Fourteen excavation squares were opened up (1 × 1 m each; Figs. 3, 4), distributed across areas where concentrations of knapping debitage had been identified (Fig. 5). The knapped flint items were found in a shallow layer (15 cm thick), beneath which were loess soil and calcareous stones that were devoid of archaeological finds.
The quality of the flint assemblage gathered at the site is very poor as a result of weathering and prolonged exposure to the sun. The items are coated with a thick dark patina (‘desert varnish’), and most of the worked items are broken. The flint is derived from local rock outcrops that are exposed to the elements, where the Mishash layer is easily quarried; fragments scattered along the ridge can also be used.
Approximately 250 flaked flint items were collected (Table 1); most of the debitage consists of flakes and primary flakes. The excavation yielded eleven cores, and another three were recovered as surface finds (Table 2). Most of the cores were flake cores, and a few could also be used for the production of blades. The majority of the cores are centripetal Levallois cores (Fig. 6:1), including one small core of this type formed on a flake (15 × 40 × 40 mm; Fig. 6:2). Two bidirectional Levallois cores (Fig. 6:3) and irregularly shaped cores were also found.
The excavation yielded 31 tools, and another six were surface finds (Table 3). Most of the tool assemblage comprises retouched items (38.7% from the excavation; Fig. 7:1–3). The second most common type includes flakes and blades with notches and denticulates (25.9%), followed by scrapers (16%; Fig. 7:4). Among the end scrapers found on the surface was a heavy-duty scraper (Fig. 7:5) made on a large flake. Three side scrapers (Fig. 7:6) and two awls were also found (Fig. 7:7); the latter probably belong to a later period.
 
Table 1. Distribution of flint finds 
 
Excavation
Surface finds
Type
N
%
N
%
Primary flake
66
22.4
 
 
Primary blade
18
6.1
 
 
Flakes
82
27.9
 
 
Blades
26
8.8
 
 
Bladelets
3
1.0
 
 
Core trimming elements
5
1.7
 
 
Overpass
2
0.7
 
 
Burin spall
1
0.3
 
 
Chunks
44
15.0
 
 
Chips
5
1.7
 
 
Total debitage
252
85.7
 
 
Tools
31
10.5
6
66.7
Cores
11
3.7
3
33.3
Total
294
100.0
9
100.0
 
Table 2. Distribution of core types
 
Excavation
Surface finds
Core type
N
%
N
%
Centripetal Levallois
6
54.5
2
66.7
Bidirectional Levallois
2
18.2
1
33.3
Irregular
3
27.3
 
 
Total
11
100.0
3
100.0
 
Table 3. Tool type distribution
 
Excavation
Surface finds
Tool type
N
%
N
%
Retouched blade
6
19.4
1
16.7
Retouched flake
6
19.4
2
33.3
Scraper
5
16.1
3
50
Side scraper
3
9.7
 
 
Awl
2
6.5
 
 
Denticulate
2
6.5
 
 
Notch
6
19.4
 
 
Unknown
1
3.2
 
 
Total
31
100.0
6
100.0
 
 
Although the hilly terrain is relatively inaccessible and affords little sustenance, the available raw material is easily collected and quarried and was thus favored by the local population in the Middle Paleolithic period. Nevertheless, the excavation indicates that the site was only sporadically occupied, as most of the finds were recovered from the surface.