tions in every large series which we have from nearly a hundred localities well spread over the range.
The interpretation of this population has become possible thru the discovery of its closest relatives, Cynips wheeleri to the north and Cynips derivatus to the south of the area occupied by erinacei (see map, fig. 63). Erinacei occurs exactly where we might expect a hybrid of wheeleri x derivatus origin.
Wheeleri is a uniformly small insect; derivatus is large; erinacei shows every gradation between the extremes. Wheeleri is largely black; derivatus is prominently rufous over most of the body; erinacei shows every sort of combination and recombination of these characters. The mesonotum of wheeleri is largely smooth and naked; the mesonotum of derivatus is rough and more hairy than in any of the related species; the mesonotum of erinacei varies from smooth and more naked to rough and hairy, again showing every gradation and combination between the supposed parents. The galls of wheeleri are ellipsoidal, polythalamous, and uniformly spiny; those of derivatus are spherical, one- or two-celled, and uniformly naked; the galls of erinacei (figs. 312-315) show these extremes and a remarkable series of every conceivable intermediate between and combination of these extremes. Detailed descriptions of the insects are given in the systematic portion of this study. Remembering that many of these characters are probably controlled by multiple factors in heredity, erinacei appears as just that variable combination of characters which we might expect from a wheeleri x derivatus cross.
This interpretation finds confirmation in such series as the 107 insects which we have from Meadville, in the northwestern corner of Pennsylvania. In this series 41 per cent of the individuals show clear evidence of wheeleri affinities, 37 per cent are practically identical with our Alabama and Georgia material of derivatus, and 21 per cent show gradations between wheeleri and derivatus that would pass as good erinacei. It is certain that erinacei is not of present-day origin, for wheeleri and derivatus are separated thruout most of their ranges by several hundreds of miles. This distance is too great to allow any present-day hybridization of pure stocks of wheeleri and derivatus. The occurrence of the
