ANEWHYPOTHESISOF DECAPODPHYLOGENY

BY

CHRISTOPHER J.DIXON 1/,SHANE T.AHYONG 2/ andFREDERICK R.SCHRAM 1,3/ 1/ Institutefor Biodiversity and Ecosystem Dynamics, University of Amsterdam, Mauritskade 57, NL-1092AD Amsterdam,Netherlands 2/ AustralianMuseum, 6 CollegeStreet, Sydney, NSW 2010,Australia

ABSTRACT

Acladisticanalysis based on external morphology was carried out on 60 taxa of decapod .An analysis with unordered characters and one with ordered characters were both in agreementregarding the major relationships. The ordered analysis gave better resolution of more advancedclades, while the unordered analysis gave better resolution of more basal clades. None ofthe traditional groups Palinura, Anomura, and Macrura is monophyletic. A newclassiŽ cation of decapodcrustaceans is proposed.Homarida and Astacida are closely related, as shownby theunique processon the ischium of their Ž rstpereiopods. Glypheoidea forms the sister group to Astacura, withinan enlarged . is the sister group to Meiura ( + Brachyura)in a newclade, Eurysternalia, characterized by auniqueantennular morphology and by theeponymous widesternum of its members. Thalassinida emerge as the sister group to Eurysternalia, in a new clade,Sterropoda, characterized by fusionof the Ž rstsegments of the thoracic limbs. The fractostern isinterpreted to beaeureptantfeature, and a possibleburrowing habitus is positedfor the ancestral Eureptantia.

RÉSUMÉ

Uneanalyse cladistique fondé e surla morphologieexterne a étémené e sur60 taxons de Crustacés Décapodes. Les deux analyses, l’ uneutilisant les caractè res non ordonné s, l’ autreles caractè res ordonnés étaienttoutes les deux en accordsur les principales relations. L ’analyseordonné e aprésenté unemeilleure ré solution des clades les plus avancé s, tandis que l’ analyse non ordonné e adonnéune meilleureré solution des clades plus basaux. Aucun des groupes traditionnels Palinura, Anomura et Macruran’ estmonophylé tique. Une nouvelle classiŽ cation des Crustacé s Décapodes est proposé e. LesHomarida et les Astacida sont é troitementapparenté s, comme dé jà montré par le processus uniquesur l’ ischium de leurspremiers pé ré iopodes. Les Glypheoidea constituent le groupe-frère des Astacura,à l’intérieur du groupe é largides Astacidea. Les Achelata sont le groupe-frère des Meiura (Anomala+ Brachyura)dans un nouveau clade, Eurysternalia, caracté risé par une morphologie uniquede l’ antennuleet par le large sternum é ponymede ses membres. Les Thalassinida é mergent commele groupe-frè re des Eurysternalia, au sein d’ un nouveau clade, les Sterropoda, caracté risé par

3/ Correspondingauthor; e-mail: [email protected] © KoninklijkeBrill NV ,Leiden,200 3 Crustaceana 76(8):935-975 Alsoavailable online: www.brill.nl 936 C.J.DIXON, S.T.AHYONG &F.R.SCHRAM lafusiondes premiers segments des appendices thoraciques. Le fractosterne est interpré té comme un caractère de vrairampant , etunhabitus fouisseur possible est supposé pour l’ Eureptantiaancestral.

INTRODUCTION

Theorder is aremarkablydiverse groupof malacostracans, bothmor- phologicallyand ecologically, and is agroupof great economicand environmental importance.Gruner et al. (1993)estimate 10,000species inthe group,but actual numbersmay be at least half againas large.The relationships withinthe order havebeen the subject ofdebatefor decades; opinionsdiffered in the early years as to whichcharacter(s) shouldbe usedto classify decapods,be it tail length(from Linnaeusuntil Boas,1880), gill type(Huxley, 1878), number of chelae (Beurlen &Glaessner, 1930),or modeof locomotion (Boas, 1880). These sometimes sim- plistic approacheshave now been replaced by moreholistic methodsusing a wider variety ofcharacters. DeSaint Laurent(1979) believed that “les seuls critères dela morphologieexterne : : : nepermettront pas derésoudre [les] relations phylétiques entre les différentes lignées deDécapodes” , butwe wouldargue that whilst larval, spermatozoal,and other characters are useful,external morphologycan be enough to workout such relationships. Wewill nonetheless makereference to otherchar- acters andtheir agreementor disagreement with the scheme weproposebased on external morphology. Boas in his seminal workof 1880divided Decapoda into Natantia forthe swim- mingforms, and Reptantia forthe walkingforms. Although Reptantia is amono- phyletic group,Natantia appearsto beless valid andas longago as 1907,Bor- radaile suggestedthat Natantia was actually paraphyletic.Burkenroad (1963) Ž - nally deconstructedNatantia, recognizingthat the three groupsof natant decapods are notclosely related. AmongReptantia, Boas recognizedsix groups:Homari- dae,Loricata, Eryonidae, Thalassinidae, Anomala,and Brachyura. Although some names havechanged, all these groupsare generallyconceded to be monophyletic (cf.Martin &Davis,2001). That belief will also betested bythis study. It is perhapsa shame that at the turnof the twentieth century,authors reverted to the oldersystem ofH. Milne Edwards(1834), dividing Reptantia into Palinura, Astacura,Anomura, and Brachyura, which have remained as traditional groups. Noneof the fourhas remainedunchallenged as amonophyletictaxon. Several authors(Abele, 1991; Forest &DeSaint Laurent,1989) have concluded that Palinura was untenableas agroup,and others haveconsidered the differences betweenThalassinida andAnomala (which collectively makeup Anomura)to be insurmountable(De Saint Laurent,1979; Martin &Abele,1986; Tudge, 1997). Scholtz &Richter (1995)separated the twoparts ofthe Astacura,and there has