The supposition which Cabanis and M. de Gobineau have taken up will, doubtless, never be realised. To admit that all human races can reach a complete hybridity, would be to admit that each race is cosmopolite, which it is not. But at least it remains true, that when two _very different_ races are united, we must not hope for anything good or durable from their union. The same phenomenon happens, with the simple difference of _intensity_, when two different _species_ of animals are united. So the monogenists are astonished at such a result in man, “a result quite contrary,” says one of them,[280] “to what one generally expects in crossing a race.”[281] The astonishment of the learned man, of whom we speak, is explained easily enough by the ideas which he holds of human races, where he only sees degenerated varieties of the original type, preserved by the European in its primitive purity. It is evident that in this monogenic hypothesis, which we shall not touch on again, the union of one of these degenerated races with the pure stock would be a sort of _hygid_[282] consanguinity, and therefore favourable to the offspring. Here there would happen something analogous to the practice of the peasants in the _crétin_ districts, who try to struggle against the scourge by seeking for marriages in the plains, in order to give purer blood to each generation. In a more general manner it is evident that if we suppose two sets of people born of the same stock, and that one of them, after various fortunes, after having undergone fatal influences, should unite itself with the other, which had remained unaltered, it is evident that the produce of such a union ought to tend to reproduce in its purity the primitive type.[283] If it is not so with the union of different races of men, the reason is simply that they do not directly descend the one from the other; and from this debasement of produce there results a new proof in favour of the ideas which we are defending. It remains for us to speak of hybridity, as applied to the propagation of a deformity or a monstrosity. We know that when we experimentally unite one of this class to the other, two individuals whose organism has equally deviated from the usual type, “nothing is more difficult than to prevent these mischances from being done away with.”[284] A stronger reason, then, for the same when one of these individuals alone is deformed, which happens always in a state of nature. The races which we can thus produce are a kind of experiment which exist, but which it would be illogical to deduce can exist naturally. Because we make in a laboratory oxygenated water, or mixtures of hydrogen and chlorine, must we admit that these bodies are to be found united in nature? Quite the contrary; we deduce from their instability that they do not, and cannot thus exist in a natural state. CHAPTER VIII. SPECIES.