Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Nov 1.
Published in final edited form as: Mol Phylogenet Evol. 2011 Jul 21;61(2):265–277. doi: 10.1016/j.ympev.2011.07.002

Molecular systematics of the Middle American genus Hypopachus (Anura: Microhylidae)

Eli Greenbaum a,*, Eric N Smith b, Rafael O de Sá c
PMCID: PMC3175309  NIHMSID: NIHMS312239  PMID: 21798357

Abstract

We present the first phylogenetic study on the widespread Middle American microhylid frog genus Hypopachus. Partial sequences of mitochondrial (12S and 16S ribosomal RNA) and nuclear (rhodopsin) genes (1275 bp total) were analyzed from 43 samples of Hypopachus, three currently recognized species of Gastrophryne, and seven arthroleptid, brevicipitid and microhylid outgroup taxa. Maximum parsimony (PAUP), maximum likelihood (RAxML) and Bayesian inference (MrBayes) optimality criteria were used for phylogenetic analyses, and BEAST was used to estimate divergence dates of major clades. Population-level analyses were conducted with the programs NETWORK and Arlequin. Results confirm the placement of Hypopachus and Gastrophryne as sister taxa, but the latter genus was strongly supported as paraphyletic. The African phrynomerine genus Phrynomantis was recovered as the sister taxon to a monophyletic Chiasmocleis, rendering our well-supported clade of gastrophrynines paraphyletic. Hypopachus barberi was supported as a disjunctly distributed highland species, and we recovered a basal split in lowland populations of Hypopachus variolosus from the Pacific versant of Mexico and elsewhere in the Mesoamerican lowlands. Dating analyses from BEAST estimate speciation within the genus Hypopachus occurred in the late Miocene/early Pliocene for most clades. Previous studies have not found bioacoustic or morphological differences among these lowland clades, and our molecular data support the continued recognition of two species in the genus Hypopachus.

Keywords: Central America, frog, Mexico, phylogenetics, sequence divergence

1. Introduction

The microhylid genus Hypopachus Keferstein, 1867, which currently includes two species in the lowlands and highlands of southeastern Texas, Mexico and Middle America, is a group of morphologically conservative, fossorial frogs. However, species of the genus Hypopachus vary considerably in coloration, intensity and distribution of pigmentation, and male advertisement calls (Nelson 1973a, 1974). Unbeknownst to earlier workers, Hypopachus is sexually dimorphic in its pattern of foot webbing (EG, unpubl.data); this intraspecific and intersexual variation may have contributed to the current 23 available synonyms for the widespread and lowland species H. variolosus (Cope, 1866) and three for the Mesoamerican highland species H. barberi Schmidt, 1939 (Frost, 2011). The former species occurs from southern Texas, USA, to northern Costa Rica on the Atlantic versant, and from southern Sonora, Mexico, to central Costa Rica on the Pacific versant (McCranie and Wilson, 2002); its habitat ranges from desert flats to rainforests (Frost, 2011; Savage, 2002).

The genus Hypopachus currently includes two species that were extensively studied by Nelson (1973a, 1974), who examined adult morphology (with multivariate statistical analyses), color pattern, and mating calls from dozens of localities throughout the range. However, Savage (2002:395) pointed out that, “[Nelson’s] analysis left unresolved whether the nominal upland form, H. barberi, was really distinct from H. variolosus, since no diagnostic feature unequivocally separates these population systems.” Some of the characters traditionally used to distinguish the highland and lowland species include size and distance between the inner and outer metatarsal tubercles (inner and outer metatarsal tubercles separated from each other by a multiple of the width of the outer metatarsal tubercle; Köhler et al., 2006); however, we have observed intra- and interspecific variation of this character, which suggests it is not a reliable diagnostic character for the species.

Compared to most other New World anurans, microhylids are relatively dull-colored, globular, rotund, and small. Many species are burrowers, with secretive habits and brief reproductive seasons—factors contributing to their rareness in collections. These characteristics might help explain why New World microhylids have received relatively little taxonomic and systematic attention from herpetologists (Duellman and Trueb, 1986). Following taxonomic treatments of New World microhylines (Carvalho, 1954; Parker, 1934), a subfamily of microhylids that formerly included taxa in the New World and Southeast Asia, Donnelly et al. (1990) and Wild (1995) used morphological data to construct phylogenies of the New World microhylids. Wild (1995) used novel morphological data and larval characters from Donnelly et al. (1990) to construct a phylogeny of all New World microhyline genera. However, Wild (1995) did not specify outgroups and assumed the monophyly of the group, which has been partially refuted by recent large-scale phylogenetic analyses (e.g., Frost et al., 2006, Van der Meijden et al., 2007). The foci of these and subsequent molecular studies (e.g., Van der Meijden et al., 2004, 2007; Frost et al., 2006; Van Bocxlaer et al., 2006), were at the level of family or higher taxonomic categories, and included a limited number of New World microhylids (only one included data from the genus Hypopachus). Van der Meijden et al. (2007) published a global-scale phylogeny of microhylid frogs that included Hypopachus variolosus, which was recovered as sister to Gastrophryne carolinensis, the type species of both genera. No molecular study to date has included H. barberi, or examined intraspecific variation within either species of Hypopachus.

Frost and Hillis (1990) predicted that most wide-ranging polytypic species of amphibians and reptiles eventually would be found to consist of several evolutionary lineages. Recent molecular studies have modified decades-old taxonomy of Middle American herpetofauna to include more species than recognized by morphology alone (e.g., Parra-Olea et al., 2004; Shaffer et al., 2004; Mendelson et al., 2005; Pfeiler and Markow, 2008; Zarza et al., 2008; Daza et al., 2009; Douglas et al., 2010). The chaotic taxonomic history (Frost, 2011) and widespread distribution of the Middle American microhylid genus Hypopachus render it an ideal taxon for testing species boundaries with molecular data.

2. Materials and methods

2.1 Taxon sampling

We sequenced eight samples of Hypopachus barberi and 35 samples of H. variolosus from across the entire known distribution of the species (Fig. 1). Based on previous studies of microhylid relationships (Frost et al., 2006; Van der Meijden et al., 2007), we included several outgroups in the study, including: three species of Gastrophryne (the sister group to Hypopachus), one species of Elachistocleis, three species of Chiasmocleis, one species of Phrynomantis (phrynomerine), and two arthroleptid species in the genera Arthroleptis and Leptopelis. Breviceps mossambicus was used to root phylogenetic trees (Loader et al., 2004; Van der Meijden et al., 2004).

Fig. 1.

Fig. 1

Open in a new tab

Map of Mesoamerica showing the known distribution of Hypopachus in gray shading, and the locations of samples used in this study. Colors of samples correspond to lineages shown in Figure 2.

2.2 Laboratory protocols

Frog tissue samples were collected as frozen whole tissue (muscle or liver), in ethanol or lysis buffer (0.5 M Tris/0.25% EDTA/2.5% SDS pH 8.2). We isolated genomic DNA using standard phenol-choloroform extractions (Hillis et al., 1996), or with DNeasy Tissue Kits (Qiagen, Valencia, CA). Although the phenol-chloroform method usually resulted in higher yields of DNA, the Qiagen kits often isolated DNA when the former method failed to do so. Double-stranded PCR amplifications were executed with 25 μl of Promega Master Mix (Promega, Madison, WI), 23 μl of purified water, 0.4 μl of forward and reverse primers, and approximately 0.5 μl of DNA (depending on strength of DNA isolation). A segment of about 400 base pairs (bp) from the 12S rRNA gene was amplified with primers 12Sa 5-AAACTGGGATTAGATACCCCACTAT-3 and 12Sb 5-GAGGGTGACGGGCGCTGTGT-3 with the following sequencing conditions: 94 °C for 2.5 min, 53 °C for 1 min, and 72 °C for 2 min, followed by 30 cycles of 94 °C for 1 min, 53 °C for 1 min, and 72 °C for 1.5 min. A segment of about 600 bp from the 16S rRNA gene was amplified with primers 16SH10 5-TGATTACGCTACCTTTGCACGGT-3 and 16SL2A 5-CCAAACGAGCCTAGTGATAGCTGGTT-3 with the following sequencing conditions: 94 °C for 2 min, 50 °C for 1 min, and 72 °C for 1.5 min, followed by 34 cycles of 94 °C for 1 min, 50 °C for 1 min, and 72 °C for 1.5 min. A segment of about 310 bp from the rhodopsin nuclear gene was amplified with primers Rhod 1A/D (Hoegg et al., 2004) with the following conditions: 94 °C for 2 min, 49 °C for 1 min, and 72 °C for 1 min, followed by 34 cycles of 94 °C for 1 min, 49 °C for 1 min, 72 °C for 1 min, and one final cycle of 72 °C for 6 min.

Products from PCR were purified with AMPure magnetic beads (Agencourt Bioscience, Beverly, MA) or the GenElute DNA Purification Kit (Sigma-Aldrich, St. Louis, MO). Purified products were cycle-sequenced with the dideoxy chain termination method with the SequiTherm Excel II DNA sequencing kit (Epicenter Technologies, Omaha, NE) or BigDye® Terminator Cycle Sequencing Kits (Applied Biosystems, Foster City, CA, USA). SequiTherm sequencing products were run on 4% acrylamide (44-cm and 66-cm plates) gels with a Licor DNA 4000L or 4300 automatic sequencer. BigDye sequencing products were run on an ABI 3130xl automated sequencer at the DNA Core Facility at the University of Texas at El Paso (UTEP). Forward and reverse sequence contigs for each sample were assembled and edited in BaseImagr, AlignIR (Licor Biotechnology) or Seqman (DNAStar) for visual inspection of data for mismatches of aligned positions to confirm or manually correct automatic readings. All sequences generated in this study were deposited in GenBank (Table 1), and the final combined alignment is available on Treebase (http://purl.org/phylo/treebase/phylows/study/TB2:S11472).

Table 1.

Anuran specimens sampled for molecular data, and their associated voucher numbers and locality information. Museum abbreviations are listed in Leviton et al. (1985), except for Museo de Zoología, Facultad de Ciencias, Universidad Autonoma de Mexico (MZFC). Uncatalogued specimen abbreviations include: RDS (Rafael O. de Sá field series) and BPN (Brice P. Noonan field series).

Sample Locality Voucher No. GenBank Accession Numbers
12S 16S Rhodopsin
Breviceps mossambicus Tanzania: Morogoro RDS 903 JF836942 JF836996 JF837044
Arthroleptis adolfifriederici Democratic Republic of the Congo: South Kivu: Kahuzi Biega National Park, Mugaba UTEP 20311 JF836941 JF836995 JF837043
Leptopelis kivuensis Democratic Republic of the Congo: South Kivu: Kahuzi Biega National Park, Mugaba UTEP 20137 JF836940 JF836994 JF837042
Phrynomantis microps Ghana: Eastern Region: Muni Lagoon KU 290437 JF836939 JF836993 JF837041
Chiasmocleis bassleri Peru: Loreto: 1.5 km N Teniente Lopez KU 222104 JF836936 JF836990 JF837038
C. hudsoni Suriname (5.18765°; −55.6521°) BPN 925 JF836937 JF836991 JF837039
C. ventrimaculata Peru: Madre de Dios: Cuzco Amazonico, 15 km E Puerto Maldonado KU 215540 JF836938 JF836992 JF837040
Elachistocleis bicolor Paraguay: Parque Nacional San Luis de la Sierra (22°40′S; 57°21′W) KU 289149 JF836935 JF836989 JF837037
Gastrophryne carolinensis 1 USA: Louisiana: Ouachita: Russell Sage Wildlife Reserve KU 289624 JF836899 JF836953 JF837006
G. carolinensis 2 USA: Alabama: Washington Co: Boykin Wildlife Management Area UTEP 19907 JF836901 JF836955 JF837007
G. carolinensis 3 USA: Florida: Putnam Co: Caravelle Ranch Wildlife Management Area UTEP 20087 JF836900 JF836954 —
Gastrophryne olivacea 1 USA: Texas: Travis Co: Austin UTEP 18274 JF836902 JF836956 —
G. olivacea 2 USA: Texas: Hudspeth Co: Indio Mountains Research Station UTEP 19815 JF836903 JF836957 JF837008
G. usta Mexico: Guerrero: Soyatepec, 17.32791 N 99.55802 W, 738 m. UTA A-60366 JF836898 JF836952 JF837005
Hypopachus barberi 1 Mexico: Chiapas: 29.7 km NE Huixtla by Motozintla de Mendoza Rd., forested slope above road summit; 2011 m MVZ 160939 JF836886 JF836943 JF836997
H. barberi 2 Guatemala: Baja Verapaz: Carretera Salama-Pantin, 1580 m UTA A-57153 JF836888 JF836945 —
H. barberi 3 Guatemala: El Quiché: Uspantán, El Chimel UTA A-50963 JF836887 JF836944 JF836998
H. barberi 4 El Salvador: Chalatenango: Cerro El Pital, 2300 m KU 291248 JF836889 JF836946 JF836999
H. barberi 5 Guatemala: El Quiché: Uspantan (within city limits), 15°15.93′N, 90°52.07′W, 1825 m UTA A-55222 JF836922 JF836976 JF837025
H. barberi 6 Honduras: La Paz: cattle pond ca. 6 km N of Guajiquiro, 2160 m (14.146537°, −87.844239°) UF 161686 JF836891 — JF837000
H. barberi 7 Honduras: La Paz: cattle pond ca. 6 km N of Guajiquiro, 2160 m (14.146537°, −87.844239°) UF 161687 JF836892 — JF837001
H. barberi 8 Honduras: Ocotepeque: Reserva Biologica El Guisayote, 2230 m (14.438562°, −89.064444°) UF 161688 JF836890 — —
H. variolosus 1 USA: Texas: southeastern Texas RDS 277 JF836893 JF836947 —
H. variolosus 2 Mexico: Nayarit: Huajimic MZFC 13101 JF836896 JF836950 —
H. variolosus 3 Mexico: Jalisco: Sta Maria de Los Angeles-Bolanos: Rio Cartagena, 21.98340N, 103.34106W, 1602 m UTA A-57702 JF836909 JF836963 JF837014
H. variolosus 4 Mexico: Jalisco: La Mascota: 20.53515N, 104.80064W, 1133 m UTA A-57706 JF836910 JF836964 —
H. variolosus 5 Mexico: Nayarit: Navarrete, 21.646940°, − 105.115280° (max error distance: 0.952 km; datum: NAD27) MVZ 144018 JF836907 JF836961 JF837012
H. variolosus 6 Mexico: Nayarit: Navarrete, 21.646940°; − 105.115280° (max error distance: 0.952 km; datum: NAD27) MVZ 144019 JF836908 JF836962 JF837013
H. variolosus 7 Mexico: Michoacán: AOR Mexican Hwy 120 between Apatzingan and Tepalcatepec, 19.15663°N, 102.47886°W, 370 m UTA A-56885 JF836913 JF836967 JF837017
H. variolosus 8 Mexico: Michoacán: AOR between Tepalcatepec and Coalcoman, 18.79473°N, 103.14625°W, 1030 m MZFC 17650 JF836914 JF836968 JF837018
H. variolosus 9 Mexico: Guerrero: San Vicente de Benites, 17.29061°N, 100.27955°W, 951 m UTA A-56888 JF836916 JF836970 —
H. variolosus 10 Mexico: Guerrero: San Vicente de Benites, 17.29061°N, 100.27955°W, 951 m MZFC 17641 JF836917 JF836971 JF837020
H. variolosus 11 Mexico: Guerrero: Carretera Rio Santiaga-San Vicente, 17.30069°N, 100.27911°W, 904 m MZFC 17642 JF836915 JF836969 JF837019
H. variolosus 12 Mexico: Yucatán: Cenote Chamac-Kinchil MZFC uncatalogued (S/D 00000 field no.) JF836897 JF836951 JF837004
H. variolosus 13 Mexico: Campeche: Zona Arqueológica de Calakmul, 18°06′26″N, 89°48′34″W No voucher (RDS tissue collection) JF836918 JF836972 JF837021
H. variolosus 14 Mexico: Tamaulipas: Hacienda Acuna, 23.20452°N, 98.43519°W, 860 m UTA A-56889 JF836911 JF836965 JF837015
H. variolosus 15 Mexico: Querétaro: Carretera El Lobo San Juan Del Rio Hwy 120, 21.25670°N, 99.25301°W, 1111 m UTA A-56886 JF836912 JF836966 JF837016
H. variolosus 16 Mexico: Quintana Roo: no specific locality MZFC uncatalogued (CA 560 field no.) JF836895 JF836949 JF837003
H. variolosus 17 Mexico: Campeche: Carretera Xpujil-Bel-Ha, 18.89567°, −89.35042°, 197 m UTA A-60454 JF836919 JF836973 JF837022
H. variolosus 18 Mexico: Campeche: Hwy. 186, 22.4 mi E (by road) Escárcega, 18.601389°; −90.421389° (max error distance: 2.458 mi; datum: NAD27) MVZ 164760 JF836920 JF836974 JF837023
H. variolosus 19 Guatemala: Petén: Aguada near Hotel Villa Maya: 1.9 km N of Santa-Elena - El Remate Hwy on road to Peténchel UTA A-53261 JF836925 JF836979 JF837028
H. variolosus 20 Guatemala: Baja Verapaz: Salama, Salama Hospital Field UTA A-50968 JF836921 JF836975 JF837024
H. variolosus 21 Guatemala: Zacapa: Cabañas, Aldea El Arenal UTA A-50971 JF836923 JF836977 JF837026
H. variolosus 22 Guatemala: Zacapa: Cabañas, Aldea El Arenal UTA A-50973 JF836924 JF836978 JF837027
H. variolosus 23 Honduras: Olancho: San Estebán, Aldea Las Trojas UTA A-50667 JF836926 JF836980 JF837029
H. variolosus 24 Honduras: Olancho: Quebrada San Lorenzo ca. 3 km S of San Lorenzo Ariba, 220 m UF 14244 JF836928 JF836982 JF837030
H. variolosus 25 Honduras: Francisco Morazán: Tegucigalpa, 14.051°N, 87.219°W, 1000 m UTA A-56790 JF836927 JF836981 —
H. variolosus 26 El Salvador: San Miguel: Volcán de San Miguel, 740 m KU 291269 JF836930 JF836984 JF837032
H. variolosus 27 El Salvador: Usulután: Cerro del Tigre, 13°28.37′N, 88°26.21′W, 1100 m KU 289842 JF836929 JF836983 JF837031
H. variolosus 28 El Salvador: San Miguel: Volcán de San Miguel, 740 m KU 291270 JF836931 JF836985 JF837033
H. variolosus 29 Nicaragua: Matagalpa: Comarca Peñas Blancas, Finca San Sebastián UTA A-51790 JF836932 JF836986 JF837034
H. variolosus 30 Costa Rica: Puntarenas: Reserva Biologia Carara; 1.9 rd mi S of Tarcoles River on Hwy 34 TCWC 83293 JF836934 JF836988 JF837036
H. variolosus 31 USA: Texas: Willacy County: N of La Sal Vieja on Rowland Co. Rd, 4.7 miles UTA uncatalogued (RM 685 field no.) JF836894 JF836948 JF837002
H. variolosus 32 Costa Rica: Guanacaste: Cafetal, Sector Santa Rosa UCR 17777 JF836933 JF836987 JF837035
H. variolosus 33 Mexico: Morelos: Mpio. Tepoztlán: San Andrés de la Cal,18.96135 N, 99.10912 W, 1462 m UTA A-60368 JF836904 JF836958 JF837009
H. variolosus 34 Mexico: Morelos: Mpio. Tepoztlán: San Andrés de la Cal,18.96135 N, 99.10912 W, 1462 m UTA A-60457 JF836906 JF836960 JF837011
H. variolosus 35 Mexico: Guerrero: Soyatepec-El Ocotito, 17.30623°N, 99.56733°W, 715 m UTA A-60367 JF836905 JF836959 JF837010
Open in a new tab

2.3 Sequence alignment, phylogenetic analyses and divergence time estimates

An initial alignment of each gene was produced in MEGALIGN (DNA Star) with the Clustal W algorithm, and manual adjustments were made in MacClade 4.08 (Maddison and Maddison, 2005). Rhodopsin data were translated to amino acids in MacClade to confirm conservation of the amino acid reading frame, ensure proper alignment, and check for premature stop codons. After preliminary phylogenetic analyses of data from each gene resulted in congruent topologies, all data were combined for subsequent analyses.

Phylogenetic relationships among the samples were assessed with maximum parsimony (MP), maximum likelihood (ML), and Bayesian inference (BI) optimality criteria in the programs PAUP* 4.0b10 (Swofford, 2002), RAxML (Stamatakis, 2006), and MrBayes 3.1 (Ronquist and Huelsenbeck, 2003), respectively. For MP analyses, the heuristic search algorithm was used with 25 random-addition replicates, accelerated character transformation, and tree bisection-reconnection branch swapping, zero-length branches were collapsed to polytomies, and indels were treated as missing data. We used non-parametric bootstraps (1000 pseudoreplicates) to assess node support in resulting topologies (Felsenstein, 1985). Comparisons of uncorrected p sequence mitochondrial divergence data were made with the complete deletion option in MEGA4 (Tamura et al., 2007).

The corrected Akaike Information Criterion in jModelTest (Posada, 2008) was used to find the model of evolution that best fit the data for subsequent partitioned ML and BI analyses. Following previous studies that demonstrated the importance of gene and codon partitions for accurate BI analyses (e.g., Castoe et al., 2004; Brandley et al., 2005), we estimated models for 12S, 16S, and each codon position of rhodopsin. RAxML analyses were executed with partitioned datasets (one each for 12S and 16S, and one for each codon position of rhodopsin using the least restrictive model) and 100 replicate ML inferences were peformed for each analysis. Each analysis was initiated with a random starting tree, and employed the rapid hill-climbing algorithm (Stamatakis et al., 2007). Clade support was assessed with 1000 bootstrap replicates, also with the rapid-hill climbing algorithm (Stamatakis et al., 2008).

Partitioned Bayesian analyses were conducted with default priors. Analyses were initiated with random starting trees and run for 20,000,000 generations; Markov chains were sampled every 1000 generations. Convergence was checked by importing the trace files (p files) from the MrBayes output to the computer program Tracer v1.3 (http://tree.bio.ed.ac.uk/software/tracer/), which plots the likelihood values against generation number. Once the graphical plot leveled off, convergence had been met; we conservatively discarded 25% of trees as “burn in.” Four separate analyses with two independent chains were executed to check for convergence of log-likelihoods in stationarity (Huelsenbeck and Ronquist, 2001; Leaché and Reeder, 2002). To test the monophyly of clades recovered in our phylogenetic analyses, we used the Shimodaira-Hasegawa (SH) and approximately unbiased (AU) tests as implemented in CONSEL V0.1i (Shimodaira and Hasegawa, 2001; Shimodaira, 2002).

Estimates of divergence times for lineages identified in the phylogenetic analyses were calculated under a Bayesian statistical framework with the program BEAST version 1.6.1 (Drummond and Rambaut, 2007). For a calibration point, we used a fossil of Gastrophryne cf. carolinensis described by Holman (1965) from the Hemingfordian North American Stage of the early Miocene (ca. 18 mya) in Florida, which was used as the minimum age for the genus Gastrophryne. To ensure BEAST analyses reached effective sample size (ESS) values > 200 for the posterior and prior values, we used unlinked substitution and clock models for the separate mitochondrial and nuclear datasets, GTR substitution and gamma site heterogeneity models for each dataset, relaxed clock (uncorrelated lognormal) and Yule process speciation tree models, and an exponential prior on the Gastrophryne fossil calibration (Ho, 2007). Each MCMC chain was run for 20,000,000 generations (sampled every 1000 generations) to ensure ESS values were higher than 200. BEAST runs were repeated four times to ensure the stability and convergence of the MCMC chains. Combined results from the four analyses were analyzed with the program Tracer version 1.5 (Rambaut and Drummond, 2007).

2.4 Haplotype networks and analyses of geographic and population structuring

We used NETWORK 4.600 (http://www.fluxus-engineering.com) to construct haplotype median-joining networks (Bandelt et al., 1999). Networks were constructed from the combined 12S and 16S dataset for all samples of Hypopachus (weights = 10 and = 0); nuclear data had minimal variation and were not included. We computed genetic diversity (nucleotide and haplotype diversity, and mean number of pairwise differences) within the different groups identified from the phylogenetic analyses with the program ARLEQUIN 3.5 (Excoffier and Lischer, 2010). This program was also used to conduct a hierarchical analysis of molecular variance (AMOVA, Excoffier et al., 1992) to assess the most probable configuration and geographic subdivision. Populations were grouped into two, three, four, five and six groups according to different geographic hierarchies that matched the lineages recovered in our phylogenetic analyses. Groupings that maximized values of among group variation (FCT) and were statistically significant indicated the most parsimonious geographic subdivisions (Zhang et al., 2010). Finally, ARLEQUIN was used to estimate mismatch distributions for several widespread clades to test for sudden population expansion (Patou et al., 2010). Assuming the infinite sites model, the mismatch distribution is smooth and often unimodal as a result of population expansion, whereas for stationary populations the distribution is ragged and often multimodal (Harpending, 1994; Harpending et al., 1998; Rogers and Harpending, 1992). We also implemented Fu’s Fs and Tajima’s D neutrality tests as additional assessments of population expansion (Fu, 1997; Zhang et al., 2010).

3. Results

3.1 Alignment, phylogenetic analyses and haplotype networks

The total alignment for the dataset included 1275 bp (12S = 394 bp; 16S = 583 bp; rhodopsin = 298 bp). Alignment of all datasets did not result in any ambiguous regions and no stop codons were detected in the rhodopsin dataset; 417 characters were parsimony-informative (32.7%) sites. Three samples of H. barberi from Honduras failed to amplify for 16S (these were excluded from NETWORK and Arlequin analyses), and two samples of Gastrophryne and seven samples of Hypopachus failed to amplify for rhodopsin (Table 1).

The following models of nucleotide substitution were selected by jModeltest for ML and BI analyses: 12S (GTR + I + Γ); 16S (TiM3 + Γ); rhodopsin 1st codon position (TPM2uf), rhodopsin 2nd codon position (F81), rhodopsin 3rd codon position (TPM2uf + Γ). If a given model was not an option in RAxML or MrBayes, the least restrictive available model (i.e., GTR) was used. The topology of trees generated from MP, ML, and BI analyses were nearly identical, with the only difference related to the position of Gastrophryne usta, which was weakly supported (50% bootstrap support) as sister to the G. carolinensis/G. olivacea clade in the MP analysis (Fig. 2). The MP analysis of the dataset resulted in 3882 most parsimonious trees (length = 1757, CI = 0.538, RI = 0.694); the likelihood score of the most optimal tree in the ML analysis was −9737.457540.

Fig. 2.

Fig. 2

Open in a new tab

Preferred maximum likelihood phylogeny (A) generated with RAxML using combined 12S, 16S and rhodopsin datasets. The phylogenetic position of three samples of Hypopachus barberi were inferred from 12S and rhodopsin data only, and are shown with a dashed line. Branch support values are maximum parsimony/maximum likelihood/Bayesian inference. Haplotype network (B) obtained from the analysis of the mtDNA dataset (12S and 16S) are labeled according to their locations and sample ID. Haplotype connecting lines represent single mutations unless indicated otherwise (in parentheses). Median vectors introduced by the NETWORK algorithm are shown as small black circles. Hb = Hypopachus barberi; Hv = Hypopachus variolosus.

Strongly supported clades (> 70% bootstrap and 0.95 posterior probability [pp]) from all three analyses (Fig. 2A) included: a monophyletic Hypopachus; a monophyletic H. barberi (from highland localities in Chiapas [Mexico], Guatemala, Honduras and El Salvador); and a monophyletic H. variolosus. Within H. variolosus, there was a well-supported, basal split between the Nayarit, Morelos, and Guerrero clades from the Pacific versant of Mexico (WMX), and the Texas, Central, Costa and Baja clades from elsewhere in the Mesoamerican lowlands (MML). The corresponding haplotype network (Fig. 2B) also showed two groups corresponding to the WMX and MML clades, differing by 26 mutations. Samples of H. barberi were separated into a third group from the Central Clade by 51 mutations (Fig. 2B). Maximum likelihood and BI analyses strongly supported Gastrophryne usta as sister to Hypopachus, and all analyses recovered moderate support for the position of a Gastrophryne carolinensis + G. olivacea clade as sister to the G. usta/Hypopachus clade. Our analyses strongly supported Elachistocleis as sister to the Gastrophyrne/Hypopachus clade, a monophyletic Chiasmocleis, and a paraphyletic Gastroprhyninae because of the well-supported inclusion of the African phrynomerine genus Phrynomantis (moderately supported as sister to Chiasmocleis). Divergence times estimated from BEAST are listed in Table 2. The earliest lineage divergence within the genus Hypopachus occurred in the late Miocene, with subsequent divergences occuring in the Pliocene and Pleistocene.

Table 2.

Results of estimated divergence times (in millions of years) from BEAST analyses. Node names correspond to clades shown in Fig. 1.

Node Mean Standard error of mean 95% HPD lower 95% HPD upper Effective sample size
Gastrophryne 18.83 0.03 18.00 20.49 67904.26
Hypopachus 11.71 0.08 7.37 15.98 859.96
WMX 6.00 0.05 3.00 9.38 1044.86
MML 6.38 0.06 3.38 9.60 831.88
Barberi 4.29 0.07 1.48 7.84 677.31
Nayarit 2.95 0.04 0.76 5.58 985.47
Morelos 0.86 0.01 0.03 2.26 4049.04
Guerrero 2.95 0.03 1.09 5.10 1345.57
Texas 4.30 0.05 1.96 6.94 843.01
Central 3.26 0.04 1.23 5.63 1195.36
Costa 0.72 0.01 0.02 2.08 3085.51
Baja 1.50 0.02 0.25 3.26 1289.07
Open in a new tab

3.2 Sequence variation and monophyly tests

The largest uncorrected p genetic distance in the mitochondrial dataset was between the outgroups Arthroleptis adolfifriederici and Breviceps mossambicus (29.4%). Among samples of Gastrophryne, the largest uncorrected p mitochondrial genetic distance was between G. usta and G. olivacea (14.3%); divergences between G. usta and G. carolinensis ranged from 13.9–14.2%, but divergences between sister taxa G. carolinensis and G. olivacea ranged from 6.9–7.1%. Variation within well-supported clades of Hypopachus ranged from 0.2–5.2% for H. barberi (maximum variation only 1.7% when the divergent Chiapas sample was excluded), whereas the divergences for H. variolusus ranged from 0.0–3.6% for WMX and 0.0–4.0% for MML. Uncorrected p mitochondrial divergences between well-supported clades of Hypopachus were: 4.7–10.0% (H. barberi vs. MML), 5.7–10.4% (H. barberi vs. WMX), and 3.5–6.5% (MML vs. WMX). We tested the monophyly of two taxa that were recovered as paraphyletic in our analyses, Gastrophryne and Gastrophryninae. The monophyly of Gastrophryne was not rejected by the AU (P = 0.090) and SH (P = 0.118) tests, but the monophyly of Gastrophryninae was rejected by the AU (P = 0.037) test, and a near-significant result for the SH (P = 0.00051) test.

3.3 Population and geographic structure

All 40 included samples of Hypopachus had unique haplotypes. Genetic diversity estimates and neutrality tests within populations of Hypopachus are shown in Table 3; populations with low sample sizes included the Morelos, Costa and Baja groups (n ≤ 3). Among all individuals, π = 0.0489, h = 0.999 and κ = 46.13, suggesting that haplotype diversity is high and nucleotide diversity is relatively low. This pattern suggests that modern populations have very low levels of gene flow among them, and it is likely that these populations evolved in relative isolation from each other. In the AMOVA (Table 4), the highest amount of genetic variance among groups (FCT = 0.406, P < 0.01) was found with three groups ([Barberi] [Morelos, Nayarit, Guerrero] [Baja, Costa, Central, Texas]), which are congruent with the three well-supported clades for H. barberi, MML and WMX recovered in the phylogenetic analyses (Fig. 2).

Table 3.

Genetic diversity estimates and tests of neutrality within Hypopachus, partitioned into genetic groups identified from the phylogenetic analyses in Figure 2.

Clade n π h κ Fu’s Fs P-value Tajima’s D P-value
Barberi 5 0.0339 1.0 31.20 1.027 0.433 −0.911 0.201
Nayarit 5 0.0122 1.0 11.60 −0.144 0.277 −0.654 0.357
Morelos 2 0.0032 1.0 3.00 1.098 0.442 0 1.0
Guerrero 6 0.0274 1.0 25.67 0.318 0.344 −0.185 0.452
Texas 10 0.0172 1.0 16.29 −2.24 0.079 0.202 0.627
Central 7 0.0103 0.952 9.81 0.117 0.428 −0.537 0.336
Costa 2 0.0011 1.0 1.00 0 0.25 0 1.0
Baja 3 0.0109 1.0 10.33 1.174 0.466 11946792.089 1.0
TOTAL 40 0.0489 0.999 46.13 −8.416 0.011 −0.4608 0.368
Open in a new tab
π

nucleotide diversity; h = haplotype diversity; κ = mean number of pairwise differences; n = samples size.

Table 4.

Hierarchical analysis of AMOVA of Hypopachus populations.

Groups FST FSC FCT % Among groups % Within populations
All populations 0.687*** — — 68.68 31.32
2 Groups: [Barberi, Baja, Costa, Central, Texas] [Morelos, Nayarit, Guerrero] 0.719*** 0.643*** 0.212* 21.24 28.14
2 Groups: [Barberi, Morelos, Nayarit, Guerrero] [Baja, Costa, Central, Texas] 0.712*** 0.642*** 0.195* 19.50 28.81
2 Groups: [Barberi] [Baja, Costa, Central, Texas, Morelos, Nayarit, Guerrero] 0.773*** 0.639*** 0.372 37.16 22.71
3 Groups: [Barberi] [Morelos, Nayarit, Guerrero] [Baja, Costa, Central, Texas] 0.727*** 0.541*** 0.406** 40.57 27.31
3 Groups: [Barberi, Baja, Costa, Central, Texas] [Nayarit] [Morelos, Guerrero] 0.706*** 0.656*** 0.145 14.49 29.42
4 Groups: [Barberi] [Morelos, Guerrero] [Nayarit] [Baja, Costa, Central, Texas] 0.719*** 0.533*** 0.398** 39.80 28.12
4 Groups: [Barberi] [Baja, Costa] [Morelos, Guerrero, Nayarit] [Central, Texas] 0.709*** 0.531*** 0.381** 38.07 29.06
5 Groups: [Barberi] [Morelos, Guerrero] [Nayarit] [Central, Texas] [Baja, Costa] 0.703*** 0.516*** 0.385** 38.47 29.75
5 Groups: [Barberi] [Morelos, Guerrero, Nayarit] [Central, Texas] [Baja] [Costa] 0.709*** 0.518*** 0.397** 39.69 29.07
5 Groups: [Barberi] [Morelos, Guerrero, Nayarit] [Central] [Texas] [Baja, Costa] 0.695*** 0.555*** 0.316* 31.55 30.47
6 Groups: [Barberi] [Morelos, Guerrero, Nayarit] [Central] [Texas] [Baja] [Costa] 0.695*** 0.538*** 0.339* 33.93 30.50
Open in a new tab
*

P < 0.05

**

P < 0.01

***

P < 0.001

Mismatch distributions were generated for geographically widespread groups with relatively large sample sizes, including the Barberi, Central, Guerrero and Texas clades (Fig. 3). All four clades showed multimodal patterns (Barberi raggedness index r = 0.2, P = 0.51; Central r = 0.12, P = 0.63; Guerrero r = 0.13, P = 0.56; Texas r = 0.06, P = 0.33), indicating stable or slowly declining populations and long-term demographic stability (Rogers and Harpending, 1992). Non-significant Fu’s Fs tests (Table 3) for these clades support sequence evolution consistent with the expectation of selective neutrality and stable demographic history for the clades.

Fig. 3.

Fig. 3

Open in a new tab

Mismatch distributions computed for the combined 12S and 16S mitochondrial dataset for the barberi, central, guerrero and texas clades. Dashed line, observed distribution of pairwise differences; solid line, expected distribution in the case of sudden demographic expansion.

4. Discussion

4.1 Phylogeny and taxonomy of Hypopachus

Our results provide the first comprehensive analysis of Hypopachus relationships based on molecular data. The well-supported clades identified in this study are geographically structured and correspond to previously recognized species of Hypopachus (Frost, 2011; Nelson 1973a, 1974). We confirm the recognition of disjunct highland populations as H. barberi, which is consistent with current taxonomy. The well-supported basal split among lowland samples of Hypopachus variolosus from western Mexico (WMX) and elsewhere in Mesoamerica (MML), and the uncorrected p mitochondrial sequence divergence between these clades (3.5–6.5%) suggests that the species diversity of Hypopachus may be underestimated. These sequence divergence values are lower than the divergences (6.9–7.1%) we noted between sister taxa of Gastrophryne, but comparable to or exceed mitochondrial gene divergences noted for valid New World anuran species of Incilius (Mendelson et al., 2005), Dendrobates (Vences et al., 2003), Rana (Pfeiler and Markow, 2008), and salamander species in the genera Bolitoglossa (Parra-Olea et al., 2004) and Pseudoeurycea (Parra-Olea et al., 2005).

Nelson (1974) examined adult morphology, color pattern, and male advertisement call data and found no differences among the lowland populations of Hypopachus. Using the data from Nelson (1973a, 1974), we grouped male advertisement call data according to the geographic clades recovered in our phylogenetic analysis, but this call data alone failed to distinguish the clades (Fig. 4). Examination of color photographs, field note remarks, x-rays, and preserved adult specimens (EG, unpubl.data) showed substantial variation (Fig. 5), and failed to identify any morphological or color pattern characters that could unequivocally distinguish the lowland clades of Hypopachus. Preliminary examination of tadpoles from different clades (ENS, unpubl.data) suggest there might be morphological differences that can distinguish the lineages, but available sample sizes are too low to determine whether these differences are intraspecific variation or diagnostic characters among valid evolutionary lineages. Because there is currently no morphological, mensural or meristic character to diagnose the western Mexico (WMX) clade from other lowland populations (MML) of Hypopachus, we choose to uphold current taxonomy and consider all lowland populations as one species, H. variolosus.

Fig. 4.

Fig. 4

Open in a new tab

Graphical representation of frequencies and fundamental/harmonic intervals of well-supported clades of Hypopachus. Data are summarized from Nelson (1973a, 1974).

Fig. 5.

Fig. 5

Open in a new tab

Photographs of selected populations of Hypopachus in life, showing extensive color pattern variation. (A) H. barberi (UTA A-50963) from El Quiché, Guatemala; (B) H. variolosus (UTA A-60367) from Guerrero, Mexico; (C) H. variolosus (UTA A-60368) from Morelos, Mexico; (D) H. variolosus (UTA A-60393) from Petén, Guatemala.

4.2 Outgroup relationships

Our well-supported relationship ((Hypopachus + Gastrophryne) Elachistocleis)) is consistent with the studies of Frost et al. (2006), Van Bocxlaer et al. (2006), and Van der Meijden et al. (2007), including the latter authors’ reanalyzed molecular data of Frost et al. (2006). Our placement of a well-supported, monophyletic Chiasmocleis as sister to the Hypopachus/Gastrophryne/Elachistocleis lineage is also consistent with the nuclear-gene phylogeny of Van der Meijden et al. (2007).

The distinct placement of Gastrophryne usta in our analyses was unexpected. Only one previous study has focused on the evolutionary genetics of the genus Gastrophryne (Makowsky et al., 2009), but most of their sampling was limited to G. carolinensis (the type species of the genus) in the eastern United States, which is remarkable for its lack of genetic diversity, possibly the consequence of a recent bottleneck. The uncorrected p mitochondrial sequence divergence between G. usta and the G. carolinensis/olivacea clade (14.3%) is comparable to divergences between both lineages of Gastrophryne to either Hypopachus (9.5–14.6%) or Elachistocleis (12.7–13.8%). Gastrophryne usta is known from open tropical deciduous formations, marshes, savannas, and dry forest from central Sinaloa, Mexico to El Salvador on the Pacific versant, and Veracruz to Oaxaca, Mexico on the Atlantic versant (Nelson 1972a; Köhler et al., 2006; Frost, 2011). Nelson (1972b) and Köhler et al. (2006) provided accounts of this species’ ecology, morphometrics, and natural history. Nelson (1972a) and Nelson and Altig (1972) noted that G. usta is distinguished from its congeners by the possession of paired metatarsal tubercles (single in congeners) in adults and tadpoles with well-developed hind limbs, presumably an adaptation for xeric habitats. Nelson and Altig (1972:382) noted that Hypopachus also have two metatarsal tubercles, but tadpoles in this genus are distinguished from all Gastrophryne by having scalloped or papillate labial flap margins and relatively longer, ventrally convergent or overlapping labial flaps. Compared to its congeners, Nelson (1972a, 1973b) also noted a markedly lower harmonic interval (70–130 Hz [range includes data from Fouquette and Rossman, 1963]) in the male advertisement call for G. usta (155–280 Hz in other species, call of G. pictiventris not quantified [Nelson, 1972c, 1973b; Savage, 2002]), and a salient difference in calling behavior—G. usta calls from concealed locations under leaves, grass, or trash, either at the edge of water or as much as 10 m away, whereas all other Gastrophryne species typically call in water with the body floating freely, but submerged from the axillae posteriorly. Because our outgroup sampling was not designed to test Gastrophryninae relationships, it would be premature to suggest a taxonomic rearrangement for G. usta, but future studies should examine more samples of this taxon in an expanded phylogenetic study of gastrophrynines.

Our strong support for the inclusion of Phrynomantis in the Gastrophyninae and moderate support (81% MP, 89% ML, 1.0 BI) for a sister relationship of Phrynomantis microps to Chiasmocleis is a novel finding. Other studies either recovered Phrynomantis in a well-supported and separate lineage from Gastrophryninae (Frost et al., 2006; Van Bocxlaer et al., 2006), or as an unresolved basal microhylid lineage (Van der Meijden et al., 2007), a paradox that the latter authors explained as “a biological polytomy, due to a fast initial radiation.”

4.3 Historical biogeography

A South American origin for the ancestor of Hypopachus was suggested by Van der Meijden et al. (2007), who commented that Hypopachus and Gastrophryne were firmly nested within their South American gastrophrynine clade, and their divergence estimate for the Hypopachus/Gastrophryne lineage was 17 (± 4 standard deviation) mya. Several fossils of Gastrophryne, some identified to extant species, are known in North America—Gastrophryne carolinensis and G. olivacea have been identified from Pleistocene (< 2 mya) deposits in Florida (Holman, 1965, 1996) and Texas (Holman, 1963), respectively. The presence of a Gastrophryne cf. carolinensis ilium fossil from Hemmingfordian Miocene deposits of Florida (Auffenberg, 1956; Holman, 1965) suggests the microhylid invasion of North America occurred at least 18 mya. This predates the mid-Miocene (ca. 15 mya) island-arc system that might have made contact with South America (Duellman, 2001) to allow a land-based invasion of the Gastrophryne-like ancestor from South America. It is possible that Gastrophryne could have rafted to Central America and further north before the Pliocene (ca. 3.5–5 mya) contact between North and South America via the Panama isthmus—this hypothesis seems likely given the Florida fossil, the basal position of the northern species of Gastrophryne relative to G. usta and Hypopachus, the Van der Meijden divergence estimate of 17 mya, and the increasing number of studies that have demonstrated transoceanic dispersal in amphibians (e.g., Heinicke et al., 2007; Measey et al., 2007; Vences, 2004). Because many microhylids (including Hypopachus and Gastrophryne) are semi-fossorial burrowers with secretive habits and brief reproductive seasons, these frogs may be especially well-suited for trans-oceanic transport inside floating flotsam.

Given the lack of morphological, color pattern, and call data that can be used to diagnose our genetically defined lineages of Hypopachus, the relatively modest sequence divergence between lowland sister clades, and the late Miocene/early Pliocene divergence estimates, it is likely that Hypopachus lineages formed relatively recently in geological time. However, given the high haplotype diversity, negative Fu’s Fs values, and multimodal distribution shapes in the examined mismatch distributions, it is also likely that the clades have not experienced secondary contact since their initial divergence. Some features of the distribution, habitat preference, and phylogenetic position of Hypopachus in this study are consistent with known events of Middle American geological history. Campbell (1999) noted that it is likely that many montane species in Middle America (e.g., H. barberi) did not exist until the late Miocene (ca. 7 mya) at the earliest because mountains were just forming at this time; the high elevations currently associated with Nuclear Central America did not occur until the Pliocene 2–3 million years later. Campbell (1999) and Duellman (2001) added that vertical oscillations in climate during the Pleistocene (≤ 1.8 mya) resulted in depression of vegetational belts (perhaps as much as 1000 m). These depressions likely led to connections between montane cloud forests, which would explain the close relationship among currently disjunct highland populations of H. barberi.

Another consistent link between Hypopachus phylogeny and geological history involves the slightly divergent, well-supported lowland lineage we recovered from our Baja Verapaz and Zacapa samples (Baja Clade), which are restricted to xeric interior Guatemalan valleys. Campbell (1999) discussed the formation of the Salamá Valley during the mid- to late Pliocene (ca. 2.5–3.6 mya) when the headwaters of the Río Polochic were captured by the Río Negro. Later in the Pleistocene (< 2 mya), renewed volcanic activity in southwestern Guatemala influenced local weather patterns to produce rain shadow conditions in this interior valley, again suggesting relatively recent events in the evolution of H. variolosus.

Explaining the disjunct distribution of the divergent lowland clades of Hypopachus in Mexico is more problematic, but there seem to be similarities with distribution and phylogeographic patterns in other amphibian and reptile groups. There is an interesting parallel to the pattern we found in Hypopachus and the distributions of sister taxa in the monotypic hylid genera Diaglena and Triprion—Diaglena occurs in the Pacific lowlands of Mexico from Sinaloa to Oaxaca, whereas Triprion inhabits the xeric habitats of the Yucatán Peninsula (Duellman, 2001; Frost, 2011). It is likely that a common ancestor to each hylid genus was confined to the subhumid habitats of Middle America and once inhabited a more widespread area, but subsequently was confined to relict populations during changes in aridity from glacial oscillations of the Pleistocene. In a study of the Neotropical snake genus Leptodeira, Daza et al. (2009) recovered similar patterns of genetic divergence in Pacific versant Mexico populations of L. maculata and L. annulata cussiliris. These authors explained this pattern as a possible consequence of Pleistocene sea level fluctuations in the Isthmus of Tehuantepec, which may have isolated vertebrate populations in the lowlands of Mexico. A second, older explanation from the snake study is either the formation of main river basins or Miocene climatic changes, which is consistent with our divergence estimates for Hypopachus.

  • We present a phylogenetic study of the Middle American microhylid frog genus Hypopachus.

  • Results confirm the placement of Hypopachus and Gastrophryne as sister taxa.

  • Gastrophryne was strongly supported as paraphyletic.

  • Phrynomantis was recovered as sister to Chiasmocleis, rendering gastrophrynines as paraphyletic.

  • Our molecular data support the continued recognition of two species in the genus Hypopachus.

Acknowledgments

The work for this project occurred at the University of Kansas (KU), University of Richmond (UR), Villanova University (VU), and the University of Texas at El Paso (UTEP). The senior author thanks Linda Trueb, Bill Duellman, Rafe Brown, Ed Wiley, Bruce Lieberman, Oliver Komar, Omar Torres-Carvajal (KU), Ken Yanek (UR), Aaron Bauer and Todd Jackman (VU), Max Shpak (UTEP) and Tony Gamble (University of Minnesota) for useful comments, assistance with preliminary data analyses, and access to laboratory resources. Tissue samples were kindly provided by Oliver Komar (formerly of KU), David Wake (MVZ), Robert Murphy (ROM), Michael R. J. Forstner (Texas State University), Andrew J. Crawford (formerly of the Smithsonian Tropical Research Institute), Oscar Flores-Villela (UNAM), Jonathan Campbell (UTA), Carl S. Lieb and Dominic I. Lannutti (UTEP), and Josiah H. Townsend (University of Florida). The senior author acknowledges a grant from the Kentucky Herpetological Society. The molecular work for this study was supported primarily by NSF award DEB-0342918 to Rafael O. de Sá and W. Ronald Heyer and also by Omar Hernandez and Ana Betancourt of the UTEP DNA Analysis Core Facility (NIH grant #5G12RR008124). We thank two anonymous reviewers for their comments on the manuscript. For work in Mexico we thank E. Beltrán Sánchez, J. C. Blancas-Hernández, A. Carbajal-Saucedo, R. Aguilar, U. O. García Vázquez, L. Canseco-Marquez, C. M. Sheehy, III, R. C. Jadin, G. A. Acevedo and O. Flores-Villela for providing support and permits (SEMARNAT). Assistance in Guatemala was provided by M. E. Acevedo, M. V. Centeno, C. L. Guirola, J. Monzón-Sierra, and permit officials (CONAP). Work in Honduras was possible thanks to J. Ferrari-Castro (ORPRAH), A. Sosa, K. Lima de Zapata and K. Calidonio (IHT), J. H. Malone, M. Sasa-Marín, C. J. Franklin, C. S. Castañeda, A. J. Crawford, and Ing. L. Cortés (COHDEFOR), for providing permits (CI-MP-012-2005). Fieldwork was supported by the National Science Foundation (DEB-0416160 to ENS and J. A. Campbell; and DEB-0613802, 9705277, 061382, to JAC), the Texas Advanced Research Program (grant 003656-001 to JAC), the Wildlife Conservation Society (to ENS), the Audubon Society (to ENS), and the Instituto Bioclon (to ENS).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Contributor Information

Eric N. Smith, Email: e.smith@uta.edu.

Rafael O. de Sá, Email: rdesa@richmond.edu.

References

  1. Auffenberg W. Remarks on some Miocene anurans from Florida, with a description of a new species of Hyla. Breviora. 1956;(52):1–11. [Google Scholar]
  2. Bandelt HJ, Forster P, Röhl A. Median-joining networks for inferring intraspecific phylogenies. Mol Biol Evol. 1999;16:37–48. doi: 10.1093/oxfordjournals.molbev.a026036. [DOI] [PubMed] [Google Scholar]
  3. Brandley M, Schmitz A, Reeder TW. Partitioned Bayesian analyses, partition choice, and the phylogenetic relationships of scincid lizards. Syst Biol. 2005;54:373–390. doi: 10.1080/10635150590946808. [DOI] [PubMed] [Google Scholar]
  4. Campbell JA. Distribution patterns of amphibians in Middle America. In: Duellman WE, editor. Patterns of Distribution of Amphibians: A Global Perspective. The Johns Hopkins University Press; Baltimore, Maryland: 1999. pp. 111–210. [Google Scholar]
  5. de Carvalho AL. A preliminary synopsis of the genera of American microhylid frogs. Occ Pap Mus Zool Univ Michigan. 1954;555:1–19. [Google Scholar]
  6. Castoe TA, Doan TM, Parkinson CL. Data partitions and complex models in Bayesian analysis: the phylogeny of gymnophthalmid lizards. Syst Biol. 2004;53:448–469. doi: 10.1080/10635150490445797. [DOI] [PubMed] [Google Scholar]
  7. Daza JM, Smith EN, Páez VP, Parkinson CL. Complex evolution in the Neotropics: The origin and diversification of the widespread genus Leptodeira (Serpentes: Colubridae) Mol Phylogenet Evol. 2009;53:653–667. doi: 10.1016/j.ympev.2009.07.022. [DOI] [PubMed] [Google Scholar]
  8. Donnelly MA, de Sá RO, Guyer C. Description of the tadpoles of Gastrophryne pictiventris and Nelsonophryne aterrima (Anura: Microhylidae), with a review of morphological variation in free-swimming microhylid larvae. Amer Mus Novit. 1990;2976:1–19. [Google Scholar]
  9. Douglas ME, Douglas MR, Schuett GW, Beck DD, Sullivan BK. Conservation phylogenetics of helodermatid lizards using multiple molecular markers and a supertree approach. Mol Phylogen Evol. 2010;55:153–167. doi: 10.1016/j.ympev.2009.12.009. [DOI] [PubMed] [Google Scholar]
  10. Drummond AJ, Rambaut A. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evol Biol. 2007;7:214. doi: 10.1186/1471-2148-7-214. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Duellman WE. The Hylid Frogs of Middle America. 2. Vol. 2. Society for the Study of Amphibians and Reptiles; Ithaca, New York: 2001. [Google Scholar]
  12. Duellman WE, Trueb L. Biology of Amphibians. The Johns Hopkins University Press; Baltimore, Maryland: 1986. [Google Scholar]
  13. Excoffier L, Smouse PE, Quattro JM. Analysis of molecular variance inferred from metric distance among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics. 1992;131:479–491. doi: 10.1093/genetics/131.2.479. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Excoffier L, Lischer HEL. Arlequin suite ver 3.5: A new series of programs to perform population genetics analyses under Linux and Windows. Mol Ecol Res. 2010;10:564–567. doi: 10.1111/j.1755-0998.2010.02847.x. [DOI] [PubMed] [Google Scholar]
  15. Felsenstein J. Confidence limits on phylogenies: an approach using the bootstrap. Evolution. 1985;39:783–791. doi: 10.1111/j.1558-5646.1985.tb00420.x. [DOI] [PubMed] [Google Scholar]
  16. Fouquette MJ, Jr, Rossman DA. Noteworthy records of Mexican amphibians and reptiles in the Florida State Museum and the Texas Natural History Collection. Herepetologica. 1963;19:185–201. [Google Scholar]
  17. Frost DR. Amphibian Species of the World: an Online Reference. Version 5.5. Museum of Natural History; New York, USA: 2011. Jan 31, Electronic Database accessible at http://research.amnh.org/vz/herpetology/amphibia/American. [Google Scholar]
  18. Frost DR, Grant T, Faivovich J, Bain RH, Haas A, Haddad CFB, de Sá RO, Channing A, Wilkinson M, Donnellan SC, Raxworthy CJ, Campbell JA, Blotto BL, Moler P, Drewes RC, Nussbaum RA, Lynch JD, Green DM, Wheeler WC. The amphibian tree of life. Bull Amer Mus Nat Hist. 2006;(297):1–371. [Google Scholar]
  19. Frost DR, Hillis DM. Species in concept and practice: herpetological applications. Herpetologica. 1990;46:87–104. [Google Scholar]
  20. Fu YX. Statistical tests of neutrality of mutations against population growth, hitchhiking, and background selection. Genetics. 1997;147:915–925. doi: 10.1093/genetics/147.2.915. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Harpending HC. Signature of ancient population growth in a low-resolution mitochondrial DNA mismatch distribution. Hum Biol. 1994;66:591–600. [PubMed] [Google Scholar]
  22. Harpending HC, Batzer MA, Gurven M, Jorde LB, Sherry ST. Genetic traces of ancient demography. Proc Natl Acad Sci. 1998;95:1961–1967. doi: 10.1073/pnas.95.4.1961. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Heinicke MP, Duellman WE, Hedges SB. Major Caribbean and Central American frog faunas originated by ancient oceanic dispersal. Proc Nat Acad Sci, USA. 2007;104:10092–10097. doi: 10.1073/pnas.0611051104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Hillis DM, Moritz C, Mable B, editors. Molecular Systematics. Sinauer Associates, Inc; Sunderland, Massachusetts: 1996. [Google Scholar]
  25. Ho SYW. Calibrating molecular estimates of substitution rates and divergence times in birds. J Avian Biol. 2007;38:409–414. [Google Scholar]
  26. Hoegg S, Vences M, Brinkmann H, Meyer A. Phylogeny and comparative substitution rates of frogs inferred from sequences of three nuclear genes. Mol Biol Evol. 2004;21:1188–1200. doi: 10.1093/molbev/msh081. [DOI] [PubMed] [Google Scholar]
  27. Holman JA. Late Pleistocene amphibians and reptiles of the Clear Creek and Ben Franklin local faunas of Texas. J Grad Res Center, South Meth Univ. 1963;31:152–167. [Google Scholar]
  28. Holman JA. Early Miocene anurans from Florida. Quart J Florida Acad Sci. 1965;28:68–82. [Google Scholar]
  29. Holman JA. The large Pleistocene (Sangamonian) herpetofauna of the Williston IIIA site, north-central Florida. Herp Nat Hist. 1996;4:35–47. [Google Scholar]
  30. Huelsenbeck JP, Ronquist F. MrBayes: Bayesian inference of phylogenetic trees. Bioinformatics. 2001;17:754–755. doi: 10.1093/bioinformatics/17.8.754. [DOI] [PubMed] [Google Scholar]
  31. Köhler G, Veselý M, Greenbaum E. The Amphibians and Reptiles of El Salvador. Krieger Publishing Company; Malabar, Florida: 2006. [Google Scholar]
  32. Leaché AD, Reeder T. Molecular systematics of the fence lizard (Sceloporus undulatus): a comparison of parsimony, likelihood, and Bayesian approaches. Syst Biol. 2002;51:44–68. doi: 10.1080/106351502753475871. [DOI] [PubMed] [Google Scholar]
  33. Leviton AE, Gibbs RH, Jr, Heal E, Dawson CE. Standards in herpetology and ichthyology. Part I Standard symbolic codes for institutional resource collections in herpetology or ichthyology. Copeia. 1985;1985:802–832. [Google Scholar]
  34. Loader SP, Gower DJ, Howell KM, Doggart N, Rödel MO, Clarke BT, de Sá RO, Cohen BL, Wilkinson M. Phylogenetic relationships of African microhylid frogs inferred from DNA sequences of mitochondrial 12S and 16S rRNA genes. Organ Diver Evol. 2004;4:227–235. [Google Scholar]
  35. Maddison DR, Maddison WP. MacClade: analysis of phylogeny and character evolution. Vers. 4.08. Sinauer; Sunderland, Massachusetts: 2005. [Google Scholar]
  36. Makowsky R, Chesser J, Rissler LJ. A striking lack of genetic diversity across the wide-ranging amphibian Gastrophryne carolinensis (Anura: Microhylidae) Genetica. 2009;135:169–183. doi: 10.1007/s10709-008-9267-5. [DOI] [PubMed] [Google Scholar]
  37. McCranie JR, Wilson LD. The Amphibians of Honduras. Society for the Study of Amphibians and Reptiles; Ithaca, New York: 2002. [Google Scholar]
  38. Measey GJ, Vences M, Drewes RC, Chiari Y, Melo M, Bourles B. Freshwater paths across the ocean: molecular phylogeny of the frog Ptychadena newtoni gives insights into amphibian colonization of oceanic islands. J Biogeog. 2007;34:7–20. [Google Scholar]
  39. Mendelson JR, III, Williams BL, Sheil CA, Mulcahy DG. Systematics of the Bufo coccifer Complex (Anura: Bufonidae) of Mesoamerica. Sci Pap Nat Hist Mus Univ Kansas. 2005;38:1–27. [Google Scholar]
  40. Near TJ, Meylan PA, Shaffer HB. Assessing concordance of fossil calibration points in molecular clock studies: an example using turtles. Amer Nat. 2004;165:137–146. doi: 10.1086/427734. [DOI] [PubMed] [Google Scholar]
  41. Nelson CE. Systematic studies of the North American microhylid genus Gastrophryne. J Herpetol. 1972a;6:111–137. [Google Scholar]
  42. Nelson CE. Gastrophryne usta (Cope), Two-spaded narrow-mouthed toad. Cat Amer Amph Rept. 1972b;123:1–2. [Google Scholar]
  43. Nelson CE. Gastrophryne pictiventris (Cope), Southern narrow-mouthed toad. Cat Amer Amph Rept. 1972c;135:1–2. [Google Scholar]
  44. Nelson CE. Systematics of the Middle American upland populations of Hypopachus (Anura: Microhylidae) Herpetologica. 1973a;29:6–17. [Google Scholar]
  45. Nelson CE. Mating calls of the Microhylinae: descriptions and phylogenetic and ecological considerations. Herpetologica. 1973b;29:163–176. [Google Scholar]
  46. Nelson CE. Further studies on the systematics of Hypopachus (Anura: Microhylidae) Herpetologica. 1974;30:250–274. [Google Scholar]
  47. Nelson CE, Altig R. Tadpoles of the microhylids Gastrophryne elegans and G. usta. Herpetologica. 1972;28:381–383. [Google Scholar]
  48. Parker HW. A Monograph of the Frogs of the Family Microhylidae. British Museum (Natural History); London, United Kingdom: 1934. [Google Scholar]
  49. Parra-Olea G, García-París M, Wake DM. Molecular diversification of salamanders of the tropical American genus Bolitoglossa (Caudata: Plethodontidae) and its evolutionary and biogeographic implications. Biol J Linn Soc. 2004;81:325–346. [Google Scholar]
  50. Parra-Olea G, Garcia-Paris M, Papenfuss TJ, Wake DM. Systematics of the Pseudoeurycea bellii (Caudata: Plethodontidae) species complex. Herpetologica. 2005;61:145–158. [Google Scholar]
  51. Patou ML, Wilting A, Gaubert P, Esselstyn JA, Cruaud C, Jennings AP, Fickel J, Veron G. Evolutionary history of the Paradoxurus palm civets—a new model for Asian biogeography. J Biogeogr. 2010;37:2077–2097. [Google Scholar]
  52. Pfeiler E, Markow TA. Phylogenetic relationships of leopard frogs (Rana pipiens complex) from an isolated coastal mountain range in southern Sonoroa, Mexico. Mol Phylogenet Evol. 2008;49:343–348. doi: 10.1016/j.ympev.2008.06.011. [DOI] [PubMed] [Google Scholar]
  53. Posada D. jModeltest: phylogenetic model averaging. Mol Biol Evol. 2008;25:1253–1256. doi: 10.1093/molbev/msn083. [DOI] [PubMed] [Google Scholar]
  54. Rambaut A, Drummond AJ. Tracer. 2007;1.5 See http://beast.bio.ed.ac.uk/Tracer.
  55. Rogers AR, Harpending H. Population growth makes waves in the distribution of pairwise genetic differences. Mol Biol Evol. 1992;9:552–569. doi: 10.1093/oxfordjournals.molbev.a040727. [DOI] [PubMed] [Google Scholar]
  56. Ronquist F, Huelsenbeck JP. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics. 2003;19:1572–1574. doi: 10.1093/bioinformatics/btg180. [DOI] [PubMed] [Google Scholar]
  57. Savage JM. The Amphibians and Reptiles of Costa Rica: A Herpetofauna Between Two Continents, Between Two Seas. University of Chicago Press; Chicago, Illinois: 2002. [Google Scholar]
  58. Shaffer HB, Fellers GM, Voss SR, Oliver JC, Pauly GB. Species boundaries, phylogeography and conservation genetics of the red-legged frog (Rana aurora/draytonii) complex. Molec Ecol. 2004;13:2667–2677. doi: 10.1111/j.1365-294X.2004.02285.x. [DOI] [PubMed] [Google Scholar]
  59. Shimodaira H. An approximately unbiased test of phylogenetic tree selection. Syst Biol. 2002;51:492–508. doi: 10.1080/10635150290069913. [DOI] [PubMed] [Google Scholar]
  60. Shimodaira H, Hasegawa M. CONSEL: for assessing the confidence of phylogenetic tree selection. Bioinformatics. 2001;17:1246–1247. doi: 10.1093/bioinformatics/17.12.1246. [DOI] [PubMed] [Google Scholar]
  61. Stamatakis A. RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics. 2006;22:2688–2690. doi: 10.1093/bioinformatics/btl446. [DOI] [PubMed] [Google Scholar]
  62. Stamatakis A, Blagojevic F, Nikolopoulos D, Antonopoulos C. Exploring new search algorithms and hardware for phylogenetics: RAxML meets the IBM cell. J VLSI Signal Processing. 2007;48:271–286. [Google Scholar]
  63. Stamatakis A, Hoover P, Rougemont J. A rapid bootstrap algorithm for the RAxML web servers. Syst Biol. 2008;57:758–771. doi: 10.1080/10635150802429642. [DOI] [PubMed] [Google Scholar]
  64. Swofford DL. PAUP*: Phylogenetic Analysis Using Parsimony (* and other methods), Version 4.0b10. Sinauer; Sunderland, Massachusetts: 2002. [Google Scholar]
  65. Tamura K, Dudley J, Nei M, Kumar S. MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) Software Version 4.0. Mol Biol Evol. 2007;24:1596–1599. doi: 10.1093/molbev/msm092. [DOI] [PubMed] [Google Scholar]
  66. Van Bocxlaer I, Roelants K, Biju SD, Nagaraju J, Bossuyt F. Late Cretaceous vicariance in Gondwanan amphibians. PLoS ONE. 2006;1:e74. doi: 10.1371/journal.pone.0000074. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Van der Meijden A, Vences M, Meyer A. Novel phylogenetic relationships of the enigmatic brevicipitine and scaphiophrynine toads as revealed by sequences from the nuclear Rag-1 gene. Proc Roy Soc Lond B (Supplement) 2004;271:S378–S381. doi: 10.1098/rsbl.2004.0196. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Van der Meijden A, Vences M, Hoegg S, Boistel R, Channing A, Meyer A. Nuclear gene phylogeny of narrow-mouthed toads (Family: Microhylidae) and a discussion of competing hypotheses concerning their biogeographic origins. Mol Phylogenet Evol. 2007;44:1017–1030. doi: 10.1016/j.ympev.2007.02.008. [DOI] [PubMed] [Google Scholar]
  69. Vences M. Phylogeography of Ptychadena mascareniensis suggests transoceanic dispersal in a widespread African-Malagasy frog lineage. J Biogeog. 2004;31:593–601. [Google Scholar]
  70. Vences M, Kosuch J, Boistel R, Haddad CFB, La Marca E, Loetters S, Veith M. Convergent evolution of aposematic coloration in Neotropical poison frogs: A molecular phylogenetic perspective. Organ Div Evol. 2003;3:215–226. [Google Scholar]
  71. Wild ER. New genus and species of Amazonian microhylid frog with a phylogenetic analysis of New World genera. Copeia. 1995;1995:837–849. [Google Scholar]
  72. Zarza E, Reynoso VH, Emerson BC. Diversification in the northern neotropics: mitochondrial and nuclear DNA phylogeography of the iguana Ctenosaura pectinata and related species. Mol Ecol. 2008;17:3259–3275. doi: 10.1111/j.1365-294X.2008.03826.x. [DOI] [PubMed] [Google Scholar]
  73. Zhang M, Rao D, Yang J, Yu G, Wilkinson JA. Molecular phylogeography and population structure of a mid-elevation montane frog Leptobrachium ailaonicum in fragmented habitat of southwest China. Mol Phylog Evol. 2010;54:47–58. doi: 10.1016/j.ympev.2009.10.019. [DOI] [PubMed] [Google Scholar]
  74. Zweifel RG. A new genus and species of microhylid frog from the Cerro de la Neblina region of Venezuela and a discussion of relationships among New World microhylid genera. Amer Mus Novit. 1986;2863:1–24. [Google Scholar]

RESOURCES