During June 2006, a trial excavation was conducted at Horbat Zafrat (Permit No. A-4819*; map ref. NIG 19705/63653–4; OIG 14705/13653–4), in the wake of damage to antiquities. The excavation, on behalf of the Antiquities Authority, was directed by R. Avner, with the assistance of R. Abu Halaf (administration), V. Essman and T. Kornfeld (surveying and drafting), T. Sagiv (field photography), N. Zak (final plans), Y. Gorin-Rosen and N. Katsnelson (glass) and E. Braun and M. Avissar (pottery reading).
The excavation was conducted along the northwestern fringes of Horbat Zafrat, next to the road leading to the settlement of Newe Shalom (Fig. 1). Prior to the excavation, numerous glass fragments were discerned on the surface. Initially, two adjacent squares were excavated but after reaching a depth of c. 0.75 m, the excavation area was reduced to two small squares (each 2.5 × 2.5 m), aiming to reach maximum depth (Fig. 2). Four strata were exposed: Stratum IV was dated to Middle Bronze II, Strata III, II, to the Early Islamic period and Stratum I, to the Mamluk period. Two potsherds that belonged to a clay lamp and glass fragments, including slag were discovered on surface and dated to the Byzantine period. These finds were not associated with a specific stratum and therefore, Gorin-Rosen and Katsnelson suggested that the glass was debris from a glass workshop that existed somewhere on the site in the Byzantine period.
Stratum IV. Red terra rossa fill on bedrock (Loci 109, 110) was discovered in the northern square. It contained fragments of pottery vessels from MB II, mostly body fragments of jars. Remains of a curved wall (W1; exposed length 1.65 m), built of small fieldstones on bedrock, were exposed. The wall, preserved to a maximum of three courses high (0.4 m), extended westward into an area that was not excavated. Six mud bricks were discerned in the southern section of the square (Fig. 2: Section 1-1). The excavation of the square’s southern balk revealed soil fill (L111) that contained fragments of pottery vessels from the Middle Bronze Age and the Early Islamic period. No complete mud bricks or remains of mud-brick walls were discovered in the balk. The excavation of the square’s western balk (L110; Fig. 2: Section 4-4) exposed mud-brick material.
Stratum III. Overlying the fill of Stratum IV in the northern square was a light gray fill that contained a few small fieldstones and potsherds from MB II and the Early Islamic period. The ceramic finds from MB II included bowls (Fig. 3:1, 2), jars (Fig. 3:3–6) and a body fragment decorated with a herringbone pattern, probably from a jar (Fig. 3:7). The fill in the stratum also yielded a bone plaque, carved with a symmetric scene of two heraldic birds on either side of a solar motif with five sun rays (Fig. 4). Based on its artistic style, the bone plaque is dated to the Early Islamic period.
Stratum II. Gray soil fill (Loci 102–105), which was discovered in the two excavation squares, contained small and medium-sized fieldstones and fragments of pottery vessels, mostly from the Early Islamic period and a few from the Mamluk period. The Early Islamic period finds included two bowls (Fig. 3:8, 9), two jars (Fig. 3:10, 11), a jug (Fig. 3:12) and a lamp fragment (Fig. 3:13). It seems that the ceramic finds from the Mamluk period penetrated from the fill of Stratum I.
Stratum I. Light gray soil fill, composed of wind-borne fine grains (Loci 100, 101), was discovered on surface in the two excavation squares. It contained fragments of pottery vessels from the Early Islamic and Mamluk periods. The finds from the Mamluk period included bowls (Fig. 3:14–15), a krater (Fig. 3:16) and body fragments of a handmade vessel that bore a red-painted decoration (Fig. 3:17).