sign in sign up
<<
Home , Carapace, Cephalocarida, Crustacean, Hoplocarida, Malacostraca, Mictacea

The University of Maine DigitalCommons@UMaine

Marine Sciences Faculty Scholarship School of Marine Sciences

6-1-2009 The lP ace of the in the Malacostracan Pantheon Les Watling University of Maine - Main, [email protected]

C. H.J. Hof

F. R. Schram

Follow this and additional works at: https://digitalcommons.library.umaine.edu/sms_facpub

Repository Citation Watling, Les; Hof, C. H.J.; and Schram, F. R., "The lP ace of the Hoplocarida in the Malacostracan Pantheon" (2009). Marine Sciences Faculty Scholarship. 134. https://digitalcommons.library.umaine.edu/sms_facpub/134

This Article is brought to you for free and open access by DigitalCommons@UMaine. It has been accepted for inclusion in Marine Sciences Faculty Scholarship by an authorized administrator of DigitalCommons@UMaine. For more information, please contact [email protected]. JOURNALOF CRUSTACEANBIOLOGY, 20, SPECIALNUMBER 2: 1-11, 2000

THE PLACE OF THE HOPLOCARIDA IN THE MALACOSTRACAN PANTHEON

Les Watling, Cees H. J. Hof, and Frederick R. Schram

(LW,corresponding) Darling MarineCenter, University of Maine, Walpole, Maine 04573, U.S.A. (e-mail: [email protected]);(CHJH) Department of EarthSciences, , Wills MemorialBuilding, Queens Road, Bristol BS8 1RJ, United Kingdom (e-mail: [email protected]); (FRS) Zoological Museum, University of Amsterdam,Post Box 94766, NL-1090 GT Amsterdam, The Netherlands(e-mail: [email protected])

ABSTRACT The stomatopodbody plan is highly specializedfor ,yet the SuperorderHoplocarida originatedfrom something other than the "lean,mean, killing machine" seen today.The record of the groupindicates that it originatedearly on froma non-raptorialancestor, with the specialized predatorymorphology developing much later. The RecentHoplocarida have been variouslyposi- tionedwithin the ,from a Subclassequal in rankto the (= Caridoida) to being placedas a Superorderwithin the Eumalacostraca.Consideration of the early fossil mor- phology,especially of the form of the ,of the positionand functioningof the articlesin the last threepairs of thoracopods,and of otherfeatures, suggests that hoplocarids are earlyderiv- ativesof a basaleumalacostracan stock that was "-like"in form.The enhancementof an ab- dominalrespiratory system most likely allowed the developmentof the anteriorthorax into the specializedraptorial system present today.

Wewant to build a phylogenetictree thatpresents the actual evolutionaryhis- tory of the organisms in it. For phylogeny to be correctly inferredfrom the wealthof data available,it is necessaryto identifyinformative homologous fea- tures that allow us to unite organisms in the tree. . . . Sorting out ho- mology from homoplasy is one of the chief pastimes of phylogeneticists. -Raff (1996)

The Hoplocaridaare best known today as Ruppertand Barnes (1994) rank the Hoplo- the Stomatopoda, the "lean, mean, carida as a Subclass of the Malacos- killing machines" of the shallow marine traca,equivalent in rankto the world. Yet, while it is obvious that the mor- and Eumalacostraca;Meglitsch and Schram phology embodied in the modem stomato- (1991) move the Phyllocaridato a new Class pods is highly specialized, the origins and Phyllopodaand rank the Hoplocaridaand Eu- affinitiesof that morphologyare not quite as malacostracaas subclasses of the Malacos- clear (e.g., Schram, 1969, 1974, 1986; traca;and Brusca and Brusca (1990) place the Hessler, 1982; Dahl, 1983a; Wheeler, 1998; Hoplocaridaas a Superorderwithin the Eu- Wills, 1998). In this paper, the base of the malacostraca,leaving Phyllocaridaas a sub- hoplocaridanclade will be examinedwith a class of the Malacostraca,essentially as Cal- view to establishing its affinities and taxo- man had done. Among treatmentsdealing nomic relationshipswith other malacostra- strictlywith ,Bowman and Abele cans. (1982), Schram (1986), and Forest (1994) Hoplocaridahave historicallybeen difficult move the Hoplocaridaout of the Eumalacos- to place. Calman(1909) arrangedthe hoplo- traca, whereas Kaestner(1970) divides the caridansalong with other groups within the Malacostracainto six superorders,one of Series Eumalacostracaalongside the Series which is the Hoplocarida.In the presentpa- Phyllocarida.There seems to be no univer- per, we treat the hoplocarids as the sister sally accepted position for the group today, of the syncarid/eucaridline (Table 1). however. Modem invertebratezoology text- From a paleontologicalas well as neonto- books have taken a variety of approaches: logical perspective, the Hoplocaridais rep-

1 2 JOURNALOF CRUSTACEANBIOLOGY, VOL. 20, SPECIALNO. 2, 20u0

Table 1. Conceptual arrangement of the malacostracan orders with stenopodous thoracic limbs (= Eumalacostraca, sensu lato), after Watling (1983, 1999) and Schram (1986), with modifications according to the suggestions in this paper. Known geological ranges are given. Taxa marked with a '?' are of uncertain rank. This schema is not to be seen so much as a functional but as a heuristic device to focus discussions and stimulate more detailed and comprehensive cladistic analyses in the future. There is much to be gained by setting aside facies concepts and focusing on identifying and rigorously defining monophyletic groups.

Hoplocarida Palaeostomatopodat (L. -U. Carboniferous) Stomatopoda Archaeostomatopodeat (M. -U. Pennsylvanian) Unipeltata (L. -Recent) ? Palaeocaridaceat (L. Carboniferous-L. ) (-Recent) (Recent) "Carida" (L. -Recent) Aeschronectidat (M. Pennsylvanian) Belotelsonideat (Carboniferous) Waterstonellideat (Carboniferous) Euphausiacea (Recent) Pygocephalomorphat (Carboniferous-Permian) (Pennsylvanian-Recent) (?Jurassic-Recent) (Recent) Spaleaogriphacea (Mississippian-Recent) (Recent) (L. Carboniferous-Recent) (?Pennsylvanian-Recent) ? (L. Eocene-Recent) ? (M. Pennsylvanian-Recent)

resentedby one modem order,the Stomato- Calman(1909) gave a brief history of the poda and the extinct Orders Palaeosto- Malacostracaand then provided definitions of matopoda(Late Devonianthrough Late Mis- the groupshe had proposedfive years earlier sissippian)and Aeschronectida(see Schram, (Calman, 1904). These include the Series 1986). The OrderStomatopoda is subdivided Leptostracaand Eumalacostraca,the latter into two Suborders,the Unipeltata (Upper containing the Divisions Syncarida, Per- Jurassicto Recent) containingfossil taxa as acarida,, and Hoplocarida. Calman's well as the modem stomatopods,and the Ar- definitionof the Eumalacostracais as follows: chaeostomatopodea(Middle Pennsylvanian to "Abdomenof six somites (the numbermay UpperPennsylvanian) containing only a small be reducedby coalescence),the last of which groupof extinctforms. Recently, Jenner et al. typically bears a pair of ,and a (1998) uncovered the possibility that the which never bears moveable furcal Palaeostomatopodamight be paraphyleticrel- rami; no adductormuscle of the carapace; ative to the monophyleticStomatopoda. thoracic limbs rarely all similar (Euphausi- acea), typicallypediform, protopodite of two ARE HOPLOCARIDSEUMALACOSTRACANS? segments except in Stomatopoda"(Calman, As a startingpoint, the early definitionsof 1909: 148). Fromthe beginning,then, the def- the Eumalacostracaand the Hoplocaridaneed initionof the Eumalacostracahas exceptions, to be examined.Schram (1969) noted, after butseems to be centeredaround Calman's con- a review of the variouslysuggested phyloge- cept of a "caridoid"facies (Hessler,1982a). netic positions, that the "definitionof a hop- Phyllocarids differ from the other mala- locarid varied from author to author, de- costracans in having polyramous phyl- pendingon the phylogeneticinterpretation he lopodouslimbs andseven abdominalsomites, adopted"(p. 277). one of which is without appendages.Exact WATLINGET AL.: PHYLOGENETICPOSITION OF THE HOPLOCARIDA 3 definitions for the non-phyllocarid malacos- If we focus on hoplocarids and eumala- tracans have always been troublesome be- costracans, Malacostraca sensu stricto are cause of the great diversity of body forms en- united by the following features: "naupliar compassed by all the orders, both living and of three cups each with three everse sen- fossil. One solution, derived from the cladis- sory cells, polyramous and stenopodous tho- tic analysis of Schram (1986), was to move racopods, , and a postcephalic cara- the phyllocarids out of the Malacostraca and pace structure that does not envelop the ab- unite them with other phyllopodous limb- domen or thoracic limbs" (Schram, 1986: bearing groups such as the cephalocarids and 528). From there Schram noted that eumala- branchiopodans. This plan generated another costracans have a uniarticulate antennal scale suite of criticisms, but it had the advantage of and strongly elaborated abdominal muscula- making the Malacostraca more homogeneous. ture to facilitate the caridoid escape reaction, By 1969, the effective definition of Eu- while hoplocarids possess triramous anten- malacostraca had become generalized to in- nules, three-articulate thoracopod protopods, clude all malacostracans except phyllocarids: four-articulate thoracopod outer branch, at "Malacostraca generally of shrimp-like form least in the fossil forms, (if one accepts the distinguished from the Phyllocarida by non- view of Claus (1871) that the orientation of bivalve of carapace and lack of sev- the limb rotates during development), and enth abdominal somite, telson without un- dendrobranchiate-like gills on the pleopods. segmented, movably articulated caudal furca" The reliance on Calman's caridoid model (Moore, 1969: R332). This definition implic- casts a functional constraint on the eumala- itly assumes that those eumalacostracans costracans that has necessitated much dis- without a carapace, e.g., syncarids, am- cussion of loss and reduction in some of the phipods, and isopods, must have lost it sec- superorders. However, it does not provide a ondarily as argued by Calman (1909), Hessler useful starting point for delimiting the dif- (1982a, 1983), and others. The only synapo- ferences between the caridoids and the hop- morphy in that definition that may have no loids, nor from the peracarid/brachycaridline, exceptions is the absence of a seventh ab- which, as shown by Watling (1999), could be dominal somite, although there is some evi- considered a third line of radiation within the dence of the seventh somite having been pres- Malacostraca. The caridoid facies is a highly ent (e.g., in lophogastrids and perhaps in developed functional model that reaches its cumaceans). full extent in the modem caridean shrimp and After being unable to find a caridoid link is not always fully present in syncarids to the hoploid fossil taxa he studied, Schram (Schram, 1986) and mysids (Watling, 1999). (1969) suggested that the Eumalacostraca, as It perhaps would be better to think of the conceived at the time, was polyphyletic. This "caridoid facies" as an apomorphic end-point lead him (Schram, 1973, 1981) to move the of the phylogeny of the caridoid line rather hoplocarids out of the Eumalacostraca and re- than as the starting point. strict the definition of the latter to those While hoplocarids were considered by Cal- groups having had a caridoid (sensu Calman) man, Siewing, and others to be poorly de- ancestor. Bowman and Abele (1982) followed veloped, "incomplete," caridoids, Schram the lead of Schram (1969) and Kunze (1981) (1969) suggested that stomatopods were de- in moving the hoplocarids out of the Eu- rived from ancestors, such as the Aeschronec- malacostraca. tida, that did not show any of the modem car- In the 1983 Issues volume nivorous specializations. On this basis he also dealing with crustacean phylogeny, the eu- proposed that hoplocarids originated sepa- malacostracans received considerable discus- rately from the other eumalacostracans, hav- sion. Of the papers treating this group, ing evolved from a phyllocarid ancestor dif- Hessler (1983) was unique in keeping the ferent from that which gave rise to the cari- hoplocarids within the Eumalacostraca, albeit doid eumalacostracans. Kunze (1981, 1983) distinct from what he terms Caridoida. Dahl extensively analyzed internal and external (1983a), Kunze (1983), and Watling (1983) anatomical features of modem stomatopods, delimited the Eumalacostraca so as to exclude including the functioning of the foregut, and the Hoplocarida. In all those papers, no list concluded that the "independent origin of the of synapomorphies was given. Hoplocarida from an early malacostracan an- 4 JOURNALOF CRUSTACEANBIOLOGY, VOL. 20, SPECIALNO. 2, 2000 cestral stock is supported" (1983: 185). She In order to understand more about the pos- viewed the early malacostracanas phyllocarid terior thoracopods, we need first to see what in design, thus implicitly supporting Schram's can be discerned from the fossil record of the contention. Schram's view was strongly re- group. For the most part thoracopod preser- futed by Burnett and Hessler (1973), who pro- vation in hoplocarid is pretty poor. posed a scheme whereby the eumalacostracan However, in the aeschronectid, Kallidecthes ancestor could be derived from a phyllocari- richardsoni, Schram (1969, Fig. 115) illus- dan, which subsequently could give rise to the trates one specimen where all three protopo- two separate caridoid and hoploid lineages. dal articles are preserved. From the third ar- Hessler (1983) proposed a sequence wherein ticle there is a very clear junction of both in- the urmalacostracangave rise to phyllocarids ner and outer branches, with the outer branch on one hand and eumalacostracans on the clearly posterior to the inner, as is seen today. other. The hoplocarids were then an early off- This ancient leg differs from the modem sto- shoot from a somewhat caridoid-looking eu- matopod thoracopod in two important ways, malacostracan. however (Fig. 1). First, the three protopodal articles are of similar length, whereas in mod- The Problem of Stomatopodan em stomatopods the middle article is much and 8 Thoracopods 6, 7, longer than either of the other two. Second, There is a strong difference in the structure the inner branch is very long and most likely of the posterior thoracopod between hoplo- composed of more than two articles, proba- carids and caridoids. Hoplocarids are con- bly four. sidered to have a 3-articulate protopod, In modem stomatopods, the first protopo- whereas caridoids and all other eumalacos- dal article (precoxa) bears insertions of mus- tracans have only a coxa and basis. Both cles from the and has a promotor-re- groups seem to have exopods on the thoracic motor movement (Table 2). The long second legs, at least primitively. In hoplocarids, the article (coxa) has an abduction-adduction exopod is reduced and ultimately lost on all movement and bears proximally the insertion but the last three pairs of thoracopods. Cari- of two large muscles that also originate well doids use the exopod for locomotory pur- inside the body on the thoracic tergites. The poses, and it is reduced only as the power of third protopodal article moves in the abduc- the abdominal appendages develop. tion-adduction direction and is capable of Stomatopods are well known for their complete flexion, forming a "knee" at its highly modified thoracic legs, sometimes re- junction with the long second protopodal ar- ferred to as maxillipeds (in fact, Hansen ticle. It is essentially fixed to the first en- (1925) makes the case that only the first of dopodal article. The junction of the first en- these should be called a maxilliped as the dopodal article to the third protopodal arti- somite to which it belongs is fused to the cle is at an angle so that the rolling movement ). The last three pairs of thoracic legs dif- produced results in bending the plane of the fer from those of the other malacostracan leg in a slight promotor-remotorfashion. This groups in their possession of three protopo- range of movement is quite similar to that de- dal articles and a reduced number of articles scribed for by Hessler in the putative endopod (inner branch). Cal- (1982b). The exopod, attached to the poste- man and subsequent authors refer to the exo- rior margin of the third endopodal article, pod and endopod as inner and outer branches, may comprise two articles, the motion of the largely because Claus (1871) raised the pos- whole ramus being anterior-posterior. sibility that the endopod and exopod reversed Based on these observations, one might ask position during development. This statement whether the homologies, so long assumed for has often been repeated but has never been the articles of these thoracopods, are incor- verified by modern workers; Hansen (1925) rect. That is, should the articles of the leg be strongly disagreed with it. In fact, the orien- termed coxa, basis, ischium, merus, and car- tation of the branches is more anterior-pos- pus, with the 2-articulate exopod originating terior, with the "exopod" posterior to the "en- from the ischium? Or, has the possible loss dopod." This orientation seems to be present of the precoxa in most malacostracans re- from the earliest larval stages where the rami sulted in a shift of the promotion-remotion are visible (Komai and Tung, 1929). function to the body-coxa articulation?On the WATLING ETAL.: PHYLOGENETIC POSITION OF THE HOPLOCARIDA 5

- st mx

-op

,A. -it.

Fig. 1. Details of the hoplocarid thoracopod. a. photograph of the basal portion of the thoracopods of the aeschronec- tid, Kallidecthes richardsoni, clearly showing three-articulateprotopod; b. videoprint of thoracopod 7 from an uniden- tified gonodactylid stomatopod; c. close-up view of the distal part of protopodal article 2, protopodal article 3, and the proximal parts of the endopod and posteriorly located exopod. Note short third protopodal article. mn, mandible; 1stmx, first maxilla; 2ndmx, second maxilla; p, precoxa; c, coxa; b, basis; ob, outer branch; ib, inner branch (a from Schram, 1969). basis of the comparative data in Table 2, we Tung, 1929), so the possibility of a shift dis- are more inclined toward the latter alterna- tally of the articulation of the exopod cannot tive. On the other hand, the development of be entirely discounted. Other alternatives these legs shows relatively weak article could be advanced here; however, a detailed boundaries in early larval stages (Komai and study of the developmental genetic control of 6 JOURNAL OF CRUSTACEAN BIOLOGY, VOL. 20, SPECIAL NO. 2, 2000

Table 2. Comparison of thoracopods 6-8 articulation and hinge movement between an anaspid syncarid and a sto- matopodan hoplocarid (extension-flexion here is in the transverse plane).

Syncarid Hoplocarid

Articulation Hinge movement Articulation Hinge movement body-precoxa promotion-remotion body-coxa promotion-remotion precoxa-coxa extension-flexion coxa-basis extension-flexion coxa-basis extension-flexion basis-ischium promotion-remotion basis-ischium promotion-remotion (rolling motion) ischium-merus extension-flexion ischium-merus extension-flexion merus-carpus extension-flexion merus-carpus absent leg segment formation in stomatopods will single , supporting a monophyletic clade begin to supply needed information. of the Malacostraca containing both Phyllo- carida and Eumalacostraca sensu lato. Results of Previous Phylogenetic Analyses Schram and Hof (1998) coded 90 charac- Using Computer-GeneratedCladograms ters for both extant and extinct crustaceans "The position of the Hoplocarida" seems to and other . Within the Crustacea, be a heading, or the topic sentence of a para- they found most large groupings to reflect graph, in nearly all papers treating malacos- what crustacean workers have long thought tracan phylogeny (e.g., Dahl, 1983a; Hessler, to be the case (Fig. 2). On the other hand, they 1982a; Schram, 1969, 1986; Wills, 1998). expressed surprise that hoplocarids occurred Most authors, as already noted above (e.g., so far up into the eumalacostracan clade, not- Dahl, 1983a; Schram, 1974, etc.), treat the ing that this placement is at odds with the Hoplocarida as a separate subclass of the views of most earlier workers on the group. Malacostraca intermediate between the Phyl- When characters representing primarily soft locarida and the Eumalacostracasensu stricto. were removed, the hoplocarids The cladistic analysis of Schram (1986) sub- moved to the base of what could be broadly sequently removed the phyllocarids to a new construed as a malacostracan clade. A tree a Class Phyllopoda, which left the hoplocari- few steps longer showed that hoplocarids dans and eumalacostracans as subclasses of could still be considered as a sister taxon to the Class Malacostraca. This move and the Eumalacostraca. Not only does the position subsequent placement of the hoplocarids of the Hoplocarida seem remarkable, but the are supported by the cladistic analyses of other malacostracan groups appear in a vari- Wheeler (1998) and Wills (1998) (Fig. 2), but ety of topological positions as well. they used Schram-derived data sets, either di- Formation and Its rectly or with modifications. However, using Carapace Bearing on Evolution 18S rDNA, Spears and Abele (1998, 1999) Hoplocarid concluded that the Phyllopoda could not be The various modes of carapace formation supported as a monophyletic unit and that the were suggested by Dahl (1991) and reviewed Phyllocarida was the sister taxon to a clade by Watling (1999). While there seem to be containing the hoplocarids, syncarids, and eu- five variations in carapace design, the rela- carids. Depending on the tree-building tionships among them are not clearly under- method used, the syncarids were either the stood. That is, if a group such as the phyllo- sister taxon to the eucarids (Fig. 2) or to the carids, for example, is typified by the pres- hoplocarids, but the hoplocarids, eucarids, ence of a dorsal fold extending loosely and syncarids were consistently grouped in a posteriorly and laterally, whether directly

Fig. 2. Selection of phylogenetic trees dealing with the position of the Hoplocarida; Schram (1986), compilation of figure 43-1 and 43-2 (B), trees based on morphological data; Spears and Abele (1998), copy of figure 14.4, tree based on maximum parsimony analysis of 232 parsimony-informativecharacters based on 18S rDNA sequences; Wills (1998), simplified compilation of figure 15.3 (b) and 15.4, strict consensus tree based on the parsimony analysis of morphological data using ordering and weighting techniques; Schram and Hof (1998), edited copy of figure 6.8, 50% majority rule consensus tree based on the parsimony analysis of unordered and unweighted morphological data. WATLING ETAL.: PHYLOGENETIC POSITION OF THE HOPLOCARIDA 7

Schram (1986) Spears and Abele (1998)

Wills(1998) Schram and Hof (1998) Amphipoda Isopoda Thermosbaenacea Cumacea Peracarida Eucarida Mictacea Tanaidacea Palaeostomatopoda 1 Aeschronectida Hoplocarida ] Hoplocarida Stomatopoda 1 Mysida Peracarida Lophogastrida Peracarida Amphionidacea Reptantia Eucadda Euzygida Euphausiacea Belotelsonidea Waterstonellidea Palaeocaridacea i i ISyncadda Syncarida 8 JOURNALOF CRUSTACEANBIOLOGY, VOL. 20, SPECIALNO. 2, 2000

co -yncanr ". .-* ___: aa__ ------00c | Lophog;astrida m Mysida i|...... ------DEuphausiacea Decapoda m Aeschronectida

-i-L117 Palaeostomatopoda l -- - i------^ -- Stomatopoda II Amnhinnda Spelaeogrip hacea ThermosbaEenacea Mictacea -- Tanaidacea Cumacea 1 Isopoda

Fig. 3. One possible phylogenetic scenario for the Eumalacostraca accommodating morphological information as detailed in the text. Discussion of upper branch arrangements are provided in Watling (1999) for all groups except the Hoplocarida, which is from Jenner et al. (1998). Characters associated with numbered bars, and for the basal eu- malacostracan, are given in the text.

fromthe cephalonas definedby Dahl, or from ate the design, if not the origin, of the cara- the posteriormargin of the cephalothoracic pace in several fossil forms. shield,could that type of carapacebe replaced Severalfossil malacostracansare preserved by segmental pleural folds? That particular in positions that suggest they possessed a argumentwas not consideredby Dahl, but he carapaceformed by a dorsal fold with bran- did (Dahl, 1983b) suggest that once a "cara- chiostegal flaps (see for example, the phyl- pace"encompassed fused thoracicsomites (as locarid, Dithyrocaris rolfei in Schram and in the case of segmentalpleural folds), it was Homer, 1978, PI. 1, Fig. 1; and the palaeo- improbablethat that type of carapacecould stomatopod, Bairdops beargulchensis in Jen- be lost andthe individualthoracic somites re- ner et al., 1998, PI. IV), whereasothers pre- stored("Dahl's Rule"). On the otherhand, a served lying sideways sometimes have the simple dorsal fold could be long or short, carapacein line with the remainingthoracic varying with the needs of the , and and abdominal somites and sometimes not could change with little consequence.New- (e.g., the archaeostomatopodeans,Tyran- man and Knight (1984) suggested that the nophontes theridion in Factor and Feldmann, transformationfrom a dorsal fold emanating 1985, fig. 6; and Gorgonophontes peleron in from the cephalic shield to a dorsalfold pro- Schram, 1984). Applying Dahl's Rule, one duced from the posterior margin of a pro- would predict that the shortenedand more gressivelyposteriorward-developing cephalo- firmly attachedcarapace of the archaeosto- thoracic shield was a simple and not unex- matopodeanswould be easily derived from pected development. In contrast to Dahl's the longerand loosely attachedcarapace seen assertion, they suggested that the posterior in the palaeostomatopods.In addition,a small marginof the cephalothoracicshield could be breakin the side of the carapaceof the De- moved more posteriorlyor retreatanteriorly vonian shrimp, newberryi with little consequence.There is thus a sig- (P1.3, Figs. 1-3, in Schramet al., 1978), re- nificantdifference between the "fusion"con- veals whatcould be intactthoracic segments, cept of Newmanand Knight(1984) and that suggesting that the carapaceof this animal of Dahl (1991) which needs resolving. Nev- consists of a long dorsal fold with lateral ertheless,there is still the potentialto evalu- branchiostegalflaps. Of course, there are WATLINGETAL.: PHYLOGENETICPOSITION OF THE HOPLOCARIDA 9 other explanations for the carapace features low only the large carapace line. This seen in these fossils, viz., the possibility of could be characterized by the following preservation artifacts (Hof and Briggs, 1997) synapomorphies (Fig. 3, bar 2): a) carapace or that the fossils are partially disarticulated developed as a large dorsal fold with bran- shed . Much more needs to be chiostegal flaps (fused at most to the first tho- known about the evolutionary development of racic somite) covering entire thorax; and b) the carapace and its regulation, perhaps uropodswidened, forming with telson a tail fan. through the discovery of appropriatehomeotic As suggested by Burnett and Hessler gene complexes. (1973), the divergence of the caridoid and hoploid lineages must have happened soon af- A VIEWOF FUNCTIONAL ter the eumalacostracan morphology became HOPLOCARIDORIGINS established. In fact, evidence from modem The fossil record as it is currently known circulatory-system design, as well as other gives significant clues to the development of features, suggests that the hoploid ground the Hoplocarida. As Schram has pointed out, plan was laid before the caridoid the earliest hoplocarids were barely distin- became too specialized. Therefore, even guishable from the earliest eucarids. In fact, though similar in overall body design (e.g., some of the fossil specimens support the no- carapace developed as a large dorsal fold tion that all early carapace-bearing malacos- covering entire thorax), hoploids differed tracans (including the Phyllocarida should from caridoids early in their evolution in sev- one choose to include them in this group) had eral important synapomorphic features (Fig. a rather loosely attached carapace consisting 3, bar 4): a) triflagellate antennule; b) thoracic of a large dorsal fold with branchiostegal endopod reduced to four articles; c) ?devel- flaps. The issue of the Phyllocarida is a com- opment of abdominal respiratory system; and plex one, and for purposes of higher level d) reduced development of thoracic arterial nomenclature, it will be assumed in this pa- system concomitant with enhanced abdomi- per that phyllocarids are malacostracans, al- nal arterial system. though the issue can be considered to be un- The caridoid synapomorphies, meanwhile, resolved, e.g., see Schram (1986), Spears and progress in the direction of thoracic special- Abele (1998), Schram and Hof (1998). ization for functions such as respiration (Fig. The morphology of the basal eumalacos- 3, bar 3): a) loss of thoracopod precoxa and tracan is assumed to have the following fea- transfer of precoxa-body articulation to body- tures (Fig. 3): a) cephalothoracic shield does coxa joint; b) development of enhanced tho- not extend beyond first thoracic somite and racic respiratory structures; c) development bears a simple dorsal fold carapace; b) of thoracic sternal artery; and d) addition of with series of lateral arteries in both thorax diagonal muscles to the pleonites. and ; c) beginning movement of ex- The position of the Syncarida at the base pression of Hox genes posteriorly so that of this lineage is problematic. Modern anas- modification of first thoracic limb is begun pidaceansyncarids possess the sternalartery typ- (see Watling, 1999, for explanation); d) ab- ical for this entire line, but there is no hint from dominal somites without diagonal muscles; e) the fossil record that the earliest syncaridsever 3-articulate thoracopod protopod; and f) first possessed a carapace of any kind. However, if and second antennae biramous. the ancestral caridoid carapace comprised a It seems possible that the basal eumala- simple dorsal fold, it could have easily been costracan developed into two lineages, the lost (Newman and Knight, 1984). Another true caridoid/hoploid line where the carapace possibility, noted by Schram and Hof (1998), becomes enlarged and confers significant hy- is that syncarids are paraphyletic and their drodynamic as well as respiratoryadvantages, true position within these lineages is likely and the peracarid line, which ultimately de- more complicated than we now recognize. velops a small carapace for respiratory uses As can be seen (Watling, 1999), the cari- (one of the possibilities of Schram and Hof, doid lineage culminates in the highly spe- 1998; also see Watling, 1999, for details of cialized decapods where the full range of cari- synapomorphies associated with bar 1 in Fig. doid features, as delimited by Calman (1909) 3). Because we are interested here in the de- and Hessler (1983), among others, can be rivation of modem stomatopods, we will fol- seen. The hoplocarid lineage undergoes a 10 JOURNALOF CRUSTACEANBIOLOGY, VOL. 20, SPECIALNO. 2, 2000

more extensive modification, culminating in Factor, D. F., and R. M. Feldmann. 1985. Systematics the modem stomatopods. In both cases, there and paleontology of malacostracan arthropods in the Bear Gulch Limestone (Namurian) of Central Mon- is a continuous change in the carapace from tana.-Annals of the Carnegie Museum 54: 319-356. a dorsal fold extending posteriorly and later- Forest, J. 1994. Les Crustaces, d6finition, formes prim- ally over the thorax to its replacement with itive et classification.-Trait6 de Zoologie, VII (1): 1-8. the posteriorly developed cephalothoracic Hansen, H. J. 1925. Studies on Arthropoda, Part II. On shield and fused folds the comparative morphology of the appendages in the segmental pleural Arthropoda.A, Crustacea.-Gyldendalske Boghandel, (sensu Dahl). In caridoids, the fused pleural Copenhagen. 176 pp. folds form a true branchial structure whereas Hessler, R. R. 1982a. Evolution within the Crustacea. in stomatopods, the fused pleural folds are re- Part. 1. General: , , and Mala- duced in number posteriorly and in size lat- costraca. Pp. 150-185 in L. G. Abele, ed. The biology of Crustacea, Volume 1, Systematics, the fossil record, erally, forming a structure with reduced and biogeography. Academic Press, New York. branchial capability and retaining some hy- . 1982b. The structural morphology of drodynamic advantages while accommodat- mechanisms in eumalacostracan crustaceans.-Philo- ing the raptorial second thoracopod. sophical Transactions of the Royal Society of London (B) 296: 245-298. ACKNOWLEDGEMENTS ? 1983. A defense of the caridoid facies; wherein the early evolution of the Eumalacostraca is dis- It is with great pleasure that we dedicate this paper to cussed.-Crustacean Issues 1: 145-164. our friend and mentor, Raymond Manning. In the early Hof, C. H. J., and D. E. G. Briggs. 1997. Decay and min- years of Les Watling's career, he did not work on either eralization of mantis (Stomatopoda: Crus- decapods or stomatopods, two of Ray's loves. However, tacea)-a key to their fossil record.-Palaios 12: Ray always kindly gave of his time to discuss whatever 420-438. crustacean issue was on Watling's mind whenever he vis- Jenner, R. A., C. H. J. Hof, and F. R. Schram. 1998. ited the Smithsonian Institution. He has benefited greatly Palaeo- and archaeostomatopods (Hoplocarida, Crus- from the advice and wisdom received. Ray's body of work tacea) from the Bear Gulch Limestone, Mississippian on stomatopods has been a source of inspiration to both (Namurian), of central Montana.-Contributions to Fred Schram and Cees Hof. This study was supported in 67: 155-185. part by the NSF PEET (Partnership to Enhance Exper- Kaestner, A. 1970. zoology. Volume tise in Taxonomy) Program grant DEB-952173 (L. 3, Crustacea. Interscience Publishers, New York. Watling and I. Kornfield, P.I.s) and a European Union 523 pp. Marie Curie Fellowship to Cees Hof. Komai, T., and Y. M. Tung. 1929. Notes on the larval stages of oratoria, with remarks on some other LITERATURECITED stomatopod larvae found in the Japanese .-An- Bowman, T. E., and L. G. Abele. 1982. Classification notationes Zoologicae Japonenses 12: 187-237. of the Recent Crustacea. Pp. 1-27 in L. G. Abele, ed. Kunze, J. 1981. The functional morphology of stomato- The biology of Crustacea. Vol. 1, Systematics, the fos- pod Crustacea.-Philosophical Transactions of the sil record, and biogeography. Academic Press, New Royal Society of London (B) 292: 255-328. York. . 1983. Stomatopoda and the evolution of the Brusca, R. C., and G. J. Brusca. 1990. . Si- Hoplocarida.-Crustacean Issues 1: 165-188. nauerAssociates, Inc., Sunderland,Massachusetts. xviii Meglitsch, P. A., and F. R. Schram. 1991. Invertebrate + 922 pp. zoology. Oxford University Press. 623 pp. Burnett, B. R., and R. R. Hessler. 1973. Thoracic Moore, R. C. 1969. Eumalacostraca. P. R332 in R. C. epipodites in the Stomatopoda (Crustacea): a phyloge- Moore, ed. Treatise on invertebrate paleontology, Part netic consideration.-Journal of Zoology 169: R, Arthropoda4. Geological Society of America, Boul- 381-392. der, Colorado. Calman, W. T. 1904. On the classification of the Crus- Newman, W. A., and M. D. Knight. 1984. The carapace tacea Malacostraca.-Annals and Magazine of Natural and crustacean evolution-a rebuttal.-Journal of Crus- History (7) 13: 144-158. tacean Biology 4: 682-687. .1909. Crustacea. Pp. 1-346 in E. R. Lankester, Raff, R. A. 1996. The shape of : genes, development, ed. A treatise on zoology. Volume 7, Part 3. Adam and and the evolution of animal form. University of Chi- Charles Black, London. cago Press, Chicago. 520 pp. Claus, C. 1871. Die metamorphoseder Squilliden.-Ab- Ruppert, E. E., and R. D. Barnes. 1994. Invertebrate handlungen der Koniglichen Gesellschaft der Wis- Zoology, 6thEdition. SaundersCollege Publishing, New senschaften zu Gottingen 16: 111-163, 35 plates. York. 1056 pp. Dahl, E. 1983a. Malacostracan phylogeny and evolu- Schram, F. R. 1969. Some Middle Pennsylvanian Hop- tion.-Crustacean Issues 1: 189-212. locarida (Crustacea) and their phylogenetic signifi- .1983b. Alternatives in malacostracan evolu- cance.-Fieldiana, Geology 12: 235-289. tion.-Memoirs of the Australian Museum 18: 1-5. . 1973. On some phyllocarids and the origin of . 1991. Crustacea Phyllopoda and Malacostraca: the Hoplocarida.-Fieldiana, Geology 26: 77-94. a reappraisal of cephalic and thoracic shield and fold . 1974. The Mazon Creek caridoid Crustacea.- systems and their evolutionary significance.-Philo- Fieldiana, Geology 30: 9-65. sophical Transactions of the Royal Society of London 1981. On the classification of Eumalacos- (B) 334: 1-26. traca.-Journal of Crustacean Biology 1: 1-10. WATLINGETAL.: PHYLOGENETICPOSITION OF THE HOPLOCARIDA 11

. 1984. Upper Pennsylvanian arthropods from Watling, L. 1983. Peracaridan disunity and its bearing black shales of Iowa and Nebraska.-Journal of Pale- on eumalacostracan phylogeny with a redefinition of ontology 58: 197-209. eumalacostracan superorders.-Crustacean Issues 1: 1986. Crustacea.Oxford UniversityPress. 606 pp. 213-228. , R. M. Feldmann, and M. J. Copeland. 1978. The . 1999. Toward understandingthe relationship of Late Devonian Palaeopalaemonidae and the earliest the peracaridanorders: the necessity of determining ex- decapod crustaceans.-Journal of Paleontology 52: act homologies. Pp. 73-89 in F. R. Schram and J. C. 1375-1387. von Vaupel Klein, eds. Crustaceans and the biodiver- , and C. H. J. Hof. 1998. Fossils and the inter- sity crisis. Brill Academic Publishers, Leiden, The relationships of major crustacean groups. Pp. 233-302 Netherlands. in G. D. Edgecombe, ed. fossils and phy- Wheeler, W. C. 1998. Sampling, groundplans, total ev- logeny. Columbia University Press, New York. idence and the systematics of arthropods. Pp. 87-96 , and J. Horner. 1978. Crustacea of the Missis- in R. A. Fortey and R. H. Thomas, eds. Arthropod re- sippian Bear Gulch Limestone of central Montana.- lationships. Chapman and Hall, London. Journal of Paleontology 52: 394-406. Wills, M. A. 1998. A phylogeny of recent and fossil Spears, T., and L. G. Abele. 1998. Crustaceanphylogeny Crustacea derived from morphological characters. inferred from 18S rDNA. Pp. 169-187 in R. A. Fortey Pp. 189-209 in R. A. Fortey and R. H. Thomas, eds. and R. H. Thomas, eds. Arthropod relationships. Chap- Arthropod relationships. Chapman and Hall, London. man and Hall, London. , and . 1999. Phylogenetic relationships RECEIVED:24 June 1999. of crustaceans with foliaceous limbs: an 18S rDNA ACCEPTED:17 December 1999. study of Branchiopoda, , and Phyllo- carida.-Journal of Crustacean Biology 19: 825-843.