Ancient Building
Three walls (W8—length 4.2 m, width 0.5 m, height 0.3 m; W12—length 2.1 m, width 0.4 m; W13—length 2.4 m; Fig. 3) were discovered in Stratum 3, the earliest layer. In addition, a section of a mosaic pavement (not drawn) was found c. 5 m west of W12. The walls were built partly on the bedrock and partly on soil fill mixed with small stones. Wall 13 was constructed of ashlars. The mosaic pavement was made of coarse white tesserae (c. 30 per sq dm). A trial trench was dug down to the bedrock south of W8 (L113; 1.5 × 2.5 m, depth 0.3 m), but no ancient floor levels were exposed. The building remains from this stratum were meager and insufficient in order to reconstruct the plan of the building. In the absence of any definite finds derived from this layer it was impossible to date the stratum.
Public Building
Remains of a building (Fig. 4) comprising three construction phases (1–3) were identified in Stratum 2.
Phase 1. In the earliest phase, a rectangular building (8.1 × 16.2 m) was constructed along a southeast–northwest axis. The building had wide longitudinal walls (W1, W2; width 1.2 m) and narrow lateral walls (W3, W6; width 0.7 m) built of two rows of large fieldstones (length 0.8–1.3 m), some of which were roughly hewn, and a core of small and medium fieldstones and soil. The walls were preserved to a height of 3–6 courses (0.8–1.8 m). The building’s walls were oriented slightly differently than those of the Stratum 3 building. Inside the building was a tamped-earth floor (L109) that incorporated the tops of W8 and W12 of Stratum 3. The entrance (width 1 m) to the structure—a threshold built of small stones bonded with plaster (debesh)—was set in the center of W2. The eastern part of W1 was built on top of W13 of Stratum 3, which served as its foundation. The considerable width of the building’s two longitudinal walls may indicate that they supported a barrel vault that supported a second story. Numerous tesserae, most of them white, were discovered in the collapse above the floor level; they serve as additional evidence that the structure had a second story.
Phase 2. A wall (W10; width c. 0.9 m) constructed in the eastern quarter of the building divided the structure into a large western room (1; 5.6 × 10.9 m) and a small eastern room (3; 2.9 × 5.6 m; Fig. 5). The wall was built on the tamped-earth floor of Phase 1. A doorway (width 0.8 m) that connected the two rooms was fixed in its center. The central section of W10 had settled, suggesting that it was built without a foundation. The floor in Room 3 was raised with a layer of tamped earth covered by a layer of crushed and tamped lime (L122; thickness c. 3 cm). A bench (L121; width 0.8 m, height 0.3 m) built of fieldstones along W1, between W3 and W10, extended as far as W10 but no further; this seems to indicate that it was either contemporary with W10 or of a later date.
Phase 3. A trapezoidal room (4; 6.2 × 7.2–7.8 m; Fig. 6) was constructed along the building’s eastern wall (W6). The walls of the room (W7, W9 and W11; width 0.6–0.7 m, height 0.4–0.9 m) were built of small and medium fieldstones. They exhibit a simpler construction method than that of the walls of the earlier phases. A tamped-earth floor (L107) was exposed on both sides of the room; the central part of the room had apparently been destroyed by mechanical equipment in the modern era. A doorway (width c. 0.8 m) consisting of one dressed threshold stone and two embedded doorjambs was set in W11; only the western doorjamb survived. Two pillars (W15—0.6 × 0.8 m, height 0.9 m; W16—0.6 × 0.9 m, height 0.2 m) were erected 5.2 m apart in the middle of Walls 6 and 7. It seems that these pillars, along with a central pillar, which was not preserved, supported one or two arches that supported the roof. In the northwestern corner of the room was a raised surface (L114; height c. 0.3 m) built of a single row of medium-sized fieldstones (W14), which enclosed a layer of soil fill. The surface was apparently built due to the natural slope of the bedrock.
Fragments of pottery vessels dating from the Late Byzantine period (sixth–seventh centuries CE) were discovered in all of the phases of the public building in Stratum 2. These included bowls with inverted rims (Fig. 7:1–3), kraters with a combed decoration below a thickened rim (Fig. 7:4), cooking pots with a flared rim and handles that are drawn from the rim to the shoulder of the vessel (Fig. 7:5, 6), Gaza jars (Fig. 7:7), bag-shaped jars (Fig. 7:8, 9) and jugs that have a ring base (Fig. 7:10).
Animal Pen
In the latest stratum (1), an elliptical animal pen (c. 8 × 11 m; not on the plan), probably for sheep/goats, was constructed on the remains of the public building of Stratum 2. The pen was built in its center with stones that were taken from its walls. The pen also utilized some of the building’s walls and extended northward, beyond it. Several fragments of black Gaza ware were discovered when the pen walls were dismantled, and within the pen was modern refuse. The finds date the pen to the Late Ottoman period and to the time of the British Mandate (nineteenth–twentieth centuries CE).
Ritual Bath and Rock-cuttings
Adjacent to the building, to its west, a rectangular ritual bath (miqveh; c. 4.5 × 5.0 m, depth 3.3 m; Figs. 8, 9) was hewn in the soft limestone bedrock and treated with five or six layers of hydraulic plaster. Natural fissures were discovered in the walls of the bath; these existed when the installation was quarried, were filled with stones and plaster, and were covered over when the bath was plastered. The quarrymen utilized the layer of hard nari above the soft chalk for the bath’s ceiling; however, it eventually collapsed, and its fragments were found inside the bath. A stairway (L129; width 1.1 m) descending from the southeast led into the bath. The upper steps were built of stones, and the lower ones were hewn in the bedrock. The staircase consisted of at least nine steps, of which six were preserved: the upper five—four built and one rock hewn—and the bottommost step, which was hewn. The bedrock descended in a stepped manner on either side of the lower, rock-hewn steps, forming three descending benches (width 0.5–0.9 m, height 0.4–0.5 m). The lowest bench continued along the southwestern side of the bath. The benches were also treated with hydraulic plaster. At the lowest part of the bath was a rectangular immersion pool (L120; 1.7 × 4.3 m); a depression (L126; depth 0.3 m) for the accumulation of silt was hewn in its northeastern part. The surface above the bath, to its north, was hewn smooth with a slight descent to the south (L130) and was probably intended to divert rainwater into the bath. On the bedrock surface above the immersion pool was an opening, through which rainwater could flow into the bath.
An accumulation of earth mixed with fragments of pottery vessels (thickness c. 1.5 m) was discovered inside the bath, under chunks of bedrock that had collapsed from the ceiling. The pottery sherds date from the Late Byzantine period (sixth–seventh centuries CE). These include fragments of red burnished bowls with a ridge below the rim (Fig. 10:1), bowls with a thickened rim (Fig. 10:2, 3), kraters with a thickened inverted rim (Fig. 10:4), a krater with a thickened flaring rim (Fig. 10:5), casseroles with a cut rim and horizontal handles (Fig. 10:6), cooking pots with a flared rim and handles drawn from the shoulder (Fig. 10:7), cooking jugs (Fig. 10:8), Gaza jars (Fig. 10:9, 10) and bag-shaped jars (Fig. 10:11). These ceramic finds postdate the use of the bath, and are most probably from the time when the bath was used as a refuse pit.   
Two square rock cuttings of similar size (L124, L125; 1.5 × 1.9 m, depth c. 1 m) were discovered in a bedrock ledge, c. 5 m south of the bath; a bedrock partition (width 0.6 m) remained between the rock cuttings. The western rock cutting (L124) was delimited on the south by several stones. The purpose of these rock cuttings is unclear.
The artifacts found in the accumulations within the bath, which date from the Late Byzantine period, seem to postdate the use of the bath. As these finds are similar in date to those from the public building, it seems reasonable that the bath functioned as a refuse pit when the public building was in use. The bath may have been contemporary with the early building, discovered beneath the public building. Judging by the size of the bath and its quality construction, it was probably a public miqveh (Reich 1990). The ashlar-built wall and the mosaic pavement of the early building may indicate that it too was a public structure. Its proximity to the public bath suggest that it was a synagogue. The public building from the Byzantine period, however, was not a synagogue, as it did not face eastward and was not contemporary with the nearby ritual bath.