Roman period (second–fourth centuries CE). A wall (W2; Fig. 3), founded on travertine sediment devoid of pottery, was exposed in the western part of the square. It had two stone faces bonded with gray, cement-based plaster. The western face of the wall was built of small and medium nari stones, while its eastern face was built of very small fieldstones, and plastered, probably at a later phase, with light gray–white plaster. The western face of the wall was deeper than the eastern one. The southern end of the eastern face was founded on buff-brown gravel fill. The wall continued to the northeast beyond the limits of the excavation. It was damaged by trial trenches that were dug prior to the excavation. Its southern part is cut by a later wall (W7; below). Below the gravel fill was a tamped floor (L9, L12; Fig. 4), made of gray travertine soil, travertine pebbles and small basalt fieldstones. A fragment of a marble slab in secondary use was incorporated in this floor. Sections of the same floor were discovered also in the southern part of the square (L19, L20). The floor was founded on travertine sediment devoid of pottery. The floor was at a lower level than the eastern, plastered face of W2. It is possible that originally the eastern face of W2 was also deeper, and the floor abutted it before it was plastered. It seems that W2 and the floor were part of an installation inside a residential building.
The pottery from this phase includes fragments of an ETS bowl (Fig. 5:1); of jars (Fig. 5:6, 7), which date to the Early Roman period; and of Kefar Hananya vessels from the Middle Roman period, such as bowls Form 3B (Fig. 5:2), bowls Form 1B (Fig. 5:3), a basin (Fig. 5:4), and a cooking pot Form 4D (Fig. 5:5).
Late Roman and Byzantine periods (fourth–sixth centuries CE). Remains of an open pool were revealed in the southwestern part of the square. The northern wall of the pool (W7; preserved height 1 m; Fig. 6) was built of small fieldstones, mainly soft limestone, bound together with gray cement-based plaster. The inside, southern face of this wall, was plastered with a thick layer of gray plaster, mixed with ribbed sherds, and over it a layer of waterproof, gray plaster mixed with crushed pottery. South of W7 was a floor foundation made of small flat fieldstones bound with gray plaster. On the foundation was a pile of collapsed stones, gray plaster and pottery sherds that probably fell from W7. On the foundation and in the collapse were fragments of pottery dating to the Byzantine period (fifth–sixth centuries CE).
Part of a wide wall (W3), adjacent to W2 of the earlier phase, and oriented northeast–southwest was exposed north of the pool remains. Wall 3 was built of small and medium fieldstones and roughly hewn stones, bound with gray plaster. It was set in a fine gravel fill, and was preserved to a height of one course. Southeast of W3 and parallel to it was another wall (W14), which was built of small and medium basalt stones, most of them fieldstones, and some roughly hewn. It was founded on buff-brown gravel. Part of the top of the wall slopes to the northwest and is coated with light gray plaster. It may have been the threshold of an opening in the wall. The northern part of W14 is not preserved, apparently it was robbed. It seems that W3 and W14 were part of an architectural unit associated with the pool.
In the southeastern part of the square, two walls (W5, W27) were aligned in a northwest–southeast direction . Wall 27 is a short, dry-construction wall, built of small and medium-sized limestones, some dressed, some fieldstones; it forms a corner with W14 (Fig. 7). Wall 5 has two faces made of small and medium-sized dressed basalt stones, with a core of small fieldstones; it continues to the southeast, beyond the boundaries of the excavation. Between Walls 5, 14 and 27 (Fig. 8), there is a floor foundation of small fieldstones, which abuts W5. It seems that these walls enclosed a paved architectural unit that was built parallel to the pool. The northwestern edge of W5 was constructed as a doorpost, parallel to the edge of W27. There was a narrow space between the two walls, and the travertine that was deposited there by flowing water indicates the this space functioned as a channel. It seems that the foundation of small fieldstones, served as a foundation for this channel as well. Fragments of a terra-cotta pipe near the channel to the northeast, may have been part of a gutter that was set between the walls and conveyed runoff water to the channel and on to the pool. Two adjacent basalt slabs were exposed slightly west of the corner formed by W14 and W27 (Fig. 9); travertine accumulated on these slabs and on the tops of the walls in the corner.
The area to the north of the pool’s wall (W7) seems to have served as a narrow passageway along the edge of the pool, which then continued southeast, to the area south of Walls 5 and 27. In the southeastern end of the passage was a floor foundation of small fieldstones, which abutted W5.
The pottery finds from this phase are from the Byzantine period, and include three bowls (Fig. 5:8–10) and a jar (Fig. 5:11).
The excavation shows that settlement at the site began during the Roman period, probably in the second century CE. Later, probably in the fourth century CE, a pool with associated buildings were constructed. The pool may have been a cistern that collected runoff water, or it may have been supplied by the aqueduct system of Scythopolis. It is also possible that the pool was used for bathing, and that the adjacent structures were dwellings. The pool seems to have remained in use until the end of the sixth century CE.