Kinsey: Gall Wasp Genus Cynips 
311 
not always distinct, and that the other characters are not a 
good basis for distinction from typical Biorhiza. My studies 
of the type material in the National Museum show that Mayr 
was correct on the antennal count, but both he and Ashmead 
overlooked the scattering hairs on the mesonotum and the 
hypopygial spine characteristic of true Cynips. 
But rather than consider mellea the type of a distinct group, 
Sphaeroteras, I am inclined to consider it in the subgenus 
Acraspis. This relation seems attested by the shape of the 
head, the indefinite malar furrow, the characteristic 13-14 
segmented antennae, the punctation and hairs of the mesono- 
tum, the complete parapsidal grooves, the reduced anterior 
parallel and lateral lines, the lack of median groove, the punc- 
tation of the mesopleuron, and the distribution of the hairs 
on the abdomen. The gall of mellea is spherical, thick-walled, 
and without a distinct larval cell, agreeing in these respects 
with the species arida and conica of our Acraspis, even tho it 
lacks the surface spines of the other species of the subgenus. 
The relation of the short-winged variety mellea to the long- 
winged insects is uniquely verified by the discovery of a varia- 
ble variety bifurca (q.v.) which shows several intermediate 
stages between the most diverse insects of the species. 
An objection to this conclusion will be that the hypopygial 
spines of the short-winged mellea and bifurca are like the 
spines of the long-winged rather than of the short-winged 
forms of Acraspis. We are forced to the conclusion that the 
blunt spine found in these other short-winged forms is not a 
necessary accompaniment of the mutation in wing character. 
Another case bearing on this question is to be found in Acras- 
pis villosa apache , which has a plough-shaped spine and wings 
one-half the normal length. 
It is true that the variety mellea has nearly simple tarsal 
claws, instead of the toothed claws found thruout the rest of 
the genus Cynips. This difficulty is cleared up, however, in 
an examination of other varieties of mellea. The galls of the 
ten varieties are inseparable; but some of the insects (Caro- 
lina, unica, rydbergiana, mellea) show tarsal claws which are 
nearly simple; others ( crassior , compta, anceps) show claws 
that are broader basally and more nearly but not quite toothed. 
The generic significance of such a “diagnostic character” as the 
tarsal claw must be questioned when we find it variable within 
a single species. 
