Area GB-North
A salvage excavation occasioned by damage caused to the mound during expansion of the Kinneret cemetery was conducted in April 2007, north of the fortified enceinte now dated to the Early Islamic period (Fig. 1). An 8 m long section-face was first cleared (Fig. 2), and a 30 sq m trench was excavated back from the face of the section. The sounding revealed a dense stratigraphic sequence, including (from top to bottom): Slope-wash with Islamic-period pottery; a Hellenistic (third–second centuries BCE) refuse pit; an Early Bronze III and II paved street, bordered by a long stretch of stone walling; a very deep series of alternating floors and fills, probably dating back to early EB I, and two superimposed refuse pits of the EB IA, belonging to the very first phase of settlement on the mound.
The street, which runs approximately north–south, was paved with a thick layer of packed gravel and sand obtained from the nearby lake-shore and topped with flat stones, of which only a few survived. One of ten similar segments of paved streets found in different parts of the mound over the years, it must have headed down the slope to a part of the mound that has disappeared into the lake. These streets – with their careful construction and orientation –indicate that there was an advanced level of municipal planning and regulation at Bet Yerah, at the very dawn of urban life.
The deep pits, cut into the soft marl bedrock, provided a rich pottery inventory that included gray-burnished bowls of the early type, a plain-ware fenestrated incense burner, and a new type of line-decorated pottery (Fig. 3). Considerable quantities of seeds and charcoal were retrieved and two radiocarbon samples from charcoal provided the following dates:
 
Sample No.
14C Age BP
Cal. BC (68.2% probability)
Cal. BC (95.4% probability)
d13C 0/00 PDB
RTT 5863
4780 +/- 45
3640 – 3620 (9.8%)
3610 – 3520 (58.4%)
3650 – 3500 (85.7%)
3430 – 3380 (9.7%)
-27.2
RTT 5864
4850 +/- 45
3700 – 3630 (55.2%)
3560 – 3530 (13.0%)
3720 – 3620 (64.7%)
3610 – 3520 (30.7%)
-24.3
 
These appear to establish the date of the earliest settlement at Tel Bet Yerah at the start of the Early Bronze Age, some time before the middle of the fourth millennium BCE.
 
Area SA-M
The main thrust of the 2007 season was a series of probes in the Circles Building (the Bet Yerah ‘granary’), continuing work begun in 2003 (see HA-ESI 117). In the east end of the southern platform, the stone floor of the surviving western half of Circle VI was removed with the aim of ascertaining whether there was an earlier phase of the structure. Material found immediately beneath the pavement included pottery that could be ascribed to early EB III. The foundations of the circle itself descended to a depth of five stone courses beneath the pavement. In the middle of the circle, however, there were remains of domestic structures cut by the circle foundations and hence predating the Circles Building (Fig. 4). Pottery associated with the domestic phase belonged to EB II. Thus, the building was set into a group of pre-existing houses. These were partially razed, their remains serving to stabilize the deep foundations of the presumed granary, which were built into carefully planned foundation trenches. A similar trench excavated in the 2009 season in the western platform (Circle II) revealed only a limited and poorly preserved EB II layer, overlying late EB I deposits that were also cut by the foundations.
In the northeastern part of the building, excavations through the floor of the late bathhouse were continued in the expected location of the eighth circle of the Circles Building and above the interior courtyard of the structure. While searching for the circle that was supposed to fill the northeast corner of the building, as reconstructed in many publications, we found instead an external eastern façade of the structure, fronted by an open area with what seemed to be an approach ramp. The façade was well-built and apparently niched (Fig. 5). Thus, instead of the nine circles shown in all previous reconstructions, the building had seven; instead of being enclosed and trapezoidal, the building now had a truncated U-shaped plan and it opened on to a large courtyard that sloped down toward the east (Fig. 6).
Within this courtyard, we were able to excavate untouched portions of its fine cobblestone floor. Parts of this pavement had been repaired in antiquity with pebbles and fragments of Khirbet Kerak Ware. These repairs clearly belonged to a late phase in the history of the structure. Our earlier research has shown that this late phase was characterized by artisanal activities carried out within the granary, after it had lost its original function.

In 2009, excavation focused on the stratigraphy of Circle VII, where the 2003 excavations had revealed two successive stone-paved floors. After dismantling a meter-wide segment of the late bathhouse wall in the northwest quadrant of the circle, which had massive foundations set in a deep trench, we were able to establish that the original slab floor of the circle sloped down sharply to the north. This necessitated the laying of a second stone floor, partly placed on fill, in order to make the space usable. Whether this repair belongs to the original construction phase of the building remains unclear; there is, however, no doubt that the last phase of use of the circle involved the joining of the east–west partitions by means of a narrow stone and mud-brick construction to form a single wall bisecting the circle.
North of the building, excavations were renewed in the presumed street that bordered this side of the structure, according to the earlier excavators. In order to retrieve a complete stratigraphic sequence, we excavated a portion of the last surviving unexcavated sector left standing between the Circles Building and the Guy–Bar-Adon excavations to its north. The topmost layer of this “island”, which is 1–2 m higher than the surrounding area, revealed badly leached deposits belonging to the Early Islamic strata and more substantial Hellenistic- period remains. Beneath those, excavations began to reveal a broad open area composed of soft lenses of ash and decomposed mud brick, alternating with harder earth surfaces. These deposits abutted a series of stair-like elements built against the north face of the Circle Building, which itself appears to have been carefully laid in alternating wide and narrow courses of stone (Fig. 7). Thus, rather than a street, we seem to be faced with a broad area of communal activity fronting a well-designed building façade. The contents of these deposits were characterized by heavy concentrations of Khirbet Kerak Ware in a very broad range of types and forms, including several that had not been previously encountered at the site (Fig. 8). Much further work remains to be done, however, before nature of this area is clarified.
 
Area GB-H
This area lies to the northeast of the Circles Building, within the confines of the late fortified structure. Here, beneath a series of Hellenistic structures excavated by Bar-Adon, we expected to recover further domestic or public areas contemporaneous with the Circles Building. Excavations were pursued both in the unexcavated “island” left by Bar-Adon (see Area SA-M, above), and in an area in which Hellenistic walls had already been removed. In the “island”, nearly sterile deposits associated with the late fortified structure covered some well-preserved portions of the Hellenistic structures that extended northward. These structures had at least two clear construction phases, and yielded a wealth of pottery (Fig. 9), including a considerable quantity of stamped imported wine amphorae.
In the eastern part of the excavation area part of an Early Bronze III structure was revealed, while to its west there was a series of midden-like deposits replete with ash lenses, animal bones, large quantities of Khirbet Kerak Ware and more than a dozen fragments of limestone maceheads. These deposits are substantially the same as those excavated in Area SA-M, so that – at this stage in the excavation – the presence of a large open area extending north of the Circles Building in the Khirbet Kerak Ware-rich phase of the EB III is indicated.
 
Area SA-S
On the western side of the Circles Building, on the far side of the paved street that borders it, excavations were renewed in a domestic area into which a sounding had been excavated in 1946. Our expectations were to recover the Early Bronze Age sequence from its very top, and especially to gain insight on activities coeval with the construction and initial use of the Circles Building. In 2007, excavations focused on the southern part of the area, revealing a nine-meter segment of the east–west street that borders the houses on the south, as well as several rooms that faced it. At least three episodes of resurfacing were identified in the street, providing a rich sample of refuse thrown out of the nearby houses.
Inside the houses, we began to uncover a sequence of rooms and courtyards belonging to domestic structures. Parts of a kitchen floor contained a fine basalt mortar and several broken jars.
In 2009, the entire space between Tower 12 of the latest fortified structure (see Area GB-T, below) and the street bordering the west side of the Circles Building was put under excavation. In the uppermost strata, remains of several phases in the two successive Early Islamic fortified structures were revealed, consisting of parts of the fortification that had been trenched by Bar-Adon, packed surfaces and pillar foundations associated with the later structure, and a well-built plastered water channel associated with the earlier structure. Beneath these were surfaces, middens and traces of walls belonging to the Hellenistic period.
The later activity appears to have removed part of the uppermost Period D (Early Bronze III) deposits in this area; those that survived – fragmentary walls and poorly preserved floors – indicate the presence of Khirbet Kerak Ware in rather small quantities and in restricted areas. This situation contrasts sharply with that revealed in Areas SA-M and GB-H. An earlier Period D phase can be connected to the architecture revealed in 2007 and 1946. The emerging plan is of a domestic complex composed of several contiguous structures bounded by the paved streets on the east and south: one occupies the southeast corner, another, partly excavated unit extends to its west, and a third to its north (Fig. 10). Excavations north of the domestic complex revealed a packed gravel pavement with a pronounced slope toward the east. The pavement is bounded by walls north and south, creating a space (width c. 4 m) that extends westward beyond the excavated area. It evidently approached the nearby paved street on the east, but erosion has removed a significant portion of it. This erosion exposed earlier remains—an east–west wall and related deposits – firmly dated to EB II.
Among the local ceramics and Khirbet Kerak Ware found upon the pavement, there was a fragment of an Egyptian ceremonial palette. The fragment – the first of its kind ever discovered outside Egypt – shows part of a well-executed scene depicting an arm and hand grasping an archaic, double-tailed ‘ankh and a was scepter (Fig. 11). It may be ascribed a late predynastic or Dynasty I date. As the fragment was clearly not in its original context when found, it is likely that it originated in the Period C (Early Bronze II) structures, parts of which were excavated nearby. 
 
Area GB-T
A new aspect of the 2009 excavations is our attempt to re-excavate and reinterpret the huge fortified complex cleared by Bar-Adon and Guy in the early 1950s but never fully published. Originally identified as a synagogue and then as a Roman or Byzantine fort, the most recent suggestion has been to identify the complex with the Early Islamic palace of al-Sinnabra. The area has been obscured for decades by the thick subtropical vegetation that characterizes the mound. Because the structure was largely dismantled in antiquity, leaving only wall and floor foundations intact, and due to the summary excavation methods used in the original excavations, our principal aim was to identify sealed or otherwise datable contexts, such as foundation trenches and subfloor deposits. Additionally, the surviving portions of the superstructure had to be revisited and recorded.
Thus far, the southwest tower of the enceinte and parts of the southern annex adjoining the large apse have been reinvestigated, two large mosaic floor segments recorded (Fig. 12), and a portion of the central floor-bed removed. Some preliminary observations may be made:
(1) The original wall foundations of the external fortifications, the adjoining bathhouse, and the central structure are all equally massive and deep, indicating a high level of investment, similar building concepts in all parts of the complex, and the likelihood that the superstructure was quite substantial.
(2) There is multiple evidence for the existence of at least two building phases in the main structure. The later phase involved wall demolition and replacement, as well as repairs in the mosaic pavements.
(3) We have begun to see evidence of earthquake damage; this could eventually aid in the dating of the structure (Fig. 13).
(4) A second fortified enclosure was built over part of the main enclosure. This later enclosure was never reported by the excavators.
(5) Although finds are sparse, the Early Islamic dating does appear to be confirmed by coins found beneath the floor of the central hall and in the foundation trench of the bathhouse.
We are therefore confident that the Umayyad palace of al-Sinnabra has been found.