The excavation was located on the bedrock sloping gently down to the north, immediately southwest of Khan Sha‘ar Ha-Gāy. The hard chalk rock, containing flint, and covered by a soft chalk nari layer, belongs to the Mount Scopus rock formation. At the outset of the excavation, the area was uncovered with a mechanical digger. The excavation exposed a rock-cut tomb, a quarry, two cisterns, a surface runoff collection system, agricultural terrace walls, a cave and rock-cuttings (Fig. 2).

The Sha‘ar Ha-Gāy Khan, controlling the important junction of the Jerusalem–Jaffa road (modern Road No. 1) and the road leading to the Judean Shephelah low hills (modern Road No. 38), was built in 1873, by the Ottoman governor of Jerusalem. In a previous excavation at Khirbat Harsis, in the area of the khan, two caves, serving as storage facilities in the Ottoman khan, were excavated; these caves damaged earlier rock-cut elements, including a cistern and a burial cave (Greenhut 2004). An excavation and survey conducted at Khirbat Harsis, north of the khan, revealed a storage cave, agricultural terraces and rock-cuttings that were probably part of the agricultural hinterland of nearby Deir Ayub. The sherds retrieved reflect human activity at the site in the prehistoric period, and in the Iron Age and Persian, Hellenistic, Roman, Mamluk and Ottoman periods (Ein Mor 2010; Kagan 2012; Landes-Nagar 2013; Monnickendam-Givon 2013). At Kh. Khatula, west of the khan, an Ottoman-period lime kiln was excavated (Radashkovsky 2015), and numerous cupmarks and rock cuttings were discovered (Grosman and Goren-Inbar 2007).

Rock-cut tomb. In the southeastern part of the excavation area, a rectangular rock cutting surrounded by a shallow groove, presumably intended for a cover stone, was uncovered (L20; 1 × 2 m; Fig. 3). This was probably part of a hewn cist tomb from the Late Roman–Byzantine period (fourth–seventh centuries CE; Avni 1997:28–29: Type 3.1). However, since the rock cutting was not excavated, it cannot be ruled out that it was an entrance into an underground space, possibly a ritual bath or cave, similarly to a Late Roman–Byzantine period rock-hewn corridor that led into a burial cave at nearby Khirbat Harsis (Greenhut 2004).
Quarry. A shallow quarry (L6; 5 × 5 m, depth 1 m; Fig. 4), with rock cutting marks indicating the technique employed, and the dimensions of the detached stones (average size 0.5 × 0.6 × 1.5 m), was exposed. The quarry apparently extended further west, where it was damaged by construction of the khan, and by an adjacent cistern (L17; see below).
Cistern. A rectangular cistern (L17; 3 × 10 m, depth c. 5 m) was uncovered next to the khan. Its upper part was built, and its lower part rock-hewn, and it had an opening on the northeastern side. Gutters descending from the roof of the khan to a plastered surface channeled the rainwater into the cistern (Fig. 5). A settling basin was discovered near the cistern. A brick from the Motza brick factory was found in the cistern’s foundation trench (L4; Fig. 6). Since the Motza factory operated from 1880 to 1882 and from 1900 to 1920, the construction of the cistern may be dated at the earliest to the 1880s, a few years after the khan was built.
Rock-cut cistern and surface runoff collection system. A rock-cut cistern (L12; diam. 0.5 m, depth of cistern neck 1.3 m, depth of cistern c. 4.0 m), was exposed, fed by a surface runoff collection system (Fig. 7). The system comprised two rock-cut settling basins (L9, L10), a shallow channel (L11) that led the water into the cistern, and an agricultural terrace wall (W19), that among other functions, retained eroding soil on the slope, prevented it from clogging the water system.
Agricultural terrace walls. Three terrace walls (W18, W19, W23; Fig. 8) were discovered, built on the bedrock of various-sized local fieldstones. Shallow rock-cuttings exposed around the walls led surface runoff to the agricultural terraces.
Cave. In the northwestern part of the excavation area, a cave with a rounded interior was exposed (L16; diam. c. 3 m, exposed height c. 2 m; Fig. 9). The eastern wall of the cave was completed with a construction of roughly dressed stones; in the northern wall, an opening (0.5 × 1.0 m) featured dressed doorposts and a lintel stone.
Rock cuttings. A rectangular rock cutting (1.0 × 2.0 m, depth 0.5 m), a cupmark (L22; diam. 0.3 m, depth 0.1 m; Fig. 10), and an ovoid rock cutting (L8) were exposed. Signs of the quarrying of flint cores were observed (L21; Figs. 11, 12), similar to marks previously identified in the area (Grosman and Goren-Inbar 2007). A recent rock cut (L13), where the hard chalk layer was removed by a dynamite blast, also damaging Channel 11 (see above), was also observed.
Despite the limited excavation area, examples of various agricultural and industrial installations characteristic in the hilly regions of Israel, were exposed. Due to the extended use of agricultural installations over a long time, it is difficult to determine their period of construction and use. It may reasonably be assumed that their final use was in the late Ottoman period, at the time of the construction and use of the khan. The cave may have been used for burial, but it cannot be dated; the burial cave opening resembles a cave previously exposed in the vicinity that was dated to the Late Roman–Byzantine period.