Area A. The southern part of the reservoir’s opening (2.5 × 3.5 m; Fig. 3) was exposed. The opening was hewn in the chalk bedrock (L105). A wall (W12; width 1–2 m) running along a higher elevation than the opening, to its south and east, was meant to reinforce the bedrock and prevent its collapse when runoff flowed into the reservoir. Wall 12 was built of dry fieldstone construction directly on the bedrock, and was preserved two–three courses high. An eight-step staircase (L107; width 1.0–1.5 m, length 3.5 m; Fig. 4) was hewn southwest of the opening. At the top of the staircase was a circular installation (L110; diam. 1.2 m, depth 0.3 m; Fig. 5). Its floor was leveled bedrock, and the walls were built of fieldstones preserved to a height of three courses. The installation may have served as a water basin or settling vat. Remains of a mortar floor bedding (L104) were discovered slightly south and east of W12. Bedding 104 was laid above a stone pavement that was partially discovered in the past, just to the south of the excavation area (Stratum XII in the excavation of the tell; Herzog 2002:20, Fig. 8). The mortar foundation may have been part of a water channel that did not survive, which conveyed runoff to the staircase and from there into the reservoir. A fieldstone wall (W16; length 3 m, width 1 m) was built adjacent to a hewn niche in the southern wall of the reservoir’s opening; it was exposed in previous excavations directed by Z. Herzog and Y. Guvrin (unpublished). Wall 16 may have been meant to support the bedrock and prevent its collapse.
In its first phase, the water reservoir’s opening seems to have been smaller, and it was widened after the bedrock walls around it collapsed. The new, hewn opening was stepped and reinforced by Walls 12 and 16. The northern continuation of W12, which was similarly built on the bedrock around the reservoir’s opening (full diam. c. 7.5 m). Wall 12 canceled the southern wall of an eighth century BCE temple (Strata XI, X; Herzog 2002: Fig. 12), suggesting that the reservoir opening was fixed after the temple was no longer in use.
 
Area B was opened near a shaft that was excavated in the past, where a water channel and a tunnel meet the western wall of the citadel (Strata X–VIII; Herzog 2002: Figs. 12, 15, 16). A soil fill and collapsed mud-bricks (L103, L106) were exposed; below them were sections of three walls (W11, W13, W15; Fig. 6). The walls were built of fieldstones preserved two–five courses high (0.3–1.0 m); the western part of W13 had slid down into the shaft. The walls were exposed in previous excavations, and over the years were partially destroyed and partially covered with soil fill.
 
Pottery. Fragments of pottery vessels dating to the Iron Age II, Persian and Hellenistic periods were discovered in the soil that filled the water reservoir opening. Several of the sherds were ascribed to the Iron IIA (Strata XII and XI on the tell; Singer-Avitz 2002). These included carinated and round bowls (Fig. 7:1–4), cooking pots with thickened and triangular rims (Fig. 7:5, 6) and a cooking jug with a wide neck (Fig. 7:7). Similar vessels were discovered in Strata V and IV at Tel Lachish (Zimhoni 2004: Figs. 25.2:6, 10; 25.28:2; 25.33:5) and in Stratum XI on Tel ‘Arad (Singer-Avitz 2002: Fig. 8:5–7). Most of the sherds discovered in the excavation date to the Iron Age IIB and IIC (Strata X–VI on the tell; Singer-Avitz 2002: Figs. 10–49). These include bowls that have straight, carinated or round sides and a rim that is either thin or thickened and folded out (Fig. 8:1–14); kraters with a thick rim that is folded out (Fig. 8:15–17); cooking vessels that have a wide neck or narrow neck with a ridge (Fig. 8:18–25); juglets (Fig. 8:26, 27); hole-mouths (Fig. 8:28, 29); and a base of a lamp (Fig. 8:30). Several fragments of pottery vessels from the Late Persian–Early Hellenistic period were discovered near the surface, including round bowls (Fig. 9:1, 2), a krater with a high neck (Fig. 9:3), jars (Fig. 9:4–6) and jugs (Fig. 9:7–10).
 
Stone Artifacts. Four round, hard limestone (diam. 5.5–7.0 cm; Fig. 10:1–4) items—possibly used for grinding foodstuffs—were discovered in both excavation areas.
 
Faunal Remains
Nuha Agha
 
The faunal remains from five excavation loci (L101, L103–L105, L107) were examined. Baskets with mostly Iron Age pottery and only less than 5% of the finds dating to the Persian period were ascribed to the Iron Age. In three other loci the finds were quite mixed and therefore were not treated. The baskets were divided into three groups according to the date of the pottery they contained: Iron Age IIA–C, Iron Age IIB–C and Iron Age IIC–Persian period (Table 1). The bones were gathered by hand; since they were well-preserved, they were washed with tap water only. The work method conforms to the protocol practiced by the Israel Antiquities Authority, which has been described in detail in previous publications (Agha 2014).
 
The assemblage comprised 109 bones, teeth and horns, of which 89 items (c. 80% of the assemblage) were identified according to species or a particular size group (Table 2). The bones were well preserved, exhibiting neither cracks or surface peeling nor any signs of damage caused by the weather—indicating they were probably not exposed on the ground prior to being buried. However, three bones bore gnawing marks caused by rodents and predators. The assemblage will be described below from earliest to latest.
 
Table 1. Animal bones
Period
Locus
Basket
Total number of bones
Iron IIA–C
105
1010
12
Iron IIB–C
103
1003
 
66
104
1004
105
1006, 1011
107
1008
Iron IIC and Persian
101
1001, 1005
11
Total
 
 
89
 
Table 2. Animal remains according to species and size groups
Period
 
Species/size
Iron IIA–C
Iron IIB–C
Iron IIC and Persian
NISP
% NISP
NISP
% NISP
NISP
% NISP
Cattle
 
 
8
12.1
5
45.5
Dromedary camel
 
 
1
1.5
 
 
Horse
 
 
1
1.5
 
 
Donkey
 
 
1
1.5
 
 
Large-sized animal
 
 
10
15.2
1
9.1
Sheep/goat
5
41.7
21
31.8
3
27.3
Goat
 
 
1
1.5
 
 
Sheep
1
8.3
2
3
 
 
Pig
 
 
1
1.5
 
 
Medium-sized animal
6
50
20
30.3
2
18.2
Total
12
100.0
66
99.9
11
100.1
 
Iron Age IIA–C. Twelve bones of sheep/goats or of a medium-sized animal were found in the soil that filled the water reservoir opening. The sheep/goat remains included six bones: two pelvis fragments, a mandible without teeth, radius, metacarpal and metatarsal. The radius belongs to a sheep that was at least ten months old. The pelvic bone bears cut-marks indicating dismemberment. The remains of the medium-sized animal include six bones: two rib fragments, a cranial fragment, two radius fragments and a tibia. On the radius and the tibia were cut-marks indicating dismemberment.
 
Iron Age IIB and IIC. This is the largest assemblage (NISP=66), with the greatest variety of species—cattle, sheep/goat, Equidae, dromedary camel, donkey and pig—probably due to its size. Most of the assemblage (NISP=42; c. 64%) was gathered from the soil fill in the opening of the water reservoir.
The cattle remains include eight items: a mandible with a sequence of first to third molars, two maxillae with the two first and second molars, a metacarpal, axis vertebra, a femur, an ulna and a scapula. Cut-marks indicative of dismemberment were noted on two of the bones. The remains include body parts that are rich in meat, probably consumption waste, along with body parts with little meat, apparently slaughtering waste. The remains represent at least one individual over 3.5 years old. In addition, ten bones belong to a large-sized animal, one of which bears a cut-mark indicating dismemberment.
The sheep/goat remains include twenty-four bones, teeth and horns. Two bones (a tibia and a calcaneus) belong to sheep that were 2.5–3.0 years old or older. A humerus belongs to a goat over ten months old. The rest of the sheep/goat remains represent at least three individuals: two adults, judging by the dental erosion and the bone fusion, and a fetus, as indicated by an unfused proximal end of a metatarsal. At least one individual was male, as indicated by the shape of the pelvis. Six of the bones bore cut-marks, of which four are signs of dismemberment and two are indicative of skinning. In addition, ten bones belonged to a medium-sized animal, presumably a sheep/goat. Four bones bore cut-marks: three butchering marks and one indicative of meat removal. Scars identified on three of the bones (a radius, a humerus and a metatarsal) are characteristic of striking the bone with a pounder in order to extract marrow. The remains of the sheep/goats and similar-sized animals include all parts of the skeleton, from the head to the feet. The fracture pattern of some of the bones indicates they were broken while still fresh in order to extract their highly caloric bone marrow.
The pig, camel and donkey are each represented by one bone. Another bone belongs to Equidae. The camel bone, an atlas vertebra, is extremely important because it bears a cut-mark that is indicative of decapitation (see below).
 
Iron Age IIC and Persian Period. Eleven Bones, teeth and horns that belong to two farm animals—cattle and sheep/goat—and to similar-sized animals were discovered in the fill above W12.
The cattle remains include five bones: an unfused distal humerus, cervical vertebra, a cranial fragment with part of a horn, a fused distal metacarpal and an unfused pelvis. The bones belonged to at least three individuals: the state of fusion of the pelvic bone seems to indicate that it belonged to an individual that was 7–10 months old at the most; the unfused distal humerus suggests that the individual it belonged to was no older than 12–18 months; the fused distal metacarpal belonged to a third individual, which seems to have been no older than 2.0–2.5 years of age. Cut-marks were discerned on three of the five bones; two of the marks are indicative of dismembering the animal and one mark is indicative of skinning. The pelvic bone of the young individual has a cut-mark indicative of dismemberment. A vertebra belonging to a large-sized animal, probably cattle, was also discovered.
The sheep/goat remains include three metatarsal bones, a maxilla fragment and part of a pelvis. One of the bones bears a cut-mark indicative of dismemberment. The first and second molars are attached to the maxilla fragment, and their erosion fits that of an adult. Two bones (a scapula and a tibia) belong to a medium-sized animal, probably a sheep/goat; one of the bones exhibits burn marks.
 
The assemblages from Tel ‘Arad are small, and represent selective gathering from mixed loci. Phalanx bones are completely absent from the assemblages, probably due to the manner by which the bones were collected. Nevertheless, an examination of the assemblages makes it possible to draw general conclusions regarding the herd interface in the Iron Age II. The representation of all parts of the skeleton and the cut-marks indicate that all of the slaughtering stages, the consumption of the meat and discarding of the waste were done on site. The mortality ages indicate that the sheep/goat and cattle herds included mature individuals along with young individuals, hence they were probably used for producing secondary products as well as meat. The assemblage from the Iron Age IIB and IIC exhibits a composition and species diversity similar to other assemblages from the Iron Age. The few bones belonging to sheep/goats identified at the level of species might indicate a preference for sheep in the herds, although goat remains were also identified. A claim that frequently appears in the research literature is that sheep were raised mainly for their wool, and that they were not adapted to arid regions due to their weak stamina. However, at Tel ‘Arad and in other Iron Age assemblages in the Be’er Sheva‘ Valley and its surroundings sheep are more prevalent than goats, even though goats are better suited to the region due to their endurance to harsh conditions. This should probably be ascribed to the resilience of a mixed herd in the face of epidemics and diseases, as well as its ability to exploit pasturage more efficiently.
 
The special importance of the assemblage lies in the discovery of the camel bone bearing a cut-mark. To the best of my knowledge, this is the first cut-mark on a camel bone from the Iron Age to be discovered in the country. Marks of this type were not reported in other Iron Age assemblages that included camel remains, such as those discovered at Tel Be’er Sheva‘, Tell Jemmeh, Jericho, Tel Rehov, Qadesh Barne‘a, Lachish, Mezad ‘En Qadis, Tel Dan, Tel ‘Ira and Tel Qasile. The camel remains from Tell Jemmeh (Wapnish 1981) were used to support the argument that they were used only as pack animals. The cut-mark now discovered for the first time on a camel bone suggests that the animal was utilized for food as well. However, this conclusion must be considered with caution, as the bone was found in a fill.

The quarrying and construction undertaken in the reservoir opening indicate that the water system, which was hewn in the Iron Age IIB, was used for a very long time. The reservoir made it possible for the Iron Age citadels to maintain an ordinary existence in an arid environment, without readily available sources of water. During the Iron Age, ‘Arad served as a military stronghold on the border of the Kingdom of Judah and as a center of trade and worship at a crossroads between Judah, the ‘Arava and Edom in the east and south and the coastal plain in the west. The combination of a military and commercial center with the availability of water in the complex reservoir facilitated the continuity in establishing a series of citadels over time.