The Roman Temple, the Church and Buildings from the Umayyad Period—Areas 67.3 and 77.1 (Fig. 2)
In previous seasons, a Roman temple, founded in the first half of the second century CE, and a church, constructed above it in the late fifth or early sixth century CE, were discovered in Areas 67.3 and 77.1, in the insula south of the decamanus and east of the cardo. Remains of other buildings, revealed in several of the excavation squares, constituted an intermediate phase after the temple was abandoned in the mid-fourth century CE and prior to the establishment of the church, some 100 years later. The church was no longer used during the Umayyad period, at which time parts of its walls were dismantled and rooms, varying in size, were built in the southeastern corner of the atrium. A small settlement was exposed dating to the Abbasid period, after which time the site was abandoned.
Work during this season focused in two areas—the western and northern areas. Several squares were opened in each subarea to clarify questions of stratigraphy that arose during the previous seasons and to supplement new details relating mainly to the church and to later construction in the Umayyad and Abbasid periods.
In Area 67.3 West, one and one-half squares were excavated, balks were removed and the excavation continued in adjacent squares from previous seasons (Fig. 3). The area was situated west of the church’s narthex, in the southeastern corner of the atrium and next to the two rooms with geometric mosaics discovered in previous seasons.
Two main construction phases were discerned, complementing the picture of the two later layers recognized in this area. The eastern end of the atrium’s southern stoa, which was paved with plaster, was exposed. It ended at the line of a wall built of one course of stones oriented north–south, and it indicates the transition point to the eastern stoa of the atrium. Two in situ pedestals were found on the line of the wall that continued south, to the point where it adjoined the church’s closing wall. Another short space, paved with plaster, was found in the area delimited by the wall with the two pedestals, which served as a stylobate, and the two rooms with the geometric mosaics to its west.
The remains that were exposed make possible a reconstruction of the southeastern corner of the atrium and its adjacent rooms in the following manner: the northern pedestal was in line with the plastered doorjamb, the column drum and the threshold discovered during the previous season at the end of the eastern stoa, at the transition point to the room to its south. This line, with the addition of the southern pedestal exposed during the current season, formed a two-sided L-shaped stoa that demarcated the southeastern corner of the atrium. One side was at the southern end of the eastern stoa in the atrium, and the other side, perpendicular to the latter, was in front of the two rooms with geometric mosaics. In the area between the two sides was a square space—probably an open courtyard—situated between the two stoas, through which one passed to reach the row of rooms adjacent to the southern wall of the church.
The construction remains attributed to the last phase in this area were connected to the two rooms from the Umayyad period discovered in the previous season in the southeastern corner of the atrium. The later construction activity was accompanied by the dismantling of the church’s walls and the excavation of a foundation trench in several of its floors. This season exposed the western wall of the two Umayyad rooms, which was apparently built on the line of the eastern wall of the two rooms with the geometric mosaics that were in the church. This partially preserved wall was built of stones in secondary use, including architectural elements that had been taken from the church—a column fragment, a lotus-leaf-shaped capital and another capital decorated on either side with crosses. Moreover, the western continuation of the plaster floor in the southern room, exposed in previous seasons, was discovered; after its removal, three coins, all dating to the Umayyad period, were found in the fill below.
The excavation in the northern part (Area 77.1) focused on completely exposing two additional shops that continued the row of shops constructed during the intermediate phase south of the decamanus, probably in the fourth century CE, when the entrance to the temenos was blocked and direct access from the decamanus to the temple temenos was canceled. The shops are almost identical in size; their western, southern and eastern walls were preserved to a height of one course, as was a foundation course that sometimes protruded from the line of the wall. Except for small sections, the floor in the two shops was not preserved.
In a later phase, a circular installation was added, adjacent to the eastern wall of the easternmost shop excavated this season in Area 77.1. The installation (diam. 1.33 m, depth c. 0.58 m; Figs. 2, 4) was embedded in the shop’s floor. Its walls were made of smooth, thick, levigated plaster, and its floor was apparently made of lime-based plaster. Two small round openings were set opposite each other at the bottom of the installation’s wall. At this stage of the excavation, their function is unclear. Judging by the many plaster fragments discovered near the installation, it was probably partly built above the current floor level. A thick burnt layer containing numerous olive pits covered large parts of the eastern shop and reflects the last time the shop was used, probably in the Abbasid period.
 
The Eastern Cardo and its Adjacent Buildings—Area 68.2 (Fig. 5)
About five squares were opened to the west and mostly to the east of the eastern cardo, to study the nature of the construction in that part of the city and determine how far it extended to the east. The excavation in this area is of great significance, not just because of its contribution to information about other buildings in the city, but primarily because it may add important data concerning the boundaries of Zippori and the development of the city in the first centuries CE.
Another segment of the eastern cardo was exposed, specifically its eastern side, so that today, there is a complete east–west cross section of the street. Five plaster floors, one above the other, were discovered above the flagstone-paved Roman street. Trenches, excavated beneath several of the plaster floors, and an examination of the fills between the floors, revealed that the early level was laid during the Byzantine period and that this thoroughfare was no longer in use during the Umayyad period. This phenomenon is familiar in the northern part of the eastern cardo, as well as in other thoroughfares in the lower city. The finds discovered in the previous season in the western half of the eastern cardo and the finds exposed during this season seem to indicate that the later route, paved with layers of plaster, was c. 3.5 m wide, slightly more than half the width of the original thoroughfare.
Other spaces belonging to the building to the west of the cardo continued to be exposed (Fig. 5). The exposure of the courtyard paved with polygonal stones excavated last season was completed (Weiss 2016), and parts of two rooms with plastered floors were found to its north. The stone pavement in the courtyard was only partially preserved, whereas the walls of the rooms that were robbed at an unknown stage survived for the most part to a height of one course (Fig. 6). In a later phase, possibly after the courtyard ceased to be used, a wall aligned north–south was added; its architectural context is presently unclear. The building was destroyed by fire, signs of which were discovered in the courtyard, particularly in the eastern of the two rooms to the north of the courtyard. A variety of artifacts were found in the burnt layer, including more than 50 coins, probably wrapped together, or in a cache that scattered in the room following the destruction. An initial examination of the coins indicates that the building was destroyed and ceased to be used in the second half of the sixth century CE.
Remains of buildings from several phases were discovered in the square adjacent to the eastern side of the cardo, but their poor state of preservation did not present a clear archeological picture. Two perpendicular walls were exposed in the southern part of the square. To their south were the remains of a plastered installation, possibly a ritual bath (miqwe) that was mostly destroyed. Leveled bedrock c. 0.7 m higher than the street was discovered east of the western wall, which was c. 1.5 m from and parallel to the line of the cardo. Despite the differences in elevations, the dressed rock seems to reflect the level of the room that was built east of the street. At this stage of the excavation, it is difficult to determine what was in the space left between the cardo and the room to its east—an especially narrow room, a sidewalk or perhaps steps that were intended to bridge the difference in elevation between the two spaces. Several changes were made to the layout of the buildings later in the Byzantine and at the beginning of the Umayyad periods; this is evident in the installation of a plaster floor, and walls erected during a later phase, some of which seem to follow the lines of earlier walls. Large quantities of Umayyad pottery found above the level of the later room indicate it was no longer in use at about this time.
Architectural remains from the Byzantine and Umayyad periods were exposed in another square that was opened 6 m to the east (Fig. 6, the furthest square to the right). The ancient remains were not well preserved, and there were generally two clear layers of construction that could be observed above them. In the first phase, a thick plaster floor that covered the ancient remains was installed. Later, probably at the beginning of the Islamic period, two walls were built on top of them and new floors of similar quality were laid alongside them. The rooms were apparently destroyed during the Umayyad period. Based on the remains that were discovered this season in the area to the east of the cardo, settlement in that part of the city apparently commenced at the beginning of the Byzantine period and continued until the Umayyad period.
 
Two Monumental Buildings from the Roman Period—Area 78.2 (Fig. 1)
Two monumental buildings from the Roman period were exposed in the area north of the decamanus and east of the cardo: the earlier building, dating to the end of the first century–beginning of the second century CE, and the later building, still dating to the Roman period (Fig. 7). So far, two wings of the earlier building have been discovered. In the center of the excavation area, next to the decamanus, part of a large stone-paved courtyard was exposed and slightly to its west was a portico with a plaster floor. To the west, in the area extending from the portico to the southwestern corner of the building, next to the intersection of the cardo and decamanus, was another complex comprising several rooms, some of them decorated with mosaics. For reasons that are still unclear, the earlier building was dismantled and a monumental structure, larger than the previous one, was erected above it; sections of this building were exposed mainly in the western and northern parts of the excavation area. It is noteworthy that the new building’s foundations encompassed and buried parts of the earlier structure in a manner that has not been observed to date elsewhere in Zippori.
The work was concentrated in the western part of the excavation area; eight excavation squares were opened and nine balks were removed. The most prominent finds were several underground cavities with partially preserved barrel vaults that belonged to the earlier monumental building. Two underground cavities with barrel vaults were discovered, in addition to the four spaces that were found in previous seasons and further explored during this season. One space was located on the western side of the earlier building, parallel to the line of the cardo, while the other five were in the building’s northern wing. The spaces varied in size, orientation and method of construction. Some were oriented along a north–south axis and others, east–west (Fig. 8). The walls of the five cavities had been treated with thick gray plaster, evidence that they served as water cisterns. The sixth vault was wider than the others (width 3.6 m, reconstructed height c. 3 m); its walls were covered with light-colored, high-quality plaster and its floor was made of concrete. The method employed in the construction of this underground space, as well as its location in the northern wing of the building, which opened to the northern decamanus, clearly suggest that it was used for other purposes that are still unclear at this stage. Although most of the underground cavities were used for storing water, it seems that the impetus behind their construction was related to another reason. The monumental building was originally constructed along the slope, and the underground cavities to the west and the north were primarily meant to level the area and facilitate the construction of the superstructure along the same level as the decamanus to its south.
Two not particularly large rooms were discovered in the western wing of the building, north and south of the underground cavities. Their floors were made of poor-quality plaster and an opening (width 1.1 m) was fixed in their common wall. The level of the two rooms was different than that of the adjacent vaults to the north and south, and lower than the rooms with the mosaics, located in the southwestern corner of the building. It is unclear at this stage whether they were constructed at the same time as the earlier building, or were added later, when that building was still in use. To the west of the earlier building were the remains of a shop that opened onto the cardo, joining those shops that were exposed in the previous season. The floor in the store was made of plaster, and preserved in the southwestern corner was a tabun. Part of a plaster floor was discovered in the space between the storefront and the line of the cardo, along the line of the sidewalk that continued in front of the stores and parallel to the street.  
When the earlier building was abandoned and its walls were partly dismantled, its building materials—stones and plaster fragments, some of them colored—were buried beneath the floor levels of the second Roman-period building. One such cluster of construction debris, which was spilled diagonally, was discovered in the western part of the excavation area. It consisted of a concentration of hundreds of plaster fragments, probably from one or more rooms in the earlier building. The patterns on the plaster fragments are varied and decorated in many colors. Geometric patterns (guilloche) were found among them, as well as lines occurring in a variety of thicknesses that delineated a lustrous color surface, evidence that the walls were adorned with rectangular panels. Other fragments have light-colored floral motifs of white on a red background or motifs occurring in various shades set on a white background. Architectural elements, columns, niches, parts of a frieze etc. are depicted on other fragments. Of significance are the fragments on which figures are portrayed—the head of a lion, an antelope, a bird, the back of a tiger etc.—usually on a black background. The body of a person holding a club is portrayed on one fragment. The study of the plaster fragments is still in its initial stages but it is already clear that at least one of the building’s rooms was adorned with figurative images, possibly an exhibition of exotic animals and birds portrayed in a variety of poses.
During this last season, other sections were discovered of the second monumental structure built later in the Roman period. The continuation of the western wall to the north and the northern wall to the west were exposed. Another monumental wall (length c. 13 m, width c. 1.49 m), much of which survived to a height of three courses, was discovered 5.7 m east of, and parallel to, the western wall. The wall was built of well-dressed ashlar stones arranged with one face as headers and the other face as stretchers, like the rest of the building’s walls. It continued north, beyond the excavation area; its southern end was truncated and it may have continued southward. The new wall and the continuation of the monumental walls exposed during this season severed the vaults of the earlier building and canceled them, a clear indication that the construction of the later structures did forego the spaces of the earlier building, which was no longer in use.
No distinct floor levels belonging to the later building were discovered. Yet, in some places, earlier fills were revealed that had been placed between the walls of the later building. These were intended to level the area and served as a bedding for a floor that was not found, such as the fresco and plaster cluster that was spilled diagonally (above), discovered in the area between the two monumental walls in the western part of the building. It is possible that the floors were destroyed by later construction activity in the Byzantine period, but the lack of any distinct habitation levels in the later building and the absence of its monumental walls, which have so far not been discovered in the southern part of the excavation area, suggest that its construction may never have been completed. This possibility will be examined in the future, after the northern and eastern areas are excavated.
Remains of walls, plaster floors and several levels whose function could not be identified, all ascribed to the Byzantine period, were exposed throughout the area. Stratigraphically, it is clear that some of these walls and levels were added afterward, also in the Byzantine period, and there was probably more than one construction phase. In any event, at this stage of the excavation it is impossible to define the nature of the Byzantine construction and its phases or indicate a more precise date. A wall that continued east–west for several meters above the remains of the northern vaults stood out prominently among the later walls. Its western end terminated in a single straight line with the line of the eastern monumental wall discovered this season, indicating that at least part of the second Roman building was standing when the adjoining Byzantine wall was erected. Part of a water channel that descended toward the cardo, which ran from east to west, was preserved south of the Byzantine wall. The continuation of the channel to the west was fairly well preserved in the vicinity of ​​the Roman shop (above). This section of the channel continued in a straight line (length c. 4.75 m), and at a certain point curved south, below the line of the cardo. Apparently, the channel was intended to drain water from inside the Byzantine building, whose plan, purpose and date are not yet known.
Later activity in the area was related to the dismantling of stones and stacking them in piles, probably during the Abbasid period. Several walls identified as stone terraces and dark brown garden soil discovered here indicate that parts of the area were cultivated. The pottery sherds found in this stratum date it to the Mamluk period, and apparently the area was cultivated until the beginning of the twentieth century CE.
 
The finds from this season of excavation are another tier in the intensive work of the Zippori expedition, which in recent years has sought to clarify structural and stratigraphic issues of some of the monumental buildings constructed along the decamanus in the Roman period. The settlement picture that has been revealed and the exposure of several public buildings along the decamanus, provide important information regarding the city’s construction and development, and show that the decamanus, rather than the cardo, served as the main entrance route to the city during the Roman period and in Late Antiquity.