Stratum III occurred in the Triangle, the two St trenches, in the Annexe, in the Funerary Cleft, and in the entrance of the main cavern. In the back of the main cavern it began at a depth of about 1·25 m. below the surface, and had an average depth of about 1.25 m., which is as extensive as stratum II and I together. In the Annexe it continued to 4·50 m. In the main cavern, stratum III was a compact mixture of sherds with lightly coloured soil, mostly sand, ashes, and disintegrated limestone. Among the very numerous sherds we found some handmills, bone tools, spindle whorls, and loom-weights, but only one formless piece of metal, apart from what was buried with the dead, though there were plenty of whetstones. The only recognisable features among this amorphous mass were the hearths. They had been built of clay, which must have come from somewhere outside the cavern, laid carefully over a layer of pebbles. This indicates that they were not used merely once to serve a casual need, and confirms the use of Manaccora as a dwelling place. Several hearths existed at the same time, which makes it likely that more than one family made use of the cavern. That would account for the quantity of material found, and would be in keeping with modern usage of Monte Gargano, where to live in a cave is still quite common. A smaller cave adjoining Manaccora was inhabited during our excavation.

On the 22nd September, 1950, whilst workmen were digging a trench for the laying of a water main at Ghajn Qajjet, near Rabat, Malta (Malta 2″ Map. Ref. 356247) at a depth of 2 ft. 8 in. below the surface of the road, the burial chamber of a rock-cut tomb was broken into (fig. 1).

This chamber (fig. 2), rectangular in plan and with a flat ceiling, measured 12 ft. in length, 10 ft. in width, and 5 ft. 7 in. in height; its long axis ran in a north-easterly direction. A slab of Globigerina limestone (pl. XII), 6 ft. 5 in. long, 5 ft. 8 in. wide, and 5 in. thick, rested horizontally on the floor and was set with two of its sides in contact with the north-east and the south-east walls; owing to the slight inclination of the floor, being higher towards the north-east, it was propped up by roughly dressed small blocks of stone. There was also a cavity, 10 in. in diameter and 6 in. deep, cut in the floor at a distance of 4 in. from the south-west side of the stone slab and quite close to the south-east wall of the chamber.

During the North African campaigns of 1941–3 numerous air-photographs of the Tripolitanian coast were taken by the R.A.F. for operational purposes, and the site of Lepcis Magna was included in the area covered. Examination of these photographs (pl. XV) showed many suggestive features relating to the defences of the ancient city, and a preliminary ground survey was later (1947–50) undertaken to establish, with a minimum of excavation, the course of the successive wall-circuits.

The results of this investigation are described below, and are discussed in relation to the historical and epigraphic evidence. It is not claimed that these results are exhaustive, or that they will not need modification in the light of future discoveries. Since, however, there is little likelihood of any early resumption of large-scale excavations at Lepcis, this preliminary study may help to illustrate the growth and subsequent decline of the city that came to be the most important centre between Carthage and Alexandria.

Gasr el-Gezira stands on high ground about one kilometre south of the edge of the escarpment overlooking Wadi el-Matmùra where it debouches on to the Gefara, the coastal plain of Tripolitania, and about four kilometres due north of kilometre-stone 166 on the Jefren—Giado road. The escarpment in the neighbourhood is over 400 metres high, and the building stands at a height of 745 metres above sea-level, on the watershed between the wadis running down to the Gefara and those feeding the affluents of the Upper Sofeggin system to the south.

The building is surrounded by scattered troglodyte dwellings and sparse olive groves, interspersed with fig gardens and more open land used for cereal cultivation. The remains of a Roman village lie some three hundred metres to the south-east, and the whole complex marks the north-eastern extremity of an area of Roman olive cultivation, roughly coinciding with the district known as ez-Zintan, and probably to be assigned to the period between the first and the fourth centuries a.d.

This paper summarises the results of work carried out during the years 1949–51 in the eastern Gebel of Tripolitania. The area had been visited, and some of the sites noted, by nineteenth-century travellers and by Italian archaeologists between the wars, and the adjacent plateau of Tarhuna has been the subject of a recent paper by R. G. Goodchild. My work was complementary to his, and I have not reproduced the details of information so lately set out except where they were immediately necessary.

I was able to spend long periods in the territory while holding the Scholarship in Classical Studies at the British School at Rome and a Research Scholarship at Trinity College, Cambridge, and I am deeply indebted both to the School and to my college for their generous support in unusual circumstances: in particular the latter, for providing a truck which carried me over fifteen thousand miles in difficult country and for a generous grant towards the cost of illustrating this paper.

This article forms part of a larger study of Roman architectural ornament undertaken during two years of residence in the British School. The subject was first suggested to me by the Director, to whom I am grateful both for his constant encouragement of my work and for a great deal of practical advice and help in the preparation of this article for press. My friend, Dr. Enrico Paribeni, has been an unfailingly generous source of information, and I wish to thank him especially for drawing my attention to the fragments in the Via del Banco di Santo Spirito and for identifying some fragments from the Hadrianeum. Mrs. Sheila Rizzello has shown great patience and skill in the preparation of the drawings and tracings; and the text could not have been completed without the help of Miss Shirley Twallin.

In the Via del Portico d'Ottavia at Rome, close by the Theatre of Marcellus, there still stands the propylaeum of the Porticus Octaviae, from which the street took its name. The inscription on the architrave records the restoration of the building by Septimus Severus and Caracalla in A.D. 203, but passes over in silence the previous history of the site. For that we are dependent largely on literary sources, chief among which is a passage of Velleius Paterculus, in which he records the building-activities of Q. Caecilius Metellus Macedonicus (cos. 143, censor 131 B.C.), the conqueror of Andriscus: Hic est Metellus Macedonicus, qui porticus, quae fuerunt circumdatae duabus aedibus sine inscriptione positis, quae nunc Octaviae porticibus ambiuntur, fecerat, quique hanc turmam statuarum equestrium quae frontem aedium spectant, hodieque maximum ornamentum eius loci, ex Macedonia detulit. … Hic idem primus omnium Romae aedem ex marmore in iis ipsis monumentis molitus <huius> vel magnificentiae vel luxuriae princeps fuit.

Velleius' statement, that the site of the Porticus Octaviae had formerly been occupied by a Porticus erected by Metellus, is clear, and adequately confirmed elsewhere. Less clear is what he says about the two temples, which we know from other evidence to have been dedicated to Juppiter Stator and Juno Regina.

My Supplement to Paestan Pottery was already in the press when, in the late summer of 1952, Dr. Sestieri began excavating in the vicinity of the so-called Temple of Poseidon at Paestum. An intensive campaign yielded a very rich harvest of finds, as well as providing clear evidence that the temple itself was part of a Sanctuary to Hera, which included a number of smaller temples and a long row of altars. Among the finds were several complete red-figure vases and an immense number of fragments which, with characteristic generosity, Dr. Sestieri allowed me to examine in detail in the basement of the new museum recently opened on the site. Nearly all of them are Paestan and provide further confirmation for the location of the fabric at Paestum; pending a fuller publication of Sestieri's important discoveries, it seemed advisable to me to give a classified list of the most significant of the new pottery finds as soon as possible, adding a few comments on outstanding pieces.