The excavation area (Area U) was opened east of the entrance to the Warren’s Shaft system, was one of numerous excavations that have been conducted in this area, due to its proximity to the Gihon Spring and the many water and fortification systems built around it (for detailed summaries of the water systems and excavations in this area, see Reich 2011). Among the important excavations are those of K. Kenyon, c. 30 m north of the current area, which unearthed c. 30 m of Jerusalem’s eastern city wall from the Iron Age II–III (Steiner 2001). East of the excavation is a complex of rock-cut rooms from the Iron Age II (Vincent 1911; Reich and Shukron 2011). The remains discovered at the site in the previous excavation seasons post-date the Iron Age (for preliminary results of these excavation seasons, see Szanton and Uziel 2015; Dan-Goor 2017; Hagbi and Uziel 2017). These remains covered and sealed several structures that were unearthed in the two excavation seasons presented here.
Eight squares were opened (4 × 35 m), revealing three well-preserved buildings (17044, 17049, 17081; Fig. 2): their walls were as high as c. 2 m, and each structure consisted of several rooms. The buildings, which were built directly on the bedrock, were preliminarily dated to the Iron Age II–III (eighth century BCE to the destruction of Jerusalem in 586 BCE); this dating parallels Strata 12–10 of Y. Shiloh’s excavation in the City of David (Shiloh 1984; De Groot and Bernick-Greenberg 2012). Two main construction phases were identified, and in some of the rooms, sub-phases were also noted, including the raising of floors throughout the Iron Age.
 
The earliest of the three buildings (17081; Fig. 3) was built above a cluster of earlier rock-cut installations, such as cupmarks. The building comprised four rooms: three longitudinal rooms (from north to south: 17129, 17063 and 17130) and one broad room (17043). The rooms were quite narrow, ranging in width from 0.7 m (Room 17043) to 1.6 m (Room 17129). Their walls (thickness 0.5–0.7 m, preserved height c. 2 m) were built of medium-size fieldstones; the lower part of the walls in the northern room (17129) was hewn in the bedrock. A coating of mud was preserved on the building stones in some of the rooms; it bore a design created by fingertips smoothing the mud. A wall (W17081; thickness 1 m) that was uncovered along most of the excavation was broader than the rest of the walls, and it was abutted by several of the earlier walls. This was apparently the wall that delineated the building on the west. Two doorways in Room 17043 were attributed to the early phase of the structure: one was set in the western wall (W17027), and the other was set right across from it, in the eastern wall (W17112). In Rooms 17063 and 17130 each was a series of ten superimposed floors. The earliest floor, from the first phase of the building, was dated preliminarily to the eighth century BCE (apparently parallel to Shiloh’s Stratum 12; Shiloh 1984; De Groot and Bernick-Greenberg 2012). A layer of crushed vessels was found in situ on this floor, attesting to the partial destruction of the building. It was apparent that changes were subsequently made to the building. These included the intentional filling of Rooms 17043 and 17129 with an earthen fill, and the sealing of the doorways (W17027a, W17112a). Rooms 17063 and 17130, however, continued in use: the superimposed floors and the pottery finds discovered in these rooms point to their continued use and to the raising of floors, beginning in the eighth century BCE until the end of the Iron Age. An assemblage of bullae, both iconographic and epigraphic and dated to the Iron Age II–III (Fig. 4; Mendel-Geberovich, Chalaf and Uziel 2020) was found in this building.
The two other buildings (17044, 17049), which belong to the second phase of construction, were built during the Iron Age II–III on a rock terrace above the cluster of rock-cut installations. Building 17044, in the center of the excavation area, north of Building 17081, cut W17081 of the earlier phase. The northern part of this building was damaged by Parker’s Trenches XV and XVII (Vincent 1911: Pl. VI). The excavation partially uncovered three rooms, whose walls (thickness 0.5–0.7 m, max. preserved height 1.8 m) were built of two rows of fieldstones. The western wall of the building (W16014) was abutted by plaster and beaten-earth floors on both the east and west. Pottery from the Iron Age II–III was found on the floors. In a later phase, the southern room was sealed with stones and went out of use. The rest of the building continued in use until the end of the Iron Age, as did the two southern rooms in Building 17081.
In Building 17049, in the northern part of the area, three longitudinal rooms were discovered; only two were excavated. Like Building 17027, this building was built over earlier rock-cut installations, as well as over a natural north–south channel in the bedrock. The building had fieldstone-built walls (width 0.5–1.0 m). The eastern wall (W15047) consisted of one row of stones, some of which were boulders (1 × 2 m). The partition walls (preserved height 2.5 m) were built of smaller fieldstones, which did not rest on the bedrock but rather on a layer of soil (thickness c. 10 cm) devoid of indicative finds. The middle room (17049) had a plaster floor with a particularly thick bedding (c. 0.3 m); it was covered by a layer of destruction that included fallen stones, ash and numerous pieces of burnt wood—apparently part of the room’s ceiling that had burned and collapsed onto an assemblage of jars that stood on the floor of the room. The jar handles were stamped with rosette impressions typical of destruction strata dated to 586 BCE (Fig. 5; for the dating of rosette-stamped handles, see Cahill 1995; Ben-Yosef et al. 2017). East of the building was a layer of stones that abutted the building’s eastern wall. The finds discovered in this layer of stones date it to the Iron Age III, after the construction of Building 17049.
 
The three buildings uncovered in the excavation are dated to the Iron Age II–III. Two construction phases may be identified from this period. Building 17081 was built first and was partially destroyed during the Iron Age II. The building then underwent architectural changes, and two more buildings were added (17044 and 17049). Only in one room (17049) was there evidence of the city’s destruction in 586 BCE. The epigraphic bullae uncovered in the excavation, particularly in Building 17081, are typical of the late Iron Age in the City of David (Shiloh 1984). These buildings, outside the route of the Kenyon-Shiloh city wall, indicate that the settlement which spread east of the city wall, into this area of the eastern slope of the City of David hill, continued to exist and develop even after the expansion of the city to the western hill. This conclusion contrasts with the conclusions that emerged from the excavations in other parts of the eastern slope of the City of David, such as Areas D and E, which were abandoned following the westward expansion, and between the walls, where construction ceased (De Groot and Bernick-Greenberg 2012).