AQUATIC INSECTS IN NEW YORK STATE 265, 
quired an unusually careful examination of the nymphs of the 
two species to detect any differences whatever. I find only the 
following, and am not assured as to how constant these will 
prove. (7’ The front border of the median lobe of the labium in 
externusS is Slightly but distinctly convex [see Bul. 47, pl. 20, 
fig.12];in fraternus it is straight or scarcely convex. 
This is probably a good character. (2) The lateral spines of the 
ninth abdominal segment are something more than two times 
the length of the tenth segment in externus; in frater- 
nws they are something less than two times the length of the 
tenth segment. 
Gomphus borealis 
I now regard this as a species, and not a variety of G. de- 
acriptus. I described it as a variety because I was unable 
to give any satisfactory character for the separation of females; 
but a more careful examination of my specimens reveals differ- 
ences in the form of the hind margin of the occiput, that appear 
to be reliable. Inthe female of borealis the occipital mar- 
gin viewed from the front exhibits two low, rounded lobes, be- 
tween which is a depression whose curvature is exactly the re- 
verse of the convexity of the lobes. In the female of descrip- 
tus the occipital margin is scarcely bilobed, and the median 
half of it is straight, not concave in the center at least. Com- 
paring the nymphs again, descriptus nymphs from Ithaca 
with borealis nymphs from Saranac Inn, I find that the 
lateral spines of the sixth abdominal segment in borealis 
are half as long as those of the seventh segment; in descrip- 
tus, about equal in length on these segments. 
GOMPHURUS 
This subgeneric name I have used as a convenient designation 
for the dilatatus group of Selys. Mr Williamson hag in- 
cluded some of its species in Stylurus!, but I prefer as yet to 
restrict that name to the very homogeneous group of species 
whose separateness was indicated by Hagen in the Transactions 
- 
1Amer. Ent, Soc. Trans. 27:205-17. 
