HERPETOLOGICA · HERPETOLOGICA VOL. 65 JUNE 2009 NO. 2 Herpetologica, 65(2), 2009, 115–128 E 2009...

HERPETOLOGICAVOL. 65 JUNE 2009 NO. 2

Herpetologica, 65(2), 2009, 115–128

E 2009 by The Herpetologists’ League, Inc.

TAXONOMIC FREEDOM AND THE ROLE OF OFFICIAL LISTS OFSPECIES NAMES

GREGORY B. PAULY1,2, DAVID M. HILLIS, AND DAVID C. CANNATELLA

Section of Integrative Biology and Texas Natural Science Center, 1 University Station C0930, University of Texas,Austin, TX 78712, USA

ABSTRACT: The sixth edition of the Scientific and Standard English Names of Amphibians and Reptiles ofNorth America (Crother, 2008, SSAR Herpetological Circular 37:1–84) is the ‘‘Official Names List’’ for thethree major North American herpetological societies. Although this publication is intended to aid users ofscientific and common names, we argue that current practices for authoring, reviewing, and using this list, insome cases, generate taxonomic chaos. By this we mean that users are uncertain of which name to use and/orthe rationale for using a particular name, and efficient communication is hindered by this confusion. Mostimportantly, through inadequate and inconsistent review of this list, the societies have endorsed unnecessaryand arbitrary name changes and are uncritically promoting individual taxonomic viewpoints when a clearchoice on the most appropriate name has not been reached by the community. This problem is exemplifiedby North American anurans for which 57 of the 100 species have scientific names (i.e., genus-speciescombinations) different from the previous version of the list. Forty-eight of these new combinations resultfrom changes to the genus name, and there is controversy over the proposed genus names for at least 43 ofthese. Despite this controversy and that a stated goal of the list is to report on such controversies, thealternative names are not discussed. As a result, for these taxa, the list fails to provide adequate informationfor users to make informed decisions on name usage. Here, we examine the role of such lists in taxonomy.Although we specifically focus on the arbitrary changes to the names of North American Bufo and Rana, thecontinuation of current practices for generating the list will promote instability and taxonomic confusion on abroader scale. We conclude with recommendations for improving the utility of such lists and for avoidingunnecessary taxonomic chaos.

Key words: Bufo; Rana; Species names; Taxonomic stability

THE LAST decade’s flood of new phylogenieshas rewritten our understanding of phyloge-netic relationships and taxonomy of amphib-ians and reptiles. Publication of the sixthedition of the Scientific and Standard EnglishNames of Amphibians and Reptiles of NorthAmerica (Crother, 2008; hereafter referred toas ‘‘the list’’) brings this issue to the forefront.The sixth edition is sponsored by the threemajor North American herpetological socie-ties: the American Society of Ichthyologists

and Herpetologists (ASIH), the Herpetolo-gist’s League (HL), and the Society for theStudy of Amphibians and Reptiles (SSAR).Although the authors, editors, and sponsoringherpetological societies have changed sincethe first edition (Collins et al., 1978), the list’sprimary motivation is the same: to standardizeand stabilize common names and therebypromote efficient communication. (We willuse ‘‘common name’’ rather than ‘‘standardEnglish name’’ as our points are relevant to alllists of common names and not just those thatstabilize standard English names; see Crother[2007] for a discussion of these terms.)Although standardizing common names wasthe reason the list was conceived (see Conant

1 PRESENT ADDRESS: Department of Evolution andEcology, and Center for Population Biology, Universityof California, Davis, California 95616, USA.

2 CORRESPONDENCE: e-mail, [email protected]

115

et al., 1956), the list also has an increasinglyimportant role relevant to scientific names.With the rapid growth of phylogenetic studiesand the concomitant increase in changes toscientific names, the list has become evermore important as a single, convenientreference to current scientific names.

We argue, however, that some currentpractices for authoring, reviewing, and usingthis list result in a disservice to the field oftaxonomy and to users of scientific names. Inparticular, there appears to be inadequate andinconsistent review of proposed changes toscientific names. Without extensive review,the list may incorporate unnecessary taxo-nomic changes at the authors’ whim (eachtaxonomic section—frogs, lizards, etc.—isauthored by one to four individuals). Becausethe list is endorsed by the three major NorthAmerican herpetological societies, a readermight assume that any changes are broadlysupported by the society members, but, as wedetail below, this is not always the case.Moreover, no evidence suggests that thechanges incorporated in the list even reflectthe consensus of the committee that compiledthe list. As a result, the list may promotetaxonomic instability and thwart the goals ofefficient communication and information re-trieval that motivated its establishment.

Although the situation described above (i.e.,society endorsement of unnecessary andsubjective name changes lacking a strongconsensus among taxonomic specialists) mayseem unlikely, we argue that this exactscenario has occurred for the North Americananuran list (Frost et al., 2008) and is resultingin unprecedented name instability for NorthAmerican frogs. More than half of the 100species listed in Frost et al. (2008) have newnames. Of the 57 new combinations of anurannames made between the 2000 and 2008 lists,43 were recommended by Frost et al. (2006a)in their monograph on amphibian phylogeny.As suggested by several authors (Hillis, 2007;Smith and Chiszar, 2006; Vences, 2007), andas we show below, genus-name changesproposed by Frost et al. (2006a) for the NorthAmerican anuran species are arbitrary andunnecessary to reflect evolutionary history.Nevertheless, these proposed changes areincorporated into the anuran section of the

list, of which Frost is also the lead author(Frost et al., 2008).

The list is one of the most far-reachingcontributions of the herpetological societies tofellow professionals and the general public. Itis an important resource to legislators, lawenforcement personnel, amateur herpetolo-gists and naturalists, and a diverse group ofprofessional biologists, including wildlife man-agers and conservation biologists, journaleditors, authors of natural history lists andpublications, and managers of biologicaldatabases (e.g., GenBank, HerpNet, Ency-clopedia of Life, etc.). Given this huge usergroup, all members of the societies have aninterest in ensuring that the list representsthe best available science and taxonomicpractice. In our opinion, this standard wasnot met, at least in the case of the anuranlist, because it promotes individual view-points that do not follow long-establishedtaxonomic principles.

This essay is intended to stimulate discus-sion of both best practices in taxonomy andhow names lists can best benefit the scientificand nonscientific user community. We firstexamine the purpose of classification and therole of stability. We argue that stability of thegenus-species combination (i.e., the ‘‘speciesname’’ or ‘‘scientific name of the species’’according to the Code; ICZN, 1999) should bemaintained, except in cases of nonmonophylyof genera. In such cases, the number ofchanges in genus-species combination shouldbe minimized. We also discuss the relation-ship between phylogeny and taxonomy. Wethen examine the purposes of standard listsand discuss the importance of broad peerreview for evaluating proposed changes in listsof scientific and common names. We demon-strate ways to name taxa and provide addi-tional phylogenetic information withoutchanging genus-species combinations; taxo-nomic problems are addressed only as theyrelate to the scientific names—the genus-species combinations, rather than supragene-ric names or controversies related to subspe-cies status. Last, we explore how these issuesof instability and official lists are exemplifiedwith changes in the species names for NorthAmerican anurans, and we make generalrecommendations to promote stability.

116 HERPETOLOGICA [Vol. 65, No. 2

STABILITY AND THE PURPOSE OF

BIOLOGICAL CLASSIFICATION

Modern biological classifications are gener-ally recognized as having two principal goals(Ashlock, 1979; Benton, 2000; Mayr, 1982;Ross, 1974; Simpson, 1961). The first is tofunction as a reference system (Ross, 1974).As such, classifications facilitate communica-tion; this is the practical or utilitarian purposeof classification (Mayr, 1982). The second goalis to reflect evolutionary history (Ashlock,1979; de Queiroz, 1988, 1997). This is thegeneral or scientific goal of classification(Mayr, 1982). The phrase ‘‘reflect evolutionaryhistory’’ requires clarification; some haveargued that both paraphyletic and monophy-letic taxa reflect evolutionary history (e.g.,Ashlock, 1979). However, we accept themodern consensus that the phrase means thattaxon names, at least those above the specieslevel, refer only to monophyletic groups(clades).

Obviously, there can be conflict in achiev-ing these two goals. As noted by Cantino et al.(1999:796), ‘‘the communication function ofnomenclature is best served if names remainstable through time.’’ This is especially truegiven that the nonexpert user community isvirtually unlimited (Kennedy et al., 2006).Changes in species names require that allusers (academic and professional biologists,amateur biologists, laypersons, etc.) learn newnames to communicate about the taxon ofinterest. Furthermore, name changes separatea species from its previous literature. There-fore, users must also maintain a record of theprevious name to access published informa-tion. Clearly, reducing stability and familiarityof classifications simultaneously diminishestheir utility.

Absolute stability, however, is not desirablebecause it would hinder the scientific goal ofreflecting evolutionary relationships. As newphylogenetic information becomes available,changes to names may be necessary to reflectan increased understanding of evolutionaryrelationships. This conflict in simultaneouslymeeting the two principal goals of classifica-tion has been long recognized by taxonomists(e.g., Mayr et al., 1953; Simpson, 1961). Forexample, Simpson (1961:112) argued, ‘‘theremust be some compromise between the

usefulness of up-to-date classifications andthe usefulness of stable classifications.’’

The scientific name of the species is theprimary token by which biologists refer to aparticular species. Given this, our view is thatchanges to the genus name should be avoided,except in cases of nonmonophyly (or wheredemanded by the Principle of Priority, whichwill not be discussed further). Even then, thenumber of changes to scientific names shouldbe minimized. We recognize that informaticssystems are increasingly better at mappingequivalencies between different taxonomicreference systems (Kennedy et al., 2006),but even the best informatics system will noteliminate the need to communicate usinggenerally understood names. To ignore thegoal of stability is to deny the role of biologicalclassifications as general reference systems forall end users.

THE CONNECTION BETWEEN PHYLOGENY

AND CLASSIFICATION

It is generally accepted that classificationshould reflect phylogeny, but classificationsdiffer from phylogenies in at least two ways.First, a classification is usually an incompleterepresentation of phylogeny, because only asubset of branches of the Tree of Life arenamed. Second, one phylogeny can yieldmany different but equally informative classi-fications.

So, which branches of a phylogeny shouldbe named? In most cases we start not from ablank slate, but from existing classifications.New hypotheses of evolutionary relationshipsmay require taxonomic changes to maintainmonophyletic taxa. How should classificationsbe revised, when change is required bynonmonophyly? A common-sense approachwas well-articulated by Simpson (1961:112):‘‘A published classification in current useshould be changed when it is definitelyinconsistent with known facts and acceptedprinciples, but only in so far as necessary tobring it into consistency.’’ In the language ofmodern phylogenetics, ‘‘known facts andaccepted principles’’ suggests well-establishedand strongly supported clades that are in-ferred under multiple analyses and fromvarious data sources, a view echoed by theTurtle Taxonomy Working Group (2007) and

June 2009] HERPETOLOGICA 117

the American Ornithologists’ Union (1998) intheir discussions of protocols for revisingclassifications. Changing names based onpoorly supported branches, or for issuesunrelated to monophyly, is ill advised andcounter to the goals of biological classification(Cantino et al., 1999; Gaffney, 1979; Godfrayand Knapp, 2004; Mayr, 1982; Simpson,1961).

But Simpson’s rule of thumb is just that—arule of thumb. It is up to the biologicalcommunity to evaluate whether proposedchanges promote the goals of classification.If they do, such proposals should be adopted.However, if proposed changes generate un-necessary instability, then they should bedisregarded. Importantly, all users of nameshave the opportunity to make these decisions.Although many users of scientific namesfollow the most recent proposal or mostrecent taxonomic revision, there is no obliga-tion to do so, and doing so, may contributeeven further to taxonomic instability.

As secondary literature, names lists, espe-cially those sponsored by scientific societies,should state the current scientific name, or, incases where there may be alternatives, statethese names and a brief summary of therationale for each with relevant citations. Weexplicitly reject promotion of an individualviewpoint by authors of lists, and especially oftheir own taxonomic proposals. Likewise, wedo not wish to replace that imposed systemwith our own. Rather, the herpetologicalcommunity at large should be invested in theprocess of evaluating alternative classifica-tions. Because numerous classifications ofdiffering taxonomic stability may result fromone phylogeny, the choice of which taxonomybest serves the purposes of biological classi-fication is too important to be left only totaxonomists.

THE ROLE OF STANDARD LISTS

Standard names lists should serve multiplefunctions that help to achieve both principalgoals of classification:

1. Standard lists stabilize common names sothat a diversity of end-users will use thesame names for the same organisms eventhough no codes govern the use of

common names. Thus, standard listsfacilitate communication, a basic goal ofclassification.

2. Lists of scientific names also help toensure that classifications reflect evolu-tionary history. Phylogenetic studies maysuggest that the existing taxonomy isinconsistent with phylogenetic relation-ships. Thus, updated classifications pro-vide an important format for disseminat-ing the results of specialized phylogeneticresearch to a broad audience (see Hillis,2007, for further discussion).

There is also a more general role for listsendorsed by professional societies or agencies.These lists are viewed as authoritative, pre-sumably because users assume that the namesreflect the general consensus of the societies’membership or at least the consensus of thetaxonomic specialists within that society. As aresult, these lists are an important resource toauthors of more regional lists and publications(e.g., conservation and management agencylists, field guides, etc.) and may guide nameusage in these works. Given this role, authors,editors, reviewers, and sponsoring organiza-tions of such lists must recognize that they area relatively small group whose decisions oncontent impact a huge audience. Thesedecisions also determine whether a list willpromote or hinder the goals of classification.Thus, it is critical that these individuals workto ensure that classifications reflect the bestinformation available and simultaneously re-spect the interests and needs of the diverseuser community. Importantly, lists should notrestrict taxonomic freedom, including futuredebate on the most appropriate taxonomy.Below, we explore these issues by examining(1) taxonomic freedom, and (2) peer review/evaluation of names lists.

THE USE OF LISTS TO LEGISLATE TAXONOMY

The fifth and sixth editions of the list(Crother, 2000, 2008) were the first designat-ed as the ‘‘Official Names List.’’ Morespecifically, the title page of the fifth editionstates that the list is ‘‘officially recognized andadopted by the SSAR, ASIH, and HL’’ whilethe phrase ‘‘Official names list of the ASIH,HL, and SSAR’’ is on the front cover of the


sixth edition. Use of the word ‘‘official’’designates that the list is a publication of thesponsoring societies and distinguishes it fromlists generated by other organizations andagencies. However, some readers may misin-terpret the word ‘‘official’’ as implying that thescientific names are regulated by the societies;that is, the list may be misinterpreted as a listof official names (e.g., Stuart, 2008).

We argue that no scientific society shouldsuggest or imply that they are regulatingscientific names because such regulation iscounter to the spirit of the International Codeof Zoological Nomenclature (ICZN, 1999).Indeed, the first principle of the Code(1999:XIX) states, ‘‘The Code refrains frominfringing upon taxonomic judgment, whichmust not be made subject to regulation orrestraint.’’

This concept of taxonomic freedom isimportant. The Code (1999:119) defines a‘‘taxonomic taxon’’ as

‘‘A taxon (e.g., family, genus, species) includ-ing whatever nominal taxa and individuals azoologist at any time considers it to contain inhis or her endeavour to define the boundariesof a zoological taxon (q.v.).’’

This delineation of the ‘‘taxonomic taxon’’ isknown as the Taxon Concept (Kennedy et al.,2006): the mapping of a taxon name to apublication stating the author’s opinion of thetaxon’s content. According to the Code, anauthor is free to decide the content of a taxon,or taxon concept, even without regard to dataor published literature. On one hand, anindividual may propose a bewildering numberof new names. Although such action would behighly irresponsible, the formal rules ofnomenclature do not prevent it. On the otherhand, taxonomic freedom also allows eachindividual to ignore published proposals.Presumably, because users want to be up-to-date with the latest research, an extremelycommon misconception is that the mostrecent taxonomy must be followed. However,nothing in the Code obliges or encouragesauthors, editors, or other users to do this.

So how do we reconcile taxonomic freedomwith the goal of stability? We feel that‘‘taxonomy by informed consensus’’ is a good

descriptor of the process by which nomencla-tural changes are accepted by the communityof biologists, and we support this process.Given two competing taxonomic arrangements,a user ideally should select the one thatpromotes the goals of classification (to facilitatecommunication and indicate evolutionary his-tory) and whenever possible maintains stabilityand familiarity. Preferably, the user shouldjustify the choice of taxon concept in publica-tion. As more users make their opinions known,a consensus will emerge.

Taxonomy by informed consensus mightcollapse for several reasons. First, nonsyste-matist users are unlikely to be familiar with, orequipped to assess, the pertinent taxonomicliterature and may therefore make ill-in-formed choices. Second, uncritical acceptanceof the most recent published classification,simply to be current, will destabilize taxonomyand hinder broader communication. Third,the enforcement of a particular name byeditors when there is debate over the mostappropriate choice prevents discussion andevaluation of taxonomic proposals.

This third point is especially important.Inclusion in the list may suggest to many usersthat only these names are acceptable for use.Although there are no formal requirements inmost society journals that authors follow thenames in the list, authors may feel obligated tocomply simply because these names are in thesociety-sponsored list. One exception is Her-petological Review, in which authors ofnatural history and geographic distributionnotes are explicitly required to follow thenames in this list (this requirement dates backto Volume 10, 1979, the first volume followingthe publication of the SSAR’s first commonand scientific names list [Collins et al., 1978]).Similarly, editors of nonherpetological jour-nals may require that authors also follow thelist. Thus, if the list presents a one-sidedviewpoint, then the authors of the list, and theendorsing societies, are effectively, if uninten-tionally, legislating taxonomy. This result iscounter to the way in which taxonomy as afield should function. In our view, the listshould strive to summarize the current state ofnomenclature for each taxon, which shouldinclude discussing multiple names if alterna-tives exist.


Interestingly, early versions of the listrecognized that such lists might restricttaxonomic freedom. The first attempt by aNorth American herpetological society tostandardize common names was undertakenby the Committee on Herpetological Com-mon Names of the ASIH; this effort followedthe ASIH-sponsored checklist of scientificnames by Schmidt (1953). The authors made‘‘the list informal and as free as possible from thelegalistic practices that have plagued binomialnomenclature’’ (Conant et al., 1956:172). Thenames in this list were given ‘‘no official status,and their use (was) in no way compulsory’’(p. 172). The goal was simply to promote moreefficient communication. The current list(Crother, 2008) is a direct outgrowth of thisASIH list. Although the current list is also notintended to be compulsory, some users may notrecognize this.

PEER REVIEW AND EVALUATION OF THE

NAMES LIST

Balancing the dual goals of stability andconveying phylogenetic information is no easytask. This challenge requires input from abroad group of biologists, because the needfor a general consensus makes this task tooimportant to be left only to taxonomists. Eventhough taxonomists have the primary expertknowledge, scientific names should serve theentire community. We suggest that theprotocols of the Committee on StandardEnglish and Scientific Names (the jointcommittee that published Crother, 2000,2008) for evaluating proposed name changesare inadequate because some individual view-points have been promoted to the exclusion ofalternatives that maintain both stability andphylogenetic information.

A list of names is a snapshot in timewhereas taxonomic debates about namesoccur over long periods. At any point, a clearconsensus on the most appropriate names maynot exist. Crother (2000, 2008) recognized theimportance of providing information on com-peting proposals and specifically included‘‘comments regarding confidence in our un-derstanding.’’ We agree that this is a veryuseful contribution. Unfortunately, such com-ments are inconsistently used across taxonom-ic sections. For example, de Queiroz and

Reeder (2008) provide informative discussionsof recent taxonomic changes to North Amer-ican skinks (Plestiodon, formerly Eumeces) andwhiptails (Aspidocelis, formerly Cnemido-phorus). However, such discussions are mini-mal for controversial changes in the anuransection (Frost et al., 2008). Of the 57 newcombinations of anuran species names madebetween the 2000 and 2008 lists, 43 wereproposed in an earlier publication by the samelead author (Frost et al., 2006a). Severalpublications have argued against these propos-als (e.g., Smith and Chiszar, 2006; Hillis, 2007),but these objections were largely ignored byFrost et al. (2008). Clearly, the list of anurannames does not reflect the current state ofanuran taxonomy or the opinions of taxono-mists who have published phylogenies ortaxonomic revisions for the relevant groups.Thus, the list fails to elucidate the availablealternatives and provides but a single view-point. In our opinion, authors of this list shouldnot arbitrate taxonomic controversies, and theherpetological societies should not endorse liststhat attempt to do so. In the case of the currentlist, and specifically Frost et al. (2008), we feelthat the inconsistent approach to dealing withproposed taxonomic changes and the failure tocatch these inconsistencies in the reviewprocess demonstrate that the protocol forauthoring, editing, and reviewing this list needsto be revised.

The protocol for evaluating proposed nomen-clatural changes has changed dramatically overthe six editions of the list. At present (Crother,2008), one to four authors establish the list fortheir subsection (e.g., frogs, lizards, etc.),including decisions about proposed changes tostandard English and scientific names. The list isthen reviewed by a small number of herpetol-ogists. As discussed above, we believe that thisreview was inadequate. Given the importance ofthis list (and other similar lists of various taxa),we suggest that the review process be expandedto ensure that the list discusses current alterna-tive names if these exist.

Our concern about lack of community inputand prevalence of individual viewpoints ex-tends to other lists as well. For example, ITISis the Integrated Taxonomic InformationSystem, a partnership of several U.S. federalagencies, including the Environmental Pro-


tection Agency, Department of the Interior,and the National Museum of Natural History.The ITIS website lists the data stewards forNorth American amphibian species as RoyMcDiarmid and Darrel Frost, two of the threeauthors of the anuran section of the list. Thegoal of ITIS

‘‘is to create an easily accessible database withreliable information on species names andtheir hierarchical classification.’’ (www.itis.gov/info.html)And further,‘‘…ITIS is meant to serve as a standard toenable the comparison of biodiversity datasets,and therefore aims to incorporate classifica-tions that have gained broad acceptance in thetaxonomic literature and by professionals whowork with the taxa concerned.’’ (www.itis.gov/standard.html)

Just as with the Crother list, the ITIS listwill likely influence other lists and publications.Such lists and the communities they serve willbenefit from a broad review and greater focuson increasing taxonomic stability.

STABILITY AND PHYLOGENETIC INFORMATION:EXAMPLES FROM BUFO AND RANA

Normally, changes to classifications resultfrom new discoveries of taxon relationships. Itis worth noting that the 43 changed speciesnames for the North American Bufo and Ranarepresent no discoveries of new clades byFrost et al. (2006a). The phylogenies of NorthAmerican Bufo and Rana had been previouslyresolved and sampled more thoroughly byPauly et al. (2004; all 22 species of Bufo) andby Hillis and Wilcox (2005; 25 of 28 extantspecies of Rana; the 3 unsampled species wereall formerly treated as subgroups of sampledtaxa). For the North American species, Frostet al. (2006a) sampled only 9 species of Bufoand 12 species of Rana.

Frost et al. (2006a) provided a generalmotivation for dismantling paraphyleticgroups and a specific rationale for proposedchanges to Bufo:

‘‘A serious impediment in amphibian biology,and systematics generally, with respect toadvancing historically consistent taxonomies,

is the social conservatism resulting in thewillingness of many taxonomists to embrace, ifonly tacitly, paraphyletic groupings, evenwhen the evidence exists to correct them.The reason for this is obvious. Recognizingparaphyletic groups is a way of describing treesin a linear way for the purpose of telling great[emphasis theirs] stories and providing favoredcharacters a starring role.’’ (p. 12).‘‘A complete remedy of the polyphyly/para-phyly of Bufo is beyond the scope of this study,although we take limited actions to start thisinevitable process. We could place all of thenames that are demonstrably derived from‘Bufo’ into the synonymy of Bufo, therebyproviding a monophyletic taxonomy. Howev-er, because much of this paraphyly wasunderstood in 1972 (various papers in Blair,1972a), it is clear that social inertia is standingin the way of progress. We judge that progresswill require the partition of ‘Bufo’ into moreinformative natural units.’’ (p. 214)

Frost et al. (2006a) considered the parti-tioning of the paraphyletic genus Bufo intosmaller generic units to be ‘‘progress.’’ Here,progress is confused with change. We feel thisproposed change demands needless taxonom-ic chaos at the expense of stability. Certainly,we do not oppose naming newly discoveredclades, but the solution offered by Frost et al.(2006a) and mirrored in Frost et al. (2008) is notthe only one. There are two general classes ofsolutions for resolving the paraphyly of Bufo.One is to break Bufo up into smaller monophy-letic taxa at the same rank. A second is to retainBufo as a large group and to subsume within itthe taxa that make Bufo paraphyletic. This canbe done either by reducing the rank of thesmaller clades or using unranked taxa.

Frost et al. (2006a) favored the first option,dismantling Bufo into smaller genera. Asquoted above, part of their justification is thatrecognition of a greater number of smallergenera is more ‘‘informative,’’ meaning thatmore clades are named. However, the secondoption also results in at least as many namedclades. Moreover, the second option containsmore information: a larger genus Bufo is nowmonophyletic as are the smaller infragenerictaxa. Both subgenera and unranked infrage-neric taxa have been used increasingly (e.g.,


FIG. 1.—Phylogenies of Bufo from (A) a maximum likelihood re-analysis of Frost et al.’s (2006a) 12S–16S data for thebufonids and (B) Frost et al. (2006a). The likelihood analysis was conducted on a secondary structure based alignment.Support values are Bayesian posterior probabilities from 16,000 post burn-in trees. The Bayesian analysis used two runsof four Markov chains each with each run lasting ten million generations and trees sampled once every 1000 generations.In (B), support values are from a parsimony jackknife with 37% deletion of Frost et al’s 12S–16S data for the bufonids(the deletion percentage was chosen to mimic Frost et al’s. [2006a] analyses, although they did not report the specific


Brandley and de Queiroz, 2004; Hillis andWilcox, 2005; Leache and McGuire, 2006),and this solution maintains stability withoutsacrificing informativeness.

The solution espoused by Frost et al. (2008)for North American Rana is even moreegregious in its subjectivity. Frost et al.(2008) arbitrarily divided the clade of Laur-asian Rana species recovered by Hillis andWilcox (2005) and Hillis (2007) into twogenera, Rana and Lithobates. This action isnot necessitated by the phylogenetic analysis,it does not remove any paraphyletic groups, itprovides no new phylogenetic informationbeyond that in Hillis and Wilcox (2005), andit unnecessarily promotes taxonomic instabil-ity and confusion (Hillis, 2007; Smith andChiszar, 2006). Furthermore, the clades calledRana and Lithobates by Frost et al. (2008) areboth inconsistently supported across analysesof various data sets (as discussed by Hillis andWilcox, 2005). In contrast, support for themonophyly of the two combined groups (Ranasensu Hillis, 2007) is consistent and strong.

Hillis (2007) suggested that Rana beretained for all North American ranids andthat Lithobates be recognized as one ofseveral subclades of Rana. This proposalretains the existing species names but alsoprovides more information about the phylog-eny of these frogs. Hillis and Wilcox (2005)originally defined Lithobates as a subcladeequivalent to the Rana palmipes speciesgroup, and treated this taxon as a subgenusunder the Linnean system. Moreover, thistaxon concept of Lithobates is much morerestricted than the genus called Lithobates byFrost et al. (2006a). The brief comments inFrost et al. (2008) give the reader no hint ofthese issues.

Alternative Proposals for Bufo and Rana

Interestingly, the various groups of authorswithin Crother (2008) chose very differenttaxonomic strategies. In contrast to Frost et al.

(2008), de Queiroz and Reeder (2008) madethe sensible decision to recognize the sub-clades of Phrynosoma proposed by Leacheand McGuire (2006). Thus, stability of thewell-known genus is retained, but additionalphylogenetic information is available (e.g.,Phrynosoma (Anota) solare).

The proposal we advocate for Bufo andRana is similar to that of Hillis (2007) and deQueiroz and Reeder (2008): a hybrid systemusing ranked scientific names to maintainstability, and unranked names to conveyadditional phylogenetic information. This ap-proach combines features of the ZoologicalCode and of phylogenetic nomenclature(Cantino et al., 1999; de Queiroz and Gau-thier, 1992). Authors who prefer completelyranked taxonomies can follow these samerecommendations, treating the unrankednames as taxa of infrageneric rank.

Bufo.—The North American bufonidstreated by Frost et al. (2008) as the generaRhinella, Ollotis, and Anaxyrus form a well-supported clade termed the New World Cladeby Pauly et al. (2004). The sister taxon to thisclade, however, is not strongly supported inany study. Despite the apparently strongsupport values reported by Frost et al.(2006a), the support for relationships amongthese clades (Fig. 1) is an artifact of theirunusual analysis parameters (low gap cost,treating gaps as a fifth character, and ajackknife deletion setting of 37%, all of whichinflate support values relative to standardapproaches such as the bootstrap and jack-knife analysis with 50% deletion). Our re-analysis of their 12S–16S bufonid data (fromD. C. Cannatella and G. B. Pauly, unpub-lished data; sequence data aligned in CLUS-TALX with modifications based on secondarystructure, as in Pauly et al. [2004] followed bymodel-based analyses in GARLI [Zwickl,2006] and MrBayes 3.1.1 [Ronquist andHuelsenbeck, 2003]) suggests that B. bufo isthe sister taxon of the New World Clade

r

value). This re-analysis was conducted on the identical sequence data as examined in (A), but the alignment is takendirectly from Frost et al. (2006a). Asterisks indicate support values of 100. Classifications recommended by us (A) and byFrost et al. (2006a, 2008) (B) are shown. Shaded boxes denote the New World Clade (sensu Pauly et al., 2004). SeeAppendix I for synonymy of Ollotis with Incilius.


(Fig. 1A). Similarly, results from Pauly et al.(2004), Pramuk (2006), and Pramuk et al.(2007), indicate that (1) the New World Cladesensu Pauly et al. (2004) is very stronglysupported; this clade includes all of the NorthAmerican species of the list; and (2) the NewWorld Clade is closely related to, or the sister-group of, the Bufo bufo species group.

Therefore, we delimit Bufo as minimallyincluding the New World Clade (115 species)and the Bufo bufo species group (Fig. 1).Thus, a large number of the species names arereturned to their previous combinations.Although our taxonomic action here recogniz-es a larger Bufo, no phylogenetic informationis lost because we regard Anaxyrus, Incilius(see Appendix I for synonymy of Ollotis asIncilius), and Rhinella simultaneously assubgenera and named clades within Bufo.We defer formal definitions of the cladenames pending a re-classification of allBufonidae in progress. At the moment, theexhaustive content of Bufo as we delimit it is notclear, because of the uncertain phylogeneticposition of some clades. But we note that Frostet al. (2006a) also retained about two dozenspecies as ‘‘Bufo’’ because of uncertain affini-ties. Our proposal is only partly satisfactory, andas taxon sampling and gene sampling isincreased, it may be that other clades arereturned to Bufo. Nonetheless, it is an improve-ment over the classification of Frost et al.(2006a, 2008) because taxonomic stability isenhanced without sacrificing phylogenetic in-formation.

To be clear, our advocacy of a larger Bufo isnot based on social conservatism (see above),sentimentality for paraphyletic groups (noneare recognized), telling great stories, or ageneral preference for traditional approaches.Rather, it is motivated solely by the desirabil-ity for stability in classifications/taxonomiesprovided that such stability does not sacrificephylogenetic information.

Rana.—For species of Rana, the taxonomyrecommended by Hillis (2007:Fig. 2) is com-pletely consistent with the Code (contrary toearlier concerns about some of these namesreported by Frost et al. [2006a]). It isimportant to note that the different recom-mendations by Hillis (2007) and Frost et al.(2008) do not reflect any differences in

phylogenetic results. Indeed, the phylogenyof virtually all species of North AmericanRana by Hillis and Wilcox (2005:Fig. 2A) andthat from a much smaller sample of NorthAmerican Rana by Frost et al. (2006a:Fig. 2B)differ in minor ways that do not affect eithertaxonomic proposal. Moreover, the cladesFrost et al. (2006a) called Rana and Litho-bates are two of the few named clades that arenot consistently supported in phylogeneticanalyses of North American ranids. Themonophyly of these groups is controversial(especially with regard to the placement ofRana sylvatica; for a review of this issue, seeHillis and Wilcox, 2005). Use of Lithobates forthe majority of North American Rana is notonly unnecessary and destabilizing for taxo-nomy, it is also highly confusing. The onlyclear phylogenetic definition of the cladeLithobates was provided by Hillis and Wilcox(2005), who defined it as ‘‘The clade stemmingfrom the most recent common ancestor ofRana palmipes Spix 1824, Rana vaillantiBrocchi 1877, Rana bwana Hillis and de Sa1988, and Rana juliani Hillis and de Sa 1988.’’Under this phylogenetic definition and thephylogenies of Hillis and Wilcox (2005) andFrost et al. (2006a), no species in the cladeLithobates occur within the United States.

The taxonomy of Frost et al. (2006a) notonly destabilizes the names of well knownspecies, it actually provides much less infor-mation on the phylogeny of North Americanranids compared to the taxonomy proposedby Hillis (2007:Fig. 2); Frost et al. (2006a)recognized two named clades while Hillis(2007) recognized nine (Fig. 2). In addition,unlike the taxonomy proposed by Frost et al.(2006a), the named clades proposed by Hillis(2007) reflect groups that are strongly andconsistently supported across multiple analy-ses and data sets. Therefore, the classificationproposed by Hillis (2007) better fulfills bothgoals of biological classification (facilitatingcommunication and reflecting evolutionaryrelationships).

SUMMARY AND RECOMMENDATIONS

Classifications are not the personal sand-boxes of taxonomists. Promoting unnecessaryname changes has a huge negative impact onthe field of taxonomy and more generally on


FIG. 2.—Phylogenies of Rana from (A) Hillis and Wilcox (2005) and (B) Frost et al. (2006a). Asterisks in (A) indicatethose species that were sampled in both studies. Roman numerals denote those relationships that were reconstructeddifferently in the two studies. In (A), the phylogeny is from Hillis and Wilcox (2005), and the taxonomicrecommendations follow Hillis (2007). Shaded boxes denote the clade that Frost et al. (2006a) termed Lithobates.


biology as a whole. It is a mistake to assumethat nonsystematists have no interest or stakein scientific names. (In this context, one of us[DCC] recognizes that his previous positionon changing generic names simply to havemore ‘‘informative’’ taxa—recognizing Silur-ana as distinct from Xenopus [Cannatella andde Sa, 1993]—was mistaken.)

Historically, a major rationale for usingscientific names over common names was topromote efficient communication becausecommon names varied widely among users,regions, and languages. Ironically, because ofthe recent flux in scientific names and thepublication of standard lists for commonnames, common names often show morestability (at least for standard English names).Although Lithobates catesbeianus and Ranacatesbeiana may confuse some, nearly all usersrecognize the American Bullfrog. If subjectiveand unnecessary changes to taxon namesprevail, the result will be not only a greatdisservice to the science of taxonomy but alsoto all who use the results of taxonomy, fromlaypersons to professional biologists. Suchchanges have already brought the ire of manybiologists (Godfray and Knapp, 2004) and onlypromote the increasing disregard of the field.In recognizing these concerns and objections,we take seriously the obligation of taxonomiststo serve the larger community with stable andinformative classifications.

These problems are not unique to herpe-tology. All biology is faced with changingclassifications and the difficulties of updatingnomenclature. Taxonomic specialists are a tinyfraction of the users of the taxonomies, andspecialists should be cognizant of the diverseuser community and its needs. Most important-ly, changes to classifications that attempt tomaximize information content must minimizeany negative impacts on utility. So in summary,we argue that taxonomic stability and mono-phyly are primary goals, but stability is not to beachieved at the expense of monophyly.

Below we make several recommendations;these are not exhaustive, but are intended tostimulate further discussion.

1. Because it is contrary to the letter andspirit of the Code, scientific societiesshould not regulate scientific names.

Lists of names should not be treated aslists of officially accepted scientificnames. To avoid confusion, sponsorshipof a list should not be indicated by theword ‘‘official.’’ However, name lists areimportant for stabilizing common names,so herpetological societies serve a vitalfunction in promulgating such lists.

2. Stable taxonomies of scientific names arean important goal, and scientific societiesshould promote the discussion and inter-action needed to achieve this. Systema-tists should accept responsibility forproviding stable and informative classifi-cations to the biological community, andthe community of users should rejectnew classifications that destabilize namesunnecessarily.

3. The scientific name of the species(genus-species combination) should onlybe changed when there is strong evi-dence that the changes are necessary toreflect evolutionary history.

4. Authors of lists should discuss alternativetaxonomic proposals when these exist.

5. Authors working on taxa for which thereare multiple recent or current proposedscientific names should explain thechoice of names, i.e., their choice oftaxon concept.

6. Name lists should not be used to restricttaxonomic freedom. Authors and editorsshould be aware that no requirementexists (in the Code or anywhere else) thatthe most recent taxonomic proposals befollowed. Editors should not establishpolicies that a particular list or taxonomicrevision be followed.

7. All name lists, and especially thosepromoted by scientific societies, shouldundergo extensive peer review. This willensure rigorous scholarship, discussionsof alternative proposals, and a broaderconsensus of the societies’ members.

Acknowledgments.—We sought comments and feed-back from numerous individuals. Although not everyoneagreed entirely with our opinions, we are grateful to all ofthe following for their helpful discussions and comments:A. J. Abrams, J. M. Brown, C. M. Edwards, O. Flores-Villela, D. M. Green, M. Guerra, J. Hanken, T. R. Jones,V. Leon-Regagnon, A. Larson, J. B. Losos, E. J. Routman,M. J. Ryan, J. C. Santos, M. J. Sredl, B. L. Stuart, J. N.Stuart, S. Swenson, R. Symula, R. J. Wassersug and


especially to R. M. Brown, F. T. Burbrink, B. I. Crother,K. de Queiroz, R. W. McDiarmid, P. C. Rosen, H. M.Smith, M. Vences, and D. B. Wake. Preparation of thispaper was supported by a University of Texas ContinuingFellowship to GBP and NSF grant EF 0334952 to DCCand DMH.

LITERATURE CITED

AMERICAN ORNITHOLOGISTS’ UNION. 1998. Check-list ofNorth American Birds, 7th ed. American Ornitholo-gists’ Union, Washington, D. C., U.S.A.

ASHLOCK, P. D. 1979. An evolutionary systematist’s view ofclassification. Systematic Zoology 28:441–450.

BENTON, M. J. 2000. Stems, nodes, crown clades, andrank-free lists: Is Linnaeus dead? Biological Reviews75:633–648.

BRANDLEY, M. C., AND K. DE QUEIROZ. 2004. Phylogeny,ecomorphological evolution, and historical biogeogra-phy of the Anolis cristatellus series. HerpetologicalMonographs 18:90–125.

CANNATELLA, D. C., AND R. O. DE SA. 1993. Xenopus laevisas a model organism. Systematic Biology 42:476–507.

CANTINO, P. D., H. N. BRYANT, K. DE QUEIROZ, M. J.DONOGHUE, T. ERIKSSON, D. M. HILLIS, AND M. S. Y.LEE. 1999. Species names in phylogenetic nomencla-ture. Systematic Biology 48:790–807.

COLLINS, J. T., J. E. HUHEEY, J. L. KNIGHT, AND H. M.SMITH. 1978. Standard common and current scientificnames for North American amphibians and reptiles, 1sted. SSAR Herpetological Circular 7:1–36.

CONANT, R., F. R. CAGLE, C. J. GOIN, C. H. LOWE, JR.,W. T. NEILL, M. G. NETTING, K. P. SCHMIDT, C. E.SHAW, AND R. C. STEBBINS. 1956. Common names forNorth American amphibians and reptiles. Copeia1956:172–185.

COPE, E. D. 1863. On Trachycephalus, Scaphiopus, andother American Batrachia. Proceedings of the Academyof Natural Sciences of Philadelphia 15:43–54.

COPE, E. D. 1889. The Batrachia of North America.Bulletin of the United States National Museum34:1–525.

CROTHER, B. I. (Ed.). 2000. Scientific and standardEnglish names of amphibians and reptiles of NorthAmerica north of Mexico, with comments regardingconfidence in our understanding. SSAR HerpetologicalCircular 29:1–82.

CROTHER, B. I. 2007. Standard language (insert languageof choice) names versus common names. Herpetolog-ical Review 38:143.

CROTHER, B. I. (Ed.). 2008. Scientific and standardEnglish names of amphibians and reptiles of NorthAmerica north of Mexico, with comments regardingconfidence in our understanding, 6th ed. SSARHerpetological Circular 37:1–84.

DE QUEIROZ, K. 1988. Systematics and the DarwinianRevolution. Philosophy of Science 55:239–259.

DE QUEIROZ, K. 1997. The Linnean hierarchy and theevolutionization of taxonomy, with emphasis on theproblem of nomenclature. Aliso 15:125–144.

DE QUEIROZ, K., AND J. GAUTHIER. 1992. Phylogenetictaxonomy. Annual Review of Ecology and Systematics23:449–480.

DE QUEIROZ, K., AND T. W. REEDER. 2008. Squamata:Lizards. Pp. 24–45. In B. I. Crother (Ed.), Scientificand standard English names of amphibians and reptilesof North America north of Mexico, with commentsregarding confidence in our understanding. SSARHerpetological Circular 37.

DUNN, E. R., AND M. T. DUNN. 1940. Generic namesproposed in herpetology by E. D. Cope. Copeia1940:69–76.

FROST, D. R., T. GRANT, J. FAIVOVICH, R. H. BAIN, A. HAAS,C. F. B. HADDAD, R. O. DE SA, A. CHANNING, M.WILKINSON, S. C. DONNELLAN, C. J. RAXWORTHY, J. A.CAMPBELL, B. L. BLOTTO, P. MOLER, R. C. DREWES, R. A.NUSSBAUM, J. D. LYNCH, D. M. GREEN, AND W. C.WHEELER. 2006a. The amphibian tree of life. Bulletin ofthe American Museum of Natural History 297:1–370.

FROST, D. R., T. GRANT, AND J. R. MENDELSON, III. 2006b.Ollotis Cope, 1875 is the oldest name for the genuscurrently referred to as Cranopsis Cope, 1875 (Anura:Hyloides: Bufonidae). Copeia 2006:558.

FROST, D. R., R. W. MCDIARMID, AND J. R. MENDELSON,III. 2008. Anura: Frogs. Pp. 2–12. In B. I. Crother(Ed.), Scientific and standard English names ofamphibians and reptiles of North America north ofMexico, with comments regarding confidence in ourunderstanding. SSAR Herpetological Circular 37.

GAFFNEY, E. S. 1979. An introduction to the logic ofphylogeny reconstruction. Pp. 79–111. In J. Cracraftand N. Eldridge (Eds.), Phylogenetic Analysis andPaleontology. Columbia University Press, New York,New York, U.S.A.

GODFRAY, H. C. J., AND S. KNAPP. 2004. Introduction.Philosophical Transactions of the Royal Society ofLondon, Series B 359:559–569.

HILLIS, D. M. 2007. Constraints in naming parts of theTree of Life. Molecular Phylogenetics and Evolution42:331–338.

HILLIS, D. M., AND T. P. WILCOX. 2005. Phylogeny of theNew World true frogs (Rana). Molecular Phylogeneticsand Evolution 34:299–314.

INTERNATIONAL COMMISSION ON ZOOLOGICAL NOMENCLA-

TURE. 1999. International Code of Zoological Nomen-clature, 4th ed. International Trust for ZoologicalNomenclature, London, U.K.

KELLOGG, R. 1932. Mexican tailless amphibians in theUnited States National Museum. Bulletin of the UnitedStates National Museum 160:1–224.

KENNEDY, J., R. HYAM, R. KUKLA, AND T. PATERSON. 2006.Standard data model representation for taxonomicinformation. OMICS 10:220–230.

LEACHE, A. D., AND J. A. MCGUIRE. 2006. Phylogeneticrelationships of horned lizards (Phrynosoma) based onnuclear and mitochondrial data: evidence for amisleading mitochondrial gene tree. Molecular Phylo-genetics and Evolution 39:628–644.

MAYR, E. 1982. The Growth of Biological Thought:Diversity, Evolution, and Inheritance. Harvard Univer-sity Press, Cambridge, Massachusetts, U.S.A.

MAYR, E., E. G. LINSLEY, AND R. L. USINGER. 1953. Methodsand Principles of Systematic Zoology. McGraw-Hill BookCompany, New York, New York, U.S.A.

PAULY, G. B., D. M. HILLIS, AND D. C. CANNATELLA. 2004.The history of a Nearctic colonization: Molecular


phylogenetics and biogeography of the Nearctic toads(Bufo). Evolution 58:2517–2535.

PRAMUK, J. B. 2006. Phylogeny of South American Bufo(Anura: Bufonidae) inferred from combined evidence.Zoological Journal of the Linnean Society 146:407–452.

PRAMUK, J. B., T. ROBERTSON, J. W. SITES, JR., AND B. P.NOONAN. 2007. Around the world in 10 million years:biogeography of the nearly cosmopolitan true toads(Anura: Bufonidae). Global Ecology and Biogeography17:72–83.

RONQUIST, F., AND J. P. HUELSENBECK. 2003. MRBAYES 3:Bayesian phylogenetic inference under mixed models.Bioinformatics 19:1572–1574.

ROSS, H. H. 1974. Biological Systematics. Addison-WesleyPublishing Company, Reading, Massachusetts, U.S.A.

SCHMIDT, K. P. 1953. A Check List of North AmericanAmphibians and Reptiles, 6th ed. University of ChicagoPress, Chicago, Illinois, U.S.A.

SIMPSON, G. G. 1961. Principles of animal taxonomy.Columbia University Press, New York, New York,U.S.A.

SMITH, H. M., AND D. CHISZAR. 2006. Dilemma of name-recognition: Why and when to use new combinations ofscientific names. Herpetological Conservation andBiology 1:6–8.

STUART, J. N. 2008. Book Review: Scientific and StandardEnglish Names of Amphibians and Reptiles of NorthAmerica North of Mexico, with Comments RegardingConfidence in our Understanding, Sixth Edition, by theCommittee on Standard English and Scientific Names(Brian I. Crother, committee chair). Bulletin of theChicago Herpetological Society 43:1–3.

TURTLE TAXONOMY WORKING GROUP [J. W. Bickham, J. F.Parham, H.-D. Philippen, A. G. J. Rhodin, H. B.Shaffer, P. Q. Spinks, and P. P. van Dijk]. 2007. Turtletaxonomy: methodology, recommendations, and guide-lines. Pp. 73–84. In H. B. SHAFFER, N. N. FITZSIMMONS,A. GEORGES, AND A. G. J. RHODIN (Eds.), DefiningTurtle Diversity: Proceedings of a Workshop onGenetics, Ethics, and Taxonomy of Freshwater Turtlesand Tortoises. Chelonian Research Monographs, 4.Chelonian Research Foundation, Lunenburg, Massa-chusetts, U.S.A.

VENCES, M. 2007. The Amphibian Tree of Life: Idealogie,chaos, oder biologische realitat? Zeitschrift fur Feld-herpetologie 14:153–162.

ZWICKL, D. J. 2006. Genetic Algorithm Approaches for thePhylogenetic Analysis of Large Biological SequenceDatasets under the Maximum Likelihood Criterion.

Ph.D. dissertation, The University of Texas at Austin,Austin, Texas, U.S.A.

.Accepted: 20 May 2009

.Associate Editor: Frank Burbrink

APPENDIX I

The Taxonomy of Middle American Bufo

Frost et al. (2006a) resurrected Cranopsis Cope 1875for the clade of Middle American Bufo identified byPauly et al. (2004:Fig. 1A). Later, Frost et al. (2006b)pointed out their oversight: that Cranopsis was preoc-cupied, that Cope himself had realized this (1889) andproposed the replacement name Cranophryne, and thatOllotis Cope 1875 had priority as the valid name for thegenus.

The story continues because another name, InciliusCope 1863, is relevant. Cope (1863) did not designate atype-species for Incilius. Frost et al. (2006a:222), actingunder the First Reviser principle (Article 24) designatedBufo cognatus Say 1823 as the type-species of InciliusCope 1863 to ensure that Incilius would be a synonym ofAnaxyrus (i.e., the Nearctic Bufo sensu Pauly et al. [2004],which includes Bufo cognatus: see Fig. 1A). However,Frost et al. (2006a) were apparently unaware that a type-species for Incilius had already been designated. Asindicated by Dunn and Dunn (1940), Kellogg (1932:29)designated Bufo coniferus Cope 1862, one of the speciesoriginally included in the genus, as the type-species ofIncilius. Bufo coniferus is in the Middle American Clade(sensu Pauly et al., 2004), which is treated as Ollotis byFrost et al. (2008).

Under Article 70.2, Kellogg’s fixation of type-speciestakes precedence over that of Frost et al. (2006a).Therefore, Incilius has priority over Ollotis as the olderavailable name for the genus comprising species in theMiddle American Clade of Bufo. However, readers shouldnot construe our statement about priority of Incilius overOllotis as being a taxonomic action (see Article 8.3 of theCode) or ‘‘synonymy by implication.’’ We do not recognizeOllotis or Incilius as the valid genus name for the MiddleAmerican Clade, in order to avoid the 33 new combina-tions that would result. Rather, as discussed and depictedin Figure 1, we consider these species in the MiddleAmerican Clade to be part of Bufo.


HERPETOLOGICA · HERPETOLOGICA VOL. 65 JUNE 2009 NO. 2 Herpetologica, 65(2), 2009, 115–128 E 2009...

Documents

Transcript of HERPETOLOGICA · HERPETOLOGICA VOL. 65 JUNE 2009 NO. 2 Herpetologica, 65(2), 2009, 115–128 E 2009...