62 BUREAU OF AMERICAN ETHNOLOGY [bull. 51 



was gained by moans of a hatchway on one side of the roof near 

 the large banquette, which occupies a position, as respects the en- 

 trance and the place supposedly occupied bj^ the ladder and the 

 fire-pit, similar to the spectator's platform of a modern rectangular 

 Hopi kiva, except that it is higher above the floor and is relatively 

 smaller. If the banquettes were depressed and enlarged- into a plat- 

 form, the form of the kiva being changed from circular to rectangu- 

 lar, thus modified the banquette would form a structure like the 

 spectator's platform of a typical modern Hopi kiva.*^ 



Perhaps of all the ceremonial roonis repaired the walls of kiva M 

 were in the most dangerous condition. The front of the northern 

 wall of room 39 had been undermined and was without foundation, 

 hanging without basal support except at the ends. A support was 

 constructed under this hanging wall, and to give additional strength 

 the foundations were rebuilt a little broader at the base than for- 

 merly, causing the wall to bulge almost imperceptibly into the kiva. 

 Although no pilasters were seen, the deep banquette on the north- 

 western side places it among the kivas of the first type. 



KIVAS OF THE SECOND TYPE • 



The architecture of the two kivas O and R are so different from 

 those already considered that they are set apart from the others in 

 a second type. The form and structure of kiva W indicate that 

 this room also may be classed as of the same type. In the side canyon 

 north of that in which Cliff Palace is situated, where water was 

 obtained throughout the summer, there is another kiva, also supposed 

 to belong to the second type.^ 



The main difference in construction between the two types of 

 kivas is the absence of pilasters, which implies the absence of a 

 roof in the second type. The suggestion that a kiva of the second 

 type is simply an unfinished form of the first type has little to 

 support it, but whether the architectural difference in the two 

 types has any functional importance or meaning is unknown. It 

 has been suggested that one type was used by the Winter, the other 

 by the Summer people.*' 



'■ The two circular kivas of Kukuchomo, near Sikyatki, liavo this large banquette and 

 in other respects resemble the ruins of Canyon dc Chclly. Kukuchomo marks the site of 

 a settlement of the Coyote clan of the Hopi in prehistoric times. 



I" As a huge rock had fallen from the roof of the cave in which this kiva lies, since it 

 was first occupied, it would appear that the place was abandoned on that account. 



<■ Nordenskiold's description of this kiva has been quoted earlier in this paper. In the 

 description of a ceremonial room of a somewhat similar or of the same type in Spruce- 

 tree House the term "warrior room"' is used; there is nothing to warrant this designa- 

 tion, however, and it would be better to consider it simply as a kiva of the second type. 



