Moreover, this line of development was, as with the structural features already reviewed, unbroken from first to last--from cave to cliff, and from cliff to round-town conditions of life; for the art remains of the round ruins, of which I recovered large numbers when conducting the excavations of the Hemenway expedition in ruins east of Zuñi, are with scarcely an exception identical, in type at least, with those of the cliff ruins, although they are more highly developed, especially the potteries, as naturally they came to be under the less restricted, more favorable conditions of life in the open plains. Everything, in fact, to be learned of the round-ruin people points quite unmistakably to their descent in a twofold sense from the cliff-dwelling people; and it remains necessary, therefore, only to account for their change of habitat and to set forth their supposed relationship finally to the modern Zuñi pueblos. In earlier writings, especially in a "Study of Pueblo Pottery,"[3] where the linguistic evidence of the derivation of the Zuñis from cliff-dwelling peoples is to some extent discussed, I have suggested that the prime cause of the abandonment of the cliffs by their ancestry was most probably increase of population to beyond the limits of available building area, and consequent overcrowding in the cliffs; but later researches have convinced me that, although this was no doubt a potent factor in the case and ultimately, in connection with the obvious advantages of life in more accessible dwelling places, led by slow degrees, as the numbers and strength of the cliff villages made it possible, to the building of contiguous pueblos both above their cliffs on the mesas and below them in the valleys, still it was by no means the only or the first cause of removal from these secure strongholds. Nor is it to be inferred from the evidence at hand that the cliff dwellers were ever driven forth from their almost inaccessible towns, either by stress of warfare or by lack of the means of subsistence, as has been so often supposed. On the contrary, it is certain that long after the earliest descents into the plains had been ventured, the cliffs continued to be occupied, at first and for a very long period as the permanent homes of remnant tribes, and later as winter resorts and places of refuge in times of danger for these latter tribes, as well as, perhaps, for their kinsfolk of the plains. [3] Fourth Annual Report of the Bureau of Ethnology, 1882-83.