During January 2006, a salvage excavation was conducted at the site of Ard el-Mihjar, within the precincts of the Ziqim army base, west of Qibbuz Karmiyya (Permit No. A-4527; map ref. 15300–55/61297–302), after ancient artifacts were discovered during earthmoving work. The excavation, undertaken on behalf of the Israel Antiquities Authority and underwritten by the Israel Defense Forces, was directed by V. Nikolsky, with the assistance of H. Lavi (administration), V. Essman and V. Pirsky (surveying), T. Sagiv (field photography), C. Amit (studio photography), I. Lidsky-Resnikov (pottery drawing), N. Amitai-Preiss (seal reading), Y. Israel, D. Varga and Y. Huster (IAA Ashqelon district) and laborers from Ashqelon.
The site is located c. 500 m from the beach and c. 3 km south of Ashqelon. Nine excavation squares were opened where a stone, used as the base of a screw press, was found. Fragmentary remains of a winepress from the Late Byzantine period, in whose collecting vat a lime pit had been installed, were exposed (Fig. 1).
A section of the winepress’ treading floor (L116), which was severely damaged during the earthmoving work, had survived. Its surface was built of small kurkar stones that were overlain with a layer of potsherds, placed vertically alongside each other (Fig. 2). A circular plastered installation (L123) was exposed in the middle of the treading floor. A rectangular depression (L127), intended for placing the stone base of a screw press, was in the center of the installation (Fig. 3). A channel (preserved length c. 1 m), paved with stone slabs, extended from the installation to a square collecting vat, preserved in its entirety (L146; 2.5 × 2.5 m; Fig. 4). The sides of the vat (W1, W2, W4, W5) consisted of cast concrete mixed with small kurkar stones and were coated with plaster. Rough kurkar stones without mortar were stacked along the eastern side of W1, probably to support it, as it was founded on sand, like the rest of the walls.
This complex type of winepress has analogies at other sites in the region, such as Mavqi‘im (HA-ESI 120), east of the Barne‘a quarter in Ashqelon (ESI 13:100–105) and in Ashqelon-Barne‘a (North; HA-ESI 122). These winepresses were attributed to the Byzantine period. It seems that the winepress was no longer in use at the beginning of the Early Islamic period.
At a later phase, a limekiln (L147) was installed within the collecting vat. It consisted of a round pit (diam. c. 2 m; Fig. 4), lined with kurkar stones, which had a slot (width 0.3 m) that served as a vent hole cut in its western wall (W4). The pit was only partially cleaned and was found filled with pieces of slag and burnt kurkar. The potsherds in the fill indicated that the use of the installation continued into the eighth century CE.
The finds consisted of numerous white tesserae that apparently originated in the pavement of the winepress’ components and were discovered scattered throughout the excavation area, as well as kurkar millstones (Fig. 5) and fragments of pottery vessels from the Byzantine and Early Islamic periods, including bowls (Fig. 6:1, 2), cooking pot lids (Fig. 6:3, 4), a cooking jug (Fig. 6:5), Gaza jars (Fig. 6:6–8), a jug (Fig. 6:14) and a flask (Fig. 6:15), as well as bag-shaped store jars from the Early Islamic period (Fig. 6:9–13) and a jar handle, dating to the ninth century CE, which bears a stamped seal impression and was found on the surface (L102; below).
A Stamped Seal Impression on a Jar Handle
The jar handle bears a stamped seal impression that is composed of an Arabic inscription above a decoration of an isosceles triangle, standing on its vertex and open toward the top, with two stripes inside it.
Inscription: من / قصبة
Transcription: min qatzba
Translation: from Qatzba [name of settlement]
It seems that the settlement name Qatzba should be identified with Khirbat Qassāba (H
orbat Suf; HA-ESI 118
: Sites 45-46), located on the banks of Nah
al Shiqma, c. 17 km east of the Ard el-Mih
jar site. Remains of an ancient settlement are known at the site, although surface finds collected during a survey were attributed to the Byzantine period. It appears that an error occurred in the transcription of the Arabic name, recorded during the British Mandate era, whereby it was written with a kāf (ك) and alif instead of a qāf (ق) and without an alif.
No similar inscription that bears the settlement name of Qatzba is known. Among the jar handles from the ninth century CE that bear stamped seal impressions with inscriptions, several site names appear, including Caesarea, Ramla, Nebi Samuel and Rehovot and a few handles bear the name of the Dir Samuil settlement. The settlement names indicate the place where the jar was manufactured and possibly the origin of its contents as well, which was probably oil or wine.