A layer of homogenous terra rossa soil (thickness 0.25–0.80 m), containing sparsely dispersed flint items, was exposed both in the trenches and in the excavation area. It lay directly on the bedrock in the lowermost terrace and on top of a natural layer of flint pebbles (thickness c. 0.4 m; Figs. 1, 2) on the upper terraces. The layer of flint pebbles accumulated as a result of the erosion of a flint layer in the chalk (soft limestone) bedrock exposed beneath them. The chalk is of the Mishash formation, which is the geological unit at the base of the surface level in the region.
No archaeological horizons containing knapped items were discovered. The items that were found were scattered in a soil layer. There was a somewhat higher concentration of items near the surface. An assemblage of 113 flint items was collected (only items greater than 2 cm were gathered), mainly consisting of weathered natural flint fragments, including several worked items: amorphous cores, flakes and primary flakes, core debitage and a few tools and retouched flakes. Most cannot be ascribed to a particular culture or period. The tools were mainly side scrapers (Fig. 3), two of which occurred on Levallois flakes that can be attributed to the Mousterian culture (Fig. 3:2, 6).
It appears that the excavation area is not an actual prehistoric site. The low concentration of items, their distribution pattern and the absence of a distinct archaeological horizon all indicate a random accumulation that is colluvial in nature. The provenance of the worked items was probably a site located higher up topographically, and presumably it was related to flint production from the Mishash formation. Such a site was exposed in 2006, c. 1 km northeast of the excavation (Barzilai, Birkenfeld and Crouvi 2009).