Stratum III. A complex of three interconnected installations (L916, L920, L922; Fig. 3) hewn into the soft limestone bedrock and dating to the Early Chalcolithic period were found directly below the modern surface. They were previously interpreted as poorly preserved remains of a rock-hewn tomb (Covello-Paran 2013); however, subsequent reappraisal deemed this analysis to be erroneous.
Installation 916 is oval (1.70 × 2.35 m, depth 0.7 m); its eastern side was not well hewn, resulting in a rough sloping wall (Fig. 4). Accumulated remains in this installation (thickness 0.4–1.3 m) included numerous stones, fragments of pottery vessels, flint tools and poorly preserved animal bones. The scant datable potsherds include holemouth vessels (Fig. 5:1, 2), a krater (Fig. 5:3), bow-rim pithoi (Fig. 5:4, 5) and a handle (Fig. 5:6), all dating to the later phase of the Early Chalcolithic period. According to Getzov (pers. comm.), they postdate the Wadi Rabah culture and precede the Ghassulian culture. The flint assemblage (Yaroshevich, below) contained numerous cores and primary trimming elements; however, no datable tools were retrieved.
Installation 920 is roughly hewn and circular (diam. 0.7 m; Fig. 6). It has a sloping floor that connects with Installation 922, which is oval (length 1.7 m, depth 0.7 m; Figs. 3: Section 1–1; 5) and only partially preserved, due to damage caused by the burial in Stratum II. Installations 920 and 922 were devoid of finds.
 
Stratum II yielded a Roman period rock-hewn tomb comprising a square courtyard (2.2 × 2.2 m, depth 1.8 m; Fig. 7) and a burial chamber. Four rock-hewn stairs in the northwestern corner of the courtyard led down into it (Fig. 8). The courtyard was delineated on the north, south and west by rock-hewn walls. The entrance to the burial chamber was on the east, and was sealed with a roll stone (diam. 1 m; Fig. 9) that was found in situ. In the wall to the left of the roll stone was a rock-hewn recess that matched its shape. When the burial chamber was opened, presumably numerous times, the recess would have held the roll stone. The roll stone was not moved, and the burial chamber was not excavated. East of the blocked entrance was an elongated area (L918), perhaps a passageway or part of the burial chamber itself. It continued beyond the southern boundary of the excavation and was only partially exposed. The hewing of L918 cut through Installation 922 in Stratum III, and the rock cutters built a wall of ashlar blocks to patch or block Installation 922 (W921; Fig. 10).
The relatively small number of potsherds that were retrieved from the area of the tomb, including cooking pots of Kefar Hananya Form 3B, date the courtyard and the burial chamber to the Middle–Late Roman periods. The pottery in the courtyard (not illustrated) was fragmentary and included small sherds from types that span the Middle–Late Roman period and an intrusive sherd from Stratum I.
 
Stratum I. The only definable feature in Stratum I is a pit (L917; diam. c. 1.3 m), which cuts into the debris that accumulated in the tomb’s courtyard (Stratum II). This pit was identified by loose debris, sporadic stones and many body sherds of non-restorable pottery vessels that date it to the Early Byzantine period. Most likely, the pit is associated with domestic structures beyond the excavation limits.
 
The Flint Assemblage
Alla Yaroshevich
 
The excavation yielded 276 flint artifacts, all from a single rock-hewn pit (L916). The assemblage is homogeneous, the items all fresh, sharp and made on high-quality beige flint of Eocene origin. The majority of the items (N=156, 56.5%) are debitage; 21 (7.6%) are tools; the remaining 99 items (35.9%) were defined as chunks.
Within the group of debitage, flakes are predominant (N=92, 59%); cores and primary elements each number 27 items (17.3%); eight blades constitute 5.1% of the assemblage; and two core-trimming elements account for the remaining 1.3%. Consistent with the composition of this group, all but two cores were used to produce flakes (Fig. 11:1). They are small–medium in size and often retain remains of cortex. The remaining two cores are mixed blade-flakes and bladelet-flakes cores (Fig. 11:2). Nine cores have one striking platform, six have two platforms, three are cores on flakes and the rest are amorphous.
Medium-sized flakes were used as blanks to produce most of the tools in the assemblage. The most numerous groups are retouched flakes (N=7) and notches/denticulates (N=6). Four awls and two scrapers (Fig. 12) were also produced on flakes. Two retouched blades constitute the remainder of the tool assemblage.
The technological composition, in particular the high number of cores and primary elements, indicates on-site knapping activity. Typologically, the assemblage represents ad-hoc industry, i.e., a collection of expedient tools. The homogeneous appearance and the fresh look of the items, in conjunction with their provenance in a pit, suggest several knapping events, subsequent production of tools for immediate use, and finally disposal of the items into the hewn pit. The absence of formal tools prevents an affiliation with any particular period or culture.
 
This small-scale excavation increases the available data regarding settlement patterns at Kafr Kanna. The built remains from Stratum III, in conjunction with the associated pottery and flint artifacts, provide evidence for an open area with circular rock-hewn installations, which is associated with a myriad of activities, including flint knapping. Similar circular rock-hewn installation are known from other sites within the Early Chalcolithic horizon (see Getzov 2015).
The hewn burial cave upholds Alexandre’s claim (Alexandre 2011) that the village center served as the burial grounds for the Roman-period settlement that lay farther west at Karm er-Ras. The fragmentary finds from the current excavation span the Middle–Late Roman periods and enable us to date the burial activities on this hill.
The scant Byzantine-period remains indicate that the area was no longer used for burial during this period. Instead, the activity was most likely related to the settlement associated with the church.