Anomura (or Anomala) is the taxon with the greatest disparity within Decapoda when the number of different body shapes – crabs, squat lobsters, symmetrical and asymmetrical hermit crabs and mole crabs (Tsang et al., 2011) – is taken into account. Apart from mole crabs, none of these groups is likely to be monophyletic. The crab-like body shape alone evolved three times independently within anomurans, and independently from the crab-like body shape in brachyurans (Fig. 1; see also Reimann et al., 2011). The term carcinization was introduced a century ago to describe the evolutionary process which led to a crab-like habitus (Borradaile, 1916; McLaughlin, 1983; McLaughlin and Lemaitre, 1997; Scholtz, 2014), though the phenomenon itself began to be investigated even earlier (e.g. Boas 1880a, b; Bouvier 1895). Consistently with the disparity found in the external morphology of anomurans, comprehensive studies into the internal anatomy of various anomuran taxa have revealed a variety of differently constituted inner anatomical structures (Jackson, 1913; Pike, 1947; MacGaw and Duff, 2008; Keiler et al., 2013, 2015a, b). Some of these inner structures appear to be phylogenetically informative, e.g. provide evidence of sister group relationships (Keiler et al., 2013, 2015a, b), and at the same time shed light on the morphological transformations which took place in the course of carcinization. Studies into king crabs (Keiler et al., 2013, 2015b) and porcelain crabs (Keiler et al., 2015a) have shown that carcinization in these two lineages was characterized by similar changes in external morphology, e.g. the pleon became flexed and started to be carried tightly under the cephalothorax, which led to the concave emargination of the posterior margin of the sternal plastron and the broadening of the carapace. Some internal organs, especially those of the nervous system and the hemolymph vascular system, were also affected by carcinization in similar ways in the different taxa. In other respects, however, these and other organs evolved differently, reflecting the convergent evolution of a crab-like habitus (Keiler et al., 2013; 2015a, b).
Fig. 1. Different phylogenies of Anomura based on molecular + morphological (Schnabel et al., 2011) and solely molecular data (Tsang et al., 2011 and Bracken-Grissom et al., 2013) implying that Lomisoidea evolved from a squat lobster-like ancestor. Coloured dots indicate body shape in the ground pattern in the respective taxon. Underlined taxa included crab-like forms.
To extend the insights into carcinization obtained by previous investigations, we studied Lomis hirta (Lomisoidea: Lomisidae) as the third and final taxon in our survey of crab-like anomurans. Lomis hirta is endemic to the intertidal of southern Australia and northern Tasmania (Poore, 2004) and the only representative of the monotypic Lomisoidea. For comparison, Aegla cholchol (Aegloidea: Aeglidae) and Kiwa puravida (Chirostyloidea: Kiwaidae), each belonging to similarly exceptional lineages within anomuran crustaceans, were also studied. Aegloidea is represented by a single genus, Aegla, whose representatives are endemic to South America and – despite a single hermit crab species (McLaughlin and Murray, 1990) a unique feature among anomuran crustaceans – live in freshwater (McLaughlin et al., 2010). Representatives of Kiwa, also known as ‘yeti crabs’, live chemosynthetic-associated at deep sea methane seeps or hydrothermal vents (Thurber et al., 2011; Roterman et al., 2013). While Lomis has a crab-like habitus, Aegla and Kiwa each resemble squat lobsters (Tsang et al., 2011) in having a carapace longer than it is broad and a pleon which is only partly flexed under the cephalothorax. The phylogenetic position of Lomis has been under discussion since the taxon was first described by Lamarck (1818) as an alleged representative of porcelain crabs. In his revision, Milne Edwards (1837) placed Lomis closer to lithodid king crabs; Stimpson (1858), in turn, closer to hapalogastrid king crabs. Later, Bouvier (1895) suggested a closer relationship to the hermit crabs Mixtopagurus (Pylochelidae) and Paguristes (Diogenidae). Subsequent authors, more or less following Bouvier, placed Lomis close to symmetrical hermit crabs (Boas, 1926; Balss, 1957; Pilgrim, 1965) or as sister group to the Paguroidea (hermit crabs + king crabs) (Richter and Scholtz, 1994; Reimann et al., 2011). In the morphology-based cladistic analysis by Martin and Abele (1986), Lomis and king crabs clustered together and formed the sister group to hermit crabs. McLaughlin (1983), who revised the external morphology of Lomis, called it a ‘highly […] specialized taxon’ representing a ‘distinct evolutionary lineage’ within Anomura. She denied that it was especially close to paguroids and placed Lomis in its own superfamily taxon without phylogenetic affinities. More recent studies, however, point to a closer relationship between Lomis and Aegla (Morrison et al., 2002; Tudge and Scheltinga, 2002; Schnabel et al, 2011; Tsang et al., 2011; Keiler and Richter, 2011; Bracken-Grissom et al., 2013), which are either sister groups or cluster in a clade together with Chirostyloidea (Bracken-Grissom et al., 2013; but see also Reimann et al., 2011). Aegla – which shares the squat lobster habitus – was traditionally classified as belonging to Galatheoidea (Henderson, 1888 (as Galatheidea); Ortmann, 1892 (as Galatheidea); Balss, 1957; Martin and Abele, 1986; Martin and Davis, 2001). In more recent analyses, the non-monophyly of the traditional Galatheoidea has been recognized and Aegla and the newly discovered squat lobster-like Kiwa species (MacPherson et al. 2005; Thurber et al. 2011) were excluded from Galatheoidea and each transferred to their own superfamily taxon (McLaughlin et al., 2007; DeGrave et al., 2009). Recently, Chirostylidae was also excluded from the remaining Galatheoidea and is now combined with Kiwa in the newly erected taxon Chirostyloidea (Schnabel et al, 2011; Bracken-Grissom et al., 2013). According to recent cladistic analyses (Fig. 1), Lomis is either a sister group to Aegla with Chirostyloidea basal to this clade (Schnabel et al., 2011; Tsang et al., 2011), or sister group to Chirostyloidea with Aegla basal to this clade (Bracken-Grissom et al., 2013). The sister group relationship between Aegla and Lomis is additionally supported by morphological data (Tudge and Scheltinga, 2002; Keiler and Richter, 2011). The clade formed by Aegloidea, Lomisoidea and Chirostyloidea is herein termed Australopoda (tax. nov.). On the basis of this evidence, the most probable scenario appears to be the derivation of Lomis from a squat lobster-like ancestor representing the Australopoda ground pattern. The crab-like habitus of Lomis is thus the derived condition, while a squat lobster habitus can be regarded as plesiomorphic within Australopoda (Tsang et al., 2011). Using micro-computer tomography and computer-aided 3D reconstruction, we studied and compared the anatomy of L. hirta, A. puravida, and K. puravida in order to reconstruct the anatomical ground pattern of Australopoda and, starting from the ground pattern of Australopoda, to reconstruct the morphological transformations which, within the context of carcinization, led to the hairy stone crab L. hirta. Our findings are compared herein with anatomical descriptions and evolutionary interpretations of other instances of carcinization within Anomura.