252 BUREAU OF AMERICAN ETHNOLOGY [Bull. 179 



On the Columbia River among the Tenino Indians the following 

 spring, they noted : 



Since we left them in the autumn they have removed their village a fevsr hun- 

 dred yards lower down the river, and have exchanged their cellars in which 

 we then found them, for more pleasant dwellings on the surface of the ground. 

 They are formed by sticks and covered by mats and straw. . . . [Ibid., vol. 2, 

 p. 63.] 



By the time they had arrived in Nez Perce territory again : 



The salmon not having yet called them to the rivers, the greater part of the 

 Chopunnish (Nez Perc6) are now dispersed in villages through this plain, for 

 the purpose of collecting quamash (camas) and cows (kouse). [Biddle, 1904, 

 vol. 2, p. 83.] 



There are numbers of references by other observers in the area, all 

 pointing to seasonal shifts in habitation which meant the movement 

 of the entire village. Even as late as 1885, the pattern was still being 

 followed by some Nez Perce on the lower Snake River. According 

 to one of the writer's informants, an early settler, the Nez Perce 

 used to spend the winter in the river bottom and move out in the 

 spring. Ray summed it up as follows: 



On the whole. Plateau life involves wintertime occupancy of river villages 

 and summertime camping at fishing, berrying and root digging spots. [Ray, 

 1939, p. 14.] 



What would be the effect of seasonal migration on the archeological 

 record? One result would be found in architecture, since only the 

 semisubterranean pit houses would be preserved. The summer mat 

 house would leave little or no trace after a few years. It is possible 

 that the multiple floor levels found in the house pits at Cold Springs, 

 Techumtas Island, and elsewhere are records of year-to-year occupa- 

 tion. After the pit became too shallow, it may have been cleaned out, 

 and some of the floor levels may record only one winter's occupation. 

 Doubtless, the seasonal migration pattern is an old one in the Plateau, 

 for the Indians held to it long after other customs had disappeared. 

 There is no way of determining how far back in time the custom was 

 followed. 



ARCHITECTURE 



One of the best descriptions of Indian dwellings in the Plateau was 

 also the earliest. In the spring of 1805, near the mouth of the Snake 

 River, Lewis and Clark wrote : 



The houses of the Sokulks are made of large mats of rushes, and are generally 

 of a square or oblong form, varying in length from fifteen to sixty feet, and 

 supported on the inside by poles or forks about six feet high : the top is covered 

 with mats leaving a space of twelve or fifteen inches the whole length of the 

 house, for the purpose of admitting light and sufifering the smoke to pass 

 through: the roof is nearly flat . . ., and the house is not divided into apart- 

 ments, the fire being in the middle of the large room and immediately under 

 the hole In the roof ; . . . [Biddle, 1904, voL 2, p. 189.] 



