lemurs. Both of the two species known are small, not probably exceeding a gray squirrel in size. Only the skulls and jaws are known (Fig. 2).
But the Condylarthra prove to be the ancestors of a still greater population of descendants. We have traced all forms of grinding teeth up to a pattern which consists of four tubercles or cones arranged within a square. But it has been possible to show that one of the four tubercles appeared after the other three, as an addition to them, so that the earliest form, of molar or grinding tooth was tritubercular, and that the quadritubercular was an outgrowth from it. Now, one of the three families of the Condylarthra has tritubercular molars, and there is little doubt that it was the ancestor of the two other families. The principal genus of this family is called Periptychus. From this family came an order of hoofed mammals, which never rose to the possession of four tubercular grinders, although the crowns became crested by the modification of the three which they possess. This order, the Amblypoda, had a short life in geological time, and did not grow in the dimensions of the brain, but developed huge skeletons with skulls that sprouted into horns and strange processes.
The Condylarthra with three tubercles are probably also the ancestors of the carnivorous orders. The lions, tigers, wolves, and bears of to-day can be shown to be descendants of animals absolutely intermediate between themselves and the animals just mentioned. These half-carnivores, or Creodonta, have, like the ancient hoofed mammals, more numerous teeth than their modern representatives, and differ from the true carnivora in just the ways, in limbs and feet, that we have seen that the ancient hoofed mammals differ from their modern descendants. Creodonta were not such dangerous animals as the carnivora, with some possible exceptions, because, although they were as large, they generally had shorter legs, less acute claws, and smaller and more simple brains.
This genealogy, it will be seen, does not show us the ancestors of the Condylarthra. This remains for future discovery. It is, however, probable that they will be found in the earlier geological periods, among some of the marsupial mammals only known thus far from the jaws and teeth. It must also be noted that a number of these ancestral groups are represented in the existing fauna by a few genera. Of the Condylarthra, a near relative exists in the Hyrax, or cony, which now inhabits Africa and Western Asia. Of the Creodonta, several genera exist in Madagascar, West Africa, and the northern hemisphere. The mastodons are late representatives of an ancient type, and their phylogeny has not yet been fully made out. But they certainly also came from the Condylarthra.
There is a remarkable likeness between the history of the development of the reptiles and that of the higher mammals, in one respect, and that is, that they have apparently all been derived from a single
