CRUSTACEANS describes the extraordinary case of ' a Carcinus with a right-handed walking-leg on the left side of the abdomen." : There are two other genera of Cancridae of which examples occur in Cornwall. Atelecyclus septemdentatus (Montagu) is reported by Cocks from ' trawl refuse, old and young in the stomach of Gadus aeglefinus, Trigla lyra, Trigla hirundo, etc. ; common.' Couch says of this species : ' Common in the stomachs of fishes, chiefly Cod fishes and Rays, from the depth of 20 to 50 fathoms : they must abound at these depths, as I have found more than thirty in a single fish, and almost every Ray opened for several days in succession was found to contain them.' 2 Bell says : ' The general form of the carapace of this species is so nearly circular, as to distinguish it at first sight from all the other brachyurous crabs of our coast. The lateral margins with the front, form somewhat more than a semi-circle, and the latero-posterior margins form three sides of a nearly regular octagon. The whole circumference is fringed with hair. The lateral margin on each side is furnished with nine teeth, which are alternately a little smaller and larger ; the front is tridentate, the middle tooth being rather the largest ; the whole of the teeth are slightly denticulate. The carapace is granular.' * There is no real objection to be made to this description, but obviously Montagu, in calling the species seven-toothed, neglected the foremost and hindmost of the lateral teeth, just as Bell, in calling the front tridentate, neglects the additional tooth at each orbit, which would make the front quinquedentate. The carinate hands of the chelipeds combine with the other characters to make it to my mind tolerably certain that Couch's Canctr incisocrenatus was really a young Atelecyclus. Pirime/a denticulata (Montagu) is recorded by Cocks from ' Harbour, Gwyllyn- vase, Castle point ; not uncommon.' This little species is at once distinguished from those of the two foregoing genera by its having only five antero-lateral teeth. All three have the pleon five- segmented in the male and seven-segmented in the female, but the two last-mentioned genera are discriminated from Cancer by having the third maxillipeds produced over the epistome. By this character they lead on to the next family, the Corystidae. Corystes cassivelaunus (Pennant) is spoken of by Couch 4 as ' scarcely common. ' Cocks notes its occurrence thus : ' Sands, low-water mark, Swanpool, Penzance, Mainporth, Bream bay, etc. ; not uncommon. My daughter procured two living specimens (females) Pendower beach, low-water mark.' Bate says : ' It is common on sandy shores at low water.' * But its habit of burrowing in the sand, noticed by all our authorities, prevents its being commonly seen alive. Dead specimens are some- times thrown up on to the shore in great numbers. Its carapace, longer than broad, the dorsal markings suggestive of human features, the long slender chelipeds of the male, and the peculiar second antennae, elongate and setose, make this ' Long crab ' or ' Masked crab ' easy to recognize. Mr. Walter Garstang has pointed out the contrivances by which in various sand-burrowing species the current which bathes the branchiae is filtered, and that when a species is completely embedded a reversal of the current takes place. During this reversal, alike in Corystes and Atelecyclus^ filtration is effected ' by an inhalant sieve-tube formed by the second antennae, with the participation of the third maxillipeds.' 6 The Catometopa are divided by Alcock into nine families, with many of which we are not here concerned. The first of them, the Goneplacidae, stand so close to the border line that some authorities have ranged its typical genus, Goneplax (Leach), among the Cyclometopa. G. angulata (Pennant), the ' Square crab ' or ' Angular crab,' is reported by Couch as ' common in moderately deep water, and often in the stomachs of fishes,' 7 and by Cocks from 'harbour, trawl refuse, stomach of the Raia clavata, R. maculata, Gadus morrhua, Gadus aeglefinus, Trigla hirundo ; common.' Like C. cassivelaunus the male has chelipeds of remarkable length, but unlike the Corystes this species has the carapace broader than long, and unlike all the other crabs of our survey thus far, it has the eye- stalks elongate. J. Cranch is Leach's authority for the repeatedly quoted statement that the members of this species ' live in excavations formed in the hardened mud, and that their habitations, at the extremities of which they live, are open at each end.' 8 There is no reason for doubting Cranch, but the difficulty of such observations is indicated by the fact that apparently for about ninety years his testimony has never been corroborated. In this species the outer orbital angle forms a prominent and acute tooth, behind which there is a smaller tooth on the antero-lateral margin. G. rhomboides (Linn.) is without the second tooth, or at most has only a little tubercle in its place. Cocks records this Mediterranean species at Falmouth, ' from the stomach of the Gadus aeglefinus, Trigla lyra, Trigla hirundo, etc. ; scarce.' But he does not specify any character on which he relies for distinguishing this species from its very intimately allied neighbour. If, as many think, they are identical, the name rhomboides has the priority. A much more considerable difficulty is 1 Journ. Marine Biol. Assoc. (1895-97), iv, (New Ser.), p. 144. - Fauna, p. 74. 3 Brit. Stalk-eyed Crustacea, p. 153. 4 Fauna, p. 74. 5 ' Revision,' p. 23. 6 Brit. Assoc. Rep. for 1896. ' Fauna, p. 72. 8 Malacostraca Podofhthalmata Britanniae, text to Plate XIII (l March, 1816). 261
