The excavation area lies within the Arab village of Qula, which until 1948 was located to the east of Giv‘at Koah Junction in the drainage basin of Nahal Mazor, a secondary tributary filled with alluvium that passes through the site. The site is now forested. Surveys conducted at the site in the nineteenth century CE documented a water-storage pool in the center of the village and buildings erected on ancient remains that probably dated from the Crusader period (Guérin 1875:390; Conder and Kitchener 1882:358). Two surveys conducted in the region during the twentieth century were the first to document the agricultural area beyond the boundaries of the Late Ottoman village, where there are cisterns, shafts, tombs, winepresses, and agricultural installations (Spivak and Kanias 2010; Kochavi and Beit-Arieh 2013). Previous excavations at the site focused on the village’s western fringes and uncovered settlement remains dating from the Hellenistic, Roman, Byzantine, and Ottoman periods (Avissar and Shabo 2000; Sklar 2003; Eshed 2011 [Fig. 1: A-5913]; Permit No. A-7512). Excavations conducted in the past to the north of the current excavation area (Milevski 2001a; 2001b) discovered burial caves from the Chalcolithic period and the Early and Intermediate Bronze Ages, as well as Roman and Byzantine tombs. A lime kiln from the Byzantine–Early Islamic periods was discovered near one of the burial complexes (Milevski and Shabo 1999).
Four excavation areas were opened (A–D; Fig. 2); Areas A–C are characterized by brittle bedrock, and Area D by tel soil. An alluvium accumulation containing small fieldstones covers much of the area, although the bedrock is exposed in places. In some cases, the alluvium hindered the detection of features at the points surveyed in these areas. A survey prior to the excavation in Areas A–C documented 76 findspots (1–76; Figs. 3–5), 25 of which were excavated (2, 6, 10, 13, 15, 25, 29, 30, 33, 35–40, 42, 44, 45, 48, 52–55, 57, 59); they include winepresses and a bodeda, cisterns, caves, roads, quarries, rock-cuttings, terraces, cupmarks, and tombs. At most of the findspots, no sealed loci were excavated; since the potsherds had been swept to the site, they were worn and non-diagnostic and therefore not drawn. Those ceramic finds from Areas A–C that were drawn come either from clearly defined loci or contribute to a better understanding of the site and its history. Eleven squares excavated in Area D contained two building complexes, southern and northern, dating from the Late Ottoman period and the British Mandate era.
Winepress 25 (Fig. 6) was hewn in Area A and followed the south–north slope of the natural rock surface. The winepress contained a treading floor (L105) and a collecting vat (L109) with a settling pit; a short channel (depth 1–5 cm) connected the treading floor to the collecting vat. Because the winepress was hewn on a slope, the treading floor’s eastern wall was higher than the other walls to prevent the spillage of liquid from the installation. Traces of a layer of grayish plaster mixed with fine gravel were preserved in the southern part of the treading floor. Remains of identical plaster were also uncovered on the walls of the collecting vat. A natural cavity (L119; 0.1 × 0.6 m, depth 0.2 m) beneath the collecting vat’s eastern wall was blocked with small fieldstones and soil. The vat was probably plastered after the cavity was blocked, although no plaster was preserved around it. The soil fill above the treading floor yielded worn pottery from various periods. No finds were recovered from the collecting vat. Cavity 119 yielded an Iron Age trefoil jug (Fig. 7:1), which may attest to the period when the winepress was constructed.
Winepress 33 (Fig. 8) was hewn near the summit of a hill in Area A. It contained a treading floor (L102), a settling pit (L121), and a collecting vat (L108) that were well-preserved; a channel (L126) connected the treading floor to the settling pit, and another short channel connected the settling pit to the collecting vat. In all parts of the winepress, remains of constructions made of small fieldstones were discovered held together with bonding material composed of gravel, small pebbles, shells and crushed pottery and used to seal natural hollows in the rock. On top of the fieldstone surface used for sealing, the treading floor was coated with a layer of smoothed mortar. The sides of the treading floor were coated with a layer of mortar (thickness c. 5–12 cm) identical to that discovered on the floor; on top of this was a thin layer (thickness 1–2 cm) of grayish plaster containing fine gravel and crushed shells. In places where fieldstones had been used for sealing, particularly thick layers of mortar were applied. A large fieldstone (0.5 × 0.6 × 0.6 m) with a shallow depression in the center was found in the western part of the treading floor; it was difficult to determine whether it was hewn or natural. It is not impossible that the stone was used in the must-extraction process, although there is no clear evidence of this. Channel 126 followed a natural fissure in the rock, plastered on its northern side and narrowed on its southern side by a row of small fieldstones. The row of stones was coated with a thin plaster layer (thickness 0.2–0.3 m). A small fieldstone found at the eastern end of the channel was set in mortar above the channel sides; it may be a preserved part of the channel’s roofing, but no other roofing stones or traces of mortar were found elsewhere above the channel sides.
The sides of the settling pit were coated with plaster identical to the treading floor. The upper part of the pit’s western side was built of medium-sized fieldstones interspersed with the same mortar; it was only partially preserved. The southern side of the settling pit was low, and a channel led from it to the collecting vat. The sides of the pit may have been built up to a uniform height, but the structure was not preserved.
The walls of the collecting vat were hewn on a slightly inward slant, and they bore evidence of two plaster layers identical to the two plaster layers uncovered on the sides of the treading floor. A breach in the vat’s southern wall led to a natural cavity in the rock (0.7 × 0.9 m, depth 0.7 m); the cavity was sealed by fieldstones visible through the breach. Different colored lichens on its upper layer made it difficult to determine the color of the plaster surface. A layer of clean lime was revealed at the bottom of the vat; the plaster on the southern wall continued on top of it, suggesting that a plaster layer that was not preserved was probably applied on top of the lime layer. Compacted calcareous material mixed with a little fine-grained soil was revealed above the bottom of the vat in its northeastern corner and at the same height across the entire width of the eastern wall. Since the calcareous material was discovered on top of the vat’s plaster layer and above the lime bedding, the material postdates the vat’s plastering. The material was probably placed in the collecting vat after it fell into disuse. The calcareous material was also found near the breach in the vat’s southern wall, raising the possibility that when the wall was breached, the calcareous matter was discarded on the vat floor. The rock between the collecting vat’s eastern wall and the treading floor was hewn in a stepped fashion, forming a shelf (width 0.4 m, height 1.36 m) above the floor of the collecting vat. A natural depression in the rock at the southern end of the shelf was filled with alluvial soil.
No remains were discovered that a screw was installed in the winepress, the use of which began in our region in the Roman period (Ayalon, Frankel and Kloner 2012:30); therefore, it is likely that it predates this period. A soil accumulation near the bottom of the collecting vat yielded an Iron Age trefoil jug (Fig. 7:2).
A surface that may have been deliberately leveled was revealed to the south of the winepress, on a higher level than the treading floor. The surface is delimited on the west by a natural fissure in the rock (depth 0.3 m). The surface may have been used as a cell for placing the grapes before the treading process. The rock fissure may have been used to drain any must from the cell.
Bodeda 57. A bodeda in Area C consisted of a treading floor (0.94 × 1.09 m) and an oval collecting vat (0.72 × 0.79 m) connected by a shallow rock-cut channel.
Three cisterns were revealed in Areas A, B, and C (10, 44, 53, respectively); each had a mouth built above an oval vertical rock-hewn shaft that opened into a hewn and plastered underground cavity.
Cistern 10. The cistern mouth (outer diam. 2.25 m, inner diam. 0.8 m) was built of a course of small fieldstones overlain by a course of large and medium-sized fieldstones. A rock-hewn channel that drained water to the cistern through a gap near the opening (L112) was uncovered to the northeast of the cistern. The fill in this gap and the area around the cistern (L111) yielded abundant pottery consisting primarily of jars and jugs, some of which date from the Roman period (Fig. 7:3–5) and most from the Late Ottoman period (Fig. 9:4, 7, 9, 10); a tobacco pipe was also found (Fig. 9:11). It seems that the cistern was used in the Ottoman period, although it is likely that it was dug earlier.
Approximately 1.5 m south of Cistern 10 were two cupmarks (diam. 0.2–0.3 m, depth 0.1–0.2 m) that may be associated with the cistern’s use. Two rock depressions whose function is unclear were also found.
Cistern 44. The square mouth (L211; 0.7 × 0.7 m) was built of a course of small and medium-sized fieldstones overlain by a course of ashlars bonded with mortar. Grooves made in the upper part of the opening contained iron bars set in cement or concrete. A hook for a bucket and rope for drawing water was attached to one of the bars. To the west of the opening, the cistern’s underground cavity was connected to an oval rock-cut cavity that may be a tomb. A few worn potsherds (not drawn) were collected from around the cistern, some dating from the Mamluk period and most from the Late Ottoman period. Based on the iron bars and the late ceramic finds, the opening was probably constructed in the Late Ottoman period, while the cistern is more ancient.
Cistern 53. The square mouth (0.4 × 0.4 m; Fig. 10) was built like that of Cistern 44. The cistern was probably originally a burial cave, with an oval rock-hewn entrance yard (54; L308) discovered to the southwest of the opening. Three roughly-cut steps in this entrance yard led down to the cavity opening. At some stage, this opening was blocked with small and medium-sized fieldstones, the blockage was plastered on the inside, and the cavity was converted into a cistern. The entrance yard was also filled with small and medium-sized fieldstones to support the stones blocking the opening. At some stage, the opening was breached, possibly by robbers, and resealed with a large boulder. A few worn potsherds were collected around Cistern 53 (L312), some dating from the Mamluk period and most from the Late Ottoman period (Fig. 9:8). Based on the iron bars and the late ceramic finds, the opening was probably constructed in the Late Ottoman period while the cistern itself is more ancient.
Cave 36 (Fig. 11). Area A contained a large entrance yard (3.0 × 3.4 m), covered in its westen part, with a narrow opening leading to the cave on its western side; it is unclear whether the entrance yard was hewn or natural. A probe trench (1.5 × 6.0 m) excavated in the entrance yard and cave area revealed two layers of alluvial soil mixed with small fieldstones (lower layer—L127, L130, thickness c. 0.4 m; upper layer—L104, L120, thickness c. 0.5 m); the upper layer was darker and more compacted. Approximately 1 m east of the cave entrance, a surface of light gray soil (thickness 3 cm) was detected in the northern section that did not extend the entire length of the probe and probably attests to a localized fire. Large and medium-sized fieldstones found near the cave entrance are likely related to some construction on both sides of the entrance, possibly intended to block it off; the inside of the cave was not excavated.
Both layers of alluvial soil contained remains of unarticulated human bones belonging to adult individuals (see Appendix: Table 2). The ceramic finds from the two layers of alluvium are identical and comprise mostly hand-made holemouth cooking pots (Fig. 9:2, 3) dating from the Late Ottoman period. A soil layer near the surface (L104) yielded a Byzantine jar (Fig. 7:8) originating from the alluvium.
An opening in the rock (Cave 37; L103; Fig. 11) discovered near Cave 36 had a square depression in its lower part (0.6 × 0.6 m) that was not fully excavated; therefore, it is unclear whether it was artificial. The opening may be another entrance to Cave 36, but the excavation was not completed.
Cave 42. A narrow complex in Area B contained a rock-cut cave (Fig. 12) accessed via a sloping south–north passage (L212) and an entrance yard. Passage 212 was roughly hewn. The northern end of the passage was filled with soil (L218), above which small fieldstones were placed on a gradient (L214). The fieldstones were overlain by dark brown soil containing smaller fieldstones (L213) and a few worn potsherds (not drawn). No quarrying marks were found in the entrance area, but they may have become eroded in the soft limestone. A small probe (L219) excavated in the entrance area revealed an accumulation of clayey soil and fieldstones. The eastern wall of the entrance area contained the remains of a shaft cut lengthwise (diam. 0.8 m, depth 1.8 m; Fig. 12: Section 1–1). In the southern part of the entrance area, c. 3 m below the surface, the cave entrance was hewn (width 1 m, exposed height 0.65 m) from where the cave opened southeastward. The inside of the cave was not excavated, but from the outside, it was evident that the cavity was oval with two hewn vaults, one on the southern side and the other on the eastern side. The cave’s excavation was not completed due to technical constraints. There are two possibilities regarding the association between the entrance area and the shaft on its eastern side. (1) The entrance area was originally underground, and the cave was entered via the shaft; although there are no traces of collapsed rock in the entrance area, as it was undergoing mechanical excavation, this may have removed any rubble. (2) The entrance area was hewn after the shaft and cut it into two. A small quarry (below) was discovered on the surface north of the entrance area.
Cave 52. Area C contained a cave entrance (width 0.8 m) leading to a narrow, low cavity (0.5 × 0.7 m, length 1.45 m) that widened as it deepened. The cavity contained an accumulation of dark brown alluvium with small fieldstones; the excavation was halted for technical reasons.
Road 13 (excavated length 5.5 m). A segment of a north–south road was excavated in Area A, on the eastern slope of a hill. The western boundary wall of the road was built of a row of medium-sized fieldstones with a fill of small fieldstones to its east. Part of the wall and the fill were placed on the bedrock. The road was probably disturbed by mechanical equipment. Although the road’s parallel boundary wall remains were not uncovered, the road’s construction resembles Road 59 (below).
Road 59 (excavated length c. 8 m; Fig. 13). In the northern part of Area A, a segment of an east–west road was unearthed in which two phases of construction were identified. In the earlier phase, the road had two parallel boundary walls (W136, W138) built of fieldstones; Wall 138 (width 1 m) was the more massive of the two. Between the walls, the road’s paving (L134) was composed of small and medium-sized fieldstones (average size 0.3 × 0.5 m) mixed with local soil (L134). In the later phase, the road was narrowed (width 1.6 m) by constructing another wall (W135; Fig. 14) built of a single row of medium-sized fieldstones and preserved to the height of a single course. A soil fill (L133; thickness 0.3 m) placed between Walls 135 and 136 and above fill L134 was overlain by paving made of small fieldstones mixed with soil (L132). A layer of soil (L131) found between W135 and W138 reached the level of the wall tops; it is unclear if this was a deliberate fill intended to support W135 or naturally accumulated soil. The pottery finds on the surface layer (L100) and soil layer L131 were mostly Late Ottoman (not drawn), dating the road’s latest phase. A shelf rim of a CRS bowl (Fig. 7:6) and a roof tile (Fig. 7:9) dating from the Byzantine period were also recovered. Due to technical limitations, a probe was not excavated in the road’s earlier fill and was therefore not dated.
Another wall (width 0.9 m, length 1.9 m) built of a row of large and medium-sized fieldstones on a roughly east–west alignment was partially excavated c. 15 m west of the road segment; an especially large boulder (0.8 × 1.0 m) was found at its eastern end. The wall may belong to a continuation of the road.
Quarry 35 (L124; 1.0 × 2.5 m, depth 0.4 m). In the quarry there are three vertical rock-cut walls and severance channels. Few stones were hewn from the quarry.
Quarry 42. The quarry (2.3 × 2.7 m; Fig. 12) extends as far as the edge of Cave 42 and therefore dates from the time of the cave or later. A single quarrying level for building stones was uncovered in the quarry. Severance channels and marks of extracted stones were identified. The quarry was not excavated for safety reasons.
Quarry 45 (L201, L202; Fig. 15) extends across a natural leveled rock outcrop, and three quarrying levels were discovered in it, sloping from west to east. It contained square and rectangular quarrying marks (length 0.7–1.6 m). In its eastern part, square quarrying (L202; 1.0 × 1.2 m) revealed at a slight distance from the others probably dates from another phase of the quarry’s operation. Worn potsherds from the Roman and Byzantine periods (not drawn) and Late Ottoman pottery, such as a Black Gaza Ware jar (Fig. 9:5), were found when the alluvium covering of the quarry was removed.
Quarry 48 (L210; 1.7 × 2.0 m). A small quarry contained severance channels and quarrying marks of a few stones. The quarry contained a single undetached stone (0.27 × 0.52 × 1.10 m).
Area A contained three rectangular rock-cuttings (2—1.1 × 2.0 m, depth 0.4–0.5 m; 29—0.5 × 2.5 m; 30—0.7 × 1.3 m, depth 0.10–0.45 m) whose function is unclear. Based on the remains of quarries discovered nearby, it is possible that these were building-stone quarries; although no clear severance channels were discerned in them, this may be due to weathering of the rock. Another option is that these were unfinished tombs resembling those discovered in the area (38, 39 below).
Terrace Wall 40. Area B contained a terrace wall (length 21.6 m, width 1.3 m) built on a north–south alignment from two rows of large–medium-sized fieldstones and preserved to the height of a single course.
Terrace Wall 55 (L307). Area C contained a terrace wall (length 11 m, width 1.2 m) built on a rocky outcrop and a roughly north–south alignment from two rows of large–medium-sized fieldstones and preserved to the height of a single course. Fragments of a Byzantine bowl (Fig. 7:7) and a Late Ottoman Black Gaza Ware bowl (Fig. 9:1) were found next to the wall.
Pit 6 (0.4–0.7 × 1.4 m, excavated depth 0.7 m). Area A contained an oval pit hewn in the bedrock with several stones inside; it is unclear whether the stones were placed deliberately in the pit.
Pit 15 (depth 0.45 m, diam. c. 1 m). Area A contained a circular pit cut into a rock step.
Two rectangular rock-hewn tombs c. 20 m apart (38, 39; Fig. 16) were discovered in Area A; they were documented without being excavated. They may have been robbed.
Building Remains (Figs. 17, 18)
The Southern Complex. A rectangular structure built on a northeast–southwest alignment was excavated in the south of Area D (Fig. 19). The building was delimited by four walls (W431, W432, W447, W449) built of two rows of large and medium-sized fieldstones and with a core of smaller fieldstones; the walls were preserved to a height of one to two courses. The walls’ outer faces were dressed smooth. Only W431 was excavated as far as its base, which was founded on bedrock (L434). The walls were coated with two layers of plaster; the upper layer was painted light blue and was preserved mainly in the building’s southeastern corner (Fig. 20) and the eastern part of the top of W431. The building’s floor (L440) was made of grayish plaster and amalgamated with a roller (rolka) with the plaster on the walls. In the center of the building, the plaster floor had been repaired. The floor sloped to the southwest, but no drainage opening for water was found there; there may have been one in the floor’s southwestern corner that was destroyed. When the building was abandoned, it was filled with a layer of brownish-gray soil overlain by building debris (L421, L436) that contained fragments of concrete and plaster and large and medium-sized fieldstones from another building that stood nearby. Outside the building and throughout Area D, rubble was found without fragments of concrete, showing that the building was filled with debris before the whole site was destroyed.
The ruined debris inside the building yielded Late Ottoman and British Mandate-era finds, including a bowl (Fig. 21:7), a jug (Fig. 22:6), and a tobacco pipe (Fig. 22:12). Modern plastic beads were also found, one bearing the inscription ‘Cargo’. A rifle cartridge manufactured in 1917 was found beneath L440 (L453; see Appendix: Table 1: No. 13:I). A key and a lid of a metal tobacco box from the ‘State Express 555’ British tobacco company were found on the surface outside the building (L435). The finds, particularly the cartridge, date the building’s construction to the post-First World War period.
Two options were raised regarding the building’s function: (1) The excavation area covers farmland where tools have been found, and the building may have been a warehouse. The plaster at the top of W431 may be the remains of a wall niche, a characteristic Late Ottoman architectural element. (2) Based on the building’s plastering, it may be more likely that it was used as a water-storage pool. A cistern (unpublished) recorded c. 20 m northeast of the building and outside the excavation area may have fed the pool, although the association between them has yet to be found. If the building was indeed a pool, the plaster on the top of the northern wall might show the full height of the original walls.
The Northern Complex. Several walls (W408, W441, W443, W457, W458) and a plaster floor (L415) uncovered in the northern part of Area D were probably part of one building. Only the foundation course of W441 was preserved, founded on bedrock; it was built of a southern row of medium-sized partially dressed fieldstones and two northern rows of smaller fieldstones. Wall 443 was built of a western row of medium-sized partially dressed fieldstones and an eastern row of smaller fieldstones; the two walls form a corner. Rows of fieldstones that had collapsed from the wall (L412, L454) were found west of W443. Wall 408 was the building’s eastern wall. This wall was not preserved and was identified by the edge of the plaster floor that abutted it (L415). In the place where the wall had been, fieldstones and a few broken ashlars were found displaced from their original location; they were probably the wall’s lower course. The wall’s other stones had collapsed westward and covered floor L415 (L402; hereinafter, the destruction layer). The building’s northern wall was not revealed; it may lie beyond the limits of the current excavation.
Plaster Floor L415 abutted the northern side of W441 and the eastern side of W443. The floor resembles the one found in the pool, but unlike that floor, it is smoothed on top. The floor plaster covered part of the row of fieldstones in W441, showing that this was the wall’s foundation and the wall itself was narrower (0.4 m). White bedding composed of small and medium-sized pebbles (L456) beneath the plaster floor abutted Walls 411 and 443. Another bedding layer (L429) was made of plaster mixed with gravel and set on top of a layer of gray soil; the bedding abutted the east end of W441, showing that the building had an additional room or courtyard in that area. Similar bedding (L450) was found to the west of W443. The excavation in the area was limited, but here too, the bedding suggests the presence of a room or courtyard floor. A layer of small fieldstones was discovered in several places in the excavation, mixed with soil placed on top of the bedrock (L428, L438, L446, L452). These stones were inserted where the bedrock surface required leveling before the buildings could be constructed.
Walls 408 and 443 collapsed westward, and the rubble was arranged so that it was possible to trace the courses that made up the walls. A layer of white calcareous powder mixed with plaster fragments (L422; max. thickness 0.2 m) that was unearthed beneath the collapsed stones from W408 probably came from the plaster that coated W408. The westward collapse, the orderly rubble, and the layer of powdered plaster covering L415 indicate that the structure was deliberately destroyed, probably by mechanical equipment.
A destruction layer (L401, L402, L404–L407, L414, L416–L419, L421) on top of the remains described above covered the entire surface of Area D and was characterized by powdery gray soil. The destruction layer was thicker toward the south, showing that the area sloped southward in the past and was not on a level as it is today. In the area of the northern structure, the layer also contained stones from the collapsed walls.
The ceramic finds from the northern structure include body fragments of Black Gaza Ware vessels (not drawn) and a twentieth-century CE glazed bowl (Fig. 21:2), both found in L429.
Fragments of worn pottery, including a Persian jug (Fig. 23:1), a Roman jar (Fig. 23:4), and a Byzantine LRC bowl (Fig. 23:6), were discovered among the fieldstones on the bedrock. The most recent pottery belongs to the Late Ottoman period; it includes an imported bowl from Albisola, Italy (Fig. 21:6), a porcelain cup (Fig. 21:9), a basin (Fig. 21:12), and jugs (Fig. 22:5, 7); a stone bead was also found.
The destruction layer, covered the remains, yielded a few fragments of ancient worn pottery, including Persian imported amphorae (Fig. 23:2, 3) and Byzantine LRC bowl (Fig. 23:5) and a base of Byzantine imported bowl (Fig. 23:7). Most of the finds in the destruction layer date from the Late Ottoman period, including glazed bowls (Fig. 21:1–3), a simple bowl (Fig. 21:4), a carinated bowl (Fig. 21:5), porcelain coffee cups (Fig. 21:8, 10) and a porcelain cup/bowl (Fig. 21:11), holemouth cooking pots (Fig. 22:1–3), a cooking pot/jar (Fig. 22:4), jars (Fig. 22:8–10) and a tobacco pipe (Fig. 22:13). Among other finds were a stone bead(?), a fragment of plaster with cloth and paint marks, the probable remains of a mat or rug, a sickle, a metal peg, and a horseshoe. The destruction layer also yielded animal bones and fragments of glassware (below).
The finds dated the construction of the northern building to the Late Ottoman period, and it probably continued to exist together with the southern building until the village of Qula was abandoned in 1948. In the 1960s, the village of Qula was deliberately destroyed as part of a government decision (Shai 2002 and see references there). Before the area’s excavation, preliminary inspections were carried out to the west of the buildings, which revealed nothing. Hence the two currently excavated buildings are the most westerly in the village.
Almost all the retrieved glass fragments—principally bottles and beakers, along with several windowpanes—were found in Area D and date from the Late Ottoman and British Mandate periods, corresponding to the rest of the finds from this area. On the other hand, few fragments were collected from Areas A, B, and C, but almost all were dated, as in Area D, to the Ottoman period or later.
Only two fragments are dated to earlier periods. One, found on the surface in Area A (L100, B1054), is a small fragment of a handle and a rim belonging to a lamp or a cosmetic vessel made of greenish glass and dated to the Byzantine period. The other, unearthed in Area B (L2018, B2029), is a small rim fragment of a bowl or a bottle made of colorless glass and decorated with two horizontal blue trails. Judging by the fabric, it may be dated to the Mamluk period.
Numerous chunks of raw glass were recovered from the topsoil in Area A. Among these chunks, two groups must be distinguished: 16 chunks that are of a light greenish hue, bear light brown weathering, and are full of small holes (Fig. 24); and 63 chunks of various sizes that are very dark, almost black, in color (Fig. 25). As all the chunks were found in topsoil and instead of any evidence of production stages or of the final glass products, these are most probably debris from modern-day production that took place somewhere in the vicinity and used the southeastern part of the site as a dump.
The excavation yielded 37 pieces of ammunition (see Appendix: Table 1), mostly shrapnel/musket, 0.303 British, 7.92 mm Mauser, 7.62 mm NATO, and 9 mm Parabellum bullets and cartridges. Several other items may also be part of military equipment. The type, manufacturer, and date of manufacture of most of the ammunition were identified.
Shrapnel/Musket Bullet (see Appendix: Table 1: Nos. 14:II; 17; 19:II; 20; 21). Four bullets of this type were found. On the whole, it is hard to differentiate between shrapnel and musket bullets. Nevertheless, based on their high-quality tooling and the resemblance between the finds, it is highly likely that these are shrapnel bullets, which were in use from 1917–1948.
0.303 British Bullet. The main ammunition used by rifles and machine guns of the British Empire in both world wars. Ten pieces were found (nine cartridges and a bullet), one of which is dated to the First World War (see Appendix: Table 1: No. 6) and eight to the Second World War (see Appendix: Table 1: Nos. 1–5; 7:I, II; 8:I; 22). Cartridge No. 7:I was manufactured in Fascist Italy; it was probably collected from a battlefield in North Africa. The other cartridges were produced by various manufacturers from the allied powers: Britain, the United States, South Africa, Canada, and Australia/New Zealand. Firing-pin impressions on several cartridges show that they were fired by Bren machine guns—Czech machine guns that were common in the armies of the British Empire during World War II and used by all parties involved in the War of Independence in Israel.
7.92 mm Mauser Bullet. The main type of ammunition used for German rifles and machine guns in both world wars. It arrived in Israel during the First World War in supplies for the Ottoman army, which used it as its main ammunition. Two cartridges dated to these years were found in the excavation (see Appendix: Table 1: Nos. 2, 3). Another cartridge is dated to the Second World War (see Appendix: Table 1: No. 4), probably arriving in the region after being taken from a battlefield in North Africa. In Israel’s War of Independence, all parties used this ammunition, which was also the state of Israel's main ammunition in its early years. The excavation yielded three such cartridges dated 1948 (see Appendix: Table 1: Nos. 9; 10; 14:I). The cartridges were manufactured in Czechoslovakia, Israel’s leading supplier of this type of ammunition at the time.
7.62 mm NATO Bullet. The main ammunition used by Western armies from the 1950s onward is still in use today. The current excavation yielded five cartridges of this type. Two were produced by the Belgian FN company (1955–1956; see Appendix: Table 1: Nos. 15:I; 16:I) and the three others by the Israel Military Industries Ltd. (1958–1959; see Appendix: Table 1: Nos. 7:III; 8:III; 15:II); they probably come from IDF training exercises in the late 1950s and early 1960s. Based on the type of cartridges, these exercises made extensive use of anti-tank rifle grenade launchers.
9 mm Parabellum Bullet. Five items were identified, three of which are cartridges produced by Israel Military Industries Ltd. (see Appendix: Table 1: Nos. 14:II; 17), made in the 1950s and probably remaining from IDF military exercises. A German cartridge was also recovered (see Appendix: Table 1: No. 19:I) and a complete bullet made in Italy (see Appendix: Table 1: No. 18:I), possibly from the War of Independence.
Unidentified ammunition that was not dated was also found, including three cartridges (see Appendix: Table 1: Nos. 8:II; 18:II; 23) and a shell fragment (see Appendix: Table 1: No. 24). Several other items that may have been in military use were also found: a brass ring used to secure fabrics and leather (see Appendix: Table 1: No. 13:II), a brass belt cover (see Appendix: Table 1: No. 14:V) and a metal canister lid (see Appendix: Table 1: No. 16:II).
The military finds unearthed in the excavation date from three periods of military activity at the site: (a) the First World War, probably between late 1917 (the ‘Battle for Jerusalem’) and September 1918 (the breaching of the line between the two ‘Auja rivers); one of the German cartridges was found beneath the floor of the southern building in Area D, which dates it to the post-World War I years; (b) the War of Independence; eight cartridges dating from 1948 were found in the excavation and historical sources record a series of fierce battles that took place at the site during Operation Danny in 1948; (c) the late 1950s and early 1960s; seven cartridges were found showing apparent evidence of IDF training in the area, which ended on the eve of the Six Day War.
The assemblage of faunal bones comprises 13 identified bones (see Appendix: Table 3) that were discovered in Area D. The commonest species in the assemblage is canine (N=5; 38.4%), including jackals, foxes, dogs, and wolves; the bones were not identified to the level of gender. The second commonest species is sheep/goat (N=4; 30.8%). The assemblage also includes rabbit, gazelle, cattle, and chicken (each N=1; 7.7%); one bone of each was found.
No cut marks or other man-made marks were observed on the bones. Non-domesticated animal species are present (rabbit, gazelle, and some of the canines). The bones are, therefore, probably those of undomesticated animals that lived wild in the region, as well as domesticated sheep/goat.
The current excavation revealed the agricultural and industrial hinterland of the village in Areas A–C. The land here was used for water storage, quarrying, building stones, burial, wine production, and other agricultural activities. Segments of access roads between the agricultural areas and between them and the settlement were also uncovered. Since most of the remains here are poorly preserved and yielded meager finds, they are difficult to date. Even in cases where finds were discovered, they usually came from alluvium. The ceramic finds include a few worn potsherds from the Iron Age and the Persian, Roman, and Byzantine periods, most of which date from the Late Ottoman period. In most cases, this cannot be used to show when the installations were constructed. The few finds discovered in Areas A–C probably provide a date range for human activity in these areas and settlement in the region, and the extensive Late Ottoman finds show that some of the installations were still used in this period. Some of the installations were used for a considerable length of time, including the water cisterns, where ancient phases of use were identified—like ancient quarrying linked to the cistern and the sealing of earlier openings—and later phases, which include the use of iron beams and concrete in the cisterns, showing that they were still functional up to the twentieth century CE. Mosaic tesserae scattered on the surface were probably used in installations dating from earlier than the Late Ottoman period, which have either not been preserved or have not yet been excavated. Glass-production waste and raw materials scattered on the alluvium in the northern part of Area A indicate that glass was produced nearby.
Area D contained remains of buildings found in the westernmost part of the village built in the Late Ottoman period and the British Mandate era and abandoned in 1948. The northern complex may have been used as housing, while a pool in the southern complex, built after World War II, provided water for the village’s farmland. Most of the area’s pottery, glass, and metal finds date from the Late Ottoman and British Mandate periods. The pieces of ammunition date from the First World War, the War of Independence, and the 1950s and are evidence of battles fought in the region.