A full account of the Society's involvement with the emergency excavations at Sidi Khrebish, Benghazi, is contained in the Annual Reports for 1970–1 and 1971–2. In November 1972, at the invitation of the Libyan Department of Antiquities, the Society sent out a fresh team of archaeologists to begin the new season's excavations. Work has since continued uninterrupted until the time of writing (1 July 1973) and is due to finish at the end of September.

The preservation of the site from redevelopment for another year owes a great deal to the keen and active interest of the late Director-General of Antiquities, Mr Awad Sadawya, and the success of the expedition is greatly indebted to his efforts. We owe our thanks also to the officials of the Department of Antiquities whose sustained good will and co-operation has helped us greatly over this long period. In particular Mr Mohammed Nemri, Acting Director-General of Antiquities, Mr Abdulhamid Abdussaid, Controller of Antiquities for the Benghazi area, and Mr Ali Salem Letrik, Deputy Controller of Antiquities for Benghazi have taken full part in what has always been a team operation.

The disused Turkish cemetery of Sidi Khrebish lies close to the sea, a short distance to the north of the bustling commercial heart of modern Benghazi (see Fig. 1). It covers part of the north-western outskirts of the city of Berenice, founded in 247 B.C., the Hellenistic and Roman successor to the Hellenic city of Euesperides. The major part of the city lies to the south and to the east of the cemetery, under the mixture of Turkish, Italian and more recent buildings which form the modern town.

The quantity of stratified coarse pottery from Sidi Khrebish has been considerable and preliminary study has of necessity been concentrated on important groups from sealed contexts which span the period from the second century B.C. to the seventh century A.D.

The earliest group is from a cistern of second century B.C. date; the next is from excavation underneath Roman period concrete floors which produced mid-first century A.D. material, while the largest group is from the infill of Roman period cisterns and destruction levels and can be dated to the mid-third century A.D. Information is somewhat scanty for the fourth and fifth centuries A.D. but groups of pottery representing the Byzantine period were recovered in some quantity from the destruction levels of the church and its cistern. A good group of Islamic glazed fine ware and coarse ware was associated with late occupation within the church.

Space does not permit more than a brief survey of the most common and distinctive coarse ware forms from the excavation.

In general, throughout the period of Berenice, the locally made coarse ware form shapes seem to have been influenced from the Eastern Mediterranean, especially in the second and third centuries A.D.

The commonest form of second century B.C. cooking pot is rounded, having a short neck with two vertical ‘strap’ handles from the shoulder merging with the rim (fig. 1). Another distinguishing feature of pottery from this period is a semicircular handle from the body with an indentation at the top where it has been pressed to the rim (fig. 2). Both types are of the distinctive local ‘fossil gritted ware’, the fabric of which ranges from orange brown to dark pink. The clay contains fairly large roughly circular flat flakes of bluish-grey grit, which, when split open, reveal segmented spiral fossil remains. This fabric is very common in all periods.

In 1971 the Libyan Department of Antiquities invited the Society to conduct a trial excavation at Ajdabiyah to find out whether further investigation might throw light on the medieval Islamic settlement. Documentary evidence showed that Ajdabiyah was already a caravan town in the ninth century A.D., that it flourished under the Fatimids and that it was destroyed by the Banu Sulaim and the Banu Hilal in 1051. Earlier excavations had revealed two major Islamic monuments; a ruined qasr and a mosque (Abdussaid 1964a). It was known, too, that in the first century A.D. a Roman detachment was stationed at Ajdabiyah, which is identified with Corniclanum, a name on the Peutinger Map.

The short season of 1971 established that remains of the medieval town survived on the south-east side of Ajdabiyah, in the area of the modern cemetery (fig. 1). A scatter of Roman material and a group of burials, apparently of Roman date, suggested that future excavation might reveal something of the history of Corniclanum, as well as of the Fatimid settlement (Blake, Hutt and Whitehouse 1971).

In 1972 we returned to Ajdabiyah. We resumed excavation at the mosque and planned in detail the remains of the qasr (Whitehouse 1972). Both buildings have considerable importance. The qasr had been stripped in 1952 (Abdussaid 1964a). We planned the building at a scale of 1: 40, drew elevations of the principal rooms and made a full photographic record; the structure now awaits final publication. At the mosque we found evidence for two periods of construction, both evidently Fatimid, the earlier of which was probably commissioned by Abu'l Qasim, the son of Obeid Allah. An inscription from the mosque (unfortunately not in situ) bears the date (3)10/922-3 or (3)20/932, possibly the date of construction (Lowick 1972, 5 and pl. VIIa). Among the masonry re-used in the mosque were no fewer than seven fragments of Roman inscriptions, including SEG ix, 773–95.

Lepcis and Cyrene in Roman times were two very different cities, each typical of the province to which it belonged—Lepcis to Africa Proconsularis, one of the most successfully ‘Romanized’ provinces of the Roman Empire, and Cyrene to the province of Crete and Cyrene which always remained part of the Greek-speaking world and retained much of its Greek character and institutions. Architecturally, Lepcis today is one of the most impressively Roman cities, with its splendid public buildings, which include the grand Severan Forum and Basilica and the massive Hadrianic Thermae, and its spacious streets. Cyrene, in contrast, preserves an image that is severely Greek, dominated by austere Doric architecture. Even in quite late Roman times there seems to have been a determined effort to preserve the old forms, so that the Capitolium, for example, on the edge of the old Agora, was re-built under Antoninus Pius as a Greek building.

On the occasion of the Libya exhibition, I have tried to find a theme to bring together these two very diverse provinces that both form part of the modern Libyan Arab Republic. The choice of Septimius Severus, emperor from A.D. 193–211, needs some explanation. He was, of course, born at Lepcis probably in 146; he was there again as legatus of his uncle, the Proconsul of Africa; he may have gone there as Emperor in A.D. 203 during a grand tour of Africa and personally planned the great building programme which transformed his home-city. There is, on the other hand, no record of his ever having been at Cyrene, though he might have gone there as a boy. It is generally believed, indeed, that his first experience of the Greek-speaking empire was when he went to Syria to serve under Pertinax in A.D. 180. He does, however, provide a suitable link in that he was commemorated by sculptured monuments in both places, and the attitudes of mind which these monuments reflect, help us, I think to understand the underlying unity as well as the diversity of the two provinces.

The Garamantes were the inhabitants of Southern Libya. Their capital Garama (‘clarissimum … caput Garamantum’) lies partly under the now deserted mud-brick Arab town of Germa in the Wadi el Agial some hundred miles west of Sebha. The oldest pottery so far recovered from the site dates to the late fifth and fourth centuries B.C., but earlier occupation, stretching back to the ninth century B.C., has been found on the nearby fortified spur-site of Zinchecra. The surrounding escarpment slopes, which form the southern side of the wadi, are dotted with literally thousands of graves, which vary from simple crouch-burials in shallow cists, sometimes covered by a stone cairn, to elaborate stepped ‘chouchet’-type monuments of stone or mud brick, square or circular in shape, mud-brick pyramids 10–15 feet high and even ashlar-built mausolea of which the so-called ‘Germa mausoleum’ is the most complete surviving example. The sedentary agricultural practice of the people is attested by the many hundreds of foggaras (underground water channels with down-shafts, akin to the Qanats of Persia) which tap the aqueous strata against the escarpment side and carry their water into the wadi centre. The foggaras and burials, taken together, show that approximately eighty miles in length of the Wadi el Agial were intensely cultivated and inhabited, and recent work has shown that sites similar to Zinchecra and Garama existed at various points along this eighty-mile length, the Garamanticae Fauces, or Valley of the Garamantes.

Work in the Wadi Bergiug, Murzuch-Zuila area of oases to the south has shown that similar remains exist there, while it can reasonably be argued that the third great band of oases, the Wadi Chatti on the north of the el Agial, almost certainly contained similar occupation.