From hermit to king, or king to hermit?
Analysis by Richter & Scholtznext section
Richter & Scholtz (1994: 188) stated unequivocally that without a phylogenetic analysis it would be impossible to decide whether the evolution from a “hermit” to a “king” or from a “king” to a “hermit” would be more likely. Similarly, they noted that the question of polarity of evolutionary change could only be solved on the basis of a phylogenetic analysis using additional characters and with a careful and detailed comparison of the characters examined to make homology possible. Characters used by Richter & Scholtz to unite the Paguridae and Lithodidae included, but not exclusively: division of the second to fifth pleonal tergites into two lateral halves; asymmetry; fusion of the first pleonal sternite with the last thoracic sternite; fusion of some basal segments of the “outer” flagellum of the antennule; absence in most males of second pleopods; and the presence of one or more accessory teeth on the ischium of the third maxilliped. The Lithodidae were nested within the Paguridae, but considered to represent a monophyletic assemblage because adults all lack uropods and males lack pleopods 3-5.
Richter & Scholtz [1994: 198 (2)], essentially following Boas’s (1926) hypothesis, considered the division of the second through fifth tergites into two lateral plates with a loss of calcification in the midline a synapomorphy uniting the pagurids and lithodids. In terms of the adult condition, this might appear to be the case. However, in addition to demonstrating that lithodid marginal plate development in tergite 2 is not homologous with marginal plate development in tergites 3-5, we have shown that these developmental processes are considerably more complex than heretofore imagined.
Richter & Scholtz (1994) commented that in correlation with the asymmetry of the pleon, some pleopods had been lost. They cited specifically the loss of pleopods 3-5 on one side of the pleon in males and females of “asymmetrical” hermit crabs and female lithodids But these authors indicated that the situation of the male first and second pleopods as more complex. Richter & Scholtz (1994: 197 (3)] described the condition of paired first and second pleopods in Paguropsis typica Henderson, 1888, Paguristes barbatus Ortmann, 1892 (= Paguristes ortmanni, Miyake, 1978) and Sympagurus dimorphus (Studer, 1883) as representing the plesiomorphic state, and the ancestral state of asymmetrical hermit crabs, just as paired pleopods on pleomeres 3-5 were considered plesiomorphic. We concur that the situation presented by paired first and second pleopods in male paguroids is not just a simple matter of paired pleonal appendages. Paired first pleopods are absent in anomuran and brachyuran decapods, unless they are developed as copulatory structures. This is precisely the development seen in the examples cited by Richter & Scholtz (1994), as well as certain other paguroids. Similarly, paired second pleopods in these particular taxa also are modified as gonopods. However, when unpaired second pleopods are present in adult male paguroids, they usually are not modified as gonopods, e.g., Diogenes, Dardanus, certain Pagurus species. Does a sexually modified pair of appendages represent a more primitive state than an unpaired and non-specialized appendage, or does loss precede specialization in the evolutionary framework? Clearly appendages arising from the second pleomere in species of one genus are positionally homologous (cf. Minelli & Schram 1994) with those of another, but are they developmentally homologous? Our juvenile data suggest that perhaps they are not.
Richter & Scholtz [1994: 205 (5)] found the occurrence in lithodid female of paired first pleopods even more difficult to interpret. They suggested that the most parsimonous assumption would be that this pair of appendages represented a secondary appearance (atavism), i.e., the reappearance of a pagurid ancestral character. In referring to the paired first female pleopods as gonopods, numerous authors, including McLaughlin & Lemaitre (1997: 114) have implied a sexual function similar to that of males for these female appendages; however, no such function ever has been documented. Pohle (1989) reported that in some lithodids, eggs were carried by these pleopods. No similar use has been recorded for other paguroids. As is pointed out in our discussion of pleopods, paired first pleopods in some female paguroids, like paired first and second pleopods in some male paguroids, do not appear to be developmentally homologous with the pleopods of the megalopa, where no first pair are ever developed. It would seem that if buds of first pleopods do occur in embryonic development as Richter & Scholtz (1994: 206) state, their appearance in lithodids where they sometimes facilitate egg-carrying would suggest the primitive state. The absence of this function and the ultimate loss of these pleopods in many paguroid genera would then reflect the apomorphic conditions.
Contrary to Richter & Scholtz’s [1994: 198 (4)] belief, not all genera of the Paguridae have one or more accessory teeth on the ischium of the third maxilliped (de Saint Laurent-Dechancé 1966; McLaughlin, 1997). Fusion of the last thoracic and first pleonal sternites undoubtedly occurs during juvenile development, but these sternites are independent in megalopae and early juveniles (pers. obs.). Of the two other characters utilized by Richter & Scholtz (1994), antennular article fusion can not properly be interpreted since articular segmentation is just beginning during the megalopal and early juveniles stages (Forest 1987; Carvacho 1988; McLaughlin et al. 1989, 1992; Crain & McLaughlin 1993). Sternal fusion may occur prior to maturity, but such fusion is not apparent in early juvenile stages.
Questions of asymmetry
The handedness, as well as the pleopod and uropod asymmetries that are seen in paguroids are commonly attributed to the frequent use of many species of dextrally coiled gastropod shells. Some early carcinologists, however, suggested that asymmetry was a predetermined condition (Rathke 1842; Agassiz 1875; Thompson 1903; Boas 1926). Although all animals have been said to be asymmetrical at the molecular level (Ageno 1972; Neville 1976), symmetry is generally considered to be the primitive or plesiomorphic condition (Bouvier 1940; Scholtz & Richter 1995; Palmer 1996). However, when symmetry is encountered among paguroids, it has been assumed to have been secondarily acquired (Bouvier 1940; Russell 1962; Elwood & Neil 1992; Richter & Scholtz 1994).
Asymmetry is not a phenomenon limited to paguroids. In fact, decapod asymmetry has been a subject of interest for nearly a century (Przibram 1905; Emmel 1908), but as noted by Oppenheimer (1974) interest in asymmetry has risen and fallen from time to time. Interest would again appear to be on the rise (Palmer 1994, 1996; Chippindale & Palmer 1994; Graham et al. 1998 Klingenberg et al. 1998), and while most studies have been theoretical or broad-based, certain very important facts have been documented. Asymmetry has evolved independently many times in higher animals, and conspicuous asymmetry, according to Palmer (1996), falls into two general categories: directional asymmetry (asymmetry toward a particular side) or antisymmetry (random asymmetry). In Palmer’s view, different ontogenetic origins of asymmetry imply different patterns of phylogenetic precedence; early or larval asymmetry is indicative of transition directly from symmetry to directional asymmetry, whereas “postlarval” asymmetry is indicative of a transition from symmetry to antisymmetry and ultimately to directional asymmetry. Of pertinence to our investigation is his observation of what he referred to as a nonrandom environmental trigger that might bias the late-developing asymmetries of left and right handed hermit crabs. Neither group appeared to have passed through the antisymmetry stage. However, Palmer’s consideration of pagurid asymmetry was limited only to handedness, which he associated with propensity of many hermits to live in asymmetrical gastropod shells. Unfortunately, the complexities of paguroid asymmetries do not appear to have any straightforward answers. For example, not all hermit crabs have asymmetrical chelipeds, pleopods or uropods; some are perfectly symmetrical. In contrast, lithodids never occupy gastropod shells, nonetheless, both sexes exhibit cheliped asymmetry. However, it is only females who manifest pleonal plate and pleopod asymmetry and this is not apparent until crab stage 4 or 5 in the species that have been studied. Slight directional asymmetry is evident in the cheliped development of megalopae of some species of lithodids, whereas pleopod and uropod loss remains symmetrical. Similarly, cheliped asymmetry may be apparent in other pagurid megalopae, but it is at crab stage 1 that asymmetrical pleopod loss usually first becomes observable. In contrast, while diogenid megalopae may have asymmetrical pleopods and uropods, numerous species also have symmetrical chelipeds. Additionally, in those genera where adult asymmetry is pronounced, it most commonly is at crab stage 1 that handedness becomes apparent, as do pleopod and uropod asymmetries. However, not all of these species are asymmetrical shell dwellers. From studies on Calcinus verrillii (Rathbun, 1901), a species known to utilize both gastropod shells and polychaete worm tubes, Rodrigues et al. (2002) have suggested that it might be easier for hermit crabs to change from symmetrical to asymmetrical uropods than the reverse. Contrarily, Harvey (1998) argued that the loss of asymmetry in the uropods of Clibanarius vittatus Bosc, 1802, when deprived of spirally coiled shells, demonstrated that the constraints against symmetry were minimal. Gherardi (1996a) apparently did not observe any uropod asymmetry when normally uropod-symmetrical Discorsopagurus schmitti (Stevens, 1927) megalopae selected gastropod shells over polychaete worm tubes as habitats. The observation made by Millett T. Thompson (1903: 195) “This question of the origin of asymmetry seems to me to be insoluble at the present day.” is as true now as it was a century ago. But it is most probable that no single genetic or environmental trigger is responsible.
Pleopod loss and gain
Paired first pleopods in females of most lithodid genera caused Boas (1924) to change his view of the pagurid ancestor from Pagurus to a genus like Nematopagurus or Pylopagurus. Neither of the latter genera were included in Cunningham et al.’s (1992) study, nor are they included here for lack of larval, megalopal, and early juvenile data. McLaughlin & Lemaitre (1997) reported the presence of paired first pleopods in adult females of all Pylochelidae, the diogenid genera Paguristes Dana, 1851 and Paguropsis Henderson, 1888, and several genera of the Paguridae. No paired pleopods are present on the first pleomere in the zoeal or megalopal stages of any paguroid, although Richter & Scholtz (1994) suggested they might be present in embryonic stages. When paired first pleopods do develop in the adult they appear to represent new structures, not modifications of existing megalopal pleopods, and they do not arise until sometime after crab stage 10 (Provenzano & Rice 1966; Sandberg & McLaughlin 1998; McLaughlin & Paul 2002).
Buds, representing developing pleopods, appear, with few exceptions, in the penultimate or ultimate zoeal stages on pleomeres two to five of paguroids. These appendages, fully developed, paired, and usually biramous are characteristic of the megalopa. They are lost entirely in lithodids either with the molt to crab stage 1 or shortly thereafter, and are replaced by newly developing pleopods on the left side only in females at approximately crab stage 5. In contrast, pleopods are reduced in the first and/or second crab stages, and ultimately lost on one side of the pleon, in subsequent juvenile stages in the majority of pagurids. It is unclear what mechanism(s) influence pleopod loss, but loss is not uniform throughout the superfamily. For example, Provenzano & Rice (1966) reported the reduction and ultimate loss of both second pleopods and third through fifth right pleopods in Paguristes sericeus A. Milne-Edwards, 1880 at crab stage 2. The left second pleopod remained absent through crab stage 10. Females developed an egg-bearing second pleopod on the left side of the pleon gradually in later stages, whereas males did not. In contrast, Brossi-Garcia (1987b, 1988) described pleopod loss beginning at crab stage 2 in two species of another diogenid genus, Clibanarius Dana, 1852; but this loss involved only the pleopods of the right side. Variations were also seen in the five species of Pagurus in the present investigation. In studies of P. venturensis Coffin, 1957 by Crain & McLaughlin (1993) and P. kennerlyi (Stimpson, 1864) by McLaughlin et al. (1989), marked reduction was found in the second pleopods on both sides at crab stage 1 and complete loss at stage 2. Similar reduction was observed for third through fifth pleopods on the right side at crab stage 1, with complete loss in crab stage 2. However, a tiny second pleopod bud was apparent on the left side of a presumed female of P. venturensis in crab stage 3, suggesting a return similar to, but more rapid than, that observed in Paguristes sericeus. Comparable second pleopod reduction in Pagurus bernhardus (Linnaeus, 1758) was reported by Carvacho (1988) and in P. ochotensis Brandt, 1851 by McLaughlin et al (1992) but without complete loss in crab stage 2. Clearly, pleopod absence and loss, like asymmetry, lacks a single straightforward explanation.
Uropod loss and transformation
The presence of a tailfan consisting of the telson and the paired appendages of the last pleomere is considered an ancestral decapod structure (Paul et al. 1985; Paul 1989). Uropods represent a pair of larval appendages on the sixth pleomere. As has been reported by Williamson (1974, 1988) and confirmed for some lithodids by Crain & McLaughlin (2000b), when uropods are present they appear in the third zoeal stage, whether or not that is the ultimate zoeal stage. Williamson (1974) suggested that lack of uropod development in the higher Brachyura resulted from ancestral stock that had only two zoeal stages. In most lithodid genera where uropods are entirely lacking, the species do have only two zoeal stages. Nonetheless, uropods never develop in the lithodid genera Acantholithodes and Cryptolithodes, each having four zoeal stages (Jensen pers. comm.; Hart 1965; Kim & Hong 2000), which certainly suggests the lack of uropod stage dependency. Uropods, when present in lithodids, are uniramous and are lost with the molt to crab stage 1 in all genera studied except Placetron (cf. Crain & McLaughlin 2000b). Since adults of P. wosnessenskii lack uropods, this is another example of heterochronic loss. In contrast, with the molt to crab stage 1 in all non-lithodids studied, the biramous megalopal uropods remain biramous but become substantially modified by the development of corneous scales on the dorsal surfaces. This development occurs whether the species is a shell-dweller or not (Forest 1987; McLaughlin et al. 1989, 1992, 1993; McLaughlin 2000). Uropod asymmetry, as suggested above, may be under genetic and/or environmental control.
Polarity – or what constitutes a primitive character state?
In their study of adult characters, McLaughlin & Lemaitre (1997) used Neoglyphea inopinata Forest & de Saint Laurent, 1975, as the out-group for testing the hypothesis that carcinization had occurred and that lithodids represented the advanced state. No information on development is available for N. inopinata, thus no parallel out-group analysis is possible.
Kluge (1985) and O’Grady (1985) both perceived ontogenetic data as providing three basic uses to phylogenetic systematics, i.e., to provide assessment of homology, to serve as an extra source of observations with which to judge historical relationships, and to serve to polarize character transformations. Although Christofferson (1987, 1988a, b, 1989) and Ng & Clark (2000) used both larval and adult characters in assessing phylogenetic relationships of the Eucarida, Caridea, and certain Brachyura, as pointed out by Hickman (1999), the exclusive use of invertebrate larval data in cladistic analyses is in its infancy. Only a few recent larval decapod studies have applied phylogenetic methods (Clark & Webber 1991; Marques & Pohle 1995, 1998). Polarity was determined by Clark & Webber (1991) and Marques & Pohle (1995) using the assumption that evolution had proceeded by oligomerization, i.e., the loss and reduction of segments and setation elements. However, Marques & Pohle (1998), using out-group analysis, tested the applicability of the oligomerization method of polarity determination, and found it unjustified in that it did not provide the most parsimonious explanation of the data set. In contrast, they believed their results demonstrated that only the use of out-group comparison to polarize character transformations would produce the most parsimonious hypotheses while allowing the researcher to recognize possible addition events. However, it should be noted that Koenemann & Schram (2002) now argue that maximum parsimony is a biased method to analysze development sequence data.
Marques & Pohle’s (1998) results not withstanding, and the out-group method’s common usage in assessing polarity in adult phylogenetic investigations (Meier, 1997), some researchers consider that the out-group method has disadvantages when applied to ontogenetic sequences (Nelson, 1978, 1985). For example, the out-group method of determining polarity assumes a knowledge of higher relationships (Nelson 1973a, b, 1978), and also is concerned, not so much with character polarity, but with connotations of ancestry (Williams et al. 1990). The selection of one or more taxa as out-groups theoretically establishes a hypothetical ancestor with all primitive characters, and character-state distributions are summarized at the out-group node (Maddison et al. 1984; Bryant 1997). Since our concern is for attributes pointing to possible carcinization events, our application of cladistic methods is once again somewhat unconventional (cf. McLaughlin & Lemaitre 1997: 96), thus determination of ancestry is not our specific aim.
An alternative to the out-group method is the ontogenetic polarity criterion, or Nelson’s Rule (Nelson, 1978). As elucidated by Bryant (1997), the distribution of character states through the ontogenies of members of the ingroup is used to infer the expected character states at the ingroup node. Specifically, Nelson’s criterion considers that the transformation of an ontogenetic character observed to be more general to one observed to be less general, represents a transformation from primitive to advanced. Nelson, however, did not interpret general and common as equivalents, as some of his subsequent critics have (Kluge & Strauss 1985; Kluge 1988; Kraus 1988). Patterson (1994, 1996) paraphrased Nelson’s (1978) criterion as “absence is more general than presence”. Unfortunately, generality is often in the eye of the beholder, and can be interpreted quite differently by different investigators (e.g., Mabee 1989, 1996; Patterson 1996). Similarly, absence does not universally precede presence (e.g., Fong et al. 1995).
In the opinion of de Queiroz (1985) “... characters do not transform in ontogeny; ontogenetic transformations are themselves the characters.” Importantly, Wheeler (1990) has made a justifiable distinction between character adjacency and character polarity. In an ontogenetic sequence, transformations in character states can be observed (character state adjacency), but such transformations do not provide information about polarity per se. Tests on the out-group criterion and “Nelson’s Rule” (Wheeler 1990; Meier 1997) have shown that both methods give approximately equally parsimonious results. Bryant (1992), in comparing the two methods, indicated that both require monophyly of the study group and comparison of equivalent sets of ontogenetic stages. Additionally, according to Bryant, the use of out-group analysis necessitates monophyly of a more inclusive group, or close relationship between the study group and particular out-groups, as well as an adequate survey of the distribution of character states among out-groups. Contrarily, Nelson’s Rule, as interpreted by Bryant, requires the retention of plesiomorphic states in the ontogenies of descendants. In the present study of megalopal and early juvenile development, available evidence suggests that all of the above criteria for the application of Nelson’s Rule have been met. However, as pointed out by Christofferson (1995), ontogenetic polarities of instantaneous characters, because of heterochrony, do not necessarily coincide with phylogenetic polarities of ontogeny. From our data, we have definitive evidence on character state adjacency. By applying Nelson’s Rule, we have, for some characters, been able to postulate polarities. However, in other character transformations, this method of polarity determination does not seem applicable. For example, there is no ontogenetic evidence to suggest that certain losses and/or transformations occur in ordered, stepwise manners among taxa. Specifically in coding pleopod reduction or loss and uropod loss or transformation, we have used the “intermediate method” of Wilkinson (1992, 1995), which hypothesizes that if a character state is intermediate in form, size or number between two other character states it is considered phylogenetically intermediate between those two other states. The rationales that have gone into our decision-making processes are presented for the characters used in our analyses.
The term semaphoront has been interpreted differently by various authors (Hennig 1966; Brooks & Wiley 1985; de Queiroz 1985; Nelson 1985; O’Grady 1985; Wheeler 1990). Our use of the term, like our application of cladistic analysis is perhaps unconventional. We use semaphoront to indicate an assemblage of individuals of a monophyletic group at an identifiable and comparable period in their life cycles. In the present investigation we have examined two semaphoronts, megalopal/juvenile and larval. The former semaphoront includes all paguroid species for which data on particular aspects of megalopal/juvenile morphology (megalopa, crab stage 1, and crab stage 2) are available.
Marques & Pohle (1998) stressed the need for developmental homology, emphasizing that if a phylogenetic hypothesis is based on non-homological semaphoronts, that hypothesis will likely be wrong. Having found no evidence to the contrary, we have followed Hennig’s (1966) auxiliary principle and assumed [developmental] stage homology in a “postlarval” (megalopa/juvenile) semaphoront. Regrettably, data are not complete for all stages beyond crab stage 1. Maddison (1993) and Hawkins et al. (1997) have reviewed the problems arising from missing data, versus missing characters and pointed out some of the erroneous results that can occur when computerized analyses encounter missing information. Consequently, we have limited our data to taxa experimentally observed, reported in the reliable literature and/or, where character adjacency has permitted, extrapolations. However, given the reversals observed in lateral and marginal plate divisions in Lithodes at crab stage 3, we have not pursued formal cladistic analyses beyond crab stage 2. This stage limitation notwithstanding, we have utilized the data from crab stage 3 and beyond in drawing our ultimate conclusions.
Confirmable reports of direct development in the Paguroidea by Barnard (1950), Dechancé (1963), and Morgan (1987) have described hatchings at the megalopal stage (pre-imago); in all other studies one or more zoeal (larval stages) also are involved. In our analyses, the megalopa is considered the basal point of the “postlarval” semaphoront whether abbreviated development is considered advanced (cf. Rabalais & Gore 1985) or primitive (Wolpert 1990, 1994). Among members of our “postlarval” semaphoront, the occurrence (insertion) of the megalopal stage (pre-imago) is universal.
Our larval semaphoront was to have included those paguroid species for which pertinent zoeal data could be paired with megalopal/juvenile information, at least at the generic level. However, we encountered two major problems. The first dealt with character selections. Phylogenetic brachyuran larval studies most frequently have placed considerable emphasis on setal differences in primary zoeal feeding appendages (Clark & Webber 1991; Marques & Pohle 1995, 1998; Clark 2000; Ng & Clark 2000). Unfortunately in paguroids, these appendages, i.e., paired mandibles, maxillules, and maxillae, may be well developed or rudimentary, depending upon whether or not the larval and/or megalopal stages are lecithotrophic (Van Dover 1982; McLaughlin et al. 2001, 2003). In the zoeal studies considered in our review, all paguroid species had type 1 (cf. Van Dover et al. 1982) maxillary scaphognathites in zoeal stage 1 and followed those authors’ type A sequence through subsequent stages; however, setation was noticeably reduced in species known to be at least facultatively lecithotrophic.
Lecithotrophy has not been seriously investigated in paguroids. However, Anger (1989)and Harvey (1996) have reported the occurrence of ‘secondary lecithotrophy’ in the megalopal stage of several Pagurus species and Harvey has indicated that both zoeal and megalopal stages of Paguristes tortugae Schmitt, 1933 are lecithotrophic. Anger (1996) and McLaughlin et al. (2001, 2003) have recorded similar lecithotrophy in species of Lithodes and Paralomis. To avoid the introduction of seemingly morphological differences that may in fact be differences in developmental patterns (cf. Chai 1974), feeding appendages had to be omitted from our consideration.
The second problem dealt with ontogenetic stage homology, a matter of critical concern in phylogenetic analyses. In the Paguroidea, although the most common number of zoeal stages is four, that number does vary among and within genera, and occasionally even within species. Such variation in some taxa simply reflects environmental influences (Cabrera Jiménez 1966; Lang & Young 1977; Brossi-Garcia & Hebling 1983; Siddiqui et al. 1991), while in others these variation actually are deletions [elimination of a stage ] (Gore 1985; Clark 2000) (or insertions; cf. Wolpert 1994). Marques & Pohle (1998), when confronted with variations in the number of zoeal stages in their study groups, were able to demonstrate that the second stage of their ingroup was homologous with the second stages of the out-groups despite variation in the total number of stages. Clark (2000) however, considered specific morphological attributes as characters across a series of stages with polarity conferred by the timing of their appearances. Specifically, a given character in all taxa presumably was homologous across all stages; its polarity was considered derived if it appeared earlier in the sequence of zoeal stages than the same character in the out-group. This reasoning requires that stages also are ontogenetically homologous among all taxa. Perhaps this is a reasonable expectation in some brachyuran groups; however it is certainly not the case in paguroids.
For example, a review of paguroid larval characters has shown that while some of the variations observed appear to be simply terminal deletions or insertions, e.g., four rather than five zoeal stages or vise versa, some more importantly would seem to be initial deletions (stage 1, or elements of it, eliminated or passed through prior to hatching). Initial deletion means that the state of a character in species hatching in a more ‘advanced’ condition will not be developmentally homologous with that same character in a species without advanced hatching. In one of the more readily recognizable scenarios, the first decapod zoeal stage (ZI) (prezoeas excluded) is defined as having sessile eyes regardless of the number of zoeal stages (Gurney, 1942: 123). However, in Lopholithodes (cf. Crain & McLaughlin 2000a), Lithodes (cf. McLaughlin et al. 2001), Paralomis (cf. McLaughlin et al. 2003), Anapagurus (cf. Ingle 1990), and Pylocheles mortensenii (cf. Saito & Konishi 2002), the eyes at hatching are stalked or at least only partially fused to the orbital wall. In Lithodes only three zoeal stages precede the metamorphic molt, suggesting perhaps that the first stage has been deleted. In Paralomis where the molt to megalopa is preceded by only two zoeal stages, it would appear as though at least one or perhaps even two zoeal stages have been deleted. In contrast, in the diogenid genus Paguristes, where two or three zoeal stages also are reported (Hart 1937; Rice & Provenzano 1965; Provenzano 1978), ZI eyes are sessile, suggesting that perhaps stage deletions in this genus are terminal. Alternatively, Ingle (1990) for the pagurid, Anapagurus chiroacanthus (Lilljeborg, 1855), and Crain & McLaughlin (2000a) for the lithodid, Lopholithodes mandtii, reported stalked or at least only partially fused eyes in ZI; both taxa have four zoeal stages, the number most common for species of Pagurus, which have sessile eyes at hatching. Do differences in ocular development among species of Paguristes and Pagurus, with two or three and four zoeal stages, respectively, but sessile eyes initially, reflect the primitive condition, while species of Anapagurus and Lopholithodes, each also with four zoeal stages but stalked eyes, reflect the derived condition in equivalent stage ZI larvae? If so, is this derived state ontogenetically homologous with the “derived” states seen in “ZI” Lithodes and “ZI” Paralomis with only three and two zoeal stages, respectively? We have found no evidence to allow us to assume that this is true. The condition of the eyes at hatching is just one of a number of stage-variable characters of this complexity that cast critical doubt on developmental stage homologies or equivalent sets of ontogenetic stages among paguroids. Without these, meaningful phylogenetic analysis cannot be expected. Consequently, we have not cladistically evaluated our larval phase data.
Megalopa/early juvenile characters and character states.
We have followed the recommendation of Wiens (2001) to explicitly document our character choices. Table 1 provides a summary of the characters and character states for the “postlarval” (megalopa/juvenile) semaphoront. Since data for crab stage 2 are more restricted, we have conducted our analysis at two levels, i.e., those taxa for which complete data are available for the three postlarval stages, and those taxa where data are available only for the first two postlarval stages. Table 2 furnishes the list of paguroid taxa examined and the respective character states for megalopa, crab stage 1 and crab stage 2.
Although much emphasis has been put on the transition to (or from) a crab-like pleon, the carapace of a shell dwelling hermit crab and the carapace of a well calcified, crab-like lithodid similarly represent a transformation from one condition to another. Additionally, the rostrum, common to the majority of decapod zoeae, persists at least through the megalopal stage (Williamson 1982), but undergoes major reduction in juvenile stages of many paguroids. We consider three characters in the category, carapace.
1a. – Calcification. Scholtz & Richter (1995) reported a strongly calcified exoskeleton for most adult reptant decapods, except for “asymmetrical hermit crabs” and some thalassinids, although they considered a soft cuticle as exhibited by “natant” decapods the “original” condition. While a calcified megalopal carapace is present in some lithodids, that is not true for all, nor is it true for most pagurids. In the subsequent juvenile stages, some degree of calcification may be gained, but if a calcified integument is initially present it is not lost, it is simply strengthened. Of itself, this restricted evidence would suggest that a chitinous cephalothoracic integument is the plesiomorphic condition. With any decapod, the postmolt exoskeleton is initially uncalcified (Greenaway 1985). However, this cannot be viewed as indicating an evolutionary precursory condition. The onset of calcification of the carapace varies, but clearly calcification is not an orderly and progressive transformation. We have interpreted an initially calcified carapace as primitive because of the comparable corresponding states of pleonal calcification (see rationale for characters 5-9).
1b. – Carapace delimitation. With relatively few exceptions, the pagurid megalopal carapace consists of a well-defined, chitinous shield and membranous posterior portion. In contrast, among lithodids there seems to be variation in the extent to which the anterior carapace is delimited, but a specific shield is not similarly identifiable in lithodids. Applying Nelson’s Rule, the more general condition would be lack of specific delineation.
1c. – Rostrum. Rostral processes are generally well, or at least better, developed in the megalopa in all families, but show gradual or substantial reduction in subsequent stages. We consider the more general, thus primitive condition, to be a well-developed rostrum.
2. Ocular acicles
The interpretation of what constitutes an ocular acicle varies among authors, although all concur that its presence is apomorphic. Jackson’s (1913: 40) definition as a spearhead shaped “squama”, or Makarov’s (1938: 126, 1962: 120) “as a small appendage ... which is usually oval with a pointed tip” has been interpreted to mean the entire calcified plate of the penultimate segment (cf. Powar, 1969) of the peduncle, as implied by McLaughlin (1974, 1983) and Martin & Abele (1986), and has been specifically defined as such by Sandberg & McLaughlin (1998) and Forest et al. (2000). In contrast, Richter & Scholtz (1994: Figs. 3A, B) and Boyko & Harvey (1999: 383, fig. 2A) have restricted the ocular acicle to the “spinose or platelike anterodorsal extensions.” The calcified plate often present on the second (penultimate peduncular segment), with or without projections, is understood here as representing the ocular acicle.
3. Thoracic appendages
Only two pairs of thoracic appendages are considered in our analysis, as these are the only ones that appear to be correlated with carcinization.
3a. Development of the fourth pereiopod is of primary importance. The more general condition, seen in lithodids, i.e., development of the fourth pereiopod as a walking leg, is considered primitive, rather than an atavism as claimed by Boas (1924), or gene replication of the third pereiopod as suggested by Richter & Scholtz (1994).
3b. The reduction of the fifth pereiopod, although apomorphic when evaluated among all decapods, is a character shared by all Paguroidea. As indicated by Pohle (1989), Richter & Scholtz (1994), and Scholtz & Richter (1995) this appendage in the adult is specially adapted for different functions among members of the superfamily. The more general usage of this appendage is as a gill cleaner, and as such is provided only with setae. We consider setae only as the primitive condition.
4. Pleon carriage.
The megalopal pleon is carried, at least initially, in a straight and fully extended position in all paguroids studied. While it may remain straight or become twisted in subsequent crab stages, it may also be flexed under the cephalothorax and held closely against the cephalothorax. We consider the extended, straight pleon more general, and therefore to represent the plesiomorphic condition.
5 -9. Pleonal tergites.
Integumental calcification of the tergites, like calcification of the carapace, is not easily polarized, although Goffinet & Jeuniaux (1994) considered decalcification of the pleon a secondary loss in paguroids. Fully calcified megalopal tergites are present in some, but not all lithodids, nor are they present in most pagurids. Following Nelson’s Rule, lack of calcification should, being more general or absent, be adjudged the plesiomorphic state. However, character state adjacency demonstrates that calcification, if initially present in megalopal tergites, may be lost, whereas if calcification is not present initially in those tergites, but is subsequently gained, it is not then again lost. Consequently we consider initial calcification of the megalopal tergites the primitive state. Nevertheless, fusion and calcification in each tergite, as well as tergal subdivisions must be considered as characters distinct for each of the first five tergites. Based on the limited data available, the megalopal/juvenile sixth tergite does not provide any significant ontogenetic information.
5. – Tergite 1. Only two characters are applicable to this tergite, calcification and tergal identity. Presence of a calcified tergite is considered plesiomorphic, as is the retention of tergal identity .
6. – Tergite 2. Three characters are needed to evaluate the changes that occur in this tergite: 1) calcification, 2) division, and 3) loss of identity. Character states for tergal division are undivided, marginal plates delineated, lateral plates delineated, and marginal and lateral plates delineated; for tergal identity: distinct, partially lost or obscured, and identity completely lost or nearly so. The primitive states are calcified (for the reason stated above), undivided, and distinct, the latter two being the universal megalopal conditions.
7-9. – Tergites 3-5. In addition to the characters of the second tergite, one additional character is needed, i.e., development of accessory marginal nodules. Lack of development of such nodules is considered the plesiomorphic condition.
Differences in telson shape and armature are observable in early juveniles. The typical, roundly subrectangular or subquadrate, and unarmed telsons of the megalopae are judged to be plesiomorphic.
After reviewing these losses in early juvenile stages among diogenids, pagurids, parapagurids and lithodids, it has become clear that the majority of species undergo a more or less gradual loss. We have, therefore, coded the complete loss and various degrees of gradual pleopod loss numerically using Wilkinson’s (1992, 1995) ranking methods, but these representations are not meant to suggest an ordered transition.
With the molt to megalopa, uropods, if initially present in lithodids, are quickly lost, while uropods in other paguroids are dramatically altered. Here again we have used Wilkinson’s (1992, 1995) method to code the character states, but not to suggest an ordered transformation.
We will consider only cheliped asymmetry here, as it is the only element not incorporated into other characters. Megalopal symmetry is adjudged primitive, however coding of right and left handedness is not an indication of presumed polarity.
14. Zoeal stages .
We have included the number of zoeal stages passed through by each taxon, whether these might be considered to have preceded the megalopa or been inserted following it because there undoubtedly is informational value in the extent of larval influence.