1. To do two large area excavations in the vicinity of the mosque, one west of the mosque, and one inside the mosque itself.
2. To refine the hypothetical map of the medieval city, showing more detailed variations in land usage and water resources around the city scape.

Topography/Geography

Tony Wilkinson undertook what he describes as a "brief field reconnaissance of the Sijilmasa Oasis" leading to what he also describes as "some preliminary notes on the region" and building on the work of Jim Miller and Dale Lightfoot. His field reconnaissance covered an area approximately 13 kilometers north to south and nine kilometers east to west. He divided the area examined into a series of "land systems" each of which exhibited distinctive characteristics recognizable on the ground or in air photographs, or both. The land systems that he identified summarize the basic character of the terrain within the area of his field reconnaissance, but time did not allow him to cover the entire oasis or to make a comprehensive statement about the presence or absence of outlying sites. Due to the unusually heavy rains during the first five days of October, enough surface soil eroded to expose archaeological features that we had not observed on the surface before. This was especially useful in doing an archaeological survey of the area known as Souk Ben Akla. Tony Wilkinson provided a useful sketch map of the Souk Ben Akla site. North of the tarmac road, his sketch shows two rectangular enclosures (low banks) and a range of low building mounds, extending over some four to five hectars. Within the enclosures, for example, were several aligned, circular features of "ghost bricks," i.e. mortar with the impressions of bricks now removed. These features must have been circular columns for a major structure. The site is bordered on the north by an eroded cemetery where some ten or more skeletons had been exposed as a result of gulley erosion leading down toward the Oued Gheris. South of the road he notes a roughly square mound, apparently fortified, extending over a total area of 90 meters E-W by 80 meters N-S, which could have served as a caravansary, or a fort.

The area excavation west of the mosque consisted initially of four adjacent 5 x 5 meter trenches (T25, T26, T27, and T28), leaving a two meter baulk between each trench. The entire area to be excavated, then, was 12 x 12 meters. Because of architectural features along the east baulk of the SE trench, we extended it another 5 meters to the east. It was recorded as a separate trench, T32. The latter exposed the corner of a concrete structure, which remained unidentified since the season came to an end before this trench was completed. Since this is an area excavation we will not report the results of each trench separately. This excavation revealed a bath complex with major canalization and what appears to be a hot water heater. Next to the bath is a public latrine.

The second area excavation T29 -T31 and T33-T39 are within the interior of the mosque beneath the surface floor. We found evidence of three earlier phases of mosque construction. We were not able to determine with any certainty the absolutely dates of these phases, but textual evidence suggests that the surface mosque was the renovation project of the Alaouite king Sidi Mohammed Ben Abd Allah in the late eighteenth century. The mosque floor found at thirty centimeters beneath that we believed to be the mosque of the Alaouite king Moulay Ismail in the late seventeenth century or early eighteenth century, although we have some conflicting textual evidence on this renovation. It uses the same floor plan and column bases, which we were now calling "Alouite column bases," as the surface mosque. There is a third level of mosque construction beneath the seventeenth century one. The column bases for this level are smaller than the Alaouite ones. These lower column bases were slightly different in the eastern half of the structure compared to the western half and were oriented slighty more to the south. This led us to believe that at this level, we had an original structure and an expansion, although we could not determine for sure which portion is the original and which portion is the expansion. We also suspected that this level dates to the Almoravid/Almohad dynasties. At this level of the mosque, we found the ties that supported the wooden rails for a movable minbar (pulpit), the only one that we know of to be found from that period in an archaeological context. That discovery confirms beyond the shadow of a doubt the identity of the mihrab discovered in the 1992 season, and hence the identity of the mosque. Beneath that level is still one, and in some spots two levels of occupation. At these lower levels, there is an abundance of cultural material of a culinary nature: hearths, ash pits, animal bones, cooking utensils, etc. In the last two days of excavation, we unearthed at the lowest level the edge of finely constructed lime-plaster floor and the corner of a stone foundation. We could not more specifically define either of these architectural features. More definitive answers would have to wait for the next season.

We did grid the Souk Ben Akla site at fifty meter intervals to do a systematic surface collection, collecting all artifacts from the 10 x 10 meter quadrant in the NE corner of each 50 x 50 meter quadrant. There was no Bhayr pottery at this site. Most of the pottery was of the Sijilmasain type, i.e. eleventh to fourteenth centuries. Most interestingly, the percentage of imported wares was higher here than in the central Sijilmasa site, suggesting that this was indeed some sort of staging area for caravans coming to and from Sijilmasa.

We now know that Sijilmasa was not a single walled city. Souk Ben Akla turns out to be a self contained urban entity, contemporary with and associated with Sijilmasa. The percentage of imported pottery at Souk Ben Akla exceeds that of Sijilmasa itself suggesting that it served as a staging area for the large caravans that crossed the Sahara. The central part of the city itself was not as large as we originally thought. It extended approximately one and a half kilometers from north to south with outlying occupation, probably individual villas that filled the elongated walled area to the north and south of the urban center. Finally, we know that various regimes that controlled Sijilmasa were intensively involved in water projects from perhaps as early as the Almoravids through the current Alaouite dynasty who trace their roots to Sijilmasa, and that those efforts, more than natural resources, affected the oasis's ability to feed an urban population and support long distance trade.

Our work no longer consists of looking for the original levels of the medieval city, but rather of defining some of the structures exposed in the most recent excavations. Kilns associated with the construction of dams must be excavated and dated. An airborne multi-spectral image will produce a more detailed layout of the city walls, streets, quarters, etc., and the Souk Ben Akla site deserves an excavation all its own. Sijilmasa, the mythical city of gold, has become demythicized