The evidence is all in favor of believing that direct mutations have occurred as the result of modifications that must ultimately be explained in terms of the physics and the chemistry of genes.
A more detailed genetic interpretation of our material must for the most part be postponed for a later paper; but we may point out that the graded series of individuals, obtained when short-winged species hybridize with long-winged stocks as with Cynips bifurca and several species in other genera on which we shall publish later, indicate that wing characters in these insects may be dependent on multiple factors or perhaps on more than one group of such factors. It also appears possible that other structural peculiarities regularly associated with wing reduction may result from the same mutations of one or two genes in groups of linked genes responsible for wing characters. A single gene mutation in a single generation of insects might then give rise to a very distinct cynipid. Such radically new species are usually placed in distinct genera, and this probably explains why systematists have so often failed to believe that mutation accounts for the origin of species in nature. When the mutations are slight, they pass as products of Darwinian variation. There is, apparently, need of a revision of taxonomic procedure in the light of genetics data.
In conclusion, attention should be drawn to the interesting case of Cynips bifurca, a variable-winged species which we have from only two localities, one in southern Mississippi and one in southern Georgia. Both of these stations, however, are located well within the range of Cynips anceps (fig. 50). The galls of bifurca and anceps are identical (figs. 294–295). The insects have the same hypopygial spine (figs. 389–390), and a peculiar tarsal claw (fig. 350) found nowhere else in the genus except among a few of the close relatives of anceps. The figures of the wings (figs. 357–360) and of the whole insects of bifurca (figs. 338–339) will show that there are two distinct types of wings involved: one which is uniformly reduced, and the other a truncate wing more like the laboratory mutants called “truncate” in Drosophila. The significant thing about the bifurca series is the occurrence of intermediate individuals with wing-body ratios ranging between 0.27 and 0.54, body lengths varying from 2.2 to 3.3 mm., and
