Of _Dasypodidae_ numerous representatives occur in the South American Tertiaries. From the higher beds many of the species are referable to existing genera, such as _Dasypus_ and _Tatusia_, although some are much larger than any living forms, the skull in one case being nearly a foot in length. In other instances, when lower formations are reached, the genera are also distinct, _Eutatus_ having the whole armour divided into movable bands, and the allied _Stegotherium_ representing the group in the Santa Cruz formation of Patagonia. Even in the Argentine Pleistocene there is an extinct genus, _Chlamydotherium_, represented by a species of the size of a rhinoceros, with grooved teeth approximating to those of the glyptodonts. The latter represent a family (_Glyptodontidae_) by themselves, and typically may be described as giant solid-shelled armadillos, although some of their smaller Santa Cruz representatives (_Propalaeohoplophorus_) approximate in some degree to true armadillos (see GLYPTODON). A very remarkable Santa Cruz armadillo, _Peltephilus_, has an altogether peculiar type of head-shield, developed into horns in front of the eyes; and, what is still more noteworthy, teeth in the front of the jaws, thereby rendering the ordinary definition of the order Edentata incorrect. It has been made the type of a distinct family, _Peltephilidae_. The past history of the armadillo group does not, however, by any means end here. True armadillos, it should be observed, are known in North America as far north as Texas, from the Pleistocene onwards; but in formations of middle Tertiary age are unrepresented. Recent discoveries apparently indicate, however, the occurrence of armadillos of a primitive type in the lower Tertiary or Eocene formations of Wyoming. The first evidence of these Eocene armadillos was afforded by portions of the jaws, which, together with a leg-bone of a totally different animal, were believed to indicate creatures nearly allied to the aye-aye (_Chiromys_) of Madagascar, and for which the name _Metachiromys_ was consequently proposed. According to modern usage, this name, in spite of its inappropriate nature, is retained for the armadillos, although in the writer's opinion it ought to be replaced. According to Professor H.F. Osborn, by whom their remains have been described, the North American fossil armadillos were closely related to the existing members of the group, from which they differ chiefly by the armour, or shield, having probably been formed of tough leathery skin instead of bony plates, by the presence of a single pair of large enamel-capped tusk-like teeth in each jaw, and by the degeneration of the other teeth. If these determinations are trustworthy, the question arises whether we should regard the armadillos of South America as the descendants of North American forms which migrated southwards before that separation of the two continents was established, which lasted for a large portion of