I I. In every race of domesticated animals, new varieties, departing more or less from the parent stock, in one or more of these charac ters, are occasionally produced ; some of these being directly traceable to the influence of external conditions, whose action upon a long succession of generations gradually mo difies the character of that part of the race which is exposed to it; whilst others originate in the production of offspring, which, from some cause not understood, present a marked departure from the parental type. In the first case, the variety is permanent, that is, it tends to hereditary transmission so long as the same conditions exist ; and thus arise the peculiar adaptations between the characters and constitutions of races, which have been dispersed through regions very dissimilar in their physical conditions, and the climate, food, &c., to which they have been respec tively habituated. In the second case, the variety is transitory, the individual peculiarity tending to disappear in the course of two or three generations, by becoming merged in the more general type ; but if it should happen that the individuals thus distinguished should breed together, the peculiarity shows a ten dency to perpetuation by hereditary trans mission ; and thus an entirely new race may originate, which remains distinct so long as it is not allowed to breed with others.
12. The several races of any kind of do mesticated animals, which, according to the foregoing criteria, are accounted as belonging to the same species, breed freely and sponta neously with each other, when allowed to do and the offspring are fertile, not only with either of the parent races, but with each other. The mixed races thus originating, it may be added, frequently surpass either of their parent stocks, not only in the advanta geous combination of different attributes, but also in general vigour, and in procreative capacity ; so that the mixture of races which are specifically identical, tends to the multi plication of the species as a whole.
13. On the other hand, although propaga tion may take place between individuals of undoubtedly distinct species, yet there is little spontaneous tendency to such admix ture; for each animal will select one of its own species for sexual intercourse, in preference to. one of another species ; and it is, con sequently, only when restricted by artificial interference, that a male and a female of different species are disposed to copulation with each other. The hybrid offspring of this act partake of the characters of both the parent stocks, but are deficient in genera tive power ; so that, although a mule may be fertile when paired with an individual of either of the parent races, it is seldom or never fertile with one of its own kind. Thus the peculiarities introduced by hybridity are speedily merged into those of the parent stocks ; and no new race has ever been known to originate from this kind of union.* 14. Among all those races which are en titled to rank as varieties only, the physiological conformity is often closer than the structural ; thus, as Dr. Prichard has pointed out " the great laws of the animal economy, all the principal facts which relate to the natural and vital functions, the periods and duration of life, the economy of the sexes, the pheno mena of parturition and reproduction, are, with slight deviations resulting from external agencies, constant and uniform in each par ticular species." 15. So, again, among the varieties of the same species, there is, with subordinate dif ferences, such as can be traced to external agencies, and particularly to human influence, a very close psychical conformity ; the capa cities of the several races being fundamentally the same, although varying in their degree of relative development.