The appearance of fixity which most species have is, therefore, illusory; in reality they exhibit a slow flux, gradually setting aside the superfluities they received from their ancestors, perfecting the important parts to more precise adaptation and greater functional capacity, and at the same time endeavouring to maintain all the parts in constant harmony. We can understand that as long as this process of gradual perfecting goes on, amphimixis will not readily be given up. Those that retain it must always, in the long run, have the preference. Moreover, as we have seen, _it cannot be given up_, when it has existed through æons, because of the power of persistence which the germ-plasm has gradually acquired in the course of such a long hereditary succession. It could only be given up if an advantage decisive as to survival were associated with its abandonment, such as can be actually recognized in most cases of parthenogenesis, among animals at least. In my opinion this indirect effect of amphimixis, that is, the increasing of the possibilities of adaptation by new combinations of individual variational tendencies, is the main one, while the direct nutritive effect of the two germ-cells upon one another is quite subsidiary. In this opinion I find myself in opposition to the views of many if not most naturalists, who assume that amphimixis has a direct, sometimes, indeed, only a direct effect, and believe that they can prove it by facts. In support of this position it has been pointed out that allogamy, that is, the mingling of individuals of different ancestry, occurs even among lowly unicellulars, and then higher up among most organisms; but the question has not been asked whether this mutual attraction of the unlike really expresses a primary characteristic of organisms, and may not possibly be a secondary acquisition adapted to ensure the occurrence of amphimixis. If we examine the facts we find that even in the lowly Algæ, such as _Pandorina_ and _Ulothrix_, only the migratory cells or swarm-spores of different cell-colonies conjugate with one another, but not those of the same lineage, and this phenomenon may be observed in many unicellular plants and animals. We are justified in concluding from this that a fairly large degree of difference between the conjugating gametes secures the best results, whether this result is to be looked for in a 'rejuvenation' or in an increased adaptive capacity; but it is erroneous to regard the stronger attraction between individuals of different descent as a direct outcome of this. To me, at least, it seems to be an adaptive arrangement. The whole of the long and complex phylogenetic history of the sex-cells, the gametes, shows clearly that we have here to deal with a succession of adaptations, and that the degree of attraction which obtains between gametes has gradually been increased and specialized in the course of the phylogeny. I need only briefly recall what we have discussed in a