph yly an d non m on oph yly S yst Zool

40 366 plusmn 375

FAITH D P AND J W O BALLARD 1994 Length

diƒerences an d topology-depe ndent tests a

response to KaEgrave llersjoEgrave e t al C ladistics 10 57 plusmn 64

FAITH D P AND P S C RANSTON 1991 C ould a

cladogram th is short have arisen by chan ce

alone O n permutation tests for cladistic struc-

ture C ladistics 7 1 plusmn 28

FARRIS J S M KAEgrave LLERSJOEgrave A G KLUGE AND C

BULT 1994 Te sting sign ireg can ce of incongruence

Cladistics 10 315 plusmn 319


BULT 1995 Constructing a signi reg cance test for

incon gruen ce S yst Biol 44 570 plusmn 572

FELSENS TEIN J 1985 Con reg dence lim its on phylo-

gen ies with a m olecu lar clock S yst Zool 34 152 plusmn

161

FOSTER S A W A C RESKO K P JOHNS ON M U

TLUSTY AND H E W ILLM OTT 1996 Pa tterns of

hom oplasy in behavioral evolution Pages 245 plusmn 269

in Hom oplasy Th e recurren ce of sim ilarity in

evolution (M J Sande rson and L Huƒord eds)

Academ ic Press S an Diego

G ITTLEMAN J L C G ANDERSON M KOT AND

H-K LUH 1996 Phyloge netic lab ility and rate s of

evolution a com parison of behavioral m orpho-

logical an d life history traits Page s 166 plusmn 205 in

Phylogenies and the com parative m e thod in

an imal beh avior (E P M artins ed) O xford Univ

Press Ne w York

H ENDY M D C H C L ITTLE AND D PENNY 1984

Com paring tree s with pendant vertice s labelled

S IAM J Appl M ath 44 1054 plusmn 1065

H ILLIS D M 1991 Discrim inatin g between phylo-

gen etic sign al and random noise in DNA

sequence s Pages 278-294 in Ph ylogen etic an alysis

of DNA seq uen ces (M M M iyam oto an d J Crac-

raft eds) O xford Univ Press O xford New York

H ILLIS D M AND J J BULL 1993 An em pirica l test

of bootstrapping as a m ethod for asse ssing con-

reg de nce in phylogene tic an alysis S yst Biol

42 182 plusmn 192

H ILLIS D M B K M ABLE A LARSON S K DAVIS

AND E A ZIM MER 1996 Nucleic acids IV


S equen cin g and cloning Page s 321 plusmn 381 in M olec-

ular system atics 2nd edition (D M Hillis C

M oritz and B K M ab le eds) S inauer Associates

S un derland M assachusetts

H UELSENBECK J P 1991 Tree- length distribution

skewne ss An indica tor of ph ylogen etic in form a-

tion S yst Zool 40 257-27 0

H UELSENBECK J P 1995 Performance of ph ylogen -

etic m e thods in simulation Syst Biol 44 17 plusmn 48

H UELSENBECK J P AND J J BULL 1996 A likeli-

hood ratio test to de tect conmacr icting phylogene tic

signa l Syst Biol 45 92 plusmn 98

INTERNATIONAL UNIO N OF B IOCHEMISTRY

N OMENCLATURE C OMMITTEE 1984 Enzym e

nom en cla ture Acade m ic Press O rlando Florida

IRWIN R E 1996 Th e ph ylogen etic conten t of avian

courtsh ip display an d son g evolution Page s 234 plusmn

252 in Phyloge nies and the com parative m ethod

in an imal beh avior (E P M artin s ed) O xford

Un iv Press New York

KAEgrave LLERSJOEgrave M J S FARRIS A G KLUGE AND C

BULT 1992 S kewne ss and perm utation C ladistics

8 275 plusmn 287

KENNEDY M H G S PENCER AND R D G RAY

1996 Hop step an d gape D o the social disp lays

of the Pelecan iforme s remacr ect phylogeny An im

Behav 51 273 plusmn 291

K IRKPATRICK M AND M J RYAN 1991 The

paradox of the lek and the evolution of m ating

prefe ren ces Nature 350 33 plusmn 38

LARSON A 1994 The com parison of m orphological

an d m olecular data in ph ylogen etic system atics

Pages 371 plusmn 390 in M olecu lar ecology an d evolu-

tion Approache s and application s (B S ch ie rwa-

ter B S treit G P W agne r and R D eS alle e ds)

B irkhaEgrave use r Verlag Basel

LUTZONI F 1997 Phyloge ny of liche n- and non -

lichen - form ing om ph alin oid m ush room s an d the

utility of testin g for com binability am ong m ultiple

da ta sets S yst Biol 46 373 plusmn 406

M ADDISO N W P AND D R M ADDISON 1992

M acC lade version 3 S inauer Associa tes S un der-

land M assach usetts

M ADDISO N W AND M S LATKIN 1990 Pa rsimon y

recon structions of ance stra l states do not depend

on the re lative distance s between linearly-

orde red character states S yst Zool 39 175 plusmn 178

M ADDISO N D R D L S WOFFORD AND W P

M ADDISON 1997 NEXUS An extensib le reg le

form at for system atic in form ation S yst Biol

46 590 plusmn 621

M ARTINS E P (ed) 1996 Ph ylogen ies and the

com parative m ethod in an imal behavior O xford


M ASO N-G AMER R J AND E A KELLO GG 1996

Testing for ph ylogen etic con macr ict am on g m olecu-

lar data sets in the tribe Triticeae (G ram ine ae )

S yst Biol 45 522 plusmn 543

M URPHY R M J W S ITES JR D G BUTH AND

C H H AUFLER 1996 Proteins Isozym e e lectro-

ph oresis Page s 51plusmn 120 in M olecu lar system atics

2nd e dition (D M Hillis C M oritz and B K

M ab le eds) S inaue r Associates S underland

M assachusetts

O LMSTEAD R G AND J A S WEERE 1994 Com bin-

ing da ta in ph ylogen etic system atics An em piri-

cal approach using three m olecular data sets in

the S olan ace ae S yst Biol 43 467 plusmn 481

PALUMBI S R 1996 Nucle ic acids II Th e poly-

m erase ch ain reaction Page s 205 plusmn 247 in M olecu-

lar system atics 2nd edition (D M Hillis C

M oritz an d B K M ab le eds) S inauer Associates

Sunderlan d M assach usetts

PENNY D AND M D H ENDY 1985 Th e use of tree

com parison m etrics Syst Zool 34 75 plusmn 82

POE S A 1996 Data set incongrue nce and the phy-

logen y of crocodilian s S yst Biol 45 393 plusmn 414

POMIANKOWSKI A 1994 Swordplay an d sen sory

bias Nature 368 494 plusmn 495

RAND A S M J RYAN AND W W ILCZYNSKI 1992

S ign al redun dan cy an d rece ive r permissive ness

in acoustic m ate recogn ition by the tuAcirc n gara frog

Physa la emus pustulosus Am Zool 32 81 plusmn 90

R ICE W R 1989 An alyzin g tab les of sta tistica l tests

Evolution 43 223 plusmn 225

RYAN M J 1980 Fem ale m ate ch oice in a n eotrop i-

cal frog S cien ce 209 523 plusmn 525

RYAN M J 1985 The tuAcirc ngara frog A study in

sexual selection and com m unica tion Univ

Ch icago Press Ch icago

RYAN M J 1990 Se nsory system s sexual selection

and sensory exploitation O xford S urveys Evolut

Biol 7 157 plusmn 195

RYAN M J 1996 Ph ylogen etics in beh avior som e

cautions and expectations Pages 1plusmn 21 in Phylo-

gen ies and the com parative m ethod in an imal

be havior (E P M artin s ed) O xford Univ Press

Ne w York

RYAN M J M D TUTTLE AND A S RAND 1982

Se xual advertisem ent and bat preda tion in a Neo-

tropical frog Am Nat 119 136139

RYAN M J R B C OCROFT AND A S RAND 1990a

Th e role of e nvironm en tal selection in intraspe-

ci reg c divergen ce of m ate recognition signa ls in the

cricket frog Acris crep ita ns Evolution 44 1869 plusmn

1872

RYAN M J J H FOX W W ILCZYNSKI AND A S

RAND 1990 b S exual se lection for sen sory e xploit-

ation in the frog Physa la emus pustulosus Nature

343 66 plusmn 67

RYAN M J AND A S RAND 1993 Se xual se lection

and sign al e volution The ghost of b iase s past

Ph ilos Trans Royal S oc S er B 340 187 plusmn 195

RYAN M J AND A S RAND 1995 Fem ale

response s to an cestral advertisem e nt ca lls in

tuAcirc ngara frogs S cien ce 269 390 plusmn 392

RYAN M J A S RAND AND L W EIGT 1996 G eo-

graph ic variation in a llozym e and advertisem e nt

calls in the tuAcirc n gara frog Physalaemus pustulosus


RYAN M J AND W W ILCZYNSKI 1988 Coevolu-

tion of se nde r and rece ive r Eƒe ct on local m ate

preference in cricket frogs Scien ce 240 1786 plusmn 1788

RYAN M J AND W W ILCZYNSKI 1991 Evolution

of in traspecireg c variation in the advertise m ent call

of a cricke t frog (Acris crep ita ns Hylidae) Biol

Journal Linn S oc 44 249 plusmn 271


S HAW K L 1996a Se quential radiation s and pat-

terns of speciation in the Hawaiian cricket genus

Laupa la infe rred from DNA seq ue nces Evolution

50 237 plusmn 255

S HAW K L 1996 b Polygen ic inh eritance of a

behavioral ph enotype Interspecireg c gen etics of

son g in the Hawaiian cricket genus Laupa la Evol-

ution 50 256 plusmn 266

S TEEL M A AND D PENNY 1993 D istributions of

tree com parison m e trics ETH S om e new results Syst


S WOFFORD D L 1991 W he n are ph ylogen y esti-

m ate s from m olecu lar an d m orphological data

incongrue nt Pages 295 plusmn 333 in Phylogene tic

an alysis of DNA seq uen ces (M M M iyam oto and

J C racraft eds) O xford Univ Press New York

S WOFFORD D L 1993 PAUP Phylogene tic ana lysis

using parsimon y version 31 Formerly distrib -

uted by the Illinois Natural History S urvey

C ham paign Illinois

S WOFFORD D L J L THO RNE J FELSENSTEIN

AND B M W IEGMANN 1996 Th e topology-

de pen de nt perm utation test for m onoph yly doe s

not test for m on ophyly S yst Biol 46 575 plusmn 579

TEMPLETON A R 1983 Ph ylogen etic in ference

from restriction en don uclease cleavage site m ap s

with particu lar reference to the hum ans and apes


W HEELER W C AND D G LADSTEIN 1992

M ALIG N Am erican M useum of Natural History

New York New York

W IENS J J 1995 Po lym orphic ch aracters in ph ylo-

genetic analysis Syst Biol 44 482 plusmn 500

W ILCZYNSKI W A S RAND AND M J RYAN 1995

The processing of spe ctra l cues by the call

an alysis system of the tuAcirc n gara frog Physa laemus

pustulosus Anim Beh av 49 911 plusmn 929

W IM BERGER P H AND A DE Q UEIROZ 1996 Com -

paring beh avioral and m orphological ch aracters

as indicators of ph ylogen y Pages 203 plusmn 233 in Ph y-

loge nies an d the com parative m e thod in an imal

behavior (E P M artins ed) O xford Un iv Press

New York

ZAR J 1974 Biostatistica l an alysis Pren tice Hall

Englewood Cliƒs New Jersey

Received 4 June 1997 a ccep ted 10 August 1997

Associa te Editor A Larson

APPENDIX 1

C olle ction localitie s for tissue sam ples Physa -

laem us co loradorum ETH E cuador Pich incha Tinalan -

dia an d vicin ity Ph ysala emus enesefae ETH Ven ezue la

C alab ozo Ph ysala emus eph ipp ifer ETH B razil ParaAcirc

vicinity of Belem Physa laemus freib ergi ETH Pe ru

M adre de Dios Tam bopata Explore rrsquos In n Ph ysa-

la emus petersi ETH E cuador Nap o Jatun Sacha Ph ysa-

la emus pustulatus ETH Ecuador El O ro ap prox 10 km

E Pasaj e Physa laem us pustulosus ETH Panam a Panam a

G am boa Ph ysala emus sp AETH Brazil Roraim a

M aca jai 66 km S Boa Vista Ph ysala emus sp

B ETH Peru Lam baye que O lmos 85 km N M otupe

Ph ysala emus sp C ETH Ecuador Esm eraldas Porto

Viejo

APPENDIX 2

M orph ologica l ch aracters used in phyloge netic

analysis follow M ost of these are discussed in C an-

n atella an d Duellm an (1984)

1 Re lative len gth of reg rst and se con d reg nger 0

First reg nger shorter than second reg n ger when

adpressed 1 First reg n ge r equal in length or

lon ge r than second whe n adpressed

2 Tarsal tubercle 0 Presen t 1 Ab sent

3 Flank gland 0 Absen t 1 Broad an d macr at con-

cealed ben eath skin 2 Narrow shorter and

protruding above skin

4 Parotoid gland 0 Ab sent 1 Prese nt

5 Skin texture 0 S m ooth at tim es with folds 1

W arty tube rculate

6 Sh ape of snout 0 S nout not protruding beyon d

tip of upper jaw 1 S nout protruding beyon d tip

of upper jaw

7 Black inguinal blotches 0 Ab sen t 1 Presen t

8 Dentigerous proce sses of vom er 0 Flat an d

wide 1 Thin and spikelike

9 Te eth on the m axilla and prem axilla 0 Presen t

1 Ab sen t

10 Sh ape of the stalk of the alary process of the

h yoid 0 S talk wide 1 S talk very narrow

11 In sertion of petroh yoideus an terior m uscle 0

Along m idlin e of hyoid plate 1 Along edge of

h yoid plate

12 Anterior proce ss of h yale 0 W ell develope d

and prom inen t 1 W e akly de velope d

APPENDIX 3

Data m atrix used in ph ylogen etic analysis follows

Analysis of the ALLO ZYM E and CALLS partitions

req uires step m atrice s wh ich are availab le in the

NEXUS reg le a t http wwwutexasedude ptssystb iol


Allozyme Morphology Calls 125 begins sp A 1311101001111111251101111 000000000000 1 0 AAAGGTTTGGTCCTAGCCTTGAA ephippifer 2112201001211142137242211 000000000000 2 AAAGGTTTGGTCCTAGCCTTGAA enesefae 6712 51102171111726226221119 000000000000 4 8 7 AAAGGTTTGGTCCTAGCCTTTAA pustulosus 531511101051207432121525544 101110011001 1 0 6 AAAGGTTTGGTCCTAGCCTTAAA petersi 221230102031105341413323322 101111111001 0 AAAGGTTTGGTCCTAGCCTTAAA freibergi 22133 0002041101331413314433 101111111001 0 AAAGGTTTGGTCCTAGCCTTAAA coloradorum 463 3 22003 061401623 226138877 011110010110 2 AAAGGTTTGGTCCTAGCCTTTAA pustulatus 3 4 3 3211140603 0632 4234 33 66 55 012110010110 1 6 AAAGGTTTGGTCCTAGCCTTGAA sp B 4 514412 00 07110152 562 5127 766 011110010110 7 4 2 AAAGGTTTGGTCCTAGCCTTGAA sp c 4 42 3 210000 811018277 8 42 9 96 8 012110010110 AAAGGTTTGGTCCTAGCCTTGAA

sp A ATCAATTATTACTTAATATACACATGCAAGTATCCGCACCCCTGTGAAAACGCCCTTTACT--CCCCC-ACGGGACAAGGAGCTGGTATCAGlCCCGAAA ephippi fer ATCAATTATTACTTAATATACACATGCAAGTATCCGCACCCCTGTGAAAACGCCCTTTACT--CCCCC-ACGGGACAAGGAGCTGGTATCAGGCCCGAAA enesefae ATCAATTATTACTTAATMACACATGCAAGTATCCGCACCCCTGTGAAAACGCCCTTTAITTT--CTC-ACGAAACAAGGAGCTGGTATCAGGCCCGAAT

pustulosus ATCAATTATTTCTTAATATATACATGCAAGTCTCAGCCCCCCTGIGAAAACGCCC-TTAAATA--CCCTCTAGGATAAGGAGCTGGTATCAGGCACGAAA petersi GTCAATTATTTCTTAATATACACATGCAAGTATCAGCCCTCCTGTGAAAACGCCC-TTAATTTTCCCCATTAGGATAAGGAGCTGGTATCAGGCACAAAA freibergi GTCAATTACTTCTTAATATACACATGCAAGTATCAGCCCTCCTGTGAAAACGCCC-TTATTTTTCCCCATTAGGGATAGGAGCTGGTATCAGGCACAAAA coloradorum GTCAATTACTTCTTAATATACACATGCAAGTATCCGCCCCCCTGTGAAAACGCCC-TTAAAT-- -CCCAATAGGATAAGGAGCTGGTATCAGGCACGAAA pus tu latus GTCAA TTACTTCTTAA T ATACACA TGCAAGTATACGCTCCCCTGTGAAAACGCCC -TT AAAT-- -CCCTATAGGATAAGGAGCTGGTA TCAGGCACGAAA sp B ATCAATTATTTCTTAACATACACATGCAAGTATTCAGCCCCCTGTTGAAACGCCC-TTAAAT---CCCTATAGGATAAGGAGCTGGTATCAGGCACGAAA sp C ATCAATTATTTCTTAATATACACATGCAAGTATCCGCTACCCTGTGAAAACGCCC-TTAAAA---CCCTATAGGATAAGGAGCTGGTATCAGGCACGAAA

sp A TCCTGCCCAAGACACCTAGCTATGCCACACCCACAAGGGAACT-CAGCAGTGATTAACATTAAGTATAAGCGACACGTTGACTTAGTCAAAGTAAAGAGA ephippifer TTCTGCCCAAGACACCTAGCTATGCCACACCCACAAGGGAACT-CAGCAGTGATTAACATTAAACATAAGCGACACGTTGACTTAGTTAAAGTAAAGAGA enesefae TTCTGCCCAAGACACCTAGCTATGCCACACCCACAAGGGAACC -CAGCAGTGATTAACATTAAACATAAGCGACACGTTGACTTAGTTAAAGTAAAGAGA pustulosus TTCTGCCCAAAACACCTAGCTATGCCACACCCACAAGGGAATT-CAGCAGTAATTAACATTGAATATAAGCGCCAGCTTGATTCAGTTAAAGAAAATAGA petersi TA-TGCCCAAAACACCTAGCTATGCCACACCCACAAGGGAACT-CAGCAGTGATTAACATTAAACATAAGCGCCAGCTTGATTTAGTTAAAGAAAATAGA freibergi TA-TCCCCAAAACACCTAACTACTCCACACCCACAAGGGAACT-CAGCAGTGATIAACATTAAATATAAGCGCCAGCTTGATTTAGTTAAAGAAAACAGA coloradorum TTCTGCCCAAGACACCTAGCTATGCCACACCCACAAGGGAACTTCAGCAGTGATTAACATTGAACATAAGCGACACGTTGACTCAGTTAAAGAAAAGAGA pustulatus CTCTGCCCAAAACACCTAGCTATGCCACACCCACAAGGGAATT-CAGCAGTGATTAACATTGAACATAAGCGACAGCTTGACTCAGTTAAAGAAGAGAGA sp B TTCTGCCCAAAACACCTAGCTATGCCACACCCACAAGGGAACT-CAGCAGTGATTAACATTGAGCATAAGCGATAGCTTGACTCAGTTAAAGAAAAGAGA sp C TTCTGCCCAAAACACCTAGCTATGCCACACCCACAAGGGAACT-CAGCAGTGATTAATATTGAGCATAAGCGTCAGCTTGACTCAGTTAAAGAAAAGAGA

sp A ACCGGCTAATCTGGTGCCAGCCGCCGCGGTTACACCAAGTGGTTCAAATTGATTCTTATCGGCGTAAAGCGTGATTAAAGTATTATATAATTGCAGTTGA ephippifer ACCGGCTAATCTGGTGCCAGCCGCCGCGGTTACACCAAGTGGTTCAAATTGATTCTTTTCGGCGTAAAGCGTGATTAAAGTATTATATAATTGCAGTTGA enesefae ACCGGCTAATCTGGTGCCAGCCGCCGCGGTTACACCAGATGGTTCAAATTGATTCTTATAGGCGTAAAGCGTGATTAAAGTATTATATAATTGTAGTTGA

pustulosus GCCGGCTAATCTGGTGCCAGCCGCCGCGGTTACACCACGTGACTCAAATTGATTTCATTCGGCGTAAAGCGTGATTTAAGCACTAAAAAATTAAAGTTAA petersi GCCGGCAAATCTGGTGCCAGCCGCCGCGGTTACACCACGTGACTCAAATTGATTTATATCGGCGTAAAGCGTGATTTAAGAGTATTAAGATTGAAATTAA freibergi GCCGGCAAATCTGGTGCCAGCCGCCGCGGTTACACCAAGTGACTCAAATTGACCTACATCGGCGTAAAGCGTGATTTAAGAGTCTTTAAATTGAAATTAA co loradorum GCCGGCAAATCTGGTGCCAGCCGCCGCGGTTACACCACGTGACTCAAATTGACCTTAGTCGGCGTAAAGCGTGAITAAAGTTTAAACAAATTAAAGTTAA pustulatus GCCGGCAAATCTGGTGCCAGCCGCCGCGGTTACACCACGTGACTCAAATTGACCTCTATCGGCGTAAAGCGTGATTAAAGTTTAAATAAATTGAAGTTGA sp B GCCGGCAAATCTGGTGCCAGCCGCCGCGGTTACACCACGTGACTCAAATTGAACTCAATCGGCGTAAAGCGTGATTAAAGTCCCACAATATTGAAGTTAA sp C GCCGGCAAATCTGGTGCCAGCCGCCGCGGITACACCACGTGACTCAAATTGAATTCAATCGGCGTAAAGCGTGAITAAAGTCTCATGACATTGAAGTTAA

s~ A ACATAAATTAAGCTGTGACACGCTTATTTATCTGAAAACCATAAACGAAAGTTACTTCAATTAACCCCACTTGAACTCACGACAGTTAGGACACAAACTG ephippifer ACATAAATTAAGCTGTGACACGCTTATTTATCTGAAAACCATAAACGAAAGTTACTTCAATTAACCCCACTTGAACTCACGACAGTTAGGACACAAACTG

enesefae ACATAAATTAAGCTGTAACACGCTTATTTATTTGAAAACCATAAACGAAAGTTACTTCAATTAACCCAACTTGAACTCACGACAGTTAGGACACAAACTG pustulosus ACTTAAACTAAGCTGTGACACGCTTCTTTTTAAGAAAACCTAATACGAAAGTTACTTTAATTATTACCACTTGAATTCACGACAATTAGAACACAAACTG

petersi ATTTTAATTAAGCTGTAACACGCTTGTTTTTAAGAAAACCAAACACGAAAGTTATTTCAATTATCTCCACTTGAATTCACGACAATTAGGATACAGACTG freibergi ATTACAATTAAGCTGTAACACGCTTGTTTGTAAGAAAACCTGATACGAAAGTTACTTCAACTTGATCTACTTGAATTCACGACAATTAGGACACAAACTG coloradorum ACTAAAATTAAGCTGTGACACGCTTATTTTAAGGAAAACCTGAAACGAAAGTTACTTTAACTTAATCTACTTGAACTCACGACAATTAGGACACAAACTG pustulatus ACTAAAATTAAGCTGTGACACGCTTATTTTAAAGAAAACCTAATACGAAAGTTACTTTAACTAAATCTACTTGAACTCACGACAATTAGGATACAAACTG sp B ACTAGAACTAAGCTGTGACACGCTTGTTCTTAAGAAAATCTTATACGAAAGTTACTCCAACCAAATCCACTTGAATTCACGACAATTAGGACACAAACTG sp C ACTAGAACTAGGCCGTGACACGCTTGTTCTTAAGAAAACCTGATACGAAAGTTACTTCAACTTGATCTACTTGAATTCACGACAATTAGGACACAAACTG

sp A GGATTAGATACCCCACTATGCCTTAACCGTAAACTTTAACTTACTCTTTAATCGCCAGGGAACTACGAGCAAAGCTTAAAACCCAAAGGACTTGACGGTA ephippi fer GGATTAGATACCCCACTATGCCTTAACCGTAAACTTTAACTTACTCTTTTATCGCCAGGGAACTACGAGCAAAGCTTAAAACCCAAAGGACTTGACGGTA enesefae GGATTAGATACCCCACTATGCCTTAACCGTAAACTTTAACTTACTCTTTAATCGCCAGGGAACTACGAGCAAAGCTTAAAACCCAAAGGACTTGACGGTA pustulosus GGATTAGATACCCCACTATTGC-TAATCGTAAACTTTAACGGACACCTTGCTCGCCCGGGAACTACGAGCAAAGGTTAAAACCCAAAGGACTroACGGTA petersi GGATTAGATACCCCACTATGCC-TAATCGTAAACTTTAATTTACACTAACATCGCCAGGGAACTACGAGCAAAGCTTAAAACCCAAAGGAC~ACGGTA

freibergi GGATTAGATACCCCACTATGCC -TAATCGTAAACTTTAATTTACACTAATATCGCCAGGGAACTACGAGCAAAGCTTAAAACCCAAAGGACTTGACGGTA

c o l oradorum GGATTAGATACCCCACTATGCC-TAATCGTAAACCTTAACMACATTATCATCGCCAGGGAACTACGAGCAAAGCTTAAAACCCAAAGGACTTGACGGTA pustulatus GGATTAGATACCCCACTATGCC-TAATCGTAJACCTTAATTTACATAAATATCGCCAGGGAACTACGAGCAAAGCTTAAAACCCAAAGGACITGACGGTA sp B GGATTAGATACCCCACTATGCC-TAATCGTAAACCTTAACTTACACCAATATCGCCGGGGAACTACGAGCAAAGCTTAAAACCCAAAGGACTTGACGGTA sp C GGATTAGATACCCCACTATGCC-TAATCGTAAACCTTAATTTACACTAACATCGCCAGGGAACTACGAGCAAAGCTTAAAACCCAAAGGACTTGACGGTA

sp A CCCCACATCCACCTAGAGGAGCCTGTCCTATAATCGATAATCCCCGCTTAACCTCACCACCTTTAGC-TACTCAGCCTGTATACCTCCGTCGTCAGCTTA ephippifer CCCCACATCCACCTAGAGGAGCCTGTCCTATAATCGATAATCCCCGCTTAACCTCACCACCTTTAGC-TACTCAGCCTGTATACCTCCGTCGTCAGCTTA enesefae CCCCATATCCACCTAGAGGAGCCTGTCCTGTAATCGATACTCCCCGCTTAACCTAACCACCTTTAGC-TACTCAGCCTGTATACCTCCGTCGTCAGCTTA

p u s tulosus CCCCAAATCCACCTAGAGGAGCCTGTCCTATAATCGATAACCCCCGTTTAACCTCACCACCCCTAG-TTACTCAGCCTGTATACCTCCGTCGTCAGTTTA petersi CCCCAAATCCACCTAGAGGAGCCTGTCCTATAATCGATAACCCCCGTTAAACCTCACCACTTCTAGCTTA-TCAGCCTGTATACCTCCGTCGTCAGCTTA

freibergi CCCCAAATCCACCTAGAGGAGCCTGTCCTATAATCGATAACCCCCGTTAAACCTCACCACTTCTAGCTTA-TCAGCCTGTATACCTCCGTCGTCAGCTTA coloradorum CCCCAAATCCACCTAGAGGAGCCTGTCCTATAACCGATACCCCCCGTTTAACCTCACCACTTTTAGCCT-CTCAGCCTGTATACCTCCGTCGTCAGCTTA pustulatus CCCCAAATCCACCTAGAGGAGCCTGTCCTATAACCGATACCCCCCGTTTAACCTCACCATTTTTAGCTT-CTCAGCCTGTATACCTCCGTCGTCAGCTTA sp B CCCCAAATCCACCTAGAGGAGCCTGTCCTATAACCGATACCCCCCGTTTAACCTCACCATTTTTAGCCT-CTCAGCCTGTATACCTCCGTCGTCAGCTTA

sp C CCCCAAATCCACCTAGAGGAGCCTGTCCTATAATCGATACCCCCCGTTTAACCTCACCATTTTTAGCCTA-TCAGCCTGTATACCTCCGTCGTCAGCTTA




methods and discuss the phylogeneticutility of the advertisem ent calls of thesefrogs

M ATERIALS AND M ETHO DS

S pecim ens were collected in the reg e ldtissues extracted and the voucher speci-mens preserved or prepared as ske le tons(Appendix 1) S pecim ens are deposited atthe United States National Museum andthe Texas M em orial M useum Universityof Texas S ome ske le tal m aterial wasborrowed from the American M useumof Natural History University ofKansas Museum of Natural History the Museum of Com parative ZoologyHarvard University and the LouisianaState University M useum of NaturalScience

Taxon Sampling

The species sam pled are listed inAppendix 1 All known valid species inthe ingroup were sampled we treated apopulation of P petersi that may be refer-ab le to the nom inal taxon P freibergi(Cannate lla and Duellm an 1984) as a dis-tinct taxon Monophyly of the ingroup issupported by four synapom orphies(Cannate lla and Duellman 1984) Out-group taxa were Physalaemus ephipp ifer Physa laemus sp A and Physa laemusenesefa e These species were chosenbecause our preliminary survey of m or-phology and calls among 75 of thespecies suggested that they are the mostsimilar to the pustulosus group in externalmorphology osteology and the generalcharacteristics of the call A more com -prehensive phylogenetic analysis ofrelationships in the genus is in progress

Data Pa rtitions

The following character se ts were desig-nated as data partitions morphologicalcharacters (n 5 12 M ORPHOLOG Y) ad-vertisem ent calls (n 5 12 CALLS ) allo-zym e electromorphs (n 5 27 ALLO -ZYM ES ) DNA seq uence of the cyto-chrom e oxidase I gene (n 5 543 COI)

and DNA seq uence of the sm all sub-unit of the m itochondrial ribosomal gene(n 5 1214 12S ) The combined dataset was designated as CO MBINED(n 5 1808)

M orphological characters (Appendix 2)were taken from dissections of wholespecim ens and alizarin -and-alcian plusmnstained skele tons (Dingerkus and Uhler1977) Although sample sizes of skeletonsfor most species were two or three asurvey of 30 skeletons of Physalaemuspustulosus (Cannate lla and Duellm an 1984) indicated no intraspecireg c polymor-phism in the characters examined andnone was noted in the present study

Advertisem ent calls were recorded inthe reg eld onto m etal tape with either aS ony TCD 5M M arantz PM D 420 orS ony Professional W alkm an using aM E-80 S ennheiser microphone with aK3-U power module and wind screen Tem peratures at the calling sites of eachfrog were recorded and usually were25 6 2degC S uch a small temperature dif-ferential has no substantial in macr uence oncall variation

The advertisement calls of the Physa -laemus pustulosus species group (exceptspecies C) and the three outgroup speciesare all sim ilar in that they are rather longfrequency sweeps W e refer to these callsas whines which describes the sound tothe hum an observer S ome species m ayadd to their call a suffix which isdescribed as a chuck TuAcirc ngara thecom m on nam e for P pustulosus is anonomatopoeia for the whine followed bytwo chucks Because the whine is thecom ponent required for species recogni-tion (Ryan 1985 Rand et al 1992 Ryanand Rand 1995) it is the only call com-ponent considered The whines diƒer intheir spectral properties (the onse t oƒse tand dom inant frequency) as well as inthe duration and shape of the frequencysweep All of the whines have upper har-m onics but in P pustulosus these harmo-nics have no inmacr uence on the calls rsquoattractiveness to fem ales (Rand et al1992 W ilczynski e t al 1995) These har-m onics are not considered here all


















































































TA

BL

E3

S

um

ma

ryo

fca

llsta

tisti

cs

for

the

Ph

ysa

laem

us

pu

stu

losu

sg

rou

pa

nd

clo

se

rela

tiv

es

Th

ev

ari

ab

les

are

tota

ld

ura

tio

no

fca

ll(T

LD

UR

m

se

c)

fre

qu

en

cy

at

on

set

of

ca

ll(I

NH

Z

Hz)

ma

xim

um

fre

qu

en

cy

(MX

HZ

H

z)

tim

eto

the

ma

xim

um

fre

qu

en

cy

(TM

MX

m

se

c)

tim

eto

mid

-fre

qu

en

cy

(TM

HF

HZ

mse

c)

fre

qu

en

cy

at

oƒs

et

of

ca

ll(F

NH

Z

Hz)

do

min

an

tfr

eq

ue

ncy

(DO

MH

Z

Hz)

du

rati

on

of

am

pli

tud

e-m

od

ula

ted

co

mp

on

en

t(A

MD

UR

m

se

c)

rise

tim

e

(RS

TM

in

mse

c)

tim

eto

mid

-ris

e(T

MH

FR

S

mse

c)

fall

tim

e(F

LT

M

mse

c)

an

dti

me

tom

id-f

all

(TM

HF

FL

m

se

c)

Th

em

ea

nis

giv

en

wit

hth

era

ng

eb

elo

w

Th

ela

tte

rin

pa

ren

the

se

sfo

llo

win

gth

em

ea

nis

the

ch

ara

cte

r-sta

teco

de

se

eM

ate

ria

lsa

nd

Me

tho

ds

Th

ev

ari

ab

les

are

dis

cu

sse

din

mo

red

eta

ilin

Co

cro

ft

an

dR

ya

n(1

99

5)

Sp

ecie

sn

TL

DU

RIN

HZ

MX

HZ

TM

MX

TM

HF

HZ

FN

HZ

DO

MH

ZA

MD

UR

RS

TM

TM

HF

RS

FL

TM

TM

HF

FL

sp

A1

339

(j)

812

(d)

876

(a)

65

(k)

16

0(j

)46

0(j

)9

83

(v)

0(

)9

46

(h)

56

6(h

)25

16

(n)

711

(h)

02

34

plusmn4

47

800

plusmn84

08

00

plusmn9

20

43plusmn

10

015

0plusmn1

87

40

0plusmn

520

767

plusmn13

62

0plusmn

06

4plusmn

11

637

plusmn8

317

4plusmn

339

31plusmn

11

6

eph

ipp

ifer

12

66

(g)

900

(h)

944

(e)

62

(j)

14

0(i

)57

6(q

)9

44

(t)

0(

)8

35

(g)

39

4(e

)17

74

(i)

606

(g)

238

plusmn30

88

40

plusmn1

00

08

40

plusmn1

04

04

3plusmn

81

11

2plusmn1

62

52

0plusmn

60

08

45

plusmn10

25

0plusmn

04

8plusmn

97

4plusmn

56

12

9plusmn

236

27plusmn

10

7

enes

efae

17

47

(z)

944

(j)

976

(g)

16

2(z

)38

6(z

)69

2(z

)9

62

(u)

38

48

(z)

30

15

(z)

166

6(z

)44

57

(z)

20

37

(z)

06

31

plusmn9

03

880

plusmn1

04

09

20

plusmn1

04

08

1plusmn

28

730

0plusmn

45

664

0plusmn

720

844

plusmn13

84

23

8plusmn

76

023

0plusmn

40

71

25

plusmn24

237

2plusmn

546

99plusmn

33

8

pu

stu

losu

s1

370

(k)

884

(h)

884

(a)

0(a

)12

4(h

)48

4(k

)7

12

(j)

223

(b)

240

(a)

79

(a)

34

28

(s)

17

50

(v)

02

52

plusmn49

68

40

plusmn9

60

840

plusmn9

60

0plusmn

08

7plusmn1

75

44

0plusmn

60

06

05

plusmn8

83

119

plusmn3

27

93

plusmn60

62

2plusmn1

77

23

6plusmn

45

01

06plusmn

25

2

pet

ersi

12

46

(f)

12

20

(x)

12

20

(w)

0(a

)2

8(b

)38

4(d

)6

28

(g)

120

(a)

137

(a)

11

5(a

)23

03

(m)

471

(e)

02

06

plusmn35

010

40

plusmn1

40

010

40

plusmn1

40

00plusmn

01

8plusmn5

032

0plusmn

48

05

96

plusmn6

93

84

plusmn2

23

60

plusmn1

88

96

plusmn14

219

4plusmn

331

82

plusmn1

61

frei

ber

gi

11

04

(a)

12

53

(z)

12

53

(z)

0(a

)1

2(a

)33

0(a

)4

82

(a)

150

(a)

193

(a)

17

6(b

)3

00

(a)

126

(a)

04

82

plusmn1

408

10

00

plusmn1

42

410

00

plusmn1

42

40plusmn

06

plusmn2

527

2plusmn

368

361

plusmn58

545

plusmn2

31

42

plusmn29

913

3plusmn

24

13

41

plusmn12

47

12

plusmn3

79

colo

rad

oru

m9

221

(e)

10

31

(o)

10

71

(m)

25

(d)

83

(e)

55

6(p

)10

07

(w)

470

(d)

534

(d)

23

3(c

)16

17

(h)

470

(e)

152

plusmn3

58

960

plusmn1

08

010

00

plusmn1

16

00plusmn

62

50

plusmn1

00

48

0plusmn

64

08

89

plusmn11

33

33plusmn

72

25plusmn

72

12

plusmn52

64plusmn

285

8plusmn

12

8

pu

stu

latu

s1

206

(d)

964

(k)

964

(f)

0(a

)8

8(f

)67

6(x

)10

62

(z)

943

(g)

995

(h)

95

0(n

)10

43

(e)

529

(f)

01

86

plusmn23

08

80

plusmn1

08

08

80

plusmn1

08

00plusmn

05

6plusmn1

18

64

0plusmn

80

08

20

plusmn12

54

731

plusmn10

76

903

plusmn1

09

477

7plusmn1

08

27

69

plusmn1

29

01

95

plusmn12

0

sp

B1

395

(l)

740

(a)

888

(a)

11

2(r

)11

5(g

)44

4(h

)8

94

(r)

921

(f)

10

51

(h)

69

4(j

)29

37

(p)

933

(k)

03

22

plusmn60

86

80

plusmn88

08

40

plusmn9

60

43plusmn

15

08

1plusmn1

62

40

0plusmn

48

08

54

plusmn98

16

5plusmn

14

95

0plusmn

15

425

plusmn12

023

8plusmn

44

41

5plusmn

16

9














Dxy 5 L(x+ y )

2 (Lx 1 Ly)
















RES ULTS







TA

BL

E4

A

llo

zy

me

ge

no

typ

es

ob

se

rve

din

10

sp

ecie

so

fP

hy

sala

emu

s

Se

eT

ab

le1

for

exp

lan

ati

on

of

ab

bre

via

tio

ns

of

loci

Ge

no

typ

es

are

foll

ow

ed

by

the

nu

mb

er

of

ind

ivid

ua

lso

bse

rve

dw

ith

the

resp

ecti

ve

ge

no

typ

e

Ifn

og

en

oty

pe

issh

ow

n

the

nn

on

eo

fth

esp

ecim

en

se

xh

ibit

ed

inte

rpre

tab

lea

cti

vit

y(t

he

se

loci

we

resco

red

as

mis

sin

gin

the

ph

ylo

ge

ne

tic

an

aly

sis

)

Sp

ecie

s

Lo

cu

sco

lora

doru

men

esef

ae

eph

ipp

ifer

frei

berg

ip

eter

sip

ust

ula

tus

pu

stu

losu

ssp

A

sp

B

sp

C

Aco

-1cc

10

ee

9d

d8

dd

8d

d1

0cc

8a

b2

dd

8cc

10

cc

4

cd

1b

b8

de

2

Aco

-2d

d8

bb

1a

a2

cc

7cc

10

bd

3b

b9

bb

10

dd

6b

d4

bf

2a

b2

dd

6e

e1

ƒ3

bb

4

Ak

a6

cc

10

cc

7cc

8cc

9a

a1

cc

10

cc

7cc

8a

a4

aa

2a

c3

cc

1cc

1

Ast-

Mb

b1

0cc

10

cc

8b

b9

cc

10

bb

9d

d1

0e

e1

0a

a9

bb

4

bb

1

Aa

t-S

bb

10

dd

10

bb

8cc

9cc

10

bb

9b

b8

bc

3a

a1

0b

b4

bc

2cc

3

Ck

cc

10

bb

10

aa

8a

a9

aa

10

bb

5b

b1

0a

a1

0b

b1

0b

b4

Ca

pa

a1

0b

b6

bb

8a

a9

bb

10

bb

9b

b1

0b

b1

0cc

8a

a4

Est-

Db

b1

0b

b1

0b

b8

bb

9b

b1

0a

b1

bb

10

bb

10

bb

10

bb

4

bb

8

Fb

pe

e8

dd

10

bb

8d

d9

dd

10

ƒ9

cc

10

bb

9b

b1

0b

b4

G6

pd

ha

a9

bb

10

bb

8a

a9

aa

5a

a9

aa

10

bb

10

aa

10

aa

4

Gp

ia

a8

cc

10

dd

6ii

9e

e1

0a

c4

gg

10

hh

10

cc

10

jj4

ac

2d

f2

cc

5

Gtd

ha

a1

0b

b1

0b

b8

aa

9a

a1

0a

a9

aa

10

bb

10

aa

10

aa

4

Gr

aa

8d

d1

0d

d7

dd

9d

d1

0cc

8b

b3

dd

10

dd

10

dd

3

bf

5

ƒ2

G3

pd

hb

b1

0b

b9

bb

8b

b9

bb

10

aa

9b

b1

0b

b1

0b

b1

0b

b4

Idh

-1d

d1

0d

d1

0e

e1

0d

d9

aa

4ƒ

9b

b1

0d

d1

0d

d1

0d

d4

ad

6

Idh

-2cc

10

aa

10

ab

1d

d9

dd

10

dd

8b

d4

bb

10

ee

10

ce

1

bb

7d

d6

ee

3

Ld

h-A

bb

10

bb

10

aa

8cc

9cc

9b

b9

cc

10

bb

10

bb

10

bb

4

cd

1

Ld

h-B

aa

7g

g9

ETHd

d9

dd

10

ƒ9

bb

10

ETHcc

10

hh

4

ae

1

ee

2

Md

h-1

bb

10

bb

10

cc

8d

d9

dd

10

bb

9a

a9

ee

10

ab

1ƒ

4

bb

9


TA

BL

E4

C

on

tin

ue

d

Sp

ecie

s

Lo

cu

sco

lora

doru

men

esef

ae

eph

ipp

ifer

freib

ergi

pet

ersi

pu

stu

latu

sp

ust

ulo

sus

sp

A

sp

B

sp

C

Md

h-2

aa

10

aa

10

ETHd

d9

dd

10

aa

3a

a1

0ETH

aa

10

ETH

ad

6

Md

hp

-1a

a1

0a

a1

0a

g1

bb

4b

b8

ee

9b

b1

0b

b1

0cc

9ƒ

4

bg

2b

c4

bc

2

gg

5cc

1

Md

ph

-2a

a4

ee

10

ee

8d

d9

dd

10

dd

8a

a6

ab

1a

a5

bb

4

ab

3a

b1

bb

8a

b3

bb

3b

b1

bb

1

ad

1

dd

1

Pe

p-A

dd

10

cc

9e

e7

bb

8cc

10

dd

9cc

10

aa

10

cc

10

cc

4

Pg

ma

a7

gg

10

cc

8h

h9

ee

10

bb

1ƒ

9g

g1

0cc

1jj

4

ab

3b

c4

reg1

cf

2

cc

4ƒ

7

Pg

dh

cc

10

aa

2a

b2

dd

9d

e1

aa

8e

g3

aa

1e

e1

0h

h4

ab

4b

b4

ee

7a

c1

gg

6a

b4

bb

4b

f2

bb

4

So

d-S

aa

10

dd

9d

d8

ƒ1

ee

2b

b5

cc

10

dd

10

bb

10

bb

4

fg4

eg

3b

i2

gg

4g

g1

ii1

gh

2

hh

1

Tp

ia

a1

0d

d1

0cc

10

gg

9g

g9

bb

9e

e3

cc

10

dd

9ii

4

gh

1e

g6

df

1

gg

1







Total tive ed- tree


C O M BIN ED 1808 442 068 060 12739 4 16098 0126 00002 00002 1

12S 1214 255 073 066 709 107 0151 00002 00002 1

C O I 543 138 060 045 425 26 0061 00002 00002 2

ALLO ZYM ES 27 25 080 053 102 95 0093 00002 00002 1

C ALLS 12 12 071 061 168 232 0138 00002 00004 2

M O RPHO L- 12 12 100 100 113 13 1000 00002 00002 1

O G Y















m atrices







C O I 00001 (3)

Allozym es 00047 (6) 00180 (7)

C alls 0127 (9) 00548 (8) 0259 (11)

M orphology 00002 (4) 00010 (5) 00047 (6) 0127 (9)









CO I 0724


Calls 0293 0440 0502

M orph ology 0719 0202 0452 0002









Alternative tree


Com bine d ETH 74 13365 118 1913 245 30475 87 12915

025 00001 00001 00042

12S ETH 38 351 63 480 149 6825 47 312

025 00001 00001 00038

CO Ia 24 125 ETH 63 7135 52 220 56 644

025 001 00001 010

Allozym es 7 95 10 17 ETH 15 0 5 25

025 010 00001b 010

Callsa 11 22 12 8 11 25 ETH 10 185

010 005 025 010


05 010 010 000098 b









Alternative tree


C om bined ETH 0210 00002 00002 00002

12S ETH 0246 00002 00002 00008

C O I 0201 ETH 00030 00002 0010

Allozym es 0206 0070 ETH 00002 00366

C alls 0014 00002 0093 ETH 0021

M orphology 0599 0045 0101 00002 ETH





D ISCUSSION

Incongruence






Alternative tre e


Com bined ETH 00004 0555 1000 0195

12S ETH 00002 0333 1000 0187

CO I 00004 ETH 0153 0914 0046

Allozym es 00004 00048 ETH 0945 00034

Calls 0047 0441 0016 ETH 0079

M orphology 00012 0072 0042 1000 ETH



















analysis













ACKNOW LEDGM ENTS












papers

REFERENCES





















26 657 plusmn 681








52 229 plusmn 232







holm



40 366 plusmn 375
















161













Press Ne w York












42 182 plusmn 192


























8 275 plusmn 287




























46 590 plusmn 621













M assachusetts


































Ne w York








1872




343 66 plusmn 67






















50 237 plusmn 255



























New York New York












New York





APPENDIX 1














Viejo

APPENDIX 2














W arty tube rculate



of upper jaw





1 Ab sen t





h yoid plate



APPENDIX 3









































































































TA

BL

E3

S

um

ma

ryo

fca

llsta

tisti

cs

for

the

Ph

ysa

laem

us

pu

stu

losu

sg

rou

pa

nd

clo

se

rela

tiv

es

Th

ev

ari

ab

les

are

tota

ld

ura

tio

no

fca

ll(T

LD

UR

m

se

c)

fre

qu

en

cy

at

on

set

of

ca

ll(I

NH

Z

Hz)

ma

xim

um

fre

qu

en

cy

(MX

HZ

H

z)

tim

eto

the

ma

xim

um

fre

qu

en

cy

(TM

MX

m

se

c)

tim

eto

mid

-fre

qu

en

cy

(TM

HF

HZ

mse

c)

fre

qu

en

cy

at

oƒs

et

of

ca

ll(F

NH

Z

Hz)

do

min

an

tfr

eq

ue

ncy

(DO

MH

Z

Hz)

du

rati

on

of

am

pli

tud

e-m

od

ula

ted

co

mp

on

en

t(A

MD

UR

m

se

c)

rise

tim

e

(RS

TM

in

mse

c)

tim

eto

mid

-ris

e(T

MH

FR

S

mse

c)

fall

tim

e(F

LT

M

mse

c)

an

dti

me

tom

id-f

all

(TM

HF

FL

m

se

c)

Th

em

ea

nis

giv

en

wit

hth

era

ng

eb

elo

w

Th

ela

tte

rin

pa

ren

the

se

sfo

llo

win

gth

em

ea

nis

the

ch

ara

cte

r-sta

teco

de

se

eM

ate

ria

lsa

nd

Me

tho

ds

Th

ev

ari

ab

les

are

dis

cu

sse

din

mo

red

eta

ilin

Co

cro

ft

an

dR

ya

n(1

99

5)

Sp

ecie

sn

TL

DU

RIN

HZ

MX

HZ

TM

MX

TM

HF

HZ

FN

HZ

DO

MH

ZA

MD

UR

RS

TM

TM

HF

RS

FL

TM

TM

HF

FL

sp

A1

339

(j)

812

(d)

876

(a)

65

(k)

16

0(j

)46

0(j

)9

83

(v)

0(

)9

46

(h)

56

6(h

)25

16

(n)

711

(h)

02

34

plusmn4

47

800

plusmn84

08

00

plusmn9

20

43plusmn

10

015

0plusmn1

87

40

0plusmn

520

767

plusmn13

62

0plusmn

06

4plusmn

11

637

plusmn8

317

4plusmn

339

31plusmn

11

6

eph

ipp

ifer

12

66

(g)

900

(h)

944

(e)

62

(j)

14

0(i

)57

6(q

)9

44

(t)

0(

)8

35

(g)

39

4(e

)17

74

(i)

606

(g)

238

plusmn30

88

40

plusmn1

00

08

40

plusmn1

04

04

3plusmn

81

11

2plusmn1

62

52

0plusmn

60

08

45

plusmn10

25

0plusmn

04

8plusmn

97

4plusmn

56

12

9plusmn

236

27plusmn

10

7

enes

efae

17

47

(z)

944

(j)

976

(g)

16

2(z

)38

6(z

)69

2(z

)9

62

(u)

38

48

(z)

30

15

(z)

166

6(z

)44

57

(z)

20

37

(z)

06

31

plusmn9

03

880

plusmn1

04

09

20

plusmn1

04

08

1plusmn

28

730

0plusmn

45

664

0plusmn

720

844

plusmn13

84

23

8plusmn

76

023

0plusmn

40

71

25

plusmn24

237

2plusmn

546

99plusmn

33

8

pu

stu

losu

s1

370

(k)

884

(h)

884

(a)

0(a

)12

4(h

)48

4(k

)7

12

(j)

223

(b)

240

(a)

79

(a)

34

28

(s)

17

50

(v)

02

52

plusmn49

68

40

plusmn9

60

840

plusmn9

60

0plusmn

08

7plusmn1

75

44

0plusmn

60

06

05

plusmn8

83

119

plusmn3

27

93

plusmn60

62

2plusmn1

77

23

6plusmn

45

01

06plusmn

25

2

pet

ersi

12

46

(f)

12

20

(x)

12

20

(w)

0(a

)2

8(b

)38

4(d

)6

28

(g)

120

(a)

137

(a)

11

5(a

)23

03

(m)

471

(e)

02

06

plusmn35

010

40

plusmn1

40

010

40

plusmn1

40

00plusmn

01

8plusmn5

032

0plusmn

48

05

96

plusmn6

93

84

plusmn2

23

60

plusmn1

88

96

plusmn14

219

4plusmn

331

82

plusmn1

61

frei

ber

gi

11

04

(a)

12

53

(z)

12

53

(z)

0(a

)1

2(a

)33

0(a

)4

82

(a)

150

(a)

193

(a)

17

6(b

)3

00

(a)

126

(a)

04

82

plusmn1

408

10

00

plusmn1

42

410

00

plusmn1

42

40plusmn

06

plusmn2

527

2plusmn

368

361

plusmn58

545

plusmn2

31

42

plusmn29

913

3plusmn

24

13

41

plusmn12

47

12

plusmn3

79

colo

rad

oru

m9

221

(e)

10

31

(o)

10

71

(m)

25

(d)

83

(e)

55

6(p

)10

07

(w)

470

(d)

534

(d)

23

3(c

)16

17

(h)

470

(e)

152

plusmn3

58

960

plusmn1

08

010

00

plusmn1

16

00plusmn

62

50

plusmn1

00

48

0plusmn

64

08

89

plusmn11

33

33plusmn

72

25plusmn

72

12

plusmn52

64plusmn

285

8plusmn

12

8

pu

stu

latu

s1

206

(d)

964

(k)

964

(f)

0(a

)8

8(f

)67

6(x

)10

62

(z)

943

(g)

995

(h)

95

0(n

)10

43

(e)

529

(f)

01

86

plusmn23

08

80

plusmn1

08

08

80

plusmn1

08

00plusmn

05

6plusmn1

18

64

0plusmn

80

08

20

plusmn12

54

731

plusmn10

76

903

plusmn1

09

477

7plusmn1

08

27

69

plusmn1

29

01

95

plusmn12

0

sp

B1

395

(l)

740

(a)

888

(a)

11

2(r

)11

5(g

)44

4(h

)8

94

(r)

921

(f)

10

51

(h)

69

4(j

)29

37

(p)

933

(k)

03

22

plusmn60

86

80

plusmn88

08

40

plusmn9

60

43plusmn

15

08

1plusmn1

62

40

0plusmn

48

08

54

plusmn98

16

5plusmn

14

95

0plusmn

15

425

plusmn12

023

8plusmn

44

41

5plusmn

16

9














Dxy 5 L(x+ y )

2 (Lx 1 Ly)
















RES ULTS







TA

BL

E4

A

llo

zy

me

ge

no

typ

es

ob

se

rve

din

10

sp

ecie

so

fP

hy

sala

emu

s

Se

eT

ab

le1

for

exp

lan

ati

on

of

ab

bre

via

tio

ns

of

loci

Ge

no

typ

es

are

foll

ow

ed

by

the

nu

mb

er

of

ind

ivid

ua

lso

bse

rve

dw

ith

the

resp

ecti

ve

ge

no

typ

e

Ifn

og

en

oty

pe

issh

ow

n

the

nn

on

eo

fth

esp

ecim

en

se

xh

ibit

ed

inte

rpre

tab

lea

cti

vit

y(t

he

se

loci

we

resco

red

as

mis

sin

gin

the

ph

ylo

ge

ne

tic

an

aly

sis

)

Sp

ecie

s

Lo

cu

sco

lora

doru

men

esef

ae

eph

ipp

ifer

frei

berg

ip

eter

sip

ust

ula

tus

pu

stu

losu

ssp

A

sp

B

sp

C

Aco

-1cc

10

ee

9d

d8

dd

8d

d1

0cc

8a

b2

dd

8cc

10

cc

4

cd

1b

b8

de

2

Aco

-2d

d8

bb

1a

a2

cc

7cc

10

bd

3b

b9

bb

10

dd

6b

d4

bf

2a

b2

dd

6e

e1

ƒ3

bb

4

Ak

a6

cc

10

cc

7cc

8cc

9a

a1

cc

10

cc

7cc

8a

a4

aa

2a

c3

cc

1cc

1

Ast-

Mb

b1

0cc

10

cc

8b

b9

cc

10

bb

9d

d1

0e

e1

0a

a9

bb

4

bb

1

Aa

t-S

bb

10

dd

10

bb

8cc

9cc

10

bb

9b

b8

bc

3a

a1

0b

b4

bc

2cc

3

Ck

cc

10

bb

10

aa

8a

a9

aa

10

bb

5b

b1

0a

a1

0b

b1

0b

b4

Ca

pa

a1

0b

b6

bb

8a

a9

bb

10

bb

9b

b1

0b

b1

0cc

8a

a4

Est-

Db

b1

0b

b1

0b

b8

bb

9b

b1

0a

b1

bb

10

bb

10

bb

10

bb

4

bb

8

Fb

pe

e8

dd

10

bb

8d

d9

dd

10

ƒ9

cc

10

bb

9b

b1

0b

b4

G6

pd

ha

a9

bb

10

bb

8a

a9

aa

5a

a9

aa

10

bb

10

aa

10

aa

4

Gp

ia

a8

cc

10

dd

6ii

9e

e1

0a

c4

gg

10

hh

10

cc

10

jj4

ac

2d

f2

cc

5

Gtd

ha

a1

0b

b1

0b

b8

aa

9a

a1

0a

a9

aa

10

bb

10

aa

10

aa

4

Gr

aa

8d

d1

0d

d7

dd

9d

d1

0cc

8b

b3

dd

10

dd

10

dd

3

bf

5

ƒ2

G3

pd

hb

b1

0b

b9

bb

8b

b9

bb

10

aa

9b

b1

0b

b1

0b

b1

0b

b4

Idh

-1d

d1

0d

d1

0e

e1

0d

d9

aa

4ƒ

9b

b1

0d

d1

0d

d1

0d

d4

ad

6

Idh

-2cc

10

aa

10

ab

1d

d9

dd

10

dd

8b

d4

bb

10

ee

10

ce

1

bb

7d

d6

ee

3

Ld

h-A

bb

10

bb

10

aa

8cc

9cc

9b

b9

cc

10

bb

10

bb

10

bb

4

cd

1

Ld

h-B

aa

7g

g9

ETHd

d9

dd

10

ƒ9

bb

10

ETHcc

10

hh

4

ae

1

ee

2

Md

h-1

bb

10

bb

10

cc

8d

d9

dd

10

bb

9a

a9

ee

10

ab

1ƒ

4

bb

9


TA

BL

E4

C

on

tin

ue

d

Sp

ecie

s

Lo

cu

sco

lora

doru

men

esef

ae

eph

ipp

ifer

freib

ergi

pet

ersi

pu

stu

latu

sp

ust

ulo

sus

sp

A

sp

B

sp

C

Md

h-2

aa

10

aa

10

ETHd

d9

dd

10

aa

3a

a1

0ETH

aa

10

ETH

ad

6

Md

hp

-1a

a1

0a

a1

0a

g1

bb

4b

b8

ee

9b

b1

0b

b1

0cc

9ƒ

4

bg

2b

c4

bc

2

gg

5cc

1

Md

ph

-2a

a4

ee

10

ee

8d

d9

dd

10

dd

8a

a6

ab

1a

a5

bb

4

ab

3a

b1

bb

8a

b3

bb

3b

b1

bb

1

ad

1

dd

1

Pe

p-A

dd

10

cc

9e

e7

bb

8cc

10

dd

9cc

10

aa

10

cc

10

cc

4

Pg

ma

a7

gg

10

cc

8h

h9

ee

10

bb

1ƒ

9g

g1

0cc

1jj

4

ab

3b

c4

reg1

cf

2

cc

4ƒ

7

Pg

dh

cc

10

aa

2a

b2

dd

9d

e1

aa

8e

g3

aa

1e

e1

0h

h4

ab

4b

b4

ee

7a

c1

gg

6a

b4

bb

4b

f2

bb

4

So

d-S

aa

10

dd

9d

d8

ƒ1

ee

2b

b5

cc

10

dd

10

bb

10

bb

4

fg4

eg

3b

i2

gg

4g

g1

ii1

gh

2

hh

1

Tp

ia

a1

0d

d1

0cc

10

gg

9g

g9

bb

9e

e3

cc

10

dd

9ii

4

gh

1e

g6

df

1

gg

1







Total tive ed- tree


C O M BIN ED 1808 442 068 060 12739 4 16098 0126 00002 00002 1

12S 1214 255 073 066 709 107 0151 00002 00002 1

C O I 543 138 060 045 425 26 0061 00002 00002 2

ALLO ZYM ES 27 25 080 053 102 95 0093 00002 00002 1

C ALLS 12 12 071 061 168 232 0138 00002 00004 2

M O RPHO L- 12 12 100 100 113 13 1000 00002 00002 1

O G Y















m atrices







C O I 00001 (3)

Allozym es 00047 (6) 00180 (7)

C alls 0127 (9) 00548 (8) 0259 (11)

M orphology 00002 (4) 00010 (5) 00047 (6) 0127 (9)









CO I 0724


Calls 0293 0440 0502

M orph ology 0719 0202 0452 0002









Alternative tree


Com bine d ETH 74 13365 118 1913 245 30475 87 12915

025 00001 00001 00042

12S ETH 38 351 63 480 149 6825 47 312

025 00001 00001 00038

CO Ia 24 125 ETH 63 7135 52 220 56 644

025 001 00001 010

Allozym es 7 95 10 17 ETH 15 0 5 25

025 010 00001b 010

Callsa 11 22 12 8 11 25 ETH 10 185

010 005 025 010


05 010 010 000098 b









Alternative tree


C om bined ETH 0210 00002 00002 00002

12S ETH 0246 00002 00002 00008

C O I 0201 ETH 00030 00002 0010

Allozym es 0206 0070 ETH 00002 00366

C alls 0014 00002 0093 ETH 0021

M orphology 0599 0045 0101 00002 ETH





D ISCUSSION

Incongruence






Alternative tre e


Com bined ETH 00004 0555 1000 0195

12S ETH 00002 0333 1000 0187

CO I 00004 ETH 0153 0914 0046

Allozym es 00004 00048 ETH 0945 00034

Calls 0047 0441 0016 ETH 0079

M orphology 00012 0072 0042 1000 ETH



















analysis













ACKNOW LEDGM ENTS












papers

REFERENCES





















26 657 plusmn 681








52 229 plusmn 232







holm



40 366 plusmn 375
















161













Press Ne w York












42 182 plusmn 192


























8 275 plusmn 287




























46 590 plusmn 621













M assachusetts


































Ne w York








1872




343 66 plusmn 67






















50 237 plusmn 255



























New York New York












New York





APPENDIX 1














Viejo

APPENDIX 2














W arty tube rculate



of upper jaw





1 Ab sen t





h yoid plate



APPENDIX 3





































Dxy 5 L(x+ y )

2 (Lx 1 Ly)
















RES ULTS







TA

BL

E4

A

llo

zy

me

ge

no

typ

es

ob

se

rve

din

10

sp

ecie

so

fP

hy

sala

emu

s

Se

eT

ab

le1

for

exp

lan

ati

on

of

ab

bre

via

tio

ns

of

loci

Ge

no

typ

es

are

foll

ow

ed

by

the

nu

mb

er

of

ind

ivid

ua

lso

bse

rve

dw

ith

the

resp

ecti

ve

ge

no

typ

e

Ifn

og

en

oty

pe

issh

ow

n

the

nn

on

eo

fth

esp

ecim

en

se

xh

ibit

ed

inte

rpre

tab

lea

cti

vit

y(t

he

se

loci

we

resco

red

as

mis

sin

gin

the

ph

ylo

ge

ne

tic

an

aly

sis

)

Sp

ecie

s

Lo

cu

sco

lora

doru

men

esef

ae

eph

ipp

ifer

frei

berg

ip

eter

sip

ust

ula

tus

pu

stu

losu

ssp

A

sp

B

sp

C

Aco

-1cc

10

ee

9d

d8

dd

8d

d1

0cc

8a

b2

dd

8cc

10

cc

4

cd

1b

b8

de

2

Aco

-2d

d8

bb

1a

a2

cc

7cc

10

bd

3b

b9

bb

10

dd

6b

d4

bf

2a

b2

dd

6e

e1

ƒ3

bb

4

Ak

a6

cc

10

cc

7cc

8cc

9a

a1

cc

10

cc

7cc

8a

a4

aa

2a

c3

cc

1cc

1

Ast-

Mb

b1

0cc

10

cc

8b

b9

cc

10

bb

9d

d1

0e

e1

0a

a9

bb

4

bb

1

Aa

t-S

bb

10

dd

10

bb

8cc

9cc

10

bb

9b

b8

bc

3a

a1

0b

b4

bc

2cc

3

Ck

cc

10

bb

10

aa

8a

a9

aa

10

bb

5b

b1

0a

a1

0b

b1

0b

b4

Ca

pa

a1

0b

b6

bb

8a

a9

bb

10

bb

9b

b1

0b

b1

0cc

8a

a4

Est-

Db

b1

0b

b1

0b

b8

bb

9b

b1

0a

b1

bb

10

bb

10

bb

10

bb

4

bb

8

Fb

pe

e8

dd

10

bb

8d

d9

dd

10

ƒ9

cc

10

bb

9b

b1

0b

b4

G6

pd

ha

a9

bb

10

bb

8a

a9

aa

5a

a9

aa

10

bb

10

aa

10

aa

4

Gp

ia

a8

cc

10

dd

6ii

9e

e1

0a

c4

gg

10

hh

10

cc

10

jj4

ac

2d

f2

cc

5

Gtd

ha

a1

0b

b1

0b

b8

aa

9a

a1

0a

a9

aa

10

bb

10

aa

10

aa

4

Gr

aa

8d

d1

0d

d7

dd

9d

d1

0cc

8b

b3

dd

10

dd

10

dd

3

bf

5

ƒ2

G3

pd

hb

b1

0b

b9

bb

8b

b9

bb

10

aa

9b

b1

0b

b1

0b

b1

0b

b4

Idh

-1d

d1

0d

d1

0e

e1

0d

d9

aa

4ƒ

9b

b1

0d

d1

0d

d1

0d

d4

ad

6

Idh

-2cc

10

aa

10

ab

1d

d9

dd

10

dd

8b

d4

bb

10

ee

10

ce

1

bb

7d

d6

ee

3

Ld

h-A

bb

10

bb

10

aa

8cc

9cc

9b

b9

cc

10

bb

10

bb

10

bb

4

cd

1

Ld

h-B

aa

7g

g9

ETHd

d9

dd

10

ƒ9

bb

10

ETHcc

10

hh

4

ae

1

ee

2

Md

h-1

bb

10

bb

10

cc

8d

d9

dd

10

bb

9a

a9

ee

10

ab

1ƒ

4

bb

9


TA

BL

E4

C

on

tin

ue

d

Sp

ecie

s

Lo

cu

sco

lora

doru

men

esef

ae

eph

ipp

ifer

freib

ergi

pet

ersi

pu

stu

latu

sp

ust

ulo

sus

sp

A

sp

B

sp

C

Md

h-2

aa

10

aa

10

ETHd

d9

dd

10

aa

3a

a1

0ETH

aa

10

ETH

ad

6

Md

hp

-1a

a1

0a

a1

0a

g1

bb

4b

b8

ee

9b

b1

0b

b1

0cc

9ƒ

4

bg

2b

c4

bc

2

gg

5cc

1

Md

ph

-2a

a4

ee

10

ee

8d

d9

dd

10

dd

8a

a6

ab

1a

a5

bb

4

ab

3a

b1

bb

8a

b3

bb

3b

b1

bb

1

ad

1

dd

1

Pe

p-A

dd

10

cc

9e

e7

bb

8cc

10

dd

9cc

10

aa

10

cc

10

cc

4

Pg

ma

a7

gg

10

cc

8h

h9

ee

10

bb

1ƒ

9g

g1

0cc

1jj

4

ab

3b

c4

reg1

cf

2

cc

4ƒ

7

Pg

dh

cc

10

aa

2a

b2

dd

9d

e1

aa

8e

g3

aa

1e

e1

0h

h4

ab

4b

b4

ee

7a

c1

gg

6a

b4

bb

4b

f2

bb

4

So

d-S

aa

10

dd

9d

d8

ƒ1

ee

2b

b5

cc

10

dd

10

bb

10

bb

4

fg4

eg

3b

i2

gg

4g

g1

ii1

gh

2

hh

1

Tp

ia

a1

0d

d1

0cc

10

gg

9g

g9

bb

9e

e3

cc

10

dd

9ii

4

gh

1e

g6

df

1

gg

1







Total tive ed- tree


C O M BIN ED 1808 442 068 060 12739 4 16098 0126 00002 00002 1

12S 1214 255 073 066 709 107 0151 00002 00002 1

C O I 543 138 060 045 425 26 0061 00002 00002 2

ALLO ZYM ES 27 25 080 053 102 95 0093 00002 00002 1

C ALLS 12 12 071 061 168 232 0138 00002 00004 2

M O RPHO L- 12 12 100 100 113 13 1000 00002 00002 1

O G Y















m atrices







C O I 00001 (3)

Allozym es 00047 (6) 00180 (7)

C alls 0127 (9) 00548 (8) 0259 (11)

M orphology 00002 (4) 00010 (5) 00047 (6) 0127 (9)









CO I 0724


Calls 0293 0440 0502

M orph ology 0719 0202 0452 0002









Alternative tree


Com bine d ETH 74 13365 118 1913 245 30475 87 12915

025 00001 00001 00042

12S ETH 38 351 63 480 149 6825 47 312

025 00001 00001 00038

CO Ia 24 125 ETH 63 7135 52 220 56 644

025 001 00001 010

Allozym es 7 95 10 17 ETH 15 0 5 25

025 010 00001b 010

Callsa 11 22 12 8 11 25 ETH 10 185

010 005 025 010


05 010 010 000098 b









Alternative tree


C om bined ETH 0210 00002 00002 00002

12S ETH 0246 00002 00002 00008

C O I 0201 ETH 00030 00002 0010

Allozym es 0206 0070 ETH 00002 00366

C alls 0014 00002 0093 ETH 0021

M orphology 0599 0045 0101 00002 ETH





D ISCUSSION

Incongruence






Alternative tre e


Com bined ETH 00004 0555 1000 0195

12S ETH 00002 0333 1000 0187

CO I 00004 ETH 0153 0914 0046

Allozym es 00004 00048 ETH 0945 00034

Calls 0047 0441 0016 ETH 0079

M orphology 00012 0072 0042 1000 ETH



















analysis













ACKNOW LEDGM ENTS












papers

REFERENCES





















26 657 plusmn 681








52 229 plusmn 232







holm



40 366 plusmn 375
















161













Press Ne w York












42 182 plusmn 192


























8 275 plusmn 287




























46 590 plusmn 621













M assachusetts


































Ne w York








1872




343 66 plusmn 67






















50 237 plusmn 255



























New York New York












New York





APPENDIX 1














Viejo

APPENDIX 2














W arty tube rculate



of upper jaw





1 Ab sen t





h yoid plate



APPENDIX 3































RES ULTS







TA

BL

E4

A

llo

zy

me

ge

no

typ

es

ob

se

rve

din

10

sp

ecie

so

fP

hy

sala

emu

s

Se

eT

ab

le1

for

exp

lan

ati

on

of

ab

bre

via

tio

ns

of

loci

Ge

no

typ

es

are

foll

ow

ed

by

the

nu

mb

er

of

ind

ivid

ua

lso

bse

rve

dw

ith

the

resp

ecti

ve

ge

no

typ

e

Ifn

og

en

oty

pe

issh

ow

n

the

nn

on

eo

fth

esp

ecim

en

se

xh

ibit

ed

inte

rpre

tab

lea

cti

vit

y(t

he

se

loci

we

resco

red

as

mis

sin

gin

the

ph

ylo

ge

ne

tic

an

aly

sis

)

Sp

ecie

s

Lo

cu

sco

lora

doru

men

esef

ae

eph

ipp

ifer

frei

berg

ip

eter

sip

ust

ula

tus

pu

stu

losu

ssp

A

sp

B

sp

C

Aco

-1cc

10

ee

9d

d8

dd

8d

d1

0cc

8a

b2

dd

8cc

10

cc

4

cd

1b

b8

de

2

Aco

-2d

d8

bb

1a

a2

cc

7cc

10

bd

3b

b9

bb

10

dd

6b

d4

bf

2a

b2

dd

6e

e1

ƒ3

bb

4

Ak

a6

cc

10

cc

7cc

8cc

9a

a1

cc

10

cc

7cc

8a

a4

aa

2a

c3

cc

1cc

1

Ast-

Mb

b1

0cc

10

cc

8b

b9

cc

10

bb

9d

d1

0e

e1

0a

a9

bb

4

bb

1

Aa

t-S

bb

10

dd

10

bb

8cc

9cc

10

bb

9b

b8

bc

3a

a1

0b

b4

bc

2cc

3

Ck

cc

10

bb

10

aa

8a

a9

aa

10

bb

5b

b1

0a

a1

0b

b1

0b

b4

Ca

pa

a1

0b

b6

bb

8a

a9

bb

10

bb

9b

b1

0b

b1

0cc

8a

a4

Est-

Db

b1

0b

b1

0b

b8

bb

9b

b1

0a

b1

bb

10

bb

10

bb

10

bb

4

bb

8

Fb

pe

e8

dd

10

bb

8d

d9

dd

10

ƒ9

cc

10

bb

9b

b1

0b

b4

G6

pd

ha

a9

bb

10

bb

8a

a9

aa

5a

a9

aa

10

bb

10

aa

10

aa

4

Gp

ia

a8

cc

10

dd

6ii

9e

e1

0a

c4

gg

10

hh

10

cc

10

jj4

ac

2d

f2

cc

5

Gtd

ha

a1

0b

b1

0b

b8

aa

9a

a1

0a

a9

aa

10

bb

10

aa

10

aa

4

Gr

aa

8d

d1

0d

d7

dd

9d

d1

0cc

8b

b3

dd

10

dd

10

dd

3

bf

5

ƒ2

G3

pd

hb

b1

0b

b9

bb

8b

b9

bb

10

aa

9b

b1

0b

b1

0b

b1

0b

b4

Idh

-1d

d1

0d

d1

0e

e1

0d

d9

aa

4ƒ

9b

b1

0d

d1

0d

d1

0d

d4

ad

6

Idh

-2cc

10

aa

10

ab

1d

d9

dd

10

dd

8b

d4

bb

10

ee

10

ce

1

bb

7d

d6

ee

3

Ld

h-A

bb

10

bb

10

aa

8cc

9cc

9b

b9

cc

10

bb

10

bb

10

bb

4

cd

1

Ld

h-B

aa

7g

g9

ETHd

d9

dd

10

ƒ9

bb

10

ETHcc

10

hh

4

ae

1

ee

2

Md

h-1

bb

10

bb

10

cc

8d

d9

dd

10

bb

9a

a9

ee

10

ab

1ƒ

4

bb

9


TA

BL

E4

C

on

tin

ue

d

Sp

ecie

s

Lo

cu

sco

lora

doru

men

esef

ae

eph

ipp

ifer

freib

ergi

pet

ersi

pu

stu

latu

sp

ust

ulo

sus

sp

A

sp

B

sp

C

Md

h-2

aa

10

aa

10

ETHd

d9

dd

10

aa

3a

a1

0ETH

aa

10

ETH

ad

6

Md

hp

-1a

a1

0a

a1

0a

g1

bb

4b

b8

ee

9b

b1

0b

b1

0cc

9ƒ

4

bg

2b

c4

bc

2

gg

5cc

1

Md

ph

-2a

a4

ee

10

ee

8d

d9

dd

10

dd

8a

a6

ab

1a

a5

bb

4

ab

3a

b1

bb

8a

b3

bb

3b

b1

bb

1

ad

1

dd

1

Pe

p-A

dd

10

cc

9e

e7

bb

8cc

10

dd

9cc

10

aa

10

cc

10

cc

4

Pg

ma

a7

gg

10

cc

8h

h9

ee

10

bb

1ƒ

9g

g1

0cc

1jj

4

ab

3b

c4

reg1

cf

2

cc

4ƒ

7

Pg

dh

cc

10

aa

2a

b2

dd

9d

e1

aa

8e

g3

aa

1e

e1

0h

h4

ab

4b

b4

ee

7a

c1

gg

6a

b4

bb

4b

f2

bb

4

So

d-S

aa

10

dd

9d

d8

ƒ1

ee

2b

b5

cc

10

dd

10

bb

10

bb

4

fg4

eg

3b

i2

gg

4g

g1

ii1

gh

2

hh

1

Tp

ia

a1

0d

d1

0cc

10

gg

9g

g9

bb

9e

e3

cc

10

dd

9ii

4

gh

1e

g6

df

1

gg

1







Total tive ed- tree


C O M BIN ED 1808 442 068 060 12739 4 16098 0126 00002 00002 1

12S 1214 255 073 066 709 107 0151 00002 00002 1

C O I 543 138 060 045 425 26 0061 00002 00002 2

ALLO ZYM ES 27 25 080 053 102 95 0093 00002 00002 1

C ALLS 12 12 071 061 168 232 0138 00002 00004 2

M O RPHO L- 12 12 100 100 113 13 1000 00002 00002 1

O G Y















m atrices







C O I 00001 (3)

Allozym es 00047 (6) 00180 (7)

C alls 0127 (9) 00548 (8) 0259 (11)

M orphology 00002 (4) 00010 (5) 00047 (6) 0127 (9)









CO I 0724


Calls 0293 0440 0502

M orph ology 0719 0202 0452 0002









Alternative tree


Com bine d ETH 74 13365 118 1913 245 30475 87 12915

025 00001 00001 00042

12S ETH 38 351 63 480 149 6825 47 312

025 00001 00001 00038

CO Ia 24 125 ETH 63 7135 52 220 56 644

025 001 00001 010

Allozym es 7 95 10 17 ETH 15 0 5 25

025 010 00001b 010

Callsa 11 22 12 8 11 25 ETH 10 185

010 005 025 010


05 010 010 000098 b









Alternative tree


C om bined ETH 0210 00002 00002 00002

12S ETH 0246 00002 00002 00008

C O I 0201 ETH 00030 00002 0010

Allozym es 0206 0070 ETH 00002 00366

C alls 0014 00002 0093 ETH 0021

M orphology 0599 0045 0101 00002 ETH





D ISCUSSION

Incongruence






Alternative tre e


Com bined ETH 00004 0555 1000 0195

12S ETH 00002 0333 1000 0187

CO I 00004 ETH 0153 0914 0046

Allozym es 00004 00048 ETH 0945 00034

Calls 0047 0441 0016 ETH 0079

M orphology 00012 0072 0042 1000 ETH



















analysis













ACKNOW LEDGM ENTS












papers

REFERENCES





















26 657 plusmn 681








52 229 plusmn 232







holm



40 366 plusmn 375
















161













Press Ne w York












42 182 plusmn 192


























8 275 plusmn 287




























46 590 plusmn 621













M assachusetts


































Ne w York








1872




343 66 plusmn 67






















50 237 plusmn 255



























New York New York












New York





APPENDIX 1














Viejo

APPENDIX 2














W arty tube rculate



of upper jaw





1 Ab sen t





h yoid plate



APPENDIX 3

























TA

BL

E4

A

llo

zy

me

ge

no

typ

es

ob

se

rve

din

10

sp

ecie

so

fP

hy

sala

emu

s

Se

eT

ab

le1

for

exp

lan

ati

on

of

ab

bre

via

tio

ns

of

loci

Ge

no

typ

es

are

foll

ow

ed

by

the

nu

mb

er

of

ind

ivid

ua

lso

bse

rve

dw

ith

the

resp

ecti

ve

ge

no

typ

e

Ifn

og

en

oty

pe

issh

ow

n

the

nn

on

eo

fth

esp

ecim

en

se

xh

ibit

ed

inte

rpre

tab

lea

cti

vit

y(t

he

se

loci

we

resco

red

as

mis

sin

gin

the

ph

ylo

ge

ne

tic

an

aly

sis

)

Sp

ecie

s

Lo

cu

sco

lora

doru

men

esef

ae

eph

ipp

ifer

frei

berg

ip

eter

sip

ust

ula

tus

pu

stu

losu

ssp

A

sp

B

sp

C

Aco

-1cc

10

ee

9d

d8

dd

8d

d1

0cc

8a

b2

dd

8cc

10

cc

4

cd

1b

b8

de

2

Aco

-2d

d8

bb

1a

a2

cc

7cc

10

bd

3b

b9

bb

10

dd

6b

d4

bf

2a

b2

dd

6e

e1

ƒ3

bb

4

Ak

a6

cc

10

cc

7cc

8cc

9a

a1

cc

10

cc

7cc

8a

a4

aa

2a

c3

cc

1cc

1

Ast-

Mb

b1

0cc

10

cc

8b

b9

cc

10

bb

9d

d1

0e

e1

0a

a9

bb

4

bb

1

Aa

t-S

bb

10

dd

10

bb

8cc

9cc

10

bb

9b

b8

bc

3a

a1

0b

b4

bc

2cc

3

Ck

cc

10

bb

10

aa

8a

a9

aa

10

bb

5b

b1

0a

a1

0b

b1

0b

b4

Ca

pa

a1

0b

b6

bb

8a

a9

bb

10

bb

9b

b1

0b

b1

0cc

8a

a4

Est-

Db

b1

0b

b1

0b

b8

bb

9b

b1

0a

b1

bb

10

bb

10

bb

10

bb

4

bb

8

Fb

pe

e8

dd

10

bb

8d

d9

dd

10

ƒ9

cc

10

bb

9b

b1

0b

b4

G6

pd

ha

a9

bb

10

bb

8a

a9

aa

5a

a9

aa

10

bb

10

aa

10

aa

4

Gp

ia

a8

cc

10

dd

6ii

9e

e1

0a

c4

gg

10

hh

10

cc

10

jj4

ac

2d

f2

cc

5

Gtd

ha

a1

0b

b1

0b

b8

aa

9a

a1

0a

a9

aa

10

bb

10

aa

10

aa

4

Gr

aa

8d

d1

0d

d7

dd

9d

d1

0cc

8b

b3

dd

10

dd

10

dd

3

bf

5

ƒ2

G3

pd

hb

b1

0b

b9

bb

8b

b9

bb

10

aa

9b

b1

0b

b1

0b

b1

0b

b4

Idh

-1d

d1

0d

d1

0e

e1

0d

d9

aa

4ƒ

9b

b1

0d

d1

0d

d1

0d

d4

ad

6

Idh

-2cc

10

aa

10

ab

1d

d9

dd

10

dd

8b

d4

bb

10

ee

10

ce

1

bb

7d

d6

ee

3

Ld

h-A

bb

10

bb

10

aa

8cc

9cc

9b

b9

cc

10

bb

10

bb

10

bb

4

cd

1

Ld

h-B

aa

7g

g9

ETHd

d9

dd

10

ƒ9

bb

10

ETHcc

10

hh

4

ae

1

ee

2

Md

h-1

bb

10

bb

10

cc

8d

d9

dd

10

bb

9a

a9

ee

10

ab

1ƒ

4

bb

9


TA

BL

E4

C

on

tin

ue

d

Sp

ecie

s

Lo

cu

sco

lora

doru

men

esef

ae

eph

ipp

ifer

freib

ergi

pet

ersi

pu

stu

latu

sp

ust

ulo

sus

sp

A

sp

B

sp

C

Md

h-2

aa

10

aa

10

ETHd

d9

dd

10

aa

3a

a1

0ETH

aa

10

ETH

ad

6

Md

hp

-1a

a1

0a

a1

0a

g1

bb

4b

b8

ee

9b

b1

0b

b1

0cc

9ƒ

4

bg

2b

c4

bc

2

gg

5cc

1

Md

ph

-2a

a4

ee

10

ee

8d

d9

dd

10

dd

8a

a6

ab

1a

a5

bb

4

ab

3a

b1

bb

8a

b3

bb

3b

b1

bb

1

ad

1

dd

1

Pe

p-A

dd

10

cc

9e

e7

bb

8cc

10

dd

9cc

10

aa

10

cc

10

cc

4

Pg

ma

a7

gg

10

cc

8h

h9

ee

10

bb

1ƒ

9g

g1

0cc

1jj

4

ab

3b

c4

reg1

cf

2

cc

4ƒ

7

Pg

dh

cc

10

aa

2a

b2

dd

9d

e1

aa

8e

g3

aa

1e

e1

0h

h4

ab

4b

b4

ee

7a

c1

gg

6a

b4

bb

4b

f2

bb

4

So

d-S

aa

10

dd

9d

d8

ƒ1

ee

2b

b5

cc

10

dd

10

bb

10

bb

4

fg4

eg

3b

i2

gg

4g

g1

ii1

gh

2

hh

1

Tp

ia

a1

0d

d1

0cc

10

gg

9g

g9

bb

9e

e3

cc

10

dd

9ii

4

gh

1e

g6

df

1

gg

1







Total tive ed- tree


C O M BIN ED 1808 442 068 060 12739 4 16098 0126 00002 00002 1

12S 1214 255 073 066 709 107 0151 00002 00002 1

C O I 543 138 060 045 425 26 0061 00002 00002 2

ALLO ZYM ES 27 25 080 053 102 95 0093 00002 00002 1

C ALLS 12 12 071 061 168 232 0138 00002 00004 2

M O RPHO L- 12 12 100 100 113 13 1000 00002 00002 1

O G Y















m atrices







C O I 00001 (3)

Allozym es 00047 (6) 00180 (7)

C alls 0127 (9) 00548 (8) 0259 (11)

M orphology 00002 (4) 00010 (5) 00047 (6) 0127 (9)









CO I 0724


Calls 0293 0440 0502

M orph ology 0719 0202 0452 0002









Alternative tree


Com bine d ETH 74 13365 118 1913 245 30475 87 12915

025 00001 00001 00042

12S ETH 38 351 63 480 149 6825 47 312

025 00001 00001 00038

CO Ia 24 125 ETH 63 7135 52 220 56 644

025 001 00001 010

Allozym es 7 95 10 17 ETH 15 0 5 25

025 010 00001b 010

Callsa 11 22 12 8 11 25 ETH 10 185

010 005 025 010


05 010 010 000098 b









Alternative tree


C om bined ETH 0210 00002 00002 00002

12S ETH 0246 00002 00002 00008

C O I 0201 ETH 00030 00002 0010

Allozym es 0206 0070 ETH 00002 00366

C alls 0014 00002 0093 ETH 0021

M orphology 0599 0045 0101 00002 ETH





D ISCUSSION

Incongruence






Alternative tre e


Com bined ETH 00004 0555 1000 0195

12S ETH 00002 0333 1000 0187

CO I 00004 ETH 0153 0914 0046

Allozym es 00004 00048 ETH 0945 00034

Calls 0047 0441 0016 ETH 0079

M orphology 00012 0072 0042 1000 ETH



















analysis













ACKNOW LEDGM ENTS












papers

REFERENCES





















26 657 plusmn 681








52 229 plusmn 232







holm



40 366 plusmn 375
















161













Press Ne w York












42 182 plusmn 192


























8 275 plusmn 287




























46 590 plusmn 621













M assachusetts


































Ne w York








1872




343 66 plusmn 67






















50 237 plusmn 255



























New York New York












New York





APPENDIX 1














Viejo

APPENDIX 2














W arty tube rculate



of upper jaw





1 Ab sen t





h yoid plate



APPENDIX 3

























TA

BL

E4

C

on

tin

ue

d

Sp

ecie

s

Lo

cu

sco

lora

doru

men

esef

ae

eph

ipp

ifer

freib

ergi

pet

ersi

pu

stu

latu

sp

ust

ulo

sus

sp

A

sp

B

sp

C

Md

h-2

aa

10

aa

10

ETHd

d9

dd

10

aa

3a

a1

0ETH

aa

10

ETH

ad

6

Md

hp

-1a

a1

0a

a1

0a

g1

bb

4b

b8

ee

9b

b1

0b

b1

0cc

9ƒ

4

bg

2b

c4

bc

2

gg

5cc

1

Md

ph

-2a

a4

ee

10

ee

8d

d9

dd

10

dd

8a

a6

ab

1a

a5

bb

4

ab

3a

b1

bb

8a

b3

bb

3b

b1

bb

1

ad

1

dd

1

Pe

p-A

dd

10

cc

9e

e7

bb

8cc

10

dd

9cc

10

aa

10

cc

10

cc

4

Pg

ma

a7

gg

10

cc

8h

h9

ee

10

bb

1ƒ

9g

g1

0cc

1jj

4

ab

3b

c4

reg1

cf

2

cc

4ƒ

7

Pg

dh

cc

10

aa

2a

b2

dd

9d

e1

aa

8e

g3

aa

1e

e1

0h

h4

ab

4b

b4

ee

7a

c1

gg

6a

b4

bb

4b

f2

bb

4

So

d-S

aa

10

dd

9d

d8

ƒ1

ee

2b

b5

cc

10

dd

10

bb

10

bb

4

fg4

eg

3b

i2

gg

4g

g1

ii1

gh

2

hh

1

Tp

ia

a1

0d

d1

0cc

10

gg

9g

g9

bb

9e

e3

cc

10

dd

9ii

4

gh

1e

g6

df

1

gg

1







Total tive ed- tree


C O M BIN ED 1808 442 068 060 12739 4 16098 0126 00002 00002 1

12S 1214 255 073 066 709 107 0151 00002 00002 1

C O I 543 138 060 045 425 26 0061 00002 00002 2

ALLO ZYM ES 27 25 080 053 102 95 0093 00002 00002 1

C ALLS 12 12 071 061 168 232 0138 00002 00004 2

M O RPHO L- 12 12 100 100 113 13 1000 00002 00002 1

O G Y















m atrices







C O I 00001 (3)

Allozym es 00047 (6) 00180 (7)

C alls 0127 (9) 00548 (8) 0259 (11)

M orphology 00002 (4) 00010 (5) 00047 (6) 0127 (9)









CO I 0724


Calls 0293 0440 0502

M orph ology 0719 0202 0452 0002









Alternative tree


Com bine d ETH 74 13365 118 1913 245 30475 87 12915

025 00001 00001 00042

12S ETH 38 351 63 480 149 6825 47 312

025 00001 00001 00038

CO Ia 24 125 ETH 63 7135 52 220 56 644

025 001 00001 010

Allozym es 7 95 10 17 ETH 15 0 5 25

025 010 00001b 010

Callsa 11 22 12 8 11 25 ETH 10 185

010 005 025 010


05 010 010 000098 b









Alternative tree


C om bined ETH 0210 00002 00002 00002

12S ETH 0246 00002 00002 00008

C O I 0201 ETH 00030 00002 0010

Allozym es 0206 0070 ETH 00002 00366

C alls 0014 00002 0093 ETH 0021

M orphology 0599 0045 0101 00002 ETH





D ISCUSSION

Incongruence






Alternative tre e


Com bined ETH 00004 0555 1000 0195

12S ETH 00002 0333 1000 0187

CO I 00004 ETH 0153 0914 0046

Allozym es 00004 00048 ETH 0945 00034

Calls 0047 0441 0016 ETH 0079

M orphology 00012 0072 0042 1000 ETH



















analysis













ACKNOW LEDGM ENTS












papers

REFERENCES





















26 657 plusmn 681








52 229 plusmn 232







holm



40 366 plusmn 375
















161













Press Ne w York












42 182 plusmn 192


























8 275 plusmn 287




























46 590 plusmn 621













M assachusetts


































Ne w York








1872




343 66 plusmn 67






















50 237 plusmn 255



























New York New York












New York





APPENDIX 1














Viejo

APPENDIX 2














W arty tube rculate



of upper jaw





1 Ab sen t





h yoid plate



APPENDIX 3






























Total tive ed- tree


C O M BIN ED 1808 442 068 060 12739 4 16098 0126 00002 00002 1

12S 1214 255 073 066 709 107 0151 00002 00002 1

C O I 543 138 060 045 425 26 0061 00002 00002 2

ALLO ZYM ES 27 25 080 053 102 95 0093 00002 00002 1

C ALLS 12 12 071 061 168 232 0138 00002 00004 2

M O RPHO L- 12 12 100 100 113 13 1000 00002 00002 1

O G Y















m atrices







C O I 00001 (3)

Allozym es 00047 (6) 00180 (7)

C alls 0127 (9) 00548 (8) 0259 (11)

M orphology 00002 (4) 00010 (5) 00047 (6) 0127 (9)









CO I 0724


Calls 0293 0440 0502

M orph ology 0719 0202 0452 0002









Alternative tree


Com bine d ETH 74 13365 118 1913 245 30475 87 12915

025 00001 00001 00042

12S ETH 38 351 63 480 149 6825 47 312

025 00001 00001 00038

CO Ia 24 125 ETH 63 7135 52 220 56 644

025 001 00001 010

Allozym es 7 95 10 17 ETH 15 0 5 25

025 010 00001b 010

Callsa 11 22 12 8 11 25 ETH 10 185

010 005 025 010


05 010 010 000098 b









Alternative tree


C om bined ETH 0210 00002 00002 00002

12S ETH 0246 00002 00002 00008

C O I 0201 ETH 00030 00002 0010

Allozym es 0206 0070 ETH 00002 00366

C alls 0014 00002 0093 ETH 0021

M orphology 0599 0045 0101 00002 ETH





D ISCUSSION

Incongruence






Alternative tre e


Com bined ETH 00004 0555 1000 0195

12S ETH 00002 0333 1000 0187

CO I 00004 ETH 0153 0914 0046

Allozym es 00004 00048 ETH 0945 00034

Calls 0047 0441 0016 ETH 0079

M orphology 00012 0072 0042 1000 ETH



















analysis













ACKNOW LEDGM ENTS












papers

REFERENCES





















26 657 plusmn 681








52 229 plusmn 232







holm



40 366 plusmn 375
















161













Press Ne w York












42 182 plusmn 192


























8 275 plusmn 287




























46 590 plusmn 621













M assachusetts


































Ne w York








1872




343 66 plusmn 67






















50 237 plusmn 255



























New York New York












New York





APPENDIX 1














Viejo

APPENDIX 2














W arty tube rculate



of upper jaw





1 Ab sen t





h yoid plate



APPENDIX 3






























m atrices







C O I 00001 (3)

Allozym es 00047 (6) 00180 (7)

C alls 0127 (9) 00548 (8) 0259 (11)

M orphology 00002 (4) 00010 (5) 00047 (6) 0127 (9)









CO I 0724


Calls 0293 0440 0502

M orph ology 0719 0202 0452 0002









Alternative tree


Com bine d ETH 74 13365 118 1913 245 30475 87 12915

025 00001 00001 00042

12S ETH 38 351 63 480 149 6825 47 312

025 00001 00001 00038

CO Ia 24 125 ETH 63 7135 52 220 56 644

025 001 00001 010

Allozym es 7 95 10 17 ETH 15 0 5 25

025 010 00001b 010

Callsa 11 22 12 8 11 25 ETH 10 185

010 005 025 010


05 010 010 000098 b









Alternative tree


C om bined ETH 0210 00002 00002 00002

12S ETH 0246 00002 00002 00008

C O I 0201 ETH 00030 00002 0010

Allozym es 0206 0070 ETH 00002 00366

C alls 0014 00002 0093 ETH 0021

M orphology 0599 0045 0101 00002 ETH





D ISCUSSION

Incongruence






Alternative tre e


Com bined ETH 00004 0555 1000 0195

12S ETH 00002 0333 1000 0187

CO I 00004 ETH 0153 0914 0046

Allozym es 00004 00048 ETH 0945 00034

Calls 0047 0441 0016 ETH 0079

M orphology 00012 0072 0042 1000 ETH



















analysis













ACKNOW LEDGM ENTS












papers

REFERENCES





















26 657 plusmn 681








52 229 plusmn 232







holm



40 366 plusmn 375
















161













Press Ne w York












42 182 plusmn 192


























8 275 plusmn 287




























46 590 plusmn 621













M assachusetts


































Ne w York








1872




343 66 plusmn 67






















50 237 plusmn 255



























New York New York












New York





APPENDIX 1














Viejo

APPENDIX 2














W arty tube rculate



of upper jaw





1 Ab sen t





h yoid plate



APPENDIX 3






























C O I 00001 (3)

Allozym es 00047 (6) 00180 (7)

C alls 0127 (9) 00548 (8) 0259 (11)

M orphology 00002 (4) 00010 (5) 00047 (6) 0127 (9)









CO I 0724


Calls 0293 0440 0502

M orph ology 0719 0202 0452 0002









Alternative tree


Com bine d ETH 74 13365 118 1913 245 30475 87 12915

025 00001 00001 00042

12S ETH 38 351 63 480 149 6825 47 312

025 00001 00001 00038

CO Ia 24 125 ETH 63 7135 52 220 56 644

025 001 00001 010

Allozym es 7 95 10 17 ETH 15 0 5 25

025 010 00001b 010

Callsa 11 22 12 8 11 25 ETH 10 185

010 005 025 010


05 010 010 000098 b









Alternative tree


C om bined ETH 0210 00002 00002 00002

12S ETH 0246 00002 00002 00008

C O I 0201 ETH 00030 00002 0010

Allozym es 0206 0070 ETH 00002 00366

C alls 0014 00002 0093 ETH 0021

M orphology 0599 0045 0101 00002 ETH





D ISCUSSION

Incongruence






Alternative tre e


Com bined ETH 00004 0555 1000 0195

12S ETH 00002 0333 1000 0187

CO I 00004 ETH 0153 0914 0046

Allozym es 00004 00048 ETH 0945 00034

Calls 0047 0441 0016 ETH 0079

M orphology 00012 0072 0042 1000 ETH



















analysis













ACKNOW LEDGM ENTS












papers

REFERENCES





















26 657 plusmn 681








52 229 plusmn 232







holm



40 366 plusmn 375
















161













Press Ne w York












42 182 plusmn 192


























8 275 plusmn 287




























46 590 plusmn 621













M assachusetts


































Ne w York








1872




343 66 plusmn 67






















50 237 plusmn 255



























New York New York












New York





APPENDIX 1














Viejo

APPENDIX 2














W arty tube rculate



of upper jaw





1 Ab sen t





h yoid plate



APPENDIX 3





























Alternative tree


C om bined ETH 0210 00002 00002 00002

12S ETH 0246 00002 00002 00008

C O I 0201 ETH 00030 00002 0010

Allozym es 0206 0070 ETH 00002 00366

C alls 0014 00002 0093 ETH 0021

M orphology 0599 0045 0101 00002 ETH





D ISCUSSION

Incongruence






Alternative tre e


Com bined ETH 00004 0555 1000 0195

12S ETH 00002 0333 1000 0187

CO I 00004 ETH 0153 0914 0046

Allozym es 00004 00048 ETH 0945 00034

Calls 0047 0441 0016 ETH 0079

M orphology 00012 0072 0042 1000 ETH



















analysis













ACKNOW LEDGM ENTS












papers

REFERENCES





















26 657 plusmn 681








52 229 plusmn 232







holm



40 366 plusmn 375
















161













Press Ne w York












42 182 plusmn 192


























8 275 plusmn 287




























46 590 plusmn 621













M assachusetts


































Ne w York








1872




343 66 plusmn 67






















50 237 plusmn 255



























New York New York












New York





APPENDIX 1














Viejo

APPENDIX 2














W arty tube rculate



of upper jaw





1 Ab sen t





h yoid plate



APPENDIX 3










































analysis













ACKNOW LEDGM ENTS












papers

REFERENCES





















26 657 plusmn 681








52 229 plusmn 232







holm



40 366 plusmn 375
















161













Press Ne w York












42 182 plusmn 192


























8 275 plusmn 287




























46 590 plusmn 621













M assachusetts


































Ne w York








1872




343 66 plusmn 67






















50 237 plusmn 255



























New York New York












New York





APPENDIX 1














Viejo

APPENDIX 2














W arty tube rculate



of upper jaw





1 Ab sen t





h yoid plate



APPENDIX 3


































ACKNOW LEDGM ENTS












papers

REFERENCES





















26 657 plusmn 681








52 229 plusmn 232







holm



40 366 plusmn 375
















161













Press Ne w York












42 182 plusmn 192


























8 275 plusmn 287




























46 590 plusmn 621













M assachusetts


































Ne w York








1872




343 66 plusmn 67






















50 237 plusmn 255



























New York New York












New York





APPENDIX 1














Viejo

APPENDIX 2














W arty tube rculate



of upper jaw





1 Ab sen t





h yoid plate



APPENDIX 3















































8 275 plusmn 287




























46 590 plusmn 621













M assachusetts


































Ne w York








1872




343 66 plusmn 67






















50 237 plusmn 255



























New York New York












New York





APPENDIX 1














Viejo

APPENDIX 2














W arty tube rculate



of upper jaw





1 Ab sen t





h yoid plate



APPENDIX 3




























50 237 plusmn 255



























New York New York












New York





APPENDIX 1














Viejo

APPENDIX 2














W arty tube rculate



of upper jaw





1 Ab sen t





h yoid plate



APPENDIX 3
























PhylogenyofFrogsofthe PhysalaemusPustulosus SpeciesGroup ... · 1998...

Documents

Transcript of PhylogenyofFrogsofthe PhysalaemusPustulosus SpeciesGroup ... · 1998...