The site environs were surveyed in the past as part of the Jerusalem Survey Map (Kloner 2003: Sites 49, 50). About 1 km east of the site are the remains of a fort dating from the Hasmonean through to the Byzantine periods (second century BCE–seventh century CE; Tzaferis 1974). About 0.5 km west of the site lie the remains of an extensive settlement dating from the Neolithic period to modern times (‘Ad and Eirikh-Rose 2021; Vardi and Khalaily 2021).
The present excavation comprised two areas (Fig. 2), located north and south of a large reservoir (external dimensions 11.0 × 21.2 m, depth c. 5 m down to silt accumulations). The northern area revealed a section of the reservoir’s wall, while the southern area revealed a plastered channel along the reservoir’s wall and the remains of an ancient road from the Roman and Ottoman periods.
 
The reservoir walls are wide (c. 1.5 m). A staircase (width c. 0.4 m) built against the western wall reaches the bottom of the reservoir. No inlet for water was identified, apart from a modern iron pipe installed in the southwestern corner of the reservoir.
 
The Northern Area (L12; 1.5 × 5.5 m) was reduced during the excavation to half its original length (L19). The excavation exposed part of the outer face of the northern wall of the reservoir (W6; Fig. 3), revealing three construction phases. The two lower courses, which were constructed of fieldstones, are attributed to the early phase; they protrude northward 0.10–0.15 m relative to the courses above them. Three courses, each about 0.35 m high and constructed of medium-sized roughly dressed chalk blocks, are attributed to the middle phase. The three upper courses, each about 0.2 m high and similarly constructed of medium-sized dressed chalk blocks, are attributed to the late phase. No pottery was found beside to the wall, and no change was observed in the soil composition, making it impossible to date the three phases.
 
The Southern Area. Trial trenches opened south of the reservoir prior to the excavation exposed a wall (W4) running parallel to the reservoir’s southern wall (W5). Between the two walls was the upper part of a water channel, which was plastered to its full height. Two excavation squares exposed the continuation of W4 (exposed length c. 12 m, width 0.8 m, height 1.8 m), which was constructed of dressed stones. Two segments of the water channel (L10 and L13; total length c. 10 m, width c. 0.5 m, depth c. 0.9 m; Figs. 4, 5) were fully exposed. The channel was found completely full of stones and silt but devoid of finds.
Adjacent to the southern face of W4 were scant remains of an ancient road, delineated on the north by a row of flat fieldstones (width 0.2 m; Figs. 6, 7) embedded in a layer of compacted earth. At the western end of this row was a large, dressed stone (Fig. 8), different from the other stones of the row. Three layers of earth and stones (L14–L16) were identified one above the other, apparently representing repairs to the road. The road seems to have been damaged following the digging of a trench for a main sewage line that passes under the modern road. It may have been part of an Ottoman road that was built over a Roman road; Charles Warren witnessed its construction and documented it in the 1860s (Warren 1876:308–309). According to Warren, the road builders laid a layer of earth mixed with stones that was compacted with a hand roller. The large, dressed stone identified in the row along the road may be a curbstone preserved from the Roman period.
 
The excavation yielded no finds that could date the reservoir and the adjacent channel. These features do not appear in the PEF map from the 1870s, but the reservoir is visible in an aerial photograph taken by the German Air Force during World War I, and it appears on the 1:20,000 map from the British Mandate period (Ain Karim sheet; Fig. 9). According to this map, the channel’s source was a few hundred meters east of the reservoir, and it continued west toward Nahal Soreq (Wadi Tulma). It is therefore possible that the reservoir and the channel, in their current form, were built between the 1870s and the beginning of World War I. However, they may have been built on top of similar earlier installations, as indicated by the lower courses of the reservoir; these ancient remains were not identified by the PEF, possibly because they were hidden by silt and vegetation. The channel may have been related to the Roman road, draining it from rainwater. Similar channels were commonly installed alongside roads in the Roman Empire (Chevallier 1976:88). Furthermore, they are mentioned in contemporaneous sources, such as the poetic description of a paved road with an adjacent drainage channel (lacuna) written by the Roman poet Statius (45–96 CE; Statius 1:220). Although the excavation did not yield a connection between the reservoir and the channel, both were probably built at the same time as part of one system. The reservoir was apparently fed by water flowing in the channel during the rainy season, with the water likely diverted from a secondary regulation channel whose remains are yet to be found. The extension of the channel toward Nahal Soreq probably served to divert the water for field irrigation in that area and may have even supplied water to the nearby settlement of Moza.
The road segment exposed in the excavation is apparently part of the ancient ‘Roman Ascent’, which was laid under the Hasmoneans to connect Yafo and Jerusalem and was enlarged in the second century CE; its remains are visible today 200–300 m east of the excavation (Fischer, Isaac and Roll 1996:95). It seems that the Hellenistic- to Byzantine-period fort uncovered about 1 km east of the site (Tzaferis 1974) commanded the stretch of road exposed in the excavation. In 1867, the road underwent major renovations on orders of the Ottoman Sultan ‘Abd al-‘Aziz, to allow the use of coaches. The work was carried out under the direct supervision of Nazif Pasha, the governor of Jerusalem (Ben-Ariyeh 1979:127–129). The renovation work was carelessly executed and necessitated frequent repairs after winter rains washed out parts of the road. The most extensive repairs were carried out in 1872, again under the direct supervision of Nazif Pasha. During 1886–1889, the Ottoman authorities laid a new road, which encircled Har Ha-Menuhot from the north, and the Roman Ascent was abandoned (Schick 1889). In recent decades, the remains of the abandoned road were heavily damaged following the laying of a main sewage pipe along most of its course.