Riv. Bas. Sur. A ey 
Pap. No. 26] SMALL SITES ABOUT FORT BERTHOLD—METCALF al 
The exterior surface of one sherd is simple-stamped, one shows a 
single cord-impressed line, three carry shallow incised lines, and two 
are smooth. It must be remembered that these sherds are very small 
and all might actually be parts of rims rather than from the body of 
a vessel. They measure from 3.5 to 4.5 mm. in thickness. 
The one sherd that is definitely from a rim measures 18.5 mm. long 
by 12 mm. in width. It has a flattened lip, slightly thickened to the 
outside, crossed by diagonal cord-impressions. Immediately below 
the thickening of the lip the exterior surface shows a single horizontal 
cord-impression. The sherd has a maximum thickness of 6.5 mm. at 
the lip with a minimum thickness below that point of 5 mm. (pl. 11, 7). 
It is impossible to assign these tiny sherds to any cultural group 
now known from the area. They in no way resemble the ware from 
Rock Village, from the Nightwalkers Butte in the Bull Pasture, or 
from the Knife River Hidatsa sites, nor do they seem at all related 
to sherds from the Slant Village or the Double Ditch sites. At pres- 
ent they stand in a group by themselves. 
The lower level of the site yielded, in addition to these sherds, 
about 2 quarts of unmodified chips and flakes, the great bulk of 
which were of Knife River flint, a local yellow-brown or caramel- 
colored chalcedony the surface of which changes to a blue, or in 
some cases, white, with age and exposure. A few chips were of light 
chalcedony and of agate, and there were a few pieces of fire-fractured 
granite. A tool from this level, made from the Knife River flint is 
deeply patinated. In shape it is asymmetrically triangular and may 
have served as a drill or perforator. It measures 4.6 cm. in length, 
with a maximum width of 2.6 em. (pl. 11, 2). 
The two end scrapers from the test are of the common snubnosed 
type. One is markedly planoconvex in cross section (pl. 11, ¢), the 
other is thin and flat. Both have rounded working ends which are 
chipped to a short, abrupt bevel. This end is asymmetrical in the 
case of the thinner specimen, one side of the tool being longer than 
the other (pl. 11, 5). 
Two fragmentary blades that may have been originally almond- 
shaped in outline are planoconvex in cross section and chipped on 
only one face (pl. 11, d, 2), while two others with the same outlines 
are flattened ellipses in cross section and bifacially flaked (pl. 11, ¢, f). 
Seven specimens show more or less retouch on one or both faces and 
on one or more edges. Two of these appear to be fragments broken 
from large blades (pl. 11, a). The others may be fragments of 
broken tools but are more probably chance flakes which have received 
only enough attention to shape them into objects which would serve 
as cutting and scraping tools. Two flakes show irregular chipping 
on their thin edges, apparently from use, and suggest flakes used for 
