The Dynomenidae are a group of small, uncommon, primitive crabs, which are often associated with corals. They inhabit depths down to around 500 m, between latitudes 40°N and 40°S. All genera and species are revised and redescribed, and the genus Dynomene Desmarest, 1823 is divided into two additional genera. As a result, there are thirteen known species belonging to five genera: Dynomene Desmarest, 1823 [D. hispida Guérin-Méneville, 1832, D. praedator A. Milne Edwards, 1879, D. pugnatrix de Man, 1889, D. filholi Bouvier, 1894, and D. pilumnoides Alcock, 1900], Hirsutodynomene gen. nov. [H. spinosa (Rathbun, 1911), and H. ursula (Stimpson, 1860)], Metadynomene gen. nov. [M. devaneyi (Takeda, 1977), M. tanensis (Yokoya, 1933), and M. crosnieri sp. nov.], Acanthodromia A. Milne Edwards, 1880 [A. erinacea A. Milne Edwards, 1880, and A. margarita (Alcock, 1899)], and Paradynomene Sakai, 1963 [P. tuberculata Sakai, 1963]. A key is provided to identify these species. In addition nine fossil genera, dating from the Upper Jurassic, are known: Stephanonietopon Bosquet, 1854, Dromiopsis Reuss, 1859, Palaeodromites A. Milne Edwards, 1865, Cyamocarcinus Bittner, 1883, Graptocarcinus Roemer, 1887, Cyclothyreus Remes, 1895, Gemmellarocarcinus Checchia-Rispoli, 1905, Glyptodynomene Van Straelen, 1944, Trachynotocarcinus Wright & Collins, 1972. Some extinct species have also been placed in the genus Dynomene. The definition of the family Dynomenidae given by Alcock (1901) is updated and expanded in order to allow fossil species to be more accurately determined. Because of overlap with the Dromiidae, there has been some uncertainty about true family affinities of some fossils. Although these genera are in need of revision, this is not undertaken in this paper. The status of Dynomene pilumnoides is established as a valid species, D. pugnatrix brevimana Rathbun. 1911 is synonymized with D. pugnatrix de Man, 1889, D. granulobata Dai, Yang & Lan, 1981 is a synonym of D. hispida, while D. sinensis Chen, 1979, D. tenuilobata Dai, Yang & Lan, 1981, and D. huangluensis Dai, Cai & Yang, 1996 are all synonyms of D. praedator. Dynomenids are reported from Australia for the first time in D. pilumnoides, and Hirsutodynomene spinosa. The status of Metadynomene tanensis (Yokoya, 1933) is established as a widespread Pacific species and shown to be part of the fauna of Japan, where it has been confused with D. praedator. Paradynomene tuberculata, previously known from Japan and New Caledonia, is now recorded from the Gulf of Aden, Indian Ocean. P. tuberculata as well as D. praedator and H. spinosa, are reported from Guam. The Atlantic Ocean and the Indo-Pacific share genera of dynomenids but not species. The biogeographic history of dynomenids is interpreted in the light of their present distribution and in relation to plate tectonics. Ancestral dynomenids are assumed to have been tethyan crabs and D. filholi and Acanthodromia erinacea, two insular Atlantic species, are shown to be tethyan relicts. By contrast, Hirsutodynomene ursula from the eastem Pacific, seems to be a species of quite recent origin. In redescribing the species particular attention is paid to some new characters: setae, gills, epipods and gill cleaning mechanisms, the subchelate structure of the last pereopods and the male pleopods. This work was undertaken using a scanning electron microscope. Differences in the gross appearance of setae can be used to separate species and there are substantial differences in setal structure at the microscopic level. The standard branchial formula for dynomenids is shown to be nineteen gills plus seven epipods. There is little variation in gill numbers but substantial variation in gill shape between species. Although dynomenid gills are often said to be "transitional" they are arranged as in phyllobranchs but with the epibranchial part divided into varying numbers of lobes which gives them a trichobranch-like appearance. Acanthodromia has gills which are almost identical to the phyllobranchs of the Dromiidae but which retain the "dynomenid notch" on each side which, in cross section, give each gill plate a violin shape. The gill cleaning mechanism in dynomenids is complex, being carried out by no less than eight appendages (long setae on the posterior margin of the scaphognatbite and the seven epipods) as well as stiff setae on the posterior hypobranchial wall of the gill chamber. In eubrachyurans only three appendages (maxillipodal epipods) are used. In dynomenids the last pereopod is very reduced (on average less than one-third the length of the fourth pereopod) and carried in a horizontal position alongside the posterolateral carapace margin above the base of the preceding pereopod. They are not, as it has been commonly described, carried subdorsally. Using a scanning electron microscope it was revealed that this limb is sexually dimorphic: in males the dactyl has the normal shape of a tiny claw, but in females the dactyl is a flattened plate, bearing five to sixteen spines which are opposable to an extension of the propodus. In both males and females the propodal extension is armed with spines but in Hirsutodynomene. Metadynomene and Paradynomene, females have a significantly larger number of spines, which are armed with tiny teeth. Males of three species have an additional small spine on the outer margin of the dactyl. This is a character, previously only known amongst the Dromiidae, which suggests that the last pereopod of dynomenids may have evolved from a camouflage-carrying limb. This limb appears to be vestigial and it is difficult to know what its function may have been amongst the dynomenid ancestors. However its most likely former role appears to be as a cleaning appendage, but certainly not for carrying pieces of camouflage as it is found amongst the dromiids and homolids. All dynomenids, except Acanthodromia, lack an effective abdominal locking mechanism and both sexes have five pairs of pleopods. The female has vestigial, uniramous first pleopods followed by four pairs of normal biramous pleopods, while the male has the normal first two pairs of pleopods as well as three pairs of rudimentary pleopods on segments three to five. These rudimentary pleopods can be uniramous or bifid. In Metadynomene tatiensis 17% of females were gynandromorphs with small male first pleopods but the remaining pleopods were normal. The diet of dynomenids seems to consist of food obtained by sieving fine sediment or perhaps coral mucus. The bunches of stiff setae on the inner margins of the cheliped fingers and third maxillipeds are probably used to separate fine organic fragments. Most of their gut contents are unidentifiable soft organic material along with small amounts of chopped chitinous fragments perhaps coming from hydroids or other crustaceans. Dynomenids appear to be deposit feeders. Dynomenids have a broadcast reproductive strategy, with indirect development, laying small eggs (mean diameter = 0.49 mm) which probably produce planktonic larvae. Dynomenid larvae have never been reported in plankton samples. Males are on average 19% larger than females which become sexually mature at 5-8 mm CW for small species, or 9-13 mm CW for large species. Egg numbers increase logarithmically with body size. Given the sister group relationship with homolodromiids (which have very abbreviated development) it is implied that dynomenids and dromiids evolved from ancestors which had large eggs and perhaps a brooding strategy. This conclusion is contrary to accepted wisdom, but it is the most parsimonious answer. Some dromiids have retained the brooding strategy but others have independently evolved a broadcast strategy. The evolution of such a strategy in both these families is probably related to their colonization of the shallow water habitat. Both dynomenids and dromiids are mostly crabs of the continental shelf whereas homolodromiids are crabs of the continental slope. Using morphological characters the phylogenetic relafionships of the Dynomenidae are examined. Both the Dynomenidae and the Dromiidae are monophylefic, sharing significant apomorphies. The resemblance of some dynomenids and dromiids is shown to be the result of convergent evolution within these families. The Homolodromiidae are also monophyletic but are defined almost exclusively by plesiomorphies. Monophyly of the Dromiacea de Haan, 1833 is supported by morphological characters with the Dynomenidae and Dromiidae together being the sister group of the Homolodromiidae. The ancestor of these three families was probably a camouflage carrying crab, using both of the last two pairs of pereopods. A controversial aspect of the sister group relationships of the dromiaceans is the need to assume that in dynomenids the fourth pereopod has reverted to a locomotory role and the fifth pereopod became a cleaning limb. Monophyly of the Podotremata Guinot, 1977 is also supported. This analysis suggests that camouflage-carrying behaviour has evolved independently in the Dromiidae (and probably in the Homolodromiidae) and the Homolidae. Dromiids carry pieces of sponges or ascidians as well as shells, using the last two pairs of pereopods, while homolids carry sponges or anemones, using only the last pair of pereopods. The ancestor of the Dromiacea and Archaeobrachyura was probably an inhabitant of deeper waters and not a camouflage carrying crab. |
