All eight dolmens were constructed on the slope, several located on spurs or stepped terraces; all had a panoramic view of the Sea of Galilee. Some of the dolmens were grouped together, whilst others stood alone. Two dolmens stood fairly close together on the upper basalt step (Nos. 51, 53; 47–50 m asl). Three stood in a row on an intermediate basalt step (Nos. 47–49; 38–40 m asl). One dolmen stood at a similar elevation further to the northeast (No. 59), 19 m west of a dolmen (No. 60) that was not excavated, as it lay beyond the area designated for development. Two dolmens stood lower down the slope, to the southeast, separated from each other (No. 35—13.0 m asl; No. 43—24 m asl).
All the dolmens were entirely dry-constructed of carefully selected, huge unworked orthostat-like boulders, which along with large and medium-sized basalt fieldstones formed the walls of the burial chambers; huge, somewhat flatter basalt slabs were used as roofs, which were overlain with huge capstones. These and other shared characteristics of the dolmens suggest that they all belong to the same basic type, which is indeed Epstein’s Type 2, subtypes a–c (Epstein 1985:26, Fig. 1), although a few additional features were observed. Dolmen 59 was selected as representative of this group of dolmens, as it exhibited all the main components; hence, it is described first and in full detail, including its dimensions. Subsequently, the other dolmens are described more briefly, noting only main features and variations.
Dolmen 59 (Figs. 4, 5) comprised a slightly sloping, long narrow corridor-like passage (L701; 0.5 × 3.8 m, height 0.85 m; Fig. 6) that extended along an east–west axis, from the outer eastern perimeter of the dolmen’s stone mantle (tumulus) into a narrow, east–west rectangular burial chamber (L703; 0.7 × 2.5 m, height 0.95 m). The passage walls, partially collapsed, were irregularly built of fieldstones. The chamber was walled on three sides by huge, well-fitted orthostat-like boulders that were slightly set into the thin, reddish basaltic earth layer (0.1 m thick) that overlay the basalt bedrock surface (Fig. 7). The chamber’s southern wall was constructed of a single, huge boulder (length 2 m, width 0.7 m, height 0.95 m); its northern wall—of two huge boulders (length 0.9 and 1.5 m, width 0.7 m, height 0.95 m); and its back, western wall—of a single large boulder (length 1 m, height 0.9 m) overlain by a smaller stone. Small basalt stones were inserted between the boulders, to fill gaps, and beneath the boulders, to level them out. The entrance from the passage into the chamber was roofed with a transverse-lain stone slab (1.0 × 1.6 m, height 0.5 m) that served as a lintel. On top of it and slightly set back from it was a huge saddle-shaped, stepped capstone (1.7 × 1.9 m, height 1.1 m) that roofed the rectangular chamber in its entirety. The chamber floor was carefully paved with stone slabs (average dimensions: 0.4 × 0.7 m) that were set over the thin layer of earth. The entire dolmen from the paved floor inside the chamber to the top of the capstone stood to a height of about 2.2 m above its immediate surroundings.
The dolmen mantle or tumulus formed a roughly circular to oval-shaped stone heap (L700; 9.5 × 11.5 m) comprising numerous large and medium-sized fieldstones, several of which could be discerned along the outer perimeter. It was possible to discern the contour of a curving structural wall within the southern and western parts of the tumulus and traces of several other wall contours; these indicate that the tumulus was formed by carefully building low stone walls, two or three courses high, for support and stabilization, and subsequently piling stones up to the base of the dolmen’s capstone. In addition, a small, shallow crater-like depression (diam. 2 m, depth 0.6 m) was purposely hollowed out—i.e., cleared of stones—within the tumulus and leaning against the outer face of the structural wall; no such feature has been previously documented in the research literature about dolmens. No finds were found among the stones surrounding the depression, and its function is not known.
The layer of accumulated earth (0.3 m thick) inside the burial chamber contained extremely meagre finds: a few very small and extremely worn Roman- and Byzantine-period potsherds that may reflect a secondary use of the chamber. No earlier artifacts that can date the construction and the original period of use of the dolmens as burials were found. A small probe (L704) excavated below the stone-slab paving of the burial chamber, through the earth underneath it and down to the bedrock yielded no sherds.
Dolmen 35 was only partially extant, as its stone mantle (L200) was previously damaged, and the eastern half of the dolmen had been cut off with the bedrock partially carved out by a mechanical digger during preparation work for laying a road (Figs. 8, 9). Nevertheless, the burial chamber (L202) remained almost intact. The extant section, once cleaned, provided a vertical section clearly illustrating the construction method of the dolmen. It was erected directly on the bedrock, which was composed of fused basalt boulders. The rectangular, east–west oriented, burial chamber (L202; Fig. 10) was walled on three sides with large orthostat-like boulders and roofed with a flat stone overlying another larger stone; the original capstone, however, was missing. The chamber floor was carefully paved with stone slabs lain over a thin layer of earth (L203) that covered the bedrock. A socketed spearhead characteristic of the Middle Bronze Age I was found between the paving stones in the southwestern corner of the chamber (see below, Fig. 29).
Dolmen 43 was a large circular dolmen (diam. 13 m; Figs. 11, 12) with a full stone mantle. The rectangular burial chamber (L802), oriented northeast–southwest, had three walls of large orthostat-like boulders and was entered via a long, narrow passage that sloped slightly toward the chamber (L801; Fig. 13). A huge, saddle-shaped capstone overlay a flattish, transverse stone that served as a lintel at the entrance into the chamber. Both the chamber and the passage had some extant stone-slab paving that overlay the bedrock (L803). As in Dolmen 59, a structural wall within the stone mantle (L800) formed an almost complete circle, and a small circular crater was hollowed out between the stones lying against this wall.
Dolmen 47 was a large, circular to oval dolmen (diam. 13 m; Figs. 14, 15). The burial chamber was rectangular, oriented east–west and accessed by a paved passage (L404), whose walls have mostly collapsed. The chamber (L405) had three walls made of huge orthostat-like boulders, with a huge capstone lain over flat stone slabs. The chamber floor was carefully paved with stones (Fig. 16) laid over a layer of earth (L410; thickness c. 0.3 m) that covered the bedrock. An opening in the upper northern wall of the chamber was probably created by tomb robbers who moved aside a large wall slab. The stone mantle (L400) was visibly a complex structure incorporating several well-built circular structural walls, comprising either several courses of medium-sized stones or a single course of larger stones. At the end of the excavation, the stones of the southern part of this dolmen’s mantle were mechanically removed, in order to view the construction of the structural walls inside the mantle and the external face of the southern orthostat wall of the burial chamber (Fig. 17).
Dolmen 48 was damaged by a mechanical digger, leaving its stone mantle only partially extant (L401; Fig. 18, 19). A particularly narrow and short passage (L406; min. width 0.4 m), partially blocked by an upstanding stone, led at a slight angle into a small, approximately east–west oriented, rectangular chamber (L407; 0.5 × 1.7 m; Fig. 20). The chamber was walled with huge orthostat-like boulders, and was roofed with a huge triangular capstone that rested on a very thin transverse stone. The chamber floor, only accessible after the capstone was removed, was partially paved; the paving lay on a thin layer of earth (L411) that covered the bedrock. A short segment of a circular structural wall was extant within the stone mantle (L401); beside it was a small, irregularly-shaped crater that was hollowed out between the stones.
Dolmen 49 (Figs. 21, 22) was built on a fairly steep bedrock step. The stone mantle (L402), the passage (L408) and the northeast–southwest oriented burial chamber (L409) were damaged, either by robbing activities or in a previous excavation.Here too the extant features were similar to those of the other dolmens, with a paved chamber floor set on a thin earth layer (L412) over the bedrock. Neither a circular structural wall nor a crater-like feature were visible in the partially extant stone mantle.
Dolmen 51 had a rather irregular circular form (Fig. 23; diam. 14 m). The stone mantle (L500) exhibited gaps between the stones that extended down to the bedrock, and only one of the two extant roof slab remained in place (Fig. 24). The capstone was missing; a huge rectangular stone (length 2 m) found lying to the southwest of the dolmen may have been the original capstone that had been ‘rolled’ off the dolmen’s side, possibly in the course of Turville-Petre’s excavations. The burial chamber, oriented east–west, was long and narrow (L512; 0.7 × 3.0 m, height 0.7 m) and walled on three sides with huge orthostat-like boulders. The access passage formed a direct continuation of the chamber, and both were paved with stone slabs, of which only a few large ones were extant in the chamber (Fig. 25). A probe excavated below the paving exposed a layer of earth (L515; thickness 0.3 m) on the bedrock. A circular structural wall surrounded the middle part of the tumulus, and a crater-like depression (diam. 1.2 m, depth 0.5 m) was hollowed out between the stones in its northeastern part.
Dolmen 53 was a large dolmen (13 × 16 m; Figs. 26, 27), with an irregular circular stone mantle (L501) and an extension or additional tumulus on its southeastern side. The burial chamber (L514), oriented east–west, was almost rectangular, narrowing slightly toward the entrance. The access passage (L513) was narrow, and its walls had mostly collapsed (Fig. 28). The chamber was paved with stone slabs over a layer of earth (L516; thickness 0.3 m) that covered the bedrock. The stone mantle was supported by a circular structural wall that stood to a height of four to six courses. An additional curving wall, comprising large basalt fieldstones, was incorporated in the southeastern extension and continued downhill beyond the dolmen. Two crater-like depressions were hollowed out between the stones of the stone mantle, one on the western side and the other within the southeastern extension.
Finds. Only a few fragmentary bones of caprines and hyraxes—the latter still abundant in the area—were retrieved from the dolmens; no remains of human bones were found.
Very small to tiny worn potsherds from the Roman and Byzantine periods were found sporadically in all the dolmen chambers and passages, and two Ottoman-period tobacco pipe fragments were found in Dolmen 53, along with a single Roman coin (IAA 154918) dated to the fourth century CE. These limited finds may reflect later agricultural activity in the broad vicinity, and the later use of the chambers, possibly as shelters for shepherds or for the homeless. It is also possible that the later artefacts arrived when the dolmens were rifled.
A couple of very small and worn non-diagnostic body sherds from Dolmen 49 and very few fragments of non-diagnostic flint blades from Dolmens 47 and 49 possibly date from the Intermediate Bronze or Middle Bronze Age.
A single artifact may contribute to the dating of the dolmens: the socketed spearhead found between the paving stones of the burial chamber in Dolmen 35 (Fig. 29). It is made of copper alloy; the blade is concave, with sloping shoulders, straight-tapering sides, a V-shaped midrib and a fairly short tube-shaped, folded-over socket (total length 87 mm, blade length 63 mm, max. blade width 22 mm). Such spearheads developed in the late-third millennium and are common in MB I–II contexts in Syria and the southern Levant (Socketed Spearhead Types 7 and 8; Philip 1989:92–95, 99–100, Figs. 23, 24). Similar socketed spearheads of various dimensions were found along with a variety of other copper weapons and MB I (i.e., MB IIA) pottery in several multiple-burial caves from the Middle Bronze Age in the Galilee, such as at Kefar Szold and at Ginosar Tomb 1 (Epstein 1974: Figs. 4:10, 7:14); at Gesher (Garfinkel and Bonfil 1990: Fig. 1:5); at Zefat (Damati and Stepansky 1996:10*–13*, Figs. 12, 13). The presence at these sites of the socketed spearhead in association with many ceramic forms reflecting Syrian affinities points to the Syrian origin of this spearhead.
Similar socketed spearheads were also found in two dolmen burials in the Golan: in Dolmen No. 10 at Abu Fula, a socketed spearhead in disintegrating condition was uncovered between the paving and the wall-stone near the closed end of the dolmen, together with poorly preserved IBA (designated there MB I) sherds (Epstein 1985:31, Fig. 2:9); and in Dolmen No. 13 at Deir Sras, a socketed spearhead was found with IBA (designated there MB I) and MB I (designated there MB II) pottery on the paved floor (Epstein 1985:32–33, Fig. 3:18).
Stone Heaps
Stone Heap 42 and 50 (Fig. 2), both large and circular and located near dolmens (Stone Heap 42—50 m southwest of Dolmen 47; Stone Heap 50—30 m east of Dolmen 49), were suspected of concealing dolmen burials. Both heaps were first dismantled manually; the remainder was removed with a mechanical digger. Neither of the stone heaps proved to be of any archaeological significance. Their stones may have been piled up for use in the stone mantles of nearby dolmens, or as a result of clearance in later periods.
Rows of Basalt Boulders
Irregular rows of large basalt boulders that were observed during the initial survey in a large area (52; Fig. 2) between the dolmens were suspected of being the walls of an ancient settlement, or possibly agricultural terrace walls. Several excavation squares were opened beside these stone rows. They yielded only sterile, blackish basaltic earth without any evidence of occupation, except for very small and extremely worn potsherds from the Roman and Byzantine periods, a few flint items and a stray coin. Moreover, the areas delimited by the stone rows were hardly large enough to be animal pens. It is therefore deduced that these rows of basalt outcrops were a geological phenomenon resulting from the volcanic activity that shaped the plateau.
The eight excavated dolmens, located in close proximity to each other on the southern Korazim plateau southeastern slope, overlooking the Lake of Galilee, exhibit many common characteristics. These include an east–west (or northeast– southwest) oriented burial chamber accessed via a narrow passage and walled with large basalt orthostat-like boulders, paved with carefully laid stone slabs and roofed with large flattish slabs overlain by a huge capstone. As the weight of similar capstones has been calculated as 15 tons or more (Stepansky 2005:43), it is suggested that the huge orthostats and capstones were dragged down the northern, higher side of the slope, and that the capstones were levered into place only after the stone mantles surrounding the burial chambers were piled up against and over the circular structural walls. Based on these predominant features, the dolmens all classify as Epstein’s Type 2 (Epstein 1985: Fig. 1). A feature never noted previously in the research on dolmens but identified in most of the dolmens examined in our excavation is the small, rounded crater-like depression hollowed out among the stones of the mantle; its function remains unexplained.   
The copper-alloy socketed spearhead found in situ between the paving stones in Dolmen 35 is characteristic of the MB I (nineteenth–eighteenth centuries BCE), suggesting that the dolmens were used during this period. Similar socketed spearheads, found along with MB I pottery in dolmens in the Golan indicate the widespread use of the dolmens for burial in this period. However, it cannot be deduced unequivocally that the dolmens were first constructed at this time, as they may have been constructed earlier and reused in the MB I.
The dolmens were used again much later, as evidenced by the small, worn Roman- and Byzantine-period pottery sherds and the Ottoman-period pipes, probably as a shelter for shepherds or possibly for the homeless, as suggested by the passage in the New Testament (Mark 5:2–5), which mentions the poor who lived in tombs around the Sea of Galilee.