ArticlePDF Available

Abstract and Figures

Phylogenetic relationships among the species of true frogs (Rana) from North, South, and Central America were investigated based on the sequences of approximately 2 kb from the mitochondrial genome, sampled from most of the described species, as well as eight undescribed species. This analysis, combined with previous studies of the phylogeny of New World Rana, served as the basis for a revised classification of the group. The American species of Rana are not monophyletic; the western North American Amerana is more closely related to the R. temporaria group of Eurasia (together, these frogs form the group Laurasiarana). The remaining species from the Americas form the monophyletic group Novirana, which includes: R. sylvatica; Aquarana (the R. catesbeiana group); Ranula (the R. palmipes group, including the mostly upland Levirana species and the mostly lowland Lithobates species); Torrentirana (the R. tarahumarae group, or Zweifelia, plus R. sierramadrensis), Stertirana (the R. montezumae group, or Lacusirana, plus R. pipiens), Nenirana (the R. areolata group), and Scurrilirana (most of the southern and tropical leopard frogs). The mitochondrial sequences supported many of the previous hypotheses of relationships of New World Rana, although there were some differences involving the placement of the species R. pipiens, R. sierramadrensis, and R. sylvatica. Parametric bootstrap analyses indicated significant support for the relationships inferred from the mtDNA sequences, and rejected the previous hypotheses of relationships for these three species.
Content may be subject to copyright.
Molecular Phylogenetics and Evolution 34 (2005) 299–314
www.elsevier.com/locate/ympev
1055-7903/$ - see front matter 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.ympev.2004.10.007
Phylogeny of the New World true frogs (Rana)
David M. Hillis¤, Thomas P. Wilcox
Section of Integrative Biology and Center for Computational Biology and Bioinformatics, University of Texas, Austin, TX 78712, USA
Received 22 July 2004; revised 14 October 2004
Abstract
Phylogenetic relationships among the species of true frogs (Rana) from North, South, and Central America were investigated
based on the sequences of approximately 2 kb from the mitochondrial genome, sampled from most of the described species, as well as
eight undescribed species. This analysis, combined with previous studies of the phylogeny of New World Rana, served as the basis for
a revised classiWcation of the group. The American species of Rana are not monophyletic; the western North American Amerana is
more closely related to the R. temporaria group of Eurasia (together, these frogs form the group Laurasiarana). The remaining spe-
cies from the Americas form the monophyletic group Novirana, which includes: R. sylvatica; Aquarana (the R. catesbeiana group);
Ranula (the R. palmipes group, including the mostly upland Levirana species and the mostly lowland Lithobates species); Torrentir-
ana (the R. tarahumarae group, or Zweifelia, plus R. sierramadrensis), Stertirana (the R. montezumae group, or Lacusirana, plus R.
pipiens), Nenirana (the R. areolata group), and Scurrilirana (most of the southern and tropical leopard frogs). The mitochondrial
sequences supported many of the previous hypotheses of relationships of New World Rana, although there were some diVerences
involving the placement of the species R. pipiens, R. sierramadrensis, and R. sylvatica. Parametric bootstrap analyses indicated signiW-
cant support for the relationships inferred from the mtDNA sequences, and rejected the previous hypotheses of relationships for
these three species.
2004 Elsevier Inc. All rights reserved.
Keywords: Phylogeny; Rana; Frogs; ClassiWcation; Evolution
1. Introduction
The approximately 250 extant species of true frogs
(Rana) are found throughout much of the world, with
the major exceptions being the polar regions, most of
Australia, and the temperate regions of South America.
About one-quarter of the species of Rana are found in
the Americas, with the largest concentration in the
southern United States and Mexico. In the New World,
species of Rana are found from Alaska and Canada
south throughout the continental United States and all
of Middle America to northwestern Peru on the west
side of the Andes and to eastern Brazil and northern
Bolivia on the east side of the Andes. Collectively, these
species are found in almost all of the major biotic prov-
inces that are inhabited by frogs—tundra, temperate
coniferous and deciduous forests, grasslands, deserts,
brackish-water marshes, freshwater streams and lakes,
semitropical cloud forests, and tropical rain forests.
Because one or another species of Rana is common
throughout much of the world, several species of Rana
have served as research subjects for a broad array of
studies in evolution, ecology, behavior, development,
genetics, and physiology. Given the large amount of
comparative biological information available among
species of Rana, this group has great potential for plac-
ing a wide range of biological studies in an evolutionary
framework, as long as phylogenetic estimates for the
group are available. This study examines the phylogeny
and diversiWcation of Rana in the New World based on
mitochondrial DNA, and tests the signiWcance of
diVerences between current and previous estimates of
New World Rana phylogeny. Our goal is to provide a
*Corresponding author. Fax: +1 512 471 3878.
E-mail address: dhillis@mail.utexas.edu (D.M. Hillis).
300 D.M. Hillis, T.P. Wilcox / Molecular Phylogenetics and Evolution 34 (2005) 299–314
comprehensive and well-supported phylogeny for the
group to facilitate comparative studies on the biology of
these frogs.
At least three species of New World Rana are thought
to have become extinct in historic times, and several of
the remaining species have undergone serious declines
and are threatened with extinction. Many of the threat-
ened species occur in the western United States and
Mexico. In addition, several New World tropical species
have not yet been described. This study includes all of
the undescribed species of which we are aware. A second
goal of this study is to provide the phylogenetic back-
ground for a revised classiWcation of the New World
Rana. The phylogeny will also be useful for determining
priorities for conservation of the biological diversity of
New World Rana.
The New World species are thought to form a mono-
phyletic group together with the Rana temporaria species
group of Eurasia (Case, 1978a; Farris et al., 1983; Hillis
and Davis, 1986; Post and Uzzell, 1981; Wallace et al.,
1973), and perhaps other Eurasian species groups as well
(Dubois, 1992). These previous studies have supported
the placement of the root of the New World Rana tree
between the Rana boylii plus Rana temporaria groups in
one clade, and the remainder of the New World Rana in
the sister clade. Therefore, we used the R. boylii and R.
temporaria groups as our outgroups to the remaining
species. We included one species in the R. temporaria
group in our study, but have otherwise restricted our
analysis to the New World species (including most of the
species in the R. boylii group). Previous studies of the
phylogeny of New World Rana (Case, 1978a; Hillis and
Davis, 1986; Hillis et al., 1983; Wallace et al., 1973) have
not been comprehensive in the taxa sampled, and the
phylogenetic estimates have been based on relatively
small data sets. In this study, we attempted to include all
extant species of New World Rana (described species, as
well as new, but as yet undescribed, species) and exam-
ined approximately 2kb of DNA sequences from three
mitochondrial genes (large and small subunits of the
ribosomal RNA genes, and the valine tRNA gene). A
few described extant species were not available to us for
study, but these few species are each thought to be
closely related to other species that we did include in our
analyses.
2. Materials and methods
2.1. Taxon sampling
We collected sequence data from 58 species of Rana,
including all the extant described species from the New
World except Wve species of leopard frogs that are
closely related to other included species, and two mem-
bers of the Rana boylii group that are variously treated
as species or subspecies. One of the leopard frogs we
failed to include was R. megapoda; the sample we col-
lected to represent this species actually represents a
related but undescribed species. Hillis et al. (1983) origi-
nally included this specimen (as R. megapoda) in a study
of allozyme variation of the R. pipiens complex, but
Webb (1996) questioned this identiWcation and instead
referred the specimen to R. montezumae. However, our
data (including allozyme data presented by Hillis et al.,
1983, and the sequence data presented in this paper)
indicate that this specimen is not R. montezumae, but
instead represents a distinct, undescribed species that is
more closely related to R. chiricahuensis than it is to R.
montezumae. A second leopard frog species we did not
include is R. miadis, a species of questionable status
known only from Little Corn Island, Nicaragua, and
which is closely related to (and may be conspeciWc with)
R. taylori. Another species missing from our analysis, R.
brownorum, has been considered a subspecies of R. ber-
landieri, but recent evidence suggests that it should be
recognized as a distinct species (Zaldívar-Riverón et al.,
2004). Two additional species of leopard frogs have been
described recently: R. lemosespinali (Smith and Chiszar,
2003), a species closely related to, and until recently
included within, R. chirichahuensis; and R. chich-
icuahutla (Cuellar et al., 1996), a species known from a
single crater lake, and which was thought by Cuellar
et al. (1996) to be closely related to R. spectabilis. There
are also three recently extinct species (R. Wsheri, R. johni,
and R. pueblae) that were unavailable for analysis
(although some of the northern populations currently
referred to R. chiricahuensis may actually be R. Wsheri;
see discussion in Section 4.6). Within the R. boylii group,
R. pretiosa and R. luteiventris are now treated as distinct
species (Green et al., 1997), although traditionally they
have been considered subspecies; our geographic sam-
pling of this complex included only populations now
referred to R. luteiventris. Also, R. a. aurora and R. a.
draytonii have been considered subspecies, but a recent
analysis (ShaVer et al., 2004) suggests that they should be
recognized as distinct species (we only sampled R. a.
aurora). We also included samples of eight undescribed
species, whose descriptions are forthcoming, as well as
additional specimens for two species whose sequences
were previously deposited in GenBank, to conWrm the
identity of these sequences. Taxon names, collecting
localities, voucher numbers, and GenBank accession
numbers are given in Table 1.
2.2. Data collection
DNA was extracted from tissues (liver, muscle, and
blood) using standard phenol:chloroform extraction
(Hillis et al., 1996a) or the DNEasy kit (Qiagen).
Extracted DNA was resuspended in ddH2O or in 0.1 mM
Tris (pH 8.0) and quantiWed via gel electrophoresis and
D.M. Hillis, T.P. Wilcox / Molecular Phylogenetics and Evolution 34 (2005) 299–314 301
Table 1
Specimens examined, collecting localities, voucher numbers, and GenBank accession numbers
Species Voucher number and specimen locality data GenBank No.
Rana areolata KU 204370: USA: Kansas: Lyon: just S of Hartsford AY779229
Rana aurora MVZ 188960: USA: California: Del Norte: Kings Valley Rd., 2.4 mi N Hwy. 199 AY779196
Rana berlandieri JSF 1136: USA: TX: Hays: San Marcos AY779235
Rana blairi JSF 830: USA: Kansas: Douglas: Lawrence AY779237
Rana boylii MVZ 148929: USA: California: Lake: along Butts Creek, 0.4 mi NW Napa County line AY779192
Rana bwana QCAZ 13964: Ecuador: Prov. Loja: Río Alamor near Zapotillo AY779212
Rana capito TNHC 60195: USA: Florida: Marion: Archibold Biological Station AY779231
Rana cascadae MVZ 148946: USA: California: Shasta: Dersch Meadows AY779197
Rana catesbeiana GenBank sequence; locality unknown GB X12841
Rana catesbeiana DMH 84-R2: USA: Kansas: Douglas: Lawrence AY779206
Rana chiricahuensis KU 194442: Mexico: Durango: Río Chico at Mexico Hwy. 40 AY779225
Rana chiricahuensis KU 194419: USA: Arizona: Apache: Apache National Forest: Three Forks AY779226
Rana clamitans JSF 1118: USA: Missouri: Montgomery: 3 km W Danville AY779204
Rana dunni KU 194527: Mexico: Michoacan: Tintzuntzan, Lago Patzcuaro AY779222
Rana forreri KU 194581: Mexico: Sinaloa: 37.9 km S Escuinapa AY779233
Rana grylio MVZ 175945: USA: Florida: Leon: Tall Timbers Research Station, Lake Iamonia AY779201
Rana heckscheri MVZ 164908: USA: Florida: Gadsen-Leon: OverXow creek of Ochlocknee River at Hwy. S-12 AY779205
Rana juliani TNHC 60324: Belize: Cayo District: Little Vaqueros Creek AY779215
Rana luteiventris MVZ 225749: USA: Washington: Pend Oreille: Colville Natl. Forest; Flowery Trail Road, 6.1 mi E 49
Degrees Ski area
AY779193
Rana luteiventris MVZ 191016: USA: Montana: Lincoln: Dry Creek at Hwy. 56 AY779194
Rana macroglossa KU 195138: Mexico: Chiapas: 7.7 km SE San Cristobal de las Casas AY779242
Rana macroglossa UTA A-17185: Guatemala: Sololá: Panajachel, Lake Atitlan AY779243
Rana maculata KU 195258: Mexico: Oaxaca: Colonia Rodulfo Figueroa, 19 km NW Rizo de Oro AY779207
Rana magnaocularis KU 194592: Mexico: Sonora: Arroyo Hondo, 15.2 km N Nuri AY779239
Rana montezumae KU 195251: Mexico: Morelos: Lagunas Zempoala AY779223
Rana muscosa MVZ 149006: USA: California: Mono: Meadows below Levitt Lake, W side Sonora Pass AY779195
Rana neovolcanica KU 194536: Mexico: Michoacan: Zurumbueno AY779236
Rana okaloosae toe clip (released): USA:Florida: Santa Rosa: 5 km E Harold, Garnier Creek (collected by Paul Moler) AY779203
Rana omiltemana KU 195179: Mexico: Guerrero: Agua de Obispo AY779238
Rana onca LVT 3542: USA: Nevada: Clark: Blue Point Spring, Lake Mead AY779249
Rana palmipes AMNH A-118801: Venezuela: Prov. Amazonas: Neblina Base Camp on Río Mawarinuma AY779210
Rana palmipes KU 202896: Ecuador: Prov. Napo: Misahuallí AY779211
Rana palustris KU 204425: USA: Indiana: Washington: Cave Creek near Campbellsburg AY779228
Rana pipiens GenBank sequence; locality unknown GBX12841
Rana pipiens JSF 1119: USA: Ohio: Ottawa: Little Portage State Park AY779221
Rana psilonota KU 195119: Mexico: Jalisco: 2.4 km NW Tapalpa AY779217
Rana pustulosa KU 200776: Mexico: Sinaloa: 2.1 km NE Santa Lucia AY779220
Rana septentrionalis TNHC tissue collection: Canada: Ontario: Grey AY779200
Rana sevosa TNHC 60194: USA: Mississippi: Harrison AY779230
Rana sierramadrensis KU 195181: Mexico: Guerrero: Agua de Obispo, 24.2 mi S Chilpancingo AY779216
Rana spectabilis KU 195186: Mexico: Hidalgo: La Estanzuela (holotype) AY779232
Rana sphenocephela JSF 845: USA: Kansas: Cherokee AY779251
Rana s. utricularia USC 7448: USA: Florida: Loop Road, Big Cypress National Preserve AY779252
Rana subaquavocalis James Platz Collection (specimens destroyed, DNA in TNHC tissue collection): USA: Arizona: Cochise:
Ramsey Canyon
AY779227
Rana sylvatica MVZ 137426: USA: New York: Tompkins; Connecticut Hill, ca. 10 mi SW Ithaca AY779198
Rana sylvatica DMH 84-R43: USA: Missouri: St. Louis: Tyson Environmental Study Area AY779199
Rana tarahumarae KU 194596: Mexico: Sonora: 14.4 km E Yecora AY779218
Rana taylori TCWC 55963: Nicaragua: Zelaya: 2.5 mi NW Rama AY779244
Rana temporaria DMH 84-R1: Switzerland: Valais Canton: 1.8 km NNE Grand St. Bernard Pass AY779191
Rana tlaloci KU 194436: Mexico: Distrito Federal: Xochimilco (paratype) AY779234
Rana vaillanti KU 195299: Mexico: Oaxaca: 5.6 mi NE Tapanatepec AY779214
Rana vibicaria MVZ 149033: Costa Rica: Prov. San José: El Empalme AY779208
Rana virgatipes MVZ 175944: USA: Louisiana: De Soto Parish; Frierson AY779202
Rana warszewitschii JSF 1127: Panama AY779209
Rana yavapaiensis KU 194423: USA: Arizona: Greenlee: Apache National Forest at Juan Miller Crossing AY779240
Rana zweifeli KU 195310: Mexico: Oaxaca: 1.6 mi S Cuyotepej AY779219
Rana species 1 QCAZ 13219: Ecuador: Prov. Esmeraldas: 5 km W Durango AY779213
Rana species 2 KU 204420: Mexico: San Luis Potosí: Rodeo AY779224
Rana species 3 KU 194559: Mexico: Michoacan: 11.4 km E junction Mexico Hwy. 51 and 15 AY779250
(continued on next page)
302 D.M. Hillis, T.P. Wilcox / Molecular Phylogenetics and Evolution 34 (2005) 299–314
AMNH, American Museum of Natural History; DMH, David M. Hillis tissue collection, University of Texas, Austin; GB, GenBank sequence; JSF,
John F. Frost tissue collection, University of Texas, Austin; KU, Museum of Natural History, University of Kansas; LACM, Los Angeles County
Museum; LVT, University of Nevada-Las Vegas tissue collection; MVZ, Museum of Vertebrate Zoology, University of California, Berkeley; TCWC,
Texas Cooperative Wildlife Collection, Texas A&M University; TNHC, Texas Natural History Collection, University of Texas; QCAZ, Museo de
Zoología de la PontiWcia Universidad Católica del Ecuador; USC, University of Southern California collection, now housed at Los Angeles County
Museum; UTA, University of Texas at Arlington.
comparison to a known standard. Approximately, 2kb
of mitochondrial DNA, spanning the region from 12S
rRNA through 16S rRNA, including the intervening
valine-tRNA (positions 442–2400 of the Rana pipiens
mtDNA, GenBank Y10945), were PCR ampliWed using a
series of nested primers (Table 2). PCR products were
gel-puriWed and directly sequenced using Xuorescent
thermal-cycle sequencing and an ABI 377 automated
sequencer (Perkin–Elmer).
2.3. Phylogenetic analysis
DNA sequences, ranging in length from 1935 to
1962 bp, were obtained from 64 individuals representing
58 species. Each sequence was examined for the presence
of conserved secondary structural elements to ensure the
sequences were valid, coding mtDNA sequences.
Sequences were aligned using Clustal W (Thompson et al.,
1994) and manually adjusted to accommodate secondary
structural elements (Cannone et al., 2002; http://www.
rna.icmb.utexas.edu/). Positional homology was uncer-
tain for 38 bp, and these positions were excluded from the
Wnal analysis (http://www.treebase.org/; SN2065-7080).
Aligned sequences were analyzed using PAUP*
(v4.0b4-b8, SwoVord, 2000) under the maximum-likeli-
hood (ML) and parsimony (MP) criteria, and using a
parallel version of a genetic algorithm for maximum
likelihood (Brauer et al., 2002). For ML analyses, the
GTR + + PINVAR model of sequence evolution, with
four gamma-distributed rate categories, was chosen for
the analysis (Appendix A). This was the preferred model
for the data found using the stepwise likelihood-ratio
test procedure described by Posada and Crandall (2001).
Starting trees for maximum likelihood searches in
PAUP* were obtained by conducting 5000 parsimony
searches from diVerent stepwise-addition trees (using
random taxon addition), followed by TBR branch swap-
ping. From these 5000 searches, 10 of the best trees
under the parsimony criterion were selected as starting
trees for the maximum likelihood searches, and TBR
branch swapping (to completion) was used to search for
optimal trees under the likelihood criterion. We also
conducted eight independent analyses using the parallel
genetic algorithm (Brauer et al., 2002), starting from ran-
domly selected trees, under the GTR + +PINVAR
model. The best likelihood solution that we found with
each searching method (branch swapping in PAUP*
from 10 parsimony starting trees, and selection under
the genetic algorithm from eight samples of random
trees) was identical. Bayesian posterior probabilities
(bpp) were estimated for each branch of our best likeli-
hood tree using MrBayes (v3.0b4, Huelsenbeck and
Ronquist, 2001), also using the GTR+ +PINVAR
model. Four analyses, with Monte Carlo Markov chain
(MCMC) length of Wve million generations each, were
conducted. The log-likelihood scores were found to con-
sistently stabilize after 500,000 generations within and
among these four independent analyses. Therefore, the
initial 500,000 generations from each run were dis-
carded, and we sampled one out of every 50 generations
from the remaining 18 million generations (across all
four independent analyses) to calculate posterior proba-
bilities for each branch in the maximum likelihood tree.
If a bipartition’s posterior probability was 795%, it was
considered signiWcantly supported. The interpretation of
Bayesian posterior probabilities for branches in a phylo-
genetic analysis is well-deWned; they represent the proba-
bility that the corresponding clade is present in the true
tree, given the data examined, the likelihood model, and
the speciWed priors (Huelsenbeck and Rannala, 2004;
Larget and Simon, 1999). In contrast, there is no clear or
widely accepted interpretation of other commonly used
Table 1 (continued)
Species Voucher number and specimen locality data GenBank No.
Rana species 4 AMNH A-124167: Panama: Chiriquí: 9 km SSE El Volcán AY779245
Rana species 5 LACM 146764: Costa Rica: Heredia: Monte de la Cruz AY779246
Rana species 6 LACM 146810: Costa Rica: Puntarenas: near mouth of Rio Barranca, 10 km E Puntarenas AY779247
Rana species 7 KU 194492: Mexico: Jalisco: Contla AY779241
Rana species 8 KU 195346: Mexico: Puebla: Río Atoyac at Mexico Hwy. 190 AY779248
Table 2
Primers used to amplify and sequence the mtDNA 12S–16S region in
this study (see also Goebel et al., 1999)
Primers 12L1, 16Sh, 12Sm, and 16H1 were used in primary ampliWca-
tions. Other primers were used for sequencing
aReverse primers also used in sequencing.
Primer name Primer sequence (5!3)
12L1 AAAAAGCTTCAAACTGGGATTAGATACCCCA
CTAT
12Sm GGCAAGTCGTAACATGGTAAG
16ScaGTRGGCCTAAAAGCAGCCAC
16Sh GCTAGACCATKATGCAAAAGGTA
16SaaATGTTTTTGGTAAACAGGCG
16H1 CTCCGGTCTGAACTCAGATCACGTAGG
D.M. Hillis, T.P. Wilcox / Molecular Phylogenetics and Evolution 34 (2005) 299–314 303
measures of phylogenetic support, such as nonparamet-
ric bootstrap proportions or decay indices, and their use
in the context of statistical testing is not straightforward
(e.g., see Hillis and Bull, 1993; SwoVord et al., 1996;
Newton, 1996; Wilcox et al., 2002; and Huelsenbeck and
Rannala, 2004). Therefore, we prefer to use Bayesian
posterior probabilities over other measures of phyloge-
netic support that have no widely accepted statistical
interpretation.
2.4. Testing alternative hypotheses
Wherever our phylogenetic Wndings diVered from
existing hypotheses about the relationships of New
World Rana, we conducted parametric bootstrap tests
on the results to determine if the diVerences between the
competing hypotheses were signiWcantly diVerent (Gold-
man et al., 2000; Hillis et al., 1996b; Huelsenbeck et al.,
1996). We prefer parametric bootstrap tests over other
(nonparametric) approaches for testing a priori hypoth-
eses in phylogenetic analysis because of the greater
power of the parametric tests (Goldman et al., 2000;
Huelsenbeck et al., 1996). In each case, we constrained
our analysis to Wt the hypothesis that was to be tested,
and found the best tree that supported the hypothesis in
question. We then estimated parameters for a model of
evolution from the observed data, and used these param-
eters and the optimized tree topology to simulate 100
replicates of the model tree (using Seq-Gen version 1.2.5;
Rambaut and Grassly, 1997). We then analyzed each
replicate and compared the best-Wt tree to the best tree
that Wt the constraint of the null hypothesis. This proce-
dure allowed us to construct expected distributions of
the test statistic (in this case, the diVerence in tree length
under the parsimony criterion) under the assumption
that the null hypothesis was true. If the observed diVer-
ence in tree length (between our observed best-Wt tree
and the tree that best Wt the hypothesis under test) was
greater than 95% of the simulated values, then the null
hypothesis was rejected.
3. Results
3.1. Phylogenetic analysis
The best-Wt tree based on the maximum likelihood
analysis is shown in Fig. 1, with all branches that have
signiWcant support (posterior probability >95%) indi-
cated by asterisks. This tree has a log-likelihood score of
¡21812.64132. All of the PAUP* and genetic algorithm
searches found this tree or another tree of similar but
slightly lower score (¡21812.66941); this latter tree
diVered only in rearrangements of some of the weakly
supported clades within the tropical leopard frogs (Scur-
rilirana). The Bayesian analysis indicated signiWcant sup-
port for 47 out of 61 internal branches. All but one of the
14 branches that lacked signiWcant support united spe-
cies within recognized species groups: inside the R. boylii
group (one branch), within the R. catesbeiana group
(three branches), and within the R. pipiens group (nine
branches). The relationships among the recognized spe-
cies groups were all signiWcantly supported, with the sin-
gle exception of the sister-group relationship of the R.
tarahumarae and R. pipiens species groups (Fig. 1).
We found four shortest parsimony trees of length
4121; these four trees were far from optimal under the
likelihood criterion (log likelihood scores for these four
trees ranged from ¡21886.27 to ¡21903.92, or about 74–
91 log likelihood units from the maximum likelihood
estimate). These four trees diVer from the maximum like-
lihood tree shown in Fig. 1 in the following ways: (1) they
join the relatively long branches that lead to R. tempo-
raria and R. boylii together, and so do not support the
monophyly of the R. boylii group; (2) they join the rela-
tively long branches that lead to R. grylio and R. virgati-
pes together (within the R. catesbeiana group); (3) they
group the Venezuelan sample of R. palmipes (another
long-branch taxon) outside of the R. palmipes group, as
sister to the R. pipiens group; (4) they show the remain-
ing members of the R. palmipes group as paraphyletic
with respect to the R. tarahumarae plus R. pipiens
groups; and (5) they show minor rearrangements involv-
ing some of the weakly supported branches within the R.
pipiens group. With the exception of the rearrangements
within the R. pipiens group (which reXect weak support
under either criterion), all of the other diVerences
between the maximum likelihood and maximum parsi-
mony trees are consistent with long-branch attraction
problems in the parsimony tree (Felsenstein, 1978; Huel-
senbeck and Hillis, 1993). Because the parsimony trees
show signiWcantly worse Wt to the data compared to the
maximum likelihood tree when the details of the model
of sequence evolution are taken into account, we con-
sider the maximum likelihood solution shown in Fig. 1 to
be our best estimate of phylogeny for the group.
Given the general correspondence of morphology,
allozymes, immunology, and DNA sequences for many
of the major clades of New World Rana (Case, 1978a;
Hillis, 1987, 1988; Hillis and Davis, 1986; Hillis and de
Sá, 1988; Hillis et al., 1983, 1984; this study), we have
presented a phylogenetic classiWcation of this group in
Appendix B (summarized in Fig. 2). This classiWcation
preserves and deWnes previously named groups within
Rana, wherever these names are applicable. In some
cases, there have not been any formal names applied to
some well-supported groups, and in these cases we have
provided new clade names. Following the principles of
phylogenetic classiWcation (de Queiroz and Gauthier,
1990, 1992, 1994), the clade names presented in Appen-
dix B are all unranked (i.e., they are not assigned to cate-
gories such as section or subgenus). However, their
304 D.M. Hillis, T.P. Wilcox / Molecular Phylogenetics and Evolution 34 (2005) 299–314
hierarchical relationships are indicated by indenting and
also can be seen in Fig. 2. Nonetheless, under the rules of
the International Code of Zoological Nomenclature
(International Commission on Zoological Nomencla-
ture, 1999), any uninominal name of a “ƒ genus-group
division of a genus, even if it is proposed for a secondary
(or further) subdivision, is deemed to be a subgeneric
name even if the division is denoted by a term such as
‘section’ or ‘division’ƒ” (Art. 10.4). Therefore, all of the
clade names within Rana that are deWned in Appendix B
are subgenera under ICZN rules, even though the clades
are nested hierarchically within one another (Hillis et al.,
2001). We recommend that Rana still be the primary
clade name used with species epithets to promote
nomenclatural stability; the other clade names, in turn,
are useful for discussing historical groups of species
within Rana. Therefore, the species names in this paper
(together with the clade name Rana) are identical under
either traditional Linnean binomial nomenclature (as
binomials), or following Option M for phylogenetic spe-
cies names as suggested by Cantino et al. (1999).
The tree in Fig. 1 supports many of the traditional
species groups that have been recognized previously,
including the R. boylii group (named Amerana by
Dubois, 1992; see Appendix B for the deWnition of this
and other clade names used in this paper), the R. cates-
beiana group (Aquarana), the R. tarahumare group
(Zweifelia), the R. palmipes group (Ranula; but see below
for discussion of the relationships of R. sierramadrensis),
the R. pipiens complex (Pantherana), the R. montezumae
group (Lacusirana), and the R. areolata group (Nenir-
ana). The R. pipiens complex (now Pantherana) was
Fig. 1. Phylogenetic tree of New World Rana based on maximum likelihood analysis of rDNA sequences. The scale bar indicates divergence. The
numbers and letters on clades represent some of the traditionally recognized groups discussed in the text. Three species (R. pipiens, R. sierramadren-
sis, and R. sylvatica) do not group with their traditional species groups in our analysis; the placement of these three species was subjected to addi-
tional testing. Branches with two asterisks were supported by posterior probabilities of >99%; branches with one asterisk were supported by
posterior probabilities of >95%.
D.M. Hillis, T.P. Wilcox / Molecular Phylogenetics and Evolution 34 (2005) 299–314 305
divided into the and divisions by Hillis et al. (1983)
on the basis of allozyme variation. The division
contained Lacusirana and Nenirana, whereas the divi-
sion contained Scurrilirana and R. pipiens. In our analy-
sis of DNA sequences, the major exception to this
proposal is that R. pipiens appears to be the sister species
of Lacusirana, rather than closely related to (or even a
member of) Scurrilirana (Fig. 1). Without the inclusion
of R. pipiens, the support for Scurrilirana is strong (100%
Bayesian posterior probability). However, even a modi-
Wed alpha division that includes R. pipiens (Nenirana
plus Stertirana) appears paraphyletic on the optimal
tree, with Nenirana more closely related to Scurrilirana
than to Stertirana (although this latter relationship is
not signiWcantly supported).
3.2. Tests of previous hypotheses
Although our phylogenetic hypothesis and classiWca-
tion shown in Fig. 1 is broadly consistent with recent
summaries of the phylogeny of New World Rana based
on other evidence (Hillis, 1988; Hillis and Davis, 1986;
Hillis and de Sá, 1988; Hillis et al., 1983, 1984), there are
three signiWcant diVerences. These diVerences do not
depend on the optimality criterion selected, as they
appear in both the maximum likelihood and maximum
parsimony trees. The phylogenetic analysis of mitochon-
drial DNA supports each of the following:
1. Rana sylvatica as the sister-group of Aquarana, rather
than as the sister-group (or a part of) Laurasiarana
(contra Farris et al., 1980, 1983, based on analysis of
immunological data; and also contra Hillis and
Davis, 1986, based on analysis of nuclear ribosomal
DNA restriction sites).
2. Rana pipiens as the sister-group of Lacusirana, rather
than as a part of Scurrilirana (contra Hillis et al.,
1983, based on analysis of allozyme data).
3. Rana sierramadrensis as the sister-group of Zweifelia,
rather than as a part of Ranula (contra Hillis and de
Sá, 1988, based on morphological analysis).
Our parametric bootstrap tests of these three a priori
hypotheses (Fig. 3) showed that each could be rejected at
p< 0.05 based on the mitochondrial DNA sequences.
However, the support for R. sylvatica as the sister-group
to Aquarana (rather than within or sister to Laurasiar-
ana) was relatively weak, with the observed diVerence in
tree length just greater than needed to reject the previous
hypothesis (pD0.046). In contrast, the previous hypothe-
ses for the relationships of R. pipiens (p<0.002) and R.
sierramadrensis (p< 0.014) were rejected easily.
4. Discussion
4.1. Relationships of Rana sylvatica
The relationship of Rana sylvatica to other species of
the New World Rana has long been controversial. In
general appearance, this species resembles species of the
R. temporaria group (of Laurasiarana), and many
authors have long assumed that R. sylvatica was simply
a North American member of this otherwise Eurasian
species group. Likewise, Amerana of western North
America has been considered to be closely related to, or
even a part of, the R. temporaria group (Farris et al.,
1980, 1983). However, Case (1978a) suggested that R.
sylvatica was more closely related to the eastern and
tropical groups of North American Rana (what is now
Novirana) than to the western American Amerana or the
Eurasian R. temporaria group, based on immunological
comparisons. This conclusion was contested by Farris
et al. (1980, 1983). Post and Uzzell (1981) made addi-
tional immunological comparisons, and suggested that
the data best supported either a relationship to the east-
ern and tropical Rana (as suggested by Case, 1978a), or
Fig. 2. ClassiWcation of New World Rana. Names are deWned and dis-
cussed in Appendix B.
306 D.M. Hillis, T.P. Wilcox / Molecular Phylogenetics and Evolution 34 (2005) 299–314
else a placement of R. sylvatica as a sister-group to all
the other New World Rana plus the R. temporaria
group. Hillis and Davis (1986) supported the placement
of R. sylvatica with the R. temporaria group on the basis
of nuclear ribosomal RNA restriction sites.
Our optimal tree (Fig. 1) places Rana sylvatica as the
sister species to Aquarana. Although this placement is not
exactly like that proposed by Case (1978a), it is closer to
that proposal than any other previously suggested. The
relationship of R. sylvatica to Aquarana was supported in
100% of the trees sampled in the Bayesian analysis, as well
as in the most-parsimonious trees. Moreover, we could
reject the alternative placement of R. sylvatica with Lau-
rasiarana (either as a sister-group, or embedded within
Laurasiarana) in the parametric bootstrap analysis
(pD0.046). However, given the apparent conXict with pre-
vious studies and other data sets, we have refrained from
suggesting any changes to the classiWcation of Rana that
involve the placement of R. sylvatica.
4.2. Amerana (The Rana boylii group)
This group is well-supported by our likelihood analy-
sis, although the relationship of Amerana to the R. temp-
oraria group of Eurasia requires further investigation
and taxon sampling (Green, 1986b). In our analysis
(with only a single representative of the R. temporaria
group, however), Amerana is monophyletic. Although
almost all previous studies have supported the mono-
phyly of Amerana (see summary presented by Macey
et al., 2001), there has been less agreement on the rela-
tionships among the species within this group (Case,
1978a,b; Farris et al., 1980, 1983; Green, 1986a,b; Hillis
and Davis, 1986; Macey et al., 2001; Post and Uzzell,
1981; ShaVer et al., 2004; Wallace et al., 1973; Zweifel,
1955). The close relationship between R. pretiosa and R.
luteiventris has been reXected in the treatment of these
two taxa as subspecies until recently. Other relationships
in the group, however, are in conXict among data sets,
including the relationship of the two taxa variously
treated as distinct species or as subspecies of R. aurora
(R. aurora and R. draytonii). Allozyme data presented by
Case (1978b) and Green (1986b), as re-analyzed by
Macey et al. (2001), support the sister-group relationship
between R. aurora and R. cascadae shown in Fig. 1, but
this relationship is contradicted by some karyotypic,
immunological, and other allozyme analyses (Case,
1978a,b; Green, 1986a). Analysis by Macey et al. (2001)
Fig. 3. Results of the parametric bootstrap tests of three hypotheses (see text). In each case, the graphs show the diVerence (in parsimony score)
between the null and test hypothesis for each of 500 replicates (histogram), as well as the observed diVerence in score for the empirical data (arrow).
In each case, the null hypothesis can be rejected at p<0.05.
D.M. Hillis, T.P. Wilcox / Molecular Phylogenetics and Evolution 34 (2005) 299–314 307
of mitochondrial genes (diVerent from the genes we sam-
pled) support a tree very similar to the tree we found for
the Amerana species. They found strong support for a
monophyletic Amerana, weak support for a relationship
between R. boylii and R. luteiventris (reported as R.
pretiosa), and strong support for a clade consisting of R.
muscosa, R. aurora, and R. cascadae (as on our tree). The
only apparent diVerence between our two trees is that we
found strong support for a sister-group relationship
between R. aurora and R. cascadae, whereas Macey et al.
(2001) reported weak support for a sister-group relation-
ship between R. cascadae and R. muscosa. However, this
diVerence appears to result from the diVerent samples we
used for “R. aurora.” Our sample represents R. a. aurora,
whereas the Macey et al. (2001) sample represented R. a.
draytonii. ShaVer et al. (2004) found recently that R.
aurora and R. draytonii are not each other’s closest rela-
tives, and should be recognized as distinct species. More-
over, ShaVer et al. (2004) found that R. aurora (sensu
stricto) and R. cascadae are sister species (as also sup-
ported by our data), but could not resolve the trichot-
omy among this pair of species, R. draytonii, and R.
muscosa. Thus, taking the diVerent taxa sampled across
each of the studies into account, our tree for Amerana
does not appear to be in conXict with the results of the
other DNA sequence studies on this group (Macey et al.,
2001; and ShaVer et al., 2004). None of our studies are in
conXict with the tree (((pretiosa, luteiventris) boylii)
(((aurora, cascadae) muscosa) draytonii)), although none
of the DNA studies to date include samples of all of
these taxa.
Rana pretiosa (together with R. luteiventris) some-
times has been supported as the sister-group to the
remaining species of Amerana (weak support from allo-
zyme data and DNA restriction analysis of nuclear
rRNA genes), although our analysis weakly supports a
relationship of these taxa to R. boylii (Fig. 1). Macey et
al. (2001) also found weak support for a relationship
between R. pretiosa/luteiventris and R. boylii. Green
(1986b) suggested some form of “mosaic evolution” to
account for the discrepancies of relationships of this
group supported among the various data sets. However,
there does not appear to be any strong conXict between
any of the data sets presented to date and the tree for
Amerana shown in Fig. 1, once taxon sampling issues
involving the R. aurora/draytonii complex are taken into
account.
4.3. Aquarana (The Rana catesbeiana group)
Our data strongly support the monophyly of the
Aquarana. This group is also supported on the basis of
immunological data (Case, 1978a), morphology (Hillis,
1985), allozymes (Pytel, 1986), and nuclear rDNA
restriction sites (Hillis and Davis, 1986). In addition to
the monphyly of Aquarana, our analysis strongly sup-
ports a close relationship between R. clamitans and R.
okaloosae, and a group that consists of these two species
plus R. catesbeiana and R. heckscheri (Figs. 1 and 2).
Each of these relationships was also supported in a more
extensive analysis of this group by Austin et al. (2003).
Austin et al. (2003) showed that R. okaloosae is phyloge-
netically embedded within R. clamitans, and suggested a
very recent origin for R. okaloosae.
4.4. Ranula (The Rana palmipes group)
Hillis and Davis (1986) reported a single nuclear
rDNA restriction site that supported the monophyly of
the Rana palmipes group, and additional sites that sup-
ported the monophyly of the lowland members of the
group and the montane members of the group, exclusive
of R. sierramadrensis. Hillis and de Sá (1988) reviewed
the R. palmipes group based on morphological analyses
of adults and tadpoles. They divided the group into two
subgroups: one containing the lowland species (R.
bwana, R. palmipes, and R. vaillanti) and the other con-
taining the montane/upland species (R. juliani, R. macu-
lata, R. sierramadrensis, R. vibicaria, and R.
warszewitschii). Our analysis supports this arrangement,
with two major exceptions. First, R. sierramadrensis does
not appear to be a member of Ranula. Instead, the
mtDNA data strongly place it as sister to the R. tarahu-
marae group (Zweifelia). Second, R. juliani appears to be
a part of the lowland clade (Lithobates), rather than the
upland clade (Levirana). Two of the morphological char-
acters that Hillis and de Sá (1988) used to place R. juliani
with the species of Levirana were features of the tadpole
mouthparts (the presence of larval marginal teeth, and
an increase in the number of upper rows of teeth from 4
to 5–7). Both of these features are correlated with tad-
poles in montane streams, and it appears that the tad-
poles of R. juliani are convergent with the tadpoles of
species Levirana (and divergent from the pond tadpoles
of the more closely related species of Lithobates). If the
mitochondrial DNA tree is correct, there is also conver-
gence in two characters of coloration: the supralabial
stripe and dark face mask. The convergence in morphol-
ogy of R. juliani is strongest with R. maculata, and
indeed the Wrst reports of R. juliani (from the Maya
Mountains of Belize; Henderson and Hoevers, 1975;
Lee, 1976) identiWed specimens of this species as R. mac-
ulata. However, the mitochondrial DNA data reject a
close relationship among the montane species, and
instead suggest that R. juliani is a secondarily montane
species otherwise related to the widespread lowland spe-
cies in Lithobates.
A number of authors have noted that the putative
widespread species Rana palmipes probably consists of
many species (Cochran and Goin, 1970; Fowler, 1913;
Günther, 1900; Hillis and de Sá, 1988). Hillis and de Sá
(1988) removed R. vaillanti from the synonymy of R.
308 D.M. Hillis, T.P. Wilcox / Molecular Phylogenetics and Evolution 34 (2005) 299–314
palmipes, and described R. bwana from southwestern
Ecuador and northwestern Peru. Ranula gollmeri was
described by Peters (1859) based on a juvenile specimen
from Caracas, Venezuela, although the species was
almost immediately considered a synonym of Rana
palmipes, even by Peters (Boulenger, 1920). The name
Rana gollmeri is available if the northern South Ameri-
can populations are determined to be distinct. We have
not sampled suYciently to answer this question,
although our sample of “R. palmipes” from Venezuela is
considerably more divergent from our Ecuadorian sam-
ple of R. palmipes than are many other recognized sister
species of Rana in our analysis (the Venezuelan sample
does not even group within Ranula in the parsimony
analyses, although this is likely an artifact of its high
degree of overall divergence). We are also aware of an
additional undescribed species in Lithobates, distributed
in northwestern Ecuador and western Colombia (species
1 in Figs. 1 and 2). This species was referred to R. vail-
lanti by Hillis and de Sá (1988), but they noted that the
tadpoles of species 1 are darkly pigmented and easily dis-
tinguished from those of R. vaillanti. Our phylogenetic
analysis suggests that this species is more closely related
to R. palmipes than to R. vaillanti. Thus, the clade Litho-
bates may contain six (or more) essentially parapatric
species: R. vaillanti throughout much of lowland Central
America, north to the state of Veracruz, Mexico; R. juli-
ani in the Maya Mountians of Belize (and presumably
adjacent Guatemala); R. palmipes in the Amazon River
basin; R. bwana along the dry PaciWc coast of northwest-
ern Peru and southwestern Ecuador; possibly R. gollmeri
in northern South America (if this species is supported
as distinct by additional analysis); and an undescribed
species (species 1) along the wet PaciWc coast of north-
western Ecuador and western Colombia. Interestingly,
the geographic ranges of these species mirror the ranges
of other widespread lowland complexes of frogs, such as
the Physalaemus pustulosus group (Cannatella et al.,
1998).
4.5. Torrentirana (The Rana tarahumarae group)
Members of the Rana tarahumarae group are found
throughout the Sierra Madre Occidental and Sierra
Madre del Sur of western Mexico, and historically north
to southern Arizona in the United States. Two species
(R. johni and R. pueblae) were also formerly found in the
Sierra Madre Oriental in the states of San Luís Potosí,
Hidalgo, and Puebla, Mexico, but neither species has
been seen since the 1970s and both are believed to be
extinct (Hillis et al., 1984). Populations of R. tarahuma-
rae have also rapidly declined, and the species has been
extirpated from its former range in southern Arizona
(Webb, 2001). In addition, populations of R. sierramad-
rensis disappeared from many former localities in Oax-
aca and Guerrero, Mexico by the mid-1980s, and some
former populations of R. pustulosa in Sinaloa, Mexico
disappeared by the mid-1990s (DMH, personal observa-
tion). The species called “southern R. tarahumarae” by
Hillis et al. (1984) was described as R. psilonota by Webb
(2001). Currently, R. zweifeli and R. psilonota may be the
only species in this group with relatively widespread and
healthy populations.
All members of this species group lack vocal sacs and
slits, have reduced or absent external tympana, and no
calls have been recorded for any of the species. However,
there are also no published observations of breeding
among any of the species in this group. Given that the
species in this group are highly aquatic and typically
occupy montane streams and plunge pools, it is likely
that calls, if they are produced at all, are produced below
the surface of the water.
4.6. Pantherana (The Rana pipiens complex)
The leopard frogs (Pantherana) have a complex sys-
tematic history (Hillis, 1988). Many of the currently rec-
ognized species were once thought to represent a single,
widespread species that ranged from Canada to Panama
(Moore, 1944). However, many of these species were
found to occur sympatrically, with little or no hybridiza-
tion (Littlejohn and Oldham, 1968; Moore, 1975; Pace,
1974). In addition, the species that were once placed into
the wide ranging “R. pipiens” are not each others’ closest
relatives. The species have since been found to be mor-
phologically, behaviorally, phylogenetically, physiologi-
cally, and ecologically distinct (Hillis, 1988). However,
there are still a large number of taxonomic problems
within the group. Several species (especially in Mexico
and Central America) remain to be described (Fig. 1; see
also Zaldívar-Riverón et al., 2004), and other currently
recognized species appear to be conspeciWc. In particu-
lar, R. neovolcanica and R. tlaloci are extremely closely
related and may be conspeciWc, although this may prove
diYcult to study because of the disjunct range of R. tla-
loci and the fact that these isolated populations may now
be extinct because of the growth of Mexico City (Hillis
and Frost, 1985). Also closely related are R. onca and R.
yavapaiensis, as well as R. sevosa and R. capito. Each of
these pairs of species are more similar in their mtDNA
sequences than are R. s. sphenocephala and R. s. utricu-
laria, which are currently recognized as subspecies.
Additional study is needed to clarify the status of these
closely related taxa. In addition, the recognition of R.
subaquavocalis as distinct from R. chiricahuensis is not
well supported. On the other hand, the populations of
leopard frogs from the Mogollon Rim in Arizona that
are currently recognized as R. chiricahuensis are mor-
phologically distinct from R. chiricahuensis in southern
Arizona, New Mexico, and Mexico, and may be refer-
able to R. Wsheri (a species described from southern
Nevada, and considered extinct by many authors). Rana
D.M. Hillis, T.P. Wilcox / Molecular Phylogenetics and Evolution 34 (2005) 299–314 309
Wsheri appears to have been closely related to the Mogol-
lon Rim populations of “R. chiricahuensis” based on
morphological similarity, and the name R. Wsheri may be
applicable to these Mogollon Rim leopard frogs. Our sam-
ple of R. chiricahuensis from Arizona is from the Mogollon
Rim, and therefore perhaps should be referred to R. Wsheri.
However, we have followed the current taxonomic practice
of referring to these frogs as R. chiricahuensis, pending
detailed analysis of the problem.
Although some of the currently described species of
Pantherana appear to be very closely related to one
another (such as the R. onca/R. yavapaiensis species pair,
the R. neovolcanica/R. tlaloci species pair, and the R. sev-
osa/R. capito species pair), there are still a number of
undescribed leopard frog species in Mexico and Central
America that are relatively distantly related to any
described species (e.g., species 2–8 in Fig. 1). More thor-
ough sampling and analysis is needed to determine the
status and distribution of many of these taxa, however.
Hillis et al. (1983) recognized two major divisions of
Pantherana (then the R. pipiens complex), which they
informally termed the alpha and beta divisions. Their
division consisted of two species groups, the R. montezu-
mae group (here named Lacusirana, Fig. 2 and Appendix
B) and the R. areolata group (here named Nenirana, Fig. 2
and Appendix B). The other leopard frogs were placed in
the division, which was further divided into the R. pipi-
ens group (not supported in our analysis) and the R. ber-
landieri group (essentially Scurrilirana of Fig. 2 and
Appendix B, but now also including R. blairi and R. sphe-
nocephala). Our results diVer from those of Hillis et al.
(1983) most clearly in the placement of R. pipiens, which
Hillis et al. (1983) considered to be a member of the divi-
sion. However, the mtDNA sequence analysis strongly
rejects a relationship of R. pipiens to Scurrili- rana (Fig. 3),
and instead places R. pipiens as the sister species of Lacu-
sirana (together these taxa comprise Stertirana).
The phylogenetic position of R. pipiens has bearing
on the evolutionary reconstruction of advertisement call
evolution in leopard frogs. The advertisement calls of
many species of Pantherana are highly complex, and
consist of many distinct elements. These elements of the
advertisement calls have been described as mating trills
or snores, chuckles, grunts, and grinds (e.g., Larson,
2004; Mecham, 1971; Pace, 1974; Schmidt, 1968). The
“mating trill” or “mating call” element (called a snore by
Larson, 2004) has been associated with mate attraction
by most authors, whereas other elements have been sug-
gested as having a territorial or aggressive function
(Frost and Platz, 1983; Littlejohn and Oldham, 1968;
Mecham, 1971; Pace, 1974; Schmidt, 1968). Larson
(2004) emphasized that these functions have not been
conWrmed experimentally, although Oldham (1974) did
show some female response to the “mating call” in R.
sphenocephala. In addition, the “mating call” element of
the advertisement call is the only element that is pro-
duced by all species of Pantherana during breeding cho-
ruses, and some species are not known to produce the
other elements. Therefore, this element seems likely to
function in female attraction, although it is possible that
other elements may also contribute to this role in some
species. All of the species of Stertirana (including R. pipi-
ens) and Nenirana have a “snore-like” mating call (see
Fig. 4, and Mecham, 1971; Schaaf, 1971; and Altig and
Lohoefener, 1983 for examples). Larson (2004) provided
an explicit deWnition of the snore type of call; in brief, a
snore consists of a rapid, relatively uniform pulse rate,
with a long series of continuous pulses (typically 30–80
in the various species) that are modulated in amplitude
and frequency. In contrast, the species of Scurrilirana all
have some form of “chuckle-like” mating call, that
superWcially resembles some of the other elements of the
advertisement call of species such as R. pipiens. The
“chuckle-like” calls of species of Scurrilirana consist of a
Fig. 4. Audiospectrograms of the “mating call” component of the advertisement calls of four species of Pantherana (adapted and redrawn from Frost
and Platz, 1983). The calls of the top two species (R. chiricahuensis and R. pipiens) are examples of the snore-type call pattern, and the calls of the
lower two species (R. blairi and R. yavapaiensis) are examples of the chuckle-type call pattern.
310 D.M. Hillis, T.P. Wilcox / Molecular Phylogenetics and Evolution 34 (2005) 299–314
series of groups of pulses of variable number (typically
4–12), each group with a slower pulse rate compared to a
snore, and greater spacing between successive groups
than between successive pulses (see examples in Fig. 4).
As in a snore, these pulses are often modulated in fre-
quency and amplitude, and the calls are sometimes lik-
ened to the sound of rapid human laughter. The chuckle
calls of Scurrilirana nonetheless vary widely among spe-
cies in pulse rate, pulse number, dominant frequency,
and spacing of note sequences. Based on the phylogeny
shown in Fig. 1, the long, continuous snore-like mating
call is ancestral for Pantherana, but has been modiWed
into a chuckle-like mating call in species of Scurrilirana.
However, Rana pipiens has been demonstrated to have a
complex call that often consists of both a snore (identi-
Wed by most authors as the mating call) as well as
chuckle-like elements and grunts (Larson, 2004). In con-
trast, only the snore has been reported in the advertise-
ment call of most species of Lacusirana and Nenirana.
The presence of the diverse elements in the advertise-
ment call of R. pipiens, combined with the phylogenetic
location of this species as the sister-species to Lacusirana
(Fig. 1), suggests the possibility that diVerent elements of
the ancestral call of Pantherana may have been selected
to function in mate attraction in the various species of
this group. The various advertisement calls of the species
of Pantherana are among the most complex and diverse
calls of any anurans, and this group has undergone rapid
speciation and diversiWcation (Fig. 1). Together, these
facts indicate that this would be an ideal group for the
study of call evolution as it relates to reproductive isola-
tion.
5. Conclusions
The phylogenetic estimate shown in Fig. 1 provides
an opportunity for the comparative study of many
aspects of the biology of species of New World Rana.
The species in this group are highly diverse in ecology,
physiology, and behavior, and many of the species are
common enough to be ideal subjects for intensive bio-
logical study. As an example, the group as a whole
would serve as an excellent model system for the study
of many aspects of call evolution in frogs. There is con-
siderable variation in the types and diversity of calls that
are produced among the major clades, with many species
exhibiting complex calls that appear to have several
diVerent functions. Within some of the clades, there are
many closely related species that have recently split, and
calls appear to be playing an important role in speciation
(Littlejohn and Oldham, 1968). The morphological
structures that are associated with call production and
detection (such as vocal sacs, vocal slits, and external
tympana) are highly variable across species groups and
species, and in some cases even the presence or absence
of vocal sacs and slits is polymorphic within species
(Hayes and Kremples, 1986; Hillis and de Sá, 1988). Spe-
cies of New World Rana vary in other aspects of call
production, including whether the calls are produced in
air or under water, and how the calls function in mate
attraction. Given all the diversity (both intraspeciWc and
interspeciWc) of the advertisement calls and associated
morphological structures, this group should serve as an
ideal model system for the study of frog call evolution,
especially now that a relatively detailed and complete
phylogenetic estimate of the various species is available.
Acknowledgments
We thank Marty Badgett and Todd Schlenke for lab-
oratory assistance. The following people provided some
of the tissues samples, assisted in the Weld, assisted with
analyses, made suggestions on the manuscript, or pro-
vided other assistance for which we are grateful: Jona-
than Campbell, Luis Coloma, Brian Crother, William
Duellman, Darrel Frost, John Frost, Jacques Gauthier,
Harry Greene, JeV Jaeger, William Lamar, Julian Lee,
Carl Lieb, Paul Moler, James Platz, John Simmons,
David Wake, and Derrick Zwickl. Support from the
National Science Foundation and the John D. and Cath-
erine T. MacArthur Foundation (to DMH) is gratefully
acknowledged. The use of Phylocluster for computa-
tional analyses was made possible by an IGERT grant in
Computational Phylogenetics and Applications in Biol-
ogy from the National Science Foundation (DGE
0114387).
Appendix A. Estimated parameters from maximum
likelihood analysis
General time-reversible model (GTR)+ gamma rate
heterogeneity ()+ invariant sites (PINVAR):
Base frequencies: A: 0.347609, C: 0.212590, G:
0.149945, T: 0.289856.
Rate matrix: AC: 2.21659; AG: 8.16426; AT: 2.35934;
CG: 0.91908; CT: 16.52561; GT: 1.00000.
Shape parameter for gamma distribution: 0.560832.
Number of categories for discrete gamma approxima-
tion: 4.
Proportion of invariant sites: 0.338644.
Log likelihood of best tree: ¡21812.64132.
Appendix B. ClassiWcation of new world
Rana
Rana (Although we consider all of the groups below
to be part of Rana, the limits of this clade are beyond the
scope of this paper. Phylogenetic deWnition of Rana must
await a world-wide phylogenetic study of these frogs).
D.M. Hillis, T.P. Wilcox / Molecular Phylogenetics and Evolution 34 (2005) 299–314 311
I. Laurasiarana, new clade name. DeWnition: The clade
stemming from the most recent common ancestor of
Rana temporaria Linne 1758 and Rana aurora Baird
and Girard 1852. Etymology: From the name of the
supercontinent of Laurasia, and the Latin word rana,
meaning “frog,” in reference to the distribution of this
clade of frogs in North America, Europe, and Asia.
Type species: Rana aurora Baird and Girard 1852.
A. Unnamed clade. This group was called the subge-
nus Rana (within the genus Rana) by Dubois
(1992), but under phylogenetic nomenclature, one
name cannot be applied to two diVerent clades.
Because there has been no phylogenetic deWnition
of Rana to date, the name could be deWned to refer
to this clade. However, we prefer to use Rana for
the more inclusive group of frogs. Dubois (1992)
considered this clade to include 27 species in the
R. arvalis, R. chensinensis, R. graeca, R. japonica,
R. tagoi, and R. temporaria groups (he included R.
sylvatica in the latter). Here, Rana sylvatica is
explicitly excluded from this group, and we have
collected data only on R. temporaria from this
clade.
B. Amerana Dubois 1992 (converted clade name).
DeWnition: The clade stemming from the most
recent common ancestor of Rana aurora Baird and
Girard 1852, Rana boylii Baird 1854, Rana casca-
dae Slater 1939, Rana muscosa Camp 1917, and
Rana pretiosa Baird and Girard 1853. Content:
Includes the species designated in the deWnition, as
well as Rana luteiventris Thompson 1913 (if that
taxon is considered a separate species from Rana
pretiosa Baird and Girard 1852), and Rana dray-
toni Baird and Girard 1852 (if that taxon is consid-
ered a separate species from Rana aurora Baird
and Girard 1852). This clade has been informally
termed the R. boylii group by previous authors.
Comments: Dubois (1992) named the section
Amerana, which included two new subgenera:
Amerana (including the species R. boylii and R.
muscosa) and Aurorana (including the species R.
aurora, R. cascadae, R. draytoni, and R. pretiosa).
By our analysis (Fig. 1), all three of these names
apply to the same clade, if the groups are made
monophyletic.
II. Novirana, new clade name. DeWnition: The clade
stemming from the most recent common ancestor of
Rana catesbeiana Shaw 1802 and Rana pipiens Schre-
ber 1782. Etymology: From the Latin words novus,
meaning “new,” and rana, meaning “frog,” in refer-
ence to the New World distribution of this clade.
Content: Includes Rana sylvatica, and all the species
in Aquarana and Sierrana (see below). Type species:
Rana pipiens Schreber 1782.
A. Rana sylvatica Le Conte 1825. Our sequence data
place this species as the sister species to Aquarana.
However, given the conXicting results between
our data and previous analyses, we have not
named the clade that includes Rana sylvatica and
Aquarana.
B. Aquarana Dubois 1992 (converted clade name).
DeWnition: The clade stemming from the most
recent common ancestor of Rana catesbeiana
Shaw 1802, Rana clamitans Latreille 1802, Rana
grylio Stejneger 1901, Rana heckscheri Wright
1924, Rana okaloosae Moler 1985, Rana septentri-
onalis Baird 1854, and Rana virgatipes Cope1891.
The species used in the deWnition are those species
that were included by Dubois within this group.
Content: Includes the species listed as speciWers in
the deWnition. This clade has been informally
termed the R. catesbeiana group by previous
authors.
C. Sierrana Dubois 1992 (converted clade name).
DeWnition: The clade stemming from the most
recent common ancestor of Rana juliani Hillis
and de Sá, 1988, Rana maculata Brocchi 1877,
and Rana sierramadrensis Taylor 1939. Content:
Includes all the species in Ranula, Torrentirana,
and Pantherana (see below). Comments: This
clade was named by Dubois as a subgenus; he
speciWed the content as including the species that
are used as speciWers in the deWnition. Note that
this deWnition (based on the content speciWed by
Dubois) results a much larger group than that
envisioned by Dubois (1992).
1. Ranula Peters 1859 (converted clade name). DeWni-
tion: The clade stemming from the most recent com-
mon ancestor of Rana palmipes Spix 1824 and Rana
warszewitschii (Schmidt) 1857. Content: Includes the
species in Levirana and Lithobates (see below). Com-
ments: This clade was consistently recognized by E.
D. Cope (as a genus) from 1866 until his death; he
considered it to be the American counterpart of
Hylarana Cope (1866). This group has been infor-
mally termed the R. palmipes group by previous
authors.
a. Levirana Cope 1894 (converted clade name).
DeWnition: The clade stemming from the most
recent common ancestor of Rana maculata
Brocchi 1877 and Rana vibicaria (Cope) 1894.
Content: In addition to the species speciWed in the
deWnition, this clade also includes R. warszewit-
schii.
i. Rana maculata Brocchi 1877.
ii. Trypheropsis Cope 1868 (converted clade
name). DeWnition: The clade stemming from the
most recent common ancestor of Rana warsze-
witschii (Schmidt) 1857 and Rana vibicaria
(Cope) 1894. Content: Includes the species spec-
iWed in the deWnition.
312 D.M. Hillis, T.P. Wilcox / Molecular Phylogenetics and Evolution 34 (2005) 299–314
b. Lithobates Fitzinger 1843 (converted clade name).
DeWnition: The clade stemming from the most
recent common ancestor of Rana palmipes Spix
1824, Rana vaillanti Brocchi 1877, Rana bwana Hil-
lis and de Sá 1988, and Rana juliani Hillis and de
Sá 1988. Content: In addition to the species speci-
Wed in the deWnition, includes Rana gollmeri
(Peters) 1859 if that species is recognized as distinct
from R. palmipes, and an undescribed species from
northwestern Ecuador and western Colombia (spe-
cies 1 in Figs. 1, 2).
2. Torrentirana, new clade name. DeWnition: The clade
stemming from the most recent common ancestor of
Rana tarahumarae Boulenger 1917 and Rana sierra-
madrensis Taylor 1939. Etymology: From the Latin
words torrentis, referring to a swift or violent stream,
and rana, meaning “frog,” in reference to the typical
habitat of many of the species in this clade. Content:
Includes Rana sierramadrensis and the species within
Zweifelia (see below). Type species: Rana tarahuma-
rae Boulenger 1917.
a. Rana sierramadrensis Taylor 1939.
b. Zweifelia Dubois 1992 (converted clade name).
DeWnition: The clade stemming from the most
recent common ancestor of Rana tarahumarae
Boulenger 1917 and Rana zweifeli Hillis, Frost and
Webb 1984. Content: This clade includes the spe-
cies named in the deWnition, as well as Rana johni
Blair 1965, Rana pueblae Zweifel 1955, Rana
pustulosa Boulenger 1883, and Rana psilonota
Webb, 2001. At least two of the species (R. johni
and R. pueblae) are thought to be extinct, and at
least one of the remaining species (R. tarahumarae)
is extinct over much of its former range. This clade
has been informally termed the R. tarahumarae
group by previous authors.
3. Pantherana Dubois 1992 (converted clade name).
DeWnition: The clade stemming from the most recent
common ancestor of Rana pipiens Schreber 1782,
Rana montezumae Baird 1854, Rana palustris Le
Conte 1825, and Rana berlandieri Baird 1854. Con-
tent: All of the species in Stertirana, Nenirana, and
Scurrilirana (see below). This clade has been infor-
mally termed the R. pipiens complex by previous
authors.
a. Stertirana, new clade name. DeWnition: The clade
stemming from the most recent common ancestor
of Rana pipiens Schreber 1782 and Rana montezu-
mae Baird 1854. Etymology: From the Latin words
sterto, meaning “snore,” and rana, meaning “frog,”
in reference to the snore-like element of the adver-
tisement call of the frogs in this group. Content:
Includes R. pipiens and the species in Lacusirana
(see below). Type species: Rana montezumae Baird
1854.
i. Rana pipiens Schreber 1782.
ii. Lacusirana, new clade name. DeWnition: The
clade stemming from the most recent common
ancestor of Rana montezumae Baird 1854, R.
megapoda Taylor 1942, and Rana chiricahuensis
Platz and Mecham 1979. Etymology: From the
Latin words lacus, meaning “lake,” and rana,
meaning “frog,” in reference to the habitat of
most of the species in this group. Content: In
addition to the species named in the deWnition,
this clade contains Rana dunni Zweifel 1957,
Rana megapoda Taylor 1942, Rana subaquavo-
calis Platz 1993, Rana lemosespinali Smith and
Chiszar 2003, and a least one undescribed spe-
cies from the Mexican Plateau (species 2 in
Figs. 1, 2). In addition, we place Rana Wsheri
Stejneger 1893 in this clade based on morpho-
logical similarity. Type species: R. megapoda
Taylor 1942.
b. Nenirana, new clade name. DeWnition: The clade
stemming from the most recent common ancestor
of Rana areolata Baird and Girard 1852 and Rana
palustris Le Conte 1825. Etymology: From the
Latin words nenia, meaning “a funeral song,” and
rana, meaning “frog,” in reference to the low,
mournful advertisement call of the species in this
clade. Content: In addition to the species speciWed
in the deWnition, this clade includes Rana capito
LeConte 1855 and Rana sevosa Goin and Netting
1940. Hillis et al. (1983) informally termed this
clade the R. areolata group. Type species: Rana
areolata Baird and Girard 1852.
c. Scurrilirana, new clade name. DeWnition: The clade
stemming from the most recent common ancestor
of Rana berlandieri Baird 1854, Rana sphenocep-
hala Cope 1886, Rana forreri Boulenger 1883, R.
spectabilis Hillis and Frost 1985, Rana omiltemana
Gunther 1900, Rana taylori Smith 1959, and Rana
magnaocularis Frost and Bagnara 1976. Etymol-
ogy: From the Latin words scurrilis, meaning “jest-
ing,” and rana, meaning “frog,” in reference to the
advertisement calls of most of the species in this
clade, which sound like chuckling laughter. Con-
tent: In addition to the species speciWed in the deW-
nition, this clade includes Rana blairi Mecham,
Littlejohn, Oldham, Brown, and Brown 1973 Rana
chichicuahutla Cuellar, Méndez-DeLaCruz, and
Villagrán-Santa Cruz 1996, Rana macroglossa
Brocchi 1877, Rana miadis Barbour and Loveridge
1929, Rana neovolcanica Hillis and Frost 1985,
Rana onca Cope in Yarrow 1875, Rana tlaloci
Hillis and Frost 1985, Rana yavapaiensis Platz and
Frost 1984, and several undescribed species (spe-
cies 3–8) in Mexico and Central America. Hillis
et al. (1983) informally termed this clade the beta
division of the R. pipiens complex. Type species:
Rana berlandieri Baird 1854.
D.M. Hillis, T.P. Wilcox / Molecular Phylogenetics and Evolution 34 (2005) 299–314 313
References
Altig, R., Lohoefener, R., 1983. Rana areolata. Cat. Am. Amphibians
Reptiles 324, 1–4.
Austin, J.D., Lougheed, S.C., Moler, P., Boag, P.T., 2003. Phylogenetics,
zoogeography, and the role of vicariance and dispersal in the evolu-
tion of the Rana catesbeiana (Anura: Ranidae) species group. Biol.
J. Linn. Soc. 80, 601–624.
Boulenger, G.A., 1920. A monograph of the American frogs of the
genus Rana. Proc. Am. Acad. Arts Sci. 55, 413–480.
Brauer, M.J., Holder, M.T., Dries, L.A., Zwickl, D.J., Lewis, P.O., Hillis,
D.M., 2002. Genetic algorithms and parallel processing in maximum-
likelihood phylogeny inference. Mol. Biol. Evol. 19, 1717–1726.
Cannatella, D.C., Hillis, D.M., Chippindale, P., Weight, L., Rand, A.S.,
Ryan, M.J., 1998. Phylogeny of frogs of the Physalaemus pustulosus
species group, with an examination of data congruence. Syst. Biol.
47, 311–335.
Cannone, J.J., Subramanian, S., Schnare, M.N., Collett, J.R., D’ Souza,
L.M., Du, Y., Feng, B., Lin, N., Madabusi, L.V., Muller, K.M.,
Pande, N., Shang, Z., Yu, N., Gutell, R.R., 2002. The comparative
RNA web (CRW) site: an online database of comparative sequence
and structure information for ribosomal, intron, and other RNAs.
BioMed. Central Bioinformat. 3, 2.
Cantino, P.D., Bryant, H.N., de Queiroz, K., Donoghue, M.J., Eriksson,
T., Hillis, D.M., Lee, M.S.Y., 1999. Species names in phylogenetic
nomenclature. Syst. Biol. 48, 790–807.
Case, S.M., 1978a. Biochemical systematics of members of the genus
Rana native to western North America Syst. Zool. 27, 299–311.
Case, S.M., 1978b. Electrophoretic variation in two species of ranid
frogs, Rana boylei and Rana muscosa. Copeia 1978, 311–320.
Cochran, D.M., Goin, C.J., 1970. The New Field Book of Reptiles and
Amphibians. G.P. Putnum’s Sons, New York.
Cope, E.D., 1866. Fifth contribution to the herpetology of tropical
America. Proc. Acad. Philadelphia 1866, pp. 317.
Cuellar, O., Méndez-DeLaCruz, F., Villagrán-Santa Cruz, M., 1996.
Rana chichicuahutla, a new species of leopard frog from the Mexi-
can Plateau, 8. Revista de Zoologia, Mexico, pp. 1–8.
de Queiroz, K., Gauthier, J., 1990. Phylogeny as a central principle in
taxonomy: phylogenetic deWnitions of taxon names. Syst. Zool. 39,
307–322.
de Queiroz, K., Gauthier, J., 1992. Phylogenetic taxonomy. Ann. Rev.
Ecol. Syst. 23, 449–480.
de Queiroz, K., Gauthier, J., 1994. Toward a phylogenetic system of
biological nomenclature. Trends Ecol. Evol. 9, 27–31.
Dubois, A., 1992. Notes sur la classiWcation des Ranidae (Amphibiens
Anoures). Bull. Société Linnéenne de Lyon 61, 305–352.
Farris, J.S., Kluge, A.G., Mickevich, M.F., 1980. Paraphyly of the Rana
boylii group. Syst. Zool. 28, 627–634.
Farris, J.S., Kluge, A.G., Mickevich, M.F., 1983. Immunological dis-
tance and the phylogenetic relationships of the Rana boylii species
group. Syst. Zool. 31, 479–491.
Felsenstein, J., 1978. Cases in which parsimony and compatibility
methods will be positively misleading. Syst. Zool. 27, 401–410.
Fowler, H.W, 1913. Amphibians and reptiles from Ecuador, Venezuela,
and Yucatan. Proc. Acad. Nat. Sci. Philidelphia 65, 153–176.
Frost, J.S., Platz, J.E., 1983. Comparative assessment of modes of
reproductive isolation among four species of leopard frogs (Rana
pipiens complex). Evolution 37, 66–78.
Goebel, A.M., Donnelly, J.M., Atz, M.E., 1999. PCR primers and
ampliWcation 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 ampli-
Wed DNA in amphibians successfully. Mol. Phylogenet. Evol. 11,
163–199.
Goldman, N., Anderson, J.P., Rodrigo, A.G., 2000. Likelihood-based
tests of topologies in phylogenetics. Syst. Biol. 49, 652–670.
Green, D.M., 1986a. Systematics and evolution of western North
American frogs allied to Rana aurora and Rana boylii: karyological
evidence. Syst. Zool. 35, 273–282.
Green, D.M., 1986b. Systematics and evolution of western North
American frogs allied to Rana aurora and Rana boylii: electropho-
retic evidence. Syst. Zool. 35, 283–296.
Green, D.M., Kaiser, H., Sharbel, T.F., Kearsley, J., McAllister, K.R.,
1997. Cryptic species of spotted frogs, Rana pretiosa complex, in
western North America. Copeia 1997, 1–8.
Günther, C.C.L.G., 1900. Biologia Centrali-Americana. Reptilia and
Batrachia, Porter, London pp. 197–236.
Hayes, M.P., Kremples, D.M., 1986. Vocal sac variation among frogs
of the genus Rana from western North America. Copeia 1986, 927–
936.
Henderson, R.W., Hoevers, L.G., 1975. A Checklist and Key to the
Amphibians and Reptiles of Belize, Central America, vol. 5. Mil-
waukee Publ. Mus, Contrib. Biol. Geol., pp. 1–63.
Hillis, D.M., 1985. Evolutionary genetics and systematics of the New
World frogs of the genus Rana: an analysis of ribosomal DNA,
allozymes, and morphology. Ph.D. Dissertation, Univ. Kansas,
Lawrence.
Hillis, D.M., 1987. Molecular versus morphological approaches to sys-
tematics. Ann. Rev. Ecol. Syst. 18, 23–42.
Hillis, D.M., 1988. Systematics of the Rana pipiens complex: puzzle and
paradigm. Ann. Rev. Ecol. Syst. 19, 39–63.
Hillis, D.M., Bull, J.J., 1993. An empirical test of bootstrapping as a
method for assessing conWdence in phylogenetic analysis. Syst. Biol.
42, 182–192.
Hillis, D.M., Davis, S.K., 1986. Evolution of ribosomal DNA: Wfty mil-
lion years of recorded history in the frog genus Rana. Evolution 40,
1275–1288.
Hillis, D.M., de Sá, R., 1988. Phylogeny and taxonomy of the Rana
palmipes group (Salientia: Ranidae). Herpetol. Monographs 2, 1–
26.
Hillis, D.M., Frost, J.S., 1985. Three new species of leopard frogs (Rana
pipiens complex) from the Mexican Plateau. Occ. Pap. Mus. Nat.
Hist. Univ. Kansas 117, 1–14.
Hillis, D.M., Chamberlain, D.A., Wilcox, T.P., Chippindale, P.T., 2001.
A new species of subterranean blind salamander (Plethodontidae:
Hemidactyliini: Eurycea: Typhlomolge) from Austin, Texas, and a
systematic revision of central Texas paedomorphic salamanders.
Herpetologica 57, 266–280.
Hillis, D.M., Frost, J.S., Webb, R.G., 1984. A new species of frog of the
Rana tarahumarae group from southwestern Mexico. Copeia 1984,
398–403.
Hillis, D.M., Frost, J.S., Wright, D.A., 1983. Phylogeny and biogeogra-
phy of the Rana pipiens complex: a biochemical evaluation. Syst.
Zool. 32, 132–143.
Hillis, D.M., Mable, B.K., Larson, A., Davis, S.K., Zimmer, E.A., 1996a.
Nucleic acids IV: sequencing and cloning. In: Hillis, D.M., Mable,
B.K., Moritz, C. (Eds.), Molecular Systematics, second ed. Sinauer,
Sunderland, MA, pp. 321–381.
Hillis, D.M., Mable, B.K., Moritz, C., 1996b. Applications of molecular
systematics and the future of the Weld. In: Hillis, D.M., Mable, B.K.,
Moritz, C. (Eds.), Molecular Systematics, second ed. Sinauer, Sun-
derland, MA, pp. 515–543.
Huelsenbeck, J.P., Hillis, D.M., 1993. Success of phylogenetic methods
in the four-taxon case. Syst. Biol. 42, 247–264.
Huelsenbeck, J.P., Hillis, D.M., Nielsen, R., 1996. A likelihood-ratio
test of monophyly. Syst. Biol. 45, 546–558.
Huelsenbeck, J.P., Rannala, B., 2004. Frequentist properties of
Bayesian posterior probabilities of phylogenetic trees under
simple and complex substitution models. Syst. Biol. 53, 907–
916.
Huelsenbeck, J.P., Ronquist, F.R., 2001. MRBAYES: Bayesian infer-
ence of phylogenetic trees. Bioinformatics 17, 754–755.
314 D.M. Hillis, T.P. Wilcox / Molecular Phylogenetics and Evolution 34 (2005) 299–314
International Commission on Zoological Nomenclature, 1999. Inter-
national Code of Zoological Nomenclature, fourth ed. Interna-
tional Trust for Zoological Nomenclature, London, United
Kingdom.
Larget, B., Simon, D.L., 1999. Markov chain Monte Carlo algorithms
for the Bayesian analysis of phylogenetic trees. Mol. Biol. Evol. 16,
750–759.
Larson, K.A., 2004. Advertisement call complexity in northern leopard
frogs, Rana pipiens. Copeia 2004, 676–682.
Lee, J.C., 1976. Rana maculata Brocchi, an addition to the herpetofa-
una of Belize. Herpetologica 32, 211–214.
Littlejohn, M.J., Oldham, R.S., 1968. Rana pipiens complex: mating call
structure and taxonomy. Science 162, 1003–1005.
Macey, J.R., Strasburg, J.L., Brisson, J.A., Vredenburg, V.T., Jennings,
M., Larson, A., 2001. Molecular phylogenetics of western North
American frogs of the Rana boylii species group. Mol. Phylogenet.
Evol. 19, 131–143.
Mecham, J.S., 1971. Vocalizations of the leopard frog, Rana pipiens,
and three related Mexican species. Copeia 1971, 505–516.
Moore, J.A., 1944. Geographic variation in Rana pipiens Schreber of
eastern North America. Bull. Am. Mus. Nat. Hist. 82, 345–370.
Moore, J.A., 1975. Rana pipiens —The changing paradigm. Am. Zool.
15, 837–849.
Newton, M.A., 1996. Bootstrapping phylogenies: large deviations and
dispersion eVects. Biometrika 83, 315–328.
Oldham, R.S., 1974. Mate attraction by vocalization in members of the
Rana pipiens complex. Copeia 1974, 982–984.
Pace, A.E., 1974. Systematic and biological studies of leopard frogs
(Rana pipiens complex) of the United States. Misc. Publ. Mus. Zool.
Univ. Michigan 148, 1–140.
Peters, W., 1859. Eine neue Gattung und eine neue Art von Froschen
aus Caracas. Monatsb. Akad. Wiss. Berlin 1859, 402–403.
Posada, D., Crandall, K.A., 2001. Selecting the best-Wt model of nucleo-
tide substitution. Syst. Biol. 50, 580–601.
Post, T.J., Uzzell, T., 1981. The relationships of Rana sylvatica and the
monophyly of the Rana boylii group. Syst. Zool. 30, 170–180.
Pytel, B.A., 1986. Biochemical systematics of the eastern North Ameri-
can frogs of the genus Rana. Herpetologica 42, 273–282.
Rambaut, A., Grassly, N.C., 1997. Seq-Gen: an application for the
Monte Carlo simulation of DNA sequence evolution along phylo-
genetic trees. Comput. Appl. Biosci. 13, 235–238.
Schaaf, R.T., 1971. Rana palustris. Cat. Am. Amphibians Reptiles 117,
1–3.
Schmidt, R.S., 1968. Chuckle calls of the leopard frog (Rana pipiens).
Copeia 1968, 561–569.
ShaVer, H.B., Fellers, G.M., Voss, S.R., Oliver, J.C., Pauly, G.B., 2004.
Species boundaries, phylogeography and conservation of the red-leg-
ged frog (Rana aurora/draytonii) complex. Mol. Ecol. 13, 2667–2677.
Smith, H.M., Chiszar, D., 2003. Distributional and variational data on
the frogs of the genus Rana in Chihuahua, Mexico, including a new
species. J. Maryland Herpetol. Soc. 39, 59–66.
SwoVord, D.L., Olsen, G.J., Waddell, P.J., Hillis, D.M., 1996. Phyloge-
netic inference. In: Hillis, D.M., Mable, B.K., Moritz, C. (Eds.),
Molecular Systematics, second ed. Sinauer, Sunderland, MA, pp.
407–514.
SwoVord, D.L., 2000. PAUP*. Phylogenetic Analysis Using Parsimony
(¤ and Other Methods). Version 4. Sinauer Associates, Sunderland,
MA.
Thompson, J.D., Higgins, D.G., Gibson, T.J., 1994. CLUSTAL W:
Improving the sensitivity of progressive multiple sequence align-
ment through sequence weighting, positions-speciWc gap penalties
and weight matrix choice. Nucleic Acids Res. 22, 4673–4680.
Wallace, D.G., King, M.-C., Wilson, A.C., 1973. Albumin diVerences
among ranid frogs: Taxonomic and phylogenetic implications. Syst.
Zool. 22, 1–13.
Webb, R.G., 1996. A systematic review of the Mexican frog Rana
megapoda Taylor (Anura: Ranidae). Southwest Nat. 41, 35–42.
Webb, R.G., 2001. Frogs of the Rana tarahumarae group in western
Mexico. In: Johnson, J.D., Webb, R.G., Flores-Villela, O.A.
(Eds.), Mesoamerican Herpetology: Systematics, Zoogeogra-
phy, and Conservation. University of Texas at El Paso, El Paso,
pp. 20–43.
Wilcox, T.P., Zwickl, D.J., Heath, T.A., Hillis, D.M., 2002. Phylogenetic
relationships of the dwarf boas and a comparison of Bayesian and
bootstrap measures of phylogenetic support. Mol. Phylogenet.
Evol. 25, 361–371.
Zaldívar-Riverón, A., León-Regagnon, V., Nieto-Montes de Oca, A.,
2004. Phylogeny of the Mexican coastal leopard frogs of the Rana
berlandieri group based on mtDNA sequences. Mol. Phylogenet.
Evol. 30, 38–49.
Zweifel, R.G., 1955. The ecology and systematics of the Rana boylii
species group. Univ. California Publ. Zool. 54, 207–292.
... Ripplinger & Wagner, 2004;Recuero et al., 2006;Lemmon et al., 2007) of Pseudacris sensu lato, in particular the H. regilla complex, we amplified partial regions of mitochondrial cytochrome b (cyt b, 600 bp) and 12S-16S (12S-16S, 726 bp). We used the primer pairs (MVZ 15+MVZ 16) from Moritz et al. (1992) and (12L1 + 16Sh) from Hillis & Wilcox (2005) following PCR temperature protocols in Ripplinger & Wagner (2004) and Hillis & Wilcox (2005), respectively. PCR products were sequenced in both forward and reverse directions using the PCR primers on a Beckman Coulter automated capillary sequencer. ...
... Ripplinger & Wagner, 2004;Recuero et al., 2006;Lemmon et al., 2007) of Pseudacris sensu lato, in particular the H. regilla complex, we amplified partial regions of mitochondrial cytochrome b (cyt b, 600 bp) and 12S-16S (12S-16S, 726 bp). We used the primer pairs (MVZ 15+MVZ 16) from Moritz et al. (1992) and (12L1 + 16Sh) from Hillis & Wilcox (2005) following PCR temperature protocols in Ripplinger & Wagner (2004) and Hillis & Wilcox (2005), respectively. PCR products were sequenced in both forward and reverse directions using the PCR primers on a Beckman Coulter automated capillary sequencer. ...
Article
The Pacific chorus frogs are a complex of three wide-ranging species (i.e. Hyliola hypochondriaca, Hyliola regilla, Hyliola sierra) whose current taxonomy remains unresolved. We conducted species delimitation analyses of these taxa using fragments of the cytochrome b and 12S–16S mtDNA genes to assess the species diversity. Importantly, we included samples from new locations throughout the range to better understand species distributions and identify potential contact zones among clades. Our analyses revealed three slightly parapatric but distinct species-level clades. Molecular dating revealed that these species began diverging in the Pleistocene c. 1.4 Mya with H. hypochondriaca and H. sierra diverging more recently c. 0.8 Mya. We found that populations from western Montana and Idaho originated recently from populations to the southwest that belong to H. sierra, rather than from H. regilla populations directly to the west. Population sizes of each species expanded c. 130–80 Kya with H. hypochondriaca exhibiting a more pronounced expansion beginning c. 100 Kya than the more gradual expansion of the other two species. The climatic niche models suggest that distributions of the three species were similar during the last interglacial (LIG) as they are today. During the Last Glacial Maximum (LGM), H. hypochondriaca and H. sierra occupied a larger range than they do today whereas H. regilla occupied a smaller refugium, shifted south from the current distribution. This study highlights the continued effectiveness of utilizing single-locus data sets for species delimitation and biogeographic analyses.
... The family Ranidae, although extremely diverse, is represented by only three described species in South America (Frost, 2020): Lithobates vaillanti (Brocchi, 1877), L. bwana (Hillis & de S a, 1988) and L. palmipes (Spix, 1824) (Hillis & Wilcox, 2005). These species originated approximately 28 million years ago, during the formation of the Panama Isthmus and the uplift of the Andes (F. ...
... To build ecological niche modelling (ENM), we gathered L. palmipes occurrence points from VertNet (Constable et al., 2010) and from classic papers on taxonomy and distribution of Lithobates species (Hillis & de S a, 1988;Hillis & Wilcox, 2005). We also used occurrence points from localities from where we have tissues. ...
Article
The Atlantic and Amazon rainforests have a shared but unclear past, with intermittent connections resulting from historical climate change. We investigate these connections by studying the phylogeography and climatic niche of the disjunct distributed frog Lithobates palmipes. We sequenced two fragments of mitochondrial DNA from Atlantic Forest (AtF) and Amazonia (AmF) individuals and evaluated how genetic diversity is distributed in space and whether past demographic changes occurred. Also, we evaluated the existence of past suitable connections between biomes for L. palmipes through ecological niche models (ENM) and tested for niche divergence. The AtF group is nested within the AmF group and closely related to individuals from eastern Amazonia, a pattern recovered in many species that used northeast connection routes. We found evidence of recurrent use of connections in different directions and time during the Pleistocene, resulting in genetic structure between biomes, with no signal of demographic change and evidence of niche divergence across both genetic groups. ENMs indicated suitable areas connecting forests throughout northeastern Brazil during the Pleistocene. Mitochondrial lineages do not match biomes exactly. One lineage is composed of AtF populations and eastern Amazonia individuals. The other is composed of western Amazonia individuals, suggesting an effect of past climatic heterogeneity within the Amazonia forest. This is the first evidence that this route drove genetic and ecological diversity for amphibians recently, a group with habits and ecological requirements different from other vertebrates that have been shown to use this putative corridor.
... Ripplinger & Wagner, 2004;Recuero et al., 2006;Lemmon et al., 2007) of Pseudacris sensu lato, in particular the H. regilla complex, we amplified partial regions of mitochondrial cytochrome b (cyt b, 600 bp) and 12S-16S (12S-16S, 726 bp). We used the primer pairs (MVZ 15+MVZ 16) from Moritz et al. (1992) and (12L1 + 16Sh) from Hillis & Wilcox (2005) following PCR temperature protocols in Ripplinger & Wagner (2004) and Hillis & Wilcox (2005), respectively. PCR products were sequenced in both forward and reverse directions using the PCR primers on a Beckman Coulter automated capillary sequencer. ...
... Ripplinger & Wagner, 2004;Recuero et al., 2006;Lemmon et al., 2007) of Pseudacris sensu lato, in particular the H. regilla complex, we amplified partial regions of mitochondrial cytochrome b (cyt b, 600 bp) and 12S-16S (12S-16S, 726 bp). We used the primer pairs (MVZ 15+MVZ 16) from Moritz et al. (1992) and (12L1 + 16Sh) from Hillis & Wilcox (2005) following PCR temperature protocols in Ripplinger & Wagner (2004) and Hillis & Wilcox (2005), respectively. PCR products were sequenced in both forward and reverse directions using the PCR primers on a Beckman Coulter automated capillary sequencer. ...
Article
The Pacific chorus frogs are a complex of three wide-ranging species (i.e. Hyliola hypochondriaca, Hyliola regilla, Hyliola sierra) whose current taxonomy remains unresolved. We conducted species delimitation analyses of these taxa using fragments of the cytochrome b and 12S-16S mtDNA genes to assess the species diversity. Importantly, we included samples from new locations throughout the range to better understand species distributions and identify potential contact zones among clades. Our analyses revealed three slightly parapatric but distinct species-level clades. Molecular dating revealed that these species began diverging in the Pleistocene c. 1.4 Mya with H. hypochondriaca and H. sierra diverging more recently c. 0.8 Mya. We found that populations from western Montana and Idaho originated recently from populations to the southwest that belong to H. sierra, rather than from H. regilla populations directly to the west. Population sizes of each species expanded c. 130-80 Kya with H. hypochondriaca exhibiting a more pronounced expansion beginning c. 100 Kya than the more gradual expansion of the other two species. The climatic niche models suggest that distributions of the three species were similar during the last interglacial (LIG) as they are today. During the Last Glacial Maximum (LGM), H. hypochondriaca and H. sierra occupied a larger range than they do today whereas H. regilla occupied a smaller refugium, shifted south from the current distribution. This study highlights the continued effectiveness of utilizing single-locus data sets for species delimitation and biogeographic analyses.
... Rana sphenocephala is part of a clade of anurans associated primarily with permanent breeding sites (Saenz 2004;Hillis and Wilcox 2005). These sites are known to have a greater variety and density of predators, and the increased diversity of predators may require a more diversified system of tadpole defenses (Woodward 1983). ...
... We suggest R. sphenocephala may be a recent member of the ephemeral pond breeding community and secondary antipredator mechanisms the species possesses are pleisiomorphic traits. Given their phylogenetic history (Hillis and Wilcox 2005), it seems plausible that R. sphenocephala possess secondary antipredator abilities that are more developed compared to other members of the winter breeding guild. ...
Article
The activity level-predation risk paradigm of community assembly is based on observable trade-offs between activity level and predation risk. Many species that live in ephemeral habitats are relatively more active, grow faster, and metamorphose sooner than those in more permanent habitats, thus enabling them to survive by escaping drying ponds. Species with high activity levels are shown to be superior competitors in temporary ponds, but are at greater risk of predation by visually cued predators. We take a comprehensive approach to testing if the activity-predation paradigm drives community assembly of a four-species guild of larval anurans through field surveys and a series of laboratory experiments quantifying their activity level, anti-predatory responses, swim performance, and escape behavior. Activity level was a good predictor of susceptibility to predation and habitat association for most species in our study, but not all species fit the pattern. Secondary anti-predator mechanisms contributed to differences in susceptibility to predation. Some species were more capable of avoiding predation through swim performance and escape behavior, thus allowing them to occupy a greater range of the water permanency continuum than was predicted by activity level alone. Because performance traits are deeply rooted, evolutionary history plays a major role in determining tadpole performance and life-history traits. Whole-organism performance traits (e.g., swim performance, escape behavior) have not generally been considered in the classic activity-predation paradigm. Our results show that accounting for individual and species-level performance reveals the understudied, yet important role of species-specific performance traits in disentangling the mechanisms driving community assembly.
... In addition, remnant populations are frequently dispersed over large areas and in habitats with difficult access and thick vegetation (U.S. Fish and Wildlife Service, 2007) which makes capturing individuals by hand challenging. By contrast, calls of southwestern leopard frog species are relatively distinct (Hillis and Wilcox, 2005). Thus, they serve as useful cues in the identification process. ...
... Although modern approaches to species delimitation incorporate gene tree discordance into their frameworks, the currently accepted taxonomy in many groups is based on earlier studies that relied on more limited data. Mitochondrial markers were particularly popular in studies of phylogeography and species delimitation for many years (Burbrink et al., 2000;Hillis and Wilcox, 2005;Kozak et al., 2005;Lemmon et al., 2007). However, even prior to the wide availability of high-throughput sequence data, the observation of mitonuclear discordance was becoming more frequent (Leaché and McGuire, 2006;Papakostas et al., 2016;Toews and Brelsford, 2012;Wiens et al., 2010). ...
Article
As DNA sequencing technologies and methods for delimiting species with genomic data become more accessible and numerous, researchers have more tools than ever to investigate questions in systematics and phylogeography. However, easy access to sophisticated computational tools is not without its drawbacks. Choosing the right approach for one's question can be challenging when presented with multitudinous options, some of which fail to distinguish between species and intraspecific population structure. Here, we employ a methodology that emphasizes intensive geographic sampling, particularly at contact zones between populations, with a focus on differentiating intraspecific genetic clusters from species in the Pantherophis guttatus complex, a group of North American ratsnakes. Using a mitochondrial marker as well as ddRADseq data, we find evidence of mitonuclear discordance which has contributed to historical confusion about the relationships within this group. Additionally, we identify geographically and genetically structured populations within the species Pantherophis emoryi that are congruent with previously described morphological variation. Importantly, we find that these structured populations within P. emoryi are highly admixed throughout the range of the species and show no evidence of any reproductive isolation. Our data support a revision of the taxonomy of this group, and we recognize two species within the complex and three subspecies within P. emoryi. This study illustrates the importance of thorough sampling of contact zones and consideration of gene flow when delimiting species in widespread complexes containing parapatric lineages.
... These revisions were based on genetic diversity of these frogs, especially for the Chinese species (Che et al., 2007). Although some molecular phylogenetic analysis on these taxa has been recorded (Bossuyt et al., 2006;Hillis & Wilcox, 2005;Matsui et al., 2006), the detail phylogenetic systematics for these taxa remains unclear. However, to gain a robust phylogeny for the family Ranidae, intensive taxa samplings are in need. ...
Article
Full-text available
The Omei wood frog (Rana omeimontis), endemic to central China, belongs to the family Ranidae. In this study, we achieved detail knowledge about the mitogenome of the species. The length of the genome is 20,120 bp, including 13 protein‐coding genes (PCGs), 22 tRNA genes, two rRNA genes, and a noncoding control region. Similar to other amphibians, we found that only nine genes (ND6 and eight tRNA genes) are encoded on the light strand (L) and other genes on the heavy strand (H). Totally, The base composition of the mitochondrial genome included 27.29% A, 28.85% T, 28.87% C, and 15.00% G, respectively. The control regions among the Rana species were found to exhibit rich genetic variability and A + T content. R. omeimontis was clustered together with R. chaochiaoensis in phylogenetic tree. Compared to R. amurensis and R. kunyuensi, it was more closely related to R. chaochiaoensis, and a new way of gene rearrangement (ND6‐trnE‐Cytb‐D‐loop‐trnL2 (CUN)‐ND5‐D‐loop) was also found in the mitogenome of R. amurensis and R. kunyuensi. Our results about the mitochondrial genome of R. omeimontis will contribute to the future studies on phylogenetic relationship and the taxonomic status of Rana and related Ranidae species. The Omei wood frog (Rana omeimontis), endemic to central China, belongs to the family Ranidae. R. omeimontis was clustered together with R. chaochiaoensis in phylogenetic tree. Compared to R. amurensis and R. kunyuensi, it was more closely related to R. chaochiaoensis, and a new way of gene rearrangement (ND6‐ trnE‐ Cytb‐ D‐ loop‐ trnL2 (CUN) ‐ ND5‐ D‐ loop) was also found in the mitogenome of R. amurensis and R. kunyuensi. Our results about the mitochondrial genome of R. omeimontis will contribute to the future studies on phylogenetic relationship and the taxonomic status of Rana and related Ranidae species.
Article
Resource partitioning within the ecological niche space in which there is a high level of overlap between species can alleviate the tendency toward competitive exclusion. When competitive ability is asymmetrical due to predation or other ecological factors, it may be more effective for the less competitive species to lessen direct competition by contracting their use of local resources. Species occurring in mixed assemblages may come into direct contact with each other throughout their respective breeding seasons. Where competition for breeding habitat and acoustic space exists, the level of interference is expected to vary widely, depending upon the ecological and breeding similarities between the species involved and the relative importance of the resource. In this study, I investigated the breeding season interspecific interactions of two species of ranid frog in eastern North America, the American bullfrog (Rana catesbeiana) and the green frog (R. clamitans). The ecological and behavioral similarities between these species combined with phylogenetic relatedness and comparable natural distributions make them an ideal system for studying interspecific dynamics related to their breeding ecology. Specifically, I examined the influence bullfrogs have on the breeding behavior of green frogs over several timescales, including physical avoidance of encounters through counter movements of green frogs away from bullfrog territories, an adjustment of green frog microhabitat use, and green frog avoidance of acoustic masking by bullfrogs. I found that green frogs defend territories and lay eggs closer to shore than bullfrogs. Also, both green frog territories and eggs are under heavier overhead cover than bullfrog eggs and territories. I found green frogs respond to bullfrog chorusing on a fine temporal scale by placing their calls between the notes of bullfrog calls.
Article
Full-text available
Climate change has important effects on patterns of population persistence, connectivity, and divergence. We used mtDNA sequences and species distribution modeling to assess the impact of climatic changes in the past (Last Glacial Maximum (LGM: 21 Kya) and Mid-Holocene (6 Kya)), recent (1970-2000), and future (2070) on the phylogeography and spatial distribution of populations of the Hyrcanian wood frog, Rana pseudodalmatina in Northern Iran. Based on two mitochondrial genes (cytochrome b and 16S ribosomal RNA), we found evidence for two regional clades that diverged in the Pleistocene (1.6 Mya) and are distributed in the eastern and western sections of the current species range. Biogeographic analyses support both vicariance (an increase of the Caspian Sea water levels) and dispersal events have been involved in shaping the genetic structure of the species. Reconstruction of the ancestral distribution of R. pseudodalmatina suggests the species’ range contracted in two independent eastern and western glacial refugia during the LGM, expanding since the Mid-Holocene to the present to continuously occupy Hyrcanian forests. According to future climate projections, the range of the species shows a tendency to shift to higher altitudes. Landscape connectivity analyses support higher population continuity in the central part of the current range, with isolated populations in the easternmost and westernmost extremes. Our integrative study of R. pseudodalmatina provides support for the “refugia-within-refugia” scenario in the Hyrcanian forests.
Article
Full-text available
Self-published taxon descriptions, bereft of a basis of evidence, are a long-standing problem in taxonomy. The problem derives in part from the Principle of Priority in the International Code of Zoological Nomenclature, which forces the use of the oldest available nomen irrespective of scientific merit. This provides a route to ‘immortality’ for unscrupulous individuals through the mass-naming of taxa without scientific basis, a phenomenon referred to as taxonomic vandalism. Following a flood of unscientific taxon namings, in 2013 a group of concerned herpetologists organized a widely supported, community-based campaign to treat these nomina as lying outside the permanent scientific record, and to ignore and overwrite them as appropriate. Here, we review the impact of these proposals over the past 8 years. We identified 59 instances of unscientific names being set aside and overwritten with science-based names (here termed aspidonyms), and 1087 uses of these aspidonyms, compared to one instance of preference for the overwritten names. This shows that when there is widespread consultation and agreement across affected research communities, setting aside certain provisions of the Code can constitute an effective last resort defence against taxonomic vandalism and enhance the universality and stability of the scientific nomenclature.
Article
Full-text available
Une nouvelle classification provisoire de la superfamille des Ranoidea, et plus particulierement de la famille des Ranidae et du genre Rana, est proposee. Le genre Rana est ici subdivise, principalement sur la base de criteres phenetiques, en trente-trois groupes, auxquels est ici provisoirement attribue le rang de sous-genres. Le but principal de cette classification provisoire, qui doit etre comprise comme un outil de travail et non pas comme une proposition definitive, est de servir de guide pour de futures etudes sur la phylogenie de cette famille tres vaste et a repartition quasi cosmopolite : cette classification aidera a selectionner des especes representatives de chaque groupe phenetique pour l'etude des etats de caracteres, et a choisir les hors-groupes appropries pour l'etablissement de la polarite des morphoclines de caracteres. Ce n'est que lorsque la phylogenie de la famille dans son ensemble sera resolue qu'il sera possible d'elaborer une classification plus stable, qui s'averera peut-etre tres differente de celle qui est presentee ci-dessous.
Article
Full-text available
We describe a new species of salamander (Eurycea) from the Barton Springs segment of the Edwards Aquifer, Austin, Texas, USA. The new species is most closely related to Eurycea (Typhlomolge) rathbuni from subterranean waters around San Marcos, Texas, and like that species lacks external eyes and shows other morphological features associated with subterranean life. The new species is easily distinguished on the basis of morphology from all previously described species of salamanders, and in particular is easily distinguished from its closest relatives, E. (Typhlomolge) rathbuni and E. (Typhlomolge) robusta, as well as the sympatric E. sosorum. We used sequences of the mitochondrial cytochrome b gene to infer the phylogeny of described species of central Texas Eurycea, and these data support the major groups reported previously on the basis of other DNA and allozyme data. We also define names for the major clades of central Texas Eurycea.
Article
This report updates information on variation and distribution of Rana megapoda Taylor. This distinctive, large frog is known to occur only in southwest central Mexico in the adjoining states of Jalisco, Nayarit, Michoacan, and Guanajuato. External morphological features are used to distinguish R. megapoda from other sympatric or nearby allopatric species of Mexican ranid frogs.