The winepress was apparently constructed in a room along the southern perimeter of a large building (estimated size c. 40 × 40 m; Fig. 2); only parts of the building, along its southern and western sides, were investigated. Pottery sherds uncovered in the foundation trenches of the structure, to the west of the treading floor, reveal that it was founded, along with the winepress, in the early Byzantine period (c. 350 CE). The western perimeter of the building was badly damaged, but it appears to have contained a plastered pool opposite the winepress. The pool may have been part of a bathing facility in the structure, a supposition supported by the discovery of at least one fragment of a ceramic tubulus found in bathhouses of this era. Pottery sherds, including the upper part of a Gaza wine jar (Fig. 3), discovered in a room further north, point to the use of the structure in the late Byzantine period (450–650 CE).
The winepress was in use only during the early part of the Byzantine period (350–450 CE). Its stone-paved treading floor (4.0 × 4.6 m) had a well for a screw-press in its center (Fig. 4). The juice from the first pressing emptied eastward, through an outlet in the southern end of the eastern wall and down through a plastered channel, into a small, plastered settling tank (1.0 × 1.2 m), and from there, into a deep, circular collecting vat (diam. 2.4 m, depth 2.1 m; Fig. 5), which had a round depression (diam. 1.5 m, depth 0.5 m) in its floor. The juice from the second pressing of the must in the screw-press flowed into a plastered channel through an opening in the middle of the eastern wall directly into the collecting vat.
In the later part of the Byzantine period (450–650 CE), while the rooms along the western perimeter of the building continued to be in use, the room containing the settling tank and the collecting vat appears to have gone out of use: the vat was filled with earth and debris to provide a dirt floor, and the settling tank was covered with flat stones, creating an even surface for activities. Finds from this area include a stone object (Fig. 6) that may have been part of the screw-press, which was discarded in the fill of the collecting vat; part of a Byzantine sandal lamp (Fig. 7), which was also discovered in the fill inside the vat; and a hair pin made of carved bone (Fig. 8), uncovered in the area of the treading floor, which was apparently abandoned.
The dimensions of the structure and the remains of a possible bathing facility suggest that this may have been a public building, such as a caravansary or a private farmhouse in the early Byzantine period, like that discovered at Oboda (‘Avdat; HA 1976; Negev 1977). The ‘Avdat farmhouse includes the earliest known winepress in the Negev Highlands, dated to the Late Roman/early Byzantine period (late third–early fifth centuries CE). The structure apparently went out of use with the massive earthquake documented in the vicinity in the early fifth century CE (Erickson-Gini 2010:94–95). Our winepress and the winepress at ‘Avdat are an early type constructed inside peripheral rooms of large structures, as opposed to the later type documented by G. Mazor (1981), which are freestanding. The later, freestanding type found in the Negev Highland sites appears to have been modelled on the winepresses known from numerous sites in the southern coastal plain of Israel that continued to function until the end of the Byzantine period, such as Be’er Shema‘ (Erickson-Gini, Dolinka and Shilov 2015:212–219) and the ‘Third Mile Estate’ near Ashqelon (Israel and Erickson-Gini 2013:189–194).