webb] ARCHAEOLOGICAL SURVEY OF NORRIS BASIN ^93 



Even with such repairs, the time eventually came when the build- 

 ing, because of decay, became unsuited for further use and a new 

 structure was necessary. It is natural to suppose that in most cases 

 the old town house was on the most desirable site for such a building 

 and probably on ground consecrated x for that purpose. In order 

 to have a new structure, the old one must either be removed or a new 

 site chosen and prepared. If the old site was to be used the old 

 structure had to be removed to make way for the new. From the 

 manner of its construction its mechanical removal would have been 

 difficult and the material of which it was composed would have had 

 no salvage value. The burning of such a structure was an easy 

 and effective way to clear the site for the new building. With the 

 collapse of the old structure, the earth about it and on its roof 

 actually raised the level of the site several feet ; a result very desir- 

 able in itself. Thus two advantages could be obtained by the simple 

 expedient of applying fire to the interior wooden content. It is 

 believed that there is abundant evidence to show that this was the 

 actual maimer in which these mounds were constructed. Each new 

 building had its roof covered with earth, which, when the building 

 was destroyed, added its increment to the height of the mound. 



The conclusion that these buildings had a heavy covering of earth 

 over the thatching is based on evidence enumerated as follows : 



1. The incomplete combustion of all carbonaceous material com- 

 posing the building clearly indicates a burning in an atmosphere 

 of reduced oxygen content, such as is produced intentionally in a 

 charcoal kiln, and such as would have been produced had these 

 structures been well covered with earth on both roof and walls. 

 Grass and small cane were reduced to charcoal along with large 

 logs, and all still retain their exact form, even to the preservation 

 of the strings and ropes used in roof construction. 



2. As was pointed out in the discussion of Site No. 9, earth which 

 had been heavily burned was found on the floor under the fallen 

 structure. The earth above it showed the effect of fire much less. 

 The structure, when it fell, did not reach the floor of the building 

 but was held up by burned earth which had previously fallen; evi- 

 dently from the roof of the structure. This condition clearly indi- 

 cates that earth was on the roof of the structure before the burning 

 started. 



3. So important is this concept of earth-covered structures to the 

 proper interpretation of the findings in the Norris Basin that one 

 naturally seeks from early historic records confirmation of this prac- 



1 See footnote 28, p. 214. 

 154676—38 14 



