During October–November 2003 a salvage excavation was conducted at the Gan Soreq site (Permit No. A-4009; map ref. NIG 17787/65100; OIG 12787/15100), prior to the construction of an industrial zone in the southern part of the Rishon Le-Ziyyon sand dunes, north of Moshav Gan Soreq, following an archaeological survey, directed by L. Barda and an antiquities inspection, supervised by D. Golan. The excavation, on behalf of the Antiquities Authority and financed by the Rishon Le-Ziyyon Corporation for Economic Development, was directed by U. ‘Ad and A. Dagot, with the assistance of A. Bechar and I. Dangor (administration), A. Hajian (surveying and drafting), T. Sagiv (photography), R. Chen (metallurgical survey) and M. Shuiskaya-Arnov (pottery drawing).
A large farmstead, situated on the highest hamra hill in the region, was discovered. The western side of the hill and the site were apparently damaged by work carried out in the past. Fifty-five excavation squares were opened (1370 sq m) in four areas (A–D). A large structure and other building remains were exposed at the top of the hill, in Areas A and B. Two winepresses were discovered on the western slope of the hill, in Areas C and D. Most of the finds from the excavation dated to the Hellenistic period, from the end of the third century until the middle of the second century BCE. In addition, potsherds from the Middle Bronze Age were found in the fill, especially in Area B.
The eastern part of a large ‘courtyard’-type building (22 × 40 m; Fig. 1) was discovered. Its plan is incomplete because the western part of the structure was damaged and the walls extended beyond the boundaries of the excavation. Most of the walls were built of mud bricks and a few were stone-built. Many of the walls had no foundations, whereas a small number of them, particularly in the southern part, were founded directly on kurkar bedrock. The stone walls were built of small and medium-sized kurkar fieldstones (the indigenous stone), bonded with dark mortar; some of the walls were preserved c. 1 m high (W10; Fig. 2).
The building comprised three or four wings, each consisting of several rooms; the fourth wing, postulated in the west, was not preserved. The wings were built around a courtyard, which had two stone-built rooms in its center (W10–W13; Fig. 3) and two floor segments (L124, L133). It seems that some of the stones used for the construction were in secondary use. The walls of the rooms, surrounding the courtyard, were usually built of sun-dried mud bricks (width 0.6–1.0 m). The eastern wall of the building (W15; Fig. 4) was exposed for nearly its entire length of 40 m. The southern wing included two rooms (L144, L146) and a floor section (L134). The eastern wing had two parallel walls (W15, W19) that were connected by a perpendicular wall (W18), suggesting the existence of rooms, whose walls were not preserved due to the kurkar bedrock proximity to surface. Four or five rooms, not arranged in a row as in the other wings, as well as sections of floors (L135, L152), were partially exposed in the northern wing. The various width of the walls in this wing reached as much as 1 m wide in several places. All the aforementioned floors and a section of another floor (L164) east of W15, consisted of mud-brick material, overlaid with fragments of pottery vessels, coins and basalt pounding and milling implements.
Two differently oriented mud-brick walls (W30, W31), which seem to be a later addition to the building, were discovered southeast of it. To their north and east of the building’s eastern wing was a round tabun (L175; diam. 2 m) that may have served as the main oven for cooking or baking.
Some 30 m east of Area A, remains of a hearth (burnt mud-brick material) surrounded with sections of mud-brick walls and numerous fragments of pottery vessels were exposed. These were probably the remains of another building, smaller than the one in Area A. To its west, an inclined surface of mud-brick material, whose northern part was founded on shells, was discovered.
Some 7 m west of the Area A building, a winepress was uncovered. It consisted of a square treading surface (7.5 × 7.5 m), which was dug into earth and coated with several layers of gray mortar and shells. A light colored hydraulic plaster over a bedding of pebbles was applied to it. The collecting vat, whose excavation was not completed, was to its west.
A winepress (Fig. 5) was discovered in the eastern part of the area. It included a treading floor (7 × 8 m) and a globular collecting vat (inner diam. 3 m, depth 1.64 m, 4.5 cu m) with one corner, whose upper part was built of fieldstones. A step was built in the corner to facilitate descending into the vat. A circular settling pit was revealed at the bottom of the vat. The treading floor and the vat were dug into the hamra soil and coated with light colored plaster.
It appears that the farmstead from the Hellenistic period consisted of the main ‘courtyard’-type building, as well as another building and two winepresses. In this period, before the area was covered with sand dunes c. 400 years later, the surrounding area was arable. It is therefore reasonable to assume that other agricultural installations were located within its domain. Three farmsteads of the same period were excavated in recent years in the region, two in the Rishon Le-Ziyyon sand dunes (Permit Nos. A-3062, A-3147; not yet published) and another in the Yavne sand dunes (Permit No. A-3731; not yet published). The importance and benefits of the current building lie in its fine state of preservation. Farmsteads of this type served as the agricultural hinterland for the urban centers in Jaffa and Yavne-Yam, providing them with farm produce.