The whole criticism is tersely conveyed in the form of two sequent propositions--namely, "Every species which exists, exists in virtue of adaptation; and whatever accounts for that adaptation accounts for the existence of the species." My answer is likewise two-fold. First, I do not accept the premiss; and next, even if I did, I can show that the resulting conclusion would not overturn my definition. Let us consider these two points separately, beginning with the latter, as the one which may be most briefly disposed of. I. Provisionally conceding that "every species which exists, exists in virtue of adaptation," I maintain that my definition of the theory of natural selection still holds good. For even on the basis of this concession, or on the ground of this assumption, the theory of natural selection is not shown to be "_primarily_" a theory of the origin of species. It follows, indeed, from the assumption--is, in fact, part and parcel of the assumption--that all species have been originated by natural selection; but why? _Only because natural selection has originated those particular adaptive features in virtue of which (by the hypothesis) species exist as species._ It is only in virtue of having created these features that natural selection has created the species presenting them--just as it has created genera, families, orders, &c., in virtue of _other_ adaptive features extending through progressively wider areas of taxonomic division. Everywhere and equally this principle has been "primarily" engaged in the evolution of adaptations, and if one result of its work has been that of enabling the systematist to trace lines of genetic descent under his divisions of species, genera, and the rest, such a result is but "secondary" or "incidental." In short, it is "_primarily_" a theory of adaptations _wherever these occur_, and only becomes "_also_" or "_incidentally_" a theory of species in cases where adaptations happen to be restricted in their occurrence to organic types of a certain order of taxonomic division. II. Hitherto, for the sake of argument, I have conceded that, in the words of my critic, "it is a necessary consequence of the theory of selection that every species must have some one or more structural or functional peculiarities" of an adaptive kind. But now I will endeavour to show that this statement does not "follow as a necessary consequence" from "the theory of selection."