A small limestone half-figure from Alan Rowe's 1957 excavations in Cyrene was recently found to have the word ΘEA lightly inscribed on the polos. The authors describe the sarcophagi with which this figure may have been associated, together with the burial artifacts found within them. It is rare to find ΘEA used in this way, although two examples from Eleusis provide useful parallels. One is a votive plaque, the other a large relief, and each portrays the Eleusinian deities whose iconography is discussed here and compared with that of the Cyrene ΘEA example and also other uninscribed Cyrene half-figures.

The neuter substantive centenarium appears in a series of documents to designate a particular type of building. The etymology and precise meaning of this word have long been debated by scholars, who have proposed a wide range of explanations, none of which appears to be entirely satisfactory. In this paper, we put forward a different solution, taking into account textual, archaeological and linguistic evidence. In our opinion, the word centenarium comes from centenum, which means a kind of a cereal; thus, centenarium indicates a ‘fortified grain-house’. It seems probable that original centenaria were military structures of varying shape and size, first created in the third century and then spreading in the Tetrarchic period. They appear to be distinctive of the African provinces, from which comes most of the evidence, but one cannot exclude that the same name was applied to similar structures in other parts of the Empire (such as in the Iberian Peninsula, for instance). The military centenaria were later imitated by private landlords – maybe even indigenous chieftains in charge of the defence of sectors of the frontier – who transferred the name to their unofficial or para-official defensive structures. This process of emulation was already advanced in Constantinian age.

Among the more striking finds made during the Society's excavations at Sidi Khrebish is the upper part of an inscribed marble stele which had been re-used, unfortunately with the inscribed face upwards, as a paving stone in the basilica. It is of white marble, probably from the Greek islands, slightly tapered towards the top, measuring 0·46 - 0·47 m. in width × 0·97 in height and 0·19 in depth. A very simple moulding runs round all four sides some 15 cms. from the top. It is chipped along all edges, badly worn on the face, especially in the central area, and pitted and cracked as well, especially in the lower part, probably by an attempt to break it up. A section of unknown size has in fact been hacked off at the bottom.

The text is often hard to read and I am sure that more could be extracted from the worn areas with time and patience; but it seems desirable to offer a provisional publication as quickly as possible.

In the summer of 1989 a new electricity cable was laid from New Shahat to the Caesareum of Cyrene, its course following the line of the road made in the 1920s through the (now largely abandoned) village of old Shahat. The construction of the road had already revealed several blocks of an engaged Doric architectural order in the unexcavated southeastern quarter of the ancient city; from the trench cut in 1989 were recovered parts of eight marble sculptures susceptible of interpretation, less identifiable fragments of other figures in marble and limestone, and a number of ashlar blocks with drafted margins cut in local limestone. Though the architectural finds suggest the presence of a substantial public building, the sculptures were apparently found packed in a row, giving the impression of having been deliberately buried during preparation for transit to a lime kiln. Without excavation no more can be said of their original context, beyond what may be gleaned from an examination of the sculptures themselves.

This paper is concerned with six of the more complete marble statues, which form an interesting, if not aesthetically distinguished group of portraits of members of the first Roman imperial family. The six survivors fall into three sub-groups, which are briefly described below; there follows an assessment of the relationship of the statues to contemporary portraits of the imperial family from Cyrene and elsewhere.

Two of the statues are carved in Pentelic marble, and are of exceptional scale. Of these the better preserved is a figure of a woman originally cut from one block to stand about 2.10 metres high but now sliced from top to base through the thicker part of the lower torso and horizontally through the whole torso into pieces 20–30 cm high, presumably to facilitate delivery, perhaps by hand, to the lime kiln (Figs 1–2). The unveiled woman is dressed in tunic and cloak, the latter swept over her left shoulder in the manner of the Kore of Praxiteles, a type well known in the Cyrenaican repertoire. The cloak is drawn far from the right side of her body, suggesting that her right forearm, now lost, was raised to hold an object (Fig. 3, right). The hair is waved to either side of a central parting, and drawn back from the face in thick tresses which are massed wide of the head to fall into a looped plait below the nape of the neck and over the edge of the cloak (Fig. 4). Around the face is a row of pin-wheel curls; there is no break in the curls at the central parting of the hair, and the curls come close to the ends of the eyebrows. The lower earlobes are exposed. The face is broad and round, with a low brow accentuated by the hairstyle, plump cheeks and a small receding chin. The nose and mouth are badly damaged, but the mouth appears small. The eyes are large, round and wide-set (Fig. 5).

The use of pottery distributions to identify trade contacts is one of the most important tools in the archaeologist's repertoire. Recent examples of such studies, concentrating on the classical period in Cyrenaica, have produced evidence which provides valuable new insights into changing trade patterns for which there is precious little historical information. There is, at present, a debate over Cyrenaica's trading contacts during the classical period in which Fulford has proposed a definite bias towards trade with the eastern Mediterranean (Fulford 1989) whilst Lloyd has challenged the lack of emphasis given to Italian imports (Lloyd forthcoming). Abulafia's reassessment of the Pisan bacini has likewise stimulated debate on the development of trade contacts, this time in the medieval period (Abulafia 1985). Such studies rely upon two points; the first is a well-documented corpus of material and the second is a detailed background knowledge of the wares and fabrics in general circulation to which the corpus can be related. On both points medieval archaeologists are clearly still at a disadvantage to those working on the classical period, and an attempt to understand Cyrenaica's medieval trade contacts from its ceramic assemblages cannot therefore achieve the same detail as, for example, Fulford's work (1989) has done. Nonetheless, an examination of the material which is available, and a discussion of its significance, is the intention of this paper.

Since 1989, the Society for Libyan Studies has been carrying out, in conjunction with the Libyan Department of Antiquities, excavation and fieldwork in El Merj (Dore 1990, 1991, 1992) which is held to be the site of two earlier towns named Barqa (also spelled Barca and Barka), one of medieval date and the other Greek, as also of a Roman-period village. In this paper I wish to review critically the evidence for the identification of El Merj as medieval Barqa.

The broad outline of the case is as follows: a town called Barqa is mentioned by a considerable number of medieval authors writing in Arabic. To judge from them the town flourished between the ninth and eleventh centuries AD but declined thereafter. The association of the names Barqa and El Merj with a single site seems to stem from one author, Ibn Sa 'id, writing in the thirteenth century, though even he is tentative in his identification (see below). After the fourteenth century there is a period which is devoid of information. By the eighteenth century the town(s) of Barqa/El Merj had disappeared (i.e. ceased to be inhabited) but local memory preserved the name and location of El Merj because Pacho visited its ruins and recorded the name in 1825 (see below). About twenty years after this a new town called El Merj began to grow up around a castle newly built by the Ottoman authorities on the remains of an earlier town. This town was called Barce by the Italians but reverted to being called El Merj after the second world war, and was finally destroyed by an earthquake in 1963.

The Qasr Al Haj Formation, comprised of alluvial fan gravels and sands with calcrete horizons, is subdivided into five members separated one from the other by surfaces of discontinuity. These surfaces, defined primarily by erosion and soil formation, can be followed northwards into the Jifarah Formation thus allowing detailed correlation of the fan-gravel deposits with the finer fluvio-aeolian sands and silts of the floodplain. Thick deposits of cemented and uncemented gravels at Wadi Ghan contain at least two horizons of basaltic lava flows. Previously reported finds of artifacts from this gravel sequence together with the possibility of radiometric dating of the lavas, may provide a much-needed chronostratigraphic framework for the Quaternary deposits of the north-west Tripolitanian region.

In winter 1987–8 Mr Ramadan Kwaider of the Department of Antiquities at Cyrene found a marble block, inscribed on three faces, in die lower levels of the Extramural Sanctuary of Demeter and Kore in the Wadi Belgadir; it had been uncovered by a combination of burrowing foxes and winter rains, and is now, by his prompt action, safe in Cyrene Museum (inv. no. 3480). Our preliminary report on it is awaiting publication in Libya Antiqua; we offer here the three texts, with discussion of some of their points of interest.

The history of the block, which was three times re-used, is a vivid reminder of the value of marble at Cyrene, all of it imported and therefore very rarely to be discarded when out of date or damaged, if it could be made to serve another turn. Its findspot solves an uncertainty about the attribution of some other inscriptions to the Wadi Belgadir Sanctuary. More significantly still, it provides a new and suggestive document relating to Roman taxation in Cyrenaica.

Only the third and final inscription on the block (Fig. 1) can be said with certainty to belong to the sanctuary. This was cut on a face which measures w. 0.97 m × ht. 0.35 m × d. 0.23 m and was dressed with a claw chisel, but not polished. The letters (ht. average 0.04 m) were lightly cut, rather narrow for their height, in a style dateable approximately to the Hadrianic and early Antonine periods; but although the layout, with quite careful centering, is respectable, the cutting is light so that the letters would only be easily legible if over-painted. The top surface, when the block is in this position, has been hollowed out, presumably for the insertion of the base carrying the statue implied by the text.

The female funerary bust, as a type of grave monument, consists of a bust or a demi-statue including hips, waist, shoulders and head. The type is unparalleled elsewhere in the history of ancient sculpture.

1. This unique unpublished find (pl. 1) of the Flavian period, which is made of grained white marble and measures 50 cm. high by 34 cm. wide, was found at El-Beggara in the S. W. Cemetery at Cyrene by Mr. Ali Bu Nuw-wara on 4 November 1971. The statue portrays a draped woman in half-length bust form. The bust terminates a short way below the waist and is worked flat at the bottom. The figure is in good condition; only three fingers of the left hand are missing and some drapery ridges are chipped or broken.

The figure holds her right arm, bent at the elbow, horizontally across her body; the left arm is bent up to the side of the face and holds part of her garment across the face as a veil. The figure is completely draped in a himation which both enfolds the body and covers the head. The left hand, with thumb, ring and little fingers missing, stretches a part of the himation across the eyes, nose and mouth. These features are indicated under the transparent folds of the mantle. Here the sculptor has precisely demonstrated the delicacy of the facial features through the veil, an effect produced by his controlled chisel work. The figure's forehead is graced with tightly curled hair which is massed over the brow like a short African haircut. The hair style is typical of the Flavian period and is not later than the time of Hadrian. The folds of the himation in front are rendered as sharp ridges crossing the body; the back is also carved with folds of the same type but worked in a more simplified manner. The figure wears a snake bracelet on her left arm; this type of armlet appears in many of these Cyrenean funerary busts and may have iconographic or symbolic significance.

A small Christian monument of centralised plan built against the West Wall of Apollonia was excavated and slightly restored by the Libyan Department of Antiquities under the directorship of the late Professor Goodchild during the sixties. In 1967 the present writer, then architect to the Michigan Apollonia Expedition, was commissioned to make a set of drawings of the monument as a basis for its publication by Professor Goodchild to appear in the Michigan Report. Professor Goodchild's untimely death in 1968 nullified the project and eventually in its stead only a cursory notice of the monument appeared. The substantive treatment of the monument, utilising the prepared set of drawings, was reserved for the long laboured Corpus of Christian Antiquities of Cyrenaica. Unfortunately in turn this design was frustrated by the death of Professor Ward Perkins in 1986 so that the detailed drawings of the monument remained unpublished for 25 years. In 1991 old prints of the drawings were recovered and are published here with a commentary.

This small square monument with a dome on four pillars giving a rudimentary cross-in-square plan is of late sixth century date. It is clearly sepulchral, whether it be a simple tomb or a martyrion. Thus in spite of its provincial guise it is of interest and significance (together with the Church at Qasr el Lebia) in the long vexed question of the origins of the ecumenical Byzantine cross-in-square plan.

Cyrenaica, excluding the Jebel Akhdar, forms part of the Libyan Sahara. It is a true desert with a mean annual rainfall generally less than 20 mm, and a land surface of sand, gravel or rock, largely devoid of vegetation (Plate 1). Yet a few thousand years ago, the Libyan Sahara, as elsewhere in North Africa, enjoyed an abundant rainfall and supported a varied flora and fauna as can be seen from pre-historic rock paintings. The abundant rainfall filled huge subterranean reservoirs which since then have been slowly draining to the Mediterranean Sea or low lying sabkhats and inland lakes. In consequence of the moderate relief of the Sahara and the wide regional extent and uniform structure of the sedimentary basins which form the reservoirs, the groundwater bodies occur in huge flow systems which may extend over several hundreds of miles. The reservoirs' thicknesses vary but the two most important series, the Continental Intercalaire or Nubian, and the Continental Terminal of mid-Tertiary to recent age, frequently exceed 3000 feet. Two major flow systems occur in Cyrenaica, in the Sirte and Kufra basins (Fig.1).

Until oil was discovered in the late nineteen fifties, groundwater development in Cyrenaica was mainly concentrated in coastal areas where the prevailing rainfall is sufficient to permit some recharge, and in a few interior scattered oases — Giarabub, Jalo-Augila, Kufra and Tazerbo, where the ground surface lies close to the upper phreatic (water table) surface of regional groundwater systems.

Oil exploration requires fairly large supplies of water for the main drilling operations. During exploration in the Sirte basin it was discovered that water of variable quality could be found virtually anywhere at moderate depths between 100 and 500 feet below ground level. Oil company practice to obtain temporary water supplies is to drill two or three simply constructed water wells to a few hundred feet below the water table, and to pump by air lift. The wells are subsequently capped, when no longer required.

The conventional view is that Malta has been on the ‘forgotten frontier’ of Christian maritime resistance to Islamic expansionism since the Islamic invasions of North Africa in the seventh century. The limited archival and archeological evidence suggests that, up to the arrival of the Order of the Knights of St John of Jerusalem in Malta in 1530, this picture is not accurate. The Islamic occupation of the Maltese archipelago in 870 created a cosmopolitan Muslim society which persisted until the mid-thirteenth century, despite the Norman conquest of the region in 1090. Indeed, the formal end of Muslim society in Malta only came in 1224, as a side-result of the Hohenstauffen suppression of a Muslim rebellion in Sicily.

Even under the Order of St John contacts with the Muslim world were far closer than is conventionally supposed. The Grand Master of the Order maintained close contacts with the Qaramanlis in Tripoli and the Beys of Tunis during the eighteenth century, despite the continuation of the corso. In reality, contacts had always existed and had been recognised as essential by the Holy See because Malta could not sustain its population once it had exceeded 10,000 persons. Sicily, the obvious source of supply, often exerted undesirable political pressure and the Barbary coast was the only other alternative. The main legacy of the close contacts between Malta and the North African Muslim world, however, is to be found, even today, in the Maltese language, which is really a Medieval variant of North African Arabic.

A detailed study of the coarse pottery of Cyrenaica, and particularly Berenice, in the Hellenistic and Roman periods (Riley, in press) indicates that, with the exception of the total disappearance of locally made amphoras, the Romanisation of Cyrenaica made no impact on the local pottery industry until around the middle of the second century A.D. Indeed, at a time of rapidly increasing prosperity in Cyrenaica from the Augustan period, the local pottery industry continued to produce poor quality goods in general. These were generally of the poor, shell rich fabrics (for these and Cyrenaican fabrics in general, see Riley, 1979), and comprised a wide range of variation within broad type groupings, which were of the characteristic, elegantly curved Hellenistic tradition.

When change came, around the middle of the second century, it was sweeping. The poorer quality shell rich fabric was superceded by the harder, sturdier, lime rich fabrics, and a completely new repertoire of angular and corrugated forms was introduced. This development appears to follow a similar pattern in the Aegean region during the early part of the second century, so that by the latter part of that century there is a striking typological similarity between the various coarse wares in Greece, Crete and Cyrenaica.

The price inscription on an Attic black-glossed lekanis is discussed. The lekanis was found during the excavations of one of the houses in the Greek colony of Euesperides. Its significance is considered alongside the small number of price inscriptions known from Cyrenaica. Price inscriptions draw attention to the low value of Attic pottery in antiquity, and the Euesperides graffito is considered against some of the literary and epigraphic evidence used in recent discussions.

The Libyan Department of Antiquities has begun to investigate the many chamber tombs cut into the rock scarps surrounding the oasis of Jeghbub (=Jarabub/Gerabub). Some notice of this work was given at the Congress of Cyrenaican Studies in Rome (1996) by the Director of the Department. This communication called to mind a visit to Jeghbub made more than forty years ago. The visit afforded occasion for a brief investigation of these tombs, which because of a mischance has hitherto not been recorded. In view of the renewed concern with them it may now be opportune at least to put on record some surviving impressions of these tombs, mentioning unlooked for contents and their background.

In late April 1909 the young Egyptologist Oric Bates led a three-week survey expedition to Cyrenaica under the sponsorship of the Archaeological Institute of America and the Museum of Fine Arts in Boston for the purpose of locating a suitable site for archaeological excavation. Participants in the expedition included Richard Norton, Allison V. Armour, and Russell C. Sturgis, Jr. At the conclusion of the expedition Bates sent a full report to the sponsors, who filed it away without acknowledgement. The report, which is published here for the first time, details this first official American expedition to Cyrenaica. Correspondence is discussed that reveals why the Archaeological Institute of America and the Museum of Fine Arts in Boston chose to ignore the report and to dismiss Bates abruptly from the project just several days after the report was submitted. The role that Norton played in Bates' dismissal is also examined.

This article considers three lithic artefacts (two sidescrapers and a gouge or plane) discovered at Augila Oasis, located in Cyrenaica along the dried riverbed of Wadi Nashoof and once directly in the line of the Western Desert trade and pilgrimage route. The three Augila tools belong to the so-called bifacial tradition, which spread in the eastern Sahara during the mid-Holocene, from c. 6000 to 4800 BC, and find parallels among the mid-Holocene production of the northern edge of the Farafra Oasis and other contexts of the Egyptian Western Desert. The presence of the bifacial tradition in the Lower Nile Valley and in the Libyan littoral could have unfolded through exchange networks during periods of favourable climatic condition in the mid-Holocene.

A central Tunisian red slip ware lamp from Sabratha, published by D. M. Bailey in 1994, should be classified as Type Salomonson I/Atlante I. Its discus decoration is a personification of Autumn, a standing female figure with a basket of fruit as a seasonal attribute and a Cupid (Eros) sitting on it. The prototype of the moulded decoration motif appears as an appliqué on a Hayes 171 el-Aouja sigillata jug of C1 quality. The lamp, which probably dates from the third quarter of the third century or the late third century AD, was subjected to chemical analysis and comparison with recently published reference groups from central Tunisian pottery-making centres showed that it was made at the central Tunisian fine-ware potteries at Henchir el Guellal near Djilma. A/D and C1-C4 sigillata as well as Type Atlante IV A, VI B, VII A1, VII A2 and VIII C1a lamps were produced there from about the second quarter of the third century until the mid fifth century AD.

Between 1946 and 1951 Richard Goodchild carried out the fieldwork that was to result in a seminal series of articles and publications on the ancient settlements of Tripolitania and Cyrenaica (Goodchild 1948; 1949a/b; 1950a/b/c/d; 1951a/b/c; 1952a/b/c; 1953; 1954c; 1971; 1976; Goodchild and Ward-Perkins 1953; Ward-Perkins and Goodchild 1949; 1953). The cartographic results appeared in 1954 as two splendid sheets in the ill-fated Tabula Imperii Romani (TIR) series at a scale of 1:1,000,000 (Goodchild 1954a/b). These twenty-two publications remain of fundamental importance to our understanding of the ancient topography of Libya.

Goodchild's map can with hindsight be seen as one of the few successes of the ill-fated TIR project. The TIR initiative aimed to produce 58 maps covering the Roman world, but huge problems have beset it all along and only 11 maps have ever appeared in definitive form. Although work continues in some areas, it must be considered improbable that this series will ever be completed (see Talbert 1992 for a thorough review of the history of the TIR).

The fact that it is now nearly 40 years since the compilation of Goodchild's two TER sheets for Libya is probably reason enough for resuming his interest in mapping ancient Libya. Much has happened in the interim to refine our knowledge of both urban and rural settlement, as a glance at the relevant volumes of Libya Antiqua, Libyan Studies and Quaderni di Archeologia delta Libia will reveal. For the study of the ancient geography and toponomy of Cyrenaica, the studies by Stucchi (1975) and Laronde (1987) are of particular importance. In addition to map corrections necessitated by the new information and perspectives, one may cite the inconvenience caused by the incompleteness of the TIR coverage to the south, east and west of the Leptis Magna and Cyrene sheets. For instance, how can we hope to understand the settlement geography of Roman Tripolitania without reference to Tunisian western Tripolitania or to the desert tribes (Phazanii, Garamantes etc)?