52 BUREAU OF AMERICAN ETHNOLOGY [BuLu. 185 
cutting and scraping purposes without intentional modification of 
their original shape (pl. 11, g, 7). 
Animal bone was not common at the site, and the major portion 
of it consisted of small unworked splinters. Only two bones, both 
from bison, were identifiable. No bone or antler artifacts were 
recovered. 
Although the yield from the site was scanty, it is felt that more 
work there would have been justified had time and funds permitted. 
The site was hard to reach and located in an isolated area where it 
would have been hard to maintain a camp, but a small party could 
undoubtedly have obtained much more artifactual material, which, in 
turn, would have thrown more light on one of the pottery-making 
groups who occupied the area in late prehistoric times. 
SUMMARY 
In the preceding pages an attempt has been made to present the 
data from 31 sites in and about the Garrison Reservoir area in North 
Dakota, particular attention being paid to that section of the reservoir 
which lies within the Fort Berthold Indian Reservation. The bulk 
of the data presented was gathered by two Smithsonian Institution 
River Basin Surveys parties in 1947 and 1950, supplemented by a 
small amount of work in 1951. Cooperating agencies have been 
drawn on for a certain amount of additional data. 
The site of the last Arikara earthlodge has been described and a 
floor plan presented. This appears to have been a ceremonial struc- 
ture and has been compared with its predecessor at Like-a-Fishhook 
Village, with a still earlier structure excavated at Star Village, and 
with the traditional account of the manner of constructing these 
lodges. Although varying in minor details from each other, all were 
closely related structurally. A difference is suggested between these 
and earlier structures in the manner of framing the sides. 
A hunting lodge which was found in the course of the surveys con- 
sisted of a four-post central foundation against which poles were 
Jeaned. These appear to have been covered with bark slabs over 
which earth was piled. It differs in no way from lodges noted by 
Maximilian in 1833, and in floor plan, it resembles structures exca- 
vated from both historic and prehistoric horizons in the central plains. 
Five sites showing depressions believed to represent former eagle- 
trapping pits are listed. As far as could be told without excavation, 
these followed the general pattern of eagle-trapping pits as they are 
described in the literature, not only for the Mandan and Hidatsa, 
but for the Northern Plains groups in general. 
