Correspondence addressed to the antiquary Sir John Clerk, 2nd baronet of Penicuik, Midlothian, allows us to document the initial recording in 1723 of Roman inscriptions and relief sculpture observed at several forts along the line of the Antonine Wall between the Forth and Clyde, and the arrangements put in place to transport them to his home at Penicuik House. Particularly valuable is a series of drawings, which are sometimes our only visual record of the stones.

The consular dated memorial of a military accountant (scrinarius) of a.d. 404 with a chi-rho monogram from Bavai (France, Nord), previously thought to be a forgery, is reconsidered. Geographically close to Britain and well-dated, it is relevant to the origins of post-Roman insular epigraphy and to the possibility of recognising specifically Christian tombstones in Roman Britain. The insular series derives from a late antique tradition introduced to Britain via the Christian Church at an uncertain date. There is little sign of continuity with claimed Romano-British Christian tombstones, but an early phase of the insular series can be recognised. Literacy and perhaps the ‘epigraphic habit’ survived in other media.

A review of epigraphic, numismatic and historical sources examines the episode of the war of Commodus in Britain. The relationship of this event to the present interpretation of the later campaign of Severus is discussed. Current views of the occupation of Hadrian’s Wall are examined in the light of Commodus’ policy in Britain.

The flat-topped hill of Burnswark broods over the surrounding landscape of southern Annandale. Its Iron Age defences enclose an area of 7 ha, modest when set against the hillforts of the neighbouring Tyne-Forth province, but without equal in Dumfriesshire. Its situation has often evoked comparison with the 10.5 ha site of Masada (Israel), but, even allowing for occasionally steep slopes rising some 70 m from its base, Burnswark's defences cannot match the formidable 250 m cliffs of the Judaean site.

In his article ‘A new look at some fourth-century mosaics’, Roger Ling has convincingly argued that the subject-matter of the mosaics at the villa of Brading, Isle of Wight, is derived from standard mythological and genre scenes. In the case of the mosaic from Room III (the so-called Abraxas or Iao mosaic) he has suggested that the two surviving rectangular panels owe their origins to African hunt-mosaics. The purpose of this article is to develop Professor Ling's explanation by looking in more detail at the mosaic and, in particular, at the two linear features associated with the griffins on the ‘Abraxas’ panel (PL. V) which have not previously been noted or discussed. These lines arguably represent millet stalks. As such they have significance for the interpretation of the mosaic as a whole, as they indicate that it is concerned with the activities of the amphitheatre.

Brandon Camp lies just south of Leintwardine in the central sector of the Welsh Marches. A Roman road running south from Wroxeter through the Church Stretton gap crosses the river Teme at Leintwardine and skirts the east side of Brandon Hill before traversing the Aymestrey gap and running on to Kenchester; thence branches lead to both Gloucester and Usk. The hill projects into the flood-plain to form a notable feature in this sector of the Teme Valley (PL. XI A), rising to a height of 165 m above O.D.; it has a broad almost level summit which is easily ascended from the south but is protected around the northern edge by low precipitous cliffs. The north-western end is occupied by a pear-shaped hillfort of about 8 acres (3.2 ha) with an excellent outlook west and north. An upstanding rampart 3 m high marks the south and east sides of the Camp, but along the steep scarp it is only just discernible.

The excavations at Greta Bridge (NZ o88 132) were sponsored by the then Ministry of Works as a rescue project occasioned by the re-alignment of a section of the A66 trunk road. The A66 follows the line of the Roman road from Scotch Corner to Carlisle very closely, to the extent that the Roman pavement lies immediately below the modern surface for most of its length. At Greta Bridge a slight southward deviation of the A66 seems to have occurred when the bridge carrying the route over the River Greta collapsed (FIG. 1). In the eighteenth century the new route was carried over the river on an elegant Palladian bridge — the subject of a widely reproduced watercolour by John Sell Cotman. At the outbreak of war in 1939 a Bailey bridge was built beside Cotman's to carry the weight of tanks and armoured vehicles used by the numerous army units stationed in the Catterick area. The army bridge remained in use until 1972 when the decision was made to straighten the road and to re-expose the Palladian bridge to full view. In choosing a new road line the engineers simply connected the ends of the known Roman route, thus putting the modern carriageway back to where it had been in the Roman period.

In a recent paper the late Dorothy Charlesworth published a valuable survey and gazetteer of finds entitled ‘Roman Carlisle’. This was the first attempt to bring together and summarize all the available evidence for Luguvalium, Rome's most north-westerly town and civitas capital. The paper dealt with both finds recovered during the previous century, many of which lack a secure provenance in the city centre, and with the results of observations and very small-scale excavations undertaken since the Second World War. At the time the article was written no major area excavations had been completed on Roman deposits and the first, Blackfriars Street, was still in progress. Since 1977, Carlisle Archaeological Unit has completed work on Blackfriars Street and is currently excavating in The Lanes, Castle Street, and Annetwell Street. The purpose of this paper is to draw attention to discoveries, mainly in the Lanes area, in advance of the first publication which is not anticipated before the late 1980's. The paper is divided into two sections of which the first (by M. McCarthy) outlines salient structural features, whilst the second (by T. G. Padley, Finds Assistant with the Unit, and Dr M. Henig, Institute of Archaeology, Oxford) deals with finds. It should be borne in mind that further work on all aspects of the site for final publication may result in the modification of some conclusions.

The fort at Birdoswald is situated on the line of Hadrian's Wall, some 25 km (15½ miles) east of Carlisle (FIG. 1). The site lies between the forts of Great Chesters 10 km (6⅓ miles) to the east, and Castlesteads 12 km (7⅓ miles) to the west.

The Roman site at Wood Lane End is situated on the east side of Hemel Hempstead (TL 082079) adjacent to the Industrial Estate and close to the MI intersection (FIG. I). It occupies the plateau on a 450 ft. contour about midway between the rivers Ver and Gade. Close by the land dips in a south-easterly direction to form the start of a shallow valley running across the Gorhambury estate towards Verulamium three miles away. The site is equidistant (2½ miles) between the Roman villas at Gorhambury outside Verulamium and Gadebridge situated alongside the Gade stream just to the north of Hemel Hempstead. The subsoil is of the Batcombe series—a pebbly loam over a bedrock of clay with flint. In places small pockets of gravel are encountered while occasionally fingers of chalk rise close to the surface. As elsewhere on the plateau the subsoil is surprisingly well drained, the drainage of the area being towards the Ver valley. The soil is suited to arable cultivation though its productive capacity is limited by the varying degrees of stoniness. The earliest cartographic evidence of the Wood Lane End area comes from the Tithe apportionment map of 1843 (DSA4/48). It would appear then that the site was located in a field called Hales Wood, but despite its name it was classified as arable, and was 7·021 acres in extent. None of the adjacent field names gives any indication of Roman or earlier features and neither do the boundaries of the 1843 map reflect the alignment of the Roman buildings. They relate solely to the roads, as shown on the nineteenth-century map.

Beading was frequently used as a decorative and strengthening finish for the rims of silver bowls and dishes in the late Roman and early Byzantine periods. There are two main types of beading:

(i) the bead-and-reel band (PL. XXA).

(ii) the single-bead band (PL. XXB).

The first consists of a series of oval or sausage-shaped solid beads alternating with smaller double-ridged reels (PL. XXC), while the second is a row of beads, often hollow, nominally of the same size, placed closely together (PL. XXE). In addition, some hybrid beaded rims, intermediate between the two, are found (for example, the two flanged bowls from the Carthage Treasure have a single row of small hemispherical beads worked almost entirely from the front). Beading is usually positioned on or close to the rims or outer edges of the dishes or vessels. Beaded decoration was particularly popular during the late Roman period, and decorates vessels found in many different locations, both within the Empire and outside it, for example fragments of beading found in a hoard of hacksilber from Coleraine, in Northern Ireland (PL. XXD), and a flanged silver bowl from the River Don in southern Russia, as well as vessels from Mildenhall, in England, Chaourse, in France, Carthage, Syria and Constantinople. The use of bead-and-reel decoration seems to be the earlier of the two techniques and was used on many of the vessels in the Chaourse Treasure in the third century A.D., while material from later hoards of the fourth to the sixth centuries, such as the Mildenhall treasure, show that the large single-beaded band had then become extremely popular. To discover the techniques employed to produce this simple beaded decoration examples of both types were examined using optical and scanning electron microscopy and X-ray fluorescence spectrometry. Models were made in copper to test the resulting theories on how the different methods of beading were carried out.

The fort at Auchendavy on the Antonine Wall (FIG. I), 3 km east of Kirkintilloch, was first reported in the preamble to the Theatrum Orbis Terrarum of Johannes Blaeu, from notes made before 1625 by the Scottish geographer Timothy Pont. Subsequent references in the antiquarian authors over the following centuries are testimony to the gradual destruction of the once impressive fort-defences. Today the traveller by road along the B 8023 from Kirkintilloch eastwards towards Twechar passes the fort-site without a glance, even though the modern road traverses it more or less on the alignment of the via principalis. No organized excavation has ever been undertaken at Auchendavy, although small finds reported over the years from the fort-site and vicinity have included pottery, a deposit of ballista balls, a nicolo paste intaglio in an iron finger-ring, and several coins. In May 1771 during construction of the Forth & Clyde Canal, which cut away some part of the southern defences of the fort, workmen recovered a group of five altars, two iron mallets (lost or disintegrated by the later 19th century) and a broken cult-statue, from a pit seemingly just outside the fort-defences to the south (below, p. 33). An aureus of Trajan (datable to a.d. 100), reported from the site, may have been found at much the same time. From the late 1940s onwards aerial reconnaissance by Professor J. K. S. St Joseph provided a useful check on the layout of the fort-defences, and recent fieldwork and ground survey have added fresh details, so that a reappraisal of our knowledge of the fort is perhaps now a worthwhile proposition.