A survey of the chamber cave revealed several thin layers, containing numerous animal bones and flint artifacts, found within a reddish brown terra rossa matrix. Preliminary observations attributed the assemblage to the Middle Paleolithic period (approximately 250,000–50,000 BP). Another cave, rich in speleothems, was discovered 5 m to the south; no archaeological remains were exposed and the relation between the two caves is unclear.
The aim of the excavation was to comprehend the basic stratigraphy of the cave and to identify the periods and nature of the human occupation. In addition, an attempt was made to reconstruct the paleoenvironment of the site’s immediate vicinity.
The current remains of the cave (6 × 10 m) suggest that it was originally large. It is divided into a southern part that consists of sterile reddish brown terra rossa and a northern part whose sediments are mostly anthropogenic. Two separate chambers were identified in the northern part: the main chamber to the south (width 3.8 m, height 7 m; depth unknown) and the smaller chamber to the north (width 1.7 m, height 2.4 m, preserved depth 1.1 m; Figs. 3–4). The original entrance to the cave is difficult to determine, although it is presumed to have been on the eastern side.
A total area of c. 20 sq m was excavated in a grid (1 × 1 m); squares were marked A-H along the south–north axis and numbered 1–6 along the east–west axis. The fieldwork was undertaken in units (5–10 cm), based on the changes observed in the color and texture of the soil. Only a small portion of the cave was excavated and all the sediment was dry sieved, while c. 20% was wet sieved, using a 1 mm mesh, to recover the small flint pieces and tiny bones. 
Based on the excavation, an attempt was made to reconstruct the geological and archaeological history of the cave, which is described from bottom to top (Fig. 4).
Phase 1: A chamber-shaped cave was formed due to intensive karstic activity, leading to the formation of speleothems.
Phase 2: Once the karstic activity in the cave had stopped, human occupation took place (Layers X–VIII). This is evidenced by the absence of in situ speleothems within the anthropogenic sediments, which are a compact light reddish yellow terra rossa with manganese oxide nodules, as well as broken carbonates nodules. The archaeological remains consisted of animal bones, flint artifacts and possibly the remains of a hearth. 
Phase 3: This phase is characterized by heaps of rock and broken angular stones (Layer VII), which originated from the collapse of the cave's roof and wall. This layer sloped from the north to the southern wall. At the bottom of the slope was an accumulation of numerous broken bones and flint items, many of them in a breccia. Archaeological remains were found above and below the collapse.
Phase 4: The sediment at the top archaeological layers (VI–II) was similar to that in Phase 2. These layers comprised thin horizons of flint artifacts and animal bones. Of particular note were the remains of a hearth with numerous burnt artifacts, including chips (Square C, Layer IV). 
Phase 5: The cave was abandoned and filled with reddish brown terra rossa with manganese nodules that formed a thick layer (I; thickness c. 2.6 m).
The Flint Assemblage
The lithic assemblage was analyzed using an attribute analysis methodology to describe the techno-typological characteristics of the items. The majority of the lithic assemblage is made of flint, with a few limestone pieces. Two main types of flint were used, the first is fine grained, semi-translucent and the second is more of a cherty variant, opaque and coarsely grained. A few of the artifacts are of a pinkish colored flint. Cores were made on both flint nodules and cobbles. Flint nodules are found within the limestone bedrock of the ed-Deeb ridge and in the Wadi Qana, where flint cobbles are also present.
The assemblage was analyzed according to the site stratigraphy; it includes Phase 2 (Layers X–VIII) and Phase 4 (Layers VI–II). Phase 3 (Layer VII), which contained only a few artifacts, was combined with Phase 4. The northern chamber was analyzed separately since its layers could not be easily correlated to the archaeological layers in the rest of the cave. The backhoe piles were separated due to their mixed contents. 
The assemblage (N=1496) comprised 813 lithic artifacts that originated from the excavated area and 683 lithic artifacts, which were retrieved from the backhoe dumps. Common to all the layers is a low percentage of cores, core trimming elements and primary flakes, indicating that the initial stages of core reduction took place elsewhere.
The lithic analysis seems to indicate that two technological systems were used alongside each other: the laminar technological system for the production of blades and the Levallois core reduction system, intended for the preparation of wider blanks.
Phase 2 (Layers X–VIII): An abundance of blades in these layers make up 21% of the debitage. Elements of the laminar system include single platform cores (Fig. 5:1), a ridge blade, two naturally backed knifes, as well as a number of thick elongated non-Levallois blades. The Levallois reduction system is 5% of the assemblage and the majority of the Levallois blanks are flakes. The Levallois component included flakes, one recurrent convergent core for blades and a Levallois blade. No Levallois points or short triangular flakes were found and only a few retouched pieces were exposed, mostly made on large and thick elongated blanks. The largest tool group is the side scrapers (Fig. 5:2), as well as retouched blades (Fig. 5:3), points and truncated pieces. One of the points resembled an Abu-Sif point, although the retouch is not as fine and as regular (Fig. 5:4).
Phase 4 (Layers VI–II): As in Phase 2, both the laminar and Levallois reduction systems are evident; however, the Levallois technological system is more common. The difference between the two phases is in the ratio of the two reduction systems. On average, the blanks are shorter and thinner than those in Phase 2. The Levallois reduction sequence is composed mainly of recurrent convergent cores for flakes (Fig. 6:1), Levallois flakes, blades and points (Fig. 6:2, 3). The laminar system includes a couple of blade cores, as well as the blades themselves. Side scrapers (Fig. 6:4) dominate the tool assemblage. Notches and truncated pieces are also present.
The lithic assemblage in the northern chamber is small; it is dominated by a high percentage of blades (Fig. 7:1, 4), as well as Levallois artifacts (Fig. 7:2). The few tools include side scrapers (Fig. 7:3), a point, notches and denticulates (Fig. 8).
The lithic assemblage from the dump is clearly mixed and includes components of all layers. All tools and cores coincide with the findings from the anthropogenic layers and include the same tool categories, as well as Levallois and blade cores.
The Fauna
A preliminary study of the faunal remains indicates that Cervids were common, including Mesopotamian fallow deer (Dama mesopotamica), roe deer (Capreolus capreolus) and red deer (Cervus elaphus). In addition, mountain gazelle (Gazella gazella) and aurochs (Bos primigenius) were also found (Fig. 9). Small game is scarce. Micromammal (rodent and insectivore) remains are present, including social vole (Microtus gunetheri) and field mouse (Mus macedonicus) that were identified.
Bone preservation is poor, animal long bones are covered with a thick layer of chalk incrustation. Better preserved are the teeth, mainly wild cattle and cervid jaws.
Wadi Qana is well known for its wealth of prehistoric sites, ranging from the Middle Paleolithic through the Natufian and Neolithic to the Chalcolithic period. The discovery of the Emanuel cave is significant, as it lies in the center of the Samaria hills, 25 km northeast of Shukbah. Up to date, Middle Paleolithic cave sites in the southern Levant are known mainly from the coastal plain and the western flanks of the Samaria hills.
Statigraphically, ten distinctive sediment layers were documented. Except for the top layer, all other layers consist of archeological artifacts and animal bones. The anthropogenic horizons are divided into two phases by a thick layer of collapsed debris. A clear difference exists between the lithic assemblages found above and below the collapse, although the lithic sample is small and the number of artifacts is low. Phase 2 (Layers X–VIII), bellow the collapse, seems to represent a predominantly laminar assemblage with large artifacts. By contrast, Phase 4 (Layers VI–II), above the collapse, reflects the use of the Levallois reduction sequence. At this stage of research, it is difficult to determine the time gap between the two phases. The lithic assemblage as a whole seems to belong to the Early Middle Paleolithic of  "Tabun Type D” . As in Hayonim Unit E, the laminar technological system coexists with the Levallois reduction sequence and the elongated blank production precedes the short blank manufacture. A TL date of 186±20 Ka has recently been assigned to Hayonim Unit E.
The faunal assemblage, rich in Cervides, particularly fallow deer, indicates a woodland habitat, whereas the exposed rodents characterize a more open grassland environment. The existence of rodents possibly reflects time periods when the cave was open and not occupied by humans. The duration of these time spans is not known. The predominance of fallow deer in the faunal assemblage points to an early Middle Paleolithic date, based on the resemblance to other early Middle Paleolithic faunal assemblages from sites, such as Qesem, Missliya and Tabun (see Fig 1). 
The latest U/Th date from the speleothems located on the cave’s bedrock is 213–191 Ka (Fig. 10). Hence, the archaeological layers postdate 191 Ka. It should be noted that artifacts and animal bones were found embedded in the surface of the speleothems crust at the bottom of the cave (Fig 11), indicating that a short time elapsed between the deposition of the speleothems and the human occupation.