3406: 39 58 (2012) article · pdf filezootaxa 3406: 39 –58 (2012) www

Accepted by S. Carranza: 18 May 2012; published: 1 Aug. 2012

ZOOTAXAISSN 1175-5326 (print edition)

ISSN 1175-5334 (online edition)Copyright © 2012 · Magnolia Press

Zootaxa 3406: 39–58 (2012) www.mapress.com/zootaxa/ Article

39

Phylogeography, geographic variation, and taxonomy of the Bent-toed Gecko Cyrtodactylus quadrivirgatus Taylor, 1962 from Peninsular Malaysia with the description of a new swamp dwelling species

CHELSÉA B. JOHNSON1, EVAN QUAH, S. H. 2, SHAHRUL ANUAR2, M. A. MUIN3, PERRY L. WOOD, JR.4,JESSE L. GRISMER5, LEE F. GREER1, CHAN KIN ONN5,6, NORHAYATI AHMAD6, AARON M. BAUER7 & L. LEE GRISMER1,6

1Department of Biology, La Sierra University, 4500 Riverwalk Parkway, Riverside, California, 92515-8247 USA. E-mail: [email protected]; [email protected]; [email protected] of Biological Sciences, Universiti Sains Malaysia, 11800 USM, Pulau Pinang, Penang, Malaysia. E-mail: [email protected]; [email protected] for Drug Research, Universiti Sains Malaysia, 11800 Minden, Pulau Pinang, Penang, Malaysia. E-mail: [email protected] of Biology, Brigham Young University, 150 East Bulldog Boulevard, Provo, Utah 84602 USA. E-mail: [email protected] 5Department of Biology, The University of Kansas, Lawrence, Kansas 66045-7561 USA. E-mail: [email protected]; [email protected] for Environment and Development (LESTARI), Universiti Kebangsaan Malaysia, 43600 Bangi, Selangor Darul Ehsan, Malaysia. E-mail: [email protected]; [email protected] of Biology, Villanova University, Villanova, Pennsylvania 19085-1603, USA. E-mail: [email protected]

Abstract

A review of the taxonomic status of the Bent-toed Gecko Cyrtodactylus quadrivirgatus Taylor, 1962 based on a molecularphylogeny, scalation, and color pattern analyses indicate that it is composed of a single, recently expanding, widespreadpopulation with weakly supported phylogeographic substructuring with no discrete morphological differentiation betweenpopulations. However, based on sampling, significant mean differences in selected scale counts occur between some pop-ulations. The molecular phylogeny and morphological analysis strongly indicate lineage independence between a subsetof individuals from the Bukit Panchor, Penang population and their closest relative C. pantiensis Grismer, Chan, Grismer,Wood & Belabut, 2008 from southern Peninsular Malaysia. Furthermore, the analyses indicate that the individuals of thissubset are conspecific and not part of C. quadrivirgatus as previously suggested. Additionally, this subset is morphologi-cally distinct from all other Sundaland species of Cyrtodactylus, and as such is described herein as Cyrtodactylus payacolasp. nov.

Key words: Malaysia, Bukit Panchor, Penang, Cyrtodactylus quadrivirgatus, pantiensis, payacola, taxonomy, new spe-cies

Introduction

The genus Cyrtodactylus is the most speciose genus of gekkonid lizards (at least 152 species) and the rate of newspecies being described each year shows no sign of tapering off (see http://www.reptile–database.org; Shea et al.2012). There are at least 27 species of Cyrtodactylus currently recognized from the Sunda Region of SoutheastAsia, and 16 of these species occur in Peninsular Malaysia and its associated islands (Grismer 2011). One of thesespecies, Cyrtodactylus quadrivirgatus Taylor, 1962 ranges from southern Thailand south of the Isthmus of Kra,throughout Peninsular Malaysia and its associated islands (Fig. 1) to Singapore and northern Sumatra (Grismer2011). C. quadrivirgatus is a forest dwelling, scansorial, habitat generalist ranging from sea level to 1400 m in ele-vation. Although, it is diagnosed by numerous, non-overlapping morphological characters, Grismer (2011) notedthat there was a significant geographic variation in color pattern that was associated with distinctive geographicregions and/or habitats in Peninsular Malaysia and Singapore. In this study, we compare the patterns of variation in

JOHNSON ET AL.40 · Zootaxa 3406 © 2012 Magnolia Press

FIGURE 1. Closed red circles are known localities of Cyrtodactylus payacola sp. nov. Closed white circles are localities fromwhich specimens were sampled for the molecular and morphological analysis of Cyrtodactylus quadrivirgatus. Closed blackcircle (Shah Alam, Selangor) was not sampled for the molecular analysis and closed black squares are the remaining knownlocalities for C. quadrivirgatus (fide Grismer 2011) from which no specimens were examined. Numbers correspond to the fol-lowing localities: 1. Pulau Langkawi, Kedah. 2. Pulau Tuba, Kedah. 3. Ulu Muda, Kedah. 4. Gunung Jerai, Kedah. 5. SungaiSedim, Kedah. 6. Pulau Pinang, Penang. 7. Bukit Larut, Perak. 8. Pulau Pangkor, Perak. 9. Temengor Forest Reserve, Perak. 10.Sungai Pelus, Perak. 11. Pulau Perhentian Besar, Terengganu. 12. Gunung Lawit, Terengganu. 13. Sungai Tembak, Terengganu.14. Kuala Aring, Terengganu. 15. Kuala Tahan, Pahang. 16. Sungai Lembing, Pahang. 17. Gunung Benom, Pahang. 18.Fraser’s Hill, Pahang. 19. Genting Highlands, Pahang. 20. Ulu Kali, Pahang. 21. Ulu Gombak, Selangor. 22. Lakum ForestReserve, Pahang. 23. Bukit Rengit, Pahang. 24. Bukit Bau, Pahang. 25. Pasoh, Negri Sembilan. 26. Gunung Ledang, Johor. 27.Pulau Tulai, Johor. 28. Pulau Tioman, Pahang. 29. Endau–Rompin, Johor. 30. Gunung Panti, Johor. 31. Singapore. 32. ShahAlam, Selangor. 33. Ulu Temiang Forest Reserve, Kelantan. 34. Serasa Forest Reserve, Kelantan. 35. Empangan Tembat,Terengganu. 36. Templer’s Park, Selangor. 37. Bukit Panchor, Penang. 38. Sungai Chikus Forest Reserve, Perak.

Zootaxa 3406 © 2012 Magnolia Press · 41CYRTODACTYLUS QUADRIVIRGATUS REVIEW WITH NEW SPECIES

squamation and color pattern with that of a mitochondrial DNA based phylogeny in order to assess the taxonomicstatus of various populations throughout its range in Peninsular Malaysia. We were unable to obtain tissues orspecimens from Sumatra or Thailand, and the Singapore population will be discussed separately (Grismer et al. inprep.). Our efforts focused on delimiting potential lineage boundaries between populations of C. quadrivirgatusand to determine if any of these lineages were deserving of separate species status using a general lineage speciesconcept (de Queiroz 1999).

Materials and methods

Morphological analysis. Color notes were taken from digital images of living specimens prior to preservation.The following measurements from the type series were taken with Mitutoyo dial calipers to the nearest 0.1 mmunder a Nikon SMZ645 dissecting microscope on the body where appropriate: snout-vent length (SVL), taken fromthe tip of snout to the vent; tail length (TL), taken from the vent to the tip of the tail, original or regenerated; tailwidth (TW), taken at the base of the tail immediately posterior to the postcloacal swelling; forearm length (FL),taken of the dorsal surface from the posterior margin of the elbow while flexed 90˚ to the base of the heel; axilla togroin length (AG), taken from the posterior margin of the forelimb at its insertion point on the body to the anteriormargin of the hind limb at its insertion point on the body; head length (HL), the distance from the posterior marginof the retroarticular process of the lower jaw to the tip of the snout; head width (HW), measured at the angle of thejaws; head depth (HD), the maximum height of head from the occiput to the throat; eye diameter (ED), the greatesthorizontal diameter of the eyeball; eye to ear distance (EE), measured from the anterior edge of the ear opening tothe posterior edge of the eyeball; eye to snout distance (ES), measured from anteriormost margin of the eye ballotto the tip of snout; eye to nostril distance (EN), measured between the anterior margin of the eyeball to the posteriormargin of the external nares; interorbital distance (IO), measured between the anterior edges of the orbit; ear length(EL), the greatest horizontal distance of the ear opening; and internarial distance (IN), measured between the naresacross the rostrum. Additional character states evaluated on the type series and comparative material (Appendix)were numbers of supralabial and infralabial scales counted from the largest scale immediately posterior to the dor-sal inflection of the posterior portion of the upper jaw to the rostral and mental scales, respectively; the presence orabsence of tubercles on the anterior margin of the forearm; the number of paravertebral tubercles between limbinsertions counted in a straight line immediately left of the vertebral column starting at the midpoint between theforelimb insertions and ending at the midpoint between the hind limb insertions; the number of longitudinal rowsof body tubercles counted transversely across the center of the dorsum from one ventrolateral fold to the other; thenumber of subdigital lamellae beneath the fourth toe counted from the base of the first phalanx to the claw; the totalnumber of precloacal pores; the presence or absence of a precloacal depression or groove; the degree and arrange-ment of body and tail tuberculation; the relative size and morphology of the subcaudal scales; the presence orabsence of a white network of lines forming a reticulum on the top of the head; color pattern on body and nape (i.e.,striped, banded, or blotched); degree of striping on the flanks and their degree of contact with a postorbital stripe;and the presence or absence of wide, dark bands on an original tail.

The meristic data were statistically analyzed using Analysis of Variance (ANOVA) and a t-test to test for sig-nificant differences between population means in populations that had a sample size of at least four. Populationswere determined a priori based on geography and sampling. No sexual dimorphism was observed in any scalecounts, but only males were scored for precloacal pores due to their absence in females. Additional specimensexamined are listed in the appendix. Institutional abbreviations are as follows: we retain ZRC (Zoologial ReferenceCollection) for USDZ (University of Singapore Department of Zoology), following conventional usage; LSUHCrefers to the La Sierra University Herpetological Collection, La Sierra University, Riverside, California, USA; andFMNH refers to the Field Museum of Natural History, Chicago, Illinois, USA.

Phylogenetic analysis. We obtained sequence data from the mitochondrial NADH dehydrogenase subunit 2(ND2) gene including the flanking tRNA genes (tRNA-trp, tRNA-ala, tRNA-asn, tRNA-cys, tRNA-tyr) from 80in-group samples and 13 out-group taxa based on Wood et al. in prep (Table 1). Total genomic DNA was isolated

from liver or skeletal muscle specimens stored in 95% ethanol using the Qiagen DNeasyTM tissue kit (Valencia, CA,USA) following the standard protocol for animal tissue. ND2 was amplified using a double-stranded PolymeraseChain Reaction (PCR) under the following conditions: 2.5 µl genomic DNA (concentration 10–30 µg of DNA); 2.5


µl light strand primer (concentration 4ppm) L4437b 5’–AAGCAGTTGGGCCCATACC–3’ (Macey and Schulte.1999); 2.5 µl heavy strand primer (concentration 4ppm) L5002 5’–AACCAAACCCAACTACGAAAAAT–3’(Macey and Schulte 1999); 7.5 µl of Qiagen Taq PCR Core Kit (Valencia, CA, USA), which contains 5 units/µl TaqDNA Polymerase, PCR Buffer 15 mM MgCl2, CoraLoad PCR Buffer 15 mM MgCl2, Q-Solution 5x solution, dNTPMix 10 mM of each dNTP, MgCl2, and 10.0 µl H2O. PCR reactions were executed on an Eppendorf Mastercyclergradient theromocycler under the following conditions: initial denaturation at 95°C for 2 min, followed by a seconddenaturation at 95°C for 35 s, annealing at 50–55°C for 35 s, followed by a cycle extension at 72°C for 35 s, for 31cycles (Greenbaum et al. 2007). PCR products were purified using AMPure magnetic bead solution (AgentcourtBioscience, Beverly, MA, USA). Purified PCR products were then sequenced through the Davis Sequencing, Inc.facility (Davis, CA). The two previous primers were used for sequencing along with two internal primers,CyrtintF1 –TAGCCYTCTCYTCYATYGCCC–3’ (Siler et al. 2010) and CyrtintR1 5’–ATTGTKAGDGTRGCY-AGGSTKGG–3’ (Siler et al. 2010). Sequences were uploaded, assembled, and edited in GeneiousTM v5.4 (Drum-mond et al. 2011). The protein-coding region of the ND2 sequence was initially aligned by eye. The flankingtRNAs were reconstructed using ARWEN v1.2 (Laslett and Canbäck 2008) and then aligned by eye. They werelater adjusted in MacClade v4.08 (Maddison and Maddison 2005). MacClade was also used to calculate the correctamino acid reading frame and to confirm the lack of premature stop codons.

TABLE 1. A collective list of taxa used in the molecular analyses, including voucher numbers, locality information, andGenBank accession numbers for the mitochondrial gene ND2. For museum collection abbreviations, see materials andmethods.

Voucher Species Locality ND2

LSUHC 8900 Cyrtodactylus semenanjun-gensis

West Malaysia, Johor, Gunung Panti FR, Bunker Trail JQ889177

LSUHC 8933 C. batucolus West Malaysia, Melaka, Pulau Besar JQ889178

LSUHC 8934 C. batucolus West Malaysia, Melaka, Pulau Besar JQ889179

LSUHC 6471 C. elok West Malaysia, Pahang, Fraser's Hill, the Gap JQ889180

FMNH 255454 C. interdigitalis Lao PDR, Khammouan Province, Nakai District JQ889181

FMNH 265812 C. intermedius Thailand, Sa Kaeo, Muang Sa Kaeo JQ889182

LSUHC 9513 C. intermedius Thailand, Chantaburi Province JQ889183

LSUHC 9514 C. intermedius Thailand, Chantaburi Province JQ889184

LSUHC 8906 C. pantiensis West Malaysia, Johor, Gunung Panti FR, Bunker Trail JQ889185

LSUHC 8905 C. pantiensis West Malaysia, Johor, Gunung Panti FR, Bunker Trail JQ889186

LSUHC 6349 C. seribuatebrisis West Malaysia, Johor, Pulau Nangka Kecil JQ889187

FMNH 265806 C. sp. Thailand, Loei, Phu Rua JQ889188

LSUHC 7685 C. sworderi West Malaysia, Johor, Endau-Rompin, Peta, Sungai Kawal JQ889189

LSUHC 10070 C. payacola sp. nov. West Malaysia, Penang, Bukit Panchor JQ889190



LSUHC 10072 C. quadrivirgatus West Malaysia, Penang, Bukit Panchor JQ889193

LSUHC 10073 C. quadrivirgatus West Malaysia, Penang, Bukit Panchor JQ889194

LSUHC 4018 C. quadrivirgatus West Malaysia, Selangor, Kepong, FRIM JQ889195

LSUHC 4823 C. quadrivirgatus West Malaysia, Selangor, Kepong, FRIM JQ889196

LSUHC 4980 C. quadrivirgatus West Malaysia, Pahang, Sungai Lembing Logging Camp JQ889197

LSUHC 5017 C. quadrivirgatus West Malaysia, Pahang, Sungai Lembing Logging Camp JQ889198

LSUHC 5022 C. quadrivirgatus West Malaysia, Pahang, Pulau Tioman, Sungai Mentawak JQ889199

LSUHC 5173 C. quadrivirgatus West Malaysia, Pahang, Pulau Tioman, Gunang Kajang JQ889200

LSUHC 5562 C. quadrivirgatus West Malaysia, Pahang, Pulau Tioman, Tekek–Juara Trail JQ889201

continued next page


TABLE 1. (continued)


LSUHC 5582 C. quadrivirgatus West Malaysia, Pahang, Pulau Tioman, Tekek–Juara Trail JQ889202

LSUHC 6146 C. quadrivirgatus West Malaysia, Pahang, Pulau Tioman, Gua Tengkok Air JQ889203

LSUHC 5633 C. quadrivirgatus West Malaysia, Perak, Temengor, PITC Logging Camp JQ889204



LSUHC 6461 C. quadrivirgatus West Malaysia, Pahang, Fraser's Hill JQ889207










LSUHC 6503 C. quadrivirgatus West Malaysia, Selangor, Templer's Park JQ889217

LSUHC 6607 C. quadrivirgatus West Malaysia, Pahang, Genting Highlands JQ889218




LSUHC 6737 C. quadrivirgatus West Malaysia, Penang, Pulau Pinang, Empangan Air Hitam JQ889222



LSUHC 6778 C. quadrivirgatus West Malaysia, Penang, Pulau Pinang, Telok Bahang RF JQ889225

LSUHC 9702 C. quadrivirgatus West Malaysia, Penang, Pulau Pinang, Botanical Gardens JQ889226

LSUHC 6863 C. quadrivirgatus West Malaysia, Kedah, Pulau Langkawi, Gunung Raya JQ889227






LSUHC 9445 C. quadrivirgatus West Malaysia, Kedah, Pulau Langkawi, Gunung Machinchang

JQ889233

LSUHC 9446 C. quadrivirgatus West Malaysia, Kedah, Pulau Langkawi, Gunung Machinchang

JQ889234

LSUHC 7723 C. quadrivirgatus West Malaysia, Johor, Endau-Rompin, Peta, Sungai Semawak JQ889235

LSUHC 8127 C. quadrivirgatus West Malaysia, Johor, Selai, Lubuk Tapah Trail JQ889236



LSUHC 8970 C. quadrivirgatus West Malaysia, Johor, Gunung Ledan NP JQ889239


LSUHC 8859 C. quadrivirgatus West Malaysia, Perak, Bukit Larut JQ889241


continued next page


TABLE 2. Best-fit models of evolution selected by Model test v3.7 (Posada and Crandall 1998) and models applied for Bayes-ian analyses.

For the phylogenetic analyses we applied a pluralistic approach using one non-parametric statistical method ofMaximum Parsimony (MP). This is optimized by the shortest number of evolutionary steps and two parametric sta-tistical model-based methods, Maximum Likelihood (ML) and Bayesian Inference (BI), which optimize the treebased on the best likelihood score. The Akaike Information Criterion (AIC) as implemented in ModelTest v3.7(Posada and Crandall 1998), was used to calculate the best-fit model of evolution for each codon position (Table 2).Maximum Parsimony (MP) analyses including bootstrap estimates for nodal support were run in PAUP* v4.0(Swofford 2002). 1000 bootstrap replicates for each heuristic search was run with 10 random additional sequence

TABLE 1. (continued)


















LSUHC 9057 C. quadrivirgatus West Malaysia, Terengganu, Pulau Perhentian Besar JQ889259

LSUHC 9058 C. quadrivirgatus West Malaysia, Terengganu, Pulau Perhentian Besar JQ889260

LSUHC 9191 C. quadrivirgatus West Malaysia, Perak, Pulau Pangkor JQ889261

LSUHC 9620 C. quadrivirgatus West Malaysia, Kedah, Sungai Sedim JQ889262








Gene Model selected Models applied for ML Models applied for BI

ND2

1st pos GTR+Γ GTR+I+Γ GTR+Γ

2nd pos GTR+I+Γ GTR+I+Γ GTR+I+Γ

3rd pos GTR+Γ GTR+I+Γ GTR+Γ

tRNAs HKY+Γ GTR+I+Γ HKY+Γ


replicates using tree bisection and reconnection (TBR) branch swapping. The gaps were treated as missing data.The 1000 bootstrap replicates were summarized as a 50% majority rule consensus tree. Maximum Likelihood anal-ysis was performed using RAxML HPC v7.2.3 (Stamatakis et al. 2008); 1000 bootstrap pseudoreplicates via therapid hill-climbing algorithm (Stamatakis et al. 2008) and was partitioned by codon position for the coding region.The tRNAs were treated as one partition. The most complicated model of evolution selected for the partitioneddataset was applied to all codon position due to the computer programing limitations in the software which limitsthe models applied (see Table 2 for models applied). A partitioned Bayesian analysis was carried out in MrBayesv3.1 (Huelsenbeck and Ronquist, 2001; Ronquist and Huelsenbeck 2003) using default priors (See Table 2 formodels applied). Two simultaneous runs were performed with eight chains per run, seven hot and one cold follow-ing default priors. The analysis was run for 10,000,000 generations and sampled every 1000 generations from theMarkov Chain Monte Carlo (MCMC). The analysis was halted after the average standard deviation split frequencywas below 0.01. The program Are We There Yet? (AWTY) (Nylander et al. 2008) was employed to plot the loglikelihood scores against the number of generations to assess convergence and to determine the appropriate numberof burnin trees. We conservatively discarded the first 25% of the trees as burnin. A consensus tree was then com-puted from the two parallel runs using TreeAnnotator v1.6.1 (Drummond 2007). Nodes that had posterior probabil-ities above 0.95 were considered significantly supported. Sequences divergences (p-distances) were calculated forwithin groups and between groups using MEGA v5.05 (Tamura et al. 2011) under the following conditions: substi-tutions included; transitions and transversions; and uniform rates among sites and gaps were treated as missing data(Table 3).

Mismatch distribution and tests of selective neutrality. Mismatch distributions based on the infinite-sitesmodel (Kimura 1969, 1971) were used for the genetic data in order to compare observed base pair differences tothose of a simulated distribution under the sudden population expansion model. Results with a ragged distributionor multimodal distribution suggest that a population is stable with respect to population size whereas a unimodal orsmooth distribution implies a population is undergoing a range expansion (Slatkin and Hudson 1991; Rogers andHarpending 1992). Harpending’s Raggedness index and the sum of squares deviations were calculated in ARLE-QUIN v3.1 (Excoffer et al. 2005) under the parameters of demographic expansion for 1000 bootstrap replicates. Tocomplement the mismatch distributions, two independent neutrality tests of Tajima (1989) and Fu (1997) were usedto test for possible selection and population expansion. Each test was performed in ARLEQUIN v3.1 (Excoffer etal. 2005) under the infinite-sites model for 1000 simulated samples. Tajima’s D statistic operates under the infinite-sites model, which assumes average heterozygosity for a pair of randomly chosen alleles and is compared with theexpected number of sites segregating in each sample (Tajima 1989). Fu’s Fs statistic was used to test the selectiveneutrality of random genetic samples under the assumptions of population growth, genetic hitchhiking, and back-ground selection (Fu 1997).

Results

No discrete (i.e. non-overlapping) differences in scalation were observed between populations of Cyrtodactylusquadrivirgatus (Table 4). Although, the ANOVA and t-test found significant differences in the mean number ofparavertebral tubercles, ventral scales, and precloacal pores between varying combinations of populations fromPulau Tioman, Pahang; Genting Highlands, Pahang; Fraser’s Hill, Pahang; Templer’s Park, Selangor; SungaiSedim, Kedah; Gunung Ledang, Johor; Pulau Langkawi, Kedah; Bukit Larut, Perak (Table 5); and Bukit Panchor,Penang.

The protein coding portion of the ND2 region contained 1044 bp (including the stop codon) and the five flank-ing tRNAs contained 461 bp (including the gaps). The total alignment (with gaps) contained 1505 bp with 473 par-simony informative sites and 97 variable, parsimony uninformative sites. Color pattern varied geographically andwas generally consistent with clades/populations delimited in the phylogeographic analysis (Fig. 2–5). For exam-ple, the well-supported Pulau Langkawi and Pulau Pinang populations in the north and the Endau-Rompin/Selaipopulation (herein after referred to as Endau-Rompin) in the south, although not closely related (Fig. 2), have ageneral color pattern composed of four dark dorsal stripes (Fig. 3). The well-supported upland clades (Fig. 2) fromthe Banjaran Titiwangsa (Fraser’s Hill and Genting Highlands) consistently have two dorsolateral stripes andmedial blotches (Fig. 4). Lizards from several other populations lack dorsal stripes and have only blotches (Figs.5). For example, the blotched lizards from the lower slopes


FIGURE 2. Bayesian Inference tree (-ln L 13794.01) based on ND2 and its flanking tRNAs showing the relationships betweenpopulations of Cyrtodactylus quadrivirgatus and the placement of C. payacola sp. nov. Bayesian topology with Bayesian pos-terior probabilities (BPP), Maximum Likelihood (ML), and Maximum Parsimony (MP) bootstrap support values, respectively(BPP/ML/MP), at the nodes. Nodes with low support are indicated with a – (i.e. BPP < 0.90, ML < 70, and MP < 70).


FIGURE 3. Upper: Cyrtodactylus quadrivirgatus from Endau-Rompin, Johor. Lower: C. quadrivirgatus from Pulau Langkawi,Kedah.

of Gunung Ledang lie only 80 km to the west across from the striped Endau-Rompin population, which was oncecontinuous habitat of southern Peninsular Malaysia (Grismer and Pan 2008). Although the Gunung Ledang popula-tion is morphologically the most significantly distinct, it shows no significant morphological differences from theEndau-Rompin population (Table 5).


TABLE 4. Meristic data of Cyrtodactylus quadrivirgatus diagnostic characters. Upper number denotes the range,middle number the mean, and the lower number the standard deviation. M=male and F=female.

TABLE 4. Continued

BukitLarut

BukitPanchor

Endau-Rompin

Fraser’sHill

GentingHighlands

GunungLedang Kepong

PulauLangkawi

Sex 9M, 9F 3M, 1F 2M, 3F 9M, 6F 3M, 2F 2M, 2F 2M, 0F 5M, 2F

Supralabials 8–119.39±0.70

10–1110.25±0.50

8–109.20±0.84

8–109.00±0.53

9–109.40±0.55

1010.00±0.00

11–1010.50±0.00

99.00±0.00

Infralabials 8–119.50±0.79

8–109.25±0.98

99.00±0.00

8–108.93±0.59

9–109.60±0.55

10–1110.25±0.50

99.00±0.00

8–109.14±0.69

ParavertebralTubercles

34–3835.94±1.43

37–3937.75±0.96

32–3735.80±2.17

33–3835.8±1.47

33–3734.30±1.67

36–3736.75±0.50

35–3734.00±2.83

33–3634.71±1.38

Ventral Scales 28–3733.17±1.43

36–4038.25±1.71

35–3736.00±0.71

30–3633.27±1.62

34–3634.80±1.10

35–3735.75±0.96

35–3732.00±1.41

31–3533.43±1.40

SubdigitalLamellae on 4th Toe

18–2320.33±1.03

19–2019.25±0.50

19–2220.80±1.30

18–2120.00±1.00

20–2120.60±0.55

19–2120.25±0.96

2020.00±0.89

19–2120.00±1.00

ContinuousPrecloacal and Femoral Pores

0–42.13±1.55

0–128.75±0.58

0–53.33±1.53

0–40.89±1.45

0–43.67±0.58

0–42.75±1.89

44.00±0.00

0–42.83±1.47

EnlargedPrecloacal and Femoral Scales

31–4236.28±3.08

40–4441.25±1.89

37–3937.60±0.89

34–3937.47±1.60

34–4539.00±3.94

38–4038.75±0.96

37–4036.00±1.41

37–3938.00±1.00

PulauPangkor

PulauPerhentian

PulauPinang

PulauTioman

ShahAlam

SungaiLembing

SungaiSedim Temengor

Sex 0M, 1F 0M, 2F 5M, 1F 5M, 4F 1M, 0F 1M, 1F 5M, 2F 0M, 3F

Supralabials 1010.00±0.00

1010.00±0.00

9–109.50±0.55

8–109.00±0.71

1010.00±0.00

9–109.50±0.71

9–119.71±0.76

9–109.67±0.58

Infralabials 1010.00±0.00

1010.00±0.00

9–119.83±0.75

9–119.56±0.73

99.00±0.00

9–109.50±0.71

9–109.71±0.49

9–109.33±0.58

ParavertebralTubercles

3636.00±0.00

32–3332.50±0.71

34–3634.83±0.98

33–3635.11±1.17

3838.00±0.00

35–3736.00±1.41

33–3533.86±0.90

33–3433.67±0.58

Ventral Scales 3535.00±0.00

31–3335.00±1.41

33–3634.67±1.21

34–3635.11±0.78

3737.00±0.00

32–3636.00±1.41

35–3836.43±1.27

29–3230.33±1.53

SubdigitalLamellae on4th Toe

2121.00±0.00

21–2020.50±0.71

19–2120.00±0.89

19–2120.00±0.87

1818.00±0.00

2020.00±0.00

19–2220.57±0.98

19–2019.33±0.58

ContinuousPrecloacal and Femoral Pores

0 = no males0 = no males0 = no males


0–43.40±0.55

0–32.20±1.10

1212.00±0.00

0–11.00±0.00

0–43.20±0.84


EnlargedPrecloacal andFemoral Scales

3838.00±0.00

35–3635.50±0.71

36–4037.67±1.37

36–4037.89±1.53

4040.00±0.00

37–4038.5±2.12

37–4038.14±1.07

33–3835.67±2.52


TABLE 5. Pairwise table of significant (p ≤ 0.05) scale count differences between Cyrtodactylus quadrivirgatuspopulations with at least a sample size of four. T = Paravertebral Tubercles, V = Ventral Scales, P = Continuous Precloacal

and Femoral Pores, * = Marginal.

A subset of individuals from the Bukit Panchor population is morphologically the most distinctive (Table 6). Itvaries from all other populations of C. quadrivirgatus in having a shallow precloacal groove as opposed to lackinga groove; having 11 or 12 continuous pore-bearing precloacal scales as opposed to having 0–5 such scales; and hasa reddish iris as opposed to a dark brown to gold iris.

TABLE 6. Meristic data and diagnostic characters of Cyrtodactylus payacola sp. nov. Measurements are in mm.

PulauTioman

Fraser’sHill

GentingHighlands

SungaiSedim

GunungLedang

PulauPinang

PulauLangkawi

BukitLarut

Pulau Tioman T T

Fraser’s Hill V V

Genting Highlands T

Sungai Sedim T V V V

Gunung Ledang T V T T T,V

Pulau Pinang T T*

Pulau Langkawi V T,V P

Bukit Larut V T* P

LSUHC 9982paratype

LSUHC 10070paratype

LSUHC 10071paratype

LSUHC 10074holotype

LSUHC 10076

ZRC 2.1127

Supralabials 10 10 10 10 10 10

Infralabials 9 8 10 9 9 7

No. of Paravertebral Tubercles 39 38 37 39 38 39

No. of Ventral Scales 46 37 50 46 44 58

No. of Subdigital Lamellae on 4th Toe 19 19 19 19 18 21

No. of Precloacal Pores Present 12 0 12 11 12 15

Sex male female female male male male

SVL 66.2 66.8 66.5 67.7 52.9

TL 82.6 84.4 80.9 90.7 70.2

TW 5.3 5.2 5.5 5.3 4.1

FL 8.5 9.1 8.9 8.7 7.4

TBL 9.5 11 10.4 9.5 7.3

AG 30.6 32.2 32.1 29.6 25.4

HL 17.1 16.6 18.9 17.9 13.9

HW 9.4 9.3 10.4 9.3 7.9

HD 6.7 6.8 6.2 6.8 5.2

ED 4.1 4.1 4.1 4.1 3.2

EE 4.6 4.9 5.1 5.1 3.8

ES 7.1 7.2 7.5 7.1 5.5

EN 5 5.1 5.1 5.6 4.3

IO 4 3.8 4 4.3 3.7

EL 1.3 1.2 1.3 1.1 1.3

IN 1.9 1.6 1.8 1.6 1.8


FIGURE 4. Upper: Cyrtodactylus quadrivirgatus from Fraser’s Hill, Pahang. Lower: C. quadrivirgatus from Genting High-lands, Pahang.


FIGURE 5. Upper: uncataloged specimen of Cyrtodactylus quadrivirgatus from Perlis State Park, Perlis. Lower: uncatalogedspecimen of C. quadrivirgatus from Bukit Larut, Perak.

The three phylogenetic analyses showed strong support for the monophyly of C. quadrivirgatus (with theexclusion of a subset of the Bukit Panchor population) and the monophyly of a more inclusive group including itssister species C. sworderi (Smith 1925). It reveals 13 well-supported groups within C. quadrivirgatus that corre-


spond to specific geographic areas, namely Pulau Tioman; Sungai Lembing; Fraser’s Hill/Genting Highalnds/Kepong/Templer’s Park; Fraser’s Hill; Pulau Langkawi; Pulau Pinang/Sedim; Bukit Panchor; Sedim; Pulau Per-hentian Besar; Temengor; Selai/Endau-Rompin; Gunung Ledang; and Bukit Larut/Pulau Pangkor, even thoughthere was no strong support for relationships between these groups (Fig. 2). The Fraser’s Hill/Genting Highalnds/Kepong/Templer’s Park clade (clade 4) showed some phylogeographic substructuring (Fig. 2) which was inconsis-tent with the distribution of these populations (Fig. 1). For example, lizards collected on the same trail from theuplands of Genting Highlands grouped with lizards from both the uplands of Fraser’s Hill (34 km to the north), thelowlands of Kepong (30 km to the southwest), and the lowlands of Templer’s Park (35 km to the southwest; Figs. 1and 2). Similarly, lizards from Sedim (clades 8 and 9) grouped both with some lizards from Bukit Panchor (38 kmto the southwest) and with lizards from Pulau Pinang (36 km to the west; Figs. 1 and 2). Lastly, the single specimenfrom Pulau Pangkor, Perak was embedded within the clade comprising the upland population from Bukit Larut (80km to the northeast; Figs 1 and 2).

Calculated mismatch distributions indicate that the observed distribution is multimodal, but is not significantlydifferent from the simulated distribution (Harpending’s Raggedness index = 0.00083; p-value = 0.90400) indicat-ing that C. quadrivirgatus is undergoing a demographic range expansion. This is further corroborated by Tajima’sD statistic (D = -0.53192, p-value = 0.34300) and Fu’s Fs statistic (Fs = -24.54095, p-value = 0.00000) (Tajima1989 and Fu 1997). Although the p-value for the Tajima’s test for selective neutrality is not significantly negative,it still has a negative value, which is indicative of a population expansion or purifying selection (Tajima 1989). Fu’sFs statistic was significant in indicating a population expansion or genetic hitchhiking (Fu 1997).

The phylogenetic and morphological analyses also showed strong support for (1) the monophyly of a subset ofthe Bukit Panchor population (four of six samples); (2) the conspecificity of the individuals of this subset in thatthere was only 0.14–3.3% pairwise sequence divergence (p-distances) between them and no morphological differ-entiation (3), the sister species relationship of this subset to C. pantiensis Grismer, Chan, Grismer, Wood & Belabut2008 of southern Peninsular Malaysia; and (4) that this subset is only distantly related to C. quadrivirgatus withwhich it was provisionally considered conspecific (Grismer 2011). Given this and its morphological distinctionfrom C. quadrivirgatus, C. pantiensis, and all other species of Sundaland Cyrtodactylus, this subset is describedherein as a new species.

Systematics

Cyrtodactylus payacola sp. nov.Figs. 6, 7

Cyrtodactylus cf. quadrivirgatus Grismer 2011:429.

Holotype. Adult male (LSUHC 10074) collected by E. Quah S. H. on 4 September 2011 from Bukit Panchor StatePark, Penang, West Malaysia 05º09.465' N, 100º32.885' E at an elevation of 47 m a.s.l.

Paratypes. All paratopotypes were collected in various combinations by M. A. Muin, E. Quah S.H., S. Anuar,C. K. Onn, and L. L. Grismer from the same locality as the holotype. LSUHC 9982 was collected on 6 March 2010;LSUHC 10070 and 10071 were collected on 29 June 2011.

Additional specimens examined. LSUHC 10076 was collected from Shah Alam, Selangor (3°05.262’N,101°31.466”E; 15 m a.s.l.) by Daicus Belabut. ZRC 2.1127 was collected from the Sungai Chikus Forest Reserve,Perak (4°09.426N, 101°00.562’E; 7 m a.s.l.) during April 1925.

Diagnosis. Cyrtodactylus payacola sp. nov. is distinguished from all other Sundaland species by having amaximum SVL of 67.7 mm; moderately sized, conical, keeled body tubercles; tubercles occurring on the occiput,forelimbs, hind limbs, and beyond base of tail; 44–51 ventral scales; no transversely enlarged, median subcaudalscales; 18–21 subdigital lamellae on fourth toe; abrupt transition between postfemoral and ventral femoral scales;no femoral pores; 11 or 12 contiguous, pore-bearing precloacal scales; shallow, longitudinal, precloacal groove; apair of posteromedially elongate, triangular to semilunar-shaped, paravertebral blotches on nape prominently out-lined in light yellow; no wide, dark, ventrolateral stripes on flanks; no white reticulum on top of head; paired, semi-transversely arranged, dark blotches on body.


FIGURE 6. Type series of Cyrtodactylus payacola sp. nov. Holotype LSUHC 10074.

Description of holotype. Adult male SVL 67.7 mm; head moderate in length (HL/SVL 0.26), wide (HW/HL0.59), flat (HD/HL 0.38), distinct from neck, triangular in dorsal profile, lores weakly inflated, prefrontal regiondeeply concave, canthus rostralis smoothly rounded; snout elongate (ES/HL 0.40), rounded in dorsal profile; eyelarge (ED/HL 0.23); ear opening elliptical, obliquely oriented, moderate in size (EL/HL 0.06), eye to eye distancegreater than diameter of eye; rostral square, punctuated with pores, partially divided dorsally by three postrostralgranular scales (=internasals), bordered posteriorly by large left and right supranasals, laterally by first supralabi-als; external nares bordered anteriorly by rostral, dorsally by a large anterior supranasal and small posterior supra-nasal, posteriorly by two large postnasals, and ventrally by first supralabial; 10 (R, L) rectangular supralabialsextending to just beyond dorsal inflection of labial margins, tapering abruptly below midpoint of eye; nine (R, L)infralabials tapering smoothly posteriorly to below orbit; scales of rostrum and lores raised, slightly larger size thangranular scales on top of head and occiput; scales of occiput intermixed with slightly enlarged tubercles; supraor-bitals smooth; dorsal superciliaries elongate, smooth; mental triangular, bordered laterally by first infralabials, pos-teriorly by left and right rectangular postmentals contacting medially for 40% of their length; one row of slightlyenlarged, elongate sublabials extending posteriorly to fifth infralabial; gular scales small, granular, grading posteri-orly into slightly larger, flatter, throat scales which grade into larger, flat, smooth, imbricate, pectoral and ventralscales.

Body relatively short (AG/SVL 0.44) with well-defined ventrolateral folds; dorsal scales small, granular, inter-spersed with moderately sized, conical, semi-regularly arranged, keeled tubercles being most dense on flanks;tubercles extend from occiput to anterior one-sixth of tail; tubercles on occiput and nape relatively small, those onbody largest; approximately 21 longitudinal rows of tubercles at midbody, 37 paravertebral tubercles on body; 51flat, imbricate, ventral scales, ventral scales larger than dorsal scales; precloacal scales not large; 11 contiguous,pore-bearing precloacal scales forming a “V” bordering a longitudinal, shallow precloacal groove.

Forelimbs moderate in stature, relatively short (FL/SVL 0.13); granular scales of forearm slightly larger thanthose on body, interspersed with large, keeled tubercles; palmar scales rounded, flat; digits well-developed,inflected at basal, interphalangeal joints; subdigital lamellae transversely expanded proximal to joint inflection,more granular distal to inflection; digits slightly more narrow distal to inflections; claws well-developed, sheathedby a dorsal and ventral scale; hind limbs more robust than forelimbs, moderate in length (TBL/SVL 0.14), covereddorsally by granular scales interspersed with large, conical tubercles, covered anteriorly by flat, slightly largerscales; ventral scales of thigh flat, imbricate, larger than dorsals; ventral tibial scales flat; femoral scales imbricate;small postfemoral scales form an abrupt union with large ventral scales on posteroventral margin of thigh; plantarscales low, flat; digits well-developed, inflected at basal, interphalangeal joints; subdigital lamellae transverselyexpanded proximal to inflections, more granular distal to inflections, digits more narrow distal to inflections; 19R,18L subdigital lamellae on fourth toe; claws well-developed, sheathed by a dorsal and ventral scale.


FIGURE 7. Cyrtodactylus payacola sp. nov. showing dark (upper; LSUHC 10074) and light (lower; LSUHC 10074) colorphases.

Tail moderate, 90.7 mm in length, tapering to a point, 5.3 mm in width at base; dorsal scales of base of tailgranular becoming flatter posteriorly; no median row of transversely enlarged, subcaudal scales; subcaudal scaleslarger than dorsal caudal scales; one pair of paravertebral and dorsolateral tubercle rows on either side of midline;paravertebral rows not widely separated; caudal tubercles decrease in size posteriorly, extending approximatelyone-sixth length of tail; one enlarged, postcloacal tubercle on left and right base of tail on hemipenial swelling; allpostcloacal scales flat.

Coloration in life. Dorsal: ground color of head, neck, body, limbs, and tail light straw-yellow; light brownmottling on top of head and rostrum; diffused, light brown, postorbital patch; paired, symmetrical, triangular tosemilunar-shaped, dark blotches on entire portion of nape prominently outline in ground color; seven dark brownrectangular-shaped bands extending from below nape to base of tail; bands counter shaded with lighter color; wide,dark, elongate markings above shoulders; irregularly shaped, dark blotches on flanks; body darkly speckled over-all; brown blotching on limbs, obscure banding on hind limbs; dark body bands extend onto tail to form brownbands alternating with cream-colored bands neither of which encircle tail; anterior caudal bands transform intoblack and white, respectively, on posterior two-third of tail. Ventral: surfaces of head, body and limbs lightly stip-pled in gray; subcaudal region darkened by fine mottling; iris reddish. Lizards lighten considerably at night (Fig.7).


Variation. The paratypotypes (Fig. 6) and the additional specimens examined approximate the holotype in allaspects of coloration. The caudal bands in LSUHC 9982 encircle only the posterior one-half of the tail. The addi-tional specimens examined (LSUHC 10076 from Shah Alam, Selangor and ZRC 2.1127 from the Chikus ForestReserve, Perak) share the species-level diagnostic characters of the holotype and paratopotypes and approximatethe holotype in general aspects of coloration although the dark, dorsal markings form semi-transversely arrangedbands as opposed to forming more paired markings. Furthermore, LSUHC 10076 has paired, non-symmetrical,semilunar-shaped, dark blotches only on the upper portion of nape as opposed to extending along the entire nape ofthe neck and the caudal bands encircle the tail along its entire length. Meristic differences are shown in Table 6.

Distribution. Cyrtodactylus payacola sp. nov. is known from the Bukit Panchor State Park, Penang; the Sun-gai Chikus Forest Reserve, Perak; and Shah Alam, Selangor, Peninsular Malaysia (Fig. 1) and most likely rangesthroughout all lowland coastal areas west of the Banjaran Bintang and Titiwangsa to as far south perhaps asMelaka, Melaka where its sister species C. pantienesis has also been reported to occur (Grismer et al. 2008). To thenorth, it may not range beyond the Thai–Malaysian border owing to the potential dispersal barrier of the BanjaranNakawan.

Natural history. The three localities from which Cyrtodactylus payacola sp. nov. is known are all wet,swampy, lowland forests no higher than 15 m in elevation (Fig. 8). The lizards from Bukit Panchor and Shah Alamwere collected at night from the surface of leaves up to 2 m above the forest floor. During this period this species’coloration lightens considerably (Fig. 7) making them easily confused with C. quadrivirgatus with which it is nar-rowly sympatric at Bukit Panchor and Shah Alam. At Bukit Panchor, C. quadrivirgatus appears to occur only up tothe edge of the swampy areas, remaining in the more elevated drier regions (LSUHC 10072–73). LSUHC 10070, agravid female carrying two eggs, was collected on 29 June indicating that breeding takes place at least during thismonth.

Etymology. The specific epithet payacola is derived from the word paya meaning “swamp” in the Malay lan-guage and the Latin suffix colo, which means to inhabit or dwell in and is in reference to microhabitat in which thisspecies is found.

FIGURE 8. Swampy microhabitat of Cyrtodactylus payacola sp. nov. at Bukit Panchor, Penang.


Comparisons. Cyrtodactylus payacola sp. nov. is readily differentiated from all other Sundaland species ofCyrtodactylus on the basis of color pattern and morphology (see Chan & Norhayati 2010: Table 1). It is separatedfrom C. quadrivirgatus, a species to which it was provisionally ascribed (Grismer 2011), based on having 11 or 12contiguous, pore-bearing precloacal scales along a shallow, longitudinal precloacal groove as opposed to having0–5 pore-bearing scales and no groove and a pair of posteromedially elongate, triangular to semilunar-shaped, par-avertebral blotches on the nape that are prominently outlined in light yellow as opposed to lacking such markings.It differs from its sister species, C. pantiensis, in having more precloacal pores (11 or 12 versus eight or nine); hav-ing fewer subdigital lamellae on the fourth toes (18–21 versus 22 or 23); and having a generally immaculate groundcolor as opposed to the dense, dark brown speckling seen in C. pantiensis.

The Shah Alam specimen (LSUHC 10076) is more similar to C. payacola sp. nov in all aspects of colorationand morphology than it is to any other species of Cyrtodactylus except for having paired dark, dorsal markings andpaired, non-symmetrical, semilunar-shaped, dark blotches on only the upper portion of nape. Although it is afemale, there are 12 large, non-pore-bearing scales homologous to those pore-bearing scales in males. Therefore,we tentatively consider this specimen C. payacola sp. nov. and await the acquisition of additional material.

Discussion

The data indicate that Cyrtodactylus quadriviragtus is a wide-ranging forest generalist that has recently undergoneand may still be undergoing a range expansion as indicated by the mismatch distributions and neutrality tests, theweak branch support at the basal nodes of the phylogeny, and the geographically incongruous composition ofclades composed of individuals from different localities (Fig. 2). This is supported by field observations whereinwe found this species to occur in microhabitats of all forested areas, disturbed or undisturbed, from sea level to1400 m in elevation. Sampling bias is hypothesized to account for the significant differences in mean scale countsbetween some of the populations (Table 4) being that they do not correlate well with the results of the phylogeneticanalysis. Cyrtodactylus payacola sp. nov. did not conform to these morphological, genetic, and ecological patternsand showed a significant genetic and morphological departure from all other individuals of C. quadrivirgatus aswell as its sister species C. pantiensis of southern Peninsular Malaysia. This study indicates that a careful examina-tion of “C. quadrivirgatus” from other types of distinctive microhabitats might reveal additional species masquer-ading under that nomen and, more generally, that a closer examination of other wide-ranging common species mayreveal similar patterns.

Acknowledgements

We wish to thank the Penang State Forestry Department for their permission to conduct research in Bukit Panchor.A research pass (40/200/19 SJ.1105) was issued to LLG by the Economic Planning Unit, Prime Minister’s Depart-ment. We wish to thank Daicus Belabut for his contribution to the sample collection. For field assistance, we aremost grateful to Eugene H. Johnson II and Micah R. Johnson. This research was supported in part by grants to LLGand LFG from the College of Arts and Sciences, La Sierra University, Riverside, California and by a UniversitiSains Malaysia grant to Professor Shahrul Anuar. AMB was supported by Grant DEB 0844523 from the NationalScience Foundation (U.S.A.). CBJ is especially indebted to Dr. Harold E. Johnson and Vivian M. Johnson for theirencouragement and financial support, without which this work would not have been completed.

References

Chan, K.O. and Norhayati, A. (2010) A new insular species of Cyrtodactylus (Squamata: Gekkonidae) from northeastern Pen-insular Malaysia, Malaysia. Zootaxa, 2389, 47–56.

De Queiroz, K. (1999) The general lineage concept of species and the defining properties of the species category. In: Wildon,R. A. (Ed.), Species, new interdisciplinary essays. The MIT Press, Cambridge, Massachusetts, pp. 49–89.

Drummond A.J. and Rambaut, A. (2007) BEAST: Bayesian evolutionary analysis by sampling trees. BMC EvolutionaryBiology, 7, 214.

Drummond, A.J., Ashton, B., Buxton, S., Cheung, M., Cooper, A., Duran, C., Field, M., Heled, J., Kearse, M., Markowitz, S.,


Moir, R., Stones-Havas, S., Sturrock, S., Thierer, T., and Wilson, A. (2011) Geneious V5.4. In: Available from http://www.geneious.com/.

Excoffier, L. G. L., and Schneider, S. (2005) Arlequin ver. 3.0: An integrated software package for population genetics dataanalysis. Evolutionary Bioinformatics, 1, 47–50.

Fu, Y.X. (1997) Statistical tests of neutrality of mutations against population growth, hitchhiking and background selection.Genetics, 915–925.507.

Greenbaum, E., Bauer, Jackman, A.M., Vences, M.T.R. and Glaw, F. (2007) A phylogeny of the enigmatic Madagascan geckosof the genus Uroplatus (Squamata: Gekkonidae). Zootaxa, 1493, 41–51.

Grismer, L.L. (2011) Lizards of Peninsular Malaysia, Singapore and their Adjacent Archipelagos.Edition Chimaira, 728 pp.Grismer, L.L., C.K. Onn, Grismer, J.L., Wood, Jr., P.L. and Belabut, D. (2008) Three new species of Cyrtodactylus (Squamata:

Gekkonidae) from Peninsular Malaysia. Zootaxa,1921, 1–23.Grismer, L.L. and Pan, K.A. (2008) Diversity, endemism, and conservation of the amphibians and reptiles of southern Peninsu-

lar Malaysia and its offshore islands. Herpetological Review, 3, 270–281.Huelsenbeck, J. and Ronquist, F. (2001) MRBAYES: Bayesian inference of phylogeny. Bioinformatics, 17, 754–755.Kimura, M. (1969) The number of heterozygous sites maintained in a finite population due to steady flux of mutations. Genet-

ics, 61, 893–903.Kimura, M. (1971) Theoretical foundation of population genetics at the molecular level. Theoretical Population Biology, 2,

174–208.Laslett, D. and Canbäck, B. (2008) Arwen, a program to detect tRNA genes in Metazoan mitochondrial nucleotide sequences.

Bioinformatics, 24, 172–175.Macey, J., Schulte, J. (1999) Molecular phylogenetics, tRNA evolution, and historical biogeography in anguid lizards and

related taxonomic families. Molecular Phylogenetics and Evolution, 12, 250–272.Maddison W.P., Maddison, D.R. (2005) MacClade 4: Analysis of phylogeny and character evolution. Version 4.08a. Sunder-

land, Massachusetts: Sinauer Associates.Nylander, J., Olsson, U., Alström, P. and Sanmartín, I. (2008) Accounting for phylogenetic uncertainty in biogeography: A

Bayesian approach to dispersal–vicariance analysis of the thrushes (Aves:Turdus). Systematic Biology, 57, 257–268.Posada, D. and Crandall, K.A. (1998) Modeltest: testing the model of DNA substitution. Bioinformatics, 14, 817–818.Rogers, A.R. and Harpending, H. (1992) Population growth makes waves in the distribution of pairwise genetic differences.

Molecular Biology and Evolution, 9, 552–569.Ronquist, F. and Huelsenbeck, J.P. (2003) MRBAYES 3: Bayesian phylogenetic inference under mixed models. Bioinformat-

ics,19,1572–1574.Shea, G., Couper, P., Wilmer, J.W. and Amey, A. (2012) Revision of the genus Cyrtodactylus Gray, 1827 (Squamata: Gekkoni-

dae) in Australia. Zootaxa, 3146, 1–63.Siler, C.D., Oaks, J.R., Esselstyn, J.A., Diesmos, A.C., and Brown, R.M. (2010) Phylogeny and biogeography of Philippine

bent-toed geckos (Gekkonidae: Cyrtodactylus) contradict a prevailing model of Pleistocene diversification. MolecularPhylogenetics and Evolution, 55, 699–710.

Smith, M.A. (1925) A new ground-gecko (Gymnodactylus) from the Malay Peninsula. Journal of the Federated Malay StatesMuseum, 10, 197–199.

Slatkin, M. and Hudson, R.R. (1991) Pairwise comparisons of mitochondrial DNA sequences in stable and exponentially grow-ing populations. Genetics, 129, 555–562.

Stamatakis, A., Hoover, P. and Rougemont, J. (2008) A rapid bootstrap algorithm for the RAxML web servers. Systematic Biol-ogy, 57, 758–771.

Swofford, D.L. (2002) PAUP*. Phylogenetic analysis using parsimony (*and other methods) version 4. Sinauer Associates,Sunderland, Massachusetts.

Tajima, F. (1989) Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics, 123,585–595.

Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M. and Kumar, S. (2011) Mega5: molecular evolutionary genetics anal-ysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Molecular Biology and Evolu-tion, 28, 2731–2739.

Taylor, E.H. (1962) New Oriental reptiles. The University of Kansas Science Bulletin, 53, 209–263.Wood, Jr., P. L., Heinicke, M. P., Jackman, T. R. and Bauer, A. M. Phylogeny of bent-toed geckos (Cyrtodactylus) reveals a

west to east pattern of diversification. Molecular Phylogenetics and Evolution, In prep.

Appendix

Cyrtodactylus quadrivirgatus. Johor: Endau-Rompin LSUHC 7706, 7733, 8127, 8185–86; Gunung Ledang: LSUHC 8969–72;Kedah: Pulau Langkawi LSUHC 6863–65, 6870, 9438, 9445–56; Sungai Sedim LSUHC 9620–22, 9624–25, 9837, 9864;Pahang: Fraser’s Hill LSUHC 6460–61, 6478–79, 6484, 8081, 9082–89, 9924; Pulau Pinang LSUHC 9057–58; Pulau Tio-man LSUHC 4813, 5022, 5101, 5173, 5517, 5562, 5582, 6136, 6146; Sungai Lembing LSUHC 4980, 5017; Perak: BukitLarut LSUHC 8859–60, 9011–16, 9864–72, 9909; Pulau Pangkor LSUHC 9191; Temengor LSUHC 5633–34, 5640;Selangor: Genting Highlands LSUHC 6503, 6607–08, 6617–18; Kepong LSUHC 4018, 4823; Terengganu: Pulau Perhen-tian LSUHC 9057–58.

3406: 39 58 (2012) article · pdf filezootaxa 3406: 39 –58 (2012) www

Documents