Each germinal cell contains the hereditary mneme of its ancestors, paternal and maternal, and the two cells united by conjugation (Fig. 17) that of the ancestors of each of them. We have spoken above of ecphorias produced according to _Mendel's_ law and reproducing characters which have been latent during one or two generations. Darwin was the first to study this interesting fact, which shows how atavism often results from the crossing of varieties. There are several varieties of fowls which do not brood; if two of these varieties, B and C, are crossed excellent brooders are obtained. _Semon_ assumes that in each of the non-brooding varieties the ancestral energy, A, of the primary species, is weaker than that of varieties B and C; we have then A > B, and A < C. But if B is coupled with A the product represents the value B + C + A + A. Then B and C are in equilibrium; and A being doubled becomes stronger than each of them and arrives at ecphoria in their place, which restores the faculty of brooding to the product of crossing. _De Vries_ has shown, in the crossing of varieties with their primary species, more or less analogous phenomena which he calls "Vicino-variations." Conjugation leads to infinite combinations and variations which the law of heredity traverses like a guiding line. The celebrated zoölogist, _Weismann_, considers that the chromatin of each germinal cell contains a considerable quantity of particles each of which is capable of forming an entire organism similar to the parents; these he calls "ides." According to _Weismann_, each ide is subdivided into "determinants" from which each part of the body is derived, being potentially predetermined in them. According to the action of a yet unknown irritation male or female determinants develop in each individual of the animal species with separate sexes. But if the determinants are disordered, either by abnormal variations or by pathological causes, hermaphrodites or monstrosities may be produced. In animals which are normally hermaphrodite (snails, etc.), there is only one kind of sexual determinant, while in polymorphous animals (ants, etc.), there are as many as the polymorphous forms. The conception of "ides" and "determinants" is only a hypothesis to which we must not attach much value. The mnemic laws established by _Semon_ give a much better explanation of the facts.