The excavation was conducted in the northern part of a lot slated for construction (c. 25×60 m), after no finds were discovered in trial trenches that were dug with the aid of mechanical equipment in the southern part of the lot. Remains of a building complex that was probably part of a residential quarter and remains of installations for conveying and storing water were uncovered in the northeastern part of the lot. In addition, four half squares were excavated in the northwestern part of the lot where floor remains (courtyards?), open channels and terra-cotta pipes were identified; it was not possible to connect these meager remains to those of the building. The remains in all the excavation squares dated to the Early Islamic period (late eighth or early ninth–mid-eleventh centuries CE). It was apparent in some of the trial trenches that were dug down to virgin soil that the remains were founded on a sand dune with patches of hamra soil. It seems that the original surface level gently sloped from the northeast to the southwest.
 
The Architectural Complex
Remains of a building section (a dwelling? Fig. 2) were identified. Most of the walls did not survive due to stone rubbery; nevertheless, for the most part, it was possible to reconstruct the plan of the rooms based on the location of the robber trenches (hereafter RT) and the floors. A sequence of building phases was identified in the complex, during which floors were raised, rooms were enlarged and rooms and installations were added or negated. The absence of walls made it difficult to identify the different building phases with certainty and it is evident that the processes of change that occurred in the complex were continuous, gradual and uneven. The structure in the main building phase consisted of several rooms that surrounded a courtyard, which contained water-related installations. After the building went out of use—a process that began gradually—a large open-air plaster surface was placed on top of the building remains.
All that survived of the northern wall (W138) that enclosed the northern row of rooms was the eastern end of a foundation, built of small stones set in mortar; the continuation of the wall to the west was robbed (RT355). The southern boundary of the northern row of rooms was marked by two side-by-side robber trenches (RT354, RT373) that are apparently indicative of a repair that was made to the wall. The division of rooms in this wing of the building is unclear; however, several sequences of plaster floors (A–C), which were separated by robber trenches, survived in it (Fig. 3). Fine quality floors were installed at some stage and the plaster was burnished to a pale shade of blue. A fragment of a stucco tile adorned with geometric decorations was found on one of the floors in Room A and had been damaged when a later pit was dug (L113; Fig. 4).
It seems that a narrow elongated space (D1, D2; corridor? storeroom?) extended in the western part of the building. Robber trenches (RT5/6, RT4/350) marked the course of the walls in this wing of the building. A single stone remained in situ (W4) in RT 4 and fragments of terra-cotta pipe sections were discovered. It appears that at some point this wing was expanded westward, as evidenced by RT15. Although no lateral walls were found in this area, these probably remained in the excavation balks.
A series of plaster floors (D1) was revealed in the northern part of the space and included a pit, in which a column was positioned (L350a), while a Kurkar pavement was found in the center of the space (D2), above which a series of plaster floors was identified.
Another room (E) might possibly be ascribed to the western wing. The eastern side of Room E was delineated by a wall (W94); of which only a short section of the core had survived, built of a row of partially hewn stones.
A sequence of plaster floors and installations was identified in the courtyard (F). It seems that initially the courtyard included the area of the spaces located to its east (G1–G3). A pipe of terra-cotta sections (P1) was set on the bottom floor (L217). The pipe was connected to a square debesh installation (L95; Fig. 5) that was constructed in the western part of the courtyard. At some point, Wall 117 that reduced the area of the courtyard and Installation 118, which leaned up against the wall, were built (Fig. 6); however, it is unclear if they were contemporary. Wall 117, built of two rows of fieldstones, was plastered on its eastern side that faced new spaces (rooms? installations?) built in the courtyard area (G1–G3). Installation 118 was built of large stones, some of which were roughly hewn. Pipe P1 was damaged when the bottom courses of Installation 118 were set, and possibly also W117 that was not removed. It is therefore clear that either the installation negated the use of the pipe or the pipe was already no longer in use prior to the construction of the installation.
Two large installations (H, I), used for some activity that required the containment of water, were built South of the courtyard; they were coated with pale red hydraulic plaster that clearly scaled the top of the installations’ walls.
The western installation (H) was surrounded by a bench or shallow plastered railing (W91, W92; height 0.5 m). The eastern installation (I; Fig. 7) was delineated by walls constructed from well-dressed building stones, with a row of small stones arranged alongside it (W127, W128, W359). An opening was set in W127; its doorjambs consisted of ashlars and the floor was plastered. Several levels of plaster floors were identified inside the installations. Plaster floors were identified south of the two installations; however, it was difficult to determine if these belonged to the building or just abutted its exterior.
Remains of installations, which were related to the conveyance and storage of water, were identified north and south of the building. Two installations and a plaster floor that extended between them were exposed in Sq E/F7. A small square well-built installation was exposed in the northwest of the square. It was built of debesh (L96; depth 1 m) and the inner faces of its walls were coated with plaster; a pipe (P2) was set in its northwestern corner. The bottom of the installation, which was dug into the ground, was not discerned. A section of a white industrial mosaic (L366) was preserved alongside the installation. A corner of another installation was exposed in the southeast of the square. Its floor (L326; Fig. 8) was composed of pale red hydraulic plaster (thickness c. 1 cm) and its walls (W391, W392; 0.5 m width) were built of small stones and gray mortar core that was lined with well-dressed stones; most of the stones lining the surface of the wall had been removed. A wall stump (W393) survived inside the room and several sections of a terra-cotta pipe were discovered. At some point, this region might have been incorporated into the structure because sections of walls and robber trenches discovered above the installation remains were aligned in correspondence to the building. 
The fragmentary remains south of the building were revealed in two narrow deep trenches; consequently they do not join up to form a clear plan. Several walls, aligned east–west and built side-by-side and on top of each other, were discovered in Squares F3/4; Wall 134 was the top wall. Water channels oriented north–south (L119, L135), were identified south of the walls; the channels were severed, probably by some of these walls. A water pipe, oriented east–west and set in a stone-built casing (P3; L120) was exposed. The continuation of the pipe (P4; Fig. 9), also enclosed within a stone-built casing, was discovered in Sq C3. In this square the pipe formed two 90° corners so that the westward continuation of the pipe followed a more northerly course. A covered water channel (L121), also aligned east–west, was exposed beneath the level of the pipe. A bone object of unclear purpose (see Fig. 13:5), was found inside the channel; the latter was leaning up against a wall (W230). Plaster floors and installations were discovered north and south of the channel and the wall: a section of a plastered square installation (L311), a bowl-shaped installation built of stones (L229) and a pit that was found filled with ash (L129); all of these related to one plaster floor (L124).
A sequence of plaster floors (L201 L208, L213) was discovered west of the building, in Sq B6; the two upper floors in the sequence were damaged by a later refuse pit. A wall (W215) aligned north–south was built on Floor 208. A large amount of ceramic finds that might have been the end of a debris heap belonging to a pottery workshop was found west of W215.
 
At some point, the building went out of use; its walls, or most of its walls, were gradually removed and a large open-air level of plaster mixed with chalk (thickness 2–5 cm) was placed on top of it. Only one floor was found in most of the squares, but in the northern and center part of the excavation area, as many as three superposed floors were discovered 1–5 cm above each other. The robber trenches that remained after the removal of the walls were filled with a thick layer of plaster (c. 5 cm) that was incorporated in the floor.
 
Pottery
Miriam Avissar
 
The pottery assemblage recovered from the excavation is characteristic of Ramla in the Early Islamic period (late eighth–eleventh centuries CE). Many of the vessels, particularly those made of buff-colored clay, were produced locally. Glazed vessels should also be included among the locally produced pots because there is no doubt that such vessels were manufactured in Ramla, as evidenced by the remains of glaze found on the ceramic rods that were used to place the vessels in the pottery kilns, which were discovered in many excavations in the city. Other vessels are imports and the origin of others is unknown.
 
Glazed Bowls (Fig. 10:1–9). Several glazed bowls (Fig. 10:1–3) were probably produced in Ramla. Delicate glazed bowls (Fig. 10:4, 5) were almost certainly not manufactured in the city but their origin is unknown. The bowl in Fig. 10:5 is decorated with incising and belongs to a type that first appeared probably not earlier than the tenth century CE (Northedge 2001:212–213). The celadon bowl (Fig. 10:6), made of hard gray clay (stoneware), is coated with a greenish gray glaze. Bowls of this kind were imported from China and first appeared in Israel only toward the end of the tenth century CE (Medley 1976:115). The glazed bowl in Fig. 10:7 was imported from Egypt; it belongs to Fayyumi Ware that dates from the ninth until the twelfth centuries CE (Scanlon 1993). Parallels of this bowl at Abu Mina were dated to the ninth century CE (Engemann 1989: Fig. 16; Pl. 7:d).
Lustre wares are represented by two specimens decorated with a monochrome lustre (Fig. 10:8, 9). The first (Fig. 10:8) is a fragment of a bowl base on which the background is covered with lustre, while the pattern—composed of strips and script separated by a floral design—is fashioned with a white zigzag. This fragment is dated to the late tenth and early eleventh centuries CE, based on bowls adorned with a similar decoration and signed by a potter named Moslim, who was active at that time (Jenkins 1968; Bahagat and Massoul 1930: Pls. XVII:2; XX:1; Philon 1980:166, Figs. 458, 459). The second (Fig. 10:9) is a bowl that belongs to a group of vessels decorated with patterns from the animal world, which dates to the late ninth and tenth centuries CE (Philon 1980:72–76).
 
Glazed Chamber Pots (Fig. 10:10, 11) were made of buff-colored clay and usually coated with a colored glazed that included a greenish yellowish background and a decoration of black and green patches; generally speaking, these vessels did not have a monochrome glaze.
 
Small and Plain Bowls (Fig. 11:1–12). Most of the small and plain bowls were locally produced and made of buff-colored clay (Fig. 11:1–8). Several Fine Byzantine Ware bowls (Fig. 11:9) and Egyptian Red Slip bowls (Fig. 11:10, 11) were found. In addition, there are large bowls (Fig. 11:12) that first appeared in the Late Byzantine period and continued to be used, almost unchanged, until the eleventh century CE.
 
Cooking Vessels (Fig. 11:13–18). Cooking pots with a neck (Fig. 11:13) and unglazed casseroles occurred in two sizes: large casseroles with very thin sides (Fig. 11:14) and small casseroles with fairly thick sides (Fig. 11:15). Partially glazed cooking vessels (Fig. 11:16–18) do not predate the mid-ninth century CE.
 
Jars (Fig. 12:1–4). Most of the jars in the excavation belong to types that are well-known at Ramla. The most common type in the Ramla assemblages (Fig. 12:1, 2) was produced in the Jerusalem region. Another type (Fig. 12:3) is a jar that is similar in form to the Jerusalem vessels but made of brown clay with mica inclusions, which suggests it was not of local manufacture. A jar with a gutter on the inside of the rim (Fig. 12:4) belongs to a type that is mostly found in the north of the country; these jars are usually adorned with a white painted decoration.
 
Jugs (Fig.12:5–14). Most of the jugs were prepared from buff-colored clay. Plain jugs (Fig. 12:5) are the most common type in the assemblage. Jugs with a filter positioned high up inside the neck (Fig. 12:6) usually date to the ninth–tenth centuries CE. This is also the date of a colander fragment (Fig. 12:7) that is decorated with incising and a puncture motif (Scanlon 1968). A mold-made jug adorned with palmettes (Fig. 12:9) and a handle decorated with split arabesques (Fig.12:8) does not predate the ninth century CE.
The jugs bearing pseudo-epigraphic incised decorations (Fig. 12:9–14), among them delicate jugs made of buff colored clay (Fig. 12:10), first occurred only toward the end of the eighth century CE and were in use until the eleventh century CE (Walmsley 2001:308–310).
 
Lamps (Fig. 13:1–3). It seems that all the lamps discovered at the site were made in Ramla, because lamps decorated with identical patterns were found in other excavations in the city.
 
Tobacco Pipe (Fig. 13:4). The sole artifact from the Ottoman period in the excavation is a fragment of a pipe that was found on the surface. The pipe dates to the eighteenth century CE.
 
Figurine (Fig. 14). A figurine in the image of a naked woman was found in the building. This is a hollow mold-made figurine fashioned from buff colored clay that was covered with green glaze. The head and legs are missing. One hand is spread beneath the chest while the other is missing. The genitals are highlighted by incising. The significance of the figurine is unclear because only a few figurines depicting naked women were found in excavations at Fustat. According to Scanlon they are late echoes of figurines of Isis and Aphrodite. All the figurines from Fustat were found in assemblages that dated no later than the year 800 CE (Scanlon 1968).
 
The Coins
Ariel Berman
 
Six coins were discovered in the excavation, three of which could not be identified. The other three, which were found on different floors in the building, are ascribed to the eighth–ninth centuries CE.
 
1. Basket 3182, Locus 380. IAA 92794.
Umayyad – post-reform, Damascus, eighth century CE.
Obverse: لااله الاالله وحده
Reverse: palm branch, . دمشق جائز
Æ, fals, 2.93 gr. 16 mm.
Cf. Walker 1956:249, No. 816.
 
2. Basket 2049, Locus 217. IAA 92793.
Umayyad – post-reform, Damascus, eighth century CE.
Obverse: worn.
Reverse: worn.
Æ, fals, 2.46 gr. 20 mm.
Uneven grooves were later scored in both sides of the coin.
 
3. Basket 1029, Locus 112. IAA 92792.
Abbasid, ninth century CE.
Obverse: illegible.
Reverse: illegible.
Æ, fals, 1.13 gr. 20 mm.
 
Summary
The ceramic and numismatic finds indicate that the structure, with the various building phases in it, and the adjacent installations existed from the late eighth until the beginning of the eleventh centuries CE. The remains in the excavation were recovered from a large residential building with an extensive courtyard, which yielded an elaborate system of irrigation, suggests that the inhabitants of the complex enjoyed economic prosperity. Despite the extensive robbery of the building stones, it was still possible to learn about the fine quality construction of the building’s walls; perhaps the robbery itself is indicative of the quality of the building stones utilized in the structure. 
It seems that the excavation area, located not far from the White Mosque and to its southeast, was situated in one of the important regions of the city in the Early Islamic period. This is also borne out by the findings of the excavations that were conducted nearby (see Fig. 1; Vilnay 1961:22–27; Peterson 1995). The absence of finds from the Crusader, Ayyubid and Mamluk periods shows that at the end of the Early Islamic period the center of the city shifted to the area that is today known as the “Old City”, where evidence of a residential quarter inhabited by a fairly affluent population has been uncovered (Luz 1996).