During November 1999 an underwater survey was conducted in the area between the Sharon and Sidni ‘Ali beaches in Herzliyya (License No. G-13/1999*; map ref. NIG 18146/67680; OIG 13146/17680), following a report by Z. Chen of stone objects lying on the seabed. The survey, on behalf of the Marine Archaeology Unit of the Antiquities Authority, was directed by E. Galili and J. Sharvit, with the assistance of D. Moskowitz, A. Ya‘aqobowitz, H. Sal‘i (divers), D. Peled, J. Turgeman (logistical assistance) and S. Ben-Yehuda (drafting and drawing).
A shortage of sand was created in the wake of the marina construction and a load of basalt millstones and copper-alloy nails that originated in a shipwreck were exposed. The finds were mapped and photographed and the site was surveyed by divers with the aid of an underwater metal detector. One upper and one lower millstones from the site were taken to the Antiquities Authority.
The site, located at a depth of c. 4 m, is c. 100 m from the shoreline and c. 300 m south of the Appollonia anchorage (ESI 10:119). A cargo of donkey millstones—six conical lower stones and five biconical upper stones—was scattered along the flat rocky seabed. Three of the upper stones and the six lower stones were concentrated in one dense group across an area of c. 6 × 6 m. One of the upper millstones was c. 5 m east of the group and another upper millstone was c. 30 m southeast of the group (Figs. 1, 2). The sides of the lower stones (diam. 0.8 m, height 0.85 m, weight of each 395 kg; Fig. 1, No. 2) are slightly concave. Their upper part is shaped like a truncated cone and a shallow depression is in their bottom. The upper stones (diam. 0.6 m, height 0.57 m, weight of each 250 kg; Fig. 1, No. 10) have two square projections (‘ears’) that are approximately opposite each other and the characteristic perforation is missing. It is apparent that the stonework on the millstones is incomplete. The absence of perforations in the ‘ears’ and the roughness of the interacting parts indicate that the millstones were never used. Several fragments of bag-shaped jars with small loop handles were discovered near the stones; they probably belonged to the ship whose cargo included the millstones.
Copper-alloy (bronze?) nails and parts of nails were discovered between the stones and near them, below a layer of sand (c. 20 cm thick; Fig. 3). All the nails have a shank with a square cross-section. Ten of the nails are complete and the three broken ones include two points of nails and a fragment of a nail with a head. The nails have a flat, mushroom-shaped head and all are bent and distorted. It seems the nails were twisted by the forces created in the disintegrating vessel wrecked on the shore. The intact nails can be divided into three size groups: three are 8–9 cm long (Fig. 3:A), five are 10–12 cm long (Fig. 3:B) and two are 13–16 cm long (Fig. 3:C, D). The maximum thickness of the nail’s shank below the head is 5–6 mm.
Some twenty meters southwest of the millstones an iron anchor (2 m long) with a rope ring (20 cm long) was discovered. The anchor stock found inside the anchor shank was ready for anchoring—one arm was embedded in the seabed and the other was broken. The crown of the anchor faced west and the ring faced the beach. The anchor probably belonged to the assemblage. The nails and the cargo are indicative of a merchant boat (c. 20–25 m long) that carried millstones from a region where basalt is located near the coast, probably the northern Levant (Syria, Turkey or possibly from farther away in the Mediterranean Sea) and was wrecked on its way to the southern shore of the country or to Egypt. When ancient boats were sailing empty they carried a load of stones (ballast) for stability. Sometimes the seamen would swap the ballast for a paying-ballast. It is reasonable to assume that after unloading the millstones the seamen planned to return with a cargo of Egyptian wheat. The finds indicate that the boat dates to the first–fourth centuries CE.