BRACHYURA. 73 



related genus Eriphia, is disregarded, and species of different natural 

 groups are brought together. It cannot be said on any ground that Rup- 

 pellia and Eudora are synonymes, and the latter name can be sustained 

 only by sustaining also the system of De Haan. Again, the genera 

 Xantho, Liagore, and Galene are described as having the inner branch 

 of the first maxillipeds terminate in a short dilatate triangle. If we 

 take Xantho with its typical species, and trace the genus through its 

 range, we find it passing into narrower forms, of the same essential cha- 

 racters, (though referred by us to Paraxanthus,) in which this triangle is 

 narrow-oblong instead of short-dilatate, and closely like that of Pilum- 

 nus, which is narrow-trigonal. Indeed, it is found that the form of this 

 branch varies directly with the breadth of the species, and is equiva- 

 lent in value, as a generic characteristic, to the breadth of the species, 

 and of no value at all in itself. The same variations take place in 

 Leach's Chlorodius as in Xantho. The character given for Pilumnus 

 would, therefore, include true Xanthos or Paraxanthi, and also, true 

 Chlorodii. Thus the true limits of groups are not defined, and per- 

 plexing ambiguities meet one at every step. Milne Edwards's system 

 left the press the following year after the publication of De Haan's 

 synopsis of genera; and it is obvious that no amount of study could 

 have enabled him to comprehend all the genera of De Haan, so as not 

 to have duplicated them in his own work. Such duplications actually 

 took place, and if the names of either author are to be retained, 

 science would most justly award the honour to him who characterized 

 them so as to be recognised by others. Still, it should be remembered 

 that the science is vastly indebted to De Haan for his researches. 

 He has developed many important distinctions. It is of much inte- 

 rest to know, that while the broad Cancroidea have generally the 

 inner branch of the first maxillipeds broad-triangular, the species of 

 the genus Cancer, which are remarkable for their breadth, have the 

 same branch narrow-triangular; for it shows a correspondence with 

 the front and inner antennary areas, which are also narrow, the latter 

 so much so, that the antennae are lonc/itmlmal, while transverse in all 

 other Cancroidea. The divergence of the genus Cancer from the other 

 genera of Cancroidea, is thus rendered more apparent. 



Among the Portunidae, other discrepancies between species and the 

 generic characters laid doAvn in De Haan's system, may be briefly 

 alluded to. Amphitrite is said to have the third joint of the outer 

 maxillipeds short and oblique. This genus is a subdivision of Lupa. 



19 



