During October 2002, a small excavation was conducted in the garden behind a private house in Metulla (Permit No. A-3754; map ref. NIG 25441/79759; OIG 20441/29759), in the wake of damage to antiquities caused by a backhoe. The excavation, on behalf of the Israel Antiquities Authority, was directed by Y. Alexandre (surveying, photography), with the assistance of E. Belashov (drafting), H. Tahan (drawing) and Y. Stepansky (archaeological inspector of the Eastern Galilee district).
Previous small-scale excavations in Metulla exposed two rock-cut burial caves dating to the Roman–Byzantine periods (‘Atiqot 8:26–30 [Hebrew]). Rock-hewn installations, possibly winepresses, are still visible in the vicinity. This area of limestone rock seems to have been exploited for agricultural processing and as a cemetery in the Roman and Byzantine periods. It was probably located adjacent to a contemporary settlement that had not yet been uncovered. To the northwest of Metulla, at Kh. Aryaq on Har Zefiya (map ref. NIG 2537/7979), a monumental Greek inscription on a marble fragment was observed (HA 51-52:1 [Hebrew]). This fragment may once have been part of a lintel stone from a Roman-period temple or an administrative building.
The excavation focused on cleaning out the part of the damaged winepress that was visible in the section. At the time of the original earthworks a small limestone incense altar was discovered in one of the winepress’ collecting vats.
The winepress consisted of a treading floor and two collecting vats (Fig. 1). The damaged treading floor, paved with white mosaic and visible in the section, was not excavated as it extended into the neighboring plot. The two collecting vats were adjacent to each other and seem to have had similar dimensions (L101—1.0 × 1.5 × 1.6 m; L102—0.8 × 1.4 × 1.6 m). Only the area of the vats visible in the section was exposed and cleaned. The walls of Vat 101 may have been partially rock-hewn, but they were lined with small stones set in mortar and covered with a layer of plaster (Fig. 2). The rock-cut floor sloped down on the west to form a depression at the bottom, c. 0.12 m lower than the rest of the floor, for the settling of the residue. The floor was paved with large white tesserae (Fig. 3). The base of the walls at the joint with the floor was rounded with plaster. Only a small area of Vat 102 was excavated, exposing a wall built of small stones in mortar, but no plaster was extant. Sporadic and non-diagnostic potsherds were collected in the excavation, datable to the Byzantine period.
The small incense altar was found in the fill of Vat 101 when the earthworks were carried out. The house owners exhibited the altar in the vat until the IAA took it for examination. The altar (0.21 × 0.15 m, height 0.3 m; Fig. 4) was of soft white limestone and had a stepped decoration in relief. A circular depression (diam. 0.1 m) on the altar surface was used for the burning of incense. This small incense altar was obviously not in situ in the fill of the collecting vat and may have been thrown in once the winepress went out of use. It is possible that the altar originated in the pagan temple at Kh. Aryaq, on the nearby Har Zefiya. Unfortunately, the site was subject to extensive earthworks that left no remains of the once thriving temple.
A large (height 0.7 m) altar with a similar stepped relief was found at Megiddo, within the Roman military camp of Legio (Permit No. A-3417).
The excavation exposed part of a small mosaic-paved winepress, which was one of several agricultural processing installations visible in this vicinity. It seems that this area was the margins of a contemporary Roman–Byzantine settlement that had not yet been uncovered. The incense altar may be a stray find of a pagan temple that once stood on Har Zefiya.