18 BUREAU OF AMERICAN ETHNOLOGY [bull. 118 



be attributed to repairs made upon the town bouse after its deteriora- 

 tion due to age had made repairs desirable. From information 

 gained on other sites it appears that attempts at such repairs were 

 occasionally made with the evident purpose of prolonging the life 

 of the town house. 2 



In order to ascertain if additional post molds existed to the east 

 of this structure, the portion of the mound to the east of the structure 

 was removed by slicing down to the primary floor. No molds were 

 encountered on the secondary floor plane, but a few scattered molds 

 were found on the primary floor in the southeast quadrant, as shown 

 on the plat of the site (fig. 2). 



The south side of the mound having already been sliced down to 

 the 2.0-2.7 face, the remainder of the mound was then cut down in 

 such a manner as to form four faces which alined themselves with 

 the four walls of the secondary structure as shown in plate 5, a. 

 This was done in order that the position of this secondary structure 

 might be noted with respect to that of the primary floor. It was 

 then found that there was a post-mold pattern on the primary floor. 

 Careful investigation revealed that the orientation of these two sys- 

 tems of post molds was very nearly identical, and further, that the 

 molds of the secondary floor were almost directly above those of the 

 primary floor. It was surprising to find two structures of almost 

 identical size, shape, and orientation, one superimposed upon the 

 other. (PL 5, l.) 



This change in the technique of excavation left as a residue a 

 rectangular block approximately 35 feet square and 4% feet thick. 

 Plate 6 shows the arrangement of post molds on the surface of this 

 residual block, after the surface of the secondary floor had been 

 carefully exposed by troweling. This floor contained in the interior 

 of the structure a number of scattered post molds of large size, which 

 were probably made by the ends of beams used to support the roof. 

 In the center of this pattern of post molds the floor was hard burned, 

 indicating long-continued and very hot fires. No burned structure 

 was found on the secondary level. The post molds on the primary 

 level were all empty, but those on the secondary level, while easily 

 discernible, were partially filled with black humus, due to the infil- 

 tration of surface earth. 



The vertical walls of the residual block in the center of the mound 

 were carefully studied for evidences of stratification. All four ver- 

 tical faces bore lenticular striations, curving upward to all four 

 corners, and seemingly indicative of the possibility that four pri- 

 mary mounds, one centered at each of the four corners of the pri- 

 mary floor, had been built. Horizontal striations on the faces be- 



2 See Sites Nos. 8, 11, and 17. 



