ROCK-SHELTER OF LA COLOMBIERE—MOVIUS 363 
dug, with the exception of a test section approximately 2 meters 
square left at the request of Dr. Franck Bourdier, Regional Director 
of Prehistoric Antiquities, and a small, apparently sterile area at the 
extreme eastern end of the shelter. 
The personnel of the expedition was as follows: Dr. Kirk Bryan, 
Professor of Physiography, Harvard University; Dr. S. Sheldon 
Judson, Jr., Department of Geology, University of Wisconsin; Louis 
Dupree, Harvard University; Carleton Pierpont, Harvard University; 
and the writer. Professor Bryan was assisted in the geologic field 
work by Dr. Judson, while the writer had Messrs. Dupree and Pier- 
pont as his assistants during the digging of the site. 
The rock-shelter of La Colombiére is 46 meters long and some 12 
meters wide in the approximate center of the site below the maximum 
projection of the rock overhang. When we arrived there on the 8th 
of June 1948, not only was the place overgrown with small trees and 
shrubs, but also it was impossible to see across it, owing to the presence 
of several enormous boulders and piles of earth. The former had 
fallen from the roof of the shelter during prehistoric times; the latter 
consisted of M. Pissot’s rather extensive dumps. 
When the tedious work of clearing away this refuse had been 
completed (pl. 3, fig. 1), the limits of the areas previously excavated 
could be established. As the result of finding the edges of the trench 
dug in 1912-13 in the eastern portion of the site, it was possible to 
identify accurately the various levels originally reported by Dr. 
Mayet and M. Pissot (compare pl. 2). During the course of these 
clearing operations, Dr. Judson successfully completed a detailed 
topographic map showing La Colombiére and the adjacent region. 
From the outset the chance that the rock-shelter had been occupied 
during the interval immediately preceding the deposition of the 20- 
to 23-meter terrace (1. e., prior to the invasion of the site by the river) 
was considered a good possibility. It was also believed likely that 
what appeared to be the top of a filled-in lower cave (see pl. 3, fig. 1) 
might yield evidence of an earlier, and possibly even more interesting, 
occupation. But the hoped-for lower cave failed to materialize, and 
there was no evidence whatsoever of an occupation layer underlying 
the deposits in question. 
Bedrock was finally reached at a depth of 11.85 meters, and a 
magnificent section was exposed through the tightly packed and fine- 
grained terrace deposits, which consisted entirely of sands and silts, 
with a basal layer of coarse sand and gravel (pl. 3, fig. 2, and pl. 4, fig. 2). 
Throughout, a very interesting fauna, especially rich in small verte- 
brates and also containing a few mollusks, was collected. 
Abundant soil samples were taken for analysis, and it is hoped that 
on the basis of these the presence of a microfauna, as well as pollen 
(in the silty and clayey layers), will be demonstrated. It is felt that 
