South American Journal of Herpetology, 1(3), 2006, 192-201 ? 2006 Brazilian Society of Herpetology GENETIC RESOLUTION OF THE ENIGMATIC LESSER ANTILLEAN DISTRIBUTION OF THE FROG LEPTODACTYLUS VALIDUS (ANURA, LEPTODACTYLIDAE) KENETH YANEK1-3; W. RONALD HEYER2 AND RAFAEL O. DE SA1-4 ' Department of Biology, University of Richmond, Richmond, Virginia 23237, USA. 2 Department of Vertebrate Zoology, National Museum of Natural History, PO Box 37012, Smithsonian Institution, %?/:mgfon, DC. 200B-7072, [/&4. 3 KY current address: Division of Transplant, Department of Surgery, Virginia Commonwealth University, Richmond, Virginia 23298-0254, USA. E-mail: kyanek@mcvh-vcu.edu 4 Corresponding Author: rdesa@richmond.edu ABSTRACT: Leptodactylus validus has an unusual distribution, inhabiting Trinidad, Tobago, and the Lesser Antilles, but not the mainland of South America. This distribution is inconsistent with other distribution patterns observed for these islands. Although slight variation in adult morphology has been observed among the different island populations of L. validus, call data suggest the presence of a single species. Calls of L. pallidirostris from Venezuela and Brazil suggested that this taxon might be conspecific with L. validus. Sequence data from the 12S and 16S mt rDNA genes indicate that L. validus represents a single species throughout its distribution and is conspecific with L. pallidirostris. Dispersal of L. validus from Trinidad and Tobago to the Lesser Antilles was likely mediated by human activities. KEYWORDS: Anura, Leptodactylus validus, systematics, synonymy, distribution. INTRODUCTION The frog species Leptodactylus validus Garman, 1888 as currently understood occurs on the continen- tal islands of Trinidad and Tobago and certain ocean- ic islands of the Lesser Antilles and is absent from the mainland of South America. This distribution is inconsistent with the distribution patterns observed for other faunal elements of this region. Heyer (1994) described four possible distribution patterns expect- ed for a species in this region: 1) occurring on the mainland of South America, Trinidad and Tobago, and the Lesser Antilles, 2) present in mainland South America, Trinidad and Tobago but not on the Lesser Antilles, 3) found only in Trinidad and Tobago, or 4) found only in the Lesser Antilles. In order to resolve the enigmatic distribution pattern of L. validus, Hey- er (1994) suggested that either L. validus was present on the South American mainland or, alterna- tively, that L. validus might represent two or more closely related, morphologically similar species throughout its distribution. The most likely candidate for a currently recognized species of Leptodactylus occurring on the mainland adjacent to Trinididad and Tobago that might be conspecific with L. validus is Leptodacytlus pallidirostris Lutz, 1930. The avail- able data (Heyer, 1994) are equivocal as to whether L. pallidirostris and L. validus are conspecific or not. The purpose of this paper is to analyze molecular data for intensive samples of L. validus throughout its distribution and to assess the relationships of L. validus with closely related species. The research protocol using sequence data from the 12S and 16S mt rDNA genes is designed to evaluate whether one of the four common distribution patterns for the Lesser Antillean fauna (see above) better describes the currently un- derstood distribution of L. validus. MATERIAL AND METHODS Frozen tissue samples (liver and muscle) of Lepto- dactylus validus from Trinidad (n = 15), Tobago (n = 2), and the Lesser Antillean Islands of St. Vincent (n = 13) and Grenada (n = 20) in addition to L. pallidirostris samples (from Brazil n = 1 and Guy- ana n = 1) were included in this study. Samples of L. podicipinus (Cope, 1862) (from Brazil n = 4) and L. wagneri (Peters, 1862) (from Brazil n = 1 and Ec- uador n = 2) were also included among the ingroup taxa from the L. melanonotus species group. Leptodacty- lus chaquensis Cei, 1950 (n = 1) (L. ocellatus spe- Yanek, K. et al. 193 cies group) and L. knudseni Heyer, 1972 (n = 1) (L. pentadactylus species group) were used as out- groups. The four species groups currently recognized in the genus - Juscus,' 'melanonotus,' 'ocellatus,' and 'pentadactylus,' are based primarily on morpho- logical data (Heyer, 1969, 1974; Maxson and Heyer, 1988; Frost, 2006). Locality data for the voucher spec- imens examined are provided in Appendix I. DNA isolation, amplification, and purification Total genomic DNA was isolated using standard protocols (Hillis et al, 1996). Fragments of mitochon- drial DNA (mtDNA) from the 12S and 16S mt rDNA genes were amplified using polymerase chain reaction (PCR; Palumbi, 1996) in an MJ Research PTC-200 thermocycler and a Stratagene SCS-2 temperature cycler. Double stranded amplifications of nearly the complete 12S (-880 base pairs) and 16S (-1,500 bp) rRNA genes were amplified using four sets of primers (Goebel et al, 1999). PCR reactions for the anterior -570 bp of the 12S gene were performed using the primers 12S Tphef 5'ATAGC(A/G)CTGAA(A/G)A (C/T)GCT(A/G)AGATG3' and 12S RdSl 5'GGTAC- CGTCAAGTCCTTTGGGTT3' with JumpStart Taq DNA Polymerase (Sigma). PCR reactions consisted of a 2.5-min denaturation at 94?C, 1-minute of anneal- ing at 55?C, and a 2-min extension at 72?C, followed by 30 cycles of a 1-min denaturation at 94?C, 1 minute of annealing at 55?C, and an extension period of 1.5-min at 72?C. A second set of primers, 12SA-L 5'AAACTGGGATTAGATACCCCACTAT3' and 12SB-H 5'GAGGGTGACGGGCGGTGTGT3', was used to amplify a second 12S gene fragment of-390 bp in length, beginning approximately 80 bp prior to the end of the fragment amplified by the 12S Tphef/RdSl primers. This latter fragment was amplified using PCR Master Mix (Promega) and lowering the annealing tem- perature to 53?C. The primers 12L13 5'TTAGAAGAG- GCAAGTCGTAACATGGTA3' (Feller and Hedges, 1998) and 16H10 5'TGATTACGCTACCTTTG- CACGGT3' (Hedges, 1994) were used to amplify a segment of the 16S rDNA gene of approximately -1,030 bp in length using Master Mix (Promega). PCR conditions consisted of a 2-minute denaturation at 94?C, a 1-min annealing period at 50?C, and a 72?C exten- sion for 1.5-min, followed by 34 cycles of a 1-min de- naturation at 94?C, 1-min of annealing at 50?C, and a 1.5-min extension at 72?C. An additional 16S fragment -550 bp long was amplified under the same reaction parameters using the primers 16SaR-L 5' CGCCTGTT- TACCAAAAACAT3' and 16Sd 5 CTCCGGTCT- GAACTCAGATCACGTAG3' with Master Mix (Promega), which overlapped the end of the L13/H10 fragment by approximately 70 bp. Amplified products were purified using the Wizard Plus Miniprep DNA Purification System (Promega). Sequencing Purified templates were sequenced using SequiTh- erm Excel II DNA sequencing kits (Epicentre) in an MJ Research PTC-200 thermocycler. Amplified frag- ments were sequenced in both directions. Single-strand- ed sequencing reactions were performed using prim- ers labeled with an infrared fluorescent dye (5' IRD800; Li-Cor). The primers used for sequencing reactions were identical in sequence to those used for amplifica- tion except 16Sd, which was replaced by the primer 16SbR-H 5' CCGGTCTGAACTCAGATCACGT3' (Palumbi et al, 1991). Sequencing reactions consist- ed of a 2.5-min denaturation at 95 ?C, followed by 30 cycles of a 30 second denaturation at 95 ?C, 30 sec- onds of annealing at 58?C, and 30 seconds of exten- sion at 70?C. Reaction products for gene fragments < 650 bp in length were run on 41 cm 6% acrylamide gels and those exceeding 650 bp were run on 66 cm 4% acrylamide gels. Gels were .25 mm in thickness and sequences were collected using a Li-Cor 4000L automated DNA sequencer. Sequence Alignment Image data from each single strand sequence, along with the chromatographs constructed by the Baselma- glR Ver.4.2 software (Li-Cor Biotechnology Division), were imported into the software program AlignIR (Li-Cor Biotechnology Division) and aligned with their complementary sequence. For each strand, bands from the aligned image files and their corresponding chro- matographs were visually inspected and corrected for mismatches. Sequences were aligned with ClustalX (Thompson et al, 1997) using the multiple alignment option. Alignment ambiguities were improved manual- ly under a parsimony criterion while considering sec- ondary structure constraints (Kjer, 1995; Hickson et al, 1996). 194 Genetic resolution of Leptodactylus validus Phylogenetic Analyses Phylogenetic analyses were performed for indepen- dent and combined data sets using PAUP* 4.0b 10 (Swofford, 2002). Identical haplotypes were represent- ed by a single sample. Pair-wise genetic distances were calculated under the general time reversible model with gamma distributed rate variation. Maximum parsimony (MP) analyses were per- formed using heuristic algorithms using the tree-bisec- tion reconnection (TBR) branch swapping option. All analyses were run with bases as unordered character states; gaps were treated alternatively as missing data or as a fifth character state, in the latter insertions and deletions (indels) represent informative evolutionary changes (Simmons and Ochoterena, 2000; Simmons et al, 2001). Weighted analyses were performed with transversional (tv) changes assigned twice the weight of transitions (ti). Stem and loop positions were identi- fied using secondary structure models for Xenopus laevis (Cannone et al, 2002). Additional MP and max- imum likelihood (ML) analyses were performed on all three datasets weighting loop positions twice relative to stem positions. Strict and 50% majority rule (50% MR) consensus trees were derived for analyses resulting in multiple equally parsimonious trees. Statistical stability of inter- nal branches was assessed via nonparametric boot- strapping (Felsenstein, 1985) based on 1000 pseudorep- licates (50% MR, heuristic search). Modeltest (Ver. 3.06, Posada and Crandall, 1998) was used to select the best-fitting model of sequence evolution for each dataset and the likelihood parame- ters to be implemented in ML analyses (Fisher, 1922; Felsenstein, 1981) using PAUP*. Hierarchical Likeli- hood Ratio Tests (hLRTs) resulted in selection of the Tamura-Nei model (1993) with among-site rate heter- ogeneity (TrN+G) for all datasets. Using the Akaike Information Criterion (AIC, Bozdogan, 1987), Modelt- est identified the General Time Reversible Model (Ro- driguez et al, 1990) with invariant sites and gamma- distributed rate heterogeneity (GTR+I+G) for the 12S and combined datasets, whereas the General Time Reversible Model with gamma-distributed rate varia- tion across sites (GTR+G) was selected for the 16S dataset. For ML analyses heuristic searches were con- ducted using TBR branch swapping and nonparamet- ric bootstrapping (100 pseudoreplicates, 50% majority rule). Bootstrap values < 75% were considered well supported, between 55% and 74% moderately support- ed, and values > 55% were considered to have low support. Additional likelihood analyses were performed based on Bayesian inference using MrBayes (Ver. 3.0b4, Huelsenbeck and Ronquist, 2001). The model of sequence evolution for each dataset was selected using Modeltest under AIC. The number of substitu- tion types was set to 6, enabling the rates to vary, thus being subject to the constraint of time-reversibility (Tavare, 1986). Seven simultaneous MCMC chains were run to determine the number of samples to dis- card based on convergence of log-likelihood values. Analyses were initiated using randomly selected start- ing trees, and topologies were sampled every 10 gen- erations for 2.0xl06 generations. The resulting 50% majority-rule consensus tree was rooted using the out- group samples. For the Bayesian analyses, credibility values for a clade were considered statistically signifi- cant when posterior support values were < 99%. RESULTS Analyses of the 12S, 16S, and combined datasets resulted in Leptodactylus validus and L. pallidirostris forming a monophyletic group. Leptodactylus wag- neri and L. podicipinus samples formed exclusives clades in all analyses. Pair-wise genetic distances among the L. validus and L. pallidirostris samples were < 1% in all datasets. Corrected pair-wise genet- ic distances among samples for each data set are giv- en in Table 1. Herein, we present the combined data set and the results of the combined analyses given that the inde- pendent analyses of 12S and 16S data overall did not differ from the combined analyses. Exclusive 12S and 16S haplotypes are given in Tables 2 and 3 respective- ly. In the combined sequence data the 50 L. validus samples grouped into 11 haplotypes (Table 4); these haplotypes have 16 (0.71%) variable sites. Haplotype A includes samples from Grenada (n = 9) and St. Vin- cent (n = 8), haplotype B consists of samples from Grenada (n = 11) and St. Vincent (n = 5), whereas haplotype C includes samples from Trinidad (n = 8) and haplotype D contains the samples from Tobago (n = 2). Haplotypes E-K each consists of a single sample from Trinidad. Plots of pair-wise uncorrected p-distances versus K2p distances for 12S and 16S are given in Fig. 1A Yanek, K. et al. 195 Table 1: Percentage values of pair-wise genetic distances for: the 12S data set (top graph), 16S data set (middle graph), and combined data set (bottom graph). Distances corrected using the general time-reversible model with gamma distributed rates for variable sites. 12S Data L. validus L. pallidirostris L. podicipinus L. wagneri L. chaquensis L. validus 0-0.49 L. pallidirostris 0.12-0.62 0.25 L. podicipinus 10.51-11.15 10.52-10.85 0.12-0.88 L. wagneri 5.04-5.69 5.04-5.72 10.53-11.21 0.24-1.52 L. chaquensis 12.23-12.81 12.19-12.68 12.18-12.72 10.95-11.81 L. knudseni 16.93-17.50 17.00-17.01 15.11-15.57 15.84-16.60 12.10 16S Data L. validus L. pallidirostris L. podicipinus L. wagneri L. chaquensis L. validus 0-0.36 L. pallidirostris 0.14-0.73 0.51 L. podicipinus 15.22-16.79 15.34-16.43 0-2.1 L. wagneri 9.12-9.92 8.84-9.57 13.98-15.39 0.22-1.73 L. chaquensis 20.43-20.96 20.20-20.79 17.66-17.99 21.25-22.31 L. knudseni 26.10-26.75 25.93-26.62 26.02-26.91 28.39-28.76 20.70 Combined Data L. validus L. pallidirostris L. podicipinus L. wagneri L. chaquensis L. validus 0-0.32 L. pallidirostris 0.14-0.64 0.41 L. podicipinus 13.5-14.54 13.45-14.21 0.05-1.64 L. wagneri 7.64-8.04 7.41-7.78 12.79-13.59 0.23-1.65 L. chaquensis 17.21-17.55 16.93-17.50 15.50-15.86 16.91-17.90 L. knudseni 22.44-22.74 22.27-22.66 21.57-22.18 23.18-23.57 17.20 Table 2: 12S rDNA sequence haplotypes for L. validus samples from: Grenada (Gren), St. Vincent (StVn), Tobago (Tobo), and Trinidad (Trin). Haplotypes A and B represent multiple samples. Asterisks indicate samples used to represent a haplotype. Haplo- types C-E each consist of a single sample as listed below the table. Gren006881 Gren006882 Gren006883 Gren006939 Grenl96978 Gren 196999 Gren 197005 Gren 197006 Gren 197017 StVn056421 StVn056490 StVn056561 StVn056562 StVn056612 *StVn056613 StVnl96895 StVnl96898 Tobo186596 Tobo186597 Trin 196729 B Gren 196977 Gren 196979 Grenl96980 Gren 197000 Gren 197001 Gren 197002 Gren 197003 Gren 197004 Gren 197007 Gren 197008 Gren 197044 StVn 196894 StVn 196896 StVn 196897 StVn 196899 StVn 196900 Trinl75410 Trinl75620 *Trin 196726 Trin 196727 Trin 196730 Trin 196731 Trin 196732 Trinl96733 Trin 196734 Trinl96735 Trinl96888 (C) Trin 175424 (D) Trin 196728 (E) Trinl96886 and 1C respectively, whereas comparison of uncor- rected p-distances with corrected GTR divergences are provided in Fig. IB and ID. These graphs show a nearly linear distribution. Phylogenetic analyses of the combined data set were performed under MP, ML, and Bayesian analy- ses as described above; overall tree topologies from MP and ML trees do not differ from Bayesian trees and consequently are not shown. The MP analysis of the combined data with gaps as missing data resulted in three equally parsimonious trees (L = 765; CI = 0.85). Among the 2247 bp aligned, 527 (23.5%) characters were variable and 349 (15.5%) were par- simony-informative. The strict consensus tree placed the L. pallidirostris and L. validus samples in a mono- phyletic group with 100% support. Within this clade, the L. pallidirostris sample from Brazil appeared basal to a well-supported clade that united the L. pallidirostris sample from Guyana with a well-sup- ported L. validus subclade. Within the L. validus sub- clade there is good support for a subclade formed by the two Grenada/St. Vincent haplotypes (A and B); relationships among other L. validus samples are un- resolved. The analysis with gaps as a fifth character recovered three minimum-length trees (L = 815; CI = 0.84); the strict consensus tree was identical to the consensus tree obtained when gaps were consid- ered as missing data. An analysis applying a 2:1 (tv:ti) weighting scheme recovered three minimum-length 196 Genetic resolution of Leptodactylus validus Table 3: 16S rDNA sequence haplotypes for L. validus samples from: Grenada (Gren), St. Vincent (StVn), Tobago (Tobo), and Trinidad (Trin). Samples AC represent multiple samples. Aster- isks indicate samples used to represent a haplotype. Haplotypes D-G each consist of a single sample as listed below the table. Gren006881 Gren006882 Gren006883 Gren006939 Gren196977 Gren 196978 Gren 196979 Gren 196980 Gren 196999 Gren 197000 Gren 197001 Gren197002 Gren 197003 Gren 197004 Gren 197005 Gren 197006 Gren 197007 Gren 197008 Grenl97017 Gren 197044 StVn056421 StVn056490 StVn056561 StVn056562 StVn056612 *StVn056613 StVnl96894 StVnl96895 StVnl96896 StVnl96897 StVnl96898 StVnl96899 StVn 196900 B Trin 175424 *Trin 196726 Trin 196727 Trinl96728 Trin 196729 Trin 196730 Trin 196731 Trin 196732 Trinl96733 Trin 196734 Trinl96888 *Tobo186596 Tobo186597 (D) Trinl96886 (E) Trin 175410 (F) Trinl75620 (G) Trinl96735 trees (L = 1041; CI = 0.856). Weighting loop positions (n = 1135) relative to stem positions (n = 1112) with gaps as missing data also resulted in three minimum- length trees (L = 1269; CI = 0.850). The strict con- sensus trees from these analyses were identical to the strict consensus trees obtained from the unweighted combined dataset. Using gaps as a fifth character un- der the same weighting scheme resulted in a single tree (L = 1379; CI = 0.839) identical to the consensus tree of the unweighted analyses. The ML analysis of the combined data under the TrN+G model parame- ters resulted in a bootstrap 50% MR consensus tree similar to the one from the MP analyses, with slight Table 4: Combined sequence haplotypes for L. validus samples from: Grenada (Gren), St. Vincent (StVn), Tobago (Tobo), and Trinidad (Trin). Haplotypes A-D represent multiple samples. Sam- ples used to represent each haplotype are indicated by an asterisk. Haplotypes E-K each consist of a single sample as listed below the table. B D Gren006881 Gren006882 Gren006883 Gren006939 Grenl96978 Gren 196999 Gren 197005 Gren 197006 Gren 197017 StVn056421 StVn056490 StVn056561 StVn056562 StVn056612 *StVn056613 StVn 196895 StVn 196898 Gren 196977 *Trin 196726 *Tobo 186596 Gren 196979 Trin 196727 Tobo 186597 Trin 196730 Trinl96731 Trinl96732 Trinl96733 Trinl96734 Trin 196888 Grenl96980 Gren 197000 Gren 197001 Gren 197002 Gren 197003 Gren 197004 Gren197007 Gren 197008 Gren 197044 StVn 196894 *StVn196896 StVn 196897 StVn 196899 StVn 196900 (E) Trin 175410 (F) Trin 175424 (G) Trinl75620 (H) Trinl96728 (I) Trin 196729 (J) Trin 196735 (K) Trinl96886 differences in relationships among samples within the L. podicipinus clade. A similar tree was obtained us- ing the GTR+I+G model parameters, with better reso- lution demonstrated among samples within the L. podicipinus clade. ML analyses considering sec- ondary structure and weighting loop:stem positions (2:1), using both evolutionary models resulted in bootstrap 50% MR consensus trees identical to the tree from the analysis using the GTR+I+G model. The Bayesian analysis was performed under the GTR+I+G model settings. Convergence of the log like- lihood values among the seven MCMC chains occurred within 40,000 generations of sampling, consequently the first 4,000 trees sampled were discarded. The re- sulting consensus tree from MrBayes (Fig. 2) was sim- ilar to the unweighted ML tree obtained using the GTR+I+G model parameters, with a better resolution of relationships among L. validus samples. However, with the exception of the subclade consisting of sam- ples from Grenada/St. Vincent (haplotypes A and B), there was little posterior support for relationships among other L. validus samples in the clade. Yanek, K. et al. 197 8 ?" ^"""#"" ^^ ??? 1 ^^ ^^z* ? 0.02 (A) 0.02 0.04 0.06 0.08 0.1 0.12 0.14 Kimura 2-parameter distance 8.0.12 S^** *"* s ^^** ^*^& ? | 0.02 (B) 0.02 0.04 0.06 0.08 0.1 Corrected GRT+G distance 0.12 .14 ^ ** ' 0.14 8 C 0.12 1 0.1 ^^ ^^ * 0.08 ^^?* | 0.02 ^^1* (C) >? 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 .1, .J Kim ura 2-parameter distance S 0.18 - - 8 0.14 ^>^~' ? 0.12 g 0.02 (D) 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2 0.22 Corrected GTR+G distance Figure 1. Plots depicting relative rates of transitional saturation for the 12S (A) and 16S (C) Gene sequences; and substitutional satu- ration for 12S (B) and 16S (D) using pair-wise genetic distances corrected under the GTR evolutionary model. DISCUSSION Morphological characters and call data have been previously used to assess variation among species within the Leptodactylus podicipinus-wagneri complex (Heyer, 1994). However, morphological data were in- sufficient to resolve all species boundaries in this com- plex. The status of L. pallidirostris and L. validus within this complex remained one of the major unre- solved problems (Heyer, 1994). Slight adult morphological variation exists between populations from the Lesser Antilles and those from Trinidad and Tobago. The overall morphologies of L. pallidirostris and L. validus are very similar to each other. Leptodactylus pallidirostris is distributed throughout Venezuela, Guyana, Suriname, French Gui- ana, and northern Brazil. Available call data for L. validus from Trinidad and Tobago are similar to calls analyzed from Brazilian and Venezuelan populations of L. pallidirostris (Heyer, 1994). The overall data presented in Heyer (1994) were equivocal as to wheth- er L. pallidirostris and L. validus were conspecific or represented distinct species. All analyses performed in this study strongly sup- port a monophyletic group consisting of the L. pallidirostris and L. validus samples; moreover, L. pallidirostris samples did not cluster as an exclu- sive monophyletic clade. These results support the conspecificity of the two taxa. However, saturation of nucleotide substitutions among samples can affect the estimation of evolutionary distances and potentially result in misleading tree topologies (Swofford et al., 1996; Page and Holmes, 1998; Nei and Kumar, 2000; de Peer et al., 2002). Our assessment of sequence saturation indicates a low degree of both transitional and overall substitutional saturation for these sequenc- es (Fig. 1). The combined analyses reveal some genetic struc- turing among populations within this clade, i.e., genet- ically distant samples are also geographically distant. For example, closer relationships are demonstrated among samples of L. validus from the Lesser Anti- lles. Likewise, the L. pallidirostris sample from Guy- ana appears more closely related to the L. validus samples than the L. pallidirostris sample from Brazil. The following maximum genetic distances were ob- tained for L. validus and L. pallidirostris samples: < 0.37% with the sample from Guyana and < 0.73% for the sample from Brazil. Also, less than 0.5% se- quence divergence is observed among the L. validus samples. Levels of sequence divergence are not abso- lute predictors of species diversity; however, these low levels of sequence divergence are consistent with L. pallidirostris and L. validus being conspecific. In this scenario, L. validus is a single species distributed throughout the Lesser Antilles, Trinidad and Tobago, and adjacent mainland South America. Advertisement calls are commonly analyzed in anu- ran systematic studies. These calls are almost always 198 Genetic resolution of Leptodactylus validus species-specific in Leptodactylus (see Heyer et ah, 2005 for an exception) and in anurans in general, and usually serve as a reliable indicator of species bound- aries in frogs (Heyer and Straughan, 1976; Heyer, 1978, 1979, 1994; Heyer etal, 1996; Wieczorek and Chan- ning, 1997; Camargo etal, 2006). The available call data from Brazilian and Venezuelan populations of L. pallidirostris are very similar to the advertisement call of L. validus (Heyer, 1994), providing additional support for the conspecific status of the two species. 100 62 100 100 100 100 100 L. validus B L. validus A L. validus C L. validus G L. validus H L. validus J L. validus E L. validus I L. validus K L. validus F L validus D L. pallidirostris (Guyana) L pallidirostris (Brazil) wagneri 10616 wagneri 177351 wagneri (Brazil) L. podicipinus UC243 L. podicipinus UC244 L. podicipinus UC278 L. podicipinus 53124 L knudseni L chaquensis Figure 2. Bayesian consensus tree of combined dataset. Posterior probability values > 50% are shown above branches. See text for clade descriptions. 100 100 75 100 58 Yanek, K. et al. 199 Heyer (1994) also indicated that some of the Venezue- lan samples examined and assigned to L. pallidirostris closely resembled Trinidad and Tobago samples desig- nated as L. validus, supporting their conspecificity. Leptodactylus pallidirostris was described by Lutz in 1930 from Kartabo, Guyana, who repeatedly referred to the species' resemblance to L. validus. This is in agreement with the present study which proposes that L. pallidirostris and L. validus represent a single tax- on. Leptodactylus validus was described by Garman in 1888 "1887", therefore this name has priority over L. pallidirostris. Consequently, L. pallidirostris Lutz, 1930 is placed in the synonymy of L. validus Garman, 1888. The taxonomic resolution of L. pallidirostris as a synonym of L. validus results in the resolution of the previous distributional enigma involving L. validus sen- su Heyer, 1994. Murphy (1997) considered all five of the Lesser Antillean amphibians and reptiles that oc- cur in Trinidad and Tobago to have been recent intro- ductions from the Lesser Antilles. Murphy (1997) indi- cated that most, if not all, of the introductions were the result of human activity. The distribution and minimal genetic variation of L. validus is consistent with hu- man mediated introductions, but in this case the direc- tion was most likely from Trinidad and Tobago to the Lesser Antilles. As Trinidad and Tobago are continen- tal islands, gene flow in L. validus from Trinidad and Tobago with the mainland populations likely occurred as late as about 20,000 years ago (Murphy, 1997). RESUMEN Leptodactylus validus tiene una distribution parti- cular, encontrandose en Trinidad, Tobago, y las Antillas Menores, pero no en America del Sur. Esta distribution es inconsistente con los patrones de distribution para otros grupos en estas islas. A pesar que se ha obervado variation en la morfologia adulta de L. validus en dife- rentes islas, los datos de canto sugieren la presencia de una sola especie. Cantos de L. pallidirostris de Vene- zuela y Brasil sugieren que esta especie podria ser co- especifica con L. validus. Datos moleculares de se- cuencias de los gene mt 12S y 16S sugieren que L. validus consiste de una sola especie en su distribu- tion y que esta especie es coespecifica con L. pallidirostris. La dispersion de L. validus de Trini- dad y Tobago a las Antillas Menores podria haber ocu- rrido atraves de actividades humanas. ACKNOWLEDGMENTS We acknowledge funding support for this study through National Science Foundation Awards #9815787 and #0342918 to RdS and WRH. KY thanks the Graduate School of Arts and Sci- ences, University of Richmond, for financial support. We thank the colleagues and institutions that kindly provided tissue speci- mens and/or helped in collecting samples. LITERATURE CITED BOZDOGAN, H. 1987. Model selection and Akaike's information criterion (AIC): The general theory and its analytical extensions. Psychometrica 52(3):345-370. CAMARGO, A., R.O. DE SA AND W.R. HEYER. 2006. Phylogenetic analyses of mtDNA sequences reveal three cryptic lineages in the widespread neotropical frog Leptodactylus fuscus (Schneider, 1799) (Anura, Leptodactylidae). Biological Journal of the Linnean Society 87:325-341. CANNONE J.J., S. SUBRAMANIAN,M.N. SCHNARE, JR. COLLETT,L.M. D'SOUZA, Y. Du, B. FENG, N. LIN, L.V. MADABUSI, K.M. MULLER, N. PANDE, Z. SHANG, N. YU AND R.R. GUTELL. 2002. The comparative RNA web (CRW) site: an online database of comparative sequence and structure information for ribosomal, intron, and other RNAs. BioMed Central Bioinformatics 3:15. URL:http://www.rna.icmb.utexas.edu/. FELSENSTEIN, J. 1981. Evolutionary trees from DNA sequences: a maximum likelihood approach. Journal of Molecular Evolution 17(6):368-376. FELSENSTEIN, J. 1985. Confidence limits on phylogenies: An approach using the bootstrap. Evolution 39:783-791. FELLER, A.E. AND S.B. HEDGES. 1998. Molecular evidence for the early history of living amphibians. Molecular Phylogenetics and Evolution 9(3):509-516. FISHER, R.A. 1922. On the mathematical foundations of theoretical statistics. Philosophical Transactions of the Royal Society A222:309-368. FROST, DR. 2006. Amphibian Species of the World: an Online Reference. Version 4 (17 August, 2006). Electronic Database accessible at http://research.amnh.org/herpetology/amphibia/ index.php. American Museum of Natural History, New York, USA. GARMAN, S. 1888 "1887". West Indian Batrachia in the Museum of Comparative Zoology. Bulletin of the Essex Institute 19:13-16. GOEBEL, A.M., J.M. DONNELLY AND M.E. ATZ. 1999. PCR primers and amplification methods for 12S ribosomal DNA, the control region, cytochrome oxidase I, and cytochrome b in bufonids and other frogs, and an overview of PCR primers which have amplified DNA in amphibians successfully. Molecular Phylogenetics and Evolution 11(1):163-199. HEDGES, S.B. 1994. Molecular evidence for the origin of birds. Proceedings of the National Academy of Sciences of the United States of America 91:2621-2624. HEYER, W.R. 1969. The adaptive ecology of the species groups of the genus Leptodactylus (Amphibia, Leptodactylidae). Evolution 23:421-428. HEYER, W.R. 1974. Relationships of the marmoratus species group (Amphibia, Leptodactylidae) within the subfamily Leptodactylinae. Contributions in Science, Los Angeles County Museum 253:1-45. 200 Genetic resolution of Leptodactylus validus HEYER, W.R. 1978. Systematics of the fuscus group of the frog genus Leptodactylus (Amphibia: Leptodactylidae). Science Bulletin, Natural History Museum of Los Angeles County 29:1-85. HEYER, W.R. 1979. Systematics of the pentadactylus species group of the frog genus Leptodactylus (Amphibia: Leptodactylidae). Smithsonian Contributions to Zoology 301:1-43. HEYER, W.R. 1994. Variation within the Leptodactyluspodicipinus- wagneri complex of frogs (Amphibia: Leptodactylidae). Smithsonian Contributions to Zoology 546:1-24. HEYER, W.R. AND I.R. STRAUGHAN. 1976. A functional analysis of the mating calls of the Neotropical frog genera of Leptodactylus complex (Amphibia: Leptodactylidae). Papeis Avulsos de Zoologia 29:221-245. HEYER, W.R., J.M. GARCIA-LOPEZ AND A.J. CARDOSO. 1996. Advertisement call variation in the Leptodactylus mystaceus species complex (Amphibia: Leptodactylidae) with a description of a new sibling species. Amphibia-Reptilia 17:7-31. HEYER, W.R., R.O. DE SA AND A. RETTIG. 2005. Sibling species, advertisement calls, and reproductive isolation in frogs of the Leptodactylus pentadactylus species cluster (Amphibia, Leptodactylidae). Russian Journal of Herpetology, 12(Supplement):35-39. HICKSON, RE., C. SIMON, A. COOPER, G.S. SPICER, J. SULLIVAN AND D. PENNY. 1996. Conserved sequence motifs, alignment, and secondary structure for the third domain of animal 12S rRNA. Molecular Biology and Evolution 13(1):150-169. HILLIS, DM., C. MORITZ AND B.K. MABLE (EDS.). 1996. Molecular Systematics, Second Edition. Sinauer Associates Inc., Sunderland, Massachusetts. HUELSENBECK, J.P. AND F. RONQUIST. 2001. MRBAYES: Bayesian inference of phylogeny. Bioinformatics 17:754-755. KJER, K.M. 1995. Use of rRNA secondary structure in phylogenetic studies to identify homologous positions: an example of alignment and data presentation from the frogs. Molecular Phylogenetics and Evolution 4(3):314-330. Lure, A. 1930. Segunda memoria sobre especies brasileiras do genero Leptodactylus, incluindo outras alliadas. Memorias do Institute Oswaldo Cruz 23(1): 1-20. MAXSON, L.R. AND W.R. HEYER. 1988. Molecular systematics of the frog genus Leptodactylus (Amphibia: Leptodactylidae). Fieldiana: Zoology, New Series 41:1-13. MURPHY, J.C.E. 1997. Amphibians and Reptiles of Trinidad and Tobago, Krieger Publishing Co, Malabar, Florida, 245 pp. NEI, M. AND S. KUMAR (EDS.). 2000. Molecular Phylogenetics and Evolution, Oxford University Press, Oxford, UK. PAGE, RDM. AND EC. HOLMES (EDS.). 1998. Molecular Evolution, A Phylogenetic Approach, Blackwell Science, Oxford, UK. PALUMBI, S.R. 1996. Nucleic Acids II: The polymerase chain reaction. In Hillis, D.M., Moritz, C. and Mable, B.K. (eds.), Molecular Systematics, Second Edition, pp. 205-221. Sinauer Associates Inc., Sunderland Massachusetts. PALUMBI, S., A. MARTIN, S. ROMANO, WO. MCMILLAN, L. STICE AND G. GRABOWSKI. 1991. The simple fool's guide to PCR, version 2.0. Privately published document compiled by S. Palumbi, Department of Zoology and Kewalo Marine Laboratory. University of Hawaii, Honolulu, HI. POSADA, D. AND CRANDALL, K.A. 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14(9):817-818. RODRIGUEZ, F, J.L. OLIVER, A. MARTIN AND JR. MEDINA. 1990. The general stochastic model of nucleotide substitution. Journal of Theoretical Biology 142:485-501. SIMMONS, M.P AND H. OCHOTERENA. 2000. Gaps as characters in sequence-based phylogenetic analyses. Systematic Biology 49:369-381. SIMMONS, M.P., H. OCHOTERENA AND T CARR. 2001. Incorporation, relative homoplasy, and effect of gap characters in sequence- based phylogenetic analyses. Systematic Biology 50:454-462. SWOFFORD, D.L. 2002. PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods). Version 4.0b 10. Sinauer Associates, Sunderland, Massachusetts. SWOFFORD, D.L., GJ. OLSEN, P.J. WADDELL AND D.M. HILLIS. 1996. Phylogenetic Inference. In Hillis, D.M., Moritz, C. and Mable, B.K. (eds.), Molecular Systematics, Second Edition, pp. 407-514. Sinauer Associates Inc., Sunderland, Massachusetts. TAMURA, K. AND M. NEI. 1993. Estimation of the number of nucleotide substitutions in the control region of mitochondria! DNA in humans and chimpanzees. Molecular Biology and Evolution 10:512-526. TAVARE, S. 1986. Some probabilistic and statistical problems in the analysis of DNA sequences. Lectures on Mathematics in the Life Sciences 17:57-86. THOMPSON, J.D., T.J. GIBSON, F. PLEWNIAK, F. JEANMOUGIN AND DC. HIGGINS. 1997. The ClustalX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acid Research 25:4876-4882. WIECZOREK, A.M. AND A. CHANNING. 1997. The taxonomic status of Broadley's reed frog. African Journal of Herpetology 46(2): 110-116 Received 01 August 2006 Accepted 01 October 2006 Yanek, K. et al. 201 APPENDIX I Museum catalogue numbers, field numbers, and locality data for the samples utilized in this study. Museum abbreviations: BWMC = Bobby Witcher Memorial Collection, Avila University, LSUMZ = Louisiana State University, Museum of Natural Science, QCAZ = Museo de Zoologia de la Pontificia Universidad Catolica del Ecuador, Quito, USNM National Museum of Natural History, Smithsonian Institution. Species Museum Number Field Number Locality L. chaquensis USNM 319708 USFS 186524 L. knudseni QCAZ 13244 L. pallidirostris USNM 535774 USFS 207682 L. pallidirostris USNM 302408 WRH 8626 L. podicipinus UC 243 L. podicipinus UC 244 L. podicipinus UC 278 L. podicipinus USNM 053124 USFS 303207 L. validus BWMC 06881 L. validus BWMC 06882 L. validus BWMC 06883 L. validus BWMC 06939 L. validus USNM 314793 USFS 196977 L. validus USNM 314794 USFS 196978 L. validus USNM 314795 USFS 196979 L. validus USNM 314796 USFS 196980 L. validus USNM 314798 USFS 197044 L. validus USNM 314813 USFS 196999 L. validus USNM 314814 USFS 197000 L. validus USNM 314815 USFS 197001 L. validus USNM 314816 USFS 197002 L. validus USNM 314817 USFS 197003 L. validus USNM 314818 USFS 197004 L. validus USNM 314819 USFS 197005 L. validus USNM 314820 USFS 197006 L. validus USNM 314821 USFS 197007 L. validus USNM 314822 USFS 197008 L. validus USNM 314831 USFS 197017 L. validus USNM 314512 USFS 56421 L. validus USNM 314513 USFS 56490 L. validus USNM 314514 USFS 56612 L. validus USNM 314515 USFS56613 L. validus USNM 314718 USFS 196894 L. validus USNM 314719 USFS 196895 L. validus USNM 314720 USFS 196896 L. validus USNM 314721 USFS 196897 L. validus USNM 314722 USFS 196898 L. validus USNM 314723 USFS 196899 L. validus USNM 314714 USFS 196900 L. validus USNM 314516 USFS 56561 L. validus USNM 314517 USFS 56562 L. validus USNM 523940 USFS 186596 L. validus USNM 523941 USFS 186597 L. validus USNM 286948 USFS 175410 L. validus USNM 286959 USFS 175424 L. validus USNM 306105 USFS 175620 L. validus USNM 314671 USFS 196886 L. validus USNM 314672 USFS 196888 L. validus USNM 314627 USFS 196726 L. validus USNM 314628 USFS 196727 L. validus USNM 314629 USFS 196728 L. validus USNM 314630 USFS 196729 L. validus USNM 314631 USFS 196730 L. validus USNM 314632 USFS 196731 L. validus USNM 314633 USFS 196732 L. validus USNM 314634 USFS 196733 L. validus USNM 314635 USFS 196734 L. validus USNM 314636 USFS 196735 L. wagneri LSUMZ H-13653 JPC 12969 L. wagneri LSUMZ H-12885 JPC 10616 L. wagneri USNM 320988 USFS 177351 Argentina: Tucuman; ca. 40 km SE San Miguel de Tucuman. Ecuador: Provincia de Orellana; Parque Nacional Yasuni. Guyana: Northwest District; Baramita. Brazil: Roraima; Igarape Cocal.. Brazil: Mato Grosso do Sul; Estancia Mimosa; Mun. de Bonito. Brazil: Mato Grosso do Sul; Estancia Mimosa; Mun. de Bonito. Brazil: Sao Paulo; Chacara Renascer; Bauru. Brazil: Sao Paulo; Fazenda Jatai. Grenada: St. Andrew; Spring Gardens Estate. Grenada: St. Andrew; Birch Grove. Grenada: St. Andrew; Birch Grove. Grenada: St. George; Beausejour. Grenada: St. George; Grand Anse Bay. Grenada: St. George; Grand Anse Bay. Grenada: St. George; Grand Anse Bay. Grenada: St. George; Grand Anse Bay. Grenada: St. George; Grand Anse Bay. Grenada: St. George; Grand Anse Bay. Grenada: St. George; inland from Grand Anse Bay. Grenada: St. George; inland from Grand Anse Bay. Grenada: St. George; inland from Grand Anse Bay. Grenada: St. George; inland from Grand Anse Bay. Grenada: St. George; inland from Grand Anse Bay. Grenada: St. George; inland from Grand Anse Bay. Grenada: St. George; inland from Grand Anse Bay. Grenada: St. George; inland from Grand Anse Bay. Grenada: St. George; inland from Grand Anse Bay. Grenada: St. George; inland from Grand Anse Beach. St. Vincent: St. Andrew; near Vermont. St. Vincent: St. George; Arnos Vale. St. Vincent: St. George; Arnos Vale. St. Vincent: St. George; Arnos Vale. St. Vincent: St. George; Arnos Vale. St. Vincent: St. George; Arnos Vale. St. Vincent: St. George; Arnos Vale. St. Vincent: St. George; Arnos Vale. St. Vincent: St. George; Arnos Vale. St. Vincent: St. George; Arnos Vale. St. Vincent: St. George; Arnos Vale. St. Vincent: St. George; Rose Cottage. St. Vincent: St. George; Rose Cottage. Tobago: St. Paul; Delaford, Louis d'Or River. Tobago: St. Paul; Delaford, Louis d'Or River. Trinidad: St. George; Simla Research Station. Trinidad: St. George; north of Simla Research Station. Trinidad: St. George; near Brasso Seco Village. Trinidad: St. George; west of Carapo. Trinidad: St. George; west of Carapo. Trinidad: St. Patrick; near Chatham Beach. Trinidad: St. Patrick; near Chatham Beach. Trinidad: St. Patrick; near Chatham Beach. Trinidad: St. Patrick; near Chatham Beach. Trinidad: St. Patrick; near Chatham Beach. Trinidad: St. Patrick; near Chatham Beach. Trinidad: St. Patrick; near Chatham Beach. Trinidad: St. Patrick; near Chatham Beach. Trinidad: St. Patrick; near Chatham Beach. Trinidad: St. Patrick; near Chatham Beach. Brazil: Acre; ca. 5km N Porto Walter, inland from the Rio Jurua. Ecuador: Sucumbios; Estacion Cientffica near Cuyabeno. Ecuador: Pastaza; Coca, 1km ENE of Tiguino.