16 THE VOYAGE OF H.M.S. CHALLENGER. 



in many instances — especially in the true deep-sea species — the eyes are wanting (a 

 matter to he discussed hereafter), it never happens that the tuhercle has totally dis- 

 appeared. Most genera have this tuhercle placed nearly in the middle between the two 

 ovigerous legs ; but in some genera [Phoxichilidium, e.g.) it is situated much nearer 

 the front of the segment. 



The thoracic part of the cephalothorax and the three following true thoracic segments 

 are furnished with lateral processes for the insertion of the legs ; these lateral processes 

 in the diiferent genera, and even in different species of the same genus, are of very differ- 

 ent lengths. The segments of the body themselves are also of very different lengths. 

 There are extremely slender forms with long segments and widely separated lateral 

 processes, and there are also forms so highly concentrated, that the lateral processes 

 are not separated at all ; and between these extremes, which are often met with in one 

 and the same genus, numerous intermediate forms are to be observed. The dorsal 

 surface of the body is either smooth or furnished with knots, spines, strong prickles, &c. 



At its extremity, between the two lateral processes for the insertion of the last 

 pair of legs, the last thoracic segment has a rudimentary abdomen of varying length, 

 which is sometimes {Colossendeis, e.g.) connected with the segment by means of an 

 articulation, and also sometimes {Zetes, Kroyer) shows traces of being divided into 

 two segments.' At its extremity the anal aperture is found. 



The legs begin at the ends of the lateral processes ; they are eight-jointed. For the 

 joints I retain the names proposed by Johnston ; these names are the same as those used 

 in entomology, but it is evident that in this case identity of name does not necessarily 

 go along with identity of meaning ; neither analogical nor homological comparison is 

 meant by it. 



The first three (the coxal) joints are as a rule very short ; the following three, the 

 thigh and the two tibial joints, are much longer (the second tibial being in most cases the 

 longest of all). The two tarsal joints are again a great deal shorter. The first tarsal 

 as a rule is shorter than the second ; in many instances it is even extremely small, its 

 function then being only to furnish a highly movable articulation to the last joint of the 

 leg. At its extremity the last joint is furnished with a claw, which is, or is not, accom- 

 panied by two accessory claws.^ In some genera {Colossendeis, e.g.) accessory claws are 

 never observed, whUe in other genera (NympJion) they occur in some species and are 

 wanting in others. Therefore the presence or absence of accessory claws alone should 

 not be made use of in establishing new genera. 



' Ehynchothorax viediterraneus, Cos., Microdoride mediterranea, Napoli, 1861, has a seven-joiuted abdomen (Addome 

 angusto e brevissimo di 7 articoli). 



2 I think there is not a single reason for calling this claw a ninth joint. At any rate the homology of the claw with its 

 accessory claws is much greater than that between the claw and the joints of the leg, and, therefore, if the claw is con- 

 sidered as a ninth joint in those cases where accessory claws are observed, we must speak of a joint ha^•ing two lateral 

 joints close to its origin, which would be absurd. 



