Records of the Western Australian Museum Supplement No. 63: 1-7 (2001).

Introduction to the checklists of the vertebrates of Western Australia

K.P. Aplin, N.K. Cooper, R.A. How, J.B. Hutchins, R.E. Johnstone and L.A. SmithWestern Australian Museum, Francis Street, Perth, Western Australia 6000, Australia

INTRODUCTIONThis publication of the Western AustralianMuseum presents, for the first time, acomprehensive listing of the vertebrate animalsrecorded from Western Australia and thesurrounding seas. The publication is primarilydesigned to satisfy a growing demand for anauthoritative listing of Western Australianvertebrates for 'official' governmental usage.However, it is also intended to satisfy a wideraudience who may require information of variouskinds regarding the fauna of the state, includingstudents at all levels, fellow taxonomists andmembers of the public with a special interest in ouruniquely fascinating native vertebrate fauna.. The task of preparing and maintaining regionalfauna checklists has long fallen to museums aroundthe world. This reflects the concentration inmuseums not only of the vast bulk of the world'sfauna collections, but also a sizeable proportion ofits community of taxonomic specialists. TheWestern Australian Museum, established in 1891, isby world standards, a relatively new institution.However, its collection of almost 300,000 regionally-derived vertebrate specimens, representing morethan 4600 taxa, has few equals in either size ordiversity. The Western Australian Museum'shistory of contribution to vertebrate taxonomy isalso highly distinguished, coloured by the careersof such prolific vertebrate taxonomists as LudwigGlauert, Glen M. Storr, Gerald R. Allen and DarrellJ. Kitchener. The present series of Checklists arededicated to the labours of these former staffmembers, who between them laid much of thefoundation of our current understanding ofvertebrate diversity in Western Australia.

Why publish a checklist?The major function of a formal checklist is to

provide an authoritative reference to the scientificnames of a particular group of organisms.However, a checklist is more than a simple list ofnames. By including a particular taxon on achecklist, the author is expressing an opinion thatthe particular scientific name is not only availableunder the terms of the International Code forZoological Nomenclature, but also that it is theearliest available and thus correct name for a valid

bi<?logical entity - usually a species or sub-species.These decisions often follow established views andusage, but in some cases they may represent anotherwise divergent opinion based on the author'sown original research or novel interpretation ofpreviously available evidence. Where the content ofthese checklists departs significantly from existingcompilations, we have endeavoured to provideadequate justification in the footnoted comments orby reference to appropriate sources.Increasingly, checklists are also coming to be

viewed as syntheses of biodiversity information,and are used extensively as such in both academicand management contexts. In certain circumstances,inclusion on an official list may even be prerequisitefor allocation of scarce conservation funds orinstigation of protective measures. In this regard,checklists are rapidly taking on a significantpolitical role in addition to their more usualscientific one.Other potential functions of a checklist,

depending on the particular style and content, areto provide an insight into contemporary systematicthinking and to stimulate, perhaps even guide,future research. In order to enhance this latterfunction, we have attempted through our footnotedcomments to highlight a number of species andhigher taxa in such cases where available evidencesupports the presence of additional taxa or the needfor revision of supraspecific taxa. However, sincemany groups of vertebrates have not been subjectto modern taxonomic study, the series ofaccompanying notes should not be taken as anexhaustive catalogue of unresolved taxonomicissues. In reality, there are probably many more asyet unidentified problem areas.

The future of the checklistsAs indicated above, the task of cataloguing thevertebrates of Western Australia is far fromcomplete. Among the various groups of fishes, frogsand reptiles, entirely new species are describedalmost every year and many new taxa awaitdescription. The description of new species of birdsor mammals occurs less frequently. However, manytaxa among these high-profile groups have not yetbeen subject to careful morphological analysis orany form of molecular genetic assessment. More

DOI: 10.18195/issn.0313-122x.63.2001.001-007

2

often than not, the results of such studies challengetraditional taxonomies based on morphological orfield studies, and there are many examples wheremolecular genetic studies have revealed additionalcomplexity in what were thought to be well-studiedtaxa.At a finer level of analysis, patterns of geographic

variation have been investigated in a fair proportionof bird species and in a few fish, reptile andmammal species. Geographically isolated 'races' arecommonly treated as subspecies, although the valueof such usage has been variably contested anddefended. Only very rarely is existing subspeciesnomenclature based on a combination ofmorphological and molecular genetic data.But the description of new species and finer

discrimination of patterns of geographic variationare not the only reasons why the checklistspresented in this volume are certain to undergofuture modifications. As any reader of the journalSystematic Biology would appreciate, the science ofclassification itself is in a process of constant reviewand change, with significant debate at present overtheoretical and practical aspects of 'species'concepts and methods of phylogeneticreconstruction, and even over our system ofclass-ification itself. These issues are not purely'academic', but rather have a very direct impact onpractical taxonomy, especially as it is mediated bythe process of peer-reviewed publication. Severalcritical issues will be discussed briefly below as ameans of introducing our own approach incompiling this checklist.

Species Concepts: in theory and in practiceThe concept of a 'species' is so fundamental to

biology that it almost seems ludicrous that it needsto be discussed. However, as will become evidentfrom even a brief foray into the vast and ever-growing literature on species concepts, the issueembodies a suite of controversial issues at variouslevels ranging from the philosophical to thepractical [see Claridge et al. (1997) and Wilson(1999) for a broad variety of approaches]. Here wewill limit our discussion to the more practicalapplication of species delineation as it pertains tovertebrate species, the great majority of which aresexually reproducing [see Echelle (1990) for anintroduction to the taxonomy of 'clonally'reproducing vertebrates].Much debate in the past has been concerned with

the question of whether the species category is a'real' as against an arbitrary unit of classification(e.g., Reippel, 1986; Kluge, 1990). Today, mosttheoretical taxonomists find it useful to think ofspecies as 'individuals' rather than 'classes', whichmeans that they are self-delimiting andindependent of human perception. Anotherimportant attribute of 'individuals' is that they are

-------~

Introduction

historical entities, with a distinct time and place oforigin and of demise, and a unique history that setsthem apart from other related individuals.,Admittedly, these are abstract notions, but theyserve to establish a theoretical framework withinwhich to formulate more practical taxonomicprocedures.At the practical level, it is useful to consider two

different biological scenarios - one in which anumber of species are found living together (i.e.,found in sympatry); and another where the variousspecies do not overlap in geographic range (i.e.,allopatric distributions). The former case is by farthe simplest scenario in which to make objectivetaxonomic decisions.Where two or more species live in a mixed

community, they are generally provided withopportunities to interbreed. If they are sufficientlydifferent in reproductive behaviour or anatomy orgeneral ecology (collectively known as 'pre-matingisolating mechanisms'), there may be few actualattempts to interbreed and hence, little or nosharing of genetic material. However, even wheremating does occur, the contributed geneticmaterials may differ in ways that either preclude itssuccessful recombination, or else lead to abreakdown of embryological development orpractices of parental care (collectively termed 'post-mating isolating mechanisms').Where the various potential isolating mechanismsdo not prevent interbreeding and successfulreproduction, the resultant transfer of geneticmaterial will tend to very quickly eliminate anygenetic and morphological differences that mighthave existed between sub-populations. Conversely,to find differences either in genetic properties or ingenetically-determined morphology between co-occurring or sympatric populations constitutes verystrong evidence of separate species. This is of coursethe basis of the traditional 'Biological SpeciesConcept' (BSC), first explicitly formulated byDobzhansky (1937) and Mayr (1942) and sincefollowed by the majority of animal taxonomists.Traditionally, the recognition of sympatric speciesdepended on the ability of a taxonomist to detectsometimes subtle differences in morphology orbehaviour. These days, such decisions are morecommonly made with reference to moleculargenetic markers, resulting in vastly improved levelsof both objectivity and certainty. Typically, asuspicion of sympatric species, aroused bymorphological studies, is then either confirmed orrefuted by analysis of genetic markers thateffectively test the null hypothesis that allindividuals in the combined population form asingle reproductive pool. Recent case studiesinvolving Western Australian vertebrates includeAplin and Adams (1998) and Donnellan et al. (2000).Where groups of closely-related fo~s have non-

Introduction

overlapping distributions, the opportunities tointerbreed are rare to non-existent. Application ofthe Biological Species Concept in such cases thusrequires the taxonomist to enter the realm of thehypothetical; specifically, to assess the likelihoodthat successful interbreeding would occur ifmembers of two geographically-isolatedpopulations were brought together under naturalconditions. Ideally, such decisions would be basedon detailed knowledge of pre- and post- matingisolating mechanisms in the group in question. Inpractice, such information is rarely available, andmost taxonomic decisions are based instead on suchsubjective procedures as comparing degrees ofmorphological divergence.Widespread dissatisfaction with the Biological

Species Concept, especially as it pertains to theproblem of allopatric populations, has resulted in aplethora of theoretical and practical contributionson the subject of alternative Species Concepts (seeAvise and Ball 1990, Frost and Hillis 1990; Mallett1995 for useful reviews). Here we will examine theimplications of what is emerging as the dominantnew paradigm - the Phylogenetic Species Concept(PSC).The fundamental tenet of the PSC is that species-

level units should equate to discrete evolutionarylineages (Nixon and Wheeler 1990; Baum 1992). Ingenera containing relatively few, well-differentiatedspecies, this condition is easily satisfied. However,problems arise where a widely distributed taxonhas given rise to distinctive local offshoots; in suchcases the 'parent' taxon would be paraphyletic if itcontained only some of its derivative lineages. Thissituation may actually be quite common inAustralia, where many species have probablyundergone major expansions and contractions inrange in response to environmental change, andwhere there are many examples of species-groupscomprising one or two widely-distributed taxatogether with variable numbers of locally restrictedendemics (e.g., the Lerista nichollsi species-group ofskinks).At the operational level, the PSC defines species

as "the smallest aggregation of populations (sexual)or lineages (asexual) diagnosable by a uniquecombination of character states in comparableindividuals (semaphoronts)" (Nixon and Wheeler1990: 218; see also Davis and Nixon 1992). A speciesclassification based on this principle would equateto the terminal nodes on a cladograin, therebyeffecting a closer integration of phylogenetic theoryand practice. However, for most groups oforganisms, practical application of this principlewould lead to a very significant increase in thenumber of species, including elevation of virtuallyall populations currently identified as subspecies tofull species level. A good example is Cracraft's(1992) phylogenetic classification of the Birds-of-

Paradise in which he recognized a total of 90 speciescompared with the 40-42 listed by classicaltaxonomists such as Mayr (1962) and Gilliard(1969).Even more problematic, however, is the potential

for the increasingly sensitive techniques ofmolecular genetics to diagnose extremely transitoryand/or local populations (e.g., demes or peripheralpopulations based on small founder groups),irrespective of the degree of divergence inmorphology or general biology. For example, underthe PSC, various isolated populations of each of theendemic southwestern Australia frog species,Geocrinia rosea, G. lutea and G. alba, could bediagnosed as a distinct species based on fixed allelicdifferences (Driscoll 1998a, 1998b). The solution, ofcourse, is for the PSC to be applied within thecontext of metapopulation theory (Hanski andGilpin 1998), thereby making allowance fordifferences in demographic structure, populationdynamics and levels of genetic variability; however,what might be lost is the essential (or perhapsnaive) simplicity of the PSC as it exists today.Not surprisingly, the Phylogenetic SpeciesConcept has thus far failed to bring aboutWidespread changes in taxonomic practice.Nevertheless, it is interesting to note an increasingfrequency of mention of the PSC in the context ofrecent systematic revisions, and a growingemphasis on its key elements of monophyly anddiagnosability. Schodde and Mason (1999), forexample, in their recent classification of Australianbirds, refrained from employing the PSC, butinstead introduced a new concept, the 'Ultrataxon',defined as "regional inter-breeding populations ofbirds that differ discontinuously from neighbouringrelatives in at least one morphological character thatis presumed to be genetically based" (Schodde andMason, 1999: 4). The resultant 'ultrataxa' are inreality equivalent to monotypic species or tosubspecies where significant regional differentiationis present, and also correspond to 'terminal taxa' orPhylogenetic Species in the sense of Nixon andWheeler (1988). The only real advantages of the'ultrataxon' label are therefore to avoid the "stigmaattached to 'subspecies' in conservation biology andelsewhere" (Schodde and Mason, 1999: 3), and toavoid "ambiguity and confusion in classification"through contrasting usage of the term 'species'. InSchodde and Mason's system, the term 'species' isretained for the larger evolutionary units that arebelieved to have attained the state of reproductiveincompatibility; Le., Biological Species sensuMayr.For the present checklists, we have retained the

more traditional use of species and subspecies. Inbroad terms, these categories are employed in thesense of the conceptual paradigm of the BSC, that isspecies are essentially biological entities maintainedby intrinsic attributes promoting reproductive

4 K.P.Aplin,N.K.Cooper,R.A.How,J.B.Hutchins,R.E.Johnstone,L.A.Smith

isolation,while subspecies are essentiallygeographicentitiesmaintained byextrinsicfactorsthateffectivelydenyreproductiveinteraction.Underthisdefinition,therecognitionofsubspecieswithinaspecieswillgenerallyindicatetheexistenceofmorphologically-diagnosable populationsinstrictallopatry.Where closely-relatedtaxaareinnarrowcontactandshowevidenceofhybridactivity,thedecisiontorecognizetheformsasspeciesorsubspeciesgenerallyrestsonevidenceregardingthefrequencyandspatialscaleofhybridizationandtheextentofintrogression.Atoneextreme,occasionalorspatiallyrestrictedhybridizationwithlimitedintrogressiongenerallyhaslittleimpactontheseparateidentityofpopulations. On theother hand, frequent,widespread hybridization and effectiveintrogressionislikelytoeliminategeneticandmorphologicaldistinctioninonlyafewgenerations.

Wherever possible,wehaveeliminatedanypriorsubspecificnomenclaturethathasexistedsolelytosignifygeographicalisolatesintheabsenceofmorphological orotherdifferentiation,ortodesignatesubdivisionsorpointsalongextendedclines.Both practiceswere ofcoursestandardpracticeduringtheearlieryearsofsystematicsinAustralia, aselsewhereintheworld (WilsonandBrown1953).

Exactlyhowthesetaxonomicprinciplesareappliedinpracticedependsonwhat kindsofevidenceareavailableforanygroupofanimals.Underidealcircumstancestaxonomicassessmentsshouldbebasedonacombinationofmoleculargenetic, ecological and ethological, andmorphologicalevidence.However,thisidealisonlyrarelyattained,andmanydecisionsarethusmadeintheabsenceofcriticalevidence.Insuchcases,somerankingofevidenceisprobablyadvisabletoassessthepotentialstabilityofanydecision.

Inmany respects,molecular geneticevidenceprovidesthemostdirectevidenceonwhichtobasetaxonomicdecisions. The levelof geneticdivergence, as measured by allozymeelectrophoresis,directlymeasurestheextentofgeneflowbetweenpopulations.Asnotedearlier,thiscanprovideapowerfultesttoeitherconfirmorrefutetheexistenceoftwoormore morphologically-similarspeciesinsympatry.Inthecaseofallopatricpopulations,thelevelofgeneticdivergencecanbeusedtoestimatetheextentofcontemporarygeneflowand,providedsomethingisknownofthepopulationdynamicsanddemographyoftheorganism,toestimatetheapproximatedurationofisolationbetweenthegenepools(expressedintermsofgenerations).However,amorefrequently-usedapproachistocomparepatternsofgeneticdifferentiationwithin andbetweenpopulations.GeorgesandAdams(1996:251),forexample,notedthat"thefourfixeddifferencesbetweenElseya

latisternumandtheundescribedformofElseyafromtheGwydirRiver....standsinstarkcontrasttotheabsenceoffixeddifferencesamongpopulationsofElseyalatisternumrangingfromtheRichmondRiverinNSWtoCapeYorkandArnhemLand".

At abroadercomparativelevel,thereisalsoawealthofempiricalevidencetosuggestthatcertainlevelsofgeneticdivergence(asmeasured byallozyrneelectrophoresis)areassociatedwith theattainmentofreproductiveisolation.Thorpe(1982)forexample, foundthatamongnon-avianvertebrates, 97% ofpublished interspecificcomparisonsgaveNei'sDinexcessof0.16.Avise(1975)reportedsimilarfiguresamongthemajorityofgroupsofbirds,butwith consistentlylowervaluesincertainfamilies.However,inallgroupsinvestigated,thedatashowabroadoverlapofvaluesbetweenintraspecificandinterspecificcontrasts,thereforenegatingtheuseofagenetic'yardstick',evenwithinasinglegroup.

Recent developmentsinthefieldofmolecularsystematicshavealsoprovidedanewtoolfortaxonomyintheformofmaternally-inheritedmitochondrial gene trees. By combininginformationfromtwoormore suchtrees,thepatternofhistoricaldescentofindividualsandpopulationscanbereconstructedwith ahighdegreeofconfidence.ThisapproachisparticularlyimportantinthecontextofthePhylogeneticSpeciesConcept,asitcanpotentiallydistinguishmonophyletic fromnon-monophyletic(para-orpolyphyletic)taxa.SiblingspeciesrecognisedunderthePSCareexpectedtosatisfythecriterionofreciprocalmonophyly. A combinedanalysisusingallozymeelectrophoresisandsomeformofmitochondrial DNA analysisislikelytoprovidethebestestimateofpopulationphylogenyandhencethemost reliableandstablespecies-leveltaxonomy.

PerhapsthenextmostusefulbodyoftaxonomicinformationisrepresentedbycomponentsofwhatPaterson(1980,1985)termedthe'SpecificMateRecognition System' (SMRS)-thesuiteofethological,anatomicalandbiochemicalattributesthattogetherservetoestablishadistinctionbetweenmoreandlessappropriatematingpartnersandtherebylimitthefieldforpotentialgeneticrecombination. The SMRS can includeadvertisementcalls,pheromones,sexualdisplays,detailsofreproductiveanatomyetc.,allofwhicharepotentiallyunderstrongsexualselectionandthuscapableofveryrapiddivergence(Amold1986;TurnerandBurrows1995).Atthepracticallevel,taxawhichdifferinoneormorekeycomponentsoftheSMRScanprobablybeassumedtohaveattainedasignificantlevelofreproductiveincompatibilityundernaturalconditions.

Theextentofgeneralmorphological divergencebetweenpopulations,despitebeing ~ traditional

Introduction

basis for most original species descriptions, inreality provides the least satisfactory basis fortaxonomic ranking. In several well-studied groupsof vertebrates, there is strong evidence forsignificant morphological sub-division withinpanmictic species, and conversely, for theattainment of reproductive incompatibility in theabsence of significant morphological divergence(e.g., Larson 1989). Particular caution must beexercised where putative taxa are based on patternsof geographic distribution of one or a small numberof morphological characters, especially where suchcharacters may represent the expression of one or afew genetic loci. In such cases, there is a cleardanger that the resultant taxonomy is in fact atypological construct based on a pattern of selectionor drift operating on a few loci, and non-representative of patterns of variation in other,possibly less visible, but in a taxonomic sense, noless significant, genetic loci.In concluding this discussion of species-level

classification it is also worth noting that the mostcompletely-investigated examples of speciesradiations, incorporating a variety of different linesof molecular and morphological evidence, rarelyever result in unambiguous, hierarchicaltaxonomies. The reason for this is that intra-specificevolutionary processes, operating at the level of thedeme or avatar (sensu Damuth 1985), areintrinsically reticulate in nature, and are thereforeincompatible with the simple models ofcladogenesis that are encapsulated within ourhierarchical Linnean system of classification. As ageneral rule of thumb, the more we learn about theevolutionary history of a group of related species,the less likely we are to be able to express thoserelationships via standard taxonomic conventions.

The meaning of higher categories: genera,families etc.Traditional arrangements of species into genera

and families were often essentially typological, asimple bringing together of groups of species on thebasis of general similarity. In some cases, thisresulted in natural evolutionary groups, but moreoften the groupings were based on suites of sharedprimitive features or common ecologicaladaptations.Today, most taxonomists agree that systems of

classification above the species-level should attemptto reflect the evolutionary relationships of theindividual species. While this is a laudable goal, formany groups 'of organisms, it remains a tall order.Among the various groups of Australianvertebrates, knowledge of even the most basicanatomical systems (e.g., skeletal anatomy) isusually far from complete, and while a rapidgrowth in the number of molecular studies isleading to improved phylogenetic knowledge in

5

many groups, these studies are often based on onlya few members of any group.For a group of animals where much of thephylogenetic information is very patchy incoverage, the dilemma for the taxonomist is toknow just when to recommend taxonomic changes.If changes are made prematurely and on the basisof inadequate evidence, they generally requiresubsequent modifications, creating a plethora ofshort-lived combinations that mar the taxonomicliterature and serve only to alienate non-taxonomists. On the other hand, conservingtraditional generic and familial classification for thesake of taxonomic stability, even when it is clearlycontradicted by a significant body of phylogeneticevidence, is also potentially counter-productive.This is because taxonomy is far more than a passivesystem of names; rather, as emphasised earlier, itplays an active role in guiding the selection ofspecies in comparative research, and more critically,in the setting of priorities for species conservation.For these reasons, it is important that classificationsare able to change to reflect genuine improvementsin phylogenetic knowledge.The conflicting need for stability and change hasprompted some contemporary taxonomists tochampion a dual system of taxonomy - one set ofunchanging names to identify species units; and asecond system for expressing the phylogeneticrelationships among the fundamental species units(de Queiroz and Gauthier 1992; Cantino e,t al. 1999).Although such suggestions seem overly radical tomany taxonomists, they are likely to gain muchsupport from other biologists for whom the mostpressing need is to unambiguously label aparticular organism and to identify otherinformation pertaining to the same organism.However, pending further development of suchalternative systems of classification, there is acontinuing need for conventional taxonomicrevision.In the present lists, generic and higher taxonomic

groupings are based on our best estimate of thephylogenetic relationships of the included taxa,based on both published works and our ownunpublished studies. In cases where theserelationships are unclear, we have generallyadopted a conservative position by grouping taxain a way that maximizes the number ofmonophyletic taxa, but avoids any arbitrarysubdivision of taxa of uncertain phylogeneticstructure.

Style and content of the individual checklistsEach of the individual checklists differ slightly in

style and content. To a large extent this reflectsdifferences in the state of taxonomic andphylogenetic knowledge of each group ofvertebrates. However, it also reflects differences in

6 K.P. Aplin, N.K. Cooper, R.A. How, J.B. Hutchins, R.E. Johnstone, L.A. Smith

the practical application of taxonomic conventionsamong animal taxonomists.The major difference between the checklists

relates to the number of categorical ranks employedand the ordering or 'sequence' of taxa. For thefishes, taxa are grouped only at familial level, withthe families arranged and numbered according to astandard convention that approximates the higherlevel phylogenetic relationships. In the mammalchecklist the taxa are grouped at three categoricallevels only (family, order and subclass) and arelisted in alphabetic order at and below each of theselevels (ie families within orders; genera withinfamilies; species within genera). The checklist ofreptiles and frogs employs a larger number ofcategorical ranks but otherwise maintainsalphabetic listing at and below each level. Finally,the checklist of birds employs an expanded numberof categorical ranks together with an arrangementof all taxa according to a 'phylogenetic sequence'incorporating information both as to the degree ofrelatedness and the relative primitiveness or degreeof specialization of the various taxa. The value ofthis approach has been defended recently bySchodde and Mason (1999: 5).

Common or vernacular namesCommon or vernacular names are provided only

for those groups where there is an established usage(e.g., for birds and mammals). In some cases,several common names are given, reflecting a lackof uniformity in current usage and the absence ofany universally-accepted criteria for choosingbetween alternatives. Common names are notgoverned by any international code ofnomenclature equivalent to that in use for scientificnames, although there is a growing push fromcertain quarters (especially among ornithologists)toward institutionalisation of 'English' names.Many taxa, especially among the reptiles and

fishes, currently do not have any 'common' name.In most cases they probably do not need one, asthey are rarely ever seen and even less oftendiscussed outside of the scientific literature. In thesecases, it is our opinion that common names are bestleft to emerge rather than being imposed in anarbitrary manner for the sake of completeness oruniformity.

REFERENCESAplin, KP. and Adams, M. (1998). Morphological andgenetic discrimination of new species and subspeciesof gekkonid and scincid lizards (Squamata: Lacertilia)from the Carnarvon Basin region of WesternAustralia. Journal of the Royal Society of WesternAustralia 81: 201-223.

Arnold, E.N. (1986). The hemipenis of lacertid lizards(Reptilia: Lacertidae): structure, variation and

systematic implications. Journal of Natural History 20:1221-1257.

Avise, J.c. and Ball, R.M. (1990). Principles ofgeneological concordance in species concepts andbiological taxonomy. Oxford Suroeys in EvolutionaryBiology 7: 45--67.

Baum, D. (1992). Phylogenetic species concepts. Trends inEcology and Evolution 7: 1-2.

Cantino, P.D., Bryant, H.N., de Queiroz, K, Donoghue,M.J., Eriksson, T., Hillis, D.M. and Lee, M.S.Y. (1999).Species names in phylogenetic nomenclature.Systematic Biology 48: 790-807.

Claridge, M.F., Dawah, H.A. and Wilson, M.R. (eds)(1997). Species: the units of biodiversity. Chapman andHall, New York.

Cracraft, J. (1992). The species of the Birds-of-Paradise(Paradisaeidae): applying the phylogenetic speciesconcept to a complex pattern of diversification.Cladistics 8: 1-43.

Damath, J. (1985). Selection among "species": aformulation in terms of natural functional units.Evolution 39: 1132-1146.

Davis, J.I. and Nixon, KC. (1992). Populations, geneticvariation, and the delimitation of phylogeneticspecies. Systematic Biology 41: 421-435.

De Queiroz, K and Gauthier, J. (1994). Toward aphylogenetic system of biological nomenclature.Trends in Ecology and Evolution 7: 27-31.

Dobzhansky, T. (1937). Genetics and the Origin of Species.Columbia University Press, New York.

Donnellan, S.c., Aplin, KP. and Dempsey, P.J. (2000).Genetic and morphological variation in AustralianChristinus (Squamata: Gekkonidae): preliminaryoverview with recognition of a cryptic species on theNullarbor Plain. Australian Journal of Zoology 48: 289-315.

Driscoll, D.A. (1998a). Genetic structure, metapopulationprocesses and evolution influence the conservationstrategies for two endangered frog species. BiologicalConseroation 83: 43-54.

Driscoll, D.A. (1998b). Genetic structure of the frogsGeocrinia lutea and Geocrinia rosea reflects extremepopulation divergence and range changes, notdispersal barriers. Evolution 52: 1147-1157.

Echelle, A.A. (1990). Nomenclature and non-mendelian("clonal") vertebrates. Systematic Zoology 39: 70-78.

Frost, D.R. and Hillis, D.M. (1990). Species in conceptand practice: Herpetological applications.Herpetologica 46: 87-104.

Georges, A. and Adams, M. (1996). Electrophoreticdelineation of species boundaries within the short-necked freshwater turtles of Australia (Testudines:Chelidae). Zoological Journal of the Linnean SocietyLondon 118: 241-260.

Gilliard, E.T. (1969). Birds of Paradise and Bower Birds.Weidenfeld and Nicholson, London.

Hanski, LA. and Gilpin, M.E. (eds) (1998). MetapopulationBiology. Academic Press, New York.

Kluge, A.G. (1990). Species as historical individuals.Biology and Philosophy 5: 417-431.

Larson, A. (1989). The relationship between speciationand morphological evolution. In D. qtte and J.A.

Introduction

Endler (eds), Speciation and Its Consequences: 579-598.Sinauer, Sunderland, Massachusetts.

Mallett, J. (1995). A species definition for the ModemSynthesis. Trends in Ecology and Evolution 10: 294-299.

Mayr, E. (1942). Systematics and the Origin of Species.Columbia University Press, New York.

Nixon, K. C. and Wheeler, Q. D. (1990). An amplificationof the phylogenetic species concept. Cladistics 6: 211-223.

Paterson, H. E. H. (1980) A comment on "materecognition systems". Evolution 34: 330-331.

Paterson, H. E. H. (1985). The recognition concept ofspecies. In E. S. Vrba (ed.), Species and Speciation: 21-

7

29. Transvaal Museum Monograph No. 4, TransvaalMuseum, Pretoria.

Reippel, O. (1986). Species are individuals: a review andcritique of the argument. Evolutionary Biology 20: 283-317.

Schodde, R. and Mason, I.J. (1999). The Directory ofAustralian Birds. Passerines. CSIRO Publishing,Collingwood, Victoria.

Turner, G.F. and Burrows, M.T. (1995). A model ofsympatric speciation by sexual selection. Proceedingsof the Royal Society, London B260: 287-292.

Wilson, R.A. (ed.) (1999). Species: New interdisciplinaryessays. MIT Press, Cambridge, Massachusetts.