and. West Texas. The gall of expositor (figs. 301–302) is only slightly different from that of acraspiformis, but the expositor gall is indistinguishable from that of alaria which occurs further to the north. One might collect from West Texas thru New Mexico and into Colorado without realizing he was collecting more than one species, and yet the insect of alaria (fig. 341) is subapterous and was described by Weld (1922) as a good Acraspis. The conclusion seems inevitable. The long-winged acraspiformis and expositor, and the short-winged apache, alaria, calvescens, and villosa are very close relatives, and the short-winged species must have originated directly from long-winged stocks which are still represented in the center of origin of the group (see pp. 66 to 67).
As another instance of the utilization of gall characters, we may cite the connection of the long-winged Cynips nubila with the short-winged pezomachoides. Nubila, occurring in the Southwest, produces a large, wool-coated gall (fig. 299) which is superficially as different from the small, naked, faceted gall of pezomachoides (fig. 312) as one might conceive. However, among the close relatives of nubila is the long-winged acraspiformis which we have just shown is close to the short-winged alaria and villosa. If reference is made to the figures of the details of gall structures of these species, one may find an interesting transition from the galls of the long-winged nubila (fig. 325), expositor (fig. 326), and acraspiformis (fig. 330) to the galls of the short-winged prinoides (fig. 327), erinacei (fig. 328 and 331) and macrescens (fig. 329). The last two of these are naked, faceted galls of the pezomachoides type. Further consideration of the plant tissues which enter into these gall structures (pp. 40 to 43) shows that the same elements are involved in all these galls, and that these elements are so developed nowhere but in the subgenus Acraspis of the genus Cynips. Thus even such superficially diverse galls as those of nubila and pezomachoides evidence the close affinities between short-winged and long-winged species of insects.
In another subgenus, Cynips guadaloupensis, insolens, and patelloides have moderately shortened wings. Weld considered these as species of Acraspis, altho he recognized that the galls are not typical for Acraspis. The three species occur on the canyon white oak (Q. chrysolepis) thruout the mountains of California. In the foothills, on the scrub white oaks
