Mar&h — North American Species of Diaptomus. 399 



the right exopodite ; in most cases this runs across the segment. 

 In all, except D. soltillimLS, there is an oblique ridge on the pos- 

 terior surface of the second segment of the right exopodite. In 

 all, except D. purpureus, the terminal segment of the exopodite 

 of the left foot ^s armed with a digitifomi proct.s and a ^len- 

 der articulated spine ; this spine in D. saltillinus is curved. The 

 fifth feet of the females, too, resemble each other. In all, the 

 exopodites are three-segmented. The endopodites are short, 

 and in all, except D. saltillinus, are armed with two rather 

 prominent spines. The female abdomens, with the exception of 

 D. asymmetricus, resemble each other; in D. dorsalis and D. 

 purpureus the first segment is especially long and slender, but 

 it is stouter in D. saltillinus and D. albuquerquensis. In D. 

 asymmetricus the general form is like that of D. purpureus, 

 but there is the marked peculiarity of the lateral process. 



D. albuquerquensis J, D. dorsalis and D. saltillinus differ from 

 the other species of the group in the peculiar dorsal processes 

 of the fifth thoracic segment. 



I have called the group the albuquerquensis group, not be- 

 cause there is any reason to think that this species is the most 

 primitive, but because it was the first described. There would 

 seem to be little question of the close affinity of all the members 

 of the group. D. saltillinus differs more widely than do the 

 others. D. saltillinus, too, is the most northern species, the 

 others being distinctly southern, D. albuquerquensis being 

 found in Colorado, New Mexico and Mexico, and D. purpu- 

 reus and D. asymmetricus in Cuba. 



It is hardly possible, with the present knowledge, even to 

 guess at the phylogenetic relationships of the group. 



D. sanguineus and D. Eiseni I am not prepared to locate, 

 even tentatively. It should be noticed, perhaps, that D. albu- 

 querquensis, dorsalis, saltillinus, Trybomi, sanguineus and sig- 

 nicauda all have a pronounced dorsal process or hirnip. This 

 may indicate some relationship, but it does not seem clear 

 enough to lead to a grouping of these species. 



Of the groups I have formed, it seems to me that the tenui- 

 caudatus group is the nearest to the primitive form; it does 



