72 BUREAU OF AMERICAN ETHNOLOGY [bull. 118 



shown in figure 30. This structure was 27 feet 8 inches long and 24 

 feet 6 inches wide. Trenches had been dug for the reception of the 

 base of vertical posts. These vertical posts were held in place by 

 a horizontal log at the bottom of the trench, on the outer side of the 

 trench. No horizontal molds appeared inside the vertical molds. 

 Corners were made by driving small poles in the earth, in the form 

 of an arc. There was no evidence of any fire within the rectangular 

 structure. A difference in the nature of the soil within and without 

 the structure was easily observable. Another definite vertical cleav- 

 age plane x appeared immediately over the line of post molds. 



Within the structure the soil was darker and mixed with more 

 charcoal, dark red and yellow clays. Outside the line of post molds 

 was a yellowish and brown clay which seemed to have been piled 

 against the outside of the structure. On the west side of the struc- 

 ture the post molds slanted to the west in the softer earth above 

 the floor in which the bases of the posts were embedded. This seemed 

 to indicate that the wall of the structure leaned outward for a con- 

 siderable time before its final destruction by decay. There was no 

 evidence of any burned structure in the mound and no evidence that 

 any other structure had been erected on the site. From the evidence 

 here, and from other sites, 2 the author is convinced that many of 

 these wooden structures in the basin were covered over with earth. 

 Some structures had earth on their roofs to a depth of 3 or more 

 feet. A close inspection of the remains of the structure in Mound 

 No. 1 seems to demonstrate that it stood as long as it could with- 

 stand the forces of nature — wind, water, frost, and gravity. When 

 the roof logs became so decayed that they could no longer sustain 

 the weight of the earth on the roof they collapsed. This let the 

 roof earth fall into the center of the mound, carrying with it the 

 log framework on which it rested. By so doing, the falling roof 

 slightly pressed outward the tops of the wall posts which still re- 

 mained, giving to them in some sections of this post-mold pattern 

 the outward slant which was observed. The falling of such a roof 

 would account for the vertical cleavage line observed, if it is under- 

 stood that the soil on the roof was of a different texture from that 

 which was piled up against the wall. This is an observed fact, 

 often repeated in this basin. The earth on the roof is the same 

 color and content as the adjacent village. Any kind of soil, even 

 with a large humus content, would lie on the roof. But the builders 

 seemed to understand that it took a clean, tough clay, free from 

 humus, to stand up well when piled against the walls outside such 

 a structure. It is an observed fact that on this site, as on others 



1 See Site No. 2. 



2 See Mound No. 2. 



