The excavation (c. 30 sq m) was conducted beneath Wilson’s Arch in Jerusalem, along the Western Wall of the Temple Mount and in the northeastern part of the previous excavations (Figs. 2–4). The excavation’s eastern boundary was the Temple Mount’s Western Wall, and in the west were wall remains that had previously been excavated: a solid cast wall (W4493) that may have served as a dam, and a wall (W4420) of a theater-like structure (Uziel, Lieberman and Solomon 2019: Strata 8 and 6).

The current excavation revealed W4493 to a depth of 3 m. A thick fill of boulders, small stones and soil (thickness c. 9 m) that was laid in two phases was found in the space between Walls 4493 and 4420. The fill’s earlier phase was dated to the late first century BCE–early first century CE; this fill was probably deposited after the construction of the Western Wall and served as a bedding for the street and the drainage channel, which were laid on top of it (Uziel, Lieberman and Solomon 2019: Stratum 7). The later fill served as a bedding for a stepped stone base that supported the stone seats (cavea) of the theater-like structure; this fill dates from the late first–early second centuries CE.

 
Early Roman Period (Fig. 2). The lower of the two fills, ascribed to the early phase, yielded pottery dated mainly to the Late Hellenistic and Early Roman periods (first century BCE–first century CE). The excavation also revealed a wall (W837; length 2.3 m, width 0.7 m) built perpendicular to the Western Wall, on a lower level than that of the foundations of the theater structure’s wall (W4420). The wall was built of medium-sized dressed stones and preserved to a height of seven courses. The lower course of W837 was not founded directly on the bedrock, but rather on a fill of stone chips, probably masonry debris (Fig. 5). It was probably a retaining wall for the earlier fill (L848), which was deposited against the western face of the Western Wall and reached a height of c. 3.5 m above the bedrock. A similar phenomenon of stone walls built perpendicular to the Western Wall and supporting fills that were placed against it is known from the southern part of the Western Wall (Hagbi and Uziel 2015; 2016), although the walls there supported soil fills, not stone fills.
The excavation of the Western Wall’s lower courses revealed that the thirteenth course below the level of the Wilson’s Arch prayer hall floor (Course O in Warren’s excavation; Warren 1884: Pl. XXXIII) was set 7 cm farther west than the course above it. A similar phenomenon was recorded in the course below (Course P in Warren’s excavation; Fig. 6), which was set 10 cm farther west than the course above it. This course is the lowest foundation course in this part of the Western Wall. It was laid in a shallow, smoothed trench (L857) cut into the bedrock (L856). The fill beside Courses O and P was replete with stone chips encrusted with an unidentified bonding material. Furthermore, the area was submerged in water that had seeped in from higher parts of the Old City and collected on the bedrock. The fill beside the foundation course yielded meager, worn finds, including a few potsherds from the Late Hellenistic and the Early Roman periods (first century BCE–early first century CE). In addition, numerous ex situ Iron Age finds were recovered from this fill, but obviously do not  date the excavated remains.
 
Roman Period (Fig. 3). The upper, later fill was composed of dressed stones and some architectural features in secondary use, as well as roughly dressed stones and fieldstones of various sizes. A small amount of soil fill was found among the stones. The finds from the fill are scarce, and include a few potsherds from the late first century CE–first half of the second century CE (cf. Regev et al. 2019) that served as a bedding for a stepped stone foundation that supported the stone seats in the theater-like structure; this foundation was dismantled during the excavation. The continuation of a wall (W4409) uncovered in previous excavations was discovered inside the fill (Fig. 7). The wall, running along an east–west axis, was built of medium-sized dressed stones that were preserved to a height of 13 courses. As the wall abutted the Western Wall in the east and W4420 in the west, it evidently post-dates both these walls. Wall 4409 probably served to retain the fill placed between W4420 and the Western Wall in the later phase. The latest finds from between the stones of W4409 date from the first half of the second century CE, like the finds in the upper part of the fill (cf. Regev et al. 2019).
The narrow foundation trench of W4420 (L811; Fig. 8), which was discovered between W4420 and the cast wall (W4493), is attributed to the same period. The trench was filled in with small fieldstones and soil, like the fill behind W4420. The foundation trench yielded a few worn potsherds from the Roman period. The excavation also found that W4420 was built on a foundation course of large stones, some of them flat—possibly paving slabs in secondary use (see Fig. 8).
 
The excavation beneath Wilson’s Arch revealed two layers of fill, one above the other, placed between the Western Wall and W4493 and W4420 of the theater-like structure. The earlier fill layer was dated to the first century CE; it was probably a construction fill deposited during the expansion of the Temple Mount and the building of the Western Wall. It is impossible to know the original height of this fill and its relation to the rooms that were discovered in the west pier of Wilson’s Arch. The later fill layer served as bedding for the supporting base of the stone seating in the theater-like structure, which was probably built in the late first century–first half of the second century CE.