2006 Dubois MPE Cautionary Tale

Molecular Phylogenetics and Evolution 42 (2007) 317–330www.elsevier.com/locate/ympev

Naming taxa from cladograms: A cautionary tale

Alain Dubois ¤

Vertébrés: Reptiles et Amphibiens, USM 0602 Taxonomie et Collections, Département de Systématique et Evolution, Muséum national d’Histoire naturelle, 25 rue Cuvier, 75005 Paris, France

Received 6 April 2006; revised 1 June 2006; accepted 7 June 2006Available onine 17 June 2006

Abstract

The recent publication of a new hypothesis of cladistic relationships among American frogs referred to the genus Rana, accompaniedby a new taxonomy and a new nomenclature of this group [Hillis D.M., Wilcox, T.P., 2005. Phylogeny of the New World true frogs(Rana). Molecular Phylogenetics and Evolution 34, 299–314], draws attention to the problems posed by the use of a “double nomencla-ture”, following both the rules of the International Code of Zoological Nomenclature (designated here as “onomatophore-based nomencla-ture”) and the rules of the draft Phylocode (designated here as “deWnition-based nomenclature”). These two nomenclatural systems,which rely upon widely diVerent theoretical bases, are incompatible, and the latter cannot be viewed as a “modiWcation” of the former.Accordingly, scientiWc names (nomina) following both systems should be clearly distinguished in scientiWc publications. Onomatophore-based nomina should continue to be written as they have been for about 250 years, whereas deWnition-based nomina should be written ina speciWc way, e.g., �LITHOBATES�. The combined use of both nomenclatural systems for the same taxonomy in the same paper requiresgood knowledge and careful respect of the rules of the Code regarding availability, allocation and validity of nomina. As shown by thisexample, not doing so may result in various problems, in particular in publishing nomina nuda or in using nomenclatural ranks invalidunder the current Code. Attention is drawn to the fact that new nomina published without diagnostic characters are not available underthe Code, and that the latter currently forbids the use of more than two ranks (subgenus and “aggregate of species”) between the ranksgenus and species.© 2006 Elsevier Inc. All rights reserved.

Keywords: Phylogeny; Taxonomy; Code; Phylocode; Onomatophore-based and deWnition-based nomenclatures; Characters; Ranks; Rana; Frogs

1. Introduction

Most biologists nowadays agree that biological classiW-cation should provide information on the phylogeneticrelationships between organisms. Although the real tree oflife will probably remain unknown, at least in its details,hypotheses can be proposed regarding the structure of thistree, and these hypotheses (cladograms) can be taken as abasis for building classiWcations (taxonomies). Once recog-nized by biologists, taxonomic units or taxa (singulartaxon) are usually denominated by “scientiWc names” ornomina (singular nomen; Dubois, 2000). Although oftenconfounded, the two processes are distinct: taxa can be

* Fax: +33 1 40 79 34 88.E-mail address: [email protected]

1055-7903/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ympev.2006.06.007

recognized but not named (e.g., in systems called “numericlatures”; e.g., Johnson, 1970) and nomina can be createdwithout designating taxa (e.g., the nomina nuda of zoologi-cal and botanical nomenclatures). In order for nomina toallow eYcient, unambiguous and universal communicationamong biologists, their allocation to taxa requires follow-ing precise and stringent rules. Several nomenclatural sys-tems have been and still are used by biologists.

In zoology, the traditional nomenclatural system isdescribed in the International Code of Zoological Nomencla-ture (Anonymous, 1999; Polaszek and Wilson, 2005), some-times abbreviated ICZN, and designated below as “theCode”. Among others, a new system of nomenclature forliving organisms, initially called “phylogenetic nomencla-ture” (De Queiroz and Gauthier, 1990, 1994), was recentlyformalised in the project of a new code called “the

mailto: [email protected]

mailto: [email protected]

318 A. Dubois / Molecular Phylogenetics and Evolution 42 (2007) 317–330

Phylocode” (Laurin and Cantino, 2004; <http://www.ohiou.edu/phylocode>; see also Dubois, 2005a: note6). This new system has elicited a number of comments,both favorable (recent references: Pleijel and Rouse, 2003;Donoghue and Gauthier, 2004; Laurin and Cantino, 2004;Laurin et al., 2005, 2006; De Queiroz, 2006) and unfavor-able (recent references: Kraus, 2004; Sluys et al., 2004;Dubois, 2005a; Pickett, 2005a,b; Rieppel, 2006). The pur-pose of the present paper is not to discuss again the respec-tive merits of the Code and of the Phylocode, but to addressa speciWc question: can the same nomina be available andvalid under both systems, and if so under which conditions?

This question is not only theoretical, as will be shownhere with the detailed analysis of the nomenclatural prob-lems posed by a recent paper (Hillis and Wilcox, 2005). TheCode and the Phylocode are widely distinct nomenclaturalsystems, as they rely on diVerent theories of nomenclature.The use of a nomenclature following the Code, either aloneor in parallel with a nomenclature following the Phylocodefor the same taxonomy, requires a good understanding andknowledge of the Code. To be valid under the Code, anomen must have strictly followed the rules in force at sev-eral distinct steps of the nomenclatural process. As anyjuridical text, the Code is not a simple, “self-speaking” textthat could be mastered by simple intuition or an impres-sionist approach, but its proper use requires spending sometime and eVort reading and understanding it (Dubois,2005a). Some of the problems raised in the example ana-lysed here are not unique to this publication and, if no careis taken from the part of editors and referees of scientiWcpublications, they may tend to become more and more fre-quent, thus contributing to undermining the universalityand reliability of zoological nomenclature. Starting fromthis paper, the following problems are discussed below: (1)the publication of nomina nuda for taxa recognized on thebasis of a cladogram, often obtained from molecular data,but for which no diagnostic characters are provided; (2) theinvalid use of junior homonyms; (3) the invalid use ofjunior synonyms; (4) the use of ranks not allowed in thecurrent Code; (5) problems in the etymological formationof nomina.

Full understanding of the text below requires a few pre-liminary clariWcations:

(1) A distinction is made between taxa and clades.Although the aim of taxonomy is to provide a classiW-catory scheme that accounts for the patterns of evolu-tion and phylogeny of organisms, a distinction ismade here between these patterns or clades, and taxa.Taxa are concepts or models used in biological classi-Wcation, that account for some particularities of theorganisms. Under a cladistic approach of taxonomy,a taxon can be recognized only for a group of organ-isms that is considered holophyletic, a concept forwhich the synonymous terms phylon (Dubois, 1991),cladon (Mayr, 1995) and phylo-taxon (Joyce et al.,2004) have been coined. The decision to recognize

such taxa for some organisms rests on hypothesesabout the real clades of the real world, but these taxaare concepts, not the clades themselves. Thus, if cladesmay be viewed ontologically as individuals, it is notthe case of taxa, which are classes of organisms, rec-ognized on the basis of criteria such as hypothesizedrelationships.

(2) Further distinctions are made between taxa and nom-ina, and between nomina and ranks. Taxa are taxo-nomic classes of organisms, whereas nomina are“just” labels given to these taxa to designate themunambiguously and universally, but without anyexplanatory, descriptive or other rôle or value bythemselves (Dubois, 2005a,b, 2006a). This distinctionis clear in the Code, but not so in the Phylocode. TheCode only provides rules for the allocation of nominato taxa, not for the deWnition of taxa, so that thereexists no such things as “ICZN-taxa” (Joyce et al.,2004), whereas in the Phylocode the deWnition of thetaxon and of the nomen are simultaneous. It is fullyconceivable to deWne taxa under the philosophy ofthe Phylocode (e.g., by node-based, stem-based orapomorphy-based deWnitions), but to name themunder the Code philosophy (using onomatophoresand nominal-series). Ranks as used in the Code aredistinct from nomina. The same nomen can be used todesignate distinct (i.e., more or less inclusive) taxa atdiVerent ranks in the nomenclatural hierarchy. This isbecause of the existence in the Code of a Rule ofCoordination, discussed in more detail below. NoRule of Coordination exists in the Phylocode.

(3) In a paper like this, adjectives or other devices areneeded to qualify the rules or principles that follow thetwo basic philosophies of nomenclature discussed here.This need appears in all recent publications devoted to adiscussion or comparison of these two systems, but nosatisfactory solution has been found so far. Phrases suchas “names governed by the Phylocode” or “names gov-erned by the preexisting rank-based codes”, as founde.g., in Article 6.1 of the draft Phylocode, are too cum-bersome to be used repeatedly in a publication. Adjec-tives directly based on the terms Code and Phylocode,such as “codian” or “phylocodian”, would be barba-risms, as are compound names based on abbreviations,such as “ICZN-taxon names” (Joyce et al., 2004). Theuse of an adjective like “phylogenetic” to designatenomenclature under the Phylocode is not appropriate asit would seem to mean that nomenclature under theCode is not compatible with a taxonomy based on cla-distic hypotheses, which is simply wrong. Lidén et al.(1997) found an elegant and unpolemical way to desig-nate the two systems, calling the Code “Charlie” and thePhylocode “Phyllis”, but these names do not allow theformation of clear, short and euphonious adjectives.Another solution to this problem may be to call bothsystems after the patronyms of their original propo-nents. The adjective “Linnaean” has been widely used

http://www.ohiou.edu/phylocode



A. Dubois / Molecular Phylogenetics and Evolution 42 (2007) 317–330 319

to designate the nomenclature following the Code, butthis is largely misleading (Moore, 2003; Dubois, 2005a)as the rules of the latter are widely diVerent from theoriginal nomenclatural rules of Linnaeus (1758). Itwould be more justiWed to call these rules “Linnaeanand Stricklandian” or “Linnaean-Stricklandian”, to dojustice to the major contribution of the so-called“Strickland’s code” (Strickland et al., 1843) to the build-ing of the current Code. A counterpart to this denomi-nation for the Phylocode would be “Queirozian andGauthierian”, or “Queirozian-Gauthierian”, althoughin this case also the system initially proposed by DeQueiroz and Gauthier (1990, 1994) has gone throughimportant changes since then. I proposed elsewhere(Dubois, 2005a, 2006a) the shorter term “Queirau-thian”, a combination–contraction of the latter denomi-nation, which in my mind was in no way “insulting”,but it was apparently understood as such by some read-ers, so I suggest abandoning it; furthermore, use of thisdenomination might be embarrassing for the authors ofthis system. Another possibility is to use terms based onsome of the particularities of the nomenclatural systemsat stake, but the choice of terms must be done carefullyas some designations may be misleading. Thus, the cur-rent Code is certainly “rank-based”, but the nomencla-tural philosophy of the Code would not be drasticallychanged if the Code abandoned ranks, as exempliWed bythe rules proposed by Dubois (2006a) for the nomencla-ture of higher taxa. The genuine basic diVerencebetween the two systems is the way nomina are allo-cated to taxa, by pointing to onomatophores in theCode and by phylogenetic deWnitions in the Phylocode(Dubois, 2005a). The Wrst system can thus be called“onomatophore-based”, and the second one “deWni-tion-based”. These designations are neuter, unpolemi-cal, and they designate accurately the main diVerencebetween the two systems, so they will be used below.Alternatively, the expressions “ostension-based” and“intension-based”, which have the same meaning (seebelow), can also be used, but they may appear lessfamiliar to many readers.

(4) In the discussion below, generic and speciWc nominafollowing the Code are written as they have been inmost publications for more than one century, i.e., insmall characters Italics: Lithobates or Lithobates Fitz-inger, 1843. In contrast, nomina following the Phylo-code are written under the format proposed byDubois (2005a): �LITHOBATES�, �LITHOBATES Hillis andWilcox, 2005� or �LITHOBATES [Fitzinger, 1843] Hillisand Wilcox, 2005�. The philosophy behind this pro-posal is discussed at more length in the Conclusion ofthe present paper.

2. Some basic features of the Code

Many recent discussions on pros and cons the currentCode stem from a widespread basic ignorance or misunder-

standing of some of its basic concepts or rules (Dubois,2005a), which is also outlined by the recent multiplicationof nomenclatural mistakes in zoological taxonomic publi-cations, even those of good quality and highly praised(Dubois, 2003). Time has come where these problemsshould be seriously considered by the editors of scientiWcperiodicals and books. Before any discussion of the prob-lems considered below, some clariWcations are needed con-cerning the Code and the Phylocode.

Proper mastering of the system of zoological nomencla-ture of the Code requires us to understand that it comprisesthree distinct operations, that can be viewed as a buildingwith three distinct “storeys” (Dubois, 2005a,b), Availability,Allocation and Validity of a nomen. All three Xoors must beexplored successively before the valid nomen of a taxon canbe established under the Code. This is not so under thePhylocode, as the Wrst two of these storeys are notseparated.

2.1. Availability of nomen

Unlike the Phylocode, the Code does not regulate allnomina, but only those from the rank subspecies to therank superfamily, excluding those of orders, classes andother higher taxa, and those of taxa below the rank sub-species. Under the Code, the nomina are distributed inthree “groups”, the “family group” (nomina of taxa ofranks family, tribe, and other ranks just above and below),the “genus group” (ranks genus and subgenus) and the“species group” (ranks species, subspecies and “aggre-gates” of the latter). Furthermore, within the two latter“groups”, only some ranks are recognized by the Code (seebelow). To avoid semantic confusions (e.g., between “spe-cies-group” as a taxon and as a nomenclatural rank), theterm nominal-series (Dubois, 2000, 2005a) is used hereinstead of “groups of names” as recognized in the Code:i.e., the species-, genus- and family-series instead of“groups”. Within each nominal-series, but not betweenthem, nomina interact concerning priority in cases of syn-onymy or homonymy. A nomen is not attached to a givenrank but, because of the Rule of Coordination, it is simul-taneously available for diVerent more or less inclusive taxaincluding its onomatophore, at all nomenclatural rankswithin one of the three nominal-series recognized by theCode. Whenever a taxon includes several subordinate taxaof the same nominal-series, the Wrst named of the latter,called “nominotypical” in the Code, bears the same nomen(with the same onomatophore, author and date) as thesuperordinate taxon: to avoid the use of the misleadingroot “type” (see below and Dubois, 2005a), the latternomen could be called epinym (from the Greek epi,“above”), and the subordinate homonymous one hyponym(from the Greek hypo, “below”). In the genus- and species-series, such a nomen is strictly the same at all ranks,whereas in the family-series, its ending has to be changedaccording to the rank. Under the Phylocode, ranks are notcompulsory, there are no nominal-series and there is no


Rule of Coordination, so any nomen is available for a sin-gle taxon, which may be more or less inclusive.

Under the Code, all nomina published after 1757 follow-ing a number of “conditions of availability” remain avail-able nowadays. Among these conditions, a deWnition of thetaxon designated by the nomen must have been provided(as a “diagnosis”, a “deWnition”, a “description” or an“indication”), but no conditions are put as to the kind ofdeWnition at stake, or as to its “quality” or “accuracy”, forreasons that were discussed elsewhere (Dubois and Ohler,1997). The only requirement is that this deWnition “states inwords characters that are purported to diVerentiate thetaxon” (Article 13.1.1), or refers to such a published state-ment. This is the only place in the Code where reference ismade to characters; the latter are not useful in the next sto-reys Allocation and Validity of the nomenclatural house.The conditions are of a diVerent kind in the draft Phylo-code, as in order for a nomen to be available (retermed“established”), the deWnition must be “phylogenetic”, buton the other hand it is not required that this deWnition men-tions characters (although this is recommended).

2.2. Allocation of nomen to taxon

Under the Code, allocation of a nomen to a taxon iscompletely distinct from its availability. It does not rely onthe deWnition of the taxon or on characters used for avail-ability of the nomen on the Wrst Xoor, but it is madethrough the use of a special tool, the “type” or better ono-matophore (Simpson, 1940, 1961). Replacement of the tradi-tional term “type” by this special technical term isnecessary to avoid the misinterpretation often repeated(e.g., recently: Pennisi, 2001; Joyce et al., 2004; Sluys et al.,2004) that the Code is “typological” (Dubois, 2005a). Thefunction of the onomatophore is not to provide charactersbut to establish an objective and permanent connectionbetween the real world of organisms and the world of lan-guage (Dubois and Ohler, 1997). Onomatophores are ofdiVerent kinds in the three nominal-series recognized by theCode: they are specimens for nomina of the species-series,species-series nominal taxa for nomina of the genus-seriesand genus-series nominal taxa for nomina of the family-series. However, ultimately, through this chain of onomato-phores, each nomen refers to one or several organisms.Thus an onomatophore-based nomen is not deWned, eitherby intension (characters or relationships) or by extension(list of included or excluded organisms or taxa), butattached to some members of the taxon (individuals ortaxa) by ostension (Keller et al., 2003; Dubois, 2005a;Kluge, 2005; Rieppel, 2006). This is not a “deWnition” of thenomen, as it does not provide limits for the taxon desig-nated by the nomen; such limits are given by the taxonomicarrangement adopted, i.e., by the other taxa recognizedwithin the same nominal-series.

Under the Phylocode, allocation of a nomen to a taxon issimultaneous with its availability and is also made throughits “phylogenetic deWnition” used on the Wrst Xoor.

Although such deWnitions rely on “speciWers”, the latter arenot equivalent to onomatophores, for two reasons at least.First, allocation of an onomatophore-based nomen to ataxon requires only that its onomatophore be included inthe taxon, without any need to associate this informationwith a cladistic hypothesis, whereas under the Phylocodesuch a hypothesis must be provided. Second, under thePhylocode, although speciWers may be specimens or taxa,they may also be characters (apomorphies, i.e., concepts).

2.3. Validity of nomen

There is no fundamental diVerence between the two sys-tems in this respect. In both systems, the valid nomen of agiven taxon depends on the taxonomic frame adopted.Under the Code, in any given taxonomic hierarchy consid-ered valid by any author, there is a single valid nomen foreach taxon. But for special exceptional cases in which anomen must be “protected” for some reasons related to“usage” (for a detailed discussion of this concept and of theproblems it raises, see Dubois, 2005a), the valid nomen isgiven automatically through the use of the Rule of Priority.In a given taxonomy and within each nominal-series,among all available nomina that potentially apply to agiven taxon (as the latter includes their onomatophores),the Wrst published is the valid one. Under the Phylocode,priority also applies, but of course only among nominaconsidered established according to this system and undertheir dates of validation in the latter. However, in somecases here also, some “well-known” nomina must be “pro-tected” to maintain “prevailing usage”.

3. Use of “the same” nomina following both sets of rules: an enlightening example

3.1. Introduction

Let us now consider the question raised above: can thesame nomina be available and valid under both systems,and if so under which conditions? Detailed analysis of arecent example will allow a better understanding of theproblems at stake. Hillis and Wilcox (2005) provided a newcladistic analysis of American frogs then referred to thegenus Rana. This analysis, based on sequences from themitochondrial genome of most known species of thisgroup, is much more thorough and reliable than all previ-ous studies. However, it is not the “Wnal word” on thisquestion, for two distinct reasons. First, although paramet-ric bootstrap analyses provided signiWcant support formost taxa, the relationships of a few species and groupswere not considered satisfactorily solved, either because ofconXicting results between the new analysis and previousstudies and other data sets, or because of Bayesianposterior probabilities below 95%. Second, and moreimportantly, this analysis was restricted to American spe-cies of the Rana assemblage, so that the limits of the taxonthat should be recognized under the latter nomen were not


known (Hillis and Wilcox, 2005: 310). More recently, Frostet al. (2006) provided a more complete analysis and newtaxonomic proposals, but their paper raises problems ofvarious kinds, in particular as the speciWc identiWcation ofsome of the voucher specimens, especially from Asia, usedin this study is questionable, and as the type-species ofmany genus-series nomina were not included in the analy-sis. This question will not be further addressed here as it isoutside the scope of this paper.

For the sake of discussion of nomenclatural problems,let us just consider the cladograms provided by Hillis andWilcox (2005, Figs. 1 and 2) as an acceptable workinghypothesis. On the basis of this cladogram, these authorsrecognized 24 distinct taxa within the undeWned “genusRana”. For 16 of these taxa, they provided deWnition-basednomina with their node-based “phylogenetic deWnitions”,but they left three taxa including several species unnamedpending further data supporting or rejecting their validity.Furthermore, following the taxonomic principles of thePhylocode against “taxonomic redundancy”, they did notprovide nomina for four identiWed taxa that are currentlybelieved to include only one species and for one that is con-sidered to include only non-American species.

An interesting feature of Hillis and Wilcox’s paper(2005), shared with another paper dealing with plethodon-tid salamanders by the same research team (Hillis et al.,2001) which will not be addressed further here but whichraises the same problems, is that it tries to provide a “dou-ble nomenclature”. The latter is believed to be valid bothunder the rules of the Code and of the proposed Phylocode.This “prudent” attitude is in contrast with that of othertaxonomists supporting the Phylocode who proposed onlynomina following the latter for their new taxa (e.g., Pleijel,1999), thus taking the “risk” to leave the possibility for oth-ers to rename the same taxa validly under the rules of theCode. However, as will be shown below, in this domain it isdiYcult or impossible to have two strings to one’s bow, asnot making a choice between both nomenclatural systemsentails the risk of making nomenclatural mistakes in atleast one of the systems.

One primary purpose of Hillis and Wilcox’s (2005) paperwas to provide a deWnition-based nomenclature for the 16taxa convincingly supported in their cladistic analysis. Fornine of these taxa, they converted genus-series nomina pro-posed by earlier authors under the Code, whereas for theother seven taxa they coined new nomina. The basic princi-ple followed for using and converting an older nomen wasto consider the overall content of the taxon originally rec-ognized under that nomen. This is a practice largely at vari-ance with nomenclature under the Code, in whichallocation of a nomen to a taxon is not based on the origi-nal extension of the taxon but on the onomatophore of thenomen (in the present case the nomen of the type-species ofthe genus-series nomen).

Besides proposing deWnition-based nomina for these 16taxa, Hillis and Wilcox (2005: 304) stated that these 16nomina should also be considered subgeneric nomina

under the rules of the Code, “even though the clades arenested hierarchically within one another”. In order for thisto be possible, these authors also designated type-speciesfor their seven new nomina, a designation that is neitherrequested nor recognized within the rules proposed for thePhylocode. Despite their doing so, however, they ignoredseveral basic rules of the Code, which results in all their newnomina being unavailable or/and invalid under this nomen-clatural system: they ignored the criteria of availability ofnomina, the Rule of Homonymy, the rules concerning ono-matophores and synonymy, and the hierarchy of ranks rec-ognized by the Code in the genus-series. These points will beconsidered separately, following the three storeys ofnomenclature as described above.

3.2. Availability of nomina: the need of characters

In traditional zoological taxonomy and nomenclature,the diagnosis of a taxon is a list or combination of charac-ters that are considered “characteristic” or “diagnostic” ofthe taxon, i.e., shared by all its members but not by mem-bers of other taxa. Some recent authors tend to use the termdiagnosis in the sense of list of apomorphies supposed to beunique to a taxon, but this is an unusual and relativelyrecent sense for the term, and for this concept the termapognosis is appropriate (Dubois, 1997). Diagnoses (or“deWnitions”, “descriptions”, “indications”, but not neces-sarily apognoses) are required by the Code for a new nomento be available and potentially valid in zoological nomencla-ture. A strong misunderstanding has developed in therecent literature regarding the rôle of diagnoses in nomen-clature. Some authors think that under the Code diagnosesare needed to deWne nomina. This is simply not true. TheCode does not give rules for “deWning” nomina, but forproviding an unambiguous way of allocating any nomen toone or several taxa in a given taxonomy. Under the Code,nomina are by no way deWned by diagnoses or characters,but pointed to by onomatophores (storey Allocation).Characters are indeed necessary, but this is for the avail-ability of nomina (storey Availability), not for the alloca-tion of nomina to taxa.

So, what is a character? The Glossary of the Code pro-vides the following deWnition: “Any attribute of organismsused for recognizing, diVerentiating, or classifying taxa”(Anonymous, 1999, 101). Therefore, the need of charactersfor the availability of a new nomen requires mention ofparticularities of the organisms themselves; otherwise, thisnomen is a nomen nudum, i.e., it does not have any existenceunder the Code, and cannot therefore be the valid nomen ofa taxon recognized in any taxonomy. To be available underthe Code, a nomen published after 1930 must be publishedassociated with a diagnosis, apognosis or description basedon characters, i.e., “be accompanied by a description ordeWnition that states in words characters that are purportedto diVerentiate the taxon”, or “be accompanied by a biblio-graphic reference to such a published statement”, or “beproposed expressly as a new replacement name (nomen


novum) for an available name” (Article 13.1). Important inthe quotations above are the terms “states in words”, “char-acters” and “published statement”. As concerns the latterpoint, Chapter 3 of the Code precisely deWnes a publicationas a work “produced in an edition containing simulta-neously obtainable copies by a method that assures numer-ous identical and durable copies” (Article 8.1.3), andexpressly states that “text or illustrations distributed bymeans of electronic signals (e.g., by means of the WorldWide Web)” does not constitute “published work withinthe meaning of the Code” (Article 9). These rules of theCode are clear enough to state that a new nomen is notavailable if based only on information appearing in anotherunpublished (in the sense of the Code) data base: thisapplies in particular to nucleotide sequences available elec-tronically on GenBank. On the other hand, there is norequirement or restriction regarding the kind of “diagnos-tic” characters, i.e., of characters “purported to diVerentiatethe taxon” to which this nomen applies. Any character ofthe organisms will do: it may be morphological, molecular,behavioral, chromosomal, physiological, etc. However, itcannot be a geographical or ecological distribution, as theseparticularities are “attached”, sometimes temporarily, tosome organisms, but are not part of them. It cannot eitherbe the place of the taxon in a cladogram, as cladograms arenot characters of the organisms, but hypotheses on the cla-distic relationships between organisms. Of course, thesehypotheses are usually based on character analysis, but thetwo are not equivalent, as with the same matrix of charac-ters, diVerent methods of data analysis may provide diVer-ent cladograms and therefore diVerent taxonomies.

Therefore, the seven new “subgeneric” ranid nomina cre-ated by Hillis and Wilcox (2005) are nomina nuda under therules of the Code. These nomina are nomenclaturallyunavailable as they were introduced with “phylogeneticdeWnitions” that refer to cladistic relationships of theirincluded species, but without any statement in words of“diagnostic” characters, or even reference to such a state-ment previously published. It could be argued that informa-tion is provided in the paper regarding some of thecharacters of the supraspeciWc taxa deWned there, in partic-ular regarding the advertisement calls of some of the speciesincluded in these taxa, but this information is provided cur-sorily in the course of the main body of the paper, not in theAppendix B, which gives the classiWcation of New WorldRana and where deWnitions are provided for the new nom-ina, so that these characters are clearly not considered rele-vant by the authors for the valid creation of these nomina.Furthermore, the characteristics of the advertisement callare given for some taxa only, not for all of them, so they donot permit diVerentiation of all these taxa. Even when pro-vided, these characters are not given as strictly “diagnos-tic”: for example, it is stated that the advertisement calls of“most of the species” of the “clade” called “Scurrilirana”“sound like chuckling laughter”, but this is not a “diagnos-tic” character as it is not common to all species of thetaxon. Even if quibbling could allow treating some of these

nomina, which are associated “by chance” with informa-tion on morphology or advertisement calls, as available, itis certainly better to consider all these new nomina asunavailable under the Code, as recently acknowledged byFrost et al. (2006: 255), who wrote about these nomina:“inasmuch as these were not associated with organismalcharacteristics that purport to delimit them, they are nom-ina nuda”. Incidentally, in the similar previous paper by Hil-lis et al. (2001) mentioned above, this problem had beenavoided as diagnoses (based on cytochrome b sequences)had been provided for the four new “subgeneric” nominaof plethodontid urodeles created. But the seven new ranidnomina introduced in Hillis and Wilcox (2005) are cur-rently nomenclaturally unavailable, and their correct useunder the Code would Wrst require their valid publicationunder the rules of the Wrst storey (Availability) of thenomenclatural building described above. However, at leastunder the cladistic hypothesis and the taxonomy of Hillisand Wilcox (2005), this will not prove necessary, for rea-sons explained below.

Whereas a GenBank Accession Number is not a charac-ter, a partial nucleotide sequence is a character, and cantherefore allow availability of a new nomen, either whenprovided in combination with other characters, e.g., mor-phological (e.g., Veith et al., 2001), or even being the onlydiagnostic character provided (e.g., Hillis et al., 2001). Theseseemingly trivial distinctions are inescapable for any taxon-omist who wishes to follow the rules of the Code, and duecare should be paid to these rules by any publisher, editoror referee who has to deal with a manuscript submitted to ascientiWc periodical or book. In ranoid frogs, two othercases of nomina nuda were recently published for taxa thatwere only characterized by the topology of cladograms: onespecies-series nomen (Emerson and Ward, 1998; Dubois,1999), and one family-series nomen (Bossuyt and Milinkov-itch, 2001; Dubois et al., 2001). In the two latter cases theauthors of the nomina nuda did not claim being supportersof the Phylocode, but they simply ignored some of the basicrules of the Code for the creation of new nomina. No doubtmany other similar examples could be found in other zoo-logical groups through a careful screening of recent litera-ture. All zoologists interested in taxonomy andnomenclature should be aware of these problems and avoidtheir repetition in the future.

3.3. Allocation of nomina to taxa: the rôle of onomatophores

The papers of Hillis et al. (2001) and Hillis and Wilcox(2005) are strange in a special way, as they have an incon-sistant approach to the Code regarding the allocation ofnomina to taxa. On one hand these papers seem to followthe Code in recognizing the need to designate “type-spe-cies” for all new genus-series nomina created. As such type-species are not requested by the Phylocode, this informationis apparently provided only to comply with the needs of theCode. On the other hand, these type-species play no rôle inthe allocation of nomina to taxa in this work. This


allocation is made only on the basis of the “phylogeneticdeWnitions” of the taxa. Thus, diVerent taxa with diVerent,more or less inclusive contents, are given diVerent nominaalthough the latter have the same type-species. Althoughthis is of course possible at the time of creation of the nom-ina (an author is free to create several synonymous nominawith the same onomatophore), this is of course not possiblein the end, within the frame of a given taxonomy: as will beshown below, when it comes to the storey Validity, thesenomina will have to be treated as objective synonyms andonly one of them, the oldest or that chosen by the Wrst-reviser in case of simultaneous nomina, will remain as thevalid one. This would not have been the case with diVerentchoices of the type-species for these diVerent taxa: in orderfor the nomina to be allocated to diVerent taxa, each nomenshould have been attached, at each hierarchical level, to atype-species included in a subtaxon diVerent from thatincluding the type-species of the epinymous nomen of thejust superordinate rank. This would not have changed theother problems pointed above (unavailability of these nom-ina) and below (use of ranks not allowed by the Code), butat least it would have been consistent with the rules of theCode regarding allocation of nomina to taxa.

3.4. Validity of nomina: homonymy, synonymy and ranks

3.4.1. Junior homonymsOne of the nomenclatural problems in the paper by Hil-

lis and Wilcox (2005) is rather trivial. It simply results fromignorance of the Principle of Homonymy, according towhich, under the Code, any genus-series nomen that is ajunior homonym of an earlier genus-series nomen is deWni-tively invalid. This applies here to the generic nomen Ran-ula Peters, 1860 (Amphibia), which, being a junior invalidhomonym of Ranula Schumacher, 1817 (Mollusca; seeDubois, 1981: 249), cannot be resurrected to designate anew or newly deWned subgenus. Furthermore, according tothe rules of the draft Phylocode, to be converted into anomen following the latter, a nomen must be “potentiallyvalid” according to the Code, which means that it cannot be“objectively invalid”, as are generic nomina that are juniorhomonyms. Therefore, the onomatophore-based nomenRanula Peters, 1860 cannot be converted into a deWnition-based nomen �RANULA [Peters, 1960] Hillis and Wilcox,2005�.

3.4.2. Junior synonymsOnce available under the Code, a nomen must be allo-

cated to a taxon, through use of its onomatophore. Thenext problem in Hillis and Wilcox (2005) comes from theconfusion noted above between “extensive deWnition ofnomen through taxonomic content” and “ostensive point-ing to nomen by onomatophore”. In deWnition-basednomenclature, a nomen is not attached to an onomato-phore, but deWned by speciWers. In Hillis and Wilcox (2005),each nomen is deWned as applying to the taxon stemmingfrom the most recent common ancestor of several species

listed as speciWers (node-based phylogenetic deWnition).Thus, two nomina may refer to taxa that include somecommon speciWers but that are distinct, one being moreinclusive than the other. This is not possible in onomato-phore-based nomenclature, as genus-series nomina are notdeWned by several included species but attached to singlespecies nomina, their onomatophores. Two generic nominathat are based on the same onomatophore are objectivesynonyms even if they were proposed for more or less inclu-sive taxa including this onomatophore. This situation isencountered in three cases in the nomenclatural proposalsof Hillis and Wilcox (2005): their nomen “Laurasiarana”, ifavailable, would be an invalid junior objective synonym ofAurorana Dubois, 1992, that has the same type-speciesRana aurora Baird and Girard, 1852; similarly, “Novirana”Hillis and Wilcox, 2005 would be an invalid junior objectivesynonym of Pantherana Dubois, 1992 (type-species of both,Rana pipiens Schreber, 1782), and “Torrentirana” Hillis andWilcox, 2005 would be an invalid junior objective synonymof Zweifelia Dubois, 1992 (type-species of both, Rana tara-humarae Boulenger, 1917).

The onomatophore-based nomenclature proposed byHillis and Wilcox (2005) also includes one case of subjectivesynonymy, between the nomina “Lacusirana” and “Stertirana”.If available, these nomina, although based on diVerent type-species (see Table 1), would be subjective synonyms, asboth these type-species are members of the exactly samelower taxon (hypogroup C1a in Table 2); thus, one of themwould be redundant to designate this taxon if the latter waslater recognized as a genus or a subgenus.

3.4.3. The rank hierarchy recognized by the CodeUnlike the Phylocode, the Code recognizes nominal-

series, and, in each of them, several distinct ranks areallowed. The situation is however diVerent in the threenominal-series. In the family-series, the number of poten-tial ranks is virtually unlimited, as this nominal-series“encompasses all nominal taxa at the ranks of superfam-ily, family, subfamily, tribe, subtribe, and any other rankbelow superfamily and above genus that may be desired”(Article 35.1). In contrast, in the other two nominal-series,the number of ranks recognized by the Code is strictly lim-ited, as the genus-series “encompasses all nominal taxa atthe ranks of genus and subgenus” (Article 42.1), and thespecies-series “encompasses all nominal taxa at the ranksof species and subspecies” (Article 45.1). Despite this for-mal deWnition given in Article 45.1, two additional ranksare allowed in the Code in the latter nominal-series, thoseof “aggregates of species” and of “aggregates of subspe-cies”, whose nomina may be interpolated in parenthesesbetween the genus and species nomina or between the spe-cies and subspecies nomina (Article 6.2) and are submit-ted to the Principle of Priority (Article 23.3.3).Nomenclatural ranks below subgenus and below subspe-cies are not recognized by the Code, which means thatnomina used for taxa at such lower ranks in these twonominal-series are not nomenclaturally valid.


To support their nomenclatural proposals, Hillis et al. that, in a given taxonomy, several subgenera can be “nested

Table 1Status of the onomatophore-based nomina of the genus-series used in Table 2 and of their synonyms (for more details, see Dubois, 1992, 1999)

Modes of designation of onomatophores (type-species): od, original designation; mt, monotypy; sd, subsequent designation; rn, new replacement nomen ornomen novum (same type-species as that of replaced nomen). Status of nomina: vn, valid nomen; os, invalid junior objective synonym; ss, invalid juniorsubjective synonym; jh, invalid junior homonym; un, unavailable nomen (nomen nudum). Following Dubois (2000), the latter nomina are presentedbetween quotation marks.

Genus-series nomen Type-species of genus-series nomen [mode of designation] (status of species-series nomen)

Status of genus-series nomen according to taxonomy of Table 2

Amerana Dubois, 1992 Rana boylii Baird, 1854 [od] (vn) vn of subgenusAquarana Dubois, 1992 Rana catesbeiana Shaw, 1802 [od] (vn) vn of subgenusAurorana Dubois, 1992 Rana aurora Baird & Girard, 1852 [od] (vn) ss of Amerana Dubois, 1992Chilixalus Werner, 1899 Chilixalus warszewiczii Werner, 1899 [mt] (os of Ixalus

warszewitschii Schmidt, 1857)ss of Lithobates Fitzinger, 1843

“Lacusirana” (Hillis & Wilcox, 2005) Rana megapoda Taylor, 1942 [od] (vn) un; ss of Lithobates Fitzinger, 1843Laevirana Günther, 1900 Levirana vibicaria Cope, 1894 [rn for Levirana Cope,

1894] (vn)os of Levirana Cope, 1894; ss of Lithobates Fitzinger, 1843

“Laurasiarana” (Hillis & Wilcox, 2005) Rana aurora Baird & Girard, 1852 [od] (vn) un; os of Aurorana Dubois, 1992; ss of Amerana Dubois, 1992

Levirana Cope, 1894 Levirana vibicaria Cope, 1894 [mt] (vn) ss of Lithobates Fitzinger, 1843Lithobates Fitzinger, 1843 Rana palmipes Spix, 1824 [od] (vn) vn of genus and subgenus“Nenirana” (Hillis & Wilcox, 2005) Rana areolata Baird & Girard, 1852 [od] (vn) un; ss of Lithobates Fitzinger, 1843“Novirana” (Hillis & Wilcox, 2005) Rana pipiens Schreber, 1782 [od] (vn) un; os of Pantherana Dubois, 1992; ss of

Lithobates Fitzinger, 1843Pantherana Dubois, 1992 Rana pipiens Schreber, 1782 [od] (vn) ss of Lithobates Fitzinger, 1843Pohlia Steindachner, 1867 Rana palmipes Spix, 1824 [mt] (vn) os of Lithobates Fitzinger, 1843Rana Linnaeus, 1758 Rana temporaria Linnaeus, 1758 [sd by Fleming, 1822]

(vn)vn of genus and subgenus

Ranaria RaWnesque-Schmaltz, 1814 Rana temporaria Linnaeus, 1758 [rn for Rana Linnaeus, 1758] (vn)

os of Rana Linnaeus, 1758

Ranula Peters, 1860 [nec Schumacher, 1817]

Ranula gollmerii Peters, 1859 [mt] (ss of Rana palmipes Spix, 1824)

jh; ss of Lithobates Fitzinger, 1843

“Scurrilirana” (Hillis & Wilcox, 2005) Rana berlandieri Baird, 1854 [od] (vn) un; ss of Lithobates Fitzinger, 1843Sierrana Dubois,1992 Rana sierramadrensis Taylor, 1939 [od] (vn) ss of Lithobates Fitzinger, 1843“Stertirana” (Hillis & Wilcox, 2005) Rana montezumae Baird, 1854 [od] (vn) un; ss of Lithobates Fitzinger, 1843“Torrentirana” (Hillis & Wilcox, 2005) Rana tarahumarae Boulenger, 1917 [od] (vn) un; os of Zweifelia Dubois, 1992; ss of

Lithobates Fitzinger, 1843Trypheropsis Cope, 1868 Ranula chrysoprasina Cope, 1866 [od] (ss of Ixalus

warszewitschii Schmidt, 1857)ss of Lithobates Fitzinger, 1843

Zweifelia Dubois, 1992 Rana tarahumarae Boulenger, 1917 [od] (vn) ss of Lithobates Fitzinger, 1843

(2001) and Hillis and Wilcox (2005) relied on Article 10.4 ofthe Code, which states that any “genus-group division of agenus, even if proposed for a secondary (or further) subdi-vision, is deemed to be a subgeneric name even if the divi-sion is denoted by a term such as ‘section’ or ‘division”’.Hillis et al. (2001) and Hillis and Wilcox (2005: 304) mis-construed this Article as meaning that all the nomina bywhich they designated taxa in their new taxonomy “aresubgenera under ICZN rules, even though the clades arenested hierarchically within one another”. This sentenceincludes two misunderstandings. First, it makes confusionbetween nomina and taxa. Although under the Phylocodephilosophy nomina and taxa are considered equivalent, thisis not true in nomenclature regulated by the Code: nominadesignate taxa but are not taxa. Second, this sentence showsconfusion between Wrst and third Xoors of the nomencla-tural building, Availability and Validity. What Article 10.4states, is that any uninominal nomen introduced for anydivision of a genus, whatever its rank below genus, isnomenclaturally available and is deemed to have been cre-ated at subgeneric rank. But of course it does not mean

hierarchically within one another”! This would be com-pletely contrary to the system of hierarchical ranks usedthroughout the Code, which requires that nomina at a givenrank always designate taxa that are subordinate to taxa ofthe just higher rank, and superordinate to taxa of the justlower rank: therefore, a subgenus cannot include subgen-era! Hillis et al. (2001) and Hillis and Wilcox (2005) did notrealize that Article 10.4 is part of Chapter 4 of the Code,entitled “Criteria of availability”, but not of Chapter 9,entitled “Genus-group nominal taxa and their names”:according to the latter Chapter, only the ranks genus andsubgenus are acceptable under the Code, and additionalranks are simply excluded from zoological nomenclature.This is similar to the invalid use by some authors of rankslike “form” or “variety” below subspecies, although Arti-cles 10.2 and 45.6 of the Code also state that an infrasubspe-ciWc nomen published before 1961 for a variety or a formmay be available for taxa at rank subspecies, but not usednowadays for infrasubspeciWc taxa.

As the Code only recognizes one subordinate rank, subge-nus, below genus in the genus-series, a given genus can only


Table 2Infrageneric taxonomy of American frogs referred to the genera Lithobates and Rana (plus the Europaean type-species of the latter, Rana temporaria),according to the cladistic hypothesis of Hillis and Wilcox (2005) and the generic taxonomy of Frost et al. (2006), and nomenclatures corresponding to thistaxonomy under the Code (onomatophore-based nomina) and the Phylocode (deWnition-based nomina)

1. Genus Lithobates Fitzinger, 1843 �NOVIRANA Hillis & Wilcox, 1985� 1.1. Subgenus Aquarana Dubois, 1992 �Unnamed taxon in Hillis & Wilcox, 1985�

1.1.1. Group clamitans Latreille, 1801 �AQUARANA [Dubois, 1992] Hillis & Wilcox, 1985�Lithobates (Aquarana) catesbeianus (Shaw, 1802)Lithobates (Aquarana) clamitans (Latreille, 1801)Lithobates (Aquarana) grylio (Stejneger, 1901)Lithobates (Aquarana) heckscheri (Wright, 1924)Lithobates (Aquarana) okaloosae (Moler, 1985)Lithobates (Aquarana) septentrionalis (Baird, 1854)Lithobates (Aquarana) virgatipes (Cope, 1891)

1.1.2. Group sylvaticus LeConte, 1825 �Unnamed taxon in Hillis & Wilcox, 1985�Lithobates (Aquarana) sylvaticus (LeConte, 1825)

1.2. Subgenus Lithobates Fitzinger, 1843 �SIERRANA [Dubois, 1992] Hillis & Wilcox, 1985� 1.2.1. Group palmipes Spix, 1824 �“RANULA [Peters, 1859] Hillis & Wilcox, 1985”�

1.2.1.1. Unnamed subgroup A �LEVIRANA [Cope, 1894] Hillis & Wilcox, 1985� 1.2.1.1.1. Unnamed infragroup A1 �TRYPHEROPSIS [Cope, 1868] Hillis & Wilcox, 1985�

Lithobates (Lithobates) vibicarius (Cope, 1894)Lithobates (Lithobates) warszewitschii (Schmidt, 1857)

1.2.1.1.2. Unnamed infragroup A2 �Unnamed taxon in Hillis & Wilcox, 1985�Lithobates (Lithobates) melanosoma (Günther, 1900)a

1.2.1.2. Unnamed subgroup B �LITHOBATES [Fitzinger, 1843] Hillis & Wilcox, 1985�Lithobates (Lithobates) bwana (Hillis and De Sá, 1988)Lithobates (Lithobates) juliani (Hillis and De Sá, 1988)Lithobates (Lithobates) palmipes (Spix, 1824)Lithobates (Lithobates) vaillanti (Brocchi, 1877)

1.2.2. Group pipiens Schreber, 1782 �Unnamed taxon in Hillis & Wilcox, 1985� 1.2.2.1. Unnamed subgroup C �PANTHERANA [Dubois, 1992] Hillis & Wilcox, 1985�

1.2.2.1.1. Unnamed infragroup C1 �STERTIRANA Hillis & Wilcox, 1985� 1.2.2.1.1.1. Unnamed hypogroup C1a �LACUSIRANA Hillis & Wilcox, 1985�Lithobates (Lithobates) chiricahuensis (Platz & Mecham, 1979)Lithobates (Lithobates) dunni (Zweifel, 1957)Lithobates (Lithobates) Wsheri (Stejneger, 1893)Lithobates (Lithobates) lemosespinali (Smith & Chiszar, 2003)Lithobates (Lithobates) megapoda (Taylor, 1942)Lithobates (Lithobates) montezumae (Baird, 1854)Lithobates (Lithobates) subaquavocalis (Platz, 1993) 1.2.2.1.1.2. Unnamed hypogroup C1b �Unnamed taxon in Hillis & Wilcox, 1985�Lithobates (Lithobates) pipiens (Schreber, 1782)

1.2.2.1.2. Unnamed infragroup C2 �Unnamed taxon in Hillis & Wilcox, 1985� 1.2.2.1.2.1. Unnamed hypogroup C2a �NENIRANA Hillis & Wilcox, 1985�Lithobates (Lithobates) areolatus (Baird & Girard, 1852)Lithobates (Lithobates) capito (LeConte, 1855)Lithobates (Lithobates) palustris (LeConte, 1825)Lithobates (Lithobates) sevosus (Goin & Netting, 1940) 1.2.2.1.2.2. Unnamed hypogroup C2b �SCURRILIRANA Hillis & Wilcox, 1985�Lithobates (Lithobates) berlandieri (Baird, 1854)Lithobates (Lithobates) blairi (Mecham, Littlejohn, Oldham, Brown & Brown, 1973)Lithobates (Lithobates) brownorum (Sanders, 1973)Lithobates (Lithobates) chichicuahutla (Cuellar, Méndez-DeLaCruz & Villágran-Santa Cruz, 1996)Lithobates (Lithobates) forreri (Boulenger, 1883)Lithobates (Lithobates) macroglossa (Brocchi, 1877)Lithobates (Lithobates) magnaocularis (Frost & Bagnara, 1976)Lithobates (Lithobates) miadis (Barbour & Loveridge, 1929)Lithobates (Lithobates) neovolcanicus (Hillis & Frost, 1985)Lithobates (Lithobates) omiltemanus (Günther, 1900)Lithobates (Lithobates) onca (Cope, 1875)Lithobates (Lithobates) spectabilis (Hillis & Frost, 1985)Lithobates (Lithobates) sphenocephalus (Cope, 1889)

Lithobates (Lithobates) sphenocephalus sphenocephalus (Cope, 1889)Lithobates (Lithobates) sphenocephalus utricularius (Harlan, 1825)

Lithobates (Lithobates) taylori (Smith, 1959)Lithobates (Lithobates) tlaloci (Hillis & Frost, 1985)Lithobates (Lithobates) yavapaiensis (Platz & Frost, 1984)

(continued on next page)


include subgenera, which are all at the same nomenclaturalrank, but no further genus-series divisions. It would of coursebe possible to recognize all the supraspeciWc taxa listed inHillis and Wilcox (2005) as subgenera of Rana (as in theworking taxonomy of Dubois, 1992) but then these taxashould be considered as having the same rank in the nomen-clatural hierarchy. But, as Hillis and Wilcox insist that theirtaxa are “nested hierarchically within another” and are allincluded in the same genus Rana, they cannot avoid thenomenclatural consequences of their arrangement, which isthat only two subgeneric nomina (Rana and Lithobates) arenomenclaturally valid, all the nomina of lower subordinatetaxa being invalid under the rules of the Code; among thesenomina some will have to be treated as junior synonyms ofRana, and others as junior synonyms of Lithobates. Underthe taxonomy more recently proposed by Frost et al. (2006),Rana and Lithobates are treated as genera, but then again,according to the Code, only one genus-series rank can be rec-ognized below the rank genus, not several as suggested by thecladogram in Hillis and Wilcox (2005).

3.5. Spelling of nomina

Although the Code does not expressly require correct ety-mological formation for availability of nomina, the latter iscertainly preferable to nomina based on incorrect roots. Asdiscussed below, under the current taxonomic scheme there is

no need for the new genus-series nomina created by Hillisand Wilcox (2005), but in case such nomina proved later nec-essary as a result of changes in the taxonomy, in some cases itmight be better to coin nomina slightly diVerent from thenomina nuda proposed in this work. This would have a sec-ond advantage, that of having slightly diVerent nomina forthe same taxa in onomatophore-based and deWnition-basednomenclature, thus completely avoiding the possibility ofconfusion between nomenclatural systems (see below). If thisis done, I would suggest the two following changes inorder to have better formed or at least shorter nomina:“Lacusirana”, based on the Latin lacus which have given theterms lacuna and lacunosus, could be replaced by the betterformed “Lacunirana”; and “Scurrilirana”, based on the Latinscurrilis derived from scurra, could be changed into theshorter “Scurrana”.

3.6. A valid nomenclature for American Rana under the Code

Hillis and Wilcox (2005) distributed all American speciesreferred to the genus Rana and the European species Ranatemporaria into 24 supraspeciWc taxa. For reasonsexplained above, they provided deWnition-based nominafor only 16 of them. If these taxa are to be named under theCode, they must all be given onomatophore-based nomina,including those that are currently considered to include asingle species. However, according to the Code, only two

Table 2 (continued)

Taxa are presented by alphabetical order of their onomatophore-based nomina, or of their deWnition-based nomina in the absence of onomatophore-based nomina, ending with unnamed taxa. Nomina in lower-case Italic letters are nomina following the Code, whereas nomina in small capitals andbetween “�ƒ�” signs are the nomina following the Phylocode that designate the same taxa.

a This species is called Rana maculata Brocchi, 1877 in Hillis and Wilcox (2005) and Lithobates maculatus (Brocchi, 1877) in Frost et al. (2006: 369) but thisnomen is invalid. Smith et al. (1966) requested suppression of the nomen Rana maculata Daudin, 1801, a senior synonym of Eleutherodactylus richmondi Stej-neger, 1904, by the International Commission on Zoological Nomenclature, in order to validate both the latter name and Rana maculata Brocchi, 1877. How-ever, as pointed out by Hillis and de Sá (1988: 10), the Commission never voted on this question. The nomina at stake cannot in the least be considered nominaof wide usage outside the specialised Weld of zoological nomenclature (see Dubois, 2005a,b), so the rules of the Code (homonymy and priority) should be strictlyfollowed: accordingly, the name Lithobates melanosoma is the valid name of the species, as already suggested by Hillis and de Sá (1988: 10).

1.2.2.2. Unnamed subgroup D �TORRENTIRANA Hillis & Wilcox, 1985� 1.2.2.2.1. Unnamed infragroup D1 �ZWEIFELIA [Dubois, 1992] Hillis & Wilcox, 1985�

Lithobates (Lithobates) johni (Blair, 1965)Lithobates (Lithobates) psilonota (Webb, 2001)Lithobates (Lithobates) pueblae (Zweifel, 1955)Lithobates (Lithobates) pustulosus (Boulenger, 1883)Lithobates (Lithobates) tarahumarae (Boulenger, 1917)Lithobates (Lithobates) zweifeli (Hillis, Frost & Webb, 1984)

1.2.2.2.2. Unnamed infragroup D2 �Unnamed taxon in Hillis & Wilcox, 1985�Lithobates (Lithobates) sierramadrensis (Taylor, 1939)

2. Genus Rana Linnaeus, 1758 �LAURASIARANA Hillis & Wilcox, 1985� 2.1. Subgenus Amerana Dubois, 1992 �AMERANA [Dubois, 1992] Hillis & Wilcox, 1985�

Rana (Amerana) aurora Baird & Girard, 1852 Rana (Amerana) boylii Baird, 1854 Rana (Amerana) cascadae Slater, 1939 Rana (Amerana) draytonii Baird & Girard, 1852 Rana (Amerana) luteiventris Thompson, 1913 Rana (Amerana) muscosa Camp, 1917 Rana (Amerana) pretiosa Baird & Girard, 1853

2.2. Subgenus Rana Linnaeus, 1758 �Unnamed taxon in Hillis & Wilcox, 1985� Rana (Rana) temporaria Linnaeus, 1758


ranks can be used between the rank genus and the rank spe-cies: the rank subgenus, which is part of the nominal genus-series, and the rank “aggregate of species”, which is part ofthe nominal species-series. Additional ranks are currentlynot allowed by the Code, and formal nomina cannot there-fore be given to taxa below “aggregate of species” andabove species.

Before applying nomina to genera and subgenera, thenomenclatural status of all genus-series nomina concerningthis group of frogs must be established (Table 1). Thisallows knowing which nomina are nomenclaturally avail-able and unavailable and which ones are deWnitively invalidfor being junior homonyms or junior objective synonyms.The remaining nomina must be confronted with the pro-posed taxonomy (Table 2). Hierarchical relationships inthis taxonomy are here presented, as suggested in Hillis andWilcox (2005: 304), by successive indenting of subordinatetaxa. Following Frost et al. (2006), the rank genus is givenhere in Table 2 to the taxa Rana and Lithobates. The result-ing taxonomic scheme uses Wve successive genus-series lev-els below genus. As this is not possible under the Code,which only accepts two possible ranks between genus andspecies, the three lowest levels are left unnamed in Table 2.

Because of the Rule of Coordination, any genus taxonwith two or more subgenera must include one hyponymoussubgenus that bears exactly the same nomen: thus the genusRana needs to have a subgenus Rana, and the genus Litho-bates a subgenus Lithobates. The nomina of the sister-sub-genera are automatically given by the Rule of Priority (andthe Rule of First-Reviser if necessary). “Nomenclaturalparsimony” may be deWned as the use of the lowest possiblenumber of nomina to provide unambiguous naming for agiven set of taxa. As well shown in this example, the nomen-clatural system of the Code allows a more parsimoniousnomenclatural system as compared with the Phylocode,without any risk of ambiguity or confusion. Thanks to theRule of Coordination, the same nomen can be used fordiVerent, hierarchically related, taxa, but association of thenomen with the designation of its rank avoids confusionbetween these taxa: the genus Rana and the subgenus Ranaare diVerent taxa although they bear the same nomen.

As concerns “aggregates of species”, following a usualpractice in Table 2 they are designated as “species-groups”,or more shortly “groups”. The Code (Article 23.3.3) statesthat the Rule of Priority applies to such nomina, so eachgroup is named after the nomen of its Wrst named species.Here also, the nomen of the taxon is given automatically bypriority.

As the Code currently does not allow naming taxabetween the rank “aggregate of species” and the rank spe-cies, in Table 2 these taxa are given the informal ranks“subgroup”, “infragroup” and “hypogroup” (following thehierarchy of preWxes discussed in Dubois, 2006a), and theyare designated by letters or combinations of letters andnumbers.

According to the Code (Article 6), both the nomina ofsubgenera and those of “aggregates of species” can be inter-

polated in parentheses between the generic and the speciWcnomina. As these interpolated nomina belong in two dis-tinct nominal-series, they should be enclosed in distinctparentheses. The Code also suggests (Recommendation 6B)that the “taxonomic meaning” of the nomen of an “aggre-gate of species” be mentioned by use of a special term: inthis case the term “group” can be used. As a result, thecomplete nomen of a species under the onomatophore-based nomenclature of Table 2 writes as follows: Lithobates(Aquarana) (group clamitans) catesbeianus Shaw, 1802. Ofcourse, such a notation is very cumbersome. It is not to beused often, but it may be quoted only once in a publication,on the occasion of the Wrst mention of a species, to providecondensed information on its taxonomic relationships.Such a detailed writing may also be useful in taxonomic orfaunistic checklists or catalogues.

The onomatophore-based nomenclature of Table 2requires using only nomina previously published. Even ifnomenclaturally available, the seven new nomina createdby Hillis and Wilcox (2005) would be useless as they applyto taxa at ranks below subgenus within the genus-series,therefore unacceptable under the Code. Furthermore, what-ever the taxonomic frame adopted (e.g., if species-groups oreven lower ranked taxa were raised to the rank genus),three of these seven nomina (“Laurasiarana”, “Novirana”and “Torrentirana”) would remain forever redundant anduseless in nomenclature under the Code, as nomina with thesame onomatophores are already available for the taxa forwhich they were coined (Table 1). The same would apply tothe nomina “Lacusirana” and “Stertirana”, which would besubjective synonyms as long as their type-species are mem-bers of the same lower taxon.

The Code currently allows naming only a few of the taxarecognized in the taxonomy of Table 2, whereas the Phylo-code allows naming them all, except those that are currentlyconsidered to include a single species. Actually, this caseraises a genuine interesting question that deserves furtherexploration: why should the current Code forbid the use ofa nomenclatural hierarchy of more than two ranks (genusand subgenus) within the genus-series? This limitation inthe number of ranks allowed by the Code is based only on“tradition”. It has no theoretical justiWcation and can nodoubt be considered contradictory to the Preamble of theCode, which states that none of the provisions of the Coderestricts the “freedom of taxonomic thought or actions”. Inthis respect, the Phylocode is no doubt superior to the Code.For this reason, the idea should be considered that the Codeshould evolve and be expanded to permit an unlimitednumber of ranks. This would allow taking into account thenew developments of taxonomic thinking and analysis, inparticular the important increase in the number of hierar-chically related ranks that result from the multiplication ofcladistic analyses in many zoological groups (Dubois,2006a). Precise proposals to modify the Code in this respectwere submitted to the International Commission on Zoo-logical Nomenclature (Dubois, 2006b). Detailed discussionof this question is beyond the scope of the present paper.


4. Conclusion

The philosophies and the rules of the Code and of thePhylocode for the establishment of new nomina, their allo-cation to taxa and their validation, are widely diVerent andlargely incompatible. As shown in detail above, because ofthese diVerent rules, valid nomina for taxa may be com-pletely diVerent under the two nomenclatural systems, evenwhen exactly the same taxonomy is adopted (e.g., based onthe same cladogram), and when the “same” nomina (butdeWned diVerently) are adopted in the two systems. Thisraises a special problem. If nomina that appear “the same”in the two systems are in fact deWned diVerently and followdiVerent rules of allocation to taxa, they are in fact not “thesame”, and confusion is likely to occur for users of publica-tions mentioning these nomina, especially when the usersare not taxonomists or phylogeneticists. As the two systemsare “non-miscible”, whenever using a nomen, an authorshould make clear under which of these two systems thisnomen is recognized as valid. Confusion between the twosystems would be very detrimental for communicationamong taxonomists, and even more so between the latterand all users of zoological nomina, not only biologists butalso laymen of various other horizons (medicine, agricul-ture, commerce, administration, law, etc.). Sluys et al. (2004)expressed a widely shared opinion when they wrote: “wedeplore a situation in which two, very diVerent, nomencla-tural codes will compete to govern the rules for namingorganisms”. However, this situation is already a reality, andmust be faced as such. The most important point now forclarity of communication is that users of the Phylocode beclearly distinguished from followers of the Code in all pub-lications using scientiWc nomina of animals or plants. With-out such a provision, the new nomenclature looking similarto the traditional one, many biologists and laymen may notrealize that such a drastic change has been brought in theway nomina are deWned and used, and these nomina maybecome progressively “entrenched in zoological literature”,thus later eligible for “protection” on the basis of “usage”.If deWnition-based nomina are not clearly distinguishedfrom onomatophore-based nomina, problems of homon-ymy between nomina following either rule will rapidlybecome a considerable source of confusion and error (Can-tino, 2000; Blackwell, 2002), and this confusion raises otherproblems (Stevens, 2002; Greuter, 2004; Dubois, 2005a).Therefore, use of clearly distinctive ways of printing thenomina of both systems should soon be implemented in allscientiWc publications. The international community of tax-onomists should consider imposing, through the editorialrules and the action of the editorial boards of scientiWc pub-lications, an obligation to use special marks to distinguishnomina used under the Phylocode, so that the latter cannotbe mistaken for scientiWc nomina following the Code.

To avoid any possible ambiguity, the best option wouldcertainly have been for this new nomenclature to use brandnew terms, not borrowed from traditional nomenclature,and possibly written in a language other than Latin, e.g.,

Esperanto (Greuter, 2004). However, as such a move isunlikely to be taken by supporters of this project, the leastthat can be asked from them is to use diVerent “kinds” ofnomina. This could be done, e.g., by giving these nominaspecial characteristics making them easily recognisable, asis the case in some other recently proposed nomenclaturalsystems (Kluge, 2000; Papavero et al., 2001). I discussed thisquestion in detail (Dubois, 2005a), and rejected the sugges-tion (Cantino, 2000; Pleijel and Rouse, 2003; Article 6.1 ofthe draft Phylocode) to add bracketed letters after nominato point to the nomenclatural system under which they arerecognized, namely [R] for “rank-based” nomina and [P]for nomina following the Phylocode. There are three dis-tinct reasons for rejecting this proposal. First, such brack-etted letters, not being part of the nomen itself, are likely tobe omitted in many non-taxonomic publications, thusrestoring an ambiguity. Second, the choice of the letter [R]for nomina following the Code is not a good one, as the useof ranks is not the most important diVerence between theCode and the Phylocode, the major diVerence being in theuse of onomatophores or deWnitions (see Introductionabove). Third, and most importantly, as onomatophore-based nomina have been in use for 250 years, “normal”writing (i.e., without any special diacritic sign) should bemaintained for these traditional nomina, although withdiVerent font types and characters to recognize the nominaof diVerent nominal-series: e.g., Homo sapiens (species-series), Homo (genus-series), Hominidae (family-series) andMammalia (class-series; see Dubois, 2005a,b, 2006a). Inci-dentally, this use of diVerent printing types would avoidany risk of confusion between nomina of diVerent nominal-series having the same spelling—so-called “hemihomon-ymy” (Starobogatov, 1991; Alonso-Zarazaga, 2005)—, thusmaking useless any special device in the spelling of nominathemselves to avoid this. As the Phylocode does not usenominal-series, a single printing convention is needed forall nomina following these rules, hence my proposal towrite them in small capitals between signs evoking a clado-genesis: e.g., �TETRAPODA� or �TETRAPODA GaVney, 1979�.Adoption of this convention or of another, better, one,would certainly clarify communication between taxono-mists, evolutionary biologists and all other users of scien-tiWc nomina of taxa.

References

Anonymous [International Commission on Zoological Nomenclature],1999. International Code of Zoological Nomenclature, fourthed. International Trust for Zoological Nomenclature, London,i–xxix + 1–306.

Alonso-Zarazaga, M.A., 2005. Nomenclature of higher taxa: a newapproach. Bulletin of Zoological Nomenclature 62, 189–199.

Blackwell, W.H., 2002. One-hundred-year code déjà vu? Taxon 51, 151–154.Bossuyt, F., Milinkovitch, M.C., 2001. Amphibians as indicators of early

Tertiary “out of India” dispersal of vertebrates. Science 292, 93–95.Cantino, P.D., 2000. Phylogenetic nomenclature: addressing some con-

cerns. Taxon 49, 85–93.De Queiroz, K., 2006. The PhyloCode and the distinction between taxon-

omy and nomenclature. Systematic Biology 55, 160–162.


De Queiroz, K., Gauthier, J., 1990. Phylogeny as a central principle in tax-onomy: phylogenetic deWnitions of taxon names. Systematic Zoology39, 307–322.

De Queiroz, K., Gauthier, J., 1994. Toward a phylogenetic system of bio-logical nomenclature. Trends in Ecology and Evolution 9, 27–31.

Donoghue, M.J., Gauthier, J.A., 2004. Implementing the PhyloCode.Trends in Ecology and Evolution 19, 281–282.

Dubois, A., 1981. Liste des genres et sous-genres nominaux de Ranoidea(Amphibiens, Anoures) du monde, avec identiWcation de leurs espèces-types: conséquences nomenclaturales. Monitore zoologico italiano(n.s.) 15 (Suppl.), 225–284.

Dubois, A., 1991. Nomenclature of parthenogenetic, gynogenetic and“hybridogenetic” vertebrate taxons: new proposals. Alytes 8, 61–74.

Dubois, A., 1992. Notes sur la classiWcation des Ranidae (Amphibiens,Anoures). Bulletin mensuel de la Société linnéenne de Lyon 61,305–352.

Dubois, A., 1997. An evolutionary biologist’s view on the science of biol-ogy. Alytes 15, 133–136.

Dubois, A., 1999. Miscellanea nomenclatorica batrachologica. 19. Noteson the nomenclature of Ranidae and related groups. Alytes 7, 81–100.

Dubois, A., 2000. Synonymies and related lists in zoology: general propos-als, with examples in herpetology. Dumerilia 4, 33–98.

Dubois, A., 2003. The relationships between taxonomy and conservationbiology in the century of extinctions. Comptes Rendus Biologies 326(suppl. 1), S9–S21.

Dubois, A., 2005a. Proposed Rules for the incorporation of nomina ofhigher-ranked zoological taxa in the International Code of ZoologicalNomenclature. 1. Some general questions, concepts and terms of bio-logical nomenclature. Zoosystema 27, 365–426.

Dubois, A., 2005b. Proposals for the incorporation of nomina of higher-ranked taxa into the Code. Bulletin of Zoological Nomenclature 62,200–209.

Dubois, A., 2006a. Proposed Rules for the incorporation of nomina ofhigher-ranked zoological taxa in the International Code of ZoologicalNomenclature. 2. The proposed Rules and their rationale. Zoosystema28, 165–258.

Dubois, A., 2006b. Should the Code limit the number of nomenclaturalranks in zoology? Manuscript submitted on 23 February 2006 toAndrew Polaszek, Secretary of the International Commission on Zoo-logical Nomenclature, for publication in the Bulletin of ZoologicalNomenclature.

Dubois, A., Ohler, A., 1997. Early scientiWc names of Amphibia Anura. I.Introduction. Bulletin du Muséum national d’Histoire naturelle 4 (18),297–320.

Dubois, A., Ohler, A., Biju, S.D., 2001. A new genus and species of Ranidae(Amphibia, Anura) from south-western India. Alytes 19, 53–79.

Emerson, S.B., Ward, R., 1998. Male secondary sexual characteristics, sex-ual selection, and molecular divergence in fanged ranid frogs of south-east Asia. Zoological Journal of the Linnean Society 122, 537–553.

Fleming, J., 1822. The Philosophy of Zoology, or A General View of theStructure, Functions and ClassiWcation of Animals (vol. 2). Constable,Edinburgh. 1–618.

Frost, D.R., Grant, T., Faivovich, J., Bazin, R.H., Haas, A., Haddad,C.F.B., de Sá, R.O., Channing, A., Wilkinson, M., Donnellan, S.C., Rax-worthy, C.J., Campbell, J.A., Blotto, B.L., Moler, P., Drewes, R.C.,Nussbaum, R.A., Lynch, J.D., Green, D.M., Wheeler, W.C., 2006. Theamphibian tree of life. Bulletin of the American Museum of NaturalHistory 297, 1–370.

Greuter, W., 2004. Recent developments in international biologicalnomenclature. Turkish Journal of Botany 28, 17–26.

Hillis, D.M., Chamberlain, D.A., Wilcox, T.P., Chippindale, P.T., 2001. Anew species of subterranean blind salamander (Plethodontidae: Hemi-dactyliini: Eurycea: Typhlomolge) from Austin, Texas, and a systematicrevision of central Texas paedomorphic salamanders. Herpetologica57, 266–280.

Hillis, D.M., De Sá, R., 1988. Phylogeny and taxonomy of the Ranapalmipes group (Salientia: Ranidae). Herpetological Monographs 2,[i] + 1–26.

Hillis, D.M., Wilcox, T.P., 2005. Phylogeny of the New World true frogs(Rana). Molecular Phylogenetics and Evolution 34, 299–314.

Johnson, L.A.S., 1970. Rainbow’s end: the quest for an optimal taxonomy.Systematic Zoology 19, 203–239.

Joyce, W.G., Parham, J.F., Gauthier, J.A., 2004. Developing a protocol forthe conversion of rank-based taxon names to phylogenetically deWnedclade names, as exempliWed by turtles. Journal of Paleontology 78,989–1013.

Keller, R.A., Boyd, R.N., Wheeler, Q.D., 2003. The illogical basis of phylo-genetic nomenclature. The Botanical Review 69, 93–110.

Kluge, A.G., 2005. Taxonomy in theory and practice, with arguments for anew phylogenetic system of taxonomy. In: Donnelly, M.H., Crother,B.I., Guyer, C., Wake, M.H., White, M.E. (Eds.), Ecology and Evolu-tion in the Tropics: A Herpetological Perspective. University of Chi-cago Press, Chicago, pp. 39–46.

Kluge, N.J., 2000. Sovremennaya sistematika nasekomyh. [Modern sys-tematic of insects]. Part 1. Lan, Saint Petersburg, 1–333. [In Russian;see <http://www.bio.pu.ru/win/entomol/KLUGE/zoo-name.htm>].

Kraus, O., 2004. Phylogeny, classiWcation and nomenclature: a reply to F.Pleijel and G.W. Rouse. Journal of Zoological Systematics and Evolu-tionary Research 42, 159–161.

Laurin, M., Cantino, P.D., 2004. First International Phylogenetic Nomen-clature Meeting: a report. Zoologica Scripta 33, 475–479.

Laurin, M., de Queiroz, K., Cantino, P.D., 2006. Sense and stability oftaxon names. Zoologica Scripta 35, 113–114.

Laurin, M., de Queiroz, K., Cantino, P., Cellinese, N., Olmstead, R., 2005.The PhyloCode, types, ranks and monophyly: a response to Pickett.Cladistics 21, 605–607.

Lidén, M., Oxelman, B., Backlund, A., Andersson, L., Bremer, B., Eriksson,R., Moberg, R., Nordal, I., Persson, K., Thulin, M., Zimmer, B., 1997.Charlie is our darling. Taxon 46, 735–738.

Linnaeus, C., 1758. Systema Naturae per regna tria naturae, secundumclasses, ordines, genera, species, cum characteribus, diVerentiis, synony-mis, locis. Editio decima, reformata. Tomus I. Laurentii Salvii, Hol-miae, [i–iv] + 1–824.

Mayr, E., 1995. Systems of ordering data. Biology and Philosophy 10,419–434.

Moore, G., 2003. Should taxon names be explicitly deWned? The BotanicalReview 69, 2–21.

Papavero, N., Llorente-Bousquets, J., Abe, J.M., 2001. Proposal of a newsystem of nomenclature for phylogenetic systematics. Arquivos deZoologia, São Paulo 36, 1–145.

Pennisi, E., 2001. Linnaeus’s last stand? Science 291, 2304–2305. 2307.Pickett, K.M., 2005a. The new and improved PhyloCode, now with types,

ranks, and even polyphyly: a conference report from the First Interna-tional Phylogenetic Nomenclature Meeting. Cladistics 21, 79–82.

Pickett, K.M., 2005b. Is the PhyloCode now roughly analogous to theactual codes? A reply to Laurin et al. Cladistics 21, 608–610.

Pleijel, F., 1999. Phylogenetic taxonomy, a farewell to species, and a revi-sion ofg Heteropodarke (Hesionidae, Polychaeta, Annelida). System-atic Biology 48, 755–789.

Pleijel, F., Rouse, G.W., 2003. Ceci n’est pas une pipe: names, clades andphylogenetic nomenclature. Journal of Systematic and EvolutionaryResearch 41, 162–174.

Polaszek, A., Wilson, E.O., 2005. Sense and stability in animal names.Trends in Ecology and Evolution 20, 421–422.

Rieppel, O., 2006. The PhyloCode: a critical discussion of its theoreticalfoundation. Cladistics 22, 186–197.

Simpson, G.G., 1940. Types in modern taxonomy. American Journal ofScience 238, 413–431.

Simpson, G.G., 1961. Principles of animal taxonomy. Columbia Univ.Press, New York. i–xii + 1–247.

Sluys, R., Martens, K., Schram, F.R., 2004. The PhyloCode: naming ofbiodiversity at a crossroads. Trends in Ecology and Evolution 19,280–281.

Smith, H.M., Lynch, J.D., Reese, R.W., 1966. Rana maculata Daudin, 1801(Amphibia): proposed suppression under plenary powers. Bulletin ofZoological Nomenclature 23, 169–173.

http://www.bio.pu.ru/win/entomol/KLUGE/zoo-name.htm

http://www.bio.pu.ru/win/entomol/KLUGE/zoo-name.htm


Starobogatov, Y.I., 1991. Problems in the nomenclature of higher taxo-nomic categories. Bulletin of Zoological Nomenclature 48, 6–18.

Stevens, P.F., 2002. Why do we name organisms? Some reminders of thepast. Taxon 51, 11–26.

Strickland, H.E., Henslow, J.S., Phillips, J., Shuckard, W.E., Richardson, J.,Waterhouse, G.R., Owen, R., Yarrell, W., Jenyns, L., Darwin, C.,Broderip, W.J., Westwood, J.O., 1843. Series of propositions for render-ing the nomenclature of zoology uniform and permanent, being the

Report of a Committee for the consideration of the subject appointedby the British Association for the Advancement of Science. Annals &Magazine of Natural History 11 (1), 259–275.

Veith, M., Kosuch, J., Ohler, A., Dubois, A., 2001. Systematics of Fejer-varya limnocharis (Gravenhorst, 1829) (Amphibia, Anura, Ranidae)and related species. 2. Morphological and molecular variation in frogsfrom the Greater Sunda Islands (Sumatra, Java, Borneo) with the deW-nition of two species. Alytes 19, 5–28.

2006 Dubois MPE Cautionary Tale

Documents