The muscular exertions of herbivorous animals, so far as they are fed on pastures, are not so materially changed. The grazing habits of cattle and sheep in domestication are about the same as the grazing habits of wild animals of the same class; but the rapid movements and the watchfulness required for protecting the herd against carnivorous animals have completely disappeared. Stable-fed animals live under highly artificial conditions, and material changes may occur in them. I think the changes due to these causes may be observed in the oldest types of domesticated animals, such as are found in the neolithic villages of Europe, in which native European species appear in domesticated form (Keller). They may also be observed in the dogs of the various continents, which differ markedly from the wild species from which they are derived. Even the Eskimo dog, which is a descendant of the gray wolf and still interbreeds with the gray wolf, differs in bodily form from the wild animal (Beckmann). Modifications may also be observed in newly domesticated animals, like the Chukchee reindeer, which differs in type from the wild reindeer of the same area (Bogoras). I think it very unlikely, judging from our knowledge of the methods of domestication of tribes like the Eskimo and Chukchee, that any material amount of selection has contributed to the modifications of form which are found in these races of primitive domesticated animals. Their uniformity is still fairly well marked, although they have assumed types different from the wild species. A more marked differentiation of domesticated forms does not seem to occur until man begins to select and to isolate, more or less consciously, particular breeds. Opportunity for such isolation has been the greater, the older the domestication of any particular species. We find, therefore, that the number of distinct breeds have come to be greatest in those animals which have been under domestication for the longest periods. The number of varieties of domesticated species has also been increased by unintentional or intentional crossing of different species, from which are derived many breeds whose ancestry it is often so difficult to unravel. It appears, therefore, that there are three distinct causes which bring about the development of different types in domesticated animals: first, the influence of change of nutrition and mode of life; secondly, conscious selection; and, thirdly, crossing.