252 M. A. Humbert on Kiphargus puteanus, var. Forelii. 



further and wished to derive them directly from Gammarus 

 pulex. I cannot share in this last opinion, which seems to 

 me to be that of a narrow Darwinism ; and 1 think that, both 

 with respect to the Niphargus of the Lake of Geneva and to 

 those of other dark ]ilaces, there are strong arguments to be 

 brought against this theory of O. pulex becoming trans- 

 foruied into Crangonyx and Niphargus. The following are 

 the chief of these objections : — 



" 1. So far as we know at present, Gammarus index only 

 descends to a small depth from the surface ; and there is a zone 

 destitute of Garamarids, extending between the lower level 

 at which we cease to find G. jndex and the upper level at- 

 tained by the Niphargus. This fact would be very difficult 

 to explain if the Niphargus originated from G. pulex. In 

 this case, on the contrary, we ought to find representatives of 

 this latter species at all depths, and even to meet with indi- 

 viduals establishing a passage between one form and the other. 



'' 2. If the Niphargi originated from Gammarus ptulex^ and 

 had in their youth, as asserted by M. de Rougemont, the form 

 of Crangonyx suhterranens, we should find ourselves face to 

 face with facts completely opjiosed to the general laws of de- 

 velopment. We know, in fact, that the characters which 

 separate two representatives of the same group are less marked 

 in youth than in the adult state. Forms which resemble each 

 other during the first phases of their development may sub- 

 sequently diverge in a very striking manner. This embryo- 

 genie and phylogenic law is particularly verified among the 

 Crustacea, in which affinities which are strongly marked in 

 the larvfe almost entirely disappear in the adult animal. Now 

 what do we see in the Gammarids before us ? 



" In the Gammai-i proper the last pair of saltatory feet are 

 biramose ; Gammarus pulex even has the two branches nearly 

 equal. The Niphargi have these branches very unequal, but 

 both of them still exist. In Crangonyx^ on the contrary, there 

 is only a single branch. The Crangonyches^ therefore, in this 

 respect, represent a type further removed from G. piulex than 

 the N{p>hargi. We could understand, therefore, a develo^^ment 

 in which the second branch inherited from the ancestor would 

 exist during youth, and afterwards disap])ear, by atropjiy, at a 

 more advanced age — in other words, a Ni2)ha7y us-pha.se after- 

 wards attaining the state of Crangonyx. The inverse of this 

 (that is to say, a metamorphosis of the kind observed by 31. de 

 Rougemont) appears to us to be discordant with all that we 

 know of the metamorphoses of the Crustacea. • 



" The same abnormal reversal of the laws of development 

 would also be observed in the case of the telson, which is 



