Stratum VI. Four pit graves, dug in the sand and aligned on an east–west axis, were ascribed to the earliest, Persian stratum. The dating was based on a Phoenician jar, typical of the Persian period (Fig. 3), which served as a funerary offering. A small bronze ladle was found inside the jar.
. The predominant stratum, dated to the Byzantine period (fourth–sixth centuries CE). Refuse from the storerooms of the Byzantine harbor (Figs. 4, 5) was exposed in Sqs 4–6. It contained a large quantity of sherds of imported jars and amphorae (see Haddad 2009
), which were apparently discarded after use. The discarded vessels were buried in a purpose-made, very wide ditch in the sand dune, which was oriented east–west. The southern boundary of the ditch was detected in the excavation, but an Ottoman disturbance that cut through the refuse concentration 10 m to the north obscured the northern boundary. An analysis of the inclination and positioning of the vessels shows that the discarded marine transport jars were deposited in the ditch from the north to south. Locally produced jars that are common to the Byzantine period include sandy-orange colored bag-shaped jars, Gaza jars and an Ashqelon jar. The imported types include LRA1 amphorae from the Gulf of Alexandretta in Turkey. Red painted Greek inscriptions were identified on four of them. The rims of jars from North Africa were also found. Preliminary sorting of the ceramic shows Gaza jars to be the most common type in the assemblage. A white-yellow organic material that was associated with them, was identified in the laboratories of the Weizmann Institute as a concentration of tiny fish bones and vertebrae.
Stratum IV. In Sqs 3 and 4 there is a yellowish kurkar fill with a moderate east–west slope. It was dated to the Crusader period on the basis of a sealed context of ceramics that included a large quantity of twelfth century imported Aegean Wares. The fill was probably an embankment that served as a boarding ramp, or a road that led from Birqat el-Qamar, the sandy bay south of Yafo port, toward the hill east of the excavation. This ramp/road was cut by the construction of a tabun, of a yet unknown date. Above this stratum there is a brown soil fill, and above it a fill that probably dates to the Late Ottoman period.
Stratum III. The finds from the Late Ottoman stratum include wall foundations oriented in an approximately north–south direction. In the northern part of the excavation area the wall makes several successive obtuse-angle turns to the northeast, forming a polygon (Fig. 6). The wall was a conglomerate of kurkar andreddish mortar. It was very poorly preserved in the southern part of the excavation. It was destroyed by the construction of the British-Mandate road and the building activities to its east in the 1930s. The fill on either side of the wall yielded numerous pottery sherds dating to the Ottoman period, and human bones from the Christian cemetery that surrounded St. George’s Church, and was damaged by the construction of the British Mandate road. The wall dates to the Late Ottoman–early Mandae period. Aerial photographs and maps show that this was the northern part of the western wall of the compound in which St. George’s Church was located. In an aerial photograph taken by the German air force on 24.1.1918, the church-compound and the western enclosure-wall are clearly visible. The southern part of the wall can still be seen today on the remains of the hill above the southeastern side of the street.
The excavation showed that the Giv‘at Andromeda Persian cemetery extended northwest and was damaged in the Byzantine period. The excavation uncovered for the first time in Yafo such a large quantity of Byzantine-period imported jars and amphorae which originally contained wine, oil and possibly fish sauce. The large quantity of small fish bones in the organic waste indicates that the common product in one of the jar types was garum, the fish sauce that was such a common delicacy in the Roman period. It seems that the rubbish was disposed of near a storeroom that could not be located. The proximity to the water—no more than 100 m—shows that the storerooms which housed the jars after they were unloaded from the boats, were situated nearby, and may come to light in future excavations.
The segment of the yellowish embankment was part of the road that led up from the beach in the Crusader period.
The western wall of the church compound was demolished several decades after its construction, when the British Mandate road was built and widened.