254 
[October 
CORDULIA. 
C. ? molesta n. sp. 9 Pale dull brownish olive, both in the living and in the 
dried specimen 5 almost immaculate. Head pilose; antennal seta brown; mouth 
and all beneath paler; behind the tubercle of the eyes some obscure yellow 
markings. Thorax pilose; dorsum of thorax with a much darker broad dorsal 
stripe, shading otf into the ground-color half-way to the humeral suture; dor¬ 
sal Carina yellowish. Pleura with a yellowish stripe on the spiracle, obsolete 
Gomphus, seems to indicate that in the entire genus Gomphus the length of the 
legs is a character not of subgeneric, but merely of specific value. For exam¬ 
ple, calculating from M. Selys’ own measurements, in G. dorsalis the hind 
wing is 4.60 of the hind femur, in intricatus % 4.58, in simillimus % 4.51, in pul- 
chellus "^4.33, in lividus % 4.13, in graslini % 4.06, \u pilipes % 3.94, \n flavipes % 
and militaris 3.93, in occipitalis % and Kurilis % 3.87, in vulgatissimus 3.86. 
in spicatus % 3.85, in melcenops % 3.80, in minutus 3.73, in adelphus % 3.33, in 
dilatatus % 3.20, in fraternus % 3.11, in spinosus % and armatus which are 
grouped separately from all the others except spoliatus as having “ excessively 
long hind femora,” no less than 3.04 or only .07 less than fraternus % ! and 
finally in externus % 3.00 and in spoliatus % 2.80, the last species but one being 
grouped as having “ hind femora of the ordinary length,” and yet actually 
having the longest hind femora of any of them, with the single exception of 
spoliatus ! — ird. As to the character drav’-n from the very remarkable armature 
of the 9 occiput in Ophiogomphus, {Mon. Gomph. Plate V. fig. 2. h), M. Selys 
himself, in the earlier part of his great work, laid it down as one of the cha¬ 
racters of the subgenus Macrogomphus, that the 9 “ had a ]3rotuberance on 
the middle of the occiput,” (p. 87,) but he subsequently allowed that “ this 
character was only a specific one.” (p. 428.) Are not the horns of the occiput 
in 9 Ophiogomphus likewise only of specific value ? We see that in the foreign 
subgenus Onychogomphus one species {Cerastes) has horns on the 9 occiput 
{Mon. Gomph. Plate IV. fig. 2, b.); yet it is not on that account placed by M. 
Selys in a different subgenus from the other 13 species which have no occipital 
horns.— ith. Considerable stress is laid, especially by Dr. Hagen, upon the fact 
that Ophiogomphus has a tooth upon the second joint of the penis, while Erpe- 
togomphus has none. But the most recent researches of Dr. Hagen have shown 
that some species of the foreign subgenus Onychogomphus have this tooth and 
some have not. {Mon. Gomph. p. 429.) It would seem, therefore, that neither 
can this character be of subgeneric value in Gomphus.— bth. It is said that the 
vesicle of the vertex is divided into two tubercles in Erpetogomphus, and is 
simple and normal in Ophiogomphus. But my Erpetogomphus rupinsulensis 
has the vertical vesicle “ scarcely emarginate.”— ^th. The comparative length 
of the two terminal abdominal joints is not, in reality, sufficiently different in 
in the two subgenera to be relied on as a distinguishing character. M. Selys 
stated originally that in Erpetogomphus % 9 the 10th joint was equal to the 
9th. {Mon. Gomph., table, p. 14 and p. 69.) In his latest word on the subject 
he says that in Erpetogomphus the 10th joint is equal to the 9th, but that 
