differences between the individuals are explainable only on the theory that they are due to inborn variations. The extent to which the individuals of a nest or litter differ in powers and qualities is often quite remarkable, and in no respect is this difference more remarkable than in fighting power. This power in dogs and fowls, as in the moose, depends on an immense number of structures, in fact, on almost all the structures of the body, and therefore, when one of a litter or brood exhibits a marked superiority over his brothers in fighting power, we must suppose that his superiority results from a favourable co-adaptive development, due to inborn variations alone, of all the structures concerned in fighting. Now if the ability to secure mates, and therefore descendants, depends largely on the fighting power, as it actually does among so many wild creatures, including the canidæ, the gallinacei, and the cervidæ, are not the differences due to inborn variations alone, which, as we have seen, are considerable, sufficient to lead, by survival of the fittest, to the evolution of the fighting power, i.e. to the co-adaptive evolution of all the co-ordinate structures on which that power depends? Would a race descended from the best fighters in each litter or brood differ from a race not so descended, from a race as regards which nature has exercised no selection in this respect? I suppose that no one would maintain that it would not differ, for to maintain such a proposition would be to maintain that a race descended from the biggest and strongest individuals alone would not differ from a race the ancestry of which included smaller and weaker individuals. Yet here we would have just such a case of co-adaptive evolution resulting from the accumulation of inborn variations alone, and involving nearly all the structures in the body, as Mr. Spencer has declared is impossible.
A thousand similar examples can be found. For
