Four hundred squares were excavated in three areas (A–C) arranged in two parallel strips (length c. 1 km; Fig. 1): along the railroad track (width 10–20 m) and 10–20 m east of the track (width 15–20 m). Area A revealed three Early Islamic strata (an Umayyad stratum—AIII, with two phases: AIIIb, AIIIa; and two Abbasid strata—AII and AI). Area B yielded a cemetery consisting of infants buried in jars dating to the third century CE; the architectural remains and tombs were discovered in an excellent state of preservation. In Area C, a complex comprised of six pottery kilns surrounded by buildings, all dating to the seventh century CE, was unearthed (Yannai 2012).
A natural shallow depression in a layer of hamra between Areas A and C contained an alluvium deposit (Stratum AV; thickness 0.3 m) with numerous pottery sherds from the Middle Bronze Age IIB, Late Bronze Age and Iron Age IIA and B. The sherds were in an excellent state of preservation, not having been abraded; it thus seems that they were discarded in the area by the inhabitants of a nearby settlement rather than eroded in from afar.
Area A
The area was divided into four sub-areas (A1–A4; Fig. 2).
Sub-Area A1. Several wide walls (0.9 m) and floors, some paved with colorful mosaics bearing geometric patterns (Fig. 3) were exposed. The remains were partially destroyed and robbed.
Sub-Area A2. The walls of two buildings (Phase AIIIb), whose walls were erected without foundation trenches, were discovered on virgin soil. An unpaved rectangular cesspit with narrow walls was built in the corner of the northern building. It was roofed with a barrel vault with a square opening on its western side (Fig. 4). Three or four successive layers of fill covered by tamped-lime floors were found inside the rooms of Stratum AIIIb building (Phase AIIIa; total height of 0.6 m). On several of the floors were in-situ pottery sherds, most of them belonging to bag-shaped jars. In this phase, the two buildings were connected by an entrance with a threshold stone set in a stone wall (Fig. 5). The threshold indicates that either the two buildings were joined together or that the open area between them was closed into a courtyard. The vaulted installation from Stratum AIIIb continued to be used during this phase. Two circular pits that were dug into the ruins of Stratum AIIIa, and are thus of a later phase, went out of use prior to the construction of the next stratum.
Several of the Stratum AIIIa walls continued to be used in the next stratum (AII), whereas other walls were newly constructed. A new square opening was installed in the vault over the rectangular cesspit, right above the original opening. Its location suggests that the underground installation continued to be used in this stratum. A large pit was unearthed to the south of the building; it reached down to the level of the clay alluvium, below the foundations of the Phase AIIIb building. Several stone walls were constructed inside the pit. Their thickness and alignment seem to indicate that they belonged to installations rather than buildings; however, their nature and purpose remain unclear. A tabun was built in one of the phases during which these foundations were in use. An in-situ jar was partially embedded in the ground c. 1 m east of the tabun. The destruction of Stratum AII was extensive, but there was no evidence of a conflagration; some of the walls collapsed while others were completely uprooted, together with their foundations. The collapse may have been caused by an earthquake.
In Stratum AI, some of the earlier walls, from Stratum AIII, continued to be use, while several new walls were constructed. The orientation and general division of the area were maintained as in the earlier strata. One of the new walls was adjoined from the south by an installation built of several short walls with a large flat stone between them. This stone was located directly above the square opening of Stratum AII, and it seems to have been intended to cover it. Hence, the cesspit that was initially constructed in Stratum AIIIb continued to be used in Stratum AI, and the occupational sequence continued despite the collapse and destruction.
Sub-Area A3. A deep trench (length 60 m, width 20 m, depth 5–6 m) was excavated. The foundations of a building that was filled with refuse were found on virgin sand at the bottom of the trench. The foundations belonged to four rooms; their underground portion was exposed. No floors or parts of the buildings that were above ground were found. The foundations were overlain with layers of occupation debris, sloping from the northwest to the southeast. The debris dates from the Late Byzantine period to the beginning of the Early Islamic period. It included dozens of oil lamps, stoppers and parts of vessels that bear witness to the material culture of ancient Yavne. It contained also a large quantity of animal bones. These finds point to the livestock raised by the residents of Yavne and to their diet during the Byzantine and Early Islamic periods.
Area A4. Meager wall foundations were exposed in three strata. The earliest of these dates to the Byzantine or Early Islamic period, whereas the uppermost wall is ascribed to the Abbasid period. No remains of floors were found nor were any in-situ vessels discovered. Architectural remains from the British Mandate period were found on the surface level.
Area B – The Cemetery
The cemetery was located at the foot of the northern slope of a kurkar hill southeast of Tel Yavne. The tombs were excavated in two parallel rows of squares separated by an unexcavated row of squares. Three burial strata (BIV–BII; Fig. 6) were exposed.
Stratum BIV—Non-In-Situ Burial Remains. Hundreds of fragments of human bones and fragments of juglets, Eastern Terra Sigillata bowls and Athenian-imported black-slipped vessels were found in a layer of sand (thickness 0.4 m) on top of the kurkar bedrock. Such an assemblage could only represent funerary offerings in tombs from the Hellenistic period; however, neither in-situ tombs or tomb components, such as stones or bricks, were found in the layer of sand, nor any articulated skeletal parts. The fine preservation of the bones and sherds, which bore no signs of abrasion or erosion, indicates that they were not damaged by tomb construction in a later phase, because they were already mixed with sand when the pit graves of the next stratum (BIII) were dug. They were probably discarded here from a nearby cemetery, or belonged to an ancient cemetery that was plowed, thus destroying the burials beyond recognition.

Stratum BIII—Pit Graves. A total of 24 pit graves were exposed in the eastern row of squares. All of the graves were dug in a southwest–northeast axis and contained articulated skeletons. An examination of the sections indicates that the graves were c. 1 m deep. The complete shape of nine graves located at the edge of the excavation could not be determine. Of the graves that were excavated, six were demarcated with mud-bricks around the body of the deceased. Probes conducted in some of these graves revealed that the mud-bricks were placed in one row, one course high, indicating that they only marked the location of the deceased and did not line the grave. Six additional graves were lined and covered with stones of kurkar and limestone; their contents were not excavated. Three in-situ graves were found without stone linings but without funerary offerings, evidence that the origin of the vessels found in Stratum BIV was not in Stratum BIII.
Stratum BII—Jar Burials (Figs. 7, 8). Most of the exposed graves were infant burials in jars, generally consisting of two jars placed side by side. They were intentionally broken and their wide parts were placed together in the center. Most of the jars are of the same type and were locally produced, but several unusual jars and imported amphorae were also found. These date the infant burials to the third century CE. The bones were in an extremely poor state of preservation; most were crumbling, and it was difficult to identify the exact sex and age of the deceased. Anthropological examinations show that the infants were either newborns or several months old. Above the graves was a layer of alluvial sand mixed with abraded pottery sherds from the Byzantine period (Stratum BI).
In addition, a soft limestone ossuary of the type commonly found in the region between Jerusalem and Ben Shemen was discovered. It contained the gathered bones of a skeleton. The ossuary was placed in the sand, above the lined pit graves of Stratum BIII; above it was a jar burial of the type found in Stratum BII. Thus its stratigraphic relationship was determined even though it was not attributed to a separate layer.

Area C – The Kiln Complex and the Surrounding Buildings
Six kilns (1–6; Figs. 9, 10) arranged in two clusters of three kilns each were uncovered. They were joined by tunnels with a common entrance that led to both clusters. Only the upper part of the three kilns in the southern cluster was excavated. In the northern cluster, Kiln 3 was excavated almost in its entirety, whereas only half of Kiln 1 and only the upper part of Kiln 2 were unearthed. They were built according to two construction methods: Kilns 3, 4 and 5 were lined with stone, whereas Kilns 1, 2 and 6 had earthen linings.

The Kiln Structure. On the basis of the kiln that was completely excavated, we learn that the units were dug into virgin soil (depth 6 m). They included a lower firebox and an upper firing chamber. Inside the firebox were rows of columns (height 1.5 m) and a ceiling, all built of mud-bricks. The ceiling served as the floor of the firing chamber; it was built of rectangular bricks with spaces between them that allowed the heat to pass from the firebox into the firing chamber. The floor in the firebox was coated with five or six thin layers of mortar. In three of the kilns, the firing chamber was lined with four layers of mortar (total width 1.2 m). The outer layer was made of stones bonded with mortar mixed with a large quantity of lime or chalk which turned into lime as a result of the high temperature of the kiln. The inner layer was composed of six or seven layers of mortar that covered the walls and the floor. Built between the outer stone layer and the inner layers of mortar were two layers of insulation of equal thickness (0.3–0.4 m): an inner layer of mud-bricks and an outer one that included fragments of bricks and pottery sherds that formed a thick, strong wall, which sloped slightly outwards. The tops of the plastered stone-casing walls in three of the kilns were exposed in their entirety; they were preserved in perfect condition. The plaster indicates that the kiln did not have a superstructure that protruded above ground level. The outer layer of stone was intended to stabilize the wall of the kiln and prevent it from collapsing inward, whereas the other layers were designed to insulate the firing chamber and maintain the internal temperature. As previously mentioned, three of the kilns had no outer stone casing; their walls comprised only the three insulation layers. The kilns were built below ground level in order to withstand the heat and pressure that were generated in the firing chambers. The inner layers of the chambers were burnt and red, evidence of intensive combustion.

The Corridors Connecting the Kilns (Fig. 11). The three kilns in each cluster were connected by a cross-shaped corridor covered by a barrel vault. A fourth corridor was connected to the entrance shared by the northern and southern clusters. The walls of the common opening at the point where the two clusters connected indicate that the entrance to the corridors was from the east. The corridors led to the fireboxes and were not connected to the firing chambers. In the ceiling of each corridor were stone-lined windows, whose upper part was framed by means of four dressed stones. Grooves were found in the framing stones, indicating that the windows were closed with grills. Three upper windows were built above the intermediate compartments, where each corridor was connected to the firebox. This location is extremely important in reconstructing the processes used to operate the kilns. A fourth window was located above the intersection of the corridors, thereby allowing air and light to enter. A partition wall was built inside the corridor that led to each of the kilns, at a distance of c. 1.3 m from the wall of the firebox, thereby forming an intermediate compartment between the corridor’s interior and the firebox. The intermediate compartments were separated from the firebox by the layers of insulation described above. No ash or fired clay was found inside the intermediate compartments, evidence that they were not affected by the heat of the kiln. The layers of insulation and the intermediate compartments allowed for operating each kiln separately and while working adjacent kilns simultaneously.
Were the Kilns Roofed? The complete preservation of the outer casing, up to the top of the kilns, raises the question of how their upper part was built. Kilns previously excavated in Ashqelon were reconstructed with a mud-brick dome ceiling (Israel and Erickson-Gini 2013: Fig. 22). The reconstruction was based on the assumption that the upper part of the kiln was sealed with a dome in order to maintain the needed temperature within the firing chamber. However, the tops of the kilns at Yavne were complete; hence, they were left open, without a mud-brick dome. This reconstruction is based on the following data:
1) In the two firing chambers that were fully excavated and in a third chamber whose upper part was completely excavated, there was no evidence of a dome or of any other mud-brick ceiling that protruded above ground level, and no evidence of collapsed mud-bricks was found in them.
2) The construction of the firing chambers below ground and isolating them from the corridors made it possible to insert and remove the vessels slated for firing only through the kiln’s ceiling.
3) The construction of a mud-brick ceiling for a kiln that is 4–5 m in diameter necessitates the building of support columns in the firing chamber and their removal before it was filled with pottery vessels and operated. The vessels prepared for firing prevented any possibility of constructing such support columns. Nor was it possible to erect heavy massive scaffolding on top of the five or six layers of jars that filled the firing chamber.
The finds at Yavne indicate that in order to operate the open kiln, the upper layer of jars in the firing chamber was covered with a layer of large pottery sherds. This covering blocked the gaps between the jars but did not seal them. This layer was covered by a thick layer of ceramic chips, kiln waste and mud-bricks. The heat and pressure generated in the firing chamber were released through the layer of sherds, but it was sufficiently sealed to maintain a hot enough temperature in order to fire the pottery vessels. A kiln with such a large volume needed to be dug in the ground because of the pressure that would build up in the firing chamber. Therefore it was necessary to insert the fuel into the firebox via the corridor, whereas the vessels were inserted through the upper part of the firing chamber. Its wide span made it possible to rapidly insert a large quantity of vessels and remove them upon completion of the firing with many laborers. This operational advantage permitted mass-production within short periods of time.
The Surrounding Compound. Four colonnadedbuildings were constructed around the kiln complex. They enclosed the complex on all sides and were an integral part of it. They were only partially excavated because the northwestern part of the complex was situated below the railroad tracks, and the center of the southern building and large parts of the kilns were located beneath an oil pipeline.
The Northern Stone-Colonnaded Building: North of the kiln was a covered building that had a wall built on its northern side and a row of stone pillars open to the south, facing the kilns. To the north, along the wall, was a stone-lined water channel that was constructed together with the building and treated with a thick layer of plaster. The water flowed east, from the vicinity of Tel Yavne to nearby Nahal Soreq; it was presumably supplied by a well. The water conveyed in the channel was important for the preparation of the raw material and manufacture of the pottery vessels. In 2000, an excavation was conducted west of the railroad track, opposite the northern building and the channel, exposing the western continuation of the northern colonnaded building and the channel (
Sion 2005).
The Southern Wooden-Colonnaded Building: A long colonnaded building was revealed to the south of the kilns (Fig. 12). Its foundations were set atop a fill (thickness 0.5 m) of compacted limestone chips and earth. It seems that its outer walls were built of stones that had been robbed, as evidenced by two parallel robber trenches along the northern and southern sides of the structure. Its interior was divided into three long spaces by two parallel rows of wooden columns (distance between the column bases 2.6–2.8 m, distance between the rows of columns and the walls 2.5–2.6 m). Each column had a foundation (0.8 × 0.9 m) of stones and mortar that was placed inside a pit dug in the ground, and a square base on top of it (0.5 × 0.5 m) built of ashlars with a square cavity (0.2 × 0.2 m) in its center. The tops of the bases were level with the tamped-chalk floor. The shape of the cavities suggest that the wooden pillars were square; alternatively, it is possible that only their base was square and the rest of the pillar was round. The original length of the building could not be determined because it extended beyond the limits of the excavation area to the east and west. Seventy meters of the building’s length were exposed. This structure was almost parallel to the northern building, and it seems that both enclosed the kiln complex from the north and south.
The Eastern Wooden-Colonnaded Building: The eastern side of the compound was damaged and robbed. Only two bases for wooden pillars survived, identical to those in the southern building. They were perpendicular to the northern building, but were not perpendicular to the southern building.
The Western Stone-Colonnaded Building: A stone colonnaded building was exposed parallel to and west of the kilns, adjacent and perpendicular to the northern side of the southern building (Fig. 13). The floor of the building was made of crushed chalk and earth, like the floor of the southern building. This structure included two enclosing walls, at the western and eastern ends, constructed of mud-bricks set on a foundation of fieldstones (c. 0.5 m deep). The space between the two walls was partitioned into three areas by two rows of stone columns standing on stone bases buried in the floor. The bottom part of the bases was square (0.4 × 0.4 m), and their upper part was round (diam. 0.30–0.35 m). Each column was composed of two drums (diam. 0.3 m, height 1.2 m), and each column carried a stone capital. The dimensions and shape of the capitals resemble the bases. The columns and their capitals were made of coarsely dressed soft limestone and were left with a rough finish. The width of the building was the same as that of the southern building, and the distances between the column bases were identical in both buildings. The building was found destroyed; all of the stone columns were toppled on its floor, perpendicular to its western and eastern walls. The bottom, western part of each column lay beside a base, and the upper, eastern end was lay next to a capital. The manner in which they fell is indicative of a powerful earthquake.
These structures made it possible to reconstruct a trapezoidal complex that surrounded the kilns along the four cardinal directions of the compass. Between the northwestern kiln and the stone colonnaded building was a floor that abutted both. The stratigraphy seems to indicate that the built complex and the kilns were constructed at the same time, according to a single plan.
The Ceramic Artifacts in the Kilns. A group of bag-shaped jars was found in situ, arranged in three tiers, one atop the other, on the bottom of the firing chamber in the eastern kiln of the northern cluster. The jars in the bottom course were placed directly on the floor of the chamber, their openings facing down; the middle course of jars was placed with their openings facing up; the fragments of two other jars indicate a third level, and they too were placed with their opening facing down. Above the jars was a layer of large pottery sherds, some of which were burnt, and above them—a layer (thickness 0.4 m) of orange-pink-red colored sherds and ceramic chips. The two layers were meant to cover the jars during the firing process. The firing chamber of the northern kiln in the northern cluster, which was fully excavated, contained no in-situ jars. However, kiln debris and hundreds of Gaza store-jar rim sherds were found in it. A heap of debris, discovered to the north of the kiln, comprised tens of thousands of sherds of Gaza jars and several used lamps. All of the vessels found inside the firing chambers and in the debris heap, including the lamps, date to the sixth–seventh centuries CE.
The Operating Process. The water channel found north of the northern colonnaded building indicates that the raw material for the jars was prepared northeast of the kilns and beyond its surrounding buildings. Presumably, the raw material was derived from the streambed of nearby Nahal Soreq, into which the water from the drainage channel and the soaking pools was discharged. In an excavation conducted in 2007, c. 30 m west of the northern cluster of kilns, two courses of Gaza jars were found set on top of each other (Segal 2011). In the bottom course, the vessel openings were facing down, whereas in the upper course they faced up. The soil that was exposed in the vicinity of the jars indicates that no firing activity took place in this area. It thus seems that the finished vessels were stored there. A similar pile of Gaza jars was found in an excavation in Nahal Bohu in the Western Negev (Israel 1995b:106). It seems that the finished jars were treated in the western part of the kiln complex. The water channel and Nahal Soreq flowed from the east, so presumably the preparation of the raw material and manufacturing of the jars were done in the eastern part of the complex. The manufacturing of the jars was done from east to west: upon completion of the firing process, the jars suitable for marketing were placed in neatly arranged piles to the west of the kilns, and the wasters were discarded in the debris heap north of the kilns. Apart from the kilns for firing the jars, no evidence of the jar production’s preparatory stages was found in the excavations. Neither unfired vessel fragments nor any signs of pits for the potter’s feet and the wheel base or objects used in the manufacturing of the vessels were found. One hard limestone fragment of the lower part of a potter’s wheel was found in the heap of debris. Potter’s wheels of this type were discovered at several sites in the Western Negev where Gaza jars were produced (Israel and Erickson-Gini 2013: Fig. 40:5, 6). The western colonnaded building was built of fine quality materials; it presumably was the administrative building in the manufacturing complex, since mass production such as this required extensive management.
The Geographic Distribution of Kilns Used to Produce Gaza Jars. Approximately twenty heaps of pottery debris that were surveyed throughout the Western Negev, from Tel Ashdod in the north to Nahal Ha-Besor in the south (Israel 1995a), reflect the wide-scale manufacture of Gaza jars. Numerous studies have researched the ceramic industry in the sixth and seventh centuries CE. The large kilns at Yavne prove that this industry existed also in the Nahal Soreq basin and that this site played a central role in it. A heap of kiln debris was also found in the excavation at Kafr ‘Ana (Ono) in the Nahal Ayalon basin (Kogan-Zehavi 2011), which is the northernmost evidence in the country for the production of Gaza jars. Complexes consisting of several kilns were found in many excavations, but each kiln was built separately, without any architectural connection with the other kilns. Clusters such as those found at Yavne are widespread in the Persian Gulf (Lancaster and Lancaster2010). Private pottery workshops, in which there were one or two kilns, were found at ‘Aqaba (Melkawi, ‘Amr and Whitcomb 1994), Jerash (Pierobon 1986) and many other sites outside of Israel. In Israel, groups of pottery workshops were found at Bet She‘an (Bar-Nathan and Atrash 2011), but even in these, despite their proximity, the firing chambers, fireboxes and entrance corridor were separate from each other. Pairs of kilns with a common corridor were discovered at Ashqelon (Israel 1995a:126) and Gan Yavne (Gadot and Tepper 2003:133–141). Up to the discovery of the kilns at Gan Yavne and Yavne, researchers assumed that Gaza and Ashqelon jars were manufactured between Tel Ashdod in the north and Gaza in the south (Israel 1995b), based on the distribution of these types of jars as it was known until then (Mayerson 1992). Recent excavations have revealed that this industry spread from what is today the Gaza Strip to the Nahal Ayalon basin. The enormous amounts of jar debris, which created large heaps in the area (Israel 1995b), seem to indicate that thousands of jars were manufactured annually in this region. This large quantity might reflect a territory of economic activity that includes all of the northern Negev and the Shephelah in the east, extending as far as the Yarkon basin in the north.

Organizational and Economic Aspects of the Manufacture of Jars. The jar industry supplied numerous containers for oil, wine and other products. Gaza jars, Ashqelon jars and Yavne jars, all of which are fairly uniform in size and shape, were produced in the kilns. They were probably a trademark of the region that stretched from Kafr ‘Ana in the north to Rafah in the south. Gaza jars were found throughout the entire Mediterranean basin: Spain, France, Italy, North Africa, Egypt, Antioch and around the Black Sea. The kilns exposed at Yavne can explain the scope of the vessels’ extensive distribution. Presumably, the manufacture of standardized jars was not in private hands but was managed by a national or regional organization under the auspices of institutional bodies. The uniformity of the products, particularly the Gaza-Ashqelon jars, may show that the entire production process, from Gaza to Kafr ‘Ana, was part of a national economic enterprise that included the production of wine and oil and manufacture of the containers for their marketing. The dimensions and layout of the kiln complex at Yavne clearly reflect a national aspect of the pottery industry in Philistia during the Byzantine period. The construction of kilns as a national initiative is a phenomenon known in Land of Israel already in the Roman period; see, for example, the kiln complex exposed near Binyanei Ha-Uma in Jerusalem, which was ascribed to the Tenth Legion garrisoned there (Arubas and Goldfus 2005:15–17; 2008:17–18). Until the excavation at Yavne, no evidence was found of a complex comprising more than two kilns from the Byzantine period. Moreover, no evidence was found of a peripheral complex constructed in conjunction with kilns.
The Earthquake and Destruction of the Kilns at Yavne. The kiln complex at Yavne was demolished by a significant earthquake that caused the complete destruction of the colonnaded buildings and the subsequent cessation of the production of jars. Earthquakes are known from both historical sources and archaeological evidence. Bet She‘an and Tiberias were destroyed by a major earthquake in 749 CE. On the basis of historical evidence, a major earthquake occurred in the south of the country in June 659 CE (Guidoboni and Ebel 2009:357; Ambraseys 2009:221). This historical information is in keeping with the pollen data sampled from inside the kiln at Yavne (D. Langgut, internal report), and adds to other evidence for a major earthquake that radically changed the pollen samples at the beginning of the Early Islamic period (Neumann et al. 2009:47–50). It is difficult to estimate the intensity of the earthquake and which geographic regions it affected, but there was a very sharp drop in the export of Gaza jars that occurred over a short period of time. The production of Gaza-Ashqelon jars lasted a very long time; it started at the end of the Roman period and expanded greatly in the sixth century CE (Ashqelon jars) and the first half of the seventh century (Gaza jars=LRA 4). During the Byzantine period, when the jars were a very common industrial product, significant changes were made in the shape and volume of the vessels. Judging by the shape of the rim and the upper part of the body, only Gaza jars that were typical of the seventh century CE were fired in the Yavne kilns. At the same time that production at Yavne flourished, Eretz Israel was conquered by the Muslims. Immediately thereafter, a significant decline is apparent in the trade of these jars and the products stored in them. These jars, which in the fifth and sixth centuries CE were the overwhelming majority of jars in Constantinople, for example, disappeared in the seventh century CE from the repertoire of pottery imported to that city (Hayes 1992:65). A sharp decline in their numbers was also noted in Israel, Egypt (Kelliaand Ostrakina) and North Africa (Benghazi). It is difficult to determine with certainty what were the causes for this, and if there was a connection between the change in rulers over Palestine and the drastic decline in the trade of products stored in Gaza jars. On the basis of the finds from the Yavne excavation, we can assume that one of the possible reasons for the sharp decline in the amount of jars is the destruction of kilns on the coastal plain following the earthquake that struck in 659 CE.
The cemetery in Area B denotes significant changes in the funerary customs of the residents of Yavne. These changes are in keeping with the historical information regarding the shift from non-Jewish inhabitants to a Jewish population in the Hasmonean period. The latter population was replaced by a pagan population during the reign of Vespasian. It seems that these changes are reflected in transformations in burial practices. The practice of burial of children in jars is found in the region of Ashqelon, Gaza and Northern Sinai, and it is perhaps the most common type of children’s burial there. However, only a few jars were found in these excavations and no anthropological study was conducted. The many burial jars from Yavne may yield much information on this subject.
The built complexes exposed in Area A indicate an occupational sequence from the end of the Byzantine period until the end of the Early Islamic period. This sequence continued despite the destruction the settlement sustained, which was probably struck by two earthquakes. The buildings were exposed in a very limited area and it is difficult to draw conclusions regarding their plans and use. Fragments of magnificent capitals, marble chancel screens and remains of decorated mosaic floors attest to the wealth of the settlement and its economic power during the Byzantine period.
The peripheral compound and kilns in Area C constituted a large industrial complex for the manufacture of bag-shaped jars and Gaza jars dating to the Byzantine period and the beginning of the Early Islamic period. Both were built, operated and went out of use at the same time, the latter due to an earthquake. The northern kiln in the northern cluster cut through the foundations of two meager stone walls, the only evidence of building activity at the site during the Byzantine period prior to the construction of the kilns. The other parts of the complex were built on top of and into virgin hamra soil. This industrial complex was bounded on the south by the southern colonnaded building; its eastern and western sides were situated beyond the borders of the excavation. An open area extended north of the northern colonnaded building, where a heap of production debris was found, containing hundreds of thousands of Gaza jar sherds, including tens of thousands of rejected jar rims. This was not a refuse heap of a settlement of the kind found in Area A3 because it contained only jar fragments. It was placed on virgin soil, and other than Gaza jars and used lamps, no animal bones, remains of mud-bricks, tabuns or pottery sherds of other vessels were found in it. Without doubt, the pile contains debris from the kilns. The pile formed a high mound, visible even today, that altered the topography north of the industrial complex.