16 A REVISION OF THE AsTACIRi:. 



It has been noted already that in the young stages of Cambarus and Asta- 

 cus the first abdominal appendages are wanting, and the telson is not divided 

 by a transverse suture. In these respects the young stages agree with the 

 adult condition of the Parastacinae. On this account, and on account of 

 the more generalized arrangement of the gills in the Parastacinae,* it would 

 seem that the Potamobiinae are a more specialized type than the Parastacinae. 

 Further, among the Potamobiine crayfishes the genus Cambarus will hold the 

 highest place by reason of the complete suppression of the pleurobranchise 

 and the high degree of specialization exhibited in the first abdominal appen- 

 dages and hooked thoracic legs of the males and the annulus ventralis of the 

 female. In the abnormal female of Cambarus Diogenes, noticed on page 13, it 

 is clearly demonstrated that an organ essentially like the first abdominal 

 appendage of Astacus is the first step in the transformation of the common 

 type of abdominal limb into the male appendage of Cambarus. The three 

 species of Cambaroides (p. 126) form a passage from Cambarus to Astacus 

 proper. 



Division of the Species of Cambarus into Subordinate Groups. — Girard divided 

 the Cambari into three groups, based upon the form of the rostrum and the 

 anterior pair of abdominal legs in the male. Dr. Hagen also concluded that 

 the genus comprehended three well-defined groups, but he based his division 

 of thi' genus upon the number of hooked thoracic legs in the male, taken 

 in connection with the shape of the rostrum (whether toothed or toothless). 

 It follows that the groups of these authors do not exactly coincide ; e. g. 

 C. pellucidus falls into the same assemblage of species with C. affinis, C. rusti- 

 cus, etc., in Girard's system, whereas by Ilagen's mode of division it is asso- 

 ciated with C. Blandingii and allied species. It seems to me that we gel a 

 more natural grouping of the species by taking the number of hooked tho- 

 racic legs in connection with the structure of the first pair of abdominal 

 appendages of the male as the basis for the division of the genus, without 

 reference to the form of the rostrum. Any division based wholly or in pari 

 upon the presence or absence of lateral rostral spines will divorce species 

 which in the totality of their organization are most closely related. If it be 

 urged againsl this mode of division that it implies a knowledge of the pecu- 

 liarities of one sex. 1 reply that the same objection applies to the groups of 

 Girard and Hagen, which are based in part upon characters found only in 



* In all the Parastaeinre excepts ; it, there are four pleurobranchise devi 



and the whole nber of gills is twenty or twenty-one. 



