Two squares were opened in the excavation area; these were dug to a depth of c. 2 m below the surface, the upper meter being removed by a backhoe. The bedrock was not reached, although it may have been fairly close. The excavated earth was visibly alluvial deposits, presumably from the overflow of the spring waters, and indeed, until recently, the pomegranate orchard was still irrigated by manual channeling from tree to tree. At a depth of c. 1.5 m below surface, a packed surface strip of very small stones was uncovered (length 11 m, width 1.1 m; Figs. 1, 2). It was evident that the stone surface continued northward, southward and westward beyond the excavation limits, while the eastern edge was visible in the excavated area as a fairly straight line, bordered by the alluvial soil. In the packed stone surface were many tiny sherds, dating to the Roman and Byzantine periods. The surface was at a similar elevation to the road, discovered in the adjacent earlier excavation, and it may possibly have been an associated living surface. 
A flimsy row of medium-sized stones was found c. 0.3 m higher than the Roman–Byzantine packed surface, with no associated floor or living surface (Fig. 3). Some Mamluk-period potsherds of a glazed bowl and a hand-painted jug that were found between the stones, permit dating this row to the Mamluk period. It may have been part of an agricultural field marker.
The limited finds in the excavation reflect activity at the site during the Roman–Byzantine and Mamluk periods.