Marsh—North American Species of Diaptomus. 395 
segmented exopodite. D. shoshone by its size is separated from 
the rest of the group, but its structural relationships seem to be 
very close. 
D. minutus is most widely distributed, being found from th a 
northern United States to Greenland i-nd Iceland, but not on 
the Eastern Continent. D. Ashlandi and D. sicilis, so far as 
known, are limited to the northern tier of states in the United 
States. D. Birgei and D. siciloides belong to warmer waters 
but probably do not occur south of the Ohio river, while D. 
shoshone is peculiar to the mountain region of the West. D. 
tenuicaudatus is a recent find, and is, so far, reported from only 
one locality, in the Saskatchewan region. D. sicilis and D. 
Ashlandi are distinctly lovers of cold water. In Green lake D. 
sicilis occurs in the winter months and D. minutus in the sum¬ 
mer months. (Marsh, ? 97, Marsh, ’03.) D. tenuicaudatus is 
considered the most primitive form because of the slender female 
abdomen without armature, the nearly equal length of the 
male fifth feet, and the two-segmented endopodite of the 
left fifth foot of the male. D. sicilis is the most nearly re¬ 
lated to D. tenuicaudatus. D. shoshone comes very close to 
it, but if it is in this line it must have been subjected 
to peculiarly favorable circumstances of food to have developed 
such an enormous size. D . Ashlandi might easily have been de¬ 
rived from D. tenuicaudatus, but the separation must have 
taken place at a comparatively remote time. D. siciloides and 
D. Birgei are somewhat more specialized forms from the same 
stock as D. sicilis. D. minutus, according to the standard I 
have set up, is the most specialized of the group; there is a 
marked difference in the lengths of the fifth feet of the male, 
and there is a striking reduction of the endopodites in the fifth 
feet of both sexes. The only noticeable primitive character is 
the three spines of the second segment of the exopodite of the 
fifth feet of the female. From its wide distribution we might 
well think of D. minutus as an early form from which the 
others have been derived, were it not for its specialized charac¬ 
teristics. As it is, we must think of it, perhaps, as not derived 
from D. tenuicaudatus, but as having a common ancestry with 
