Three areas were excavated (A1–A3; Fig 3). Partial remains of two structures were found in Areas A1 and A2. Both had courtyards with shafts leading to underground chambers (Fig. 4). In Area A3, a water reservoir was exposed; it was converted into the main underground ‘concealment hall’ of the hiding complex. The unexcavated parts of the complex were surveyed and plotted, as were a water reservoir and a cistern that were found near the remains of the structures (Fig 3: F4, F5). Four burial caves situated 350 m northwest of the site were also surveyed, but only one was fully plotted.
 
The structure in Area A2 yielded cooking pots (Fig. 5:1, 2), a cooking jug (Fig. 5:3), storage jars (Fig. 5:4–7), flasks (Fig 5:8, 9), a miniature bottle (Fig. 5:10) and oil lamps (Fig. 5:11, 12), all typical to the end of the Early Roman period. A coin found under a collapse in the courtyard of this structure dates to the second year of the first revolt against the Romans (67/68 CE). The last occupation phase at the site would therefore appear to date before the destruction of Jerusalem in 70 CE.
 
The hiding complex extends over an area of 550 sq m. During its construction, four earlier installations were intentionally sealed and converted into rooms for concealment. Unlike the majority of the hiding complexes known from the Judean Shephelah, the cross section of the burrows in this complex are not uniform in shape and measurements, as the burrows follow natural fissures. The hiding complex does, however, exhibit all the known features and characteristics of similar complexes found in the Judean Shephelah, the Hebron and Bet El Mountains and the Galilee (Kloner and Zissu 2009): shafts, narrow passages and changing levels were all used in order to secure the system from intruders. The sealing stone for the entrance of a burrow leading to the hiding complex was found at the bottom of a shaft (Fig 6). Finds from within the complex were dated to the first centaury CE, although some pottery types could be assigned to the first half of the second centaury CE.
 
The reservoirs that were surveyed and those that were incorporated into the concealment system suggest that the site was a village. The limited architectural remains that were unearthed at the site reflect an occupation phase, dated by ceramic and numismatic finds from secure loci to the end of the second century BCE and the first centaury CE; the settlement came to an end during the Jewish War. The finds indicate continuous occupation of the rural settlement throughout the period: all the major coin groups are in evidence, without gaps or peaks, and the relative quantities of the various types correspond to their production quantities. The scope of the excavation and the fact that some pottery types could be assigned to the second half of the second century CE make it impossible, however, to rule out a later phase at the site. The exact time frame in which the hiding complex was in use remains uncertain. To date, it is the hiding complex closest to Jerusalem from the west, and the only one surveyed and excavated in the western Jerusalem mountains.
 
Further remains that are probably part of the same site were unearthed in an excavation conducted in 2014, approximately 200 m to the north of Areas A1 and A2 (Permit No. A-7051). This excavation revealed the remains of a few rooms, in which three phases dating between the first century BCE and the second century CE were discerned. Among the finds was a hoard of 114 coins dating to the fourth year of the revolt against the Romans (P. Betzer and E. Marco, pers. comm.).
Four Roman inscriptions had previously been found at Abu Ghosh, and attest to the presence of a Roman camp, most probably of a Roman cohort that was established there no later than the year 73 CE (Fischer, Issac and Roll 1996; Eck 1999). Establishing the chronological relationship between the Roman camp and the adjacent village can contribute to our understanding of the events that unfolded in this rural area west of Jerusalem during the two revolts against the Roman Empire (66–73 and 132–135/6 CE).
The village and the Roman camp had a strategic advantage, located as they were in close proximity to the main road to Jerusalem, with abundant springs around and within Abu Ghosh, and access to fertile arable lands in the broad section of the upper Kesalon Valley. This location suggests that the settlement may have serviced pilgrims on their way to Jerusalem during the Hasmonean and Roman periods.