ig22.J S. Kemp: Notes on Crustacea Decapoda. 137 



species with a conoidal cornea do not necessarily form a natural 

 group. 



Elsewhere in the genus Periclimenes other modifications of the 

 eye are sometimes found. In P. seychellensis there is a papilla on 

 the eyestalk and in two of the three species of Periclimenaeus the 

 cornea has a circular cup-shaped depression. The evidence we 

 possess at present tends to show that the structure of the eye, 

 when unsupported by other characters, does not afford a valid 

 basis for subgeneric division. 



For these reasons I am unable to accept the subgenera pro- 

 posed by Borradaile. I recommend instead an arrangement in 

 which the primary division is based on the structure of the dactyli 

 of the last three pairs of peraeopods, whether simple or with an 

 accessor lobe or claw. The structure of the dactyli in these 

 limbs is of generic importance in the more highly specialized 

 Pontoniinae and the character is of established value in other 

 Caridea. 



Whether the arrangement leads to a natural grouping of the 

 species on a phylogenetic basis, is a question that cannot be 

 answered in the present state of our knowledge. I incline to the 

 view that it does. In some species, however, the additional dacty- 

 lar claw is reduced to a mere process or lobe, 1 and there is thus a 

 possibility that certain specialized species in which the dactylus is 

 simple may have been derived from forms in which it was once 

 biunguiculate.* 



In P. scriptus, the type-species of Periclimenes, the dactyli 

 are biunguiculate, and the subgenus to which this species belongs 

 may thus be termed Periclimenes s.s. For the more primitive 

 forms with simple dactylus Schenkel's Ancylocaris may be em- 

 ployed, though in a different sense to that in which it has hitherto 

 been used. 



Borradaile's Periclimenaeus, of which his Periclimenes subgeu. 

 Hamiger is a synonym, is at most a subgenus of Periclimenes. In 

 the three known species the dactyli of the hinder peraeopods are 

 biunguiculate, thus resembling Periclimenes s.s., but the hepatic 

 spine of the carapace, which is invariably present in the latter, is 

 here absent. The chelae of the second peraeopods are more mas- 

 sive in Periclimenaeus than in Periclimenes s.s., though the species 

 of the latter subgenus exhibit a very great range of variation in 

 this respect. 



The characters of the three subgenera that I propose may 

 be summarized thus : — 



Dactyli of last three peraeopods biunguiculate or with 



an accessory process or lobe behind terminal claw. ['39- 



Hepatic spine present .. ... ... Periclimenes s.s., p. 



Hepatic spine absent ... ... ... Periclimenaeus, p. 166. 



Dactyli of last three peraeopods simple [Hepatic spine 



usually present] ... ... ... ... Ancylocaris, p. 167. 



' eg;., Periclimenes rex and P. noverca. 

 2 Of this P. frater is perhaps an example. 



