548 FEWKEs: rxiT type of pueblo architecture 



account for a later rebuilding of a room, rather than abandon- 

 ment and secondary occupation of a cliff-house, but even that is 

 a doubtful interpretation, for foundation walls were often more 

 roughly constructed than those built upon them, at the same 

 epoch. It is still an open question whether such a ruin as Com- 

 munity House, on the Mesa Verde, was abandoned earlier than 

 the neighboring Cliff Palace. The author has seen no adequate 

 evidence to prove that the Mesa Verde cliff -houses were deserted 

 later than the villages on the Mesa.^ 



In southern Colorado and northern Arizona, where caves are 

 especially commodious and abundant, pre-puebloan man moved 

 into them for protection from foes, building his single-roomed 

 isolated dwelling in one of the most convenient.*^ Later his 

 clan or family was joined by others or increased naturally in 

 numbers. Rooms to accommodate the increase multiplied until 

 the floor space of the cave, which at first was ample for founda- 

 tions of habitations, became too small, and the houses were 

 crowded together; later the people were so hard pushed for space 

 upon which to build their habitations that they were obliged, 

 when cave floors failed, to erect rooms on the roofs of houses 

 already occupied, thus imparting a terraced form to the structure. 



' This of course does not mean that now and then there has not been a second- 

 ary occupation of certain well known cliff-houses; thus the Asa clans of the Hopi 

 inhal)ited caves in Chelly canyon well into the historic epoch. The pre-puebloan 

 houses were of course abandoned when their inhabitants moved into the cliffs. 



^ The cliff-pueblos of the Navajo National Park, in northern Arizona, have 

 several architectural features different from those of the Mesa Verde, but show 

 the transition from a pre-puebloan habitation to the "unit type." The construc- 

 tion of rooms and kivas in this region is much ruder, the walls as a rule being made 

 of undressed stones, or of clay or adobe laid on wattles, supported by upright 

 logs. A circular ceremonial opening corresponding to a Hopi or Mesa Verde 

 sipapu has not yet been described or figured from this region, although there 

 are depressions in the floors of some of the rooms which have been given that 

 name. The feature in a kiva which the author identifies as the same as the sipapu 

 of the Hopi kiva is a small circular hole having a diameter of a few inches, situated 

 in the floor about midway between the fireplace and the kiva wall, on the side 

 opposite the deflector. Other holes or depressions are also found in kiva floors. 

 Some of these, as fire-holes, may be ceremonially known as si papas. These are 

 not the same, however, as the sii)iipus of the Hopi kivas, nor have they the same 

 syml)olic interpretation. 



