These conclusions are not based on a study of the ruins in Canyon de Chelly alone, which illustrate only one phase of the subject, but of all the pueblo remains, or rather of the remains so far as they are now known. They imply a rather sparsely settled country, occupied by a comparatively small number of tribes and subtribes, moving from place to place under the influence of various motives, some of which we know, others we can only surmise. It was a slow but practically constant migratory movement with no definite end or direction in view. The course of this movement in a geographical way does not as yet reveal a preponderance in any one direction; tribes and subtribes moved from east to west and from west to east, from north to south and from south to north, and many were irregular in their course, but the movements, so far as they can now be discerned, were all within a circumscribed area. There is no evidence of any movement from without into the pueblo group, unless the close relation of the Hopi (Moki) language to the other Shoshonean dialects be such evidence, and none of a movement from within this area out of it, although such movements must have taken place, at least in the early history of the region. It must be borne in mind in this discussion that while we can assign approximate boundaries to the ancient pueblo region on the north, east, and west, no limit can as yet be fixed on the south. The arid country southward of Gila river and northward of the Mexican boundary would be a great obstacle to a movement either north or south, but little as we know about that region we do know that it was not an insurmountable obstacle. The Casas Grandes of Janos, in Chihuahua, closely resemble the type of ruins on the Gila river, in Arizona, of which the best example we now have is the well-known Casa Grande ruin. We know that there are cliff ruins in the Sierra Madre, but beyond this we know little. Concerning the immense region which stretches from Gila river to the valley of Mexico, over 1300 miles in length, we know practically nothing. In that portion of the pueblo region lying within the United States migratory movements have, as a rule, been confined to very small areas, each linguistic family moving within its own circumscribed region. Some instances of movement away from the home region have taken place even in historic times, as, for example, the migration of a considerable band of Tewas from the Rio Grande to Tusayan, where they now are, and moreover, this movement probably occurred en masse and over a considerable distance; but there is little doubt that the usual procedure was different.