Area A
Quarry (c. 100 sq m; Figs. 2, 3). Three quarried rock-surfaces were exposed beneath a fill of stones and tamped earth: one in the south (L11, L14, L15) showed marks of hewing and detachment of large rectangular stones (max. size c. 0.5 × 2.0 m); and two smaller ones in the north (L2, L10), with less obvious quarrying marks. A small karstic cave (L16) was exposed in the northwestern corner of the main surface; no datable artifacts were discovered.
Burial Cave. The northern surface of the quarry (L2), was used in a later phase as an open courtyard (c. 1 × 2 m; Fig. 3) in front of the entrance to a burial cave. In the west, the rock face rose to a maximum height of 2 m and two adjacent openings were revealed in it: one is rectangular and enclosed within a decorative stepped frame (L3; width 0.75 m, height 0.65 m; Figs. 4, 5); the other (L8) was lower, and sealed with a roll-stone (diam. c. 0.5 m). The cave was not excavated, and it is unclear if the openings belonged to a single burial cave or two separate ones. A rectangular recess hewn in the floor of the quarry (L12; 0.5 × 1.0 m, depth c. 0.5 m) was exposed slightly to the east.
 
Area B
A large structure (c. 10 × 15 m; Figs. 6–8) was partially excavated, mainly along the tops of the walls, and it is possible to reconstruct its external plan. The outer walls (W1, W6–W8; width c. 1 m) were constructed of two rows of ashlars. A hewn channel (L24, length 0.15 × 3.00 m, depth 0.15 m) was exposed near W8, outside the building. Several stones (L27) were revealed east of W7, possibly part of a floor, or collapsed stones. Part of a wall (W16; length c. 2.5 m) was uncovered inside the structure. Remains of floors made of tamped earth mixed with small stones were exposed in the southwestern (L25) and northeastern (L34) corners inside the building, as well as outside it to the northeast (L28). A square pool (L19; 4.5 × 4.5 m, depth 1.0–1.5 m; Fig. 9) was exposed in the northwestern corner of the building. Its walls (W2–W5) were rock-hewn and built, and with the exception of the southeastern corner, they were almost entirely preserved. Three round steps (L33) were discerned in the northeastern corner of the pool. A perforation (L30; diam. 0.1 m, length c. 0.5 m) in the northern wall, c. 0.5 m from the upper step, was apparently drilled to drain excess water from the pool directly to the fields. The floor of the pool was paved with a coarse white mosaic (size of the tesserae 1.5–2.0 cm); part of the floor was preserved in situ, and many of the tesserae were found scattered in the earth fill inside the pool. A smaller pool (L22; 1.5 × 1.5 m, max. depth 1.2 m) was built along the eastern side of Pool 19; three of its walls (W5, W9, W10) were rock-hewn, and the southern wall did not survive. Sections of a mosaic floor (L22) were preserved. Layers of pink plaster (thickness 1.5–2.0 cm) containing large amounts of grog were applied to the walls and the steps of both pools. An opening (L25; diam. 0.2 m, depth c. 0.15 m; Fig. 6: Section 1–1) at the bottom of the northern wall of the small pool conveyed water to the fields through a rock-hewn channel (L23; 0.5 × 2.2 m, depth c. 0.15 m), which was supported by built walls (W14, W15; length 2 m, width 0.25 m, height 0.4 m). A section of a stone floor (L29) was exposed at the corner between the channel and the small pool. A hewn opening of a large cavity (L7; c. 2.5 × 2.5 m, depth c. 1.2 m) was revealed south of the small pool, next to the wall that did not survive. The edges of the opening had collapsed. The cavity, which was covered with modern debris, was not excavated, and only its upper part was documented; it seems that it was used as a well. On the surface of the walls it is possible to see hewn sections, as well as hollows and depressions that were formed by the karstic dissolution processes. Judging by the little that was left of the hewn wall, it seems that the well was round (diam. c. 1.2 m). Remains of rectangular notches (width 0.4 m, depth 0.3 m) were identified in the eastern and western sides of the well. They were hewn opposite each other, and probably accommodated a chain of jars. A trapezoidal basin with rounded corners (L6; max. length 1.0 m, max. width 0.7 m, max. depth 0.4 m, wall thickness 0.1 m; Fig. 10), made of hard limestone, was discovered to the southwest.
Pottery. The earth fill over the floor of the large pool and the floors of the building contained numerous fragments of saqiye jars (Fig. 11:20–24), as well as sherds of bowls (Fig. 11:1–8), a cooking pot (Fig. 11:9), jugs and jars (Fig. 11:10–19), all dating to the Byzantine and Early Islamic periods.
Numismatic Finds. A floor (L16) was exposed west of W1, and a Roman provincial coin from the second–third centuries CE (IAA 75810) was found below it. A coin dating to 378–383 CE (IAA 75812), was discovered on the eastern side of W7 (L27). A folles (IAA 75811), dating to the first half of the seventh century CE, was found inside the channel (L23) and a post-reform Umayyad fals (697–750 CE, Damascus mint; IAA 75813) was found above the floor north of W6 (L28).
 
The various elements that make up the complex—the building, the two pools and the well—indicate a pumping and storage system of an ancient saqiye well. The pools that were exposed resemble those of the saqiye installation that was excavated at Tel Ashdod (Baumgarten 1999), where the large pool was used to collect the water, and the small one for irrigation. On the basis of the numismatic finds it seems that the installation was constructed in the Late Roman period, and was used in the Byzantine and Early Islamic periods. The ceramic finds date the final use of the irrigation system to the Early Islamic period. No datable artifacts were discovered in the quarry, and the burial cave was not excavated. It was therefore not possible to determine the relationship between the two areas.