126 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 93 



HOUSE I 



In 1930 this house pit was round, having a diameter of 33 feet. The 

 depression was very shallow, perhaps 6 inches in depth and was barely 

 noticeable. We ran a 3-foot trench from the center to the south end of 

 the pit, keeping just to the east of the center line. Since the earlier 

 excavation had been carefully filled in there was no way to tell which 

 half Sterns had excavated other than by Dr. Gilmore's memory. At 

 a depth of 3 feet 2 inches a floor line was encountered, and in the north- 

 west corner of the trench (i. e., the center of the pit and presumably 

 of the old house) a fire area 3 to 4 inches in thickness was found at 

 the same level. The only aboriginal artifacts recovered were a few 

 flint chips and a half dozen potsherds of the same type as those found 

 later in houses 2 and 3. However, near the center of the trench we 

 found a post containing iron nails, indicating that we were in Sterns' 

 former excavations. Since there seemed no way to avoid duplicating 

 much of this earlier work, and as the untouched house pits had just 

 been located, we stopped work in house i and filled in. Trial pits in 

 house 2 yielded the surest evidence of occupation, so we transferred 

 our activities to that house pit. 



HOUSE 2 



This house was indicated in the brushy area at the south end of the 

 cemetery by a faint depression, perhaps 8 inches in depth and roughly 

 30 feet in circular outline. Work began here April 12, 1930, and was 

 carried on three other week ends during the spring. On the summer 

 expedition of that year we worked in house 2 from June 11 to June 18, 

 when the excavation was completed and the pit filled in. The method 

 of excavation by trenching has already been described in regard to the 

 Lost Creek site. Plate 12, figure i, illustrates this work in progress 

 on the southeast side of house 2. The presence of trees and brush with 

 the resulting network of subsoil roots made excavation difficult. 



As seen in the diagram (fig. 14), the house outline was approxi- 

 mately square with rounded corners, measuring 18 feet from northeast 

 to southwest and 17 feet from northwest to southeast. The edges of 

 the old floor line were difficult to determine with exactitude, except 

 on the southeast side, where there was a clear line of demarcation. 

 This is indicated in the diagram (fig. 14) by solid lines where the 

 boundaries were perfectly clear and dotted lines where they are 

 approximate. The charred remnants of five vertical posts and one post 

 mold, each about 5 inches in diameter, were encountered on and just 

 below the floor line. These are apparently remnants of the inner row 



