Six excavation areas were opened to the west of Road 66 (A–F; Fig. 1). In Area A (25 sq m), a settlement occupation level and a well dating to the Pre-Pottery Neolithic C period were exposed. In Area B (50 sq m), a bedrock outcrop with hewn and built tombs was unearthed; jars set beside the tombs (funerary offerings?) date to the Middle Bronze Age. A burial cave containing artifacts dating to the Intermediate and Late Bronze Ages was also revealed in this area. In Area C (c. 75 sq m), built and rock-cut tombs were found, but could not be dated. In Area D (25 sq m), a burial cave with loculi that dates to the Roman period was exposed. In Area E (50 sq m), part of a cave with a collapsed ceiling was excavated; the accumulated fills inside the cave date to the Byzantine period. In Area F (50 sq m), an occupation level dating to the Late Chalcolithic period was exposed.
Area A. One square was excavated, yielding an occupation layer and a well dating to the PPNC. Several potsherds from the Roman period were found in the uppermost stratum along with modern finds. A level of small stones and crushed chalk over bedrock—a prehistoric occupation level—was exposed at a depth of one meter below the surface; the occupation level extended beyond the limits of the excavation. A probe excavated into the stone level yielded flint tools, including sickle blades with deep denticulation and adzes dating to the PPNC (Fig. 2). In the center of the occupation level, which was partially excavated, were two large stones perpendicular to each other that constituted part of the opening at the top of a well (Fig. 3). They were set on top of an elliptical wall (at an elevation of 127 m asl), bonded together and into a foundation of limestones (the well-head). The stones protruded c. 0.2 m above the prehistoric occupation level. An installation was exposed to the west of the well, at the same elevation as that of the occupation level. The installation, resembling a built channel, comprised a row of basalt and limestone slabs flanked by additional slabs that lay on their side; it may have been used as a foundation for a wall that did not survive (Fig. 4). The architectural remains were dated to the PPNC.
The upper part of the well was constructed (Fig. 5), whereas the lower part was hewn through the soft limestone bedrock down to a friable and impermeable layer of chalk at a depth of 8.5 m (118.5 m asl). Four components were discerned in the installation of the well: a hewn shaft, construction above its lowest course, a built wall at the top of the shaft, and the stones that make up the well-head.
1. The hewn shaft was divided into four components according to their shape and type of rock. The upper three meters or so comprise a square shaft (Fig. 6). It was hewn asymmetrically, according to the characteristics of the rock, probably along vertical and horizontal cracks, some of which may have been fissures in the limestone bedrock that were enlarged (Fig. 7). The bottom part of the square shaft becomes narrow where it transitions to softer bedrock. Below that, the lower part of the shaft, dug to a depth of 7.5 m., is elliptical and gradually becomes narrower, as quarrying became irregular and followed cracks in bedrock. Finally, the bottom of the shaft is concave, hewn in friable rock (c. 1.5 m deep). On the wall of the well, above the constriction, was a round rock-cutting; above it were four arched-shaped slots (max. depth 2 cm; Fig. 8).
2. A feature consisting of two large, flat hewn blocks of hard limestone was built c. 1 m above the bottom of the well. The stones, trapezoidal and partly dressed, were arranged to conform to the dimensions of the shaft (Fig. 9). They were most likely intended to filter mud and dirt that accumulated at the bottom of the well, as can be seen in modern wells.
3. The hewn well-head consisted of an elliptical wall built of small- and medium-sized stones that survived to a height of 4–10 courses. The wall was built along a series of natural recesses at the top of the shaft. The stone courses in the northeastern part of the elliptical wall were constructed with a slight inward incline, resulting in the narrowing of the shaft’s opening (Fig. 10).
4. The well-head stones were set perpendicular to each other on top of the elliptical wall, and were held in place by a fill of small stones. Two of the original stones survived, both on its southeastern side. One large stone (length 1.45 m, width 0.45 m) was curved on the side facing the opening. It had three vertical grooves, probably resulting from the repeated friction of drawing ropes tied to water jars from the bottom of the well (see Figs. 3, 5).
The soil and stone fill within the shaft (c. 15 m3) consisted of four layers:
1. An accumulation of wet, muddy soil at the bottom of the well, beneath the constructed filtration element (No. 2 above). Charcoal, notched flint tools, a cow bone and lumps of clay were found in this layer, which is ascribed to the use-phase of the well.
2. An accumulation of crumbly soil above the filtration element (No. 2 above); the moisture in this layer decreases toward the top. It contained numerous limestone rocks, medium-sized stones of chalk and several basalt slabs similar to those found in the built installation on the occupation level outside the well. Many archaeological finds were discovered in this layer, among them animal bones, human bones, remains of charcoal and organic matter, and flint tools, including sickle blades with deep denticulation and a fragment of an ‘Amuq-type arrowhead. Other finds include stone objects such as a bowl fragment, limestone pounders, a basalt grinding vessel and a pumice object. The animal-bone assemblage includes a jaw bone of a pig. Three individuals could be identified among the human bones: a woman c. 19–20 years of age, a man c. 30–40 years old and a third individual of undetermined sex and age (Fig. 11). Two gravestones were found at the top of the layer: one cylindrical (Fig. 12) and the other rectangular. Although the skeletons were not articulated, the archaeological finds—especially the gravestones—suggest that these were burials.
3. Compact soil containing many stones of various sizes in the elliptical, middle part of the shaft, below the point where it narrows. The finds were meager; they included charcoal, organic material and several stone objects, among them two limestone hammerstones. The most significant find from this layer is the animal bones, particularly those of sheep and goat. Some of the bones exhibited butchering marks.
4. An accumulation from the upper square, rock-cut part of the shaft to the top of the elliptical wall. The soil was very compact, rendering it difficult to excavate. Many stones of various sizes and basalt slabs, similar to those from the installations from the top occupation level, were discovered in it. At the bottom of this layer were large limestone pieces that probably collapsed from the well-head. Charcoal, organic matter and flint tools were found, including several blades with deep denticulation (Fig. 13) and an ‘Amuq-type arrowhead.
In conclusion, the ‘Enot Nisanit well was hewn in soft limestone down to a layer of friable chalk that was impermeable to groundwater. The well received its water from the aquifer, a reservoir of groundwater in an unsaturated region, above a water-bearing layer of bedrock. Wells of this kind, which were hewn into the aquitard (a layer of bedrock allowing the slow passage of water) and characterized by relatively low-volume water flow and large seasonal fluctuations, apparently supplied a sufficient quantity of water for the residents of the site. The well was part of a PPNC agricultural settlement. This date is based on the assemblage of flint tools and the absence of pottery vessels. The burial within the well, the fill that was excavated in the bottom layer of the shaft and the accumulations in the middle part of the shaft, all date to the same period. The PPNC site apparently continued to exist for some time after the well was abandoned. The upper layer of accumulations in the shaft probably dates to the last days of the site, or shortly after it was abandoned.
Area B. Cist tombs and rock-cut tombs that could not be dated were excavated in two squares aligned in a general northeast–southwest direction were excavated perpendicular to the road. The lower of the two squares was situated along the edge of a modern drainage ditch, dug alongside the road, while the upper square was situated on a higher bedrock terrace. Several tombs in the upper square and in the balk between the two squares were not excavated, but their top was revealed. Potsherds dating to the Intermediate, Middle and Late Bronze Ages, as well as the Roman and Byzantine periods, were discovered in the accumulations around the tombs. Modern finds, including a platoon commander’s pin bearing the symbol of the Haganah, were collected from the upper level. These items date to the mid-twentieth century CE, the time of the Mansi immigrant transit camp. A concave bedrock surface that might have been the collapsed roof of a cave was also found. In the southern part of the square was a rectangular rock-cutting. At its bottom was a level of stones and alongside it was a cist tomb covered with stone slabs. Another cist tomb—both rock-cut and built, and covered with stones—was discovered in the northern part of the square; only its covering stones were exposed. Two jars dating to the MB were discovered; one was complete (Fig. 14) and contained a bead, several bones that could not be identified and small animal bones, including crab claws. It seems that the jars were part of the funerary offerings belonging to a tomb that was destroyed by later tombs. Other architectural remains were also exposed, but were not excavated; they were likely to have been used for burial.
Part of an elliptical cavity hewn in the soft chalk was exposed in the bottom square. It seems that this was part of a cave whose ceiling is missing; it might have been removed during land development (Fig. 15). The cavity was used for burials from the Intermediate Bronze Age to the LB. A level ascribed to the early burial phase carried pottery vessels—bowls, a red burnished teapot and amphoriskoi—dating to the Intermediate Bronze Age. Fragments of vessels, including pieces of an LB Cypriot bilbil jug, were associated with the late burial phase.
Area C. Four half-squares aligned in a general southeast–northwest direction according to the ground cover in the area (a grove of trees) were excavated. An elliptical cavity, hewn in soft chalk and missing its ceiling, was almost entirely exposed in the southeastern part of the area. An arch built of ashlars was revealed in the eastern part of the cavity. The piers of the arch abutted the bedrock. At the bottom of the arch were large, hard limestone ashlars. Potsherds dating to the Early Roman period were found in the accumulations excavated inside the cavity. Although neither any evidence of a grave nor bones were found, it is possible that the space was used for burial.
A similar elliptical cavity was exposed in the northwestern part of the area; it was partially excavated and its floor was not exposed due to the limitations of the excavation. In the cavity’s section, beginning almost at the surface, one could discern quarrying debris of small chalkstones. An east–west oriented cist tomb, which was not excavated, cut the quarrying debris and was built into it. A built wall, probably belonging to a second tomb, was located on the eastern edge of the rock-cut cavity. These two tombs are somewhat later than the quarrying debris.
Some fourteen elements identified as tombs were exposed between the two cavities, but were not excavated. They were all oriented in an east–west direction. Some of them were rock-hewn pit graves, whereas others were built and covered with stone slabs. The finds around these tombs were mixed and included potsherds that range in date from the Bronze Age to the present. EB and MB potsherds were found on the bedrock.
Area D. One square was excavated, exposing an elliptical loculus cave hewn in soft chalk bedrock. The cave’s ceiling was missing; it either collapsed or was removed in an earlier phase. Five loculi that opened to the north, northeast, south, southwest and west were documented; their location indicates that they were situated to conform to the contour of an existing cavity (Fig. 16). The loculi were not excavated; however, they were probably not long enough to accommodate a coffin or a corpse. It is therefore likely that they were used for secondary burial, possibly with ossuaries. Potsherds dating to the Roman period were found in the cave’s interior; it was customary during this period for the deceased to be interred in secondary burials. Part of a hewn elliptical cavity was exposed above the cave, in the southern part of the square. With the discovery of human bones on the upper level of this cavity and inside the loculus cave, the excavation was halted.
Area E. Two squares oriented in an east–west direction were excavated. No ancient finds were discovered in the eastern square, probably because of the land development conducted there over the years. In the western square, part of an underground cavity hewn in the limestone bedrock was exposed; its ceiling had collapsed. Potsherds dating to the Roman and Byzantine periods were gathered from the fill below the collapsed ceiling. The excavation did not reach the floor of the cave. It was thus impossible to determine the purpose of the cave or to date it with certainty.
Area F. Two squares aligned in a general north–south direction were excavated near the perennial spring of ‘En ‘Uzi and a road that separates the source of water from an ancient site. A layer of topsoil that yielded Roman-period potsherds was excavated in the northern square, close to the road. Beneath this layer was an occupation level with small and medium-sized stones, sherds, flint tools and fragments of basalt bowls that were dated to the Late Chalcolithic period (Ghassulian phase). In the eastern part of the square were several large stones that were possibly part of a wall. The occupation level and architectural remains extended beyond the boundaries of the excavation.
An occupation level that sloped from west to east and north to south and continued outside the limits of the excavation was found in the southern square, below topsoil. It contained small stones, crushed chalk and burnt stones. A rectangular limestone installation was built in the center of the level (Fig. 17); beside it was a fragment of a stone weight. Two superimposed floors were found in a section excavated in the eastern part of the square. Walls were exposed in the north, below the upper floor. Cattle bones and several human bones were found beside them. In the southern part of the section, below the level of the second floor, was a rock-hewn installation that contained several olive pits. Above the occupation level, between the floors and below them, were flint tools, potsherds and stone objects dating to the Late Chalcolithic period, including fragments of basalt bowls with lug handles and zigzag decorations on the inside of the rims.
The excavation yielded prehistoric settlement levels in Areas A and F. Later cemeteries, which were used for extended periods, were uncovered in Areas B–E. The prehistoric sites that were exposed were single-period settlements. It seems that the ancient inhabitants of the region recognized the advantage of locating their permanent settlements along the interface between the valley and the mountain, exploiting the available sources of water for their needs: the springs alongside Areas A and F. They even accomplished hewing an artificial well down to the level of the water table (Area A). The settlements were based on agricultural, as indicated by the remains of olive pits (Area F) and flint tools, especially the blade artifacts in Area A. A preliminary analysis of the animal bones suggests that the residents of the site made their living by raising sheep/goat and cattle.
The well in Area A dates to the PPNC. Until now, Pre-Pottery Neolithic wells were known mainly from Cyprus and in only two sites in Israel, both of which are currently underwater in the Mediterranean Sea: ‘Atlit Yam and Kefar Samir. Other wells are known in the country from later periods: at Sha‘ar Ha-Golan, dating to the Pottery Neolithic period; at Tel Zaf, dating to the Chalcolithic period; and in the western part of the Megiddo, where a water system dating to the Late Bronze or Early Iron Age was found. The ‘Enot Nisanit well reflects an impressive quarrying capability as well as extensive knowledge in the geology and hydrology of the region. Such knowledge made it possible to contain the source of water within the boundaries of the permanent site prior to the use of pottery and metallic vessels. The enormous investment in quarrying the well in a region featuring numerous natural springs, attests to the importance attributed to water resources for the needs of the local community and its livelihood. The components of the ‘Enot Nisanit well also show how important it was to ensure a supply of clean water, free of dirt from runoff (the well-head) and from the aquifer (the filter). It seems that the components of the ‘Enot Nisanit well is a prehistoric prototype, as they are similar in detail to those found in later wells, both historic and modern.
The extensive burial site is located on a rocky spur and along its fringes. The nature of the soft rock apparently dictated the extent of the site. Burial caves and several types of built and rock-cut tombs were noted. On the basis of potsherds gathered on the bedrock surface, the early phase of burial should be dated to the Early Bronze Age. Elliptical-shaped burial caves were first hewn in the Intermediate Bronze Age. Some of them were found to have been reused in the Late Bronze Age and possibly during the Roman period as well, usually for burial purposes. The cemetery served the residents of the region during historic periods for many generations. Its proximity to Tel Megiddo, which was an important national, regional and urban center in the Bronze and Iron Ages, would seem to suggest that at least some of the residents of the nearby tell were interred in the cemetery at ‘Enot Nisanit. The tombs from the Roman and Byzantine periods are also connected to the urban centers of the region, such as Legio-Maximianopolis, and probably also to the burials along the main, paved roads leading toward them (Tepper 2002; 2003). The soft rock and the ancient burial features were exploited again in the modern era, probably as late as the time of the nearby village of al-Mansi, in the middle of the twentieth century CE.
The excavation has contributed to our understanding of the prehistoric settlement pattern along the fringes of the Jezreel Valley and of the distribution of the Tel Megiddo and Legio-Maximianopolis necropolis over extended historic periods. Continued analysis of the finds from the well, including botany, microfauna, fauna, physical anthropology and other studies, will add significant knowledge about the history of the fauna, flora and humans in the region during the Neolithic period.