88 BUREAU OF AMERICAN ETHNOLOGY [BuLL. 185 
and at the Nightwalkers Butte in the Bull Pasture site (82ML39). 
At the Leavenworth site the length of the entrance was nearly three 
times its width and each side was outlined by a row of postholes, a 
type of vestibule entrance of some antiquity in the central Plains, 
where it reaches back to Upper Republican times. 
The use of leaners to support the earthen sides of the lodge seems to 
be a rather late trait on the Plains. The Pawnee, Omaha, Otoe, Arik- 
ara, Mandan, and Hidatsa all used the leaner in the historic period. 
For the immediately preceding period there is little evidence. Lower 
Loup houses in general show an outer circle of 16 posts and although 
leaner remains have not been recognized, they were presumably pres- 
ent. 
In 1951 Donald J. Lehmer excavated 10 houses at the Phillips Ranch 
site (839ST14), near Pierre, S. Dak., for the Missouri Basin Project of 
the River Basin Surveys. The site is undocumented but yielded trade 
material and horse remains (Lehmer, 1954). All houses uncovered 
here were circular in outline, with central fireplaces, vestibule type en- 
trances, and with four single or multiple primary roof supports set 
to form a rectangle about the fireplace. In general these center posts 
were set at, or close to, an even distance from the fireplace and show 
a strong tendency to be halfway between the center of the house and 
the wall. Entrances are unlike those at Star Village and resemble 
those at the Leavenworth site in being of the long vestibule type out- 
lined with closely spaced postholes. The greatest difference between 
the structures here and the later ones lies in the outer ring of posts, 
which at the Phillips Ranch site were closely set at fairly regular in- 
tervals. The small, closely spaced posts were generally interspersed 
with heavier posts which might or might not be regularly spaced. 
Since all these wall members were set vertically, it would appear that 
leaners were not used here. The suggestion that leaners were not pres- 
ent may be offset by evidence gathered by Lehmer in 1950 at the nearby 
Dodd (39ST30) site, where he also uncovered the floors of 10 round 
structures. These he considers to be earlier than those at the Phillips 
Ranch site since trade goods were less abundant and horse remains 
were lacking. At one house, Feature 35, charred material found lying 
on the floor led Lehmer to postulate a wall of four separate layers, 
the inner one of which was composed of a “series of overlapping split 
wood ‘bats’ roughly 2 inches thick which apparently leaned against 
the house framework” (ibid., pp. 12-18). It would seem at least as 
probable that the charred material found here might have represented 
roofing rather than the remains of the wall of the house. It is possible 
that at the Dodd and Phillips Ranch sites leaners were used with their 
butts resting on the ground surface outside the house pit, although 
Lehmer in describing the charred remains at Feature 35 did not be- 
lieve this to have been the case. At Rock Village (82ME15) the posi- 
