Crangon and Galathea. 53 



DIVISIONS OF CRANGON. 



Among the species agreeing in these general characters, types of variations 

 are found in the form of the anterior tooth of the carapace (rostrum), in the 

 characters of the surface of the carapace, and in those of the surface of the ab- 

 dominal somites. "With these are conjoined less generally appreciable, though 

 possibly more important characters, derived from the comparative development 

 and proportions of certain of the cephalo-thoracic somital appendages, which 

 point to a further subdivision of the genus — such, indeed, as was effected long 

 since by Leach and Eisso, but rejected by subsequent authors. 1 am well 

 aware of the evil of carrying generic division too far ; but when, as here, we 

 find a number of sets of characters, any one of which enables us to distinguish 

 a group of animals, these groups, no matter which set of characters is chosen, 

 -being, as compared type with type, identical as to species contained beneath 

 them, and as to numbei', surely the principles of a natural classification are not 

 violated by giving distinct names to groups thus marked out, and forming from 

 the wreck of one genus several. It should never be forgotten that, whether we 

 believe genera to be actual existences or not, in practice the intention of them 

 is to enable an author, by the use of a single word or initial letter, to express an 

 assemblage of common characters, and thus avoid, in the account of individual 

 species, the repetition of characters common to a number ; and that the use of 

 the word " or," in a generic description, shows one of two things — either that 

 genera are confounded, or else, as is more often the case, that the characters 

 given are specific, and not generic. 



Keeping this canon in view, let us then examine the characters in which the 

 most fixed variation prevails among the members of the genus Crangon as con- 

 stituted by M. Edwardes, Bell, Dana, Stimpson, and others ; and I hope to 

 establish that at least three subdivisions are called for, of which, Cra7igon vul- 

 garis of Fabricius, Pontophilus spinosus of Leach, and yEgeon loricatus of 

 Risso, are the representatives ; and although to some the distinctive characters 

 given may seem unimportant when we look to the whole Decapoda, yet, taking 

 into consideration their constancy in Crangonidas, it must be acknowledged that 

 they have at least convenience of definition and diagnosis of species on their 

 side, and ought therefore to be adopted. 



VOL. XXV. I 



