The distribution of kivas in the ruins of De Chelly affords another indication that the occupancy of that region was quiet and little disturbed, and that the ruins were in no sense defensive structures. Kivas are found only in permanent settlements, and the presence of two or three of them in a small settlement comprising a total of five or six rooms implies, first, that the little village was the home of two or more families, and, second, that there was comparative if not entire immunity from hostile incursions. If the conditions were otherwise, these small settlements would have combined into larger ones, as was done in other regions. Probably these small settlements with several kivas mark a late period in the use of outlying sites. The position of the kivas in some of the settlements on defensive sites, and their arrangement across the front of the cove, suggest that such sites were first used for outlooks, and that their occupancy by regular villages came at a later period. All of the now available traditions of the Navaho and of the Hopi Indians support the conclusions reached from a study of the intrinsic evidence of the ruins, that they represent a comparatively late period in the history of pueblo architecture. It appears that some at least of the ruins are of Hopi origin. It is certain that the ruins were not occupied at one time, nor by one tribe or band. As criteria in development or in time the cliff ruins are valueless, except in a certain restricted way. They represent simply a phase of pueblo life, due more to the geological character of the region occupied than to extraordinary conditions, and they pertain partly to the old villages, partly to the more modern. Apparently they reached their greatest (not their highest) development in the period immediately preceding the last well-defined stage in the growth of pueblo architecture, a stage in which most of the pueblos were at the time of their discovery by the Spaniards, and in which some of them are now. Reliance for defense was had on the site occupied, and outlying settlements for horticultural purposes were very numerous, as they must necessarily be also in the last stage--the aggregation of many related villages into one great cluster.