Area A (Figs. 1–3), a long excavation trench (4.5 m deep), was opened along a northeast to southwest axis. The southwestern part of the trench (Fig. 2:1–1) was excavated beneath modern walls (W300–W302) that are set into a series of accumulations (L100, L104, L108). A late Ottoman period cup produced in Kütahya in western Turkey (Fig. 4:2; Anna de Vincenz, pers. comm.) and a spindle whorl (Fig. 4:3) were found in Accumulation 104, and a kiln separator (Fig. 4:4), that probably dates to the Mamluk period, was recovered from Accumulation 108.
Approximately 2.5 m from the southwestern end of the trench, along its northern balk, was a core of a wall (W307; exposed length 2.5 m, width at least 1 m, preserved height c. 1.5 m; Fig. 5), comprised of medium- to small-sized fieldstones and rubble. The southern face of W307 had been robbed out in antiquity; a robber trench (width c. 0.4 m) ran along its southern side, separating it from a flagstone pavement (L113) that continued southward, beyond the excavation trench. A fals of el-Nasir Abu el-Sa‘adat Faraj (1399/1400 CE, minted in Aleppo; IAA 143126) was retrieved from Pavement 113. The flagstones, roughly rectangular in shape (e.g., 0.3 × 0.5 m), sealed a fill (L116), in which a single Iron Age II jar sherd (Fig. 4:1) was found.
The southwestern end of W307 abuts a wall running in a northwest–southeast direction (W306; width c. 0.90–1.35 m; Figs. 2:1–1; 6) that carried the eastern side of a barrel vault. Its western face was constructed of well-dressed and smoothed boulders; its eastern face—of roughly dressed, medium-sized stones. The vault seems to have covered a passageway that ran in a general north–south direction and continued beyond the limits of the excavation, and whose western side was cut by a modern wall (W301), making it impossible to determine its dimensions. The plaster floor (L114) of the passageway was covered with an accumulation of earth and rubble (L112) that filled the entire height of the passageway.
A water channel (L112A; internal width c. 0.3 m, external width 1.35 m, depth 0.35), was set into Accumulation 112 along a north-south axis, about 0.25 m from the top of the accumulation and one meter above Floor 114. The channel, which extended the entire length of the excavated portion of the passageway, was coated on the inside with a layer of plaster and was covered with stone slabs (e.g., 0.30 × 0.45 m).
A plaster floor (L105; thickness up to 5 cm; Fig. 2: 2–2, 3–3) in the northeastern part of the trench was covered with an accumulation (L101). Near Floor 105, the accumulation contained Ottoman-period bowls (Fig. 7:6, 7) and jars (Fig. 7:8, 9). Beneath Floor 105 was an accumulation (L106) with Mamluk-period pottery—a basin (Fig. 7:1), bowls (Fig. 7:2–4) and a jar (Fig. 7:5)—and a bronze coin (IAA 143123) of Autonomous Cyprus or Sicily(?), dated to 1200 CE.
Accumulation 106 came up to the upper, and possibly later, courses of an east–west wall (W305, W305A; exposed length 3 m, width at least 0.7 m; Figs. 8–10). The wall, built of dressed boulders, was preserved six courses high (2.5 m). An opening (2 m wide) in the wall was blocked with roughly dressed boulders (see Fig. 9). A flagstone pavement of a courtyard or a wide street (L125A; Fig. 3:3–3) ran up to W305A. The flagstones were non-uniform in both size and shape (e.g., 0.25 × 0.40 m; 0.10 × 0.15 m); two especially large rectangular slabs (0.15 × 0.75 × 0.80 m on the west; 0.15 × 0.72 × 0.90 m on the east) were incorporated into the pavement right in front of the blocked entrance in W305A (see Fig. 8). This wide street or courtyard was evidently the continuation of Pavement 113 in the southwestern end of the excavation trench; the difference in elevation between the two is due to the mild eastward slope of the original surface.
Pavement 125A was covered with an accumulation of soil and rubble (L109, L110, L123; thickness 0.65 m). These accumulations contained Mamluk-period pottery, including a bowl (Fig. 11:1; L110), a basin (Fig. 11:4; L123), jars (Fig. 11:2, 5; L110, L123 respectively), jugs (Fig. 11:6, 7; L123), the base of a jar or a jug (Fig. 11:3; L110) and a stone weight (Fig. 11:11; L109). A jar (Fig. 11:8) and jugs (Fig. 11:9, 10), from the Mamluk period as well, were retrieved from the fill (L125) that lay right over Pavement 125A.
A water channel (L115; 0.35 m wide, 0.3 m deep; see Figs. 8, 9) covered with roughly hewn stone slabs (e.g., 0.15 × 0.30 × 0.50 m; Figs. 8–10) lay 0.5 m beneath Pavement 125A. The fill (L128) within Channel 115 included a small amount of residual, late first-century to early second-century CE pottery, such as a basin (Fig. 12:1) and jar (Fig. 12:2). The water channel was set in an accumulation (L129; thickness c. 0.5 m; Figs. 2:2–2; 10) that lay on bedrock.A few Byzantine sherds, including bowls (Fig. 12:3, 4) and basin (Fig. 12:5), were found in Fill 129.
Area B (3.5 × 4.0 m; Figs. 13, 14). No architectonic remains were found, except for a child’s grave (0.70 × 1.15 m) with covering slabs that was unearthed about one meter below the surface; it was left unexcavated. Three coins were recovered from the fills above and near the grave: a manghir of Sulayman II Qanuni from 1523 CE (Fill 107; IAA 143125); a one-para coin of Mahmud I, from 1730 CE (L107; IAA 143124); and a 40-para coin of ‘Abd el-Medjid from 1858/9 CE (L102).
The recovered pottery and the coins suggest that the area was not occupied before the Byzantine period. In this period, a courtyard or wide street (L113, L125A) was built that led to a building to its northeast (W305A) and a vaulted passage (L114) to the southwest. In the Mamluk period, a wall (W305) was probably erected atop the remains of the Byzantine building, preserving the earlier building’s orientation and sealing its entrance that faced northwest. Also in the Mamluk period, a water channel was installed within the earlier Byzantine passageway. A child’s grave found in the eastern square belongs to a chance burial that was laid later than the Mamluk period, apparently toward the end of the Ottoman period.