368 
GAMMARID/E. 
confounded together ; thirdly, from the misapplication 
of clear descriptions or figures to different species ; and 
fourthly, from a total disregard of the rules of nomen- 
clature on the part of subsequent writers who have 
endeavoured to revise the names of these animals. 
The genus, even in its restricted state, has need of 
subdivision with a view to greater facility in the deter- 
mination of the species, and which has been attempted 
with greater or less success by various carcinologists. 
M. Milne Edwards, in his classical work upon the 
Crustacea, after the removal of those species which have 
the eyes of a circular form, divides the species into — 
A. Those which have the first three segments of the 
tail straight, and not produced in the middle of the 
hind margin into a tooth ; these are subdivided into, 
a , those with spines on the fourth and fifth segments of 
the tail (G. locusta , fluviatilis } marinus , not of Leach; 
Olivii, affinis, & c.) : and, a a, those which have no 
spines on the hinder portion of the tail (G. pulex, 
Ermanni, Othonis , pinguis , &c.) ; whilst the second di- 
vision, B., comprises those species which have the third, 
and generally the first and second also, armed with a 
strong tooth in the middle of the hind margin of each, 
G, Sabinii , mucronatus , and appendiculatus. According to 
this arrangement our common fresh-water species, found 
in rivulets, belongs to the section A. a ., but the only 
fresh-water species which M. Milne Edwards introduces 
into that section is regarded by him as identical with 
the insect figured by Roesel, which has a row of strong 
teeth along the back, and which M. Gervais, who first 
identified the species, described under the name of 
G. Roeselii . We presume this must be an error on the 
part of M. Milne Edwards, and are accordingly inclined 
rather to consider our common English species to be 
