Carcinization in the Anomura - Fact Or Fiction? I

Contributions to Zoology. 67 (2) 79-123 (1997) SPB Academic Publishing bv. The Hague Carcinization in the Anomura - fact or fiction? I. Evidence from adult morphology

Patsy A. McLaughlin' & Rafoel Lemaitre^ ' Shannon Point Marine Center, Western Washington University. Anacortes, Washington. USA.; * Depart- ment of Invertebrate Zoology. National Museum of Natural History. Smithsonian Institution. Washington, D.C.. U.S.A.

Keywords: Carcinization, Anomura, Paguroidea, Galatheoidea, Hippoidea, Lomoidea, Paguridae, Li- tbodidae, adult morphology, phytogeny

Abstract crabe existe certainement, mais que les intetprtations tradi- tionnelles de ce phinomtoe, basbes sur des donates ioad^ Caicinizatioo, or the process of becoming a cnb, has been, and quates et souvent depourvues de pricisioa, sont errontes; (2) continues to be, a focal point of anomunm evolutionaiy hypoth- que les crabes lithodides n'ont pas ôlui par caicioisation i eses. Ttaditional examples of carcinization in the Anomura are partir d'un anctoe Pagure. most celebrated among hennit crabs but certainly are not limited to this group. CaicinizatioD, if it has occuned, has done so independently in all major anomuran taxa. Introduction In this critique, die traditional examples of carcinizatioo in the Anomura are reviewed and more modem variations oo the theme assessed. Potential pathways of carcinization are exam- A number of decq>ods look like crabs, but do not ined from perspectives of adult moiphology in the Paguroidea, qualify as "true" crabs (Brachyura) because of Galatteoidea, Hijîdea and Lomoidea, with emphasis on die adult morphological inconsistencies or larval Paguroidea. Specific attention is given to the theoretical trans- characteristics. Presumably, to become a "true formation of a hermit crab-like body foim into a "king crab"- crab" requires that a reptant deo^Hxl undergo like lidiodid aib. Resulting coercive evidence indicates: (I) carcinization (Borradaile, 1916) or brachyu- diat while the evolution of a crab-like body form ceitainly occurs, the tnditional applications, based on inadequate and often rization (StevQi®, 1971). Altiiough the two terms inaccurate data, are flawed; and (2) diat lithodid crabs did not appear here to be synonymous, we believe that arise from a hermit crab predecessor through the process of car- not all authors who have enq)loyed "carcini- cinization. zation" or "brachyurization" have had quite the same phenomenon in mind. For example, Martin & Abele (1986) defined carcinization as the reduction and folding of the abdomen beneath the La carcinisation (processus par lequel on devient un crabe) a thorax, whereas Sluys (1992) used carcinization He, et continue d'tee, un point focal des hypodieses sur revolution des Anomures. Les exemples les plus connus de to mean the evolution of a crab-like appearance carcinisation cbez les Anomures sont founiis par les Pagures, as in lithodids. To Blackstone (1989) hermit crabs mais le phteomine n'est certainement pas limite i ce groupe. became carcinized through broadening of the car- S'il y a eu carcinisation, celle-ci s'est indipendamment pro- âce and reducing the habitation of shells. In duite dans tous les grands groupes d'Anomures. this first article, we address anomuran carcini- Dans le present aper^ critique on passe en revue les exemples traditionnels de carcinisation dans ce groupe et on (value zation only from the perspective of adult mor- les variations plus tecentes sur ce theme. Dans la perspective de phology. la morphologie des adultes on examine les voies potentielles de We accept that the Anomura, as defined by la carcinisation chez les Paguroidea, Galatheoidea, Hippoidea McLaughlin (1983b) and McLaughlin & Holthuis et Lomoidea, en insistant sur les Paguroidea. Une attention sp«- (1985), include four major taxa: Galatheoidea, ctale est accordee a la transformation - en thiorie - d'un habitus du type Pagure dans celui d'un crabe lithodide. L'evidence Hippoidea, Lomoidea, and Paguroidea. We rec- qui s'impose indique: (1) qu'une evolution vers un habitus de ognize Forest's (1987) elevation of the Paguroi- 80 P.A. McLaughlin & R. Lemaitre - Carcinization in the Anomura dea to Section Paguridea with superfamilies of the crab-like lithodids had been explored in Paguroidea and Coenobitoidea; however, hierar- detail eariier both by Boas (1880a, b) and Bou- chical ranking has no bearing upon our discus- vier (1894a-c; 189Sa, b; 1897). Additional exam- sion. We are aware of the propositions made re- ples of presumed paguroid carcinization were cently by Williamson (1992) and by Spears et al. reported by Bouvier (1896) and Borradaile (1916) (1993), but disputed by Scholtz & Richter (199S), in Ostraconotus A. Milne Edwards, 1880, and that the Dromioidea are actually anomunms. Sim- Tylaspis Henderson, 1885; in Probeebei Boone, ilarly, we note, but do not concur widi, the re- 1926b and Labidochirus Benedict, 1892 by Wolff introduction of "Anomala" by the latter audiors. (1961b) and Blackstone (1989); in Solitariopagu- Boiiadaile (1916), who first proposed the term rus TQrkay, 1986, by Turkay (1986); and in carcinization, implied a quite specific meaning, Pagurus hirsutiusculus (Dana, 18S1) by Black- i.e., "... one of die many attempts of Nature to stone (1989). As envisioned by most caicinolo- evolve a crab", and used for his example of car- gists, the evolutionary processes involved the cinization the hermit crab Porcellanopagurus sp.; modification of an anomuran having an inflated, StevQi® (1971), on the other hand, proposed the elongate cephalothorax into one that was broad- term brachyurization as a "more general term ened and somewhat flattened. Similarly, die long, relevant to all crabs as a whole, widiout regard to frequently articulated, and relatively weakly cal- shape and organizational level". Wolff (1961b), cified abdomen was converted into one that was Guinot (1979), and Turicay (1986) appear to have not only reduced, but also bent under the thorax used carcinization, as Williamson (1992) and and ultimately pressed closely against the ster- Martin & Abele (1988) have used brachyu- num. Such transformations required modifica- rization, in the sense of independent, convergent, tions not only to the thorax and abdomen them- evolutionary phenomena leading to a crab-like selves, but also to the entire animal. Morpholog- body form. In this review we use carcinization in ical diversity within die major anomuran taxa the sense of Wolff (1961b), Guinot (1979), and precludes any sweeping generalizations. Among Turkay (1986). The only decapods that have these taxa, carcinization would appear to have demonstrated these apparently evolutionary ten- been an independent phenomenon that proceeded dencies of carcinization are members of the at varying rates. Anomiuia and Brachyura. Presumably, in the Paguroidea carcinization would ultimately lead to Historical review of reported instances of the evolution of the lithodid body form, and in the carcinizatioB Galatheoidea, to the evolution of the porcellanid body form. The Hippoidea and Lomoidea would Paguroidea appear to have already reached a certain level of carcinization but with no distinct pathways evi- The most well-known of these theories are those dent. Our intent here is not to discuss anomuran of Boas (1880a, b, 1924), Bouvier (1894a, c, phylogenetic relationships in great depth, but 189Sb, 1897), Cunningham et al. (1992), and radier to investigate the evolutionary pathways Richter & Scholtz (1994) who perceived evolu- that siqqrasedly have given rise to the crab-like tion of Lithodes Latreille, 1806 (including Para- body forms of anomuran taxa, with particular lithodes Brandt, 1848) from a species of Pagurus emphasis on the Paguroidea. Fabricius, 1775 (Fig. 1), or another related genus. As previously indicated, the term carcinization Although earlier authors (e.g., H. Milne Edwards, was introduced by Borradaile (1916) to describe 1837; H. Mihie Edwards & Lucas, 1841; Brandt, the phenomenon he perceived to be the transfor- 1848. 1850, 1851) had proposed a relationship mation of a typical shell-dwelling hermit crab between Lithodes and Pagurus, Boas (1880a, b) into a crab-like animal no longer dependent upon was the first one who attempted to provide mor- a spiraled gastropod shell for protection of its phological data to substantiate such a relation- membranous abdomen. However, the phytogeny ship. These included similarities in the structure Contributions to Zoology, 67 (2) - 1997 81

Fig. I. Hypothetical tnnsformation of Pagurus Fabricius, I77S to Uthodes LatreiUe, 1806. To explain the asymmetiy of bennit cnbs, Piiez (1934) used a fictitious example of a person holding a package while descending on a dextrally coiled staiiwell similar to a snail shell Oeft). The person would find it more comfortable to cany the package on the left hand because there is more space on that side. The implication is that use of dextral shells by hermit crabs led to an abdominal asymmeuy fovoring Ae left side. Typical shell- dwelling hemut crabs such as Pagurus aleuticus (Benedict, 1892) (bottom, after Benedict, 1901), presumably underwent tiansfotma- ti

7^«

•A-W/

Fig. 2. Stages in the transfonnation of the abdomen from a pagurid to a lithodid as proposed by Bouvier (189Sb, 1897): a, pagurid precursor, b, Hapahgaster Bnuidt, 1850, dorsal (top) and ventral (bottom) views; c, Dermaturus Btandt, 1850, posterior (top) and dorsal (bottom) views; d, Neotithodes A. Milne Edwards & Bouvier, 1894b, posterior (top) and dorsal (bottom) views; e, Paralithodes Biandt, 1848; f, Uthodes Latreille, 1806, posterior (top) and dorsal (bottom) views; g, Lopholithodes Brandt, 1848, posterior (top) and dorsal (bottom) views; h. Paralomis White, 1856; i. Rhinolitodes Brandt, 1848; j, Crypiolithodes Brandt, 1848. Abbreviations: i, intercalary rods; 1, lateral plates; m, paired marginal plates; M, unpaired median plates. Numbers indicate abdominal somites (1-6) and telson (7). (Modified firom Bouvier, 1897.)

plates were homologous with the tergal plates of gation of the fourth pereiopods reflected a return the pagurid In Nematopagurus, and Pagunis, the to typical decapod structure. Before concluding tergal plates are joined to the sternal plates of the his study, Boas had the opportimity to examine somites by membranous strips, and Boas conclud- specimens of Pylopagurus, a genus with gonopod ed that it was from these that the small marginal bearing females, but one in which the males, like plates seen in Paralithodes had developed. Boas male lithodids, were not equipped with sexual reasoned that the loss of uropods in lithodids was tubes. His conclusion therefore became one in a direct result of loss of function; and that elon- which he evolved Uthodes from an ancestor 84 P.A. McLaughlin & R. Lemaitre - Carcinization in the Anomura

Fig. 3. Tnnsfonnadon of the pagurid abdomen to the liAodid as hypodiesized by Boas (1924): a, dissected abdomen of Nemato- pagurus A. Milne Edwaids & Bouvier, 1892, showing diffiaent parts; b, abdomen of ParaJUhodes Brandt, 1848. Abbreviations: a, acetabula; d, tngite; f. first pleopod; h, setose edge ("Haaigebrinie"); s, stemite. Numben denote abdominal somites and telson (modified fiom Boas, 1924).

closely related to Nematopaguns and Pylopagu- Pagurus. There is no doubt that a genetic rela- rus. tionship exists between pagurids and lithodids. More recently, Paralithodes was not only de- However, the study by Cunningham et al. (1992) rived from Pagurus (Fig. 4) but the proposition was based on data derived from only eight spe- was made by Cunningham et al. (1992) that P. cies from a gene pool in excess of 900 species. In camtschatica (Tilesius, 181S) and Lithodes aequi- view of the number of morphologically, and spina Benedict, 1895 were more closely related to seemingly phylogenetically, more closely related Pagurus bemhardus (Linnaeus, I7S8) and Pagu- intermediate taxa within ihc Paguroidea, the ap- rus acadianus (Benedict, 1901) than were the proach of Ctmningham et al. is in need of an latter two species to Pagurus pollicaris Say, 1817 expanded data base. and P. longicarpus Say, 1817. These authors The latest proposal for derivation of the Litho- went so far as to propose that the results from didae frt>m die Paguridae is that of Richter & their ribosomal RNA studies justified the inclu- Scholtz (1994). While not suggesting any direct sion of these two lithodid species in the genus line(s) of descent, these authors similarly con- Contributions to Zoology, 67 (2) - 1997 85

too Ptturut pctllctrlt (Ntw Cn9ljn0) Ptguna polllctrls («utr 01 ntaKil 100 r Ptgurut lengUtrput (New (nfiaM) PtgaruM Icnglcvpia 100^ is«iif MntiKtl -^ Ptgurui btmhtrtus • (C

ae CllOtiurlut vnutvs (Cult M nriKc)

(Cwif M n»K«i ArtnnI* stlMt

0.1 units

Fig. '<. Ev

eluded that the Lithodidae represent a group of nopagurus Filhol, 1885 (Fig. 5) as a phenomenon secondarily free living "asymmetrical" hermit "quite independent" of carcinization in other crabs. In Aeir study, they presented a series of anomurans. Borradaile interpreted the widening ei^t characters that separated their probably" of the stemite of the third maxillipedes seen in panq>hyletic "symmetrical" Pylochelidae ftom the Paguridae and Parapaguridae as the first step the monophyletic "asynunetrical" hermit crabs, in the carcinization process. Further advances which included Diogenidae + Coenobitidae, Para- included broadening of the cephalothorax and paguridae, and Paguridae-Lithodidae. Their study accompanying calcification, as depicted by Labi- addressed the issue of carcinization only in their dochirus splendescens (Owen, 1839) (Fig. 6a, b), conclusion that the lithodid characters they con- Ostraconotus (Fig. 6c), Tylaspis (Fig. 6d), Por- sidered reflected a change toward a free living cellanopagurus (Fig. 6e), and the semi-terrestrial habit. coconut crab, Birgus latro (Linnaeus, 1767) (Fig. Within the Paguridae, several other instances 6f). To emphasize his argiunent for independent of carcinization have been reported. Whereas pathways of carcinization in these taxa, Borradai- Bouvier (1894c, 1895b, 1897) and Boas (1880a, le (1916) noted a number of dissimilarities among b, 1924) placed primary emphasis on changes in them, e.g., male gonopod development, chelipede the abdomen and pleopods, Borradaile (1916) and pereiopod structure, development of a ros- emphasized changes in the cephalothorax. He trum, etc.; he gave only passing attention to the described carcinization in a species of Porcella- differences in reduction and/or calcification of 86 P.A. McLaughlin & R. Lemaitre - Carcinization in theAnomura

projections (Ip) of the typical pagurid shield (Fig. 7b). The first two of the three lateral wing-like expansions appear to be expansions of the lateral margin of the shield itself. The third, and most posterior lobe, which is cleariy delineated from the other two by the cervical groove (eg), is probably a calcified and expanded portion of the posterior can^Mice. The hard strip of cuticle present directly behind the midpoint of the posterior margin of die carapace was considered by Borradaile (1916: 114, fig. 13a) to be a portion of the last thoracic somite. Forest (195la) reported that the last thoracic tergite was represented by a narrow chitinous band, and we concur. As may be seen in Fig. 7d, a transverse calcified band is also appattax behind the posterior can^Mce margin. In our specimens there is a clear demarcation between the chitinous last dioracic segment (as determined by the positions of the acetabula of the fifth pereiopods) and the tergite containing the calcified strip, which we interpret to be a calcified portion Fig. 5. PorcelUm^gurus Filhol, 188S, "probably" P. edward- of the tergite of the first abdominal somite. Bor- si Filhol, 1885, as viewed by Bonadaik (1916): a, ovigeious radaile (1916) reported two tergal sclerites ftmale, donal view, b, same after removal of most eggs (end of present on the second abdominal somite and sin- fiiUi leg shown enlarged). Abbreviations: c, cervical groove; r, gle plates on the left third and fourth somites, rostnim. Numbers are "sidetobes" of cephalotborax. (After with only faint thickenings on the right. However, Borradaile, 1916.) both Forest (1951a) and we were able to distin- guish well-formed, entire (albeit not calcified), the abdomens. He set Porcellanopagurus apart by tergal plates on somites 2-5 (Fig. 7d) in bodi what he perceived to be the unique dorsal posi- males and females. A more significant point is die tion of the female pleopods, and began his dis- structure and placement of die female unpaired course with a detailed description of a female pleopods. Borradaile (1916) and Forest (1951a) attributed "probably" to Porcellanopagurus ed- both stated that the female pleopodal endopods wardsi Filhol, 188S. Forest (19Sla, b) redescribed are well developed and the exopods reduced. P. edwardsi and discussed its systematic position Borradaile (1916) also considered die pleopods to within die Paguridae, but did not address any be dorsal in position, with pl2 almost median and aspects of carcinization. the other two positioned successively more to the For the purpose of the current disoission, we left, indicating a position analogous to that of a present a brief morphological overview of Por- spirally twisted abdomen. One might well be cellanopagurus, based on our own examination of drawn to this conclusion if the abdomen were P. edwardsi. The cephalothorax (Fig. 7a, c) is partially contracted, as in many preserved speci- broadened, flattened and drawn out into a series mens. However, when the abdomen is preserved of lobes, the centermost being the prominent ros- in a relatively relaxed state, it is clear that the trum (r). Borradaile (1916) and Forest (1951a) pleopods (pl2-pl4) as well as the uropods sit in a identified the remaining lobes only by number; straight line on a symmetrical, albeit greatly however, we believe that the anterolaterai lobes swollen abdomen (Figs. 6e, 7d). In a small female represent extraordinarily well-developed lateral specimen with the pleopods aligned in the normal Contributions to Zoology, 67 (2) - 1997 87

Fig. 6. Examples of advanced cases of carcinization in hennit crabs according to Bonadaile (1916). Labidochina splendescens (Owen, 1839): a, whole animal; b, cephalothorax and cephalic appendages (after McLaughlin, 1974); c, Ostraconotus spatulipes A. Milne Edwards, 1880 (after A. Milne Edwards & Bouvier, 1893); d, Tylaspis anomala Henderson, 1888 (after Lemaitie, in press); e, Porcellanopagurus edwardsi Filhol, I88S (after Forest, 19Sla); f, Birgus latro (Linnaeus, I7S8) (after Alcock, 1905). Not to scale. 88 P.A. McLaughlin & R. Lemaitre - Carcinization in the Anomura

Fig. 7. Morphology of PorceUanopagunaedwardsi Filbol, 188S, (a, c, d); b: Pagurus Fabricius, 177S: a, donal view (peieiopods US not shown); b, left side of-typical pagurid shield, branchiostegite and left half of cephalodiorax as seen in Pagurus (after Pilgrim, 1973); c, right side of cephalothorax, lateral view; d, abdomen, dorsal view. Abbreviations: eg, cervical groove; Ip, lateral projection; It, linea transversalis; optp, outer ptetygostomial plate; r, rostrum. Numbers indicate lateral lobes of carapace (a, c) or abdominal somites (d); arrow points to calcified tod of abdominai somite 1. Not to scale. lateral position, it is the endopod that is reduced. is the exopod that has undergone reduction. When the pleopods have eggs attached, the pleo- Boone (1926a, b) described what she believed pods are twisted to allow the eggs to be carried to be a new and primitive macriiran representing dorsally; therefore the impression is given that it a new family of decapods. In his redescription of Contributions to Zoology. 67 (2) - 1997

Alaiiu^tagunis Abunop^uroides

Fig. 8. Classical exaiq|>les and recently described candidates of carcinized hennit ciabs: a, b, Probeebei mirabilis Boone, 1926b (after Wolff, l%lb); c, Solitariopagurusprofimdus TMcay, 1986 (after TQrkay, 1986); d, Alainopagurus crosnieri Lemaitre & McLau^in, 199S (modiiied from Lemaitre & McLaughlin, I99S); e,Alainopaguroides lemaitrei McLaughlin, 1997 (after McLaughlin, 1997). Not to scale.

Probeebei mirabilis Boone, 1926b, Wolff (1961a, within the group. Although he relied heavily on b) correcdy recognized this unusual, deep-water Borradaile's (1916) carcinization hypothesis, species as a paguroid (Fig. 8a, b), and he inter- WolfT (1961b) did address abdominal changes in preted it as another example of carcinization considerable detail. He noted that Labidochirus 90 P.A. McLaughlin & R. Lemaitre - Carcinization in the Anomura splendescens (as Pagurus) and Probeebei mira- pod shells of successively smaller size in popula- bilis shared the same carapace grooves and gen- tions from south to north. His findings led him to eral calcification of the cephalothorax. However, suggest that: (1) some hermit crabs possess intrin- the abdcHnen of L. splendescens, albeit reduced, sic characters that can lead to carcinization; (2) was membranous. The abdomen of P. mirabilis ecological conditions favoring carcinization may was more calcified than in any other paguroid; exist at polar, or at least northern, latitudes the development of the second abdominal tergite [northern populations have broader, often more was seen as more parallel to Lithodes. exposed can^Mices]; and (3) size increase [as seen The recently described genera, Solitariopa- in the more northern populations] may be associ- gurus (Rg. 8cX Alainopagurus (Lemaitre & ated widi reduced shell-living and carcinization. McLaughlin, 1995) (Fig. 8d), and Alainopagu- Provocative as his hypothesis is, the fitct that his roides McLaughlin (1997) (Fig. 8e) all bear cer- southern and northern populations actually repre- tain marked resemblances to Ostraconotus and sent distinct species [P. hirsutiusadus (Fig. 9a-c) Porcellanopagurus, and similarly might be con- and P. venturensis Coffin, 1957 (Fig. 9d-i) (cf sidered candidates for the tnuditional carcini- Grain & McLaughlin, 1994)] raises questions re- zation hypothesis. In feet, SoUtariopagurus was garding the validity of his conclusions. considered intermediate between the latter two genera by its author (Turicay, 1986), who regard- ed it as an another example of a hermit crab that Galatheoidea had abandoned its gastropod shell and had become at least partially caicinized. Only in SoU- Far less attention has been directed to carcini- tariopagurus and Porcellanopagurus is the very zation in the Galatheoidea, despite the presence well-calcified shield flattened and drawn out lat- in the group of a number of characters that in erally into three broad projections; however, in paguroids have been identified as attributes of all five genera the stemites of the second and carcinization. The body forms of galatheoids third thoracic segments are broadened. While (Fig. lOa-f) vary from the elongate lobster-like neaiiy complete calcification of the posterior car- Chirostylus Ortnumn, 1892 and Munida Leach, apace occurs in the monotypic Ostraconotus, only 1820, through the more "squat lobster" forms of varying degrees of weak calcification are seen in Uroptychus Henderson, 1888 and Munidopsis the other four genera. Species of all five genera Whiteaves, 1874, to the pseudo crab-like AegUi have reduced abdomens; males all lack pleopods Leach, 1820 and crab-like Petrolisthes Stinq>son, and females have pleopods on only somites two 1858. Frequently early carcinologists failed to duough four. The abdomens of Porcellanopa- recognize die relationships among these ttuee gurus, Ostraconotus, and Alainopaguroides are body forms (e.g., De Haan, 1850; Dana, 1852b). membranous and the second through fifth tergites Boas (1880b) and Henderson (1888) solidified are only faintly indicated. However, in SoUtario- the classification of the Galadieoidea. However, pagurus and the monotypic Alainopagurus these it was A. Milne Edwards & Bouvier (1894a) who tergites are chitinized or calcified to some extent. provided the first critical review of what might be The similariy mcmotypic Alainopaguroides ap- evolutionary processes within the superfamily. pears to be the only one to inhabit a gastropod Despite Bouvier's (1894a, b, c) concurrent work shell. to derive the crab-like lithodids fi-om the pagurids An interesting hypothesis of carcinization in- through a series of morphological analogies, A. volving Pagurus was proposed by Blackstone (1989). In what he reported to be southern and northern populations of the common, eastern Pa- Fig. 9. Pagunis hirsutiusculus (Dana, 1851) (a-c): a, animal in shell (after Schmitt, 1921); b, animal removed from shell; c, cific, intertidal hermit crab, Pagurus hirsutiuscu- telson (after Hart, 1982). Pagurus venturensis Coffin, 19S7 lus (Dana, 1851) (Fig. 9a, b), Blackstone docu- (d-i) (after Grain & McLaughlin, 1994): d, shield and cephalic mented the progressive broadening of the anterior appendages; e, right cbelipede; f, left chelipede; g, right second calcified carapace (shield), and the use of gastro- pereiopod; h, left third pereiopod; i, telson. Not to scale. Contributions to Zoology. 67 (2) - 1997 92 P.A. McLaughlin & R. Lemaitre - Carcinization in theAnomura

Milne Edwards & Bouvier (1894a) explained the pidea, Miers (1879) expressed the belief that die morphological variations seen in Uie galatheoids, hippoids (Fig. lOg-i) exhibited considerable re- for the most part, as adiqytations to habitat and semblances to the brachyuran oxystomes. Snod- depth. For example, the well-calcified cepha- grass (19S2) made the observation that "when an lothorax and abdomen of "primitive'* galatheids Emerita Scopoli, 1777 is stretched out at fidl were seen by Nfibie Edwards & Bouvier to sug- lengdi it is seen to be a conq>lete crab". Guinot gest a certain relation to macmran-lBce stock, as (1979) attributed the similarities of hippoid body had been proposed by earlier carcinologists (e.g., form and car^>ace with various burrowing Brach- H. Mibe Edwards, 1831). Reduction of the abdo- yura as another example of convergence. men, according to these authors presumably was brought about by loss of the original primitive function of swimming. The similarity of diese Lomoidea galatheoids to the symmetrical pylochelids of the Paguroidea suggested to A. Mihie Edwards & Lomis hirta (Fig. II) shares, with several other Bouvier (1894a) a possible ancestral relationship. anomuran families, mcHphological characters at- However, Aey speculated that the pagurids had tributable to carcinizaticm. In his original descrip- sought out shelters (crevices or shells) in which to tion, Lamarck (1818) placed it in die porcellanid hide their abdomens, dius rendering them soft and genus PorceUana Lamarck, 1801. The lack of asymmetric, whereas the Galadieidae (including ocular orbits, die cylindrical antennular pedun- at diat time the Porcellanidae) ad24>ted the posture cles, hirsute antenna! fh^Uum, and pediform of folding the abdomen against the stemiun. third maxillipedes were among the characters that Thus, A. Milne Edwards & Bouvier concluded convinced H. Milne Edwards (1837) to establish that both adaptive patterns led to reduction in the the genus Lomis for diis distinctive animal. Lomis size of the abdomen. Additionally, the porcel- was considered to be related to Lithodes by H. lanids' preference for shallow-water habitats and Milne Edwards and most subsequent zoologists. seclusion under stones accentuated the abdominal It was only when diis relationship was critically regression, bringing their body form in close examined by Bouvier (1894a, 189Sb) diat its dis- alignment with the Brachyura. Guinot (1979) spe- tinctiveness was recognized. However, Bouvier cifically noted die tendoacy of porcellanids to- was convinced that Lomis had evolved from an at ward carcinizati(m. least partially asymmetrical ancestor of pagurid The Aeglidae, because of their Recent exclu- lineage. Borradaile (1916) considered Lomis a sively freshwater habitat, were reviewed only prime exanq>le of carcinization. Pilgrim (1965), briefly by A. Mihie Edwards & Bouvier (1894a), McLaughlin (1983a), and Martin & Abele (1986) but have received considerable scrutiny in recent years by Martin & Abele (1986, 1988). These Fig. 10. Examples of body fonns of galatheoids (a, Chirosiylus four audiors agreed that the aeglids exhibited re- Octmann, 1892; b, Munida Leach, 1820; c, Uroptychus Hender- maricable similarities to pagurids, although none son, 1888; d, Aegfa Leach, 1820; e, Munitl^is Whiteaves, related them specifically to the litfaodids. The dis- 1874; f, PetroUsthes Stisipson, 1858), and hippoids (g, Albunea oovoy of Haumuriaegla glaessneri Feldmatm, WAer, 1795; h, Emerita Scopoli, 1777; i, Hippa Fabricius, 1984 from the Late Cretaceous of New Zealand 1787). Species shown: Chirostylus affinis Chace, 1942 (after Chace, 1942); Munida schrotderi CHiace. 1939 (after Chace, now places ancestral aeglids in the marine envi- 1942); Uroptychus fomicatus Chace. 1942 (after Chace, 1942); ronment. Munidopsis bermudezi Chace, 1939 (after Chace, 1942); Aegia platensis Schmitt, 1942 (top, side view, after Martin & Abele, 1988), and A. schmitti Hobbs, 1979 (bottom, donal view, after Hippoidea Hobbs, 1979); Petrolisthes amoenus (Guirin-Meneville, 1855) (after Gore, 1974); Albunea paretii GuMn-Mineville, 1853 (after Williams, 1965); Emerita benedicti Schmitt, 1935 (after We are not aware of any early authors that actu- WilUams, 1984); Hippa testudinarius (Heibst, 1791), dorsal ally accredit hippoid morphology to possible car- (left) and ventral (right) views [after Monod, 1956, as Hippa cinization; however, in his revision of the Hip- cubensis (De Saussure, 1857)]. Not to scale. Contributions to Zoology, 67 (2) - 1997 93

EmerU» 94 P.A. McLaughlin & R. Lemaitre - Carcinization in the Anomura

%^M^.

Fig. 11. Mor|A(dogy of lomu hirta (Lmuck. 1818). a, c, abdomen, veatnl view (after Pilgrim, 196S): a, male; c, fismale; b, wbole animal (female), donal view, details of left vpendages only (after McLaughlin, 1983a). Nottoicale. all reviewed the phylogenedc relationship of Lo- Boas' (1880a, b, 1924) hypotheses are based on mis with other anomuians. Pilgrim (196S) consid- the initial siq>position that carcinization proceed- ered Lomis as representative of a distinct £unily ed as a consequence of hermit crabs leaving their of primitive paguroids, most closely allied to gastropod shell-shelters and adopting fiee-living Mixtopagurus A. Milne Edwards, 1880, as "... an existences. Boiradaile's (1916) hypothesis fol- ofbhoot from the very base of the Paguridea to- lows similar lines; however, he concludes that not gether with the Pylochelidae, but has undergone all species have the genetic constituents regard- Uie process of carcinization - it is thus the earliest less of circumstance. anomuran to do so". McLaughlin (1983a) re- For any of these hypotheses to be tenable, first, moved Lomis to a separate superfamily (Lomoi- the assumption must be made that hermit crabs dea); whereas Martin & Abele (1986) related the initially had membranous abdomens that were Lomidae to the Lithodidae as a sister groiq> of the protected by use of a sinistral gastropod shell. Paguroidea. In more recent studies, Richter & However, what evidence is there to support this Scholtz (1994) also aligned Lomis with pagu- assunqrtion? Perhq>s that Aristotle found two roids, whereas Tudge (1997) suggested a basal q>ecies in gastropod shells, and remarked on the (»igin fon which we can all agree is that females. utilizing a shell, or some other covering protec- Boas's (1880a; 1924) derivation of Uihodes tive of the soft abdomen of most paguroids, con- from a Nematopagurus- or Pylopagurus-Wkt an- fers a survival advantage. How then could not cestor requires more catastrophic morphological owning a shell confer a greater advantage? What changes, although paired first abdominal female the traditional hypoAeses of carcinization require pleopods are common to both. Despite the facX is either diat a gastropod shell-dwelling hermit that Boas' (1880a) early "rags to riches" e]q>lana- crab Mrith a totally membranous abdomen for tion for the transformation of die Pagurus abdo- some inexplicit reason chose to forsake mobility men to the Lithodes abdomen was reinforced by with safety, or that an unexplainable dis^>pear- direct comparison of sternal and tergal develop- ance of shells compelled some hermit crabs to ment in Nematopagurus (Boas, 1924), he provid- ad^t to adverse conditions by becoming free liv- ed no substantive evidence diat such a qKmtane- ing. We suggest that there is no factual evidence ous transformation could actually occur. The gen- to st4>port the proposition Aat hermit crabs were eral morphological similarities he ascribed to first and foremost shell-dwelling crustaceans, but Pagurus, Nematopagurus and Lithodes ^>ply to rather that comments such as Bosc's, generated the majority of the Paguridae. Boas attributed frtnn casual observations of common littoral spe- development of the fourth pereiopod as a "return cies, became "fact" through repetition. to the ordinary condition of the Decapoda"; the If pagurid morphology is reexamined, Pagurus rostrum simply developed more strongly. Again, is the most simplistic of all paguroids, lacking reversibility in a conq>lex structure such as a most integumental calcification; having a very walking leg, is considered highly iiiq)robable reduced or virtually nonexistent rostrum; having (Rensch, 1959), although Richter & Scholtz no sexually modified ^)pendages other than the (1994) suggest that repetition of genetic informa- regular fismale egg-bearing pleopods, and having tion for size and spine pattern of the second and no penes (or sexual tubes), but sharing with most third pereiopods might have shifted posteriorly to other hermit crabs a specialized adjq>tation for be similariy expressed in the lithodid fourth housing stability, i.e., highly developed rasps on pereiopods. The considerable number of paguroid the uropods and frequently also on the propodi of taxa now known that have morphological struc- the reduced fourth and fiM pereiopods. For Bou- tures intermediate between a Nematopagurus' vier's (1894c, 189Sb, 1897) hypothesis to be Pylopagurus ancestor make Boas* gargantuan plausible, a Pagunis-]ik.e ancestor must have de- evolutionary transfoimati(Mi doubtfiil. veloped an expanded, well-calcified, and usually well-armed cephalothorax, created a series of calcified abdominal plates, generated a well-de- Evidence veloped rostrum, materialized a walking leg from a highly specialized fourth pereiopod, and devel- To find su;)porting evidence for the existing, or oped a pair of pleopods on the female's first ab- for one or more alternative hypotheses of car- dominal somite. Granted, in Bouvier's hypothe- cinization, we have compared adult morphologi- sis, abdominal plate calcification took place grad- cal characters in all the major anomuran taxa with ually, as did rostral development, but conversion a series of characters purported to be indices of of the specialized fourth pereiopod to a typical carcinological trends. Recent studies (e.g., Mor- walking leg would be much like "the return fix)m gan & Forest, 1991; Lemaitrc, 1993, 1994; Le- Equus to Hyracotherium ... a practical impossibil- maitre & McLaughlin, 1995; Poupin & McLaugh- 96 P.A. McLaughlin & R. Lemaitre - Carcinization in the Anomura lin, 1996; McLaughlin, 1997) have enabled us to general reptant characters to establish polarities, include characters from a number of paguroid whereas Martin & Abele (1986) used the charac- taxa not available to earlier caicinologists. Al- ter states found in die pelagic shrimp genus Pe- diough we have used exanq)les from galatheoids, naeus Fabricixis, 1798. Neither q>proach is appto- hippoids and Lomis, as well as paguroids, this is priate to the present investigation. In the first in- not meant to imply that a continuimi is to be stance, McLaughlin's (1983b) assessment was expected It is quite probable that if carcinization essentially intuitive, not a disciplined cladistic does ocaur, it does so through convergent path- analysis. Richter & Scholtz (1994) based dieir ways. out-groiq) conq>arison on information gathered from fht Galatfaeoidea, Hippoidea, Lomoidea, and other, unq>ecified reptant deciqxxls. Since we Characters are omisidering these tluiee siq>erfiuiiilies in our investigation, they represent conqxments of our Morj^logical attributes of carcinization (i.e., overall "in-group", thus they are not ^jpropriate leading to or producing a tnachyuran-like crab to serve as out-groups. Martin & Abele (1986) body form) have been compiled from Bourne used Penaeus as an out-groiq) **because the char- (1922X Steviic (1971), and Guinot (1979) and acters of the Dendrobranchiata are conceded by include 19 categories. Five additional attributes most workers to be primitive relative to odier of those proposed by Richter & Scholtz (1994) infiraorders ...**. This undoubtedly is trxie, but in have been examined. While our major morpho- reviewing Martin & Abele's discussion of charac- logical categories have been selected specifically ters, we were unable to find any syuq>omoTiês because of their presumed evolutionary role in shared by Penaeus widi die taxa under consider- carcinization, the pertinence of individual ele- ation that would qualify it as an out-group in our ments may not always be clearly evident. For search for evolutionary pathways of carciniza- example, paired first and second male, and paired tion. Rather we have chosen Neoglyphea inopi- first female pleopods would not appear particu- nata Forest & De Saint Laurent, 1975 as our out- larly relevant. However, the total array of charac- gro»q>. ters utilized is essential in allowing a clearer pic- Specific information on N. inopinata has been ture of evolutionary pathways to emerge. gleaJMd from Forest & De Saint Laurent (1981); Our assessment of attributes enq>loys cladistic inteipretation and choice of an out-group is based methods, but it must be pointed out that not only on the guidelines of Nixon & Carpenter (1993). is this only a preliminary and partial analysis, but That many of N. inopinata's moiphological at- also we are aware of the rather unusual ^plica- tributes are plesiomorphic was denKmstrated by tion of cladistics to this study. By examining the Poore (1994). However, N. inopinata is not a sis- possible evolutionary pathways of a set of adult ter group to die Anomura (cf. De Saint Laurent, moiphological characters, we are atten]|>ting to 1979), nor is it considered as "The" ancestor to ascertain whether or not carcinization in the the Anomura. Nevertheless, the synapomorphy, Anomura is an audientic phenomenon rather than reduction in the length of the fiftti pereiopods, determine now the precise phylogenetic relation- shared by N. inopinata and all anomurans siq>- ships among taxa. A thorough assessment of ports our choice of it as our out-group. Table I these pathways can only be made when larval and provides a summary of characters and states. juvenile characters are similarly addressed, as Table II furnishes, widi N. inopinata as the out- they will be in the second part of this review. group, a list of anomuran taxa examined and their For our present study, we have used the "out- respective character states. group" concept to root our analysis. In studies of In the discussion that follows, references are evolutionary relationships within and among made to various genera, but it must be pointed out anomunm taxa, McLaughlin (1983b) and Richter diat in many cases, particularly among galatheoid & Scholtz (1994) utilized other anomuran and and hippoid taxa, data have been obtained Contributions to Zoology. 67 (2) - 1997 97 through personal examinations of a limited is moderately deep, and has generally rounded number of individual species. When possible, carapace margins; (2) **vaulted*', in which the rel- these data have been siq>plemented with pub- atively straight lateral faces of the carq)ace and lished accounts of related taxa. Characters of branchiostegites provide the inq>ression of* an ar> paguroid genera reflect the overall sense of the ciform cara^nce; (3) "flattened", rqwesenting die genera as currently interpreted, although intra- state iêre there is relatively litde lateral depth generic variations do exist For example, the dio- to the cephalothorax. genid genus Paguristes Dana, 1851 is character- Within die remaining examined paguroid gtae- ized by the presence of paired first and second n^Pylocheles A. Mihie Edwards, 1880 and Chei- pleods modified as gonopods in males and roplatea Bate, 1888 among die Pylochelidae, paired, similariy modified, first plecqxKls in fe- Cancellus H. Mikie Edwards, 1836 among die males; however, we recognize that there are ex- Diogenidae, and Pylopagurus and Xylopagurus ceptions. Such incongruities reflect deficiencies A. Milne Edwards, 1880 among die Paguridae, in the alpha level taxonomy of the groiips but do still retain a somewhat subcylindrical carqiace not present any major obstacles to the interpreta- shape; however, it is most probable diat habitat tions of carcinizational pafliways. preferences have strongly influenced car^Mce shape in these genera, rather than retention of the plesiomorphic condition. All generally are petri- Characters and/or character states colous, xylicolous, spongicolous or scaphopod shell dwellers. As may be seen in Table n, the 1. Broadening and flattening of carapace, sternal evolution to a more '*crab-like** form is reflected plates, buccal frame and eventually also the in those taxa in which the canpace is globular or branchial chamber. Six characters are considered vaulted, i.e., Birgus Leach, 1815 and Coenobita in this category, most of which are generally self Latreille, 1829 in die Coenobitidae, Paguropsis explanatory. The exception is that of carapace Henderson, 1888 and Tisea Morgan & Forest, shape. The character states of carapace shape 1991 in the Diogenidae, the presumably "partially given in this account may differ significantly carcinized" Paguridae, Solitariopagurus, Porcel- from other published accounts. For example, lanopagurus, Ostraconotus, Alainopagurus, Alai- Martin & Abele (1988) referred to the carapace nopaguroides, Icelopagurus McLaughlin, 1997, of Aegla as being extremely depressed, giving the Labidochirus, and Lithopagurus Provenzano, animal an overall flattened appearance, vîereas 1968, as well as all of die Parqiaguridae and we have scored the Aegla caiapace as vaulted. Lithodidae. la. - Carapace shape. We have recognized Although commonly thou^t of as having sub- four character states. The primitive subcylindrical cylindrical carapaces, not all hippoids do exhibit canqMice is most closely approached in the hip- this character. Carapaces of two of the four gen- poid fiunily Hippidae and die similarly sand- era san^led had the more "crab-like" globular to dwelling Euceramus Stinq>son, 1860 of the gala- vaulted canqrace structure. Representatives of the dieoid family Porcellanidae. It is probable that the Galatheoidea are even more variable. In the por- buTowing habits of these animals may be as cellanids, the car^Mce is subcylindrical in Euce- much a contributing factor to their carapace ramus, but globular in Polyonyx Stimpson, 1858, shape as the possible retention of the primitive and flattened in Petrocheles Miers, 1876 and character state. Virtually all other anomurans ex- Petrolisthes Stimpson, 1858. Carapace shîe is hibit some more extensive degree of dorsoventral seen as globular in the chirostylid Gastroptychus conq)ression of the cephalothorax. We have de- CauUery, 1896, but vaulted in Uroptychus, fined three character states to depict the transition whereas in the galatheids the distinction was less away firom the primitive subcylindrical cephalo- definitive. Carapace shape in Galathea Fabricius, thorax: (I) "globular", in which the cephalotho- 1793, Munida and Munidopsis is neither globular, rax usually is not considerably longer than broad. nor vaulted, but rather intermediate. As previous- 98 P.A. McLaughlin & R. Lemaitre - Carcinization in the Anomura

Table I. Chancters for carcinization analysis. Paragraph numerals correspond to attributes of carcinization discussed in text (see Giancters and/or character states). Character state symbols are in parentheses. Out-grotq>" Neoglyphea inopiiuaa Forest & De Saint Laurent; data from Forest & De Saint Laurent (1981). Abbreviations: mxp, maxillipede; P, pereiopod.

\. Broadening and flattening of the carapace, sternal plates, buccal frame and eventually also the branchiai dumber: te. Carapace shqw: subcylindrical (0); globular (1); vaulted (2); flattened (3). Ibk Rostrum: elongate, broad basally and tq)eriiig to acute tip (0); elongate cylindrical or subcylindrical (1); long to moderately short, triangular or subtriangular (2); very short, bluntly triangular, rounded, obsolete, or absent (3). le. Lateral projections (postotbital spines): obsolete or absent (0); weakly developed (1); moderately well developed (2); elongate, prominent (3). Id. Stemite of the third maxillipedes: nanow, bases of corresponding appendages contiguous or neariy so (0); broad, bases of corresponding appendages separated (1). le. Stemite of chelipedes (fourth thoracic stemite): narrow, bases of corresponding appendages contiguous or neariy so (0); broadened, but maximum much less than half total carapace widtii, correq>onding appendages neither approximate nor very widely separated (1); very broad, maximum neariy half total carapace or greater, corresponding appendages very widely separated (2). If. Stemite of tiiird pereiopods (thoracic stemite 6): narrow, bases of corresponding appendages contiguous or neariy so (0); broadened, but much less than half total carapace widtii, corresponding qipendages neither qvproximate nor very widely separated (1); very broad, neariy half total carapace or greater, corresponding qipendages very widely separated (2). 2. Increased integumental calcification: 2a. Anterior carapace (shield): strongly caldiied (0); moderately well calcified throughout (1); with areas of very weak or no calcification (2). 2b. Posterior carapace: entirely calcified (0); posterolateral and/or posteromedial calcification (1); entirely membranous or with only scattered small areas of calcification (2). 2c. Branchiostegites: calcified (0); partially calcified (1); membranous (2). 3. Abdomen: 3a. Abdominal orientation: elongate, straight (0); elongate, twisted (1); reduced, straight or twisted (2); reduced, folded beneath cephalothorax (3). 3b. Abdominal segmentation: 6 deariy defined somites (0); 6 distinguishable, but not cleariy defined somites (1); fewer tiian 6 distinguishable somites (2). 3c. Tergite calcification: completely calcified (0); partially calcified (1); membranous or neariy so (2). 4. Uropods: Uropod structure: foliaceous, symmetrical (0); q)ecialized, symmetrical (1); specialized, asymmetrical (2); absent (3). 5. Female pleopods: Sa. First pleopods: paired and modified as simple "gonopods" (0); paired, unmodified (1); unpaired (2); absent (3). Sb. Second pleopods: paired (0); unpaired, right or left (1); unpaired left only (2); absent (3). 5c. Third and fourth pleopods: paired (0); unpaired, right or left (1); unpaired left only (2); absent (3). Sd Fifth pleopods: paired (0); unpaired right or left (1); unpaired left only (2); absent (3). 6. Posterior abdominal appendages of males: 6a. Third and fourth pleopods: paired (0); unpaired, right or left (1); unpaired left only (2); rudimentary or vestigial (3); absent (4). 6b. Fifth pleopods: paired (0); unpaired, right or left (1); unpaired left only (2); rudimentary or vestigial (3): absent (4). 9. Bronchia: 9a. Arthrobranch foranuU: rudiment (mxpl), 1 (mxp2), 2 (mxp3), 2(P1), 2(P2), 2(P3), 2(P4), 0(P5) (0); 0,0,2,2,2,2,2,0 (1); 0,0,2,2,2,2,2,1 (2); 0.0,1,2,2,2^0 (3); 0,0,0,2,2,2,2.0 (4). 9b. Pleurobranch formula: 0.0,0,0.1(P2), l(P3), 1(P4). 1(PS) (0); 0,0,0.0.0.0.1.1.2 (1); 0.0,0,0,1,1,1,0 (2); 0,0,0,0,0,1.1,0 (3); 0.0.0.0,0.0.0.1.1 (4); 0.0.0,0.0.0.1.0 (5); 0.0.0.0.0.0.0.1 (6); 0.0.0.0.0.0.0,0 (7). 9c. Gill structure: trichobtanchiate (0); intermediate (1); phyllobranchiate (2). 10. Antennal scale (acicle/exopod): Antennal acicle: elongate, triangular, marginally armed (0); well developed, cylindrical, or blade-shaped, unarmed (1); reduced (2); absent (3). 12. Expansion ofischia and meri of third maxillipedes to form a plate-like covering over the other mouthparts: 12a. Expansion of ischium and merus: no expansion (0); some degree of expansion to partially cover other mouthparts (1); broadly expanded to form plate-like covering over other mouthparts (2). 12b. Accessory tooth (teeth): with accessory tooth (0); without accessory tooth (I); absent (2). 13. Ocular orbit development with accompanying reduction in the size of the antennuies and antennae: 13a. Ocular orbits: no orbits developed (0); partial development of orbits (1); well-developed orbits (2). 13b. Antennular and antennal reduction: well developed, elongate (0); one or other reduced (1); both reduced (2). Contributions to Zoology, 67 (2) - 1997 99

Table I. Cont

14. Fusion of last thoracic somite: not fused to either abdomen or cephalothorax (0); fused to first abdominal somite (1); fused to cq>halothorax (2). 15. Position of dactyls ofchelipedes: simple (0); subchelate (1); chelate, fingers opening horizontally (2); chelate, fingers opening obliquely (3); chelate, fingers opening vertically (4). 17. DeveU^>ment of first two pairs ofpleopods in males: 17a. Pint pleopods: modified as complex copulatoiy appendages (0); present, not modified complex copulatory sUuctures (1); absent (2). 17b. Second pleopods: paired, natatory, with appendix masculina (0); paired, modified as copulatoiy stiuctures (1); unpaired, reduced (2); absent (3). 18. Development of male penes: Male gonopoies: paired, unmodified (0); one or both masked by tuft of pvoroinent setae (1); one or both with vas defereas produced as sexual tube (2).

Supplemental characters (§tevii£, 1971; WolfT, 1961b: Richter & Schohz, 1994): 23. Development Cffthe fourth pereiopod: 23a. Size and structure of third and fourth pereiopods: generally similar (0); «ii««iniil«r (i). 23b. Ventrolateral maigin of propodus of fourth peieiopod: unarmed (0); row of q>iiies (I); row of corneous scales or tubercles (2): multiple rows of coneous scales or tubercles (3). 23c. Fourth pereiopod termination: simple (0); subchelate (1); semichelate (2); chelate (3). 24. Position and structure of the fifth pereiopod: 24a. Fifth peieipod teimination: simple (0); subchelate (1); semichelate (2); chelate (3). 24b. Propodus of fifth pereiopod: rows of setae or denticles (0); rasp of corneous scales (1). ly noted, we consider the aeglid carapace to be projections, or post-antennal spines, of die cara- vaulted. The carapace of Lomis, like the gala- pace has not been fully investigated for any theids, is neither strictly globular nor strictly anomuran group. In some paguroid genera, devel- vaulted. opment of the lateral projections appears to be lb. - Rostnim. Among anomuran taxa there is correlated with carapace calcification, whereas in a great range of development. An elongate, basal- some of the galatheoids such as the porcellanids ly broad, and taptnng rostnun, such as seen in it would appear to correlate more directly with Neoglyphea Forest & De Saint Laurent, 1975 is the development of ocular oibits. Since lateral considered plesiomorphic, with reduction indica- projections and/or post-orbital q>ines are lacking tive of more advanced conditions. Retention of in Neoglyphea (presumably plesiomorphic), we diis type of rostrum in paguroids is seen only in have scored their development as apomotphic. Probeebei among the parapagurids and a few of Among paguroids, lateral projections are best the lithodid genera. It is reduced, but nonetheless developed in some of the lithodid genera, and piominoit in Birgus. Throughout the Paguridae obsolete or entirely lacking only in Cancellus, there is a clear tendency toward reduction in ros- Probeebei and Tylaspis. tral length and strength. Among hippoids, galatheoids and Lomis, later- The rostrum in hippoids is never very well al projections vary from total absence to relative- developed, varying from moderately short in ally strong development. buneids to very short in hippids. In contrast, the Id. - Stemite of the third maxillipedes. In the primitive elongate, triangular rostrum is seen in paguroid genera of the families Pylochelidae, Aegla, Munidopsis, Galathea, and Uroptychus, Coenobitidae, and Diogenidae we see the as- widi only moderately reduction in the other gala- sumed primitive condition of a very narrow ster- theid and chirostylid genera. However, the ros- nite, with the corresponding appendages basally trum is moderately short to obsolete in all of the approximate, seen. In genera of the Paguridae, porcellanid genera examined. In Lomis the ros- Parapaguridae, Lithodidae, all hippoids, and gala- trum is moderately short as well. theoids, as well as Lomis, the bases of the third Ic. - Lateral projections. The role of the lateral maxillipedes are separated. However, in Gastrop- 100 P.A. McLaughlin & R. Lemaitre - Carcinization in the Anomura tychus and Galathea, it is the articulating con- such as Ciliopagurus Forest, 1995, with species dyles that are separated; the basal segments of the that occupy the very slender whorls of cone maxillipedes themselves are nearly îproximate. shells, exhibit markedly broad sternal plates, le. - Stemite of chelipedes (founh thoracic which reflect microhabitat influence; however, stemite). While it might be expected that if the that phenomenon sinq>ly indicates the plasticity bases of the third maxillipedes were q)proximate, of hermits to ad^t to their environment. the bases of die following îpendage pair, the Only in die hijîd Blepharipoda Randall, chelipedes, would be similarly positioned or vice 1840 is a modest broadening of the fifth and sixth versa, and this is for the most part true in pagu- thoracic stemites seen, which might or might not roids; however, there are exceptions. In the dio- be associated with carapace shi^. In contrast, genid genera Clibanarius Dana, 18S2a and Tisea, Lomis and all galatheoids exhibit exceptionally the chelipedes are separated, while the maxilli- broad sternal plates regardless of carapace sh^>e. pedes are approximate. In contrast, in the parapagurids Sympagums Smidi, 1883 (sensu Lonai- 2. Increased integumental calcification. The three tie, 1989), Parapafftrus Smith, 1879, Btvahopa- characters considered in this category reflect, in gurus Lemaitre, 1993, and Tstmogaipagurus Os- part, the division of die cephalothorax seen in awa, 1995, and the pagurid genera Enneopagu- paguroids. Widi the exception of some lidiodids, rus MrT-anghlin, 1997, Pylopaguropsis Alcock, the paguroid cephalothorax is clearly delineated 1905, and Xylopagurus, the chelipede bases are into an anterior carapace (shield) separated from approximate even though the maxillipedes are the posterior carapace by the cervical groove and lnx>adly separated by the sternal plate. There does linea transversalis (\\) (Fig. 7a, b). In galatheoids, not ^pear to be any clear correlation between hippoids, and Lomis, the comparable anterior por- carapace shq>e and the separation of either third tion of the cephalothorax is termed the "anterior" maxillipedes or chelipedes. carapace. The lateral portion of the carapace (Fig. The stemite of the chelipedes is narrow in all 7c) that covers the gills (branchiostegite) is well of the hippoids examined, and the chelipedes defined in galatheoids, in some, but not all, correspondingly ^>proximate basally. Thus, it paguroids, but is very poorly defined in some would appear that the breadth of this fourth ster- hiiq>oids. We have considered only the degree of nite in hi{îds is not a function of carq>ace shape calcification of the portions of the carapace, not either the stemite is narrow whether die car^>ace sh^>e or spinatioiL is either subcylindrical or globular-vaulted. In all 2a. - Anterior carapace. Calcification of the galatheoid taxa examined, the fourth thoracic shield in paguroids varies among families and stemite is broad, even in porcellanid genera genera. In the Pylochelidae, the shield is usually where the can4)ace varies from subcylindrical very well calcified, as it is in coenobitids and dio- (Eticeramus) to flattened {Petrolisthes and Petro- genids such as Tisea, Paguropsis, and Cancellus. cheles). A similar broadening of this stemite is In the Parapaguridae, Probeebei, Tylaspis, and also seen in Lomis. Bivalvopagurus all have well-calcified shields, If. - Stemite of third pereiopods (thoracic ster- whereas species of the other genera are frequent- nite 6). In paguroids there does seem to be a re- ly characterized as having areas of reduced calci- lationship between canq)ace sh^>e and the extent fication. Lithodids all have well-calcified anterior of the broadening of the stemite of the third carapaces, while only a few genera of the Pagu- pereiopods among pagurids and lithodids. With ridae exhibit more than moderate calcification. few exceptions, those taxa having globular or We have fotmd only two other taxa among the truncate carapaces similarly had very broad sixth remaining anomurans that show any reduction in stemites. Moderate broadening was seen in Coe- calcification of die anterior carapace, Petrocheles nobita. That was not the case however, for pylo- and Gastroptychus, and in the case of the latter, chelids, diogenids or parapagurids. Some taxa this may be an artifact of preservation in the spec- Contributions to Zoology. 67 (2) - 1997 101 imens we examined. Previous authors (e.g., Baba, character 20, emphasized by Richter & Scholtz 1988) make no mention of reduced calcification (1994). in species of this genus. 3a. - Abdominal orientation. An elongate, 2b. - Posterior carapace. In contest to the cal- straight abdomen is characteristic of die paguroid cification of the shield or anterior carq>ace in family Pylochelidae, the pagurid genera Paguritta most paguroids, very few exhibit complete calci- Melin, 1939, Discorsopagurus McLaughlin, ficati(m of the posterior carapace, except for li- 1974, Orthopagurus Stevens, 1921, Eimeophyllus thodids. In the Paguridae, completion of calcifica- McLaughlin, 1997, Pylopagurus, Xylopagurus, ti(Mi is seen only in Ostraconotus, and nearly and a few individual species of other pagurid conq>lete in Labidochirus; all other genera have genera, as well as the pan^Migurid genus Tsuno- only areas of partial calcification, if at all. Among gaipagurus; most are inhabitants of tubular struc- the Diogenidae, only Tisea, and among the Para- tures, such as hollow pieces of wood, polychaete paguridae, otAy Probeebei and Tylaspis have com- tubes, sciq)hopod shells, etc. Exclusive of die pletely calcified posterior carapaces. The poste- Lithodidae and the coenobitid Birgus, elongate, rior carapace is well calcified in Birgus; there is but twisted abdomens are found in the vast major- only partial calcification in Coenobita. Litde, if ity of paguroid genera, reflecting dieir practice of any reduction in calcification is seen in the re- occtq)ying spiraled gastropod shells. As can been maining anomuran groups. seen in Table n, a few non-lithodid paguroids 2c. - Branchiostegites. Calcification of die have simply reduced abdomens; however, in branchiostegites, when it occurs in paguroids, is these, while there may be a tendency to fold the most commonly restricted to the outer pteiygosto- sixdi somite, uropods, and telson ventrally, the mial plate region (optp) (Fig. 7b, c). Exceptions abdomen itself is not similarly flexed. In contrast, include Placetron Schalfeew, 1892 among die the lithodid abdomen and that of Birgus are not Hiq;>alogastrinae lidiodids and all of the Litho- only reduced, but distinctly folded under the dinae genera, die coenobitid Birgus, and possibly cephalothorax. the diogenid Tisea. In all other anomurans, there is some degree of Some degree of decalcification has been ob- reduction and explicit folding of the abdomen. served in the posterior and/or ventral portions of Among hippoids this is maximized in members of the branchiostegites in the hippoids Lophomastix the Hippidae where the elongate telson is firmly Benedict, 1904, Albunea Weber, I79S, and Ble- pressed against the thorax. In the galatheoids, pharipoda, as well as in die porcellanids Petro- optimum flexure is seen in the porcellanids. Lo- listhes and Euceramus, and in Lomis. mis too has Ac abdomen folded under the cephalothorax. 3. Abdomen. The three characters examined in 3b. - Abdominal segmentation. Varying de- diis category pertain only to the general shape grees of visible segmentation of the abdomen can and orientation of the abdomen, delineation of the be found among paguroid families and genera. In abdominal somites, and the extent to which these the Pylochelidae, all six somites are clearly de- have changed fix>m the primitive elongate, fined as diey are in the Coenobitidae. Only in the straight, well-calcified, six-segmented abdomen diogenid Tisea, the pounds Solitariopagurus and of Neogfyphea. Internal symmetry is not ad- Alainopagunis, and the parapagurid Probeebei dressed, as it does not appear relevant to carci- can this presumed plesiomorphic condition be nization; however, asymmetry as it pertains to the clearly observed. In certain other diogenids, e.g., abdominal appendages is considered in subse- Paguristes, in virtually all other parapagurids, quent sections. Similarly, the abdominal position- and pagurids such as Porcellanopagurus, Ostra- ing of organs, such as seen in paguroids other conotus, Pylopaguropsis, and Xylopagurus, six than lithodids, is not addressed in this analysis. somites often can be distinguished as slight to Character 3c (abdominal tergites 1-S) embodies appreciable integumental thickenings or at least 102 P.A. McLaughlin & R. Lemaitre - Carcinization in the Anomura identifiable transverse segmental bands of fibrils. burrowing, but are not provided with rasps, Altfaou^ five clearly delineated segments are whereas all galatheoids retain plesiomorphic fo- common at least to lithodids of the Lithodinae, liaceous un^xxis. The uropods of females of Lo- the first is usually unrecognizable as a distinct mis similarly are foliaceous, while those of the somite because of fusion. males are vestigial. Six well-defined abdominal somites are characteristic of the Hiiq>oidea, Galatheoidea, and S. Female pleopods. We have divided the devel- Lomis; however, there often is appreciable reduc- c^mient of female pleopods into four separate tion in the size of the first somite. characters, as considerable diversity occurs in the 3c. - Tergite calcification. Again, it is the Paguroidea. Pleopods on the first abdominal Paguroidea that manifests dramatic calcification somite of females in the Anomura occur only in loss. As the data in Table II demonstrate, the the adult, and commonly are referred to as gono- primitive condition of strong tergal calcification pods, aldiough no sexually related fimction has is seen in members of the Pylochelidae, the coe- ever clearly been demonstrated for them. Second nobitid Birgus, die parapagurid Probeebei, and in pleopods, aldiough usually ui^Mured, are symmet- some genera of the Lithodidae. According to rically or asymmetrically paired in some taxa, or Lemaitre (1995) die strong calcification of the rarely are absent entirely. sixth teigite seen in such pagurid genera as Xy- Throughout die anomuran genera we exam- lopagurus and Discorsopagurus is clearly an ad- ined, the third and fourth pleopods were present, aptation for tube closing. Partial calcification is either as paired or impaired appendages, specifi- more common among species of the remaining cally adi^ted for egg carrying. Development of pan^gurid genera, die h^>alogastrinid lithodids fifdi pleopods was variable. and the pagurids Solitariopagurus, Alainopa- Sa. - First pleopods. In Neoglyphea inopinata gurus, and Porcellanopagurus. In the vast major- the paired first pleopods of the female are two- ity of diogenids and pagurids, the sixth tergite is segmented, moderately long appendages. The dis- weakly calcified, while the remainder are mem- tal segment is flattened, flagellifonn, with the branous. distal portion more or less divided into distinct In contrast, there is litde or no substantive re- articles. According to Forest & De Saint Laurent duction of calcification in die first five abdominal (1989), these qipendages are provided with a tergites in the galatheoids, hippoids, or Lomis. relatively precise arrangement of sinqile, plumose and barbed setae. Paired first ple<^x)ds are report- 4. Uropods. We have considered the morpholog- ed for all genera of the Pylochelidae; those of ical structure of the uropods only in very general Mixtopagurus showing annulations (Forest, 1987: terms. For example, die primitive state, "folia- fig. 7Se) reminiscent of N. inopinata. Within the ceous" refers to a biramous appendage in which Diogenidae, only in die genera Paguropsis and the rami are broad, flattened, and marginally se- Paguristes is the devel<^ment of female first ple- tose, in contrast to more qiecialized appendages opods retained, although this condition has been that are either symmetrical or asymmetrical. In lost in some species of the latter genus. When paguroids this specialization takes the form of present, these paired first pleopods are commonly rasps of corneous spines or scales. Contrary to the provided with marginal sinqile or pltimose setae, rather sweeping statement made by Martin & but appear to lack the complexity seen in N. in- Abele (1986), lithodids are not the only paguroids opinata or the Pylochelidae. Among the Paguri- to lack uropodal rasps. For example, although the dae, paired first pleopods are developed in sever- uropods of the diogenid Paguropsis are, in struc- al genera, including Nematopagurus and all of the ture, typical of odier genera in the family, no Pylopagurus-\ike genera (cf McLaughlin, 1981). rasps are present. Uropods are present but they However, among those genera considered poten- are very gready reduced in Birgus. tial candidates for a paguroid carcinization path- The uropods of hippoids are specialized for way, (Hily in Alainopaguroides are paired first Contributions to Zoology. 67 (2) - 1997 103 pleopods developed. Paired first pleopods are Ostraconotus, Alainopaguroides, Lithopagurus, lacking in all coenobitids. and Xylopagurus. Unpaired left fifth pleopods Lomis was the only non-paguroid anomuran in were present in all of the lithodids we examined. ^f^ch we found paired first pleopods developed, Fiftti pleopods are absent in females of Hippa but diese are rudimentary. Fabricius, 1787, but present in other hippoids. Sb. - Second pleopods. Among paguroids, fe- Lomis and most galatheoids have paired fifth ple- males of all pylochelids and parapagurids of the opods, although we found diem lacking in Uro- genera Parapagurus and Bivalvopagurus have ptychus pubescens Faxon, 1893. paired pleopods on the second abdominal somite; ttieir occurrence is variable in species of Sym- 6. Posterior abdominal appendages of males. The pagurus (sensu Lemaitre, 1989) and the right sec- occurrence, in males, of paired or luqiaired pleo- ond is well developed or rudimentary in Tylaspis. pods developed on the diird through fifth abdom- UiqMured second pleopods can occur on either the inal somites is variable, but when present, these right or left side of the abdomen in the diogenid appendages are not modified as gonods. These are symmetrical many of the galatheoids, and Lomis. They are in the pylochelids, but strongly asymmetrical in absent in all porcellanids and some chirostyiids, the parapagurids. Unpaired left tfiird and fourth and reduced in some ^)ecies of Munida and pleopods are the rule for the other genera of the Galathea. Parapaguridae, except Probeebei where they are Sc. - Third and fourth pleopods. Female pagu- vestigial or absent. These are lacking in Birgus, roid third and foiuth pleopods are paired in the and reduced or absent in coenobitids. Similarly, Pylochelidae and the parapagurids Bivalvopagu- pleopods are completely lacking in the diogenid rus and Tylaspis, aldiough they become rudimen- genera Paguropsis and Cancellus, unknown for tary with increasing animal size (Lemaitre, in Tisea, but uiq>aired oa die left in the remaining press). In all other paguroids, including the li- genera. Loss of third and fourdi pleopods is char- thodids, unpaired pleopods occur on the left side acteristic of all lithodids and several pagurid gen- only, excqrt in the diogenid genera Paguropsis era and at least one species of Pagurus (see Table and Cancellus, yihett uiq>aired pleopods can oc- II). cur on either side of the body. 6b. - Fifth pleopods. Among male paguroids, Paired third and fourth pleopods occurred in all paired fifth pleopods are found only in the Pylo- hippoids and galatheoids we examined, as well as chelidae and the parapagurids Bivalvopagurus in Lomis. and Tylaspis. Most commonly an unpaired left Sd. - Fifth pleopods. As with the third and appendage is present, often somewhat smaller in fourth pleopods, we found paired female fifth size than the preceding two. In genera such as pleopods only in the paguroid family Pyloche- Porcellanopagurus, Solitariopagurus, etc., where lidae and the parîagurids Bivalvopagurus and the third and fourth pleopods are absent, the fifth Tylaspis. Unpaired fifth pleopods occur on either also is missing; however, this pleopod is occa- the right or the left side in species of the diogenid sionally missing in pagurids that retain the third genus Cancellus, but on the left side only in all and fourth. The fifth, like the fourth and third, are other diogenid genera except Paguropsis, where absent in all lithodids. The fifth is always absent they are absent. Loss of the fifth pleopod is uni- in coenobitids. versal in the coenobitids, and not uncommon in Pleopod reduction and/or loss is common in the Paguridae, being observed in Solitariopa- male hippoids, porcellanids, and numerous other gurus, Alainopagurus, Porcellanopagurus, galatheoids. They are vestigial in Lomis. 104 P.A. McLaughlin & R. Lemaitre - Carcinization in the Anomura

7. Union of pterygostomial region of the cara- whereas Lomis and all of the galatheoids we ex- pace with the epistome (not included in Tables I amined retain a full complement of pleuro- and II). Such complete fusion is rarely, if ever, branchs. approached in anomuran taxa. 9c. - Gill structure. Although dendrobranchiate gills are considered to reflect the primitive condi- & Adherence of the branchiostegites to the tho- tion, these are thought to have given rise to the racic epimeres (not included in Tables I and 11). trichobranch gills of reptants (Martin & Abele, Again, rarely approached in anomurans. 1986). Therefore the primitive gill structure is trichobranchiate for our analysis, and this is the 9. Bronchia. We have used arthrobranch and gill structure of Neogfyphea inopinata (Forest &. pleurobranch formulae as they have been report- De Saint Laurent, 1981) and all pylochelids (For- ed for various anomuran taxa as separate charac- est, 1987). Of die diogenid genera examined, in- ters, but have not included epipods since their termediate gills woe found in Paguropsis. Gills occurrence is restricted to some galatheoids and in q)ecies of Paguristes varied from tricho- pylochelids. Gill type is the third character con- branchiate to phyllobranchiate, while the latter sidered. were common to Clibanarius and Tisea. Phyllo- 9a. - Arthrobranchia. Neariy all paguroids are branchiate gills are typical of the coenobitids and {Hovided widi five pairs of arduobranchs on each the majority of pagurid genera, however, there are genera whose species have either trichobran- side of the body, one pair on the artibrodial mem- chiate or intermediate gill types. As far as we brane of the third maxillipede and one on each of know, all lidiodids have phyllobranchiate gills. the first four dioracic q)pendages. There are, A few hippoids have trichobranchiate gills, however, three genera of the Paguridae in which others are phyllobranchiate, as are all galatheoids loss of arthrobranchs has occurred, Enneobran- observed. The gills of Lomis are trichobranchiate. chus- Garcia Gdmez, 1988, Enneopagurus, and Enneophyllus. Five pairs of aitiirotemchs are 10. Antennal scale (acicle/exopod). Reduction in present in Birgus and Coenobita, but those of the the length and armature of the anteimal acicle is third maxillipede and chelipedes are rudimentary considered evidence of carcinization; however, and presumably non-fimctional in species of the major reduction in the antennal acicle is observed latter genus. in relatively few paguroids. It is shortened in Five pairs of arthrobranchs on each side of the Birgus and Coenobita, as well as in diogenid body is also characteristic of all the hippoids and genera such as Cancellus and Clibanarius, and in galatheoids examined, as well as Lomis. pagurids such as Solitariopagurus, Porcellano- 9b. - Pleurobranchia. The primitive pleuro- pagurus, and Xylopagurus. Although spination, branch condition is represented by a single gill on including aciclar develc^ment, is strongest in ju- the body wall between the aithiobranchs of perei- venile lithodids, the most significant reductions opods two through five. Among paguroids, this we observed were in adults of Paralithodes and condition is seen in the Pylochelidae, Birgus, and Neolithodes. several genera of the Diogenidae. Generally, Appreciable reduction in the antennal acicle when gill loss occurs, it is loss of the pleuro- was noted in some of the hippoid genera, the branchs, beginning with those of the second and majority of galatheoids, and Lomis (Table II). third pereiopods. Most panqugurids have lost all bxit Ate pleurobranch of the fourth pereiopod; 11. Presence of a sella turcica posterior (not in- however, some species of Sympagurus (sensu cluded in Tables I and II). Although cited by Lemaitre, 1989) retain a remnant of a gill on the Bourne (1922) as one of the characters of carci- fifth pereiopod as well. Only in some genera of nization, a sella turcica is not present in all brach- the Paguridae are pleurobranchs lost entirely. yurans and, to our knowledge has not been iden- Pleurobranch loss in the hippoids is variable. tified in any anomurans. Contributions to Zoology, 67 (2) - 1997 105

12. EjqKmsion of the ischia and meri of third or two teeth. The litfaodid Placetron has no acces- meudUipedes to form a plate-like covering over sory tooth on the crista dentata. the other mouthparts. There is considerable vari- b hippoids we found the crista dentata present, ation in the develqnnent of the endopod and ex- but lacldng an accessory toodi in Lophomastrix (q>od of diird maxillipedes, as well as in setal and Blepharipoda; it is totally absent in die exam- types; however, only the e;q>ansion of the is- ined q)ecies of Hippa and Albunea. Among the cUum and menis has been correlated to carci- Galatheoidea we reviewed, the primitive state nization. We have added as a second character, was observed only in Aegla. The crista dentata the presence on the ischium of the tooth-like was absent in all porcellanids and in at least ansion of ischium and merus. No orbits in paguroids, if such occurs, is very slight, e^qiansion of the ischium and merus of the third and limited to a few litfaodid taxa (Table U). maxillipedes was observed in any of the paguroid Among the otfier anomurans, only die Gala- genera examined. theoidea have weakly or well-developed ocular Minor expansion of these segments was ob- orbits, most specifically in some species of the served in Hippa, but only in the porcellanids was Porcellanidae. any notewortiiy expansion seen. 13b. - Antennular and antennal reduction. No I2b.-Accessory tooth (teeth). Among the Pylo- ai^nnciable reduction in antennular or anteimal chelidae, Mixtopagurus was the only one of three size was observed in any paguroid. However, in genera we observed that retained the accessory Birgus and Coenobita the antennules are strongly tooth; however, Forest (1987) noted its presence modified to function in die aerial environment in Pomatocheles Miers, 1876 and reported two Reduction in size, and nuniber of segments, teeth in Trixocheles Forest, 1987 and several in particulariy in the antenna, was indicated only in Cancellocheles Forest, 1987. We know of no galatheoids, notably in porcellanids. coenobitid, diogenid or parapagurid genus in which the crista dentata is provided with an ac- 14. Fusion of last thoracic somite. Poore (1994) cessory tooth, although the cristate structure itself rated Neoglyphea iîomorphic in die fusion of the is usually well developed. In the Paguridae, the last thoracic somite, as in this taxon it is not fused situation is different. Richter & Scholtz (1994) to the seventh thoracic somite as it is in two of his were unsure whether one or more teeth should be other out-groups, Enoplometopus A. Milne Ed- considered plesiomorphic or apomoipbic, where- wards, 1862 (sometimes considered an axiid) and as we have rated its occurrence a plesiomorphic Nephropsis stewarti Wood-Mason, 1873 (a character. While species of the majority of pagu- nephropid lobster). Since there is neither fusion rid genera have a well-developed crista dentata between the seventh and eight thoracic sterna in and one or more accessory teeth, in a few (e.g., Neoglyphea nor in many of the Anomura, a lack Ostraconotus, Enneopagurus, Iridopagurus De of fusion is considered, for this analysis, the Saint Laurent-Dechance, 1966, and Decaphyllus plesiomorphic state. De Saint Laurent, 1968) the accessory tooth is No fusion of the last thoracic somite was a^ absent. Additionally, in a few genera the crista parent in pylochelids, coenobitids, or diogenids. dentata itself is reduced, occasionally to only one In parapagurids there appeared to be a tendency 106 P.A. McLaughlin & R. Lemaitre - Carcinization in theAnomura toward sternal fusion of the last thoracic and the pods modified as gonopods are present in all first abdominal somites in species of most genera, pylochelids (Forest, 1987), in the diogenid genera incltiding Sympagurus dimorphus Studer, 1883, a Paguropsis and Paguristes, and in most para- species scored by Richter & Scholtz (1994) as not pagurids. The occurrence of paired first pleopods fiised. Althou^ fusion between the stemites of in die Paguridae is rare (e.g., Lithopagurus, Xy- the last thoracic and first abdominal somites is lopagurus), and has not been reported for any common among pagurids, it is not complete in lithodid or coenobitid. some genera, e.g., Labidochirus, Phimochirus Among the other anomurans, paired copulatory McLaughlin, 1981, and Xylopagurus. In the litho- first pleopods occur in some galatheoids and Lo- did genus Cryptolithodes there appears to be fu- mis (Table II). sion of the last thoracic somite with both the first 17b. - Second pleopods. No natatory male sec- abdominal and the preceding thoracic somites. ond pleopods are known in anomurans. In the No fusion of these somites was observed in any pylochelids and some pan^Mgurids, these ^>- of the hippoids, galatheoids, or Lomis. pendages are modified as gonopods, as they are in Paguropsis, Paguristes, toad Strigopagurus For- 15. Position of dactyls of chelipedes. The brach- est, 1995, but unpaired and unmodified in coeno- yuran position of die dactyls perhaps has phy- bitids. With few exceptions, second pleqHxls of logenetic significance; however, this character pagurids occur only on the left side of the abdo- presumes a chelate appendage. In N. inopinata men, if diey are present at all. Although second die first pereiopods have simple dactyls. Some of pleopods are reportedly absent in all lithodids, we the hippids have subchelate first pereiopods. In have found a vestigial left i^q)endage in Rhino- lithodes. paguroids, all first pereiopods are chelate, al- Second pleopods are absent in hippoids. though the manner in which diey are carried may Among galatheoids, only in Aegla is the second vary within any given genus. It is doubtfiil that in pair reduced or absent (Martin & Abele, 1988). the Anomura this character contributes to any Lomis also has paired second pleopods modified meaningful evaluation of the carcinization hy- as gonopods. pothesis. 18. Development of male penes. No development 16. Females -with a receptacultan seminis (not in- of penes is reported for Neoglyphea; therefore we cluded in Tables I and 11). No female anomurans consider die presence of paired, unmodified go- that we know of have a receptacle for spcsm stor- nopores O^cldng extensions of the vas deferens) age. the primitive conditicm. Sexual tubes arising fix>m die goncqiores as 17. Development of first two pairs ofpleopods in extension of the vas deferens are reported in a males. The presence, in N. inopinata, of paired number of paguroid genera, members of the Hip- first male pleopods specialized as gonopods is poidea, and in the galatheoid Aegla (Table 11). On taken to indicate the plesiomorphic condition. the basis of sexual tube development in the Pagu- Reduction and/or loss of one or both members of ridae, we have rated the presence of distinct tufts the pair is apomorphic. Paired, but natatory sec- of setae masking the gonopores as an intermedi- ond pleopods are found in N. inopinata (Forest & ate condition, such as seen in Pagwixus. While De Saint Laurent, 1981). The specialization of no true sexual tubes are developed in Birgus, these pleopods followed by reduction and ulti- there are slight protrusions of the vas deferens mate loss are considered apomorphic states. Our from each gonopore in preserved specimens. At view of polarity in these male pleopods is contra- least in some species of Coenobita, the vas defe- dictory to the interpretation of Martin & Abele rens is produced into a well-calcified sexual tube. (1986). 17a. - First pleopods. Paired male first pleo- 19. Inhalant opening lies primarily before the Contributions to Zoology. 67 (2) - 1997 107 base of the chelipede (not included in Tables I (1997), a distinction is made between subchelate and n). A rare, and most probably analogous sit- fourth pereiopods and semichelate fourth pereio- uation occurs in some anomurans. pods. In the former, the pereiopod is developed as a prehensile structure by Ae folding back of the 20. Calcification oftergites of abdominal somites dactyl against the propodus (e.g.. Fig. 12e, f). In 2-5 (see 3c, not included in Tables I and II). the latter, the ventral margin of the propodus is produced beneath the dactyl to such an extent that 21. Fusira of basal segments of upper antennular flexion of the dactyl becomes much more aldn to flageUum (not included in Tables I and II). Rich- the action of a dactyl against a fixed finger of a ter & Scholtz (1994) suggested that the basal ar- chelate appendage (e.g.. Fig. 12i-k). A cbelate ticles (segments) of die outer antenniilar fla- fourth pereiopod has the propodus developed into geUum fuse in a characteristic manner in the a characteristic *iixed finger", widi which the Paguridae and Lithodidae. However, they exam- dactyl fimctions as a true chela. ined only six species. They observed similar fu- 23a. - Size and structure of third and fourth sion in two of six diogenid species inspected. pereiopods. As in A/', inopinata, the fouitii pereio- Although we made no q)eciflc effort to determine pod is elongate and structurally similar to the the degree of fusion in antennular articles in spe- third pereiopod in litbodids, galatheoids, hip- cies of the numerous genera we examined, casual poids, and Lomis. The fourth pereiopod in all observation suggests that, while this character these taxa (Fig. 12a, c, d) functions as a walking may be significant in some phylogenetic applica- leg, with a simple terminal dactyl, except in the tions, it does not ^pear to be directly related to hippoids (Fig. 12b) where it is specialized, like carcinization. the third, for burrowing. In Birgus, the fourth pereiopod is approximately one-half length of die 22. Displacement of antennular aesthetascs (not third, armed with scattered small spines or included in Tables I and II). Richter & Scholtz (1994) similarly proposed that the positioning of spinules, but not ambulatory. In all non-lithodid the rows of antennular aesthetascs might demon- paguroids, the fourth pereiopod is reduced and strate a phylogenetic relationship among the differs substantially from the third (Fig. 12e-o). Paguridae and Lithodidae. Here iâin, their sam- 23b. - Ventrolateral margin of propodus of ple size was too small to permit meaningful eval- fourth pereiopod. The ventrolateral margin of the uation, but it is doubtful that this character has propodus in lidiodids and galatheoids may <»: may any direct q>plication to the present study. not be provided with row(s) of înes. This margin is ^so armed in Lomis, but typically unarmed 23. Development of the fourth pereiopod. Of the in hippoids. In many, but not all non-lithodid duee characters considered, two require clarifica- paguroids, this propodal margin may be ^lecial- tion, i.e., the ventrolateral margin of the propodus ized to form a propodal rasp of one to several (23b) and the termination of the appendage (23c). rows of corneous înes or scales, often extend- The primitive condition seen in Neoglyphea is a ing onto the lateral face (Figs. 12h-k, 13a-e). simple ambulatory leg with an unarmMi propodal However, it is unarmed in Birgus, in the diogenid ventral margin. However, even when the fourth Paguropsis, in the parapi^^rids Probeebei (Fig. pereiopod functions as an ambulatory leg, the 13f) and Tylaspis (Fig. 13g), and in die pagurid propodal margin may be armed with spines or Ostraconotus (Fig. 12e). The rasp of Alainopagu- tubercles. roides consists of two or three small corneous A simple fourth pereiopod indicates that the scales or spines (Fig. 12g). dactyl functions as a typical terminal leg segment, 23c. - Fourth pereiopod termination. The analogous to the condition seen in the more ante- fourth pereiopod terminates simply in all litho- rior ambulatory legs. As recommended by dids, galatheoids, hippoids, and Lomis. Martin & McLaughlin (1997) and Sandberg & McLaughlin Abele's (19S6) general statement that in all her- 108 P.A. McLaughlin & R. Lemaitre - Carcinization in the Anomura

Fig. 12. Fourth pociopods of various anomumis: a, Lithodes CMifipukns Macphenon, 1988 (after Macpherton, 1988); b, Emertta talpoida (Say. 1817); c, Galathea consobrina De Man, 1902 (after Bai>a, 1988); d. Petrolisthes heterockrous Kropp, 1986 (after Kropp, 1986); e, Ostraconotus spatuUpes A. Milne Edwards, 1880 (after A. Milne Edwards &. Bouvier, 1893); f, SoUtariopagunis proftindus TOrkay, 1986 (after TOrkay, 1986); g, Alamopaguroides Umaitrei McLaughlin, 1997, propodus and dactyl (after Mcljiughtiii, 1997); h, PoreeUam^gurus ip^ i, Tomopaguna wassi McLaughlin, 1981 (after McLauîlin, 1981); j, Paguna trigoHocheinis (Stimpson, 18S8) (after McLau^din, 1974); k, Xylopagurus tayrona Lemaitre & Cantos, 1993 (after Lemaitre & Campos, 1993); 1, MtPiidopagums macrocheles (A. Milne Edwards, 1880) (after Proveozano, 1971); m, AUdnopagurus crosnieri Lemaitre & McLaufîlin, 1995 (after Lemaitre & McLauîlin, 199S); n, o, DecaphyUus banmajaya McLaughlin, 1997, left (n) and right (o) propodus and dactyl (after Mclaughlin. 1997). Not to scale. mit crab families the fourth leg is subchelate, al- roids, it is also simple in the pagurid genera beit barely so in some, is not entirely accurate, Munidopagurus A. Milne Edwards & Bouvier, even when the tenn semichelate is substituted for 1893 (Fig. 12 1), Alainopagurus (Fig. 12m), and their "subchelate". Among non-lithodid ps^- DecaphyUus (Fig. 12n). It is truly subchelate in Contributions to Zoology. 67 (2) -1997 109

Fig. 7J.Fouiaperek>pod$ofPan|M8nrkiae:a,5l»v6op^giini5gnic^^ \99\);}),Sympagitnispoyiriiuheaa&at, 1994 (after Lcmaiire, 1994); c, Oncopagurus eidaris Lemaitre, 1996 (after Lemaitre, 1996); d, Syn^agunis dimorpkus (Studer. 1883); e, Partqtagunis sp.; f, Probeebei mirabilis Boone, 1926b; g, Tylaspis eaiomala Hendenon, 1888 (after Lemaitre, in press). Not to scale.

Solitariopagurus (Fig. 12f), and chelate in Birgus is unquestionably a significant reduction in fifth and the diogenid Paguropsis. leg size in all anomurans, but that all non-shell inhabiting anomurans carry the fifth legs within 24. Position and structure of the fifth pereiopod. the branchial chamber is not as universal as these A character that unites the Anomura as a mono- authors suggest. Among the small sample of phyletic entity is the reduction of the fifth pereio- galatheoids we examined, it is questionable pod, and presumably its position in the branchial whether Galathea rostrata A. Mihie Edwards, chamber for gill cleaning (Scholtz & Richter, 1880 actually carries the fifth leg in the branchial 199S). The exception, according to these authors, chamber; species of Uroptychus and Chirostylus is found only in shell-dwelling paguroids. There probably do not, and it is certain that the porcel- 110 P.A. McLaughlin & R. Lemaitre - Carcinization in the Anomura

lanid Petrocheles spinosus Miers, 1876 does not. rasps are present, in varying degrees of complex- Neither do non-shell inhabiting paguioids, such ity, in most paguroid genera. Exceptions include as species of Pagwopsis and Tisea among the Birgus among the Coenobitidae, Tisea among the Diogenidae, Ostraconotus, Porcellanopagurus, Diogenidae, Probeebei and to a lesser extent Ty- Solitariopagurus, Alainopagurus, and Munido- laspis of the Part^Miguridae, and all the Litfaodi- pagurus among the Paguridae, and Bivalvopagu- dae. Richter & Scholtz (1994) reported that rus and Tylaspis among the Parapaguridae, al- among the Lithodidae, Hapalogaster lacked a dxxigh Probeebei and the coenobitid Birgus ap- rasp, but Lithodes maja (Linnaeus, 1758) had a pear to do so. vestigial rasp. We found no lithodids with a rasp The dactyls and propodi of the fifth pereiopods on the pnqxKhis of this appendage. However, in of Neoglyphea inopinata are very distinctively some cases ytben setae had been broken or worn developed and carry a variety of precisely im- off, the setal pores did give the siq>erficial im- planted setal types (Forest & De Saint Laurent, pression of a minute ni^. Pohle (1989) described 1981), and while these appendages are not carried the propodi of the fifth pereiod8 of lithodids as under the canqiace this might sin^>ly be attribut- being provided with four types of setae, including able to the shortness of the branchiostegites, some described as toodied. which appear to just cover the gills. Consequent- ly, it is equally possible that the plesiomorphic state for the reduced fifth pereiopod is one of a Discussion gill cleaning function. Because of polarity ambi- guity we have not included this character in our Without doubt, characters associated with the analysis. proposed carcinization phenomonon are present Like the more anterior {q>pendages, we have in all anomuran major taxa. However, they are scored the simple termination of the fifth pereio- more discernible in die Paguroidea than in either pods plesiomorphic, and chelate the most ad- die Galatheoidea or Hippoidea, and the former vanced. The presence of setae or denticles on the groiq) will be the focal point of the following dis- propodus of the fifth pereiopod is judged more cussion. A corollary to die carcinization hypoth- primitive than the development of a rasp of cor- eses of virtually all previous carcinologists has neous scales. been the concept that hermit crabs, for some rea- 24a. - Fifth pereiopod termination. Forest son, left the shelter of their gastropod shells for a (1987) reported that pylochelids had "more or free existence, ^^ch in turn "awoke" genetic less" subchelate fifth pereiopods. Judging from mechanisms that brought forth various aspects of his figures (Forest, 1987: fig. 7a-f) he used the carcinization. While this is an interesting theory, term subchelate to reflect both truly subchelate it is founded on the premise that those crabs that and semichelate conditions. He did not figure this underwent carcinization evolved from hermits ^>pendage for Cheiroplatea, which we score as that were first and foremost asymmetrical animals tending toward a chelate condition. Chelate fifth that inhabited gastropod shells. pereiopods were found in all of the diogenid gen- The question of asymmetry has not been intro- era we examined, whereas in some parapagurids duced to this point in our consideration of charac- this ^^ndage terminated in a semichelate struc- ters of carcinization, except in relation to pleopod tiuT. Among pagurid genera, the fifth pereiopod loss. Paguroid asymmetry, in its most observable is subchelate in Solitariopagurtis and Alainopagu- forms, is seen in unequal chelipedes, unequal rus, semichelate in Porcellanopagurus and nearly specialized uropods, and the absence of pleopods so in Alainopaguroides, and chelate or nearly so on the right side of the body. Nonetheless, a sub- in the other genera and species examined (Table stantial number of paguroid genera have equal or II). In all other anomurans investigated it is nearly equal chelipedes. Numerous genera also chelate. have equal, and in a few cases, completely un- 24b. - Propodus of fifth pereiopod. Propodal modified uropods. In other genera, uropod asym- Contributions to Zoology, 67 (2) - 1997 111 metry may be a function of habitat (cf. McLaugh- ment of adult copulatory appendages in males, lin & Gunn, 1992). and are fiequently absent on both sides in genera Bouvier (1940: 114) viewed pagurid synunetiy where copulatory iq)pendages do not develop. as being represented by two series of paguroids: The atKsence of right pleopods remains as the the first, die "primitive", symmetrical Pylocheli- principal example of asymmetry. The proposition dae, and the second, those paguroids tiiat pos- that this pleopod loss is a direct result fitMn the sessed paired first and second abdominal pleo- habit of many paguroids to occupy dextial gastro- pods. He proposed that asymmetry evolved, pod shells seems simple enough to the casual through shell use, in the second series, but observer, but is that really true? De^ite the ab- symmetry was secondarily re-acquired, albeit sence of convincing experimental evidence to imperfectly, in those paguroids that achieved a explain how use of dextral shells by hermit crabs crab-like form. could lead to loss of pleopods on the right side, The Pylochelidae are almost exclusively sym- diese crustaceans have become a long-standing metrical, and have been placed between the Tha- classical example of how habitat can influence lassinidea and asymmetrical pagurids by many body plan (e.g., Henderson, 1888; P6rez, 1934; carcinologists (e.g., Stebbing, 1893; Borradaile, Bouvier, 1940; Russell, 1962). This traditional 1916; Wolff, 1961b). The exception is the mono- interpretation, however, was challenged by Rab- typic subfiunily Mixtopagurinae, represented by aud (1941), based on detailed observations of the Mixtopagurus A. Milne Edwards, 1880, which is behavior of various hermit crab species vis-&-vis asymmetrical and utilizes a gastropod shell. Bou- their shell housing. He categorically concluded vier (1940) considered Mixtopagurus paradoxus that the traditional interpretation of pleopod loss, intermediate between his two series. Forest as well as other changes presumably brought (1987) has demonstrated that although symmetri- about by shell use (e.g., handness, migration of cal in virtually every sense of the word, the pylo- internal organs to the abdomen, abdominal mus- chelids morphologically are not primitive pagu- culature, decalcification), are unsubstantiated. It rids, at least in an ancestral connotation. It is our should also be noted that not all paguroids loose belief that their origin was a neotenous one that (or fail to develop) the pleopods of the right side. will be discussed in greater detail in the larval Asymmetrically paired pleopods are reported in segment. the parapagurids Bivalvopagurus and Tylaspis, Martin & Abele (1986: S9S) made the state- and in odier parapagurids total loss of right side ment that "die pleopod of the female first abdom- pleopods is not universal. The diogenids Pagu- inal somite has been lost in many anomurans". ropsis and Cancellus have no unpaired male ple- Richter & Scholtz (1994) found correlations be- opods, but those of the female can occur on either tween an asymmetrical abdomen and pleopod the right or the left side. While it is abundantiy loss, but considered "the situation regarding the clear that we still can not adequately explain the 1st and 2nd pleopods more complex". It is of overwhelmingly more fiequent occurrence of un- primary importance to point out that in the paired left pleopods, it is equally clear diat what Anomura, pleopods do not occur on the first ab- might appear to be the obvious explanation might dominal somite in either sex unless they arise in not be the accurate one. For example, Sandberg & the maturing adult as sexually modified append- McLaughlin (1997) report that pleopods are com- ages. The concept that paired first pleopods are pletely absent in very juvenile Lithodes maja, and lost through shell use is incorrect. These gono- development of pleopods on the left side in fe- pods are never developed in eidier sex in many males is preceded by slight asymmetry of the genera, including several non-shell dwelling abdomen; female gonopods develop considerably paguroids, e.g., Solitariopagunis, Alainopagurus, later. These authors suggest that abdominal asym- Porcellanopagurus, and Ostraconotus. Similarly, metry and asymmetrical pleopod development paired second pleopods, present in megalopae, may simply be a fimction of egg bearing. Given are usually reduced or lost prior to the develop- the fact that male lithodids have symmetrical ab- 112 P.A. McLaughlin & R. Lemaitre - Carcinization in the Anomura

Table 11. List of taxa (59) and cfaancten (37) used in the analysis. Numerals are the same as for diancten in Table and text (see Charactei or character states). Character states: 0-7; missing data: T". Numbers separated by dash indicate muhqtle character states.

Charaeter la lb U If 2b 2e 3a 3b Sa Se 6a TdSM

w M 24 M 143

U »1

M 04

•4

M M

2 2 2 M va Vi I 1 24 04 1 23 0 23 »3 1-3 e e 0 0 0 0 0 0 I I 0-1 0 I I 0 0 0-1 0 0-1 0 O-I 0 0 »3 0 Contributions to Zoology, 67 (2) - 1997 113

Table II. OxaiBuea.

Charaeur 9a 9c 10 Ua 12b 13a 13b 14 15 17a 17b 18 23a 23b 23e 24a 24b TaxoH

04-2

0 0 04 44 M 1-3 M

ace formation and references. Suffice it to say here shape and integumental calcification, to the typi- that calcium needed for cuticular growth is sup- cal pagurid with an entirely membranous abdo- plied in part through diet (e.g., Lafon, 1S>48) or men virtually devoid of segmentation. Concurrent absorption fix>m the ambient water (e.g., Miyaza- with the loss of calcified plates would be in- ki & Jozuka, 1964). The diet of the lithodid Par- creased absorption of fluids to maintain body tur- alithodes camtschatica in the eastern Bering Sea gor, as is seen in the bulbous abdomens of species consists of foods rich in calcium, i.e., mollusks of the Hapalogastrinae, and Porcellanopagurus, and echinoderms (McLaughlin & Hebard, I960), Alainopagurus, etc., among the pagurids. The whereas pagurids and diogenids, as far as known, elongation of the abdomen then well might be are macrophagous and detritivore scavengers justifiably be considered an adaptation to shell- (e.g., Kunze & Anderson, 1979; Schembri, 1982). living. That this paguroid abdomen is not a phy- Contributions to Zoology, 67 (2) - 1997 115 logenetic rejq)pearance of the elongate abdomens al margin has a row of corneous scales. The pro- of other replants is demonstrated by the fact that podal rasps of the fourth pereiopods are most fre- in paguroids (except lithodids) the reproductive quently diought of as "friction pads" of scales organs and elements of the hepatopancreas are (Russell, 1962); however, the single row of pro- located there, a condition not found in other de- podal scales in a substantial number of pagurid capods (McLai^in, 1980). genera, as noted above, could not provide much friction. Perusal of Table II will show that even among our small saiq}le, the propodus of the Fourth and fifth pereiopods fourth pereiopod quite conunonly is provided with only a single row of corneous spines or All anomurans have reduced fifth pereiopods that scales, not the multiple-row rasp familiar to in- may, or may not be carried under the carapace. vestigators of many of the shallow-water species. However, among paguroids, only the lithodids Although the reduction in the fourth pereiopods have retained the fotuth pereiopod as a walking and development of propodal rasps may not q)- leg. The presence of this primitive character was pear directly related to carcinization, both are considered an atavism by Boas (1924), and pos- additional morphological attributes demonstrating sibly an example of a posterior shift in genetic an evolutionary trend away ftom the primitive information for size and spine pattern of anterior elongate and sinq>le appendage to a reduced and leg by Richter & Scholtz (1994). Supportive ev- highly specialized one. idence for the latter claim may have inadvertently come ftom McLaughlin (1983b) who cited Harms (1932,1938) and Wolff (1961b) when she report- Habitat ed, sight unseen, diat ''Birgus abandons its shell shelter, undergoes a moderate degree of carcini- Crypticism or protectionism is characteristic of zation, and takes vcp a free-living existence. With nearly all paguroids except adults of the larger these changes the fourth pereiopods grow and and more mobile genera of the Lithodinae and take on an ambulatory fimction". Now having Probeebei, and for good reason. An even partially personally examined Birgus, we can report that calcified or chitinized integument could be fiitally its chelate fourth pereiopods cannot be equated to vuhierable. Members of the Hapalogastrinae are the ambulatory fourth legs of the lithodids, but commonly found in rocky crevices, as are species rather represent a specialization for its terrestrial of some non-lithodid paguroid genera. O&er life-style. paguroids have adapted to a host of available The fourth pereiopods in Probeebei and Ty- protection mechanisms. For Solitariopagurus and laspis are unquestionably reduced, and when Porcellanopagurus, a clam shell serves the pur- compared with the exceptionally elongate second pose nicely. For Paguropsis its chelate fourth and third perei(^)ods, seem markedly so. Howev- pereiopods, and for Munidopagurus its spinose er, both are simple appendages without rasps on uropods, enable them to conveniently carry an the propodi and widi claw-like dactyls. A pattern anemone above Aeir vulnerable abdomens, al- of specialization can then be followed. The pro- though this is not necessarily an obligate relation- podus of the fourth pereiopod in Ostraconotus is ship (Provenzano, 1971). Pieces of wood provide greatly expanded into a flattened, ovoid structure; an excellent shelter for Xylopagurus species, as the lateral face is covered with minute granules. do scaphopod shells for Pylopagurus and mem- The fourth pereiopods in Solitariopagurus show a bers of other genera. None of these protection hint of a propodal rasp in having three club- devices require asymmetrical uropods or twisting shaped and three spiniform stiff bristles on the of the abdomen. In fact, for some, there has been ventral surface. Alainopaguroides has only two or no need for propodal rasps either. With its greatly three small scales, while Alainopagurus has a row reduced abdomen, Labidochirus splendescens of small, simple spines. Further modification is makes very good use of small gastropod shells seen in Porcellanopagurus where the ventrolater- that are readily covered by a growth of hydracti- 116 P.A. McLaughlin & R. Lemaitre - Carcinization in the Anomura

MDoilyiMH nias. Its simple fouith peieiopods that are larely NMOMDMINMMfphM MpNlila able to be carried within the small shell ate each provided with a small and relatively insignificant rasp, practically obscured by long setae. But for the vast majority of paguroids, what better safe- guard could there be than abandoned gastropod shells to provide a movable shelter? The concept that shell resources are limiting &ctors for certain hermit crab populations is undoubtedly accurate, particularly in the often studied intertidal populations (see Elwood & Neil, 1992 for review). However, as reported by Duns (1992), lack of available shells usually results in selection of a variety of alternative Alters. From die evidence presented, we can find no siq>porting data indicating that this limitation would cause paguroids to evolve a crab-like body form.

Cladograms

Initial iq>plication of PAUP 3.1 to the entire mor- phometric matrix resulted, as might be expected, in a staggering number of 17,000 trees of 469 steps when the analysis was halted. We illustrate (Fig. 14) a majority rule consensus tree. The majority rule provides a good overall idea of wUch taxa form clades. One can thus see the direction of evolutionaiy pathways, although, as indicated earlier, these are only results of a preliminary and partial analysis of die total array of possible data. The patterns of relationships among Fig. 14. Cladognm of anomunm lau geaeiated by majority paguroid genera here does not siq>port the tradi- nik conacnsiit (* •• lilfaodid clade). tional view that evolutionary pathways proceeded from pagurids to lithodids, but radier the reverse. The exception is the Pylochelidae, which in this does the evidence substantiate even gradual car- context appear primitive. However, this is not cinization of a membranous, outstretched pagurid surprising, since the evolutionary pathways being abdomen to ultimately produce the calcified, investigated are only padiways of caroinization. strongly flexed abdomen of the Lithodinae. How- The pylochelids exhibit few, if any, characters ever, a Bouvierian transition, if viewed in re- associated with carcinization. verse, may present a viable hypothesis to corrob- orate a relationship between pagurids and lithodids. Certainly there is no single and direct An alternative hypothesis pathway from lithodid to pagurid, nor is there any particularly cogent one. However, transition from With the addition of a number of morphologically a "crab-like" body form to a "hermit-crab" body intermediate paguroid genera, it is clear that car- form does s^pear feasible and could probably cinization could not have provided a direct tran- have included the following events: sition from a Pagurus-\ike genus to Lithodes. Nor 1. Gradual change in the adult carapace shape Contributions to Zoology, 67 (2) - 1997 117 from a crab-like, entirely calcified lithodid ceph- etc.). Odiers continued to use reclusive rock or alothorax, thioiigh various phases of decalcifica- cotal shelters (e.g., Paguritta, Cancellus), and tion to the ultimate weakly calcified, generally even sponges (cf Sandford, 1994), while devel- flattened canq>ace of Pag}irus-\i)tt genera. oping specialized adaptations. However, with the 2. Gradual reduction in the rostnun fiom the adoption of gastropod shells, hermit crabs gained well-developed, frequently îniform lithodid ros- not only a vastly more abundant concealment re- trum that usually overreaches the eyes, to mark- source, but were provided greatly enhanced mo- edly reduced and/or obsolete rostra of other pagu- bility, and with that, profound and nîd specia- loids. tion. 3. Relaxation of abdominal somites three to six fit>m their strong application against the sternum in the Lithodinae, tlm>ugh gradual loss of calcifi- Acknowledgements cation in die abdominal integument, with concurrent reduction and finally obliteration of segmen- We are deeply indeblBd to Dr. Robot H. Gore for allowing us to use him as a "soundiiig boanT for our ideas. Gahrtfaeoid ipeci- tation and accm a humble hermit her- Benedict, J.E., 189S. Descriptions of new genera and species itê. Rather it were lithodids that began to expe- of crabs of the family Lithodidae widi notes on the young of Lithodes camtschaticus and Lithodes brevipes. Proc. rience integumental calcium loss, for reasons tbat U.S. natn. Mus., 17: 479-488. at this point can only be hypothesized. It then Benedict, J.E., 1901. The hennit crabs of the Pagurus bem- follows that paguroids, over time, left the con- hardus type. Proc. U.S. nam. Mus., 23: 451-466. cealments common-place to the Hapalogastrinae. Benedict, J.E., 1904. A new genus and two new species of Some covered the now somewhat soft abdomen crustaceans of the &inily Albuneidae from the Pacific Ocean; with remarks on the probable use of the antennu- with objects such as halves of bivalve shells (e.g., lae in Albunea and Lepidopa. Proc. U.S. natn. Mus., 27: Porcellanopagurus, Solitariopagurus), anemones 621-625. (some parapagurids, Paguropsis, \funidopagurus. Blackstone, N.W., 1989. Size, shell-living and carcinization 118 P.A. McLaughlin & R. Lemaitre - Carcinization in the Anomura

in geognq>hic populatioas of a hennit crab, Pagunis hir- der Jahie 1843 und 1844, 2. Zoologie (1): 77-14, pis. V- sutiusculus. J. Zool. Load, 217:477-490. VII (Kais. Akad. Wiss.. St Petenburg). Boas, J.E.V., 1880a. 0. Wissenschaftliche Mittheilungen. 1. CauUery, M.. 1896. Crastacis Schizopodes et Dte^Mdes. In: Lidiodes und Pagunu. Zool. Anz., 3: 349-352. R. Koehler, R^suhats scientifiques de la Campagne du Boas, J.E.V., 1880b. Studier over decapodemes Slcgtskabs- "Caudan" dans le Golfe de Gascogne, aoOt-sqMmbre, forhold. K. dansk. Vidensk. SelsL Skr., (S)6:25-210. 1895. Annls. Univ. Lyon. 26:365-419. Boas. J.E.V., 1924. Die veiwandtachafUiche Stellung der Gat- Chace, F.A., Jr., 1939. Report on the scientific resultt of the tung Lidiodes. K. dansk. Vidensk. Selsk. Skr., 4<4): 1-34. first Atlantis Expedition to dte West Indies, etc. Prelimi- Boone, L., 1926a. Unusual deep-sea Crustacea - some fomu naiy descriptions of one new genus and seventeen new secured by the Arcturus OManographic Expedition. New species of decq>od and stomatopod Crustacea. Memorias Yoric zool. Soc. Bull., 29: 69-72. Soc. cub. Hist nat, 13: 31-54. Boone, L., 1926b. A new £unily of Crustacea. Preliminary Chace, F.A.. Jr.. 1942. Reportt on the scientific resutts of the technical description. New York zool. Soc. Bull., 29: 73. Atlantis Expeditions to tte West Indies, under the joint Borradaile, L-A., 1916. Crustacea. Part II. PotceUanopagu- auqnces of the University of Havana and Harvard Uni- rus: An instance of carcinization. In: British Antarctic versity. Twreia, 11: 1-106. (Terra Nova") Expedition, 1910. Natural History Report Coflin, H.G., 1957. A new soudiem fbnn of Tagurus hirsu- Zoology, 3(3): 111-126. tiuscuhu" (Dana) (Crustacea, Decapoda). Walla Walla Bourne, G.C., 1922. The Raninidae: A study in careinology. CoU. Pubb. Dq>. biol. Sci., 21:1-8. J. Linn. Soc. London, Zool., 35:25-79. Crain, J.A. & PA. McLaughlin, 1994. Larval, postlarval, and Bosc, LA.G., 1802. Histoire Naturelle des Ciustac^ conte- early juvenile development in Pagurus venturensis CofRn, nant leur description et leurs moeurs; avec figures des- 1957 (Decapoda, Anomura, Paguridae) reared in the Ubo- sinies d'^nes nature, 2:1-296, pis. 9-18 (De Ouilleminet, ratoiy, with a redescription of tte adult Bull. mar. Sci., Paris). 53 (3) "1993": 985-1012. Bouvier, E.L., 1894a. Sur les caractiies et revolution des Cunningham, C.W.. N.W. Bladcstone & L.W. Buss, 1992. Lomisines, nouveau groupe de Crustac^ anomures. C.r. Evolution of king crabs fiom hermit crab anceston. Na- hebd. Sdmc. Acad. Sci., 118: 1353-1355. ture, Lond., 355: 539-542. Bouvier, E.L., 1894b. Sur la signification de Hqtalogaster Dana, J.D., 1851. Conspectus Cmstaoeoram ({uae in oibis dans revolution des Lithodinis. C.r. somm. Sianc. Soc. tentarum circumnavigatione, Carolo WiOws e classe philom. Paris, 18: 1-5. leipublicae foederatae duce, lexit et descrqtsit (Preprint Bouvier, E.L., 1894c. Sur la transformation des Paguriois en from) Proc. Acad. nat Sci., Philad., 5: 267-272. crabes anomures de la sous-famille des Litfaodines. C.r. Dana. J.D., 1852a. Conq>ectus Crustaoeorum. &c. Conq>ec- hebd. Seanc. Acad. Sci.. 119: 350-352. tus of the Crustacea of die Exploring E]q>edition under Bouvier, E.L., 1895a. Sur les Lidtodinis hapalogastriques Capt Wilkes, U.SJ>I.. including the Paguridea. continued, H^xJogaster et Dermaturus. C.r. somm. S^nc. Soc. phi- the Megalopidea, and die Macroura. Paguridea. contin- lom. Paris, No. 18:56. ued, and sttbtribe Megalopidea. (Preprint from) Proc. Bouvier, E.L.. 1895b. Recherches sur les affinitis des Lith- Acad. nat Set Philad., 6:6-28 (1854). odes & des Lomis avec les Pagurides. Annls. Sci. nat, (7) Dana, JJ>.. 18S2b. Crustacea, part II. United States Bxpkx^ 18: 157-213. ing Expeditioa. During the years 1838,1839,1840,1841, Bouvier, E.L., 1896. Sur la classification des LiAodinis et 1842. Under the command of Chartes Wilkes, U.S.N., 13: sur leur distribution dans les oceans. Annls. Sci. nat, (8) 686-1618 (C. Sherman, Philadelphia; reprinted in 1972 by 1: 1-46. Antiquariaat Junk, Lochem, The Netherlands). Bouvier, E.L., 1897. La transfoimatioa des Bernards Duris, Z., 1992. On a small collection of Crustacea Decapo- rEimite en Uthodes. Naturaliste, (2) 19 (230): 41-43. da from the Bellsund region, Spitsbergen. In: J. Repe- Bouvier, E.L., 1940. Faune de France, 37. Dteapodes lewska-Pekalowa &. K. Pekala (eds.). Wyprawy Ge- Marcheurs: 1-404 (P. Lechevalier, Paris). ografiezne na Spitsbergen: 121-154 (UMCS, Lubin, Po- Brandt, J.F., 1848. Die Gattung Lithodes Latreille nebst vier land). neuen ihr verwandten von Wosnessenski entdeckten, als Elwood, R.W. & SJ. Neil, 1992. Assessments and decisions. Typen einer besonderen Unterabtbeilung (Tribus Litho- A study of information gathering by hermit crabs: 1-192 dea) der Edward'schen Anmnuren. Bull. Cl. phys.-n)ath. (Chapman A Hall, haoAaa etc.). Acad. Sci. St Pitersbourg, 7: 171-176. Fabricius, J.C., 1775. Systems entomologiae, sistens Insecto- Brandt, J.F., 1850. Vorlaiifige Bermeikungen uber eine aus ruffi classes, ordines, genera, species, adiectis synonymis, zwei noch unbeschriebenen Gattungen und Arten gebilde- locis, descriptionibus, observationibus: [i-xxxii], 1-832 te Unterabtbeilung (H^Mlogastrica) der Tribus Lithodea, (Officina Libraria Kortii, Flensburgi et Lipsiae). begieitet von einer Charakeristik der eben genannten Tri- Fabricius, J.C, 1787. Mantissa Insectorum sistens eoium bus der Anomuren. Bull. Cl. phys.-math. Acad. Sci. St. species nuper detectas adjectis characteribus genericis, Pitersbourg, 8 (16-17): 266-269. differentiis specificis, emendationibus, observationibus, Brandt, J.F., 1851. Krebse. In: A. Th. v. MiddendorfT, Reise 1: i-xx, 1-348 (C.G. Proft, Hafhiae). in den aussersten Norden und Osten Sibiriens wihrend Fabricius. J.C, 1793. Entomologia systeniatica emendata et Contributions to Zoology, 67 (2) - 1997 119

uicta, secundum classes, ordines, genera, species, adjec- gic et biochimie. In: P.-P. Giassi &. J. Forest (eds.), Traiti tis synonimis, locis, observationibus, descriptionibus, 2: de zoologie. Anatomie, systtaiatique, biologie, 7(1): 49- i-viii, 1-S19 (C.G. Proft, Hafiiiae). 90 (Masson. Paris etc.). Fabricius, J.C, 1798. Supplementum entomologiae systemat- Gore, R.H., 1974. On a small collection of porcellanid crabs icae: [i-iv], 1-S72 (C.G. Proft, Hafiiiae). from the Caribbean Sea (Crustacea, DecqMda, Anomura). Faxon, W., 1893. Reports on the diedging operations off the BuU. mar. Sci.. 24: 700-721. west coast of Central America to the Galapagos, to the GuMn-Mteeville, F.E., 1853. m. Melanges et nouvelles. west coast of Mexico, and in the Gulf of California, in Revue Mag. Zool., (2) 5:45-48. charge of Alexander Agassiz, carried out by the U.S. Fish GuMn-Mfaeville, F.E., 1855. In: R. de la Sagra (ed.), Histo- Commission steamer "Albatross" during 1891, Lieut- ria flsica, politica y natural de la isla de Cuba. Vol. 8, At- Commander Z.L. Tanner, U.S.N., commanding, VI. Pre- las de Zooiogia. Crusticeos, Aricnidos y Insectos: pis. 1- liminary descriptions of new species of Crustacea. Bull. 20 (A. Bertrand, Paris). Mus. comp. Zool., Harv., 24 (7): 149-220. Guinot, D., 1979. Donnies nouvelles sur la morphologie, la Feldmami, R.M., 1984. Haumnriaegla glaessneri n. gen. and phylogenise et la taxonomie des Crustac^ Dteqwdes sp. (Decapoda; Anomura; Aeglidae) from Hmmurian Brachyoures. Mem. Mus. natn. Hist nat, (A, Zool.) 112: (Late Cretaceous) rocks near Cheviot, New Zealand. N.Z. 1-354. J. Geol. Geophys.. 27: 379-385. Haan. W. de, 1833-1850. Crustacea. In: P.F. von Siebold, Filbol, H., 1885. Description d'un nouveau genre de Crus- Fauna Japonica sive descriptio animalium, quae in itinere taci provenant de la Nouvelle-Zilande. Bull. Soc. philom. per J^xmiam, jussu et auspiciis superioium, qui summum Paris, (7) 9:47-48. in India Batava Imperium tenent, sutcepto, annis 1823- Forest, J., 195 la. Ccmtribution i I'itude du genre Poroellano- 1830 coUegit, notis, observationibus et adumbtatiottibus pagurus Filbol (Paguridae), 1. Description de P. edwardsi illustravit. Fasc. 1-8: i-xvi, vii-xvii, i-xxxi, 1-243. pis. I- Filhol. Bull. Mus. natn. Hist, nat, (2) 23: 82-90. LV, 1-2, A-Q (apud auctorem, Lugduni Batavoium ft J. Forest, J., 19Slb. Contributiim i I'itude du genie Porcellan- MiUler ft Co. Amstelodami). opagurus Fiihol (Paguridae), 2. Remanques systimatiques Harms, J.W., 1932. Die Realisation von Geaen und die con- et biologiloraiory voyage of H.M.S. taci Dteapode Glyphtide. In: Risultats des Campagnes Challenger, (Zoology) 27: i-xi, 1-221, pis. I-XXI (Her MUSORSTOM, 1. Philippines (18-28 mars 1976). Mini. Majesty's Stationary Office, Edinbur^ etc.). ORSTOM. 91: 51-84. Herbst J.F.W., 1791-96. Versuch einer Naturgeschichte der Forest J. & M. de Saint Laurent 1989. Nouvelle contribu- Krabben und Krebse nebst einer systematischen Be- tion i la connaissance de Neoglyphea inopinata Forest & schreibung ihrer verschiedenen Arten, 2. Krebse: i-viii, 1- de Saint Laurent i propos de la description de la femelle 226, pis. XXI-XLIV (G.A. Lange, Berlin/Stialsund). adulte. In: J. Forest (ed.), Risultats des Campagnes MU- Hobbs. H.H., III, 1979. A new species of the endemic South SORSTOM, 5. yUm. Mus. natn. Hist, nat, (Zool.) 144: American genus Aegla from Parani, Brazil (Crustacea: 75-92. Anomura: Aeglidae). Proc. biol. Soc. Wash., 91(4): 982- Garcia-Gdmez, J., 1988. A new genus and three new species 988. of hermit crabs (Crustacea: Decapoda: Paguridae) from Kropp, R.K., 1986. A neotype designation for Petrolisthes the western Atlantic Ocean. Bull. mar. Sci., 42(1): 44-64. tomentosus (Dana), and description of PetrolisAes hete- Glaessner, M.F., 1957. Evolutionary trends in Crustacea rochrous, new species, from the Mariana Islands (Anomu- (Malacostraca). Evolution, 11: 178-184. ra: Porcellanidae). Proc. biol. Soc. Wash., 99(3): 452-463. Goffinet, G. & Ch. Jeuniaux, 1994. Le tegument: morpholo- Kunze, J. & T. Anderson, 1979. Functional morphology of 120 P.A. McLaughlin & R. Lemaitre - Carcinization in the Anomura

the mouthpam and gastric mill in the hermit crabs Cliba- scriptions of four new q>ecies. In: A. Crosnier (ed.), nahus taeniatus (Milne Edwards), Clibanarius viiescens Risultats des Campagnes MUSORSTOM, 12. Mim. Mus. (Krauss), Paguristes squamosus McCulloch and Dardanus natn. Hist nat., (Zool.) 161:375-419. setifer (Milne Edwards) (Anoda: Anomura) from Australia: animaux sans vertibres, ou tableau gteiral des classes, Species of Strobopagurus Lemaitre. 1989. Sympagurus des ordres et des genres de ces animaux; pr6sentant les Smith, 1883 and two new genera. Rec. Aust Mus.. 48: caractires essentiels et leur distribution, d'aprts la con- 163-221. sideration de leurs rapports naturels et de leur organisa- Lemaitre. R., (in press). Crustacea Decapoda: Revisiting Ty- tion, et suivant I'amngement itabli dans les galeries du laqns anomala Henderson, 1885 (Parapaguridae), with Mttsium d'Hist. Naturelle, parmi leur d^pouilles con- comments on its relationships and evohitionary signifi- serves; pricide du discours d'ouverture du cours de zo- cance. Zoosystema. ologie, donni dans le Museum National d'Histoire Na- Lemaitre, R. &. N.H. Campos, 1993. Two new hermit crabs turelle I'an 8 de la R^blique: i-viii, 1-432 (chez (Crustacea: Decapoda: Paguridae) frnn the Caribbean I'auteur & Dterville, Paris). Sea. Proc. biol. Soc. Wash.. 106(3): 554-565. Lamarck, J.B.PA. [Monet, chevalier de], 1818. Histoire na- Lemaitre, R. A P>. McLaughlin, 1995. Alainopagurus cros- turelle des animaux sans vertibres, prfewaitant leurs nieri n. gen., IL sp. (Decapoda: Anomura: Paguridae) from caractires gtainaa. et particuliers de ces animaux, leur the western Pacific. Bull. Mus. natn. Hist nat, (4) 17A distribution, leurs classes, leurs ftmiUes, leurs genres, et (3-4): 273-282. la citation des principales esptees qui s'y lappottent; pri- Linnaeus, C, 1758. Systems naturae per regna tria naturae, ddie d'une introduction offiant la ditetmination des smmdimi classes, ordines, genera, species, cum charac- caractires essentiels de I'animal, u distinction du vigital teribus. difierentiis. syixmymis, locis (ed. 10). 1: i-iv, 1- et des autres corps naturels, enfin, I'eiqMMition des princi- 824 (Laurentii Salvii, Holmiae). pes fondamentaux de la zoologie, 5: i-iii, 1-612 (Diter- Linnaeus, C, 1767. Systems naturae per r^na tria naturae, ville & Verdi6re, Paris). secundum classes, ordines, genera, species, cum charac- Latreille, PA., 1806. Genera Crustaceoium et Inaectorum se- teribus, differentiis, synonymis, locis (ed. 12), 1(2): 533- cimdum ordinem naturalem in familias disposita, iconibus 1327, [1-33] (Laurentii Salvii, Hohniae). exemplisque plutimis explicata. 1: 1-302 (Treuttel et Macpherson, E., 1988. Revision of die ftmily Lithodidae Wiirtz, Parisiis et Argentorati). Samouelle. 1819 (Crustacea, Decapoda, Anomura) in the Latreille, [P.A.], 1829. Crustac^ Arachnides et partie des Atlantic Ocean. Monografias de Zoologia marina, 2: 1- Insectes. In: [C] Cuvier, Le tigne animal distribui 153. d'apris son organisation, pour servir de base i I'histoire Man, J.G. de, 1902. Die von Herm Professor KOkenthal im naturelle des animaux et d'introduction i Tanatomie com- Indischen Archipel gesammelten Dekapoden und Stoma- pare (6d. 2), 4: i-xxvii. 1-S84 (Ditervilie & Crochard, topoden. Abh. Senckenb. naturforsch. Ges., 25: 467-939, Paris). pU. XIX-XXVU. Laurent, R.F., 1983. Irreversibility: A comment on Mac- Mtttin, J.W. & L.G. Abele. 1986. Phylogenetic relationships beth's interpretations. Syst. Zool., 32: 75. of the genus Aegla (Decapoda: Anomura: Aeglidae), with Leach, W.E., 181S. A tabular view of the external characters comments on anomuran pfaylogeny. J. crust Biol., 6: 576- of four classes of animals, which Linni arranged under 616. Insects; with the distribution of the genera composing Martin, J.W. & L.G. Abele, 1988. External moti^iology of three of these classes into orders, &c. and descriptions of the genus Aegla (Crustacea: Anomura: Aeglidae). Smitii- several new genera and species. Tians. Linn. Soc. Lond, son. Contr. Zool., 453: 1-46. 11:306-400. McLaughlin, P.A., 1974. The hennit crabs (Crustacea Deca- Leach, W.E., 1820. Galatiadto. Dictionnaire des Sciences poda, Paguridea) of northwestern North America. Zool. Naturelles, 18: 49-S6 (F.G. Levrault & Le Notmant, Stras- Vetfa. Uiden. 130: 1-396. bourg/Paris). McLaughlin, P.A., 1980. Comparative morphology of Recent Lemaitre, R., 1989. Revision of the genus Parapagurus Crustacea: 1-177 (W.H. Freeman & Co., San Francisco/ (AnomurK Paguroidea: Parapaguridae), including rede- New Yoric). scriptions of the western Atlantic species. Zool. Verb. McLaughlin, P.A., 1981. Revision of Pylopagurus and To- Uiden, 253: 1-106. mopagurus (Crustacea: Decapoda: Paguridae), with the Lemaitre, R., 1993. A new genus of Parapaguridae (Decapo- descriptions of new genera and species: Part I. Ten new da: Anomura). Crust. Res., 22: 11-20. genera of the Paguridae and a redescription of Tomo- Lemaitre, R., 1994. Crustacea Decapods: Deep-water hennit pagurus A. Milne Edwards and Bouvier. Bull. mar. Sci., crabs (Parapaguridae) from French Polynesia with de- 31(1): 1-30. Contributions to Zoology, 67 (2) -1997 121

McLaugfalin, P.A., 1983a. A review of the phylogenetic posi- cueillis pendant I'exp^tion. Mem. Mus. comp. Zool. tion of the Lomidae (Cnutacea: Decapoda: Anomala). J. Harv., 14(3): 5-172. cnut Biol.. 3:43M37. Mihie Edwards, A. ft E.L. Bouvier, 1894a. Coosidirttions Mclaughlin, PA.. 1983b. Hennit crabs - Are they really gteirales sur la famille des Galathiid6s. Annls. Sci. nat, polypbyktic? J. cnut Biol.. 3: 608-621. 14: 191-327. McLaughlin. P.A.. 1997. Cnistacea Dec^Mda: Hennit crabs Milne Edwards, A. ft E.L. Bouvier, 1894b. Troisiteie cam- of the ftmily Paguridae from the KARUBAR e3q>edition pagne du yacht rHirondelle, 1887. Neolidwdes, genre in Indonesia. In: A. Crosnier &, P. Bouchet (eds.). nouveau de la sous-£unille des Litfaodiiiis. Bull. soc. Risultats des Campagnes MUSORSTOM, 16. Mim. Mus. Zool. Fr., 19: 120-122. natn. Hist nat, (Zool.) 172: 433-S72. Milne Edwards. H., 1831. Recherches sur I'oiganisation et la Mclaughlin, P.A. & S.W. Gunn, 1992. Revision of Pyio- classification naturelle des Crustac^ Dtcxpodet. Annls. pagiinu and Tomopagunis (Cnistacea: Decapoda: Paguri- Sci. nat, (1) 25: 298-332. dae). with the descriptions of new genera and species: Milne Edwards, H., 1836. Observations zoologiques sur les Part IV. Lophopagunis McLaughlin and Australeremus Pagures et description d'un nouveau genre de la tribu des McLaughlin. Mem. Mus. Victoria. S3(I): 43-99. Paguriens. Annls. Sci. nat., (Zool.) (2) 6:257-288. McLaughlin. P.A. St, J.F. Hebaid, 1960. Stomach contents of Mihie Edwards, H., 1837. Histoire naturelle des Craitac^ Bering Sea long crabs. Spec, scient. Rep. (Fish.), 293: 1- comprenant Tanatomie, la physiologie et la classification 5. de ces animaux, 2: [i-iii], 1-532 (Ubrairie Eacy- Mclaughlin, P.A. A L.B. Holtfauis. 198S. Anomura venus clopMique de Roret, Paris). Anomala. Cnistaceana. 49: 204-209. Milne Edwards, H. ft H. Lucas, 1841. Descrqition des Crus- Melin. G.. 1939. Paguriden und Galatheiden von Prof Dr. tacfe nouveaux ou peu connus conserve dans la collec- Sixten Bocks Expedition nach den Bonin-Inseb 1914. tion du Mustem. Archs. Mus. Hist nat Paris, 2:461-483. KungL svenska VetenskAkad. Handl., (3) 18 (2): 1-119. Miyazaki, K. ft K. Jozuka, 1964. Uptake and turnover of Mien, EJ., 1876. Descriptions of some new species of Cnis- *'Ca by the crayfish, Procambanis clatkiL Annual Rep. tacea, chiefly from New Zealand. Ann. Mag. nat Hist., Noto mar. Lab. Fac. Sci. Univ. Kanazawa, 4:111-120. (4) 17:218-229. Monod, Th., 1956. Hippidea et Brachyura ouest-africains. Mien. EJ.. 1879. Revision of the Hippidea. J. Linn. Soc., Mim. Inst. fir. Afr. noire. 45: 1-674. (Zool.) 14: 312-336. pi. 5. Milne Edwards, A., 1862. Faune carcinologique de I'lle Morgan, GJ. ft J. Forest. 1991. A new genus and q>ecies of Bouibon (Extrait). Annls. Sci. nat., (Zooi.) (4) 17: 362. hermit crab (Crustacea. Anomura, Diogenidae) from the Mifaie Edwards, A., 1880. Repon on the results of dredging, Timor Sea, north Australia. Bull. Mus. natn. Hist nat, under the siqiervisioa of Alexander Agassiz, in the Gtilf Paris (4) (A, Zool.) 13: 189-202. of Mexico, and in the Caribbean Sea, 1877, 78. 79. by the Nixon, K.C. & J.M. Carpenter, 1993. On outgnx^n. Cladis- United States Coast Survey steamer "Blake", Lieut.-Com- tics, 9: 413-426. mander CJ>. Sigsbee, U.S.N., and Commander J.R. Ban- Ottmann, A., 1892. Die Decapoden-Krebse des Stiassburger lett, U.S.N.. commanding. VIII. Etudes priliminaires sur Museums, mit besonderer Betiicksichtigung der von les Ctustac^. Bull. Mus. comp. Zool. Harv., 8(1): 1-68. Herm Dr. DSderiein bei Jafta und bei den Liu-Kiu-Insebi Mihie Edwards, A., 1891. Pagurides nouveaux des Azores. gesammelten und zur Zeit im Strassburger Museum auf- Campagnes scientifi({ues de S.A. Le Prince de Monaco bewahrten Formen, IV. Die Abtheihmgen Galatbeidea sur le yacht I'Hirondelle. Bull. Soc. zool. Fr., 16: 131- und Paguridea. Zool. Jb.. (Syst.) 6:241-326. pis. 11-12. 134. Osawa, M., 1995. A new par^Mgurid genus, Tsunogaipagu- Mibie Edwards, A. &. E.L. Bouvier, 1891. Observations rus, for Syaq>agunis chuni (Balss, 1911) (Crustacea: De- gtoinles sur les Paguriens recueiUis dans la mer des An- capoda: Anomura). Proc. Japan Soc. syst Zool., (53): 62- tilles et le Golfie du Mexique, par le Blake et le Hassler, 70. sous la direction de M. Alexandre Agassiz. Bull. Soc. phi- Owen, R., 1839. Cnistacea. In: F.W. Beechey (ed.). The zo- lomath. Paris, (8) 3(1): 102-110. ology of Captain Beechey's voyage; comp. from the col- Mihie Edwards, A. & E.L. Bouvier, 1892. Observations pri- lections and notes made by Captain Beechey, the ofiicen liminaires sur les Paguriens recueillis par les exp^tions and naturalist of the expedition to the Pacific and Befa- du Tnvailleur et du Talisman. Annls. Sci. nat., Zool. ring's straits performed in His Majesty's ship "Blossom", Paltont, (7)13: 185-226. under the command of C^itain F.W. Beechey ... in the Milne Edwards, A. ft E.L. Bouvier, 1893. Repons on the re- yean 1825,26.27 and 28: 77-92 (H.G. Bohn. London). sults of dredging, under the supervision of Alexander Perez, C, 1934. Expose de biologie zoologique, 1. Les Agassiz, in the Gulf of Mexico (1877-78), in the Carib- Pagures ou Bernard Termite (un example d'adaptation). bean Sea (1878-79). and along the Atlantic coast of the Actuality scientifiques et industrielles, Paris, 101: 1-33. United States (1880). by the U.S. Coast Survey Steamer Pilgrim. R.L.C., 1965. Some features in the moiphology of "Blake", Lieut.-Commander CD. Sigsbee, U.S.N. and Lomis hina (Lamarek) (Crustacea: Decapoda) and a dis- Commander J.R. Bartlett, U.S.N., commanding. XXXIII. cussion of its systematic position and phylogeny. Aust J. Description des Cnistaces de la famille des Paguriens re- Zool., 13: 545-557. 122 P.A. McLaughlin & R. Lemaitre - Carcinization in the Anomura

Pilgrim, R.L.C., 1973. Axial skeleton and musculature in the of Mexico, and its hermit crab associates. In: R.W.M. van thotax of the hennit crab, Pagunis benihardus [Anomuia: Soest, T.M.G. van Kempen &, J.-C. Braekman (eds.). Paguridae]. i. mar. Biol. Ass. U.K., 53: 363-396. Sponges in time and space. Biology, chemistry, paleontol- Pohle, G.. 1989. Gill and embryo grooming in litfaodid ciabs: ogy. Proceedings of the IV International Porifiera Confer- Comparative fimctiCrustacean Issues, 6: 75-94 (A.A. Balkema, nouveaux de TAmfaique tropicale. Revue Mag. Zool., (2) Rotterdam). 9: 501-505. Poore, G.C.B., 1994. A |diylogeny of the £unilies of Tha- Say, T., 1817. An account of the Crustacea of die United lassinidae (Crustacea: Decapods) with keys to families States. J. Acad. nat Sci. Miilad., 1: 155-169. and genera. Mem. Mus. Victoria, 54: 19-120. SchaJfeew, P., 1892. Carcinologische Bemeikungen aus dem Poupin, J. &, P.A. McLaughlin, 1996. A new species of Soli- Zoologischen Museum der Kaiserlichen Akademie der tariopagurus TQtkay (Decapoda: Paguridae) from French Wissenschafien. Bull. Acad. imp. Sci. St Pitersbouig, 35: Polynesia. Bull. Mus. natn. Hist, nat, Paris, (4) 18 (A, 331-342. Zool.) (1/2): 211-224. Schembri, PJ., 1982. Feeding behaviour of fifteen species of Powell, G.C., 1974. Gregarious king ciabs. Sea Frontiers, 20: hermit crabs (Crustacea: Decapoda: Anomura) from the 206-211. Otago region, southeastern New Zealand. J. nat Hist, 16: Provenzano, AJ., Jr., 1968. Biological investigations of the 859-878. deep sea, 37. Lithopagurus yucatanicus, a new genus and Schmitt, W.L., 1921. The marine deaqpod Crustacea of Cali- q)ecies of hermit crab with a distinctive larva. Bull. mar. fornia with special reference to the decapod Crustacea Scl, 18: 627-644. collected by the United States Bureau of Fisheries Steam- Provenzano, AJ., Jr., 1971. Rediscovery of Munid(q>agurus er "Albatross" in connection with die biological survey of macrocheles (A. Milne-Edwaids, 1880) (Crustacea, Deca- San Francisco Bay during the years 1912-1913. Univ. poda, Paguridae), with a description of the first larval CaUf. Pubis. Zool., 23:1-470. stage. Bull. mar. Sci., 21(1): 256-266. Schmitt, W.L., 1935. Crustacea Mactura and Anomura of Rabaud, E., 1941. Recheiches sur I'ad^rtation et le compor- Porto Rico and the Virgin Islands. Scientific survey of tement des Pagures. Archs. Zool. exp. gte., 82: 181-285. Potto Rico and die Virgin klands, New York Acad. Sci., Randall. J.W., 1840. Catalogue of the Crustacea brought by 15(2): 125-227. Thomas Nuttall and J.K. Townsend, from the west coast Schmitt, WJ.., 1942. The species of Aegla, endemic Soudi of North America and the Sandwich Islands, with descrip- American freshwater crustaceans. Proc. U.S. natn. Mus., tions of sticb species as are ^>parently new ... J. Acad. 91:431-520. nat. Sci. Philad., 8: 106-147. Scholtz, G. & S. Richter, 1995. Miylogenetic systematics of Rensch, B., 1959. Evolution above the species level: 1-419 the reptantian Deoqioda (Crustacea, Malacostraca). ZooL (Methuen &. Co., Ltd., London). J. Linn. Soc.. 113:289-328. Richter, S. & G. Scholtz, 1994. Morphological evidence for Scopoli, G.A., 1777. lotroductio ad historiam naturalem sis- a hermit crab ancestry of lithodids (Crustacea, Decapoda, tens genera Iqridum, plantarum, et animalium hactenus Anomala, Paguroidea). Zool. Anz., 233 (5/6): 187-210. detecta, caracteribus esaentialibus donata. in tribus divisa, Russell, E.S., 1962. The diversity of animals. An evolution- subinde ad leges naturae: [i-x], 1-506, [1-34] (W. Gerie, ary study. Acta biotbeor., 11: 1-151. Pragae). Saint Laurent, M. de. 1968. Revision des genres Cau- Sluys, R., 1992. Evolutionary side-steps. Bijdr. Dierk., paguroides et Cestopagunis et description de quatre gen- 62(2): 121-126. res oouveaux, I. Catapaguroides A. Milne Edwards et Smith, S.I., 1879. The stalk-eyed Crustaceans of the Atlantic Bouvier et Deci4>hyllus nov. gen. (Crustacis Dicapodes coast of North America north of Cape Cod. Trans. Conn. Paguridae). Bull. Mus. mun. Hist nat., (2) 39 (5) Acad. Arts Sci., 5(1): 27-136. ("1967"): 923-954. Smith, S.I., 1883. Preliminary report on the Brachyura and Saint Laurent, M. de, 1979. Vers une nouvelle classification Anomura dredged in deep water off the south coast of des Crustac^ Dteapodes Reptantia. BulL Off. natn. Pteh. New England by the United States Fish Commission in Tunisie, 3: 15-31. 1880,1881, and 1882. Proc. U.S. nam. Mus., 6(1): 21-57. Saint Laurent-Dechanci, M. de, 1966. Iridopagurus, genre Snodgrass, R.E., 1952. The sand crab Emerita talpoida (Say) nouveau de Paguridae (Crustacis Dteapodes) des mers and some of its relatives. Smitbson. misc. Coll., 117(8): tropicales amMcaines. Bull. Mus. nam. Hist, nat., (2) 1-34. 38(2): 151-173. Spears, T., L.G. Abele & W. Kim, 1993. The monophyly of Sandberg, L. & P.A. McLaughlin, 1997 (in press). Crustacea, brachyuran crabs: A phylogenetic study based on 18S DecqK>da, Paguridea. Marine invertebrates of Scandina- rRNA. Syst. Biol., 41 ("1992"): 446-461. via, 10 (Universitetsforlaget, Oslo). Stebbing, T.R.R., 1893. A history of Crustacea. Recent Ma- Sandford, P., 1994. The Florida heimit-crab sponge, a little lacostraca: 1-466 (Kegan Paul, Trench, Trfibner & Co. known "mobile" sponge from the N.E. comer of the Gulf Ltd., London). Contributions to Zoology. 67 (2) - 1997 123

Steviid, Z., 1971. The main features of brachyuian evolu- Tudge, C.C, 1997. Spetinatological evidence supports the tion. Syst Zool.. 20:331-340. taxonomic placement of the Australian cadonic hairy Stevens, BA., 1927. Oithopagunis, a new genus of Paguri- stone crab, Lomis hirta (Decapoda: Anomura: Lomidae). dae from the Pacific coast. Publs. Puget Sound mar. biol. Mem. Mus. Viet., 56(1): 235-244. Stn.. 5:245-252. TQrkay, M., 1986. Crustacea Decapoda Reptantia der Tie£we Stevenson. J.R., 1985. Dynamics of llie integument. In: D.E. des Roten Meeres. Senckenberg. marit, 18(3/6): 123-185. Bliss A LJI. Mantel (eds.). The biology of Crustacea, 9. Weber, F., 1795. Nomenclator entomologicus secundum en- Integument, pigments, and hormonal process: 1-42 (Aca- tomologiam systematicam ill. Fabricii, adjectis speciebus demic Press, Inc., New York). recens detectis et varietatibus: i-viii, 1-171 (^Hid Caro- Stimpson, W., 1838. Prodromus descriptimiis animalium lum Emestum Bohn, Chilonii [Kiel] et Hambutgi). evertebratotum, quae in expeditione ad Oceanum Pacifi- White, A., 1856. Some remarks on Crustacea of Ae genus cum SqMentiionalem, a Rq>ublica Federate missa. Cad- Lithodes, with a brief description of a species qiparently waladaro Ringgold et Johanne Rodgers ducibus, ob- hitherto unrecorded. Proc. zool. Soc. Lond., 24: 132-135, setvavit et descripsit, VII. [Prq>rint (December 1858) pi. XLn. from] Proc. Acad. nat. Sci. Philad., 10:225-252. Whiteaves, J.F., 1874. Rqxnt on further deep-sea dredging Stimpson, W., 1860. A trip to Beaufort, North Carolina. operations in the Gulf of St Lawrence. Amer. J. Sci., Amer. J. Sci. Arts, (2)29: 442-445. (3)7: 210-219. Studer, Th., 1883. Verzeicbniss der wihrend der Reise Williams, A.B., 1965. Maritte decapod crustaceans of the S.M.S. 'Xiazelle'' an der WestkOste von Afiika, Ascension Carolinas. Fish. Bull, 65(1): i-xi, 1-298. und dem Cap der Guten Hoffiiung gesammelten Crusta- Williams, A.B., 1984. Shrinqis, lobsters, and ciabs of the At- ceen. AUi. Akad. wiss. Berlin., Phys. Abh., 2 "1882": 1- lantic coast of the eastern United States, Maine to Florida: 32. i-xvii, 1-550 (Smithsonian Institution Press, Washington, Swammerdam, J., 1737. Biblia naturae; sive historia Insecto- D.C.). lum, in classes ceitas redacta, nee non exemplis, et ana- Williamson, D.I., 1992. Larvae and evolution. Toward a new tomico variorum animalculorum examine, aeneisque tab- zoology: 1-223 (Chapnum & HaU, New York). ulis illustrata. Insertis numerosis rariorum naturae obser- Wolff, T., 1961a. A bathyal-abyssal hermit crab with a calci- vationibus. Bybel der Natuure, of historic der insecten, tot fied asymmetrical abdomen. Nature, 190 (4779): 931-932. zeekere zootten gebracht: ioot vootbeelden, ontleedkun- Wolff, T., 1961b. Description of a remarkable deep-sea her- dige ondeizoekingen van veelethande kleine gediertens, mit crab with notes on the evolution of the Paguridae. als ook door kunstige kopere plaaten <^>geheldett. Ver- GaUtheaRep.,4: 11-32. rijkt met ontelbaare waamemingen van nooit ontdekte Wood-Mason, J., 1873. On Nephropsis Stewarti, a new ge- zeldzaamheden in de natuur, 1: [i-lxii], 1-550 (I. Severi- nus and q>ecies of macrurous crustaceans, dredged in nus, B. van der Aa & P. van der Aa, Leyden). deep water off the eastern coast of the Andaman Islands. Tilesius, W.G., 1815. De cancris camtschaticis, oniscis, en- i. Asiatic Soc. Bengal. 42(2): 39-44. tomostracis et canceUis marinis microscopicis noctilucen- tibus, cum tabulis IV: aenaeis et appendice aditexo de acaris et ricinis Camtschaticis. Conventui exhibuit die 3 Febniarit 1813. Mdn. Acad. iiiq>^. Sci. St Pitersbourg, 5: 331-405. Received: 22 March 1997