Daniels et al., Evolution of Afrotropical freshwater crab lineages obscured by morphological convergence (2006).pdf

Molecular Phylogenetics and Evolution 40 (2006) 227–235

www.elsevier.com/locate/ympev

Evolution of Afrotropical freshwater crab lineages obscured

by morphological convergence

Savel R. Daniels a,b,¤ , Neil Cumberlidge c , Marcos Pérez-Losada a , Saskia A.E. Marijnissen d ,

Keith A. Crandall a,e

a Department of Integrative Biology, Brigham Young University, Provo, UT 84602, USA

b Department of Botany and Zoology, University of Stellenbosch, Private Bag X1, Matieland 7602, South Africa

c Department of Biology, Northern Michigan University, 1401 Presque Isle Ave, Marquette, MI 49855, USA

d Institute for Biodiversity and Ecosystem Dynamics, University of Amsterdam, Mauritskade 57, 1092 AD, Amsterdam, The Netherlands

e Monte L. Bean Museum, Brigham Young University, Provo, UT 84602-5255, USA

Received 17 October 2005; revised 21 February 2006; accepted 22 February 2006

Available online 18 April 2006

Abstract

nities among all freshwater crab

families. With an emphasis on the Afrotropical fauna that includes Africa, Madagascar, and the Seychelles, we test the proposed Gon-

dwanan cladogenesis of the group. Phylogenetic results demonstrate that contemporary distribution patterns of freshwater crab lineages

are incongruent with the expected area cladogram of continental fragmentation. Instead, our phylogenetic estimate and divergence time

estimation indicate a post-Gondwanan, early Cretaceous cladogenesis for freshwater crabs implying that the acquisition of a freshwater

lifestyle was achieved more recently. A dispersal hypothesis as opposed to vicariance appears to best explain the contemporary distribu-

tion pattern of this group. However, our results do not explicitly disprove a Gondwanan origin for the Afrotropical freshwater crabs.

Alarmingly, these results suggest that most of the currently recognized freshwater crab families are unreliable taxonomic groupings since

virtually no Afrotropical freshwater crab families formed monophyletic units thus obscuring inferred biogeographic relationships. Con-

vergence in characters associated with the terminal segment of the mandibular palp is clearly a pervasive obstacle in the taxonomy of this

group.

We use sequence data derived from six DNA gene loci to examine evolutionary and biogeographic a

Keywords: Freshwater crabs; Biogeography; Cladogenesis; Dispersal; Vicariance; Convergence; Gondwana

1. Introduction

agascar separated (84–96 Mya), followed by the separation

of the granitic Seychelles archipelago from India

(68–65 Mya) while India continued drifting northeastward

towards the Eurasian plate, colliding with it 55–42 Mya,

during the early Eocene ( Briggs, 2003 ). Africa’s separation

from South America began 106 Mya and was completed by

84 Mya, during the late Cretaceous ( Goldblatt, 1993 ). Pro-

longed historic isolation of both Madagascar and the Sey-

chelles Islands in the Indian Ocean have led to near

complete endemicity in biotic assemblages that include

some spectacularly diverse adaptive radiations, resulting in

these areas being biodiversity hotspots. While the coloniza-

tion history of Madagascar’s vertebrate fauna has lately

er an ideal opportunity with which to test

biogeographic hypotheses derived from. Fragmentation of

Gondwana began during the mid-Mesozoic, 158–165 mil-

lion years ago (Mya), with the rifting of the Indo-Madagas-

car continent from Africa ending 121 Mya ( Rabinowitz

et al., 1983; Storey, 1995a,b ). Subsequently, India and Mad-

Corresponding author. Fax: +27 21 808 2405.

E-mail addresses: srd@sun.ac.za , sdaniels@byu.edu (S.R. Daniels).

1055-7903/$ - see front matter

doi:10.1016/j.ympev.2006.02.022

The distribution pattern of extinct and extant taxa asso-

ciated with fragmentation and rifting of ancient continental

landmasses o

228

S.R. Daniels et al. / Molecular Phylogenetics and Evolution 40 (2006) 227–235

cant attention recently ( De Queiroz, 2005;

Nagy et al., 2003; Raxworthy et al., 2002; Sparks and

Smith, 2004; Vences et al., 2003; Yoder et al., 2003 ), how-

ever there are few studies focusing on the island’s endemic

invertebrate taxa. Consequently, phylogenetic and biogeo-

graphic a

ences are derived under the assumption that the morpho-

logical characters used to unite each of the three

superfamilies are synapomorphic and to suggest that

freshwater crabs are monophyletic. In the absence of a

phylogenetic analysis for the group, the character state

remains unclear and may re

nities among invertebrates that also appear to

have undergone a radiation on Madagascar remain enig-

matic and the subject of vigorous debate ( Bott, 1955, 1965;

Ng et al., 1995; Ortmann, 1902; Rodriguez, 1986; Von

Sternberg et al., 1999 ).

The distribution and phylogenetic a

ect sympleisiomorphy.

Noticeably, several taxonomic authorities have ques-

tioned the validity of the superfamilies and their stability

( Cumberlidge, 1999; Holthuis, 1979; Ng, 2004; Ng and

Sket, 1996; Ng and Tan, 1998; Von Sternberg et al., 1999 )

further fueling the biogeographic debate over freshwater

crab origins. This hypothesis suggests that ancient conti-

nental fragmentation from a ‘centre of origin’ is the main

factor that explains the contemporary distribution of

freshwater crabs.

Conversely, hypothesis two suggests a recent post-

Gondwana (Cretaceous) evolution of freshwater crabs from

a marine ancestor ( Carriol and Secretan, 1992; Colosi, 1921;

Glaessner, 1969; Von Sternberg et al., 1999 ). In addition this

post-Gondwana hypothesis implies dispersal from a wide-

spread monophyletic but polymorphic ancestral taxon to

explain the contemporary distribution of the group. Further-

more, the fossil record for freshwater crabs is relatively

young, with the oldest fossils being from the upper Miocence.

Nevertheless, aquatic invertebrates, including freshwater

crabs, have a notoriously poor paleontological record, and

the absence of geologically ancient fossils cannot be unequiv-

ocally used to disprove Gondwanan origins.

In an attempt to test the two hypotheses proposed for

the origin and radiation of these freshwater crabs, we col-

lected partial sequence data from six gene loci that included

four mitochondrial genes (two ribosomal loci, 16S rRNA

and 12S rRNA, two protein coding loci, cytochrome c oxi-

dase subunit I (COI), and cytochrome c oxidase subunit II

(COII)), as well as two nuclear (nDNA) loci, one protein

coding nuclear gene locus (histone H3), and one ribosomal

locus (18S rRNA) for all three freshwater crab super fami-

lies (Potamoidea, Gecarcinucoidea, and Pseudothelphusoi-

dea). Our study includes 65 taxa representing Potamoidea

(46 species), Gecarcinucoidea (18 species), and the Pseud-

othelphusoidea (1 species).

nities of other

freshwater crustaceans, such as freshwater cray

sh, sup-

port the Gondwanan origin hypothesis ( Crandall et al.,

2000 ). Nevertheless, despite constituting 15% of brachyu-

ran diversity (Cumberlidge pers. comm.), evolutionary

relationships among freshwater crab families have

remained unresolved. Contemporary freshwater crabs

complete their life cycles exclusively in freshwater habi-

tats, are incapable of surviving prolonged exposure in

brackish or saline environments, and are characterized by

the production of a small number of yolk-rich (lecitho-

trophic) eggs followed by direct development and the

absence of free-swimming planktonic larval phases with

low assumed dispersal capacity ( Von Sternberg et al.,

1999 ). Two main hypotheses have been proposed to

explain the contemporary distribution of freshwater

crabs. Hypothesis one proposes an ancient Gondwanan

origin for the three freshwater crab superfamilies (Pota-

moidea, Gecarcinucoidea, and Pseudothelphusoidea) ( Ng

et al., 1995; Rodriguez, 1986 ). Morphological similarities

in diagnostic features of the terminal segment of the man-

dibular palp shared between African and South American

freshwater crab families and between African, Madaga-

scan, and Indian taxa have been used as evidence to sup-

port a Gondwanan origin ( Rodriguez, 1986 ). In addition,

Madagascar’s freshwater crab fauna includes taxa that

have long been assumed by some authors to be represen-

tative of transitional forms a

ected by continental frag-

mentation, with some of the island’s taxa apparently

showing evolutionary a

nities with both African and

Indian taxa ( Bott, 1965; Ng et al., 1995 ; but see Cumber-

lidge and Von Sternberg, 2002 for an alternative view).

The presence of a single freshwater crab species on the

Seychelles Islands, placed in the same family present on

the Indian subcontinent, would imply an ancient biogeo-

graphic link between these areas, further supporting the

assumed Gondwanan origin of freshwater crabs ( Ng et al.,

1995 ). However, earlier taxonomic work undertaken by

Bott (1955) placed the single Seychelles freshwater crab

species as a congener of the East African Deckenia (as D.

alluaudi in the family Deckeniindae), underscoring the

taxonomic instability of the group. Additionally, the

occurrence of freshwater crabs on continental islands

(such as Madagascar, Seychelles, Socotra, and Sri Lanka)

and their complete absence from oceanic islands such as

the Comoros, Reunion, and Mauritius would be further

evidence of an ancient origin. These biogeographic infer-

2. Materials and methods

2.1. Sample collection

Representatives of all the described Malagasy freshwater

crab samples (excluding only Boreas uglowi ) were obtained

from the Field Museum of Natural History, Chicago

(USA), from the Northern Michigan University collection

or from donations to authors by individuals. The following

received signi

ve Afrotropical families were included: (1) four African

genera in Potamonautidae (23 Potamonautes , 1 Potamone-

mus , 2 Sudanonautes , and 1 Liberonautes ), these included

sequence data from 14 southern African Potamonautes spe-

cies from an earlier study ( Daniels et al., 2002 ), which were

downloaded from GenBank (12S rRNA, 16S rRNA, and

S.R. Daniels et al. / Molecular Phylogenetics and Evolution 40 (2006) 227–235

229

COI mtDNA) and included in the present analysis, and

sequenced for the outstanding three gene fragments. Five

Malagasy genera in Potamonautidae (1 Madagapotamon , 4

Hydrothelphusa , 1 Marojejy , 2 Malagasya , 1 Skelosophusa ,

and 1 Foza raimundi ), the single genus in the Platythel-

phusidae (1 Platythelphusa ), the single genus in Deckenii-

dae (1 Deckenia ); Potamidae (1 Socotra —the only

Afrotropical genus in the family) and Gecarcinucidae (1

Seychellum —the only Afrotropical genus in the family). In

addition, we included representatives of the families Pota-

midae (1 Potamon , 1 Isolapotamon , 1 Johara , and 2 Geothel-

phusa ), the Oriental Gecarcinucidae (4 Barythelphusa , 1

Gecarcinucus , 1 Gubernatoriana , 1 Sartoriana , and 1 Phric-

othelphusa ), and the Oriental and Australasian family Para-

thelphusidae (1 Ceylonthelphusa , 2 Parathelphusa , 2

Rouxana , 1 Sayamia , and 1 Oziothelphusa ) as well as a sin-

gle representative of the Neotropical Pseudothelphusidae,

Guinotia dentata . A list of the voucher specimens of the taxa

sequenced in this study, together with their museum cata-

logue numbers and localities is provided in Table 1 .

Table 1

List of voucher specimens of freshwater crab taxa sequenced in the pres-

ent study and collection localities

Species

Museum catalogue number Locality

Johora tiomanensis

ZRC 1999.0899

Malaysia

Isolapotamon consobrinum

ZRC 2000.2215

Borneo

Geothelphusa albogilva

ZRC 1999.0202

Taiwan

Celonthelphusa rugosa

ZRC 1998.606

Sri Lanka

Parathelphusa maculata

ZRC 2000.2110

Singapore

Sayamia sexpunctata

ZRC 2001.1101

Malaysia

Phricotelphusa limula

ZRC 2005.107

Thailand

Skelosophusa eumeces

FMNH 11059

Madagascar

Hydrothelphusa

bombetokensis

FMNH 6878

Madagascar

Marojejy longimerus

FMNH 4656

Madagascar

Madagapotamon humberti

FMNH 11049

Madagascar

Hydrothelphusa goudoti

FMNH 4652

Madagascar

Hydrothelphusa

madagascariensis

FMNH 7591

Madagascar

Foza raimundi

FMNH 7438

Madagascar

Gecarcinus jaquemonti

NMU A28.9.2004.1

India

Barythelphusa cunicularis 1

NMU B28.9.2004.1

India

Barythelphusa jaquemonti 1

NMU 06.10.2004.1

India

Barythelphusa jaquemonti 2

NMU 19.11.2004.1

India

2.2. Outgroup selection

Barythelphusa jaquemonti 3

Unaccessioned

India

Gubernatoriana gubernatorus NMU 20.11.2004.1

India

Barythelphusa cunicularis 2

NMU 10.4.2004.1

India

erent families were used as out-

group taxa; these included the freshwater crab species Dilo-

carcinus laevifrons (Trichodactylidae), and two marine

crabs Pachygrapsus marmoratus (Grapsidae) and Carcinus

maenas (Portunidae). A number of authors ( Martin and

Davis, 2001; Rodriguez, 1992; Von Sternberg, 1997; Von

Sternberg et al., 1999; Von Sternberg and Cumberlidge,

2003 ) have suggested that Trichodactylidae is sister to Por-

tunidae while the remainder of the three freshwater super

families are potentially monophyletic, hence Trichodactyli-

dae along with two additional marine taxa were used as

outgroups.

DNA was extracted from leg muscle or gill tissue using

the Qiagen DNeasy tissue kit followed by polymerase chain

reaction (PCR) with the following primer pairs; 16S rRNA

( Cunningham et al., 1992 ); 12S rRNA ( Kocher et al., 1989 ),

cytochrome c oxidase subunit I—COI ( Folmer et al., 1994 ),

cytochrome c oxidase subunit II—COII ( Simon et al.,

1994 ), histone three (H3) ( Colgan et al., 1998 ), and 18S

rRNA ( Whiting et al., 1997 ). The DNA was denatured for

94 °C for 5 min, followed by 40 cycles at 94 °C for 40 s,

annealing temperatures between 48 and 42 °C for 45 s and

an extension at 72 °C for 1 min, with a

Barythelphusa sp. 1

NMU 11.10.2004.1

India

Barythelphusa sp. 2

NMU 11.10.2004.1

India

Oziothelphusa senex

NMU 07.10.2004.1

India

Sartoriana spinigera

NMU 09.09.2003.1

India

Rouxana minima

NMU A21.09.2000.1

Papua New

Guinea

Rouxana ingrami

NMU B21.09.2000.1

Papua New

Guinea

Parathelphusa sp.

NMU 16.08.2003.1

Malaysia

Sudanonautes Xoweri

NMU 03.2004.1

Gabon

Potamonautes ecorssei

NMU 07.01.2003.1

Mali

Potamonautes platynotus

NMU 23.04.2003.6

Tanzania

Guinotia dentata

NMU 04.01.2004.1

West Indies

Potamonautes odhnerhi

NMU 14.07.2004

Kenya

Potamonautes new sp.

NMU 18.10.2003

Kenya

Deckenia imitatrix

NMU 1998.1

Kenya

Sudanonautes aubryi

NMU 23.04.1984A

Nigeria

Potamonemus sachi

NMU 09.04.1983

Cameroon

Liberonautes rubigimanus

NMU 18.04. 1989

Liberia

Seychellum alluaudi

MRAC 53.653

Seychelles

Socotra pseudocardisoma

NMU 10.1998.1

Socotra

Potamonautes raybouldi

ZMA Crust. De. 204684

Tanzania

Potamonautes lirrangensis

ZMA Crust. De. 204681

DRC

Potamonautes emini

ZMA Crust. De. 204680

DRC

Potamonautes niloticus

ZMA Crust. De. 204683

Uganda

Platythelphusa armata

ZMA Crust. De. 204685

Tanzania

nal extension phase

at 72 °C for 10 min. PCR products were gel puri

Carcinus maenas

BYUKC 3044

Spain

ed with a

Pachygrapsus marmoratus

BYUKC 3045

Spain

Dilocarcinus laevifrons

NMU 13.12.2003.1

Peru

cation kit followed by direct sequenc-

ing on an ABI 3730 XL automated sequencer. DNA

sequences for all six loci have the following GenBank

Accession Nos., 12S rRNA (AY803490–AY803529,

AY919073–AY919085), 16S rRNA (AY803530–

AY803567, AY919086–AY919095), COI (AY803568–

AY803593, AY919108–AY919116), COII (AY803594–

AY803635, AY919117–AY919125), 18S rRNA

(AY803636–AY803672, AY919096–AY919107), and H3

(AY803673–AY803723, AY919126–138).

Potamon Xuviatilis

Unaccessioned

Italy

Zoological Reference Collection of the RaZes Museum of Biodiversity

Research, National University of Singapore (ZRC); Field Museum of Natu-

ral History, Chicago (FMNH); Northern Michigan University, Marquette

(NMU); Zoological Museum of Amsterdam, Crustacean Department

(ZMA); KC, Brigham Young University, Utah (BYU); and Musée Royal

de l’Afrique Central, Tervuren (MRAC). The 14 southern African Potamo-

nautes species sequenced were not deposited in a museum collection and

sequences were taken from GenBank for 12S, 16S, and COI mtDNA. The

abbreviation DRC stands for Democratic Republic of the Congo.

Three crabs from di

Qiagen PCR gel puri

230

S.R. Daniels et al. / Molecular Phylogenetics and Evolution 40 (2006) 227–235

2.3. Alignment and phylogenetic analysis

imated twice under the Bayesian approach and included a

burn-in period of 5 £ 10 5 steps, after which 10 6 samples

were collected every 100 accepted states. The similarity of

these two approximations indicated that the Markov

chains successfully converged. Likelihood estimations were

performed also twice using small values of 1e¡6 and

0.5e¡6 in the di

The three ribosomal gene sequences 16S rRNA, 12S

rRNA mtDNA, and 18S rRNA (652 bp) were aligned using

ClustalX ( Thompson et al., 1997 ) and manually adjusted

where mismatches were made, while the two protein coding

loci COI and COII (COI—610 bp and COII—447 bp) as

well as the nuclear markers histone three (H3) (301 bp) were

aligned manually. Large insertion and deletions that could

not be aligned with con

erence approximation of derivatives and

similar results were also encountered. Three independent

calibrations were used specifying upper and lower bounds

for the Bayesian approach and

dence for 12S and 16S rRNA

mtDNA were deleted from the phylogenetic analysis, these

resulted in two short sequences that were 272 and 352 bp,

respectively. Phylogenetic trees were estimated with

PAUP*4.0b10 ( Swo

xed ages for the likelihood

approach. A detailed discussion of the method is outlined

in Yang and Yoder (2003) and used by Nagy et al. (2003),

Porter et al. (2005), Vences et al. (2003), Yang (2004), and

Yoder et al. (2003) . Calibration point 1 (on Fig. 1 ) repre-

sents the maximum age estimation of the Potamon fossil

from the upper Miocene (24 Mya), calibration point 2 (on

Fig. 1 ) represents the geological age of the Seychelles Island

at 70 Mya, while calibration point 3 (on Fig. 1 ) represents

the Potamonautes niloticus fossil from six My old sediments

from the lower Miocene of Kenya. The two fossil calibra-

tion points were set as the lower limit, while the geological

age of the Seychelles was used as the upper limit in the

divergence time estimations.

ord, 2002 ), using parsimony (MP) and

maximum likelihood (ML) and with MrBayes 3.0b4 ( Ron-

quist and Huelsenbeck, 2003 ) using Bayesian inferences

coupled with Markov chain Monte Carlo techniques.

For the MP and ML analysis, heuristic searches with

TBR branch swapping and 100 random additions of taxa

were performed. Nodal support for MP was estimated by

bootstrapping analyzing 10,000 pseudo-replicates of data

sets, while due to computational constraints only 100

pseudo-replicates were performed for ML. For each Bayes-

ian analysis 10 Markov chains were run, with each chain

started from a random tree and running for six million gen-

erations and sampling from the chain every 5000 trees. In

the combined Bayesian analyses, data sets were partitioned

and independent models were selected for each gene using

unlinked parameters. A 50% majority rule consensus tree

was generated from the trees retained (after the burn-in

trees were discarded), with posterior probabilities for each

node estimated by the percentage of time the node was

recovered. Convergence and mixing were monitored using

the program Tracer v1.2 ( Rambaut and Drummond, 2004 ).

For the ML and Bayesian analyses, MODELTEST

( Posada and Crandall, 1998 ) was used to

3. Results and discussion

nd the appropri-

ate substitution model of evolution for each gene and the

combined data set. The best-

Phylogenetic analysis performed on the mitochondrial

and nuclear genes prior to combination into a single data

matrix recovered highly congruent topologies (independent

mt and nDNA tree topologies not shown). For MP, 956

parsimony informative characters were present that recov-

ered a single tree with a tree length of 7346 steps, with a

consistency index (CI) of 0.24 and a retention index (RI) of

0.47. The MP topology was identical to the ML and Bayes-

ian topologies. For the combined data set the substitution

model GTR (General Time-Reversible) + I (proportion of

invariable sites D 0.4587) +

(gamma distribution shape

parameter D 0.6915) had the following parameters, base fre-

quencies A D 0.3378, C D 0.1402, G D 0.2017, and

T D 0.3203, the rate matrix R (a) [A–C]D 0.4722, R (b)

[A–G] D 6.2458, R (c) [A–T]D 1.5699, R (d) [C–G]D 1.5106,

R (e) [C–T] D 2.5940, and R (f) [G–T] D 1.00.

Results derived from several phylogenetic methods (par-

simony, maximum likelihood, and Bayesian inference)

( Fig. 1 ) demonstrate that the Afrotropical freshwater crabs

samples are monophyletic, based on our taxon sampling.

However, none of the Afrotropical and Oriental families

formed natural groups, suggesting the arti

t maximum likelihood score

was chosen using the Akaike Information Criterion (AIC)

since this reduced the amount of unnecessary parameters

that contribute little to describing the data by penalizing

more complex models ( Bernham and Anderson, 2002;

Nylander et al., 2004 ). The Shimodaira and Hasegawa

(1999) test and posterior probability as implemented in

PAUP*4.0b10 ( Swo

ord, 2002 ) was used to test the tradi-

tional taxonomic designations (comparing the uncon-

strained tree to the null hypothesis).

cial nature of

the current taxonomic designations. For example, the

monophyletic Malagasy Potamonautidae formed a distinct

clade from the African Potamonautidae while the East

African Platythelphusidae was deeply nested within the

African Potamonautidae. We did not recover a sister taxon

relationship between the Seychellian and Indian freshwater

crabs despite their present assignment to the same family

(Gecarcinucidae); that is further incongruent with a Gon-

dwanan fragmentation hypothesis. Instead, the Seychelles

2.4. Divergence time estimation

Divergence times were estimated under the multi-locus

Bayesian method of Thorne et al. (1998) and Thorne and

Kishino (2002) , as implemented in the multidivtime pack-

age ( http://statgen.ncsu.edu/thorne/multidivtime.html ), and

the likelihood heuristic rate-smoothing algorithm of Yang

(1997) , as implemented in PAML3.14b3 ( Yang, 1997; Yang

and Yoder, 2003 ). The posterior distributions were approx-

S.R. Daniels et al. / Molecular Phylogenetics and Evolution 40 (2006) 227–235

231

Pachygrapsusmarmoratus

Dilocarcinuslaevifrons

Carcinus maenas

OUTGROUPS

100 / 1

Barythelphusasp 1

Bar ythelphusacunicularis1

Ba rythelphusasp 2

Barythelph usa cunicularis2

Baryth elphusa jacquemontii1

89 / 1

79 / 1

100

69 / 1

100 / 1

Barythelphusajacquemontii2

Ba rythelphusa jacquemontii3

100

India

63/ 1

Sartori anaspinigera

Gecarcinucusjacquemontii

Gubernatorianagubernatoris

Ceylonthelphusarugosa

Oziothe lphusasenex

Parathelphusamaculata

Parathelphusasp

Sayamiasexpunctata

Rouxana ingrahmi

Rouxanaminima

Phricotelphusa limula

Guinotiadentata

S ocotrapseudocardisoma

Johora tiomanensis

Geothelph usasp

Geothelphusaalbogilva

Potamonfluviatilis

Isolapotamon consobrinum

Deckenia imitatrix

Seychellumalluaudi

Madagap otamonhumberti

Foza raimundi

Hydrothelphusagoudoti

Hydrothelphusaagilis

Hydrothelphusamadagascariensis

Hy drothelphusabombetokensis

Ma rojejylongimerus

S kelosophusaeumeces

Ma lagasyaantongilensis

Malagasyagoodmani

Potamonemussachsi

Sudanonautesfloweri

S udanonautesaubryi

Liberona utesrubigimanus

Potamonautescalcaratus

Potamonautesobesus

Pota monauteslirrangensis

P otamonautesplatynotus

Potamonautes raybouldi

Platythelphusaarmata

P otamonautesniloticus

Potamonautesemini

P otamonautesnew sp

P otamonautesodhneri

Potamonautesecorssei

Po tamonautesclarus

Potamonautesdepressus

P otamonautesbrincki

Potamonautesparvispina

P otamonautesparvicorpus

Pota monautesdentatus

Potamonauteslividus

Potamonautesgranularis

Potamonautesperlatus

Potamonautessidneyi

Potamonautesbayonianus

Potamonautesunispinus

Potamonauteswarreni

6 9 / 1

76 / 1

Asia and

Papua New

Guinea

64 /1

100 / 1

Freshwater crab families

Gecarcinucidae

Parathelphusidae

Potamindae

Pseudothelphusidae

Decke niidae

Potamonaut idae

Platythelphu sidae

India

* *

West Indies

Socotra

Asia

79 / 1

Europ Asia

Africa

100 / 1

100

Seychelles

77 / 1

100 / 1

82 / 1

94 /1

ML / BI

100

87 / 1

74 / 1

Madagascar

81 / 1

Afrotropical fauna

100 / 1

75 / 1

99 / 1

75 / 1

1 00 / 1

99 / 1

Africa

70 / 1

100 / 1

100

97 / 1

98 / 1

9 0 / 1

98 / 1

100 / 1

90 / 1

100 / 1

100

0.05 substitutions / site

100 / 1

Fig. 1. A ML topology derived from the model GTR + I + , with a likelihood score of 36685.95. The tree was rooted using three outgroups ( Dilocarcinus

laevifrons , Pachygrapsus marmoratus , and Carcinus maenas ). The values above each node represent the bootstrap values for ML and posterior probability

for Bayesian inferences respectively, while the values below each node represent the MP bootstrap values. Nodes marked with an asterisk had bootstrap

values of <60% and posterior probabilities of <0.95.

taxon was closely related to an East African family (Decki-

niidae), and formed a clade sister to the Malagasy freshwa-

ter crabs. Among the Indian and Asian Gecarcinucidae and

Parathelphusidae we further found no evidence of distinct

family groups. The estimated phylogram ( Fig. 1 ) is com-

pletely irreconcilable with the expected area cladogram of

continental fragmentation. Enforcing a strict Gondwanan

fragmentation area cladogram consistently recovered a

topology that was statistically signi

dence ( Table 2 ) given these data. The

evolutionary distinctiveness of the Afrotropical families

from the Neotropical (South American and Meso-Ameri-

can) and the Oriental and Australasian families provides

further evidence for a post-Gondwanan evolution of these

freshwater crabs. A post-Gondwanan origin of Afrotropi-

cal freshwater crabs is further supported by the shallow

corrected sequence divergence values between the African

and Malagasy freshwater crab families. Divergence time

cantly worse, and can

be rejected with con

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: