Nathusius has remarked, and the observation is an interesting one, that the peculiar form of the skull and body in the most highly cultivated races is not characteristic of any one race, but is common to all when improved up to the same standard. Thus the large-bodied, long-eared, English breeds with a convex back, and the small-bodied, short-eared, Chinese breeds with a concave back, when bred to the same state of perfection, nearly resemble each other in the form of the head and body. This result, it appears, is partly due to similar causes of change acting on the several races, and partly to man breeding the pig for one sole purpose, namely, for the greatest amount of flesh and fat; so that selection has always tended towards one and the same end. With most domestic animals the result of selection has been divergence of character, here it has been convergence.[19] The nature of the food supplied during many generations has apparently affected the length of the intestines; for, according to Cuvier,[20] their length to that of the body in the wild boar is as 9 to 1,—in the common domestic boar as 13·5 to 1,—and in the Siam breed as 16 to 1. In this latter breed the greater length may be due either to descent from a distinct species or to more ancient domestication. The number of mammæ vary, as does the period of gestation. The latest authority says[21] that “the period averages from 17 to 20 weeks,” but I think there must be some error in this statement: in M. Tessier’s observations on 25 sows it varied from 109 to 123 days. The Rev. W. D. Fox has given me ten carefully recorded cases with well-bred pigs, in which the period varied from 101 to 116 days. According to Nathusius the period is shortest in the races which come early to maturity; but the course of their development does not appear to be actually shortened, for the young animal is born, judging from the state of the skull, less fully developed, or in a more embryonic condition,[22] than in the case of common swine. In the highly cultivated and early matured races the teeth, also, are developed earlier. The difference in the number of the vertebræ and ribs in different kinds of pigs, as observed by Mr. Eyton,[23] and as given in the following table, has often been quoted. The African sow probably belongs to the _S. scrofa_ type; and Mr. Eyton informs me that, since the publication of this paper, cross-bred animals from the African and English races were found by Lord Hill to be perfectly fertile.