It must be admitted, however, that Emery was probably right when he directed attention to the possibility of a 'conflict between germinal and personal selection.' It is quite conceivable that in cases of useful variations, that is, of adaptations, the processes of selection within the germ-plasm may lead to excessive developments, which personal selection cannot control, because, on account of their earlier usefulness, they have in the course of a series of generations and species become fixed not only in a majority of ids, but in almost all the ids of the collective germ-plasm of the species. In this case a reversal must be difficult and slow, for the gathering together of ids with relatively weaker determinants can only take place slowly, and it is questionable whether the species would survive long enough for the slow process to take effect. But, apart from the question of time, such a reduction of an excessive development would sometimes be quite impossible, for the simple reason that there is nothing for personal selection to take hold of. Döderlein has pointed out that many characters go on increasing through whole series of extinct species, and ultimately grow to such excess that they bring about the destruction of the species, as, for instance, the antlers of the giant stag or the sabre-like teeth of certain carnivores in the diluvial period. I shall have to discuss this in more detail in speaking of the extinction of species; it is enough to say here that such long-continued augmentations in the same direction can never be referred _solely_ to germinal selection, since it is hardly conceivable that a species--much less a whole series of species--should arise with injurious characters; they would have become extinct while they were still in process of arising. Although we see that the Irish stag, with his enormous antlers over ten feet across from tip to tip, was heavily burdened, we are hardly justified in concluding that the size and weight of the burden on his head tended to his destruction from the first--for in that case the species would never have developed at all--but it may well be that at some time or other the life-conditions of the species altered in such a manner that the heavy antlers became fatal to it. In this case the variation-direction which had gained the mastery in all ids could no longer be sufficiently held in check by personal selection, because the variations in the contrary direction would be much too slight to attain to selective value. Sudden, or at least rapidly occurring changes in the conditions of life, such as the appearance of a powerful enemy, exclude all chance of adaptation by the slow operation of personal selection.