The Installation
Two phases were identified in the installation (8.2×8.2 m; Figs. 1, 2). Phase 1, the first, dated to the Late Byzantine period (sixth century CE) and the Phase 2 dated to the transition from the Late Byzantine to the Early Islamic periods (seventh century CE).
Phase 1. The installation was enclosed by walls (W10–W13) and divided into two equal cells (I, II) by a north–south aligned wall (W14; Fig. 3). Wall 10 (length 8.2 m, width 0.9 m, height 0.11 m) was built of neatly dressed stones (0.11×0.33×0.55 m) that were incorporated in an outer face of debesh and small fieldstones (0.06–0.09×0.10×0.18 m); its inner face was coated with white hydraulic plaster. Walls 11–13 were built in a similar manner to W10. Walls 12 and 13 were damaged by modern activity. It seems that W11 was preserved a single course high, as seen in a probe trench east of the wall (L136).
A circular vat (L141; diam. 1.17 m, depth 0.55 m) was discovered in Cell II, slightly off center to the southwest. Hydraulic plaster mixed with grog, which rendered it a reddish pink hue, was applied to the sides and bottom of the vat. A smaller shallow pit (L142; diam. 0.73 m, depth 0.1 m; Fig. 4) whose sides and bottom were coated with similar hydraulic plaster was cut in the floor of Vat 141. The use of hydraulic plaster and the shape of Vat 141 and Pit 142 suggest that the installation was used in the production of liquids, such as wine or oil. The reduced size of Vat 141 indicates that the installation was initially used to produce large amounts of the product and was later adapted to produce smaller quantities.
Phase 2. Mosaic pavements were installed in Cells I and II above thick foundation layers and cells were added onto the exterior of the installation. Cell I was paved with a slovenly set white industrial mosaic (L120; size of tesserae 2×2 cm; Fig. 5) whose state of preservation is mediocre and its northern part was severely damaged. The northwestern part of the floor was repaired with fragments of pottery vessels, mostly jars that dated to the Late Byzantine period; the multiple repairs to the floor points to its prolonged use. A foundation was discovered in a probe (L133), opened in the north of the cell. The top of the foundation was hydraulic plaster mixed with grog (thickness 4 cm), below it was sandstone with inclusions of charcoal and shell (thickness 0.16 m) and the base consisted of small fieldstones (thickness 0.2 m); these levels are technical and have no chronological significance.
A mosaic pavement (length 10 cm) was preserved in just the southwestern corner of Cell II, whose corners were somewhat rounded due to the hydraulic plaster that lined them. A bedding (L132) similar to the one in Cell I (L133) was exposed in the rest of the cell’s area (L134).
Three small cells that abutted its walls were exposed around the installation. Two cells were added to the south of Wall 10: the western (L144; 1×1 m, depth 0.19 m) was square and built of small fieldstones (0.2×0.2×0.2 m) with a floor of hydraulic plaster mixed with reddish pink grog. The eastern cell (L117; 1×1 m, depth 0.41 m), whose interior was coated with pale pink hydraulic plaster, was connected to Cell II by a lead pipe (diam. 8 cm, length 0.5 m; Fig. 6). A circular cell (L143; width c. 0.9 m), built of small fieldstones (0.14×0.15×0.14 m) and coated with white hydraulic plaster on the interior, was added to the west of W11. The cells were apparently used for storing liquids.
The installation is associated with the production of liquids and its plan resembles that of winepresses from the Byzantine period. The use of auxiliary cells and lead pipes is also known from the wine industry.
 
Work Surfaces
The foundation of a floor (L116, L128; Fig. 7), composed of small and medium fieldstones, was exposed north of and c. 6.5 m from W12; white plaster, probably the remains of a floor, was preserved above it in some places. A similar floor foundation (L118), damaged by modern activity, was exposed east of W13. It seems that the paved areas around the installation were used as work surfaces.
 
Wall Remains
A wall (W15; length 2.5 m, width 0.6 m, height 0.6 m) oriented east–west was exposed northeast of the installation. It was built of dressed kurkar stones and small and medium-sized fieldstones were incorporated in it, without mortar. A cluster of fieldstones to the east of W15 was probably collapse that originated from this wall, which was partly preserved due to modern damage (L130).
A wall foundation hewn in kurkar (W16, length 3.4 m, width 0.6 m, height 0.4 m), whose upper courses did not survive, was exposed east of the installation. A floor bedding (L131) of small fieldstones was exposed along the southern side of the foundation.
Walls 15 and 16 could not be integrated in the overall architectural plan of the site due to their poor state of preservation.
 
Habitation Level
A habitation level that consisted of pale gray soil and was overlaid with fragments of potteryvessels, including jars, bowls and jugs from the Late Byzantine period, was excavated in Squares F1 and F2 (not on the plan). It seems that this was the edge of the site where minimal activity had transpired.
 
The Finds
The probe trenches excavated in and around the installation contained mostly jar fragments (Fig. 8:1, 2) and a jug (Fig. 8:3) that dated to the Byzantine period. The accumulation that covered the installation and the work surfaces yielded fragments of bowls (Fig. 8:4, 5), jars (Fig. 8:6–8), saqiye vessels (Fig. 8:9–11), a jug (Fig. 8:12) and an amphora base (Fig. 8:13) that dated to the Late Byzantine–Early Islamic periods.
Three coins were discovered in the accumulation over the installation; a bronze coin of Alexander the Great (336–323 BCE; IAA 119151; Fig. 9) and a copper Abbasid fals (ninth century CE; IAA 119152) were identified, but they are insufficient for dating the remains.
The glass finds included mostly modern material from disturbed loci.
 
The size of the installation indicates that large quantities of probably wine were produced. The nature of the installation’s construction and its maintenance reflect its prolonged use. The fact that it ceased to be used during the transition from the Byzantine to the Early Islamic periods should be viewed in light of the decline in wine production at this time.