During August–September 2002, an excavation was conducted in the Shimshit Forest (Permit No. A-3705*; map ref. NIG 2331–7/7384–8; OIG 1731–7/2384–8), aiming to expose a Roman road and a quarry. The excavation, on behalf of the Antiquities Authority and financed by the Jewish National Fund, was directed by L. Porat, with the assistance of V. Essman and V. Pirsky (surveying and drafting), H. Tahan (pottery drawing), D. Avshalom-Gorni (ceramic consultation) and D. Syon (numismatics).
Two areas were opened on the northern slope of the site. A section of the Roman road from Legio to Sepphoris (Fig. 1), between Mile XIII and Mile XIV (Yediot
25: 175–186) was exposed in Area A. A winepress and agricultural installations that were hewn in a stone quarry after it was no longer in use were excavated in Area B. A previous excavation that revealed remains of an Ottoman floor and farming terraces was located between the current excavation areas (HA-ESI 118
). Potsherds that ranged in date from the Iron Age to the Byzantine period were recovered from a probe that was cut in an adjacent natural cave.
Area A. A section of the Roman road (length 12 m, width c. 6 m; Fig. 2) paved with small fieldstones and oriented east–west was exposed. Large and medium-sized curbstones (length 0.30–0.55 m) were preserved along its northern edge and a row of medium-sized stones ran parallel to and south of them. The southern edge of the road was supported on a bedrock ledge and had no curbstones. A few jars fragments from the Roman period and a coin of ‘Abd al-Hamid I (1774–1789) were discovered.
Area B is located on a nari slope where ashlar stones were first hewn, followed by a winepress (Fig. 3) that consisted of a treading floor (3.5 × 5.0 m) and a collecting vat (1.5 × 1.5 m, depth 1.89 m) to its west. The square treading floor (L24) was coated with a thick layer of pink plaster embedded with potsherds and had a recess for the press bed (L26; 1.1 × 1 .2 m, depth 0.45–0.53 m; Fig. 4) hewn in its center. The press bed, which was not found, had a curved western side and three stones found there were used to stabilize it. A channel covered with stone slabs (length 1.2 m, width 0.4 m) led from the press bed to the collecting vat, whose walls were coated with a layer of gray plaster (thickness 5–10 cm) and six hewn steps along its northern and eastern walls descended to its bottom (Fig. 5). The floor of the vat was paved with a coarse white mosaic (c. 30 tesserae per decimeter) and a circular plastered settling pit (diam. 0.5 m) was cut in its northwestern corner. Three square cavities that may have been used to support an awning and a hole to hitch a donkey to were in the southern wall of the treading floor. Another hole was noted in a large stone lying west of the winepress. The potsherds embedded in the plaster dated to the Roman period (second–fourth centuries CE), whereas the potsherds recovered from the collecting vat (Fig. 6) dated to the Byzantine period (sixth–seventh centuries CE). It seems that the winepress was established in the Late Roman period and was used until the end of the Byzantine period.