In April 2014, a trial excavation was conducted in a cave c. 200 m southeast of the Har’el Interchange (Permit No. A-7106; map ref. 214235–51/633810–25), prior to widening Highway 1. The excavation, undertaken on behalf of the Israel Antiquities Authority and financed by the Department of Public Works, was directed by D. Ein Mor, with the assistance of N. Nahama (administration), A. Hajian (surveying and drafting), A. Marco (antiquities inspection) and A. Frumkin (geology).
A karstic cave was discovered in a bedrock terrace on the southern bank of Nahal Et-Tantūr (Fig. 1); quarrying marks inside the cave indicate man-made extension (0.5–2.0 × 12.5 m, max. height 3.5 m, Fig. 2). Apparently, this cave was one of several 'refuge caves' used at the site during the Early Roman period. Agricultural terraces and remains of traditional farming were observed in the vicinity of the cave.
Mechanical equipment was used to rjaremove modern debris from the area in front of the cave prior to the excavation. Chisel marks, some of them covered with lichen, were discerned on the bedrock terrace east of the cave. The opening of the cave (width 1.75 m, height c. 1 m; Fig. 3) faced north, and a natural rock step (height 0.8 m) led up to it. Irregular rock-cuttings, probably made with both sides of a pick or axe, and perhaps even a chisel, were discovered along the eastern wall of the cave. Some of the marks were narrow (length 8 cm, width 0.5 cm; Fig. 4) and some were wide, including square (2 × 2 cm) and triangular ones. A natural depression in the bedrock (L100; 1.20 × 1.95 m, height 1.9 m; Fig. 5) near the mouth of the cave was filled with terra rossa soil from a fissure in the ceiling. The fill was mixed with stone fragments from the ceiling of the cave, plant roots and construction debris. Fragments of a modern glass bottle were found 0.15–0.20 m above the bedrock. A hearth (diam. 0.2 m; Fig. 6) was discovered in the depression, directly on the bedrock, and around it were several pieces of charcoal and a few non-diagnostic pottery sherds. Two natural niches were exposed in the walls of the cave on either side of the depression. The eastern one (L101; Fig.7) was extended by quarrying; quarrying debris was discovered inside that niche, directly on the bedrock. South of the depression (L100) the bottom of the cave rose gently toward the south (L102; 0.5–1.9 × 4.0 m; Fig. 8). Quarrying debris (average thickness 0.3 m; c. 0.5 cu m) was discovered in the middle of this part of the cave, some directly on the bedrock and some on top of accumulated weathered rock and clay soil (L103). A non-diagnostic pottery body fragment was found in this accumulation. Judging by the fabric and firing, it was presumably of a jar from the Roman period. A shaft (L104; 0.6 × 1.7 m, depth 1.2 m; Fig. 9) with an elliptical opening was discovered farther south. The bedrock was breccia, and it was therefore hard to determine if the shaft was natural or man-made. Even if the shaft was hewn, it would be difficult to see the quarrying marks because quarrying would cause lumps to break-off from the rock. The shaft extended down below the natural level of the cave, thus it seems that it was hewn. The eroded rock from the cave’s ceiling, some clay soil and roots, and a few fragments of a modern tin container had accumulated in the shaft. South of Shaft 104, the level of the cave rose southward, and the cave ends in a natural niche (L105; 0.5–1.0 × 3.5 m, height 0.9–1.5 m). Fine-grained soil that was formed from the weathering of the rock accumulated in the niche; the soil was not excavated down to bedrock (excavation depth 0.5 m). The color of the soil was cream-white at the top, and light red lower down, probably due to iron oxide in the bedrock. On a bedrock step above the opening of the cave was a an agricultural terrace wall (W10; length 2.2 m, width 0.5 m), constructed of two rows of stones, in an east–west direction. Its northern face was constructed directly on the bedrock and was preserved to a height of four courses (height 1.4 m); of the southern face, only the tops of the stones were exposed. The wall delimited an agricultural terrace that extended to its south (width 2 m).
Some 5 m above the cave were two caves that were excavated during a second season of excavations at the site (September 2014). A few meters to the west of the cave, a burial cave(?) was exposed but not investigated due to modern construction fills that blocked its entrance.
The excavated cave was open long before the excavation, and probably served as a temporary shelter in the modern era. The hearth on the bedrock near the opening may be related to the quarrying activity. The findings from the second season of excavations at the site support the assumption that during the Early Roman period the cave served as a 'refuse cave'. Despite the absence of material from the cave, its characteristics and its close proximity to the other cavea allow us to identify it as part of a complex. The quarrying debris was not removed from the cave, possibly so as not to disclose its location.
The Geology of the Cave
A natural cave was excavated, which was enlarged by quarrying brecciated dolomite-limestone that apparently belongs to the Kesalon formation. Breccia bedrock is composed of lumps of angular rocks ranging from several centimeters to one meter in length. The same rock is also apparent on the bedrock terrace outside the cave; however, not far away the bedrock appears to be marine dolomite. The breccia is probably the remains of an ancient karstic cavity that collapsed (paleokarst). The cave was probably predisposed to develop along the breccia. The walls of the cave are brecciated and crumbled, and there is no distinct morphology to the mass. The morphology is primarily due to weathering, particularly on the western wall and the ceiling, and to quarrying, mainly of the eastern wall but also of the western wall near the opening. The cave extends in an almost straight line, with small niche-like crevices in the walls and ceiling. Calcite deposition is apparent in ancient cracks. The sediment is usually harder than the bedrock around it, and forms a boxwork-like structure. During the weathering process the bedrock disintegrated into stones and dust, and erosion products are visible on the bottom of the cave and along its edges. Rhizofossils are common, as well as roots that penetrated into the cave. Along the bottom of the cave there are sections of rock indicative of an ancient ground level or a transition between layers. Pockets of terra rossa soil are common near the opening but appear also inside the cave. Localized color-spots of reddish iron oxides appear here and there.