Horbat Bohu is situated on a low hill (c. 135 m asl) near the eastern bank of Nahal Bohu. The many archaeological remains in the area date mostly from the Byzantine, Early Islamic and Mamluk periods. In 1998, a survey was carried out by S. Lander (License No. G-149/1998), and in 2014, a Byzantine-period cemetery was excavated north of the current excavation (Permit No. A-7042). In 2015, a trial excavation was carried out at the site, revealing remains from the Byzantine to Mamluk periods, the main complex comprising a building that was probably a Byzantine-period monastery with several large wine presses, as well as water cisterns and other installations (Peretz and Michael 2017). In 2019, an excavation conducted c. 750 m to the west of the current excavation, revealed a pottery workshop, two large winepresses, cisterns and several tombs; most of the remains date from the Byzantine period, with only a few finds attributed to the Early Islamic and Ottoman periods (Permit No. A-8553). An excavation at Givʽot Etun, c. 2 km west of Horbat Bohu, uncovered a rural settlement comprising several buildings, an alley, two cisterns, loess quarries and refuse pits dating from the Byzantine and Early Islamic periods (Seriy 2017).
Eleven excavation squares were opened on the low hilltop, exposing fragmentary walls, floors, installations, habitation levels and refuse pits (Fig. 2). The remains are attributed to the Mamluk period, but their poor state of preservation precluded establishing a clear stratigraphy. Byzantine and Early Islamic-period pottery sherds and glass fragments appeared residually in the Mamluk stratum.
Parts of two stone-paved floors (L120, L128; Fig. 3) were composed mainly of dressed stones in secondary use and broken grinding stones along with fieldstones. Other rooms or spaces may have had beaten-earth floors, no longer preserved, as supported by the find of several extant clay oven (L137, L147, L149, L156). The preserved wall segments, which were built of fieldstones with incorporated dressed stones in secondary use, were preserved up to three courses. However, it was neither possible to reconstruct the missing walls nor to determine the plan of the structure or structures they belonged to. A column fragment incorporated in a wall (W125) possibly came from the monastery excavated in 2014 (Peretz and Michael 2017). A thick destruction layer (0.10–0.15 m) was found in several parts of the excavation, indicating that the structures were destroyed by fire.
A rather small pottery assemblage was retrieved in the excavation; most of the pottery dates from the Mamluk period (thirteenth–fifteenth centuries CE), and only a few sherds are from the Byzantine or Early Islamic period. The Mamluk-period pottery includes several deep crude bowls (Avissar and Stern 2005:82, Type II.1.1.3)—a thick-walled carinated bowl with a flat base (Fig. 4:1), a bowl with ribbed rounded walls and a slightly everted rim (Fig. 4:2), and a bowl with flared walls and a slightly everted rim (Fig. 4:3); three large bowls or basins with heavy everted rims (Fig. 4:4–6), one of which is squared (Fig. 4:5); several storage jars with long necks and thickened rims (Fig. 5:1–4; Avissar and Stern 2005:102, Type II.3.1.4); a very large jar with a similar neck and rim profile (Fig. 5:5), probably a pithos; a complete jug with a worn reddish brown painted decoration (Fig. 6) and additional sherds (Fig. 7) belonging to handmade vessels with a geometric painted decoration; and several sherds of various glazed bowls (Figs. 8, 9). The few Byzantine or Early Islamic-period sherds—most probably residual—include two lamps: a sandal lamp (Fig. 10:1) and the molded handle of a lamp (Fig. 10:2), the latter similar to Magness’ Form 3, Variant D (Magness 1993:251–254).
A few bronze finds—a ring, a knife, a bell and a large pendant—and several iron nails were found in the excavation. The ring (diam. 25 mm, thickness 4 mm) is a simple flattened circle. The knife (length c. 16 cm) was found on a beaten earth floor (L150) and can therefore be attributed to the Mamluk period; its handle was not preserved. The bell (L163, B1213; Fig. 11:1), retrieved in a Mamluk layer, is similar to bells from other excavations dating mainly from the Roman and Byzantine periods; hence, it may have been residual in the Mamluk layer. The pendant is in the form of a flower (L162, B1207; max. length 3.5 cm; Fig. 11:2) is an unusual artifact.
Glass Finds
Tamar Winter
The excavation yielded 16 poorly preserved glass fragments, including several hollow lamp-stems, characteristic of the Byzantine period (L128, L135, L137; not illustrated). The vessels attributed to the Late Islamic–Ottoman periods include a small wall fragment of a bowl made of dark purple glass and decorated with vertical mold-blown ribs (L136, B1176; Fig. 12:1). Bowls of this type, occasionally also bearing white horizontal trails, are characteristic of the Mamluk period. A similar, nearly complete bowl was discovered at Ramla (Gorin-Rosen 2009: Fig. 13:3).
Other finds from the Late Islamic–Ottoman periods include a bracelet with a circular cross-section, made of translucent glass and adorned with spirally-twisted, opaque black, yellow and white trails (L109, B1019; exterior diam. 7 cm; Fig. 12:2). Multicolored, spirally twisted bracelets of this type are characteristic of the Late Islamic and Ottoman periods. Bracelets adorned with spirally-twisted trails of opaque red, yellow and white glass are classified by Spaer as Type C2b and are associated mostly with the Ottoman period (Spaer 1992:4950, Fig. 4). Similar bracelets, unearthed at the site of Mary’s Well in Nazareth, were associated with the Mamluk period (Alexandre 2012:101–103, Fig. 4.10:5). Several plain glass bracelets with a D-shaped cross-section from the Late Islamic–Ottoman periods were also recorded at the site (topsoil, L127, L132; not illustrated).
Robert Kool
Thirteen coins were retrieved in the excavation. All the coins are copper fulus dated to the Mamluk period (IAA 161737–161749). Most are poorly struck, cut coins that can be roughly dated to the fourteenth and fifteenth centuries CE. A few coins that could be properly identified date from the reigns of Al-Nâsir Nâsir Muhammad (third reign; AH 709–741/1310–1341 CE; IAA 161742), Al-Ashraf Nâsir ad-Dîn Sha‘bân II (AH 764–778/1363–1377 CE; IAA 161741, 161747) and al-Ẓāhir Barqūq (AH 792–801/1390–1399 CE; IAA 161737). These coins complement a similar-dated group of a dozen Mamluk fulus excavated nearby (Peretz and Michael 2017; studied by the author). The coin finds clearly support the excavators’ conclusions that the hill site seems to have consisted of a single-phase Mamluk-period settlement.
Excavations at contemporary rural sites show the widespread and intensive use of small copper denominations for daily use in southern Bilād al-Shām during the fourteenth and fifteenth centuries CE. This widespread availability of small petty cash seems to have been a salient feature of the economy of the Mamluk state’s Syrian provinces.
Animal Remains  
Ma‘ayan Lev
The faunal assemblage includes a total of 63 bones (see Appendix: Tables 1, 2). The bones were identified using the osteological reference collection of the Archaeozoology Laboratory at the University of Haifa and were recorded using a standard diagnostic zone (Davis 1992). All the identified bones were measured (see Appendix: Table 3) using a digital caliper to the closest mm point, following von den Driesch (1976) and Davis (1996). Various taphonomic phenomena, such as burning, carnivore gnawing and weathering, were categorized according to Behrensmeyer (1978).
The 63 bones were identified to taxon (see Appendix: Tables 1, 2). The assemblage is dominated by cattle (Bos taurus, N=20, 35%), caprines (sheep, Ovis aries, and/or goat, Capra hircus, N=13, 23%) and equids (probably donkey, Equus asinus, N=13, 23%), with the presence of chicken (Gallus gallus domesticus, N=7, 12%) and camel (Camelus dromedaries, N=3, 5%). A single raptor bone was also identified (2%).
Taphonomic observations (see Appendix: Table 4) indicate that about half of the assemblage (N=29, 49%) exhibits bone weathering, with 10% exhibiting severe bone weathering (Stage 3). These observations imply that some of the bones in the assemblage were subject to prolonged subaerial exposure before burial, perhaps due to slow rates of garbage accumulation. The varying intensities of weathering may be due to bone depositions originating from various sources. None of the bones were burnt, apart from one carbonized caprine limb bone, suggesting cooking rather than roasting. Four bones exhibit fractures that affected the bone while it was fresh, and 17 bones show morphological attributes of dry breakage, the result of post-discard processes, such as trampling and crushing by sediment weight. A single butchery chop mark was identified on a camel limb bone. Carnivore gnawing marks are uncommon (N=1).
The bones retrieved in the assemblage suggest a livestock economy at the site, although the small size of the assemblage precludes a clear analysis of the site economy.
The hilltop site was settled in the Mamluk period, and several finds from the earlier Byzantine and Early Islamic periods were residual or incorporated in secondary use in the Mamluk stratum. It is possible that the Mamluk-period settlement was built on earlier remains, in the course of which the earlier remains were destroyed. Whilst the poor state of preservation did not permit an understanding of the nature of the excavated structures, they were evidently part of the previously excavated small rural settlement from the Mamluk period, where Mamluk floors, wall stumps and habitation levels were found in almost all the excavated areas (Peretz and Michael 2017).