Approximately one-half of the collection was washed, labeled (with square and identification number) and organized by Bordes by condition, but from our work, especially in producing section plots of the material, it became clear that there were problems in the way this material was organized. For instance, some pieces had been placed in the wrong type box, and, more seriously, some drawers were labeled with an incorrect stratigraphic layer.
The figure on the right shows a sample of artifacts with problems in the level assignments as indicated on drawers of material. This is a frontal view of material marked as coming from either level F (in red dots) or I (in yellow dots). Notice that much of the material assigned to level I actually plots with the level F material, while many pieces assigned to levels F or I are actually derived from lower levels.
For the material not studied by Bordes, a portion was washed and labeled but unsorted either by level or category. The principal challenge with this material, which had never before been analyzed, was to determine the stratigraphic layer. This was accomplished with reference to the field notebooks (where level was noted during excavation) and by plotting the artifacts according to their x, y, and z coordinates.