The remains of a rock-hewn cave were exposed (Figs. 1, 2). Two cisterns and two columbarium caves were exposed c. 100 m to its west and documented in the past (Permit No. A-5326). Soil accumulations that contained mixed pottery fragments, including a krater (Fig. 3:2), a jar (Fig. 3:5) and a cooking pot (Fig. 3:8) from the Hasmonean period, a jar from the Herodian period (Fig. 3:11) and numerous potsherds from the Byzantine period (not drawn), were found next to the cave.
 
The plan of the cave is not clear (see Fig. 1); its southern part had been destroyed prior to the excavation and these remains were only partially excavated because of safety considerations. The cave was hewn in friable limestone bedrock and the quality of the rock cutting is inferior. Only part of the cave’s floor (L13, L16) was exposed. It was covered with fill (min. thickness c. 0.35 m) that consisted of fine-grain light gray material—the result of the cave’s ceiling disintegration. The entrance to the cave was probably in its northwestern side, farther along a cavity with an arched ceiling and a fill that accumulated inside it and was not cleaned (L11 [west]; width 3.4 m).
 
A rectangular rock-hewn cavity (Loci 11 [East]; 3.6 × 4.5 m) in the eastern part of the cave had also an arched ceiling. A small niche (0.35 × 0.50 m, height 0.4 m) was hewn at floor level on the southern side of the cavity. The potsherds found in the fill on the floor of the cavity (L13) included cooking pots from the Hasmonean period (Fig. 3:7), a jar from the Herodian period (Fig. 4:10) and kraters with bow-shaped rims (Fig. 3:14, 16), jars (Fig. 3:18, 20) and a jug (Fig. 3:21) from the Byzantine period. Several tesserae embedded in plaster that probably originated outside the cave were found in the fill, as well as a coin whose date in the middle of the fourth century CE (IAA 115439) was based on its shape.
 
An arch-shaped niche (L12; 0.70 × 2.65 m, height 1.5 m; Fig. 4) was hewn in the bedrock wall south of the cave’s western wing; due to the damage in the cave it was difficult to determine if this was indeed part of the cave, although judging by its location and the nature of the rock cutting, this was probably the case. The niche was blocked with fill of terra rossa soil mixed with crumbled bedrock and small fieldstones. It contained pottery fragments that included a bowl (Fig. 3:1), jars (Fig. 3:3, 4), a cooking pot (Fig. 3:6) and a jug (Fig. 3:9) from the Hasmonean period; a jar (Fig. 3:12) from the Herodian period; a Late Roman C plate (see Fig. 3:13), kraters with bow shaped rims (Fig. 3:15, 17) and a jar (Fig. 3:19) from the Byzantine period.
 
A ceramic fragment incised with a cross-like mark, which may have belonged to a jar, was found in the niche (Fig. 5). The central vertical line, which extends down from the point of intersection, is almost twice as long as the right vertical line, and the left horizontal line is curved—probably the cursive form of the Hebrew letter Kof. It is possible that this letter specified the jar’s contents, as written in the Mishnah, Ma‘aser Sheni, 4, 11: “If a man found a vessel and on it was inscribed a Kof, this is Korban (sacrifice); if a Mem it is Maaser (Tithe); if a Daleth it is demai-produce (pro­duce not certainly tithed); if a Tet it is Tebel (produce certainly un-tithed); and if a Tav it is Terumah (Heave-offering)”. Another less plausible interpretation is proposed by J. Naveh and B. Mazar for similar engravings that were found on the potsherds that dated to the Persian and Hellenistic periods. This proposal says that this is the Phoenician letter Mem, which symbolizes the god Marnas, the principal deity of Gaza (IEJ 17:139, Pl. 32; Naveh, J. 2000. Hebrew and Aramaic Inscriptions. In D.T. Ariel (ed.). Excavations at the City of David 1978-1985 Directed by Yigal Shiloh, Vol. VI: Inscriptions [Qedem 41]. Jerusalem. P. 12, Inscription 26).