Phylogenetic and taxonomic relationships of the …xa.yimg.com/kq/groups/20809606/100126701/name/Zoological...Phylogenetic and taxonomic relationships of the Polypedates leucomystaxcomplex

Zoologica Scripta

Phylogenetic and taxonomic relationships of the Polypedatesleucomystax complex (Amphibia)NORIHIRO KURAISHI, MASAFUMI MATSUI, AMIR HAMIDY, DAICUS M. BELABUT, NORHAYATI AHMAD,SOMSAK PANHA, AHMAD SUDIN, HOI S. YONG, JIAN-PING JIANG, HIDETOSHI OTA, HO T. THONG &KANTO NISHIKAWA

Submitted: 7 April 2012Accepted: 3 July 2012doi:10.1111/j.1463-6409.2012.00562.x

ª 2012 The Authors d Zoologica Scripta ª 2012 The Norwegian

Kuraishi, N., Matsui, M., Hamidy, A., Belabut, D. M., Ahmad, N., Panha, S., Sudin, A., Yong,

H. S., Jiang, J.-P., Ota, H., Thong, H. T. & Nishikawa, K. (2012). Phylogenetic and

taxonomic relationships of the Polypedates leucomystax complex (Amphibia). —Zoologica Scripta,

00, 000–000.

We investigated the phylogenetic and taxonomic relationships and estimated the history of

species diversification and biogeography in the Asian rhacophorid genus Polypedates, focus-

ing on the Polypedates leucomystax complex, whose members are notoriously difficult to clas-

sify. We first estimated phylogenetic relationships within the complex using 2005-bp

sequences of the mitochondrial 12S rRNA, tRNAval and 16S rRNA genes with maximum

parsimony, maximum likelihood (ML) and Bayesian methods of inference. Polypedates

exhibits well-supported monophyly, with distinct clades for P. otilophus, P. colletti, P. macula-

tus and the P. leucomystax complex, consisting of P. macrotis, and the Malay (Polypedates sp.

from Malay Peninsula), North China (P. braueri), South China (Polypedates cf. mutus 1),

Indochina (P. megacephalus), Sunda (P. leucomystax) and Laos (Polypedates cf. mutus 2) clades.

In a subsequent phylogenetic analysis of 4696-bp sequences of the nuclear brain-derived

neurotrophic factor (BDNF), sodium ⁄ calcium exchanger 1 (NCX), POMC, Rag-1, Rhod and

Tyr genes using Bayesian methods of inference, all of these clades were recovered. Some

clades of the P. leucomystax complex occur sympatrically and show high genetic diversity or

morphological and acoustic differences. Similar tendencies were observed between some

allopatric clades. Therefore, we consider each of these groups to be distinct specifically.

We also estimated absolute divergence times within the genus using Bayesian methods.

Divergence in Polypedates began with the divergence of a primarily South Asian Clade from

the common ancestor of secondarily South-East Asia P. maculatus and South-East Asian

members. The divergence between the latter occurred much later. The P. leucomystax com-

plex diverged in the Pliocene, much later than other congeners, and seems to have been

greatly affected by human-related dispersal after the Pleistocene.

Corresponding author: Masafumi Matsui, Graduate School of Human and Environmental

Studies, Kyoto University, Kyoto 606-8501, Japan. E-mail: [email protected]

Norihiro Kuraishi, Graduate School of Human and Environmental Studies, Kyoto University,

Kyoto 606-8501, Japan. E-mail: [email protected]

Amir Hamidy, Graduate School of Human and Environmental Studies, Kyoto University, Kyoto

606-8501, Japan; Museum Zoologicum Bogoriense (MZB), Research Center for Biology, Indone-

sian Institute of Sciences, Gd. Widyasatwaloka, Jl. Raya Jakarta Bogor km 46, Cibinong West

Java, Indonesia. E-mail: [email protected]

Daicus M. Belabut, Institute of Biological Sciences, Faculty of Science, University of Malaya,

50603 Kuala Lumpur, Malaysia; Institute for Environment and Development (LESTARI),

Universiti Kebangsaan Malaysia, 43600 UKM Bangi, Selangor, Malaysia. E-mail: daicus@

um.edu.my

Norhayati Ahmad, Institute for Environment and Development (LESTARI), Universiti Keb-

angsaan Malaysia, 43600 UKM Bangi, Selangor, Malaysia; Faculty of Science and Technology,

Universiti Kebangsaan Malaysia, 43600 UKM Bangi, Selangor, Malaysia. E-mail: yati_68

@yahoo.co.uk

Academy of Science and Letters 1

Somsak Panha, Department of Biology, Faculty of Science, Chulalongkorn University, Bangkok

10330, Thailand. E-mail: [email protected]

Ahmad Sudin, Institute for Tropical Biology and Conservation, University Malaysia Sabah, Kota

Kinabalu 88999, Sabah, Malaysia. E-mail: [email protected]

Hoi Sen Yong, Institute of Biological Sciences, Faculty of Science, University of Malaya, 50603

Kuala Lumpur, Malaysia. E-mail: [email protected]

Jian-Ping Jiang, Chengdu Institute of Biology, Chinese Academy of Sciences, Chengdu 610041,

China. E-mail: [email protected]

Hidetoshi Ota, Institute of Natural and Environmental Sciences and Museum of Nature and

Human Activities, University of Hyogo, Yayoigaoka 6, Sanda, Hyogo 669-1546, Japan. E-mail:

[email protected]

Ho Trung Thong, Institute Hue University of Agriculture and Forestry, 102 Phung Hung, Hue,

Vietnam. E-mail: [email protected]

Kanto Nishikawa, Graduate School of Human and Environmental Studies, Kyoto University, Kyoto

606-8501, Japan. E-mail: [email protected]

Phylogenetic study on the Polypedates leucomystax complex d N. Kuraishi et al.

IntroductionRecent molecular phylogenetic analyses have revealed the

presence of cryptic species in many lineages of wide-rang-

ing animals, including frogs from South-East Asia, a cen-

tre of anuran diversity. Some examples of taxa found to

contain cryptic taxa include Fejervarya limnocharis (Toda

et al. 1998; Veith et al. 2001), Microhyla ornata (Matsui

et al. 2005), Staurois tuberilinguis (Matsui et al. 2007), Hyla-

rana chalconota (Stuart et al. 2006; Inger et al. 2009) and

Limnonectes kuhlii (Matsui et al. 2010c).

The Polypedates leucomystax complex (Matsui et al. 1986;

Dubois 1987 as Rhacophorus; Orlov et al. 2001) encom-

passes frogs that are notoriously difficult to identify and

distinguish from P. leucomystax. Frogs of this complex

occur widely from Nepal, through southern East Asia, to

all of South-East Asia. Several South-East Asian species

like P. macrotis and P. mutus and some South Asian (Indian

and Sri Lankan) species like P. maculatus were once associ-

ated with P. leucomystax, although all of them are currently

considered distinct species. Based on external morphology,

karyotype and mating calls, Matsui et al. (1986) separated

P. megacephalus as represented by the Taiwanese popula-

tions from P. leucomystax (see Kuraishi et al. 2011).

Subsequently, the presence of cryptic species in the

complex has been reported. The occurrence of two syn-

topic and surely independent species has been described

from Vietnam (Inger et al. 1999; Ziegler 2002) and the

Malay Peninsula (Narins et al. 1998). However, the taxo-

nomic relationships of these species pairs have never been

resolved, leaving the taxonomy of the P. leucomystax com-

plex unclear. Although the family Rhacophoridae, includ-

ing some species of Polypedates, has been included in

recent molecular studies on anuran phylogeny (e.g. Rich-

ards & Moore 1998; Marmayou et al. 2000; Wilkinson

et al. 2002; Delorme 2004; Frost et al. 2006; Li et al.

2008), few attempts have been made to elucidate phyloge-

2 ª 2

netic relationships among members of the P. leucomystaxcomplex.

Recently, Brown et al. (2010) analysed the phylogenetic

relationships among samples of the P. leucomystax complex

using mitochondrial DNA (mtDNA) genes. However, the

majority of their samples were from the Philippines and

Sulawesi, where frogs now identified as P. leucomystax pre-

dominate (see Riyanto et al. 2011 for the taxonomic status

of the Sulawesi population), and samples from the conti-

nent, where more taxa constitute difficult taxonomic prob-

lems, were limited. Specifically, no samples from Hong

Kong, the type locality of P. megacephalus (a key species

for understanding the relationships within the complex),

were included in their study. However, Brown et al. (2010)

did not resolve the phylogenetic and taxonomic problems

of the complex. More recently, Sheridan et al. (2010) com-

pared one population each of the P. leucomystax complex

from Thailand and Singapore using mtDNA and acoustic

characteristics, but reached no definite taxonomic

conclusions.

Some studies have suggested the sympatric occurrence

of two taxa of the P. leucomystax complex in Vietnam,

Malaysia and the Philippines (Narins et al. 1998; Inger

et al. 1999; Ziegler 2002; Brown et al. 2010). This leads to

the hypothesis that similar sympatric occurrences take

place at other localities within the wide range of this spe-

cies complex. The sympatric distribution of more than

one genetically distinct lineage can be strong evidence in

favour of their distinct specific status. Therefore, deter-

mining their distribution pattern would be useful for solv-

ing taxonomic problems of this complicated species

complex. In this study, by analysing mitochondrial and

nuclear DNA genes, we clarify the phylogenetic relation-

ships of samples of the P. leucomystax complex and allied

species collected over a wide range in South-East and

South Asia to obtain basic information for use in future

012 The Authors d Zoologica Scripta ª 2012 The Norwegian Academy of Science and Letters

N. Kuraishi et al. d Phylogenetic study on the Polypedates leucomystax complex

taxonomic revision of the complex. We also discuss the

historical biogeography of the genus Polypedates in these

regions.

Materials and methodsSampling design

Our own samples of Polypedates consisted of 205 speci-

mens ⁄ tissues obtained from China, Indochina, Sundaland,

the Philippines, Sulawesi, Okinawa and India (Fig. 1),

including the following taxa: P. macrotis, P. maculatus,

P. colletti, P. otilophus and the P. leucomystax complex

(Table S1; P. iskandari synonymized with P. leucomystax

following Kurniati 2011). We used Rhacophorus norhayatii,

Buergeria buergeri and Rana japonica as out-group taxa. We

obtained mtDNA sequences for B. buergeri from GenBank

(AB127977). Rhacophorus is very close to Polypedates (Liem

1970), although their sister relationship has not yet been

demonstrated unambiguously (Li et al. 2008; Yu et al.

2009), whereas Buergeria is a sister genus to the clade con-

sisting of all other rhacophorids, including Rhacophorus and

Polypedates (Richards & Moore 1998; Li et al. 2008; Yu

et al. 2009), and Ranidae (represented by Rana) is the sister

Fig. 1 Map of South-East Asia showing the Polypedates sampling

localities included in this study. Sample numbers correspond to

those given Table S1.

ª 2012 The Authors d Zoologica Scripta ª 2012 The Norwegian Academy of Science and Letters

group of Rhacophoridae plus its sister family Mantellidae

(Frost et al. 2006).

We first examined partial sequences of the 12S rRNA,

tRNAval and 16S rRNA mtDNA genes for all 208 samples

mentioned above (Table S1). Then, we analysed seven

nuclear (nu) DNA genes: BDNF, NCX, proopiomelano-

cortin A (POMC), recombination activating protein 1

(Rag1), rhodopsin (Rhod), solute carrier family 8 member

3 (SCF) and tyrosinase (Tyr) for 21 selected samples, rep-

resenting genetic groups recognized in the mtDNA genea-

logical tree and three out-group taxa (Table S2).

Voucher specimens ⁄ tissues are stored in the BORNE-

ENSIS collection, the University Malaysia Sabah (BORN),

Chengdu Institute of Biology (CIB), Kyoto University,

Graduate School of Human and Environmental Studies

(KUHE), Kyoto University, Department of Zoology

(KUZ), the Museum of Vertebrate Zoology, the Univer-

sity of California, Berkeley (MVZ), MZB, the United

States National Museum and the Zoological Institute,

Saint Petersburg.

DNA preparation, PCR and DNA sequencing

We obtained tissues from frozen or ethanol (95–99%)-

preserved specimens and extracted total genomic DNA

using a standard phenol ⁄ chloroform procedure (Hillis

et al. 1996). The tissues were homogenized in 0.6 mL of

STE buffer containing 10 mM Tris ⁄ HCl (pH 8.0),

100 mM NaCl and 1 mM EDTA (pH 8.0). Proteinase K

(0.1 mg ⁄ mL) was added to the homogenized solutions,

and the proteins were digested for 4–12 h at 55 �C. The

solution was treated with phenol and chloroform ⁄ isoamyl

alcohol, and DNA was precipitated with ethanol. The

DNA precipitates were dried and then resuspended in

0.6 mL TE (10 mM Tris ⁄ HCl, 1 mM EDTA, pH 8.0), and

1 lL was subjected to the polymerase chain reaction

(PCR).

For mtDNA, the PCR included an initial denaturation

for 5 min at 94 �C and 33 cycles of 15 s at 94 �C, 15 s at

53 �C and 2 min 40 s at 72 �C. The PCR primers are

shown in Table S3. The PCR products, purified using

polyethylene glycol (13%) precipitation, were used directly

as templates for cycle sequencing reactions with fluores-

cent dye-labelled terminator (BigDye Terminator v.3.1

Cycle Sequencing Kit; Applied Biosystems, Foster City,

CA, USA). The sequencing reaction products were puri-

fied by ethanol precipitation following the manufacturer’s

protocol and then run on an ABI PRISM 3130 Genetic Ana-

lyzer (Applied Biosystems). All samples were sequenced in

both directions using the same primers as for PCR and 10

additional sequencing primers (Table S3).

Sequence data obtained for each sample were adjusted

manually by eye using Chromas Pro (Technelysium,

3


Tewantin, Australia). The resulting sequences were depos-

ited in GenBank (AB564278–AB564288, AB727997–

AB728192). The sequences of each gene region were

aligned using the ClustalW option of Bioedit (Hall 1999).

After testing consistency among gene partitions using

incongruence length difference tests with 1000 randomized

partitions (Farris et al. 1994) and confirming no significant

heterogeneity, we obtained an alignment matrix with 2450

nucleotide sites (919, 72 and 1459 for 12S rRNA, tRNAval

and 16S rRNA, respectively). To exclude gaps and ambig-

uous areas, we further revised the alignment with Gblocks

0.91b (Castresana 2000) using the default parameters. The

revised alignment was 2005 bp long (817, 58 and 1130 bp

for 12S rRNA, tRNAval and 16S rRNA, respectively).

For nuDNA, we extracted DNA, amplified it using the

primers shown in Table S3 and sequenced these regions

using a process similar to that used for mtDNA. The

obtained sequences were deposited in GenBank

(AB728193–AB728323). We obtained an alignment matrix

with 4696 nucleotide sites (508 sites for BDNF, 1000 sites

for NCX, 571 sites for POMC, 780 sites for Rag1, 316

sites for Rhod, 1076 sites for SCF and 445 sites for Tyr;

Table S4). Using these sequences, we conducted two anal-

yses. First, we estimated the phylogenetic relationships

among the groups detected using mtDNA (14 samples

from 10 in-group clades and three out-group samples). No

heterogeneous sites were found in these samples. We

reconstructed a phylogenetic tree based on the method

used for the mtDNA analysis (see below). Second, we

examined our data for evidence of reproductive isolation

(absence of common nuDNA haplotypes) between the two

mitochondrial lineages of the P. leucomystax complex

[Indochina Clade (five samples) and Sunda Clade (six sam-

ples), see below]. We obtained heterogeneous sites in the

genotypes of BDNF and Rag 1 and inferred a haplotype

pair from them using PHASE ver. 2.1 (Stephens et al. 2001).

Phylogenetic analysis

For the mtDNA data, we used three methods for estimat-

ing phylogenetic relationships: maximum parsimony (MP)

using a heuristic search (tree-bisection-reconnection

branch-swapping algorithm with 100 random addition rep-

licates) with an equal weighting option; ML based on the

substitution model for each partition (12S rRNA, tRNAval

and 16S rRNA) and phylogenetic parameters, chosen by

the program Kakusan3 (Tanabe 2007) based on the Akaike

information criterion (AIC); and Bayesian inference (BI;

Rannala & Yang 1996; Huelsenbeck et al. 2001) with the

models derived from Kakusan3 based on the AIC using

four simultaneous Metropolis-coupled Monte Carlo

Markov chains for 6 000 000 generations and sampling a tree

every 100 generations. Convergence of the log-likelihood

4 ª 2

scores was checked using TRACER ver. 1.4 (Rambaut &

Drummond 2007). As we found that the log-likelihood

scores stabilized after 3 000 000 generations, we discarded

the first 30 000 trees (3 000 000 generations) as burn-in

and used the remaining 30 001 trees to estimate the phy-

logeny and Bayesian posterior probabilities (BPPs).

MP analyses were conducted using PAUP*4.0b (Swof-

ford 2002). Pairwise comparisons of uncorrected sequence

divergences (p-distance) in 16S rRNA were also made

using PAUP. ML analysis was conducted using TREEFIND-

ER ver. Oct. 2008 (Jobb 2008). Bayesian analysis was con-

ducted using MRBAYES v3.1.2 (Huelsenbeck & Ronquist

2001). The robustness of the MP and ML trees was tested

using bootstrap analyses (Felsenstein 1985) with 1000 rep-

licates. We regarded tree topologies with bootstrap values

(BS) ‡ 70% as sufficiently supported (Hillis & Bull 1993).

For the Bayesian analysis, we considered BPP ‡ 0.95 as

significant support (Huelsenbeck et al. 2001; Leache &

Reeder 2002; Huelsenbeck & Rannala 2004).

For the nuDNA tree based on the first data set (see

above), we used BI with the optimum substitution model

for each codon position (1st, 2nd or 3rd position) for each

nucleotide site derived from Kakusan3, based on the AIC.

Estimating divergence time

To estimate the age of each genetic group, we prepared

three different data sets: (i) 2005 bp (12S and 16S rRNA,

and tRNAval) of 36 mtDNA sequences were selected from

our original data set, including 29 individuals from the

P. leucomystax complex, three other South-East Asian Po-

lypedates (P. macrotis, P. colletti and P. otilophus), one South

Asian Polypedates (P. maculatus), two other rhacophorid

frogs (R. norhayatii and B. buergeri) and one ranid (R. japon-

ica); (ii) 933 bp (12S and 16S rRNA) of 17 mtDNA

sequences were selected from the original data set and

GenBank, including 10 samples of Polypedates, each repre-

senting genetic groups obtained from phylogenetic analy-

ses, three South Asian Polypedates [P. fastigo

(Meegaskumbura et al. 2002: AY141802, AY141848),

P. eques (Meegaskumbura et al. 2002: AY141800,

AY141846), and P. cruciger (Bossuyt & Milinkovitch 2000:

AF240928, AF249045)] and four out-group taxa, adding

one mantellid [Mantella madagascariensis (Bossuyt & Milin-

kovitch 2000: AF249005, AF249049)] to those used in data

set #1; and (iii) 4696 bp (BDNF, NCX, POMC, Rag1,

Rhod, SCF and Tyr) of 16 nuDNA sequences selected

from our original data set for 14 samples of Polypedates,

R. norhayatii and B. buergeri.

For all data sets, we estimated the divergence times

using a Bayesian relaxed molecular clock calculated in

BEAST (Drummond & Rambaut 2007) using 100 million

generations, discarding the first 10 million generations as



burn-in. Parameter values were sampled every 1000 gener-

ations, under a HKY substitution model and uncorrelated

lognormal ‘relaxed’ clock rate model (Drummond et al.

2006). Parameter estimates and convergence were checked

using TRACER ver. 1.4 (Rambaut & Drummond 2007).

For data set # 1, each external and internal calibration

point was used to estimate the dates of cladogenic events.

The divergence time Roelants et al. (2007) assumed to be

49.7 [95% credibility interval (CI) 36.6–57.6] million years

before present (MYBP) between Buergeria and other mem-

bers of Rhacophoridae was used as an external calibration

point, while an internal calibration point was set at the

divergence of the P. leucomystax complex populations from

Java and Sumatra as 2.8 (CI 1.4–4.5) MYBP, the values

estimated for the corresponding two populations of the

megophrid Leptobrachium hasseltii (Matsui et al. 2010b).

For data set #2, two external calibration points were used;

the divergence time of 73.1 (CI 53.6–92.6) MYBP between

Mantellidae and Rhacophoridae (Bossuyt & Milinkovitch

2000) was added to the divergence between Buergeria and

other members of the Rhacophoridae. For data set #3, the

divergence time between Buergeria and other members of

Rhacophoridae (Roelants et al. 2007) was used as an exter-

nal calibration point.

ResultsSequence statistics

Sequence statistics for the three mtDNA gene fragments

and for the combined alignment including all nucleotide

positions are provided in Table S4. The aligned 12S

rRNA, tRNAval and 16S rRNA data set consisted of 2005

characters, of which 823 sites were variable and 583 were

potentially phylogenetically informative. MP analysis

yielded 2400 most parsimonious trees of 2185 steps, a

consistency index of 0.543 and a retention index of 0.876.

The best substitution models estimated by Kakusan3 were

J2+ gamma shape parameter (G) of 0.245, J2 + G

(a = 0.269) and HKY (Hasegawa et al. 1985) + G

(a = 0.498) for the 12S rRNA, 16S rRNA and tRNAval,

respectively. For the Bayesian analyses, GTR (Tavare

1986) + G (a = 0.273) was selected as the best substitution

model for the 12S rRNA, GTR + G (a = 0.288) for the

16S rRNA and K80 + G (a = 0.535) for the tRNAval. The

ML and BI analyses produced topologies with lnL

)13536.09 and )13706.18, respectively (nucleotide fre-

quencies: A = 0.33670, C = 0.22394, G = 0.19895 and

T = 0.24042).

Sequence statistics for the seven nuDNA gene fragments

and for the combined alignment are given in Table S4.

For 24 samples including out-group taxa, we obtained

4696 bp of nuDNA, of which 746 were variable and 305

were parsimony informative. The best substitution models


selected for the Bayesian analyses by Kakusan3 were as

follows: K80 + G (a = 0.289), GTR and JC69 for

BDNF_p1, BDNF_p2 and BDNF_p3, respectively;

HKY85 + G (a = 1.296), HKY85 + G (a = 0.005) and

F81 + G (a = 0.005) for NCX_p1, NCX_p2 and NCX_p3,

respectively; HKY85 + G (a = 1.079), GTR + G

(a = 0.005) and HKY85 for POMC_p1, POMC_p2 and

POMC_p3, respectively; F81 + G (a = 0.046), F81 + G

(a = 0.005) and HKY85 + G (a = 1.778) for Rag1_p1,

Rag1_p2 and Rag1_p3, respectively; HKY85 + G

(a = 0.190), SYM and F81 for Rhod_p1, Rhod_p2 and

Rhod_p3, respectively; HKY85 + G (a = 1.597), GTR + G

(a = 0.005) and HKY85 for SCF_p1, SCF_p2 and

SCF_p3, respectively; GTR + G (a = 0.953), HKY85 + G

(a = 0.005) and HKY85 + G (a = 0.290) for Tyr_p1,

Tyr_p2 and Tyr_p3, respectively. The BI analyses of con-

catenated genes produced topologies with lnL –13706.18

(nucleotide frequencies: A = 0.23580, C = 0.22523,

G = 0.27249 and T = 0.26647).

Phylogenetic relationships

All mtDNA gene analyses resulted in essentially the same

topologies and differed only in associations with poorly

supported nodes. The Bayesian tree (Fig. 2) infers the fol-

lowing sets of relationships. Monophyly of Polypedates with

respect to Rhacophorus, Buergeria and Rana was supported

nearly fully in all trees (MPBS = 99%, MLBS = 100% and

BPP = 1.00). In the Polypedates clade, P. otilophus formed

the sister species to the clade of the remaining samples

(99%, 100%, 1.00). In the latter clade, P. colletti diverged

first, although the monophyly of the remaining samples

was not substantially supported by MLBS (71%, 55%,

0.96). Of the latter samples, P. maculatus diverged from

the clade of the P. leucomystax complex (84%, 83%, 1.00).

Within the P. leucomystax complex clade, 117 unique hapl-

otypes (Fig. 2) fell into three major clades with unresolved

relationships: P. macrotis, Polypedates sp. from the Malay

Peninsula (Malay Clade) and the remaining P. leucomystax

complex. The last major clade got substantial support for

monophyly only in MP (80%, 60%, 0.93) and split into three

clades: samples of the P. leucomystax complex from southern

China (Guangxi and Hainan) and Vietnam (Vinh Phu and

Kon Tum) (South China Clade: 100%, 100%, 1.00); samples

of the P. leucomystax complex from central, western and

southern China (Anhui, Zhejiang, Jiangxi, Sichuan, Guangxi

and Yunnan) and Taiwan (North China Clade: 100%,

100%, 1.00); and samples of the P. leucomystax complex

widely occurring from southern China (Hong Kong,

Guangxi and Hainan) to South-East Asia [Vietnam, Laos,

Myanmar, Thailand, the Malay Peninsula, Sumatra, Java,

Borneo, Sulawesi, the Philippines and Okinawa (intro-

duced)] (Indochina-Sunda-Laos Clade: 100%, 100%, 1.00).

5

Fig. 2 Bayesian tree using 2005 bp of the

mitochondrial 12S rRNA, tRNAval and

16S rRNA genes for the Polypedates

samples included in this study. The

sample numbers correspond to those

given in Fig. 1 and Table S1. The

numbers above the branches are the

bootstrap support for the MP and

maximum likelihood (ML) inferences and

the Bayesian posterior probability (MP-

BS ⁄ ML-BS ⁄ BPP). Asterisks indicate

nodes with MP-BS and ML-BS ‡ 70%

and BPP ‡ 0.95.


Of these, the Indochina-Sunda-Laos Clade was split fur-

ther into a wide-ranging Indochina-Sunda Clade (99%,

100%, 1.00) and a Laos Clade (100%, 100%, 1.00). The

Laos Clade consisted solely of some samples from Laos,

while the Indochina-Sunda Clade consisted of Indochina

and Sunda Clades. The Indochina Clade (95%, 98%,

1.00) is found in southern China (Guangxi, Hong Kong

and Hainan), Vietnam (Vinh Phu, Thua Thien, Hue and

Kon Tun), Laos (Vientiane) and Thailand exclusive of

southern regions, while the Sunda Clade (73%, 96%,

1.00) occurs in Myanmar, southern Thailand, the Malay

Peninsula, Sundaland, Sulawesi, the Philippines and

Okinawa.

The sequence divergences (uncorrected p-distance in

16S rRNA) among these genetic groups are shown in

Table S5. The divergences between P. macrotis and the

four major clades of the P. leucomystax complex were large

6 ª 2

(8.2–10.8%), as were those between Polypedates sp. from

the Malay Peninsula (Malay Clade) and the remaining

three clades (7.8–10.2%). The divergences between the

South and North China Clades were moderate (4.3–

6.4%), but those between the Indochina-Sunda and Laos

Clades (3.3–5.1%), and between the Indochina and Sunda

Clades were lower (2.2–4.3%).

The Bayesian tree based on concatenated nuDNA genes

(Fig. 3) inferred relationships essentially similar to those

obtained from the mtDNA analyses, except for the order

of divergence in the South-East Asian P. leucomystax com-

plex. Unlike the relationships recovered with mtDNA, Po-

lypedates sp. from the Malay Peninsula (mtDNA Malay

Clade) diverged first, followed by P. macrotis. The mono-

phyly of all clades had significant support (BPP ‡ 0.96),

except for the sister group relationship of the mitochon-

drial Indochina and Sunda Clades (0.63).


Fig. 3 Bayesian trees using a 2005-bp

sequence of the mitochondrial 12S

rRNA, tRNAval and 16S rRNA genes

(left) and a 4696-bp sequence of the

nuclear brain-derived neurotrophic factor

(BDNF), sodium ⁄ calcium exchanger 1

(NCX), POMC, Rag1, Rhod, SCF and

Tyr genes (right) for selected samples of

Polypedates species. The sample numbers

correspond to those given in Fig. 1 and

Table S1. The numbers above the

branches are the bootstrap support for

the MP and maximum likelihood (ML)

inferences and the Bayesian posterior

probability (MP-BS ⁄ ML-BS ⁄ BPP) in the

mtDNA tree (left) and the BPP in the

nuDNA tree (right).


We detected six variable sites from 508-bp fragments of

BDNF and 19 sites from 333 bp of Rag1, in five selected

samples from the Indochina Clade and six from the Sunda

Clade. From these fragments, we detected six haplotypes in

BDNF and nine in Rag1. The geographical distributions of

the haplotypes are shown in Table S6 and Fig. 4. The

mitochondrial Indochina and Sunda Clades did not have

identical haplotype compositions, although one individual

from the Indochina Clade from the base of the Malay Pen-

insula, southern Thailand (Sample 42), possessed a haplo-

type predominant in the Sunda Clade (BDNF-4).

Moreover, this individual shared a unique haplotype (Rag1-

5) with an individual from the Indochina Clade from near

the Isthmus of Kra, southern Thailand (Sample 48).

Divergence time

In all data sets, the divergence time estimations revealed

large degrees of overlap in the confidence intervals sur-

rounding the estimated values between many lineages, and

this precludes a confident conclusion regarding the timing

of colonization of the present areas of distribution. The

estimated divergence times of the two mtDNA data sets

(#1 and #2) were close to each other, while those of the

nuDNA (data set #3) were generally much younger.


Our data suggest that Polypedates diverged from Rhacopho-

rus between 42.5 (CI 28.6–56.4; data set #2) and 26.6 (CI

14.0–40.4; data set #3) MYBP. From the analysis of data set

#2, the first divergence within Polypedates was estimated to

have occurred 32.4 (CI 19.9–44.5) MYBP, when two major

clades split: the primarily South Asian Clade leading to the

group of Sri Lankan P. fastigo and P. eques, and the predomi-

nantly South-East Asian Clade, consisting of all of the other

lineages including the secondarily South Asian Clades. The

most recent common ancestor (MRCA) of the primarily

South Asian Clade is estimated at 6.1 (CI 1.3–12.2) MYBP,

while the secondarily South Asian Clade, P. cruciger and

P. maculatus, split earlier at 12.2 (CI 4.2–20.5) MYBP.

The divergence of the predominantly South-East Asian

Clade began with the separation of P. otilophus from the

remaining clades, including the South-East Asian and sec-

ondarily South Asian Clades, between 26.2 (CI 15.7–37.1;

data set #2) and 18.1 (CI 9.0–28.1; data set #3) MYBP. Sub-

sequent divergence occurred when P. colletti split from the

remaining clades between 18.4 (CI 10.6–26.6; data set #2)

and 11.4 (CI 5.4–17.8; data set #3) MYBP, and the second-

arily South Asian Clade (P. maculatus) split from the

remaining South-East Asian Clades between 18.3 (CI 10.6–

22.6; data set #2) and 8.4 (CI 3.9–13.0; data set #3) MYBP.

7

Fig. 4 Distributions of brain-derived neurotrophic factor (BDNF)

and Rag1 haplotypes for the mitochondrial groups of the

Indochina (P. megacephalus) and Sunda (Polypedates leucomystax)

Clades. Sample numbers correspond to those given in Fig. 1 and

Table S1. The white arrow indicates the Isthmus of Kra.


In the remaining South-East Asian Clades, divergence

of P. macrotis from the others began between 14.1 (CI 7.4–

20.9; data set #2) and 4.8 (CI 2.4–7.6; data set #3) MYBP,

that of the Malay Clade from the others began between

12.8 (CI 6.9–19.2; data set #2) and 6.5 (CI 3.1–10.2; data

set #3) MYBP, and the divergence of the Indochina and

Sunda Clades began between 3.7 (CI 1.8–5.6; data set #1)

and 2.6 (CI 1.4–3.9; data set #3) MYBP, following the

split of their common ancestor from the Laos Clade

between 4.7 (CI 2.4–7.3; data set #1) and 2.8 (CI 1.5–4.4;

data set #3) MYBP.

DiscussionPhylogenetic relationships within the genus Polypedates

The genus Polypedates occurs in both South-East Asia and

South Asia (Inger 1999). However, the only the South

Asian member we could study was P. maculatus. To clarify

the patterns of differentiation across the extent of its geo-

8 ª 2

graphical range fully, we constructed a phylogenetic tree

using about 800 bp of 12S and 16S rRNA from 36

sequences representing each genetic group recognized in

this study and six sequences of South Asian taxa from

GenBank, in addition to the three sequences used for esti-

mating divergence time (P. maculatus: AF215184 and

AF215358, AY880607 and AY880518; P. cruciger:

AY141799 and AY141845; P. eques: AY141801 and

AY141847, AY880469 and AY920531; P. fastigo:

AY880604 and AY880518). As in several other studies (e.g.

Delorme et al. 2005; Frost et al. 2006; Li et al. 2008; Yu

et al. 2009), we recovered a monophyletic Polypedates. The

genus Polypedates was once treated as a part of Rhacophorus

(see Matsui & Wu 1994; Matsui & Panha 2006), but its

distinct generic status is now beyond doubt.

Our analysis suggests a primary division between a clade

including Sri Lankan P. eques and P. fastigo, and another

clade encompassing the remaining taxa from South-East

and South Asia. Within the latter clade, the phylogenetic

relationships among taxa did not differ from those

obtained in the analyses of a much larger mitochondrial

data set, except that Sri Lankan P. cruciger formed a group

with P. maculatus. From mtDNA and morphological evi-

dence, Delorme et al. (2005) recognized the P. eques group

(P. eques and P. fastigo) and the P. leucomystax group (all

other species) within Polypedates. Indeed, Meegaskumbura

et al. (2010) recently established a new genus, Taruga, to

accommodate the P. eques group of Delorme et al. (2005),

and our results support this division. These patterns sug-

gest that the common ancestor of the genus Taruga

diverged from the common ancestor of the remaining lin-

eages of Polypedates, probably in southernmost South Asia,

and that subsequent differentiation proceeded in more

northern latitudes.

In the course of this latter differentiation, the diver-

gence of South and South-East Asian lineages does not

seem to have been the first event. Rather, the divergence

of South-East Asian P. otilophus and P. colletti preceded the

divergence of South Asian P. maculatus and P. cruciger

from the mainly South-East Asian P. leucomystax complex.

Consequently, the group of P. maculatus and P. cruciger

likely invaded South Asia secondarily. As other species

exist that are apparently related to the P. leucomystax com-

plex in South Asia (e.g. P. zed and P. teraiensis; Dubois

1987), evolution of the genus in this region can be prop-

erly discussed only after sufficiently studying such taxa.

Differentiation of the genus Polypedates within South-East

Asia

Among South-East Asian Polypedates, P. otilophus differs

greatly in external morphology from the others (Inger

1966) and has never been confused taxonomically with



them. In this study, that species proved to be markedly

different genetically from the other species and showed

basal divergence in both the mtDNA and nuDNA trees.

Conversely, P. colletti was once confused with P. leucomystax

(Wolf 1936), but then proved to be clearly differentiated

from P. leucomystax by external morphology (Inger 1966)

and overlaps the latter species in distribution (Inger &

Stuebing 1997). In both the mtDNA and nuDNA trees,

P. colletti showed basal divergence from the P. leucomystax

complex, indicating their remote genetic relationships.

Although P. macrotis has previously been treated as a

subspecies of P. leucomystax (Wolf 1936; Inger 1954), sub-

sequent studies demonstrated that it was distinct morpho-

logically (Inger 1966) and distributed sympatrically with

that taxon (Inger 1966; Inger & Stuebing 1997). The

genetic divergences between P. macrotis and the other six

clades of the P. leucomystax complex were as high as 8.7–

11.6%, and the species is also regarded as a valid species

genetically. Our mtDNA analyses failed to resolve the

phylogenetic relationships between P. macrotis and various

genetic groups within the P. leucomystax complex. Previ-

ously, Brown et al. (2010) demonstrated a close relation-

ship between P. macrotis and the P. leucomystax complex. In

our analyses, when the Polypedates sp. from the Malay Pen-

insula (Malay Clade, which was not included in Brown

et al. 2010) is removed, our Bayesian tree analyses suggest

the basal divergence of P. macrotis from the remaining spe-

cies. In addition, in our Bayesian analysis of nuDNA

sequences, P. macrotis was inferred to be sister to the P. leu-

comystax complex.

Phylogenetic relationships within the P. leucomystax

complex and the taxonomic assignment of each genetic

group

When the phylogenetic trees from 800 bp of mitochon-

drial 16S rRNA are compared between Brown et al. (2010)

and our study, Clades 1, 2, 3 plus 4, and 5, and the clade

of the remaining haplotypes in Brown et al. (2010) obvi-

ously corresponded to our South China Clade, Laos

Clade, Indochina Clade, North China Clade and Sunda

Clade, respectively, and their relationships were essentially

identical. The phylogenetic relationships estimated using

mtDNA and nuDNA, genetic divergences in 16S rRNA

and distribution pattern strongly supported the sympatric

occurrence of more than one taxon of the P. leucomystax

complex, and the complex is thought to encompass at least

the following four distinct taxa, in addition to P. macrotis:

South China Clade. This clade occurs in southern China

(Guangxi and Hainan) and Vietnam and is supposedly not

conspecific with the North China (P. braueri) or Indo-

china-Sunda-Laos Clades because they are sympatric. We


also consider the South China Clade to be heterospecific

with the allopatric Malay Clade because of the large

genetic divergence between these putative species. Unfor-

tunately, we could not examine male specimens of the

South China Clade. However, based on its distribution,

we consider it likely to be one of the two types of P. leuco-

mystax reported from Vietnam (Inger 1999; Ziegler 2002)

that are differentiated by the presence or absence of male

vocal sac openings.

Unfortunately, we could not examine the sequences cor-

responding to the South China Clade samples used for the

DNA analyses. However, the males of the Indochina

Clade possess vocal sac openings and are thought to be

the non-striped type of P. leucomystax of Inger et al. (1999)

or P. leucomystax of Ziegler (2002). Therefore, the South

China Clade seems to be another striped type of P. leuco-

mystax of Inger et al. (1999) or Polypedates sp. of Ziegler

(2002). Indeed, comparisons of our data with the GenBank

data on Ziegler’s Polypedates sp. (AF285221–4) revealed

that they are nearly identical. Similarly, the sequence of

P. leucomystax from Ha Tinh, Vietnam (DQ283048; Frost

et al. 2006), was identical to our South China Clade, indi-

cating an invalid original identification.

Ziegler (2002) did not refer to the presence or absence of

vocal openings in his Polypedates sp., but he later identified it

as P. mutus (Hendrix et al. 2008), suggesting that he inferred

the absence of vocal sac openings in this population. This

estimation has been confirmed by an examination of the

specimens used (T. Nguyen and T. Ziegler, personal com-

munication on 18 October 2010). Inger et al. (1999) also

noted that males of their ‘striped type’ lack vocal openings,

suggesting the absence of these structures in the South

China Clade. In the genus Polypedates, a lack of vocal open-

ings is known only in P. mutus from Myanmar (Smith

1940), but this name is not applied directly to the South

China Clade because the Laos Clade also lacks openings.

We believe that the South China and Indochina-Sunda-

Laos Clades are not conspecific, but must await application

of the name until P. mutus from the type locality can be

studied. Our sample from Hainan Island differed slightly

from samples from Guangxi and Vietnam. The morphology

of this population must also be studied in the future.

North China Clade. This clade occurs in Taiwan and

widely in continental China (Anhui, Zhejiang, Jiangxi,

Sichuan, Guangxi and Yunnan), but is absent from Hong

Kong. The Taiwan population was once assigned to

P. megacephalus, which was originally described from Hong

Kong (e.g. Matsui et al. 1986; Zhao & Adler 1993). How-

ever, the populations from Taiwan and Hong Kong clearly

differ in their morphometric and mating call characteris-

tics and are highly divergent genetically, as shown here,

9


with high levels of divergence between the North China

and Indochina-Sunda-Laos Clades (the latter contains

samples from Hong Kong). Therefore, the Taiwan popu-

lation has been classified as a distinct species: P. braueri

(Kuraishi et al. 2011). Our results indicate that the North

China and Indochina-Sunda-Laos Clades likely overlap in

distribution within Guangxi, supporting their heterospeci-

fic relationships.

Pope (1931) considered the population of the P. leucomys-

tax complex from Fukien (=Fujian) to be P. leucomystax

megacephalus. Unfortunately, we could not examine sam-

ples from Fujian, but based on distribution, we consider it

likely that this lineage be assigned to P. braueri. Further-

more, Pope noted that skin in the Fujian specimens was

free from the skull, a trait found in the Taiwan P. braueri

samples, but not in the Hong Kong P. megacephalus sam-

ples (Kuraishi et al. 2011).

However, the calls noted by Pope (1931) seem to differ

from those of topotypic P. braueri from Taiwan (Kuramo-

to 1986; Matsui et al. 1986). Frogs of the P. leucomystax

complex within this wide range exhibit variation in their

acoustic and karyological characteristics (see Matsui et al.

1986), and taxonomic identification of the mainland Chi-

nese populations still requires caution. Detailed future

studies on adult and larval morphology and bioacoustic

characteristics might reveal the presence of cryptic taxa

within the populations presently assigned to P. braueri.

Yang & Rao (2008) described P. impresus and P. spinus

from Yunnan, southern China. Of these, P. spinus should

be placed in Rhacophorus because of its green body (see

Matsui & Panha 2006), while P. impresus surely represents

a member of the P. leucomystax complex. However, as

P. impresus is known only from the type locality and no

genetic information is available, we cannot discuss the

exact status of this species. The species was described

without comparison with other Polypedates species, render-

ing its taxonomic identification difficult. In our results, a

single sample of the North China Clade from Yunnan dif-

fered from the other samples with relatively high diver-

gences (2.0–2.7%). These values overlap those observed

between the heterospecific Indochina and Sunda Clades

(2.2–4.3%, see below). Unfortunately, our specimen is a

juvenile and lacks taxonomically important features,

including the distinct postero-cranial depression and white

lip that are said to be unique to P. impresus (Yang & Rao

2008). Clarifying the relationships of our Yunnan sample

and P. impresus would be highly desirable.

Indochina-Sunda-Laos Clade. The Indochina-Sunda-Laos

Clade contained samples that occurred widely in South-

East Asia from the Sunda region in the south to southern

China (Hong Kong, Guangxi and Hainan) in the north

10 ª 2

and Myanmar in the west. The clade was genetically dis-

tinct from the others and contained independent Indo-

china-Sunda and Laos Clades. The Laos Clade, whose

males lack vocal sac openings, is obviously not conspecific

with the Indochina-Sunda Clade, in which males possess

the openings. The Indochina-Sunda Clade is further split

into allopatric Indochina and Sunda Clades, occurring

north and south of the Isthmus of Kra in southern Thai-

land, respectively. Exceptionally, the sample from Myan-

mar was included in the Sunda Clade, despite its

geographical proximity to the Indochina Clade. This one

aberrant sample may indicate the presence of gene flow

along the west coast of the Malay Peninsula.

The Indochina Clade included samples from Hong

Kong, which is the type locality of P. megacephalus, while

the Sunda Clade contained samples from Java, the type

locality of P. leucomystax. Polypedates leucomystax and

P. megacephalus have long been distinguished chiefly on

morphological grounds (Pope 1931; Inger 1966; Matsui

et al. 1986), but these previous reports should be reas-

sessed carefully because specific identification of the popu-

lations, without consideration of P. braueri, is not always

correct. Both of them are variable morphologically and are

actually very similar to each other. Furthermore, exact

comparisons using topotypic specimens have rarely been

made (Kuraishi et al. 2011). However, they certainly differ

in their acoustic characteristics. Calls reported for males

assigned to the Indochina Clade are variable (Heyer 1971;

Trepanier et al. 1999; Ziegler 2002; Sheridan et al. 2010),

but never identical to that of the Sunda Clade. Although

the call of the Sunda Clade also shows slight variation, it

is essentially similar in different populations (e.g. Matsui

1982; Brzoska et al. 1986; Narins et al. 1998; Sheridan

et al. 2010).

Taylor (1962) recognized Rhacophorus (=Polypedates) l. leu-

comystax throughout Thailand and Rh. l. sexvirgatus from

the southern part of the country. Simultaneously, he sug-

gested that these subspecies differ only in colouration and

that Rh. l. sexvirgatus may be a colour variety of Rh. l. leuco-

mystax. According to our results, Taylor’s (1962) Rh. l. leu-

comystax and Rh. l. sexvirgatus should be treated as

P. megacephalus (Indochina Clade) and P. leucomystax (Sun-

da Clade), respectively. The Hong Kong samples of

P. megacephalus formed two mitochondrial haplotype

groups. Furthermore, as noted above, variation appears to

exist in the acoustic characteristics among populations of

the Indochina Clade. At present, however, collectively call-

ing this group P. megacephalus would be best.

Like P. megacephalus from Hong Kong, the P. leucomystax

samples from Java in the Sunda Clade were split into three

geographical mitochondrial phylogroups (western, central

and central ⁄ eastern groups), but these had low genetic



diversities (0.0–2.1%). Moreover, because details of the

type locality within Java are difficult to trace and no infor-

mation on their variation other than genetic features

exists, the Sunda Clade is best called P. leucomystax. Brown

et al. (2010) briefly reviewed old names that possibly

applied to their clades, which correspond to our Sunda

Clade, and rejected splitting P. leucomystax taxonomically.

As they pointed out, no basis exists for splitting the species

at present, and further studies based on characters other

than genetic data will be necessary to arrive at a possible

future taxonomic revision.

One sample from the Indochina Clade (Sample 42)

shared nuclear BDNF and Rag1 haplotypes with one sam-

ple from the Sunda Clade (Sample 48; Table S6). This

suggests incomplete reproductive isolation of these two

clades at the boundary of their distributions and may be

interpreted in three possible ways: (i) introgression of

mtDNA genes from the Indochina Clade to the Sunda

Clade, (ii) introgression of nuDNA genes from the Sunda

Clade to the Indochina Clade and (iii) deep coalescence or

incomplete lineage sorting (Maddison 1997). Of these

three, the first and second hypotheses do not seem to be

supported by the distributions of the BDNF and Rag1

haplotypes. The BDNF-4 haplotype occurs widely in the

range of the Sunda Clade, and the first hypothesis requires

the occurrence of the Sunda Clade at the base of the

Malay Peninsula (near the locality of Sample 42), where

introgression of mtDNA genes from the Indochina Clade

occurred. However, despite intensive sampling efforts, we

failed to collect samples of the P. leucomystax complex in

the area between there and south of the Isthmus of Kra

(near the locality of Sample 48), and this hypothesis seems

implausible. From the nature of the transmission of nuD-

NA genes (e.g. Ballard & Whitlock 2004), the second

hypothesis also seems implausible. Therefore, we prefer

the deep coalescence hypothesis. Presumably, the Indo-

china and Sunda Clades split relatively recently and have

maintained ancient polymorphisms. From these lines of

evidence, the Indochina (P. megacephalus) and Sunda (P. leu-

comystax) Clades are again judged to represent different

species, although the degree of genetic divergence between

them (3.0%) is at the border of the proposed intraspecific

threshold for frogs (around 3% in 16S rRNA; Fouquet

et al. 2007).

The Laos Clade occurs in Laos, overlapping part of the

range of P. megacephalus (Indochina Clade). Our Laos

Clade corresponds to Clade 2 of Brown et al. (2010),

which contains samples from Guangxi in southern China.

Therefore, the sympatry of the Laos Clade and P. mega-

cephalus (Indochina Clade) does not seem to be limited to

Laos. Males of the Laos Clade match P. mutus in their

lack of vocal openings. Although the South China Clade


also lacks openings, as stated above, the Laos and South

China Clades are clearly different phylogenetically and are

undoubtedly heterospecific. As the sampling sites for the

Laos Clade are much nearer to Myanmar, where P. mutus

was described (Smith 1940), it seems more probable that

the Laos Clade represents P. mutus, although it is also

probable that neither the Laos Clade nor the South China

Clade actually corresponds to the true P. mutus. We tenta-

tively call the South China Clade Polypedates cf. mutus 1

and the Laos Clade Polypedates cf. mutus 2 to allude to

their lack of vocal openings in males.

Malay Clade. The Malay Clade (Polypedates sp.), which

occurs in the Malay Peninsula, first diverged from the

other genetic groups of the P. leucomystax complex in the

mtDNA tree and even faster than P. macrotis in the nuD-

NA tree. It is not conspecific with the South or North

China Clades because of the large genetic divergence, or

with the Sunda Clade because of the sympatric ⁄ syntopic

distribution. Narins et al. (1998) reported the syntopic

occurrence of two morphs (morphs A and B) of the P. leuco-

mystax complex in the Malay Peninsula that differ in mor-

phology, calls and allozyme characteristics. Based on

morphological and acoustic characteristics (M. Matsui, N.

Kuraishi & H. S. Yong, unpublished data), the Malay

Clade is thought to correspond to morph B of Narins et al.

(1998). Those authors found morph A in the range of our

Sunda Clade and reported that calls of morph A were simi-

lar to those of Philippine populations of P. leucomystax. As

our Sunda Clade occurs widely from the Malay Peninsula

to the Philippines, morph A of Narins et al. (1998) is most

probably identical to our Sunda Clade (P. leucomystax).

Evolutionary history

The evolutionary histories of various anuran lineages have

been studied in South-East Asia, including those of Eu-

phlyctis and Hoplobatrachus (Alam et al. 2008), Ansonia (Mat-

sui et al. 2010a), Leptobrachium (Matsui et al. 2010b) and

Paa (Che et al. 2010). However, only a few Polypedates have

been included as a part of large-scale divergence date esti-

mation of anuran lineages (e.g. Roelants et al. 2007; Wiens

et al. 2009), and few studies specifically estimate the age of

the genus (Brown et al. 2010).

Few stratigraphic events can be associated with the

divergence of the genetic groups recognized in Polypedates.

Therefore, to estimate divergence dates, we used the

divergence between Mantellidae and Rhacophoridae 73.1

MYBP (Bossuyt & Milinkovitch 2000) and between Buer-

geria and other Rhacophoridae 49.7 MYBP (Roelants et al.

2007) as external calibration points, and the divergence

between the Java and Sumatra populations of P. leucomystax

2.8 MYBP as an internal calibration point. This last

11


calibration point was set following the divergence time

between the Java and Sumatra populations of L. hasseltii

(Matsui et al. 2010b). As Leptobrachium is a lotic breeder

and has an evolutionary history much different from that

of Polypedates, this calibration may be inappropriate and

should be treated with great caution, but no internal cali-

bration point is presently available within Polypedates.

Using mtDNA, the divergence between Polypedates and

Rhacophorus was estimated to have occurred in the late

Eocene (42.5 MYBP; data set #2 in Table S7), which is

similar to the date in Wiens et al. (2009), while that esti-

mated using nuDNA was much later in the late Oligocene

(26.6 MYBP; data set #3). All members of Polypedates

whose breeding habits are known invariably lay a foamy

egg nest on or near still water (Taylor 1962; Inger 1966;

Alcala 1986; Dutta 1997; Iskandar 1998; Maeda & Matsui

1999), a reproductive trait shared by many species of Rhac-

ophorus and Chiromantis. Therefore, formation of a foam

nest was likely acquired before the separation of Polypedates

and Rhacophorus. Some Rhacophorus species (e.g. Rh. anguli-

rostris, Rh. cyanopunctatus and Rh. gauni; Malkmus et al.

2002) have lotic breeding habits, while Polypedates lacks

this and is thought to be more conservative than Rhacopho-

rus with regard to breeding.

Our data suggest at least two dispersal events within

the South Asian Polypedates. Separation of the primarily

South Asian Clade and predominantly South-East Asian

Clade, including the secondarily South Asian Clade, was

Fig. 5 Estimated times million years before present (MYBP) of the m

those shown in Table S7. Dotted lines indicate divergences estimated o

12 ª 2

the first event in Polypedates evolution. Our estimated

mean date for this divergence was the early Oligocene

[32.4 (19.9–44.5) MYBP; data set #2; mtDNA]. Bocxlaer

et al. (2009) reported that some bufonids endemic to

Western Ghats in India and Sri Lanka arose as a result of

the ‘into India’ movement of Laurasian species after the

Indo-Asia collision. This also seems to be the case in our

primarily South Asian Clade of Polypedates, which also

occurs from Western Ghats to Sri Lanka. However,

Bocxlaer et al. (2009) estimated that specialized endemic

bufonids such as Ghatophryne (formerly part of Ansonia)

and Pedostibes on the Indian subcontinent arose

22.0–22.4 MYBP, after separation from South-East Asian

lineages 23.5–24.2 MYBP. These estimates are much

younger than our estimate from short mtDNA sequences

(data set #2). Unfortunately, we have no nuDNA data,

but extrapolation of the estimated dates in data set #3 for

events that occurred before and after this event (separa-

tion of Rhacophorus and Polypedates 26.6 MYBP and the

divergence of P. otilophus from the remaining clades 18.1

MYBP) yielded 21.3 MYBP for the divergence of the

primarily South Asian Clade of Polypedates (Fig. 5), which

agrees well with estimates for bufonids by Bocxlaer et al.

(2009). These events seem to correspond to putative geo-

logical movement in southern Tibet and the Himalayan

region (40–19 MYBP; Che et al. 2010; Matsui et al.

2010b), which also induced drastic climate change in the

region (An et al. 2001; Harris 2006).

ain divergences of Polypedates. The node numbers correspond to

nly from mtDNA and not from nuDNA data sets.



The secondarily South Asian Clade should have sepa-

rated from the South-East Asian Clade much later, and

this was estimated to have occurred in the early Miocene

(18.3 MYBP) using data set #2 or the late Miocene (8.4

MYBP) using data set #3 (Fig. 5). As shown above, the

estimates from nuDNA (data set #3) seem to be more rea-

sonable than those from mtDNA (data sets #1 and #2) and

are consistent with the following scenarios. Even so, note

that the MRCA estimated from data set #2 was much

younger in the primary Clade (6.1 MYBP) than in the sec-

ondary Clade (12.2 MYBP), although the lack of other

South Asian samples prohibits further discussion.

Within the predominantly South-East Asian Clade,

P. otilophus first diverged in the early Miocene (18.1

MYBP; data set #3). This species occurs on Sumatra, Bor-

neo (Inger 1966) and Java (Riyanto et al. 2009), but is not

recorded from Peninsular Malaysia (Berry 1975). This sug-

gests the rapid dispersal of ancestral lineages and a present

relict distribution of P. otilophus. Later in the late Miocene,

P. colletti diverged from the remaining predominantly

South-East Asian Clade (5.4–17.8 MYBP), followed by the

divergence of the secondarily South Asian Clade (3.9–

13.0 MYBP, see above). Unlike P. otilophus, P. colletti

occurs in Borneo, Sumatra, Peninsular Malaysia and Thai-

land (Taylor 1962; Inger 1966; Berry 1975), but not in

Java (Iskandar 1998). Also, the species is also reported

from Vietnam (Orlov et al. 2002). This wide distribution

contrasts that of P. otilophus, but the known continental

localities are fragmented, suggesting a relict distribution

like P. otilophus. This wide distribution also suggests the

presence of intraspecific genetic divergence in P. colletti,

which is currently under study.

Although the order of divergence is ambiguous, Polype-

dates sp. from the Malay Peninsula (Malay Clade) and

P. macrotis diverged from the late Miocene to the early

Pliocene, earlier than the remaining members of the

P. leucomystax complex, which diverged nearly simulta-

neously in the Pliocene in wider regions of East and

South-East Asia. Polypedates macrotis occurs more widely

than P. colletti, being recorded from the Philippines, Bor-

neo, Sumatra, Peninsular Malaysia and Thailand, but not

from Java (Iskandar 1998). Although the species habitat

tends to be less widely disturbed than that of P. leucomystax

(see below), its occurrence in the Philippines may be

related to recent human activities. Future dense sampling

is required to test this hypothesis. Polypedates sp. from the

Malay Peninsula seems to have had a long history, at least

since the late Miocene, independent from P. leucomystax

with which it has long been confused (Narins et al. 1998).

This divergence is not very deep compared with the other

frog groups endemic to the Peninsula (e.g. Ansonia spp.;

Matsui et al. 2010a), but is regarded as sufficiently long


among members of the P. leucomystax complex. The spe-

cies is currently known only from central to southern Pen-

insular Malaysia and is less common than P. leucomystax

where they occur together. This restricted distribution and

the reduced intraspecific genetic divergence in this popula-

tion suggest that the species experienced past demographi-

cal or population reduction (e.g. Matsui et al. 2008).

From the current distribution, the ancestral lineage of

the remaining P. leucomystax complex seems to have occu-

pied much wider ranges in South-East Asia than Polype-

dates sp. from the Malay Peninsula or P. macrotis. This

ancestral lineage began to diverge in the northern regions

around 3.9 MYBP, which led to Polypedates cf. mutus 1.

Then, the ancestral lineage giving rise to P. braueri from

the northern regions split around 3.8 MYBP from the

common ancestor leading to Polypedates cf. mutus 2, P. leu-

comystax and P. megacephalus, which occupied more south-

ern regions. The absence of Polypedates cf. mutus 1 and

P. braueri in the southern Sundaland, which is thought to

have long been connected with the region including Indo-

china, suggests past ecological competition between these

clades and the clade ancestral to the remaining P. leucomys-

tax complex.

The divergence of Polypedates cf. mutus 2, which lacks

vocal sac openings, and the common ancestor of P. leuco-

mystax and P. megacephalus, both with vocal openings,

occurred in the southern region in the late Pliocene (1.5–

4.4 MYBP). The common ancestry of groups with and

without vocal openings indicates the importance of acous-

tic differentiation in speciation. Differentiation in vocaliza-

tion and acoustic characters may have facilitated rapid

diversification in the P. leucomystax complex.

The most recent divergence in the southern group

(P. leucomystax and P. megacephalus) occurred slightly later

in the late Pliocene or early Pleistocene (1.4–4.0 MYBP).

Their current allopatric distribution on both sides of the

Isthmus of Kra suggests that speciation was affected by a

short disjunction of the Malay Peninsula and Sunda

Islands by the South China Sea in the early Pliocene

(about 5 MYBP; Inger & Voris 2001). Although the evolu-

tionary history of P. megacephalus is short, the species

already had a wide range between the ranges of the more

northern P. braueri and more southern P. leucomystax.

Most of the intraspecific divergence in P. leucomystax

occurred very late, but divergence between Myanmar and

other samples seems to have occurred much earlier in the

evolution of this species, suggesting a relict status of the

Myanmar population. Furthermore, at least two groups

were recognized in the samples from Java, although the

degree of divergence is low. Notwithstanding such varia-

tion, a wide range of distribution, but low genetic diver-

gence, characterizes P. leucomystax. These traits seem to be

13


related not only to very recent dispersal via land connec-

tions, such as is known to have occurred during the Last

Glacial Maximum (0.021 MYBP; Sathiamurthy & Voris

2006), but also to human activities.

For example, some samples from Sabah and Sarawak,

Borneo, had unique haplotypes, while others shared identi-

cal haplotypes with samples from the Malay Peninsula. In

contrast, the samples from Kalimantan, Borneo, had hapl-

otypes common to samples from central and southern

Sumatra. Inger & Voris (2001) suggested that P. leucomystax

now found in Borneo was introduced through human

activity in the last few thousand years. Frogs of the P. leuco-

mystax complex are notorious for being frequently trans-

ported artificially (e.g. Brown & Alcala 1970; Wiles 2000;

Kuraishi et al. 2009; Brown et al. 2010). Most probably,

P. leucomystax now found in Sabah and Sarawak is a mix-

ture of populations with two different origins. One has

origins at an older time, probably in the Pleistocene, and

has diversified within the island. The other has younger

origins and was transported from the Malay Peninsula via

human-mediated activity. In contrast, frogs found in Kali-

mantan seem to include descendents originally introduced

artificially from Sumatra (and vice versa).

The population genetic structure in commensal frogs

like P. leucomystax, P. megacephalus, Hylarana erythraea and

H. nicobariensis (Inger 1966) should be studied carefully; in

particular, frogs of the P. leucomystax complex require cau-

tion because these frogs, with strongly adhesive digital

disks and a high ability to resist desiccation, are more eas-

ily transported unintentionally than other anurans.

AcknowledgementsWe are grateful to the following for their encouragement,

permission to conduct research or companionship in the

field: L. Apin, K. Araya, Md. S. Azman, A. Bogadek, B.

Bounkhoun, S.-L. Chen, the late A.-A. Hamid, T. Hikida,

A. Iizuka, S. Iwanaga, T. Kusano, D. Labang, M.B.

Lakim, M.W. Lau, A.V. Le, J.-T. Lin, P. Lin, M. Maryati,

K. Mizuno, the late J. Nabitabhata, H. Nagaoka, Le.V.

Nguyen, Lu.V. Nguyen, T. Niizato, T. Shimada, T. Suga-

hara, T. Tachi, the late I. Takiguchi, U. Tanaka, M.

Toda, A. Tominaga, the late T. Utsunomiya, H. Wakaha-

ra, C.-H. Wang, G.-F. Wu and M. Yoshizumi. For tissue

samples, we are also indebted to M. Hori, M. Kato, N.

Orlov, T. Papenfuss (for the MVZ tissue collection) and

G.R. Zug. T. Nguyen and T. Ziegler kindly provided

information on Vietnamese specimens. The Economic-

Planning Unit (formerly the Socio-Economic Research

Unit) of Malaysia, the State Government of Sarawak,

Sabah Parks, The National Research Council of Thailand

and the Royal Forest Department of Thailand kindly

allowed MM to conduct the project, and the University

14 ª 2

Malaya, Universiti Malaysia Sabah, Universiti Kebangsaan

Malaysia (UKM), JICA, the Forest Department, Sarawak,

Center for Agriculture and Forestry Research and Devel-

opment of Hue University of Agriculture and Forestry,

and Chulalongkorn University kindly provided facilities

for conducting research. Field trips by MM were made

possible by grants from The Monbusho International Sci-

entific Research Program (Field Research, 01041051,

02041051, 04041068, 06041066, 08041144), The Mon-

bukagakusho through the Japanese Society for the Promo-

tion of Sciences (JSPS: Field Research, 10041166,

15370038, 20405013, 23405014), UKM (OUP-PLW-14-

59 ⁄ 2008) and TJTTP-OECF. AH thanks M.D. Kusrini, J.

McGuire and M.I. Setiadi for providing tissue samples and

the Monbukagakusho for scholarship funding. Research

export permits were obtained from the Forest Department,

Sarawak, and the Chinese Academy of Sciences.

ReferencesAlam, M. S., Igawa, T., Khan, M. R., Islam, M. M., Kuramoto,

M., Matsui, M., Kurabayashi, A. & Sumida, M. (2008). Genetic

divergence and evolutionary relationships in six species of

genera Hoplobatrachus and Euphlyctis (Amphibia: Anura) from

Bangladesh and other Asian countries revealed by

mitochondrial gene sequences. Molecular Phylogenetics and

Evolution, 48, 515–527.

Alcala, A. C. (1986). Guide to Philippine Flora and Fauna 10,

Amphibians and Reptiles. Quezon City: J. M. C. Press.

An, Z.-S., Kutzbach, J. E., Prell, W. L. & Porter, S. C. (2001).

Evolution of Asian monsoons and phased uplift of the

Himalaya-Tibet plateau since late Miocene times. Nature, 411,

62–66.

Ballard, J. W. O. & Whitlock, M. C. (2004). The incomplete

natural history of mitochondria. Molecular Ecology, 13, 729–744.

Berry, P. Y. (1975). The Amphibian Fauna of Peninsular Malaysia.

Kuala Lumpur: Tropical Press.

Bocxlaer, I. V., Biju, S. D., Loader, S. P. & Bossuyt, F. (2009).

Toad radiation reveals into-India dispersal as a source of

endemism in the Western Ghats-Sri Lanka biodiversity hotspot.

BMC Evolutionary Biology, 9, 131.

Bossuyt, F. & Milinkovitch, M. C. (2000). Convergent adaptive

radiations in Madagascan and Asian ranid frogs reveal

covariation between larval and adult traits. Proceedings of theNational Academy of Sciences of the United States of America, 97,

6585–6590.

Brown, W. C. & Alcala, A. C. (1970). The zoogeography of the

Philippine Islands, a fringing archipelago. Proceedings of theCalifornia Academy of Sciences, 38, 105–130.

Brown, R. M., Linkem, C. W., Siler, C. D., Sukumaran, J.,

Esselstyn, J. A., Diesmos, A. C., Iskandar, D. T., Bickford, D.,

Evans, B. J., McGuire, J. A., Grismer, L., Supriatna, J. &

Andayani, N. (2010). Phylogeography and historical

demography of Polypedates leucomystax in the islands of

Indonesia and the Philippines: evidence for recent human-

mediated range expansion? Molecular Phylogenetics and Evolution,

57, 598–619.



Brzoska, J., Joermann, G. & Alcala, A. C. (1986). Structure and

variability of the calls of Polypedates leucomystax (Amphibia:

Rhacophoridae) from Negros, Philippines. Silliman Journal, 33,

87–103.

Castresana, J. (2000). Selection of conserved blocks form multiple

alignment for their use in phylogenetic analysis. Molecular

Biology and Evolution, 17, 540–552.

Che, J., Zhou, W.-W., Hu, J.-S., Yan, F., Papenfuss, T. J., Wake,

D. & Zhang, Y.-P. (2010). Spiny frogs (Paini) illuminate the

history of the Himalayan region and Southeast Asia. Proceedings

of the National Academy of Sciences of the United States of America,

107, 13765–13770.

Delorme, M. (2004). Phylogenie des Ranidae Rhacophorinae:confrontations des analyses moleculaires et morphologiques, et etude de

caracte‘res. Unpublished thesis, Museum National d’Histoire

Naturelle, Paris.

Delorme, M., Dubois, A., Grosjean, S. & Ohler, A. (2005). Une

nouvelle classification generique et subgenerique de la tribu des

Philautini (Amphibia, Anura, Ranidae, Rhacophorinae). BulletinMensuel de la Societe Linneenne de Lyon, 74, 165–171.

Drummond, A. J. & Rambaut, A. (2007). BEAST, BayesianEvolutionary Analysis Sampling Trees, Version 1.4.2. Available via

http://beast.bio.ed.ac.uk/.

Drummond, A. J., Ho, S. Y. W., Phillips, M. J. & Rambaut, A.

(2006). Relaxed phylogenetics and dating with confidence. PLoSBiology, 4, 699–710.

Dubois, A. (1987). Miscellanea taxinomica batrachologica (I).

Alytes, 5, 7–95.

Dutta, S. K. (1997). Amphibians of India and Sri Lanka (Checklist

and Bibliography). Bhubaneswar: Odyssey Publishing House.

Farris, J. S., Kallersjo, M., Kluge, A. G. & Bult, C. (1994).

Testing significance of incongruence. Cladistics, 10, 315–319.

Felsenstein, J. (1985). Confidence limits on phylogenies: an

approach using the bootstrap. Evolution, 39, 783–791.

Fouquet, A., Gilles, A., Vences, M., Marty, C., Blanc, M. &

Gemmell, N. J. (2007). Underestimation of species richness in

Neotropical frogs revealed by mtDNA analyses. PLoS ONE, 2,

e1109.

Frost, D. R., Grant, T., Faivovich, J. N., Bain, R. H., Haas, A.,

Haddad, C. F. B., De Sa, R. O., Channing, A., Wilkinson, M.,

Donnellan, S. C., Raxworthy, C. J., Campbell, J. A., Blotto, B. L.,

Moler, P., Drewes, R. C., Nussbaum, R. A., Lynch, J. D., Green,

D. M. & Wheeler, W. C. (2006). The amphibian tree of life.

Bulletin of the American Museum of Natural History, 297, 1–370.

Hall, T. A. (1999). BioEdit: a user-friendly biological sequence

alignment editor and analysis program for Windows

95 ⁄ 98 ⁄ NT. Nucleic Acids Symposium Series, 41, 95–98.

Harris, N. (2006). The elevation history of the Tibetan Plateau

and its implications for the Asian monsoon. Palaeogeography

Palaeoclimatology Palaeoecology, 241, 4–15.

Hasegawa, M., Kishino, H. & Yano, T. (1985). Dating of the

human-ape splitting by a molecular clock of mitochondrial

DNA. Journal of Molecular Evolution, 22, 160–174.

Hendrix, R., Nguyen, Q. T., Bohme, W. & Ziegler, T. (2008).

New anuran records from Phong Nha – Ke Bang National

Park, Truong Son, Central Vietnam. Herpetology Notes, 1, 23–

31.

Heyer, W. R. (1971). Mating calls of some frogs from Thailand.

Fieldiana: Zoology, 58, 61–82.


Hillis, D. M. & Bull, J. J. (1993). An empirical test of

bootstrapping as a method for assessing confidence in

phylogenetic analysis. Systematic Biology, 42, 182–192.

Hillis, D. M., Mable, B. K., Larson, A., Davis, S. K. & Zimmer,

E. A. (1996). Nucleic acids IV: sequencing and cloning. In D.

M. Hillis, C. Moritz & B. K. Mable (Eds) Molecular Systematics

(pp. 321–378). Sunderland: Sinauer.

Huelsenbeck, J. P. & Rannala, B. (2004). Frequentist properties

of Bayesian posterior probabilities of phylogenetic trees under

simple and complex substitution models. Systematic Biology, 53,

904–913.

Huelsenbeck, J. P. & Ronquist, F. (2001). MRBAYES: Bayesian

inference of phylogeny. Bioinformatics, 17, 754–755.

Huelsenbeck, J. P., Ronquist, F., Nielsen, R. & Bollback, J. P.

(2001). Bayesian inference of phylogeny and its impact on

evolutionary biology. Science, 294, 2310–2314.

Inger, R. F. (1954). Systematics and zoogeography of Philippine

Amphibia. Fieldiana: Zoology, 33, 183–531.

Inger, R. F. (1966). The systematics and zoogeography of the

Amphibia of Borneo. Fieldiana: Zoology, 52, 1–402.

Inger, R. F. (1999). Distribution of amphibians in southern Asia

and adjacent islands. In W. E. Duellman (Ed.) Patterns of

Distribution of Amphibians: A Global Perspective (pp. 445–482).

Baltimore: Johns Hopkins University Press.

Inger, R. F. & Stuebing, R. B. (1997). A Field Guide to the Frogs ofBorneo. Kota Kinabalu: Natural History Publications.

Inger, R. F. & Voris, H. K. (2001). The biogeographical relations

of the frogs and snakes of Sundaland. Journal of Biogeography,

28, 863–891.

Inger, R. F., Orlov, N. & Darevsky, I. (1999). Frogs of Vietnam: a

report on new collections. Fieldiana: Zoology, New Series, 92, 1–46.

Inger, R. F., Stuart, B. L. & Iskandar, D. T. (2009). Systematics

of a widespread Southeast Asian frog, Rana chalconota

(Amphibia: Anura: Ranidae). Zoological Journal of the LinneanSociety, 155, 123–147.

Iskandar, D. T. (1998). The Amphibians of Java and Bali. Bogor:

Puslitbang Biologi, LIPI.

Jobb, G. (2008). TREEFINDER Version of October 2008. Available

via http://www.treefinder.de.

Kuraishi, N., Matsui, M. & Ota, H. (2009). Estimation of the

origin of Polypedates leucomystax (Amphibia: Anura:

Rhacophoridae) introduced to the Ryukyu Archipelago, Japan.

Pacific Science, 63, 317–325.

Kuraishi, N., Matsui, M., Ota, H. & Chen, S.-L. (2011). Specific

separation of Polypedates braueri (Vogt, 1911) from

P. megacephalus (Hallowell, 1861) (Amphibia: Anura:

Rhacophoridae). Zootaxa, 2744, 53–61.

Kuramoto, M. (1986). Call structures of the rhacophorid frogs

from Taiwan. Scientific Report of the Laboratory for Amphibian

Biology, Hiroshima University, 8, 45–68.

Kurniati, H. (2011). Vocalization of Asian striped tree frogs,

Polypedates leucomystax (Grevenhorst, 1829) and P. iskandariRiyanto, Mumpuni & McGuire, 2011. Treubia, 38, 1–13.

Leache, A. D. & Reeder, T. W. (2002). Molecular systematics of

the eastern fence lizard (Sceloporus undulatus): a comparison of

parsimony, likelihood, and Bayesian approaches. SystematicBiology, 51, 44–68.

Li, J.-T., Che, J., Bain, R. H., Zhao, E.-M. & Zhang, Y.-P.

(2008). Molecular phylogeny of Rhacophoridae (Anura): a

15


framework of taxonomic reassignment of species within the

genera Aquixalus, Chiromantis, Rhacophorus, and Philautus.

Molecular Phylogenetics and Evolution, 48, 302–312.

Liem, S. S. (1970). The morphology, systematics, and evolution

of the Old World treefrogs (Rhacophoridae and Hyperoliidae).

Fieldiana: Zoology, 57, 1–145.

Maddison, W. P. (1997). Gene trees in species trees. SystematicBiology, 46, 523–536.

Maeda, N. & Matsui, M. (1999). Frogs and Toads of Japan, Revised

Edition. Tokyo: Bun-ichi Sogo Shuppan.

Malkmus, R., Manthey, U., Vogel, G., Hoffman, P. & Kosuch, J.

(2002). Amphibians and Reptiles of Mount Kinabalu (North Borneo).

Koenigstein: A.R.G. Gantner K.G., Koeltz Scientific Books.

Marmayou, J., Dubois, A., Ohler, A., Pasquet, E. & Tillier, A.

(2000). Phylogenetic relationships in the Ranidae (Amphibia,

Anura). Independent origin of direct development in the genera

Philautus and Taylorana. Comptes Rendus de l’Academie desSciences. Serie III, Life Sciences, 323, 287–297.

Matsui, M. (1982). Amphibians from Sabah II. Acoustic

characteristics of three common anuran species. Contributions

from the Biological Laboratory, Kyoto University, 26, 123–129.

Matsui, M. & Panha, S. (2006). A new species of Rhacophorus from

eastern Thailand (Anura: Rhacophoridae). Zoological Science, 23,

477–481.

Matsui, M. & Wu, G.-F. (1994). Acoustic characteristics of

treefrogs from Sichuan, China, with comments on systematic

relationship of Polypedates and Rhacophorus (Anura,

Rhacophoridae). Zoological Science, 11, 485–490.

Matsui, M., Seto, T. & Utsunomiya, T. (1986). Acoustic and karyotypic

evidence for specific separation of Polypedates megacephalus from

P. leucomystax. Journal of Herpetology, 20, 483–489.

Matsui, M., Ito, H., Shimada, T., Ota, H., Saidapur, S. K.,

Khonsue, W., Tanaka-Ueno, T. & Wu, G.-F. (2005).

Taxonomic relationships within the Pan-Oriental narrow-mouth

toad Microhyla ornata as revealed by mtDNA analysis (Amphibia,

Anura, Microhylidae). Zoological Science, 22, 489–495.

Matsui, M., Maryati, M., Shimada, T. & Sudin, A. (2007).

Resurrection of Staurois parvus from S. tuberilinguis from

Borneo (Amphibia, Ranidae). Zoological Science, 24, 101–106.

Matsui, M., Tominaga, A., Liu, W.-Z. & Tanaka-Ueno, T.

(2008). Reduced genetic variation in the Japanese giant

salamander, Andrias japonicus (Amphibia: Caudata). MolecularPhylogenetics and Evolution, 49, 318–326.

Matsui, M., Tominaga, A., Liu, W.-Z., Khonsue, W., Grismer, L.

L., Diesmos, A. C., Das, I., Sudin, A., Yambun, P., Yong, H.-

S., Sukumaran, J. & Brown, R. M. (2010a). Phylogenetic

relationships of Ansonia from Southeast Asia inferred from

mitochondrial DNA sequences: systematic and biogeographic

implications (Anura: Bufonidae). Molecular Phylogenetics and

Evolution, 54, 561–570.

Matsui, M., Hamidy, A., Murphy, R. W., Khonsue, W., Yambun,

P., Shimada, T., Norhayatii, A., Daicus, B. M. & Jiang, J.-P.

(2010b). Phylogenetic relationships of megophryid frogs of the

genus Leptobrachium (Amphibia, Anura) as revealed by mtDNA

gene sequences. Molecular Phylogenetics and Evolution, 56, 259–

272.

Matsui, M., Panha, S., Khonsue, W. & Kuraishi, N. (2010c). Two

new species of the ‘‘kuhlii’’ complex of the genus Limnonectesfrom Thailand (Anura: Dicroglossidae). Zootaxa, 2615, 1–22.

16 ª 2

Meegaskumbura, M., Bossuyt, F., Pethiyagoda, R.,

Manamendra-Arachchi, K., Bahir, M., Milinkovitch, M. C. &

Schneider, C. J. (2002). Sri Lanka: an amphibian hotspot.

Science, 298, 379.

Meegaskumbura, M., Meegaskumbura, S., Bowatte, G.,

Manamendra-Arachchi, K., Pethiyagoda, R., Hanken, J. &

Schneider, C. J. (2010). Taruga (Anura: Rhacophoridae), a new

genus of foam-nesting tree frogs endemic to Sri Lanka. Ceylon

Journal of Science: Biological Science, 39, 75–94.

Narins, P. M., Feng, A. S., Yong, H. S. & Christensen-Dalsgard,

J. (1998). Morphological, behavioural and genetic divergence of

sympatric morphotypes of Polypedates leucomystax in Peninsular

Malaysia. Herpetologica, 54, 129–142.

Orlov, N. L., Lathrop, A., Murphy, R. W. & Ho, T. C. (2001).

Frogs of the family Rhacophoridae (Anura: Amphibia) in the

northern Hoang Lien Mountains (Mount Fan Si Pan, Sa Pa

District, Lao Cai Province), Vitenam. Russian Journal ofHerpetology, 8, 17–44.

Orlov, N. L., Murphy, R. W., Ananjeva, N. B., Ryabov, S. A. &

Ho, T. C. (2002). Herpetofauna of Vietnam. A checklist. Part

I. Amphibia. Russian Journal of Herpetology, 9, 81–104.

Pope, C. H. (1931). Notes on amphibians from Fukien, Hainan,

and other parts of China. Bulletin of the American Museum ofNatural History, 61, 397–611.

Rambaut, A. & Drummond, A. J. (2007). Tracer v1.4. Available

via http://beast.bio.ed.ac.uk/Tracer.

Rannala, B. & Yang, Z. (1996). Probability distribution of

molecular evolution trees: a new method of phylogenetic

inference. Journal of Molecular Evolution, 43, 304–311.

Richards, C. M. & Moore, W. S. (1998). A molecular

phylogenetic study of the Old World treefrog family

Rhacophoridae. The Herpetological Journal, 8, 41–46.

Riyanto, A., Kusrini, M. D., Lubis, M. I. & Darmawan, B. (2009).

Preliminary comparison of file-eared tree frogs, Polypedatesotilophus (Boulenger, 1893) (Anura: Rhacophoridae) from Java

and other Sundaic islands, Indonesia. Russian Journal ofHerpetology, 16, 217–220.

Riyanto, A., Mumpuni & McGuire, J. A. (2011). Morphometry of

striped tree frogs, Polypedates leucomystax (Gravenhorst, 1829)

from Indonesia with description of a new species. RussianJournal of Herpetology, 18, 29–35.

Roelants, K., Gower, D. J., Wilkinson, M., Loader, S. P., Biju, S.

D., Guillaume, K., Moriau, L. & Bossuyet, F. (2007). Global

patterns of diversification in the history of modern amphibians.

Proceedings of the National Academy of Sciences of the United States

of America, 104, 887–892.

Sathiamurthy, E. & Voris, H. K. (2006). Maps of Holocene sea level

trangression and submerged lakes on the Sunda Shelf. The NaturalHistory Journal of Chulalongkorn University, Supplement, 2, 1–43.

Sheridan, J. A., Bickford, D. & Su, K. F.-Y. (2010). An

examination of call and genetic variation in three wide-ranging

Southeast Asian anuran species. The Raffles Bulletin of Zoology,

58, 369–379.

Smith, M. A. (1940). The amphibians and reptiles obtained by

Mr. Ronald Kaulback in upper Burma. Records of the Indian

Museum, 42, 465–486.

Stephens, M., Smith, N. J. & Donnelly, P. (2001). A new

statistical method for haplotype reconstruction from population

data. The American Journal of Human Genetics, 68, 978–989.



Stuart, B. L., Inger, R. F. & Voris, H. K. (2006). High level of

cryptic species diversity revealed by sympatric groups of

Southeast Asian forest frogs. Biology Letters, 2, 470–474.

Swofford, D. (2002). PAUP*: Phylogenetic Analysis Using Parsimony

(*and Other Methods), Version 4. Sunderland, MA: Sinauer

Associates.

Tanabe, A. S. (2007). Kakusan: a computer program to automate

the selection of a nucleotide substitution model and the

configuration of a mixed model on multilocus data. MolecularEcology Notes, 7, 962–964.

Tavare, S. (1986). Some probabilistic and statistical problems in

the analysis of DNA sequence. Lectures on Mathematics in the

Life Sciences, 17, 57–86.

Taylor, E. H. (1962). The amphibian fauna of Thailand. The

University of Kansas Science Bulletin, 63, 265–599.

Toda, M., Matsui, M., Nishida, M. & Ota, H. (1998). Genetic

divergence among Southeast and East Asian populations of

Rana limnocharis (Amphibia: Anura), with special reference to

sympatric cryptic species in Java. Zoological Science, 15, 607–613.

Trepanier, T. L., Lathrop, A. & Murphy, R. M. (1999).

Rhacophorus leucomystax in Vietnam with acoustic analyses of

courtship and territorial calls. Asiatic Herpetological Research, 8,

102–106.

Veith, J., Kosuch, M., Ohler, A. & Dubois, A. (2001). Systematics

of Fejervarya limnocharis (Gravenhorst, 1829) (Amphibia, Anura,

Ranidae) and related species. 2. Morphological and molecular

variation in frogs from the Greater Sunda Islands (Sumatra, Java,

Borneo) with the definition of two species. Alytes, 19, 5–28.

Wiens, J. J., Sukumaran, J., Pyron, R. A. & Brown, R. M. (2009).

Evolutionary and biogeographic origins of high tropical diversity

in Old World frogs (Ranidae). Evolution, 63, 1217–1231.

Wiles, G. J. (2000). Recent records of reptiles and amphibians

accidentally transported to Guam, Mariana Islands. Micronesica,

32, 285–287.

Wilkinson, J. A., Drewes, R. C. & Tatum, O. L. (2002). A

molecular phylogenetic analysis of the family Rhacophoridae

with an emphasis on the Asian and African genera. Molecular

Phylogenetics and Evolution, 24, 265–273.

Wolf, S. (1936). Revision der Untergattung Rhacophorus

(ausschliesslich der madagaskar-Formen). Bulletin of the RafflesMuseum, 12, 137–217.

Yang, D.-T. & Rao, D. (Eds) (2008). Amphibia and Reptilia ofYunnan. Kunming: Yunnan Science and Technology Press.

Yu, G.-H., Rao, D.-Q., Zhang, M.-W. & Yang, J.-X. (2009).

Re-examination of the phylogeny of Rhacophoridae (Anura)


based on mitochondrial and nuclear DNA. MolecularPhylogenetics and Evolution, 50, 571–579.

Zhao, E. M. & Adler, K. (1993). Herpetology of China.

Contributions to Herpetology, 10, 1–522.

Ziegler, T. (2002). Die Amphibien und Reptilien einesTieflandfeuchtwald-Schutzgebietes in Vietnam. Munster: Natur und

Tier-Verlag.

Supporting InformationAdditional Supporting Information may be found in the

online version of this article:

Table S1. Samples used in this study. L, larva; UN,

unnumbered. See text for voucher abbreviations.

Table S2. Samples used for the nuDNA gene analyses.

Sample numbers correspond to those shown in Fig. 1.

Table S3. Primers used in this study.

Table S4. Alignment statistics for fragments of the

mitochondrial 12S rRNA, tRNAval, and 16S rRNA genes

and the nuDNA genes; number of base pairs (bp); number

of variable sites (vs.); number of parsimony informative

sites (pi); and the transition ⁄ transversion ratio given for in-

groups only (ti ⁄ tv).

Table S5. Genetic distances (mean p-distance in %, fol-

lowed by ranges) in 16S rRNA between genetic clades of

Polypedates.

Table S6. A list of the Rag1 and BDNF haplotypes

observed in the mitochondrial Indochina (P. megacephalus)and Sunda (P. leucomystax) Clades. Only the variable posi-

tions are shown. For the sample numbers, refer to

Table S1.

Table S7. Estimated divergence times (MYBP) of main

divergences within Polypedates. Node numbers are shown

in Fig. 5. See text for additional details and description of

the data sets.

Please note: Wiley-Blackwell are not responsible for the

content or functionality of any supporting materials sup-

plied by the authors. Any queries (other than missing

material) should be directed to the corresponding author

for the article.

17

Phylogenetic and taxonomic relationships of the …xa.yimg.com/kq/groups/20809606/100126701/name/Zoological...Phylogenetic and taxonomic relationships of the Polypedates leucomystaxcomplex

Documents