One of the difficulties facing the archaeologist when trying to use the evidence of ancient writers is to know when that writer (historian or otherwise) is relating fact, and when he is merely involved with ‘topoi’–the stock literary descriptions or situations inseparable from history as rhetoric. This note is concerned with two ‘topoi’ in the Agricola.

Except for the interest aroused by their cemeteries, the extramural territories of the Roman cities of Britain have never aroused sufficient interest for them to have been subjected to systematic exploration. Nevertheless, as a sensitive barometer to chart their changing fortunes, the margins of a city have considerable potential, as Esmonde Cleary's review of the evidence from the cities and towns of Roman Britain amply demonstrated. Much of what we know derives from the results of adventitious rescue excavation carried out in the context of inter- and post-War city and town development, but comparatively little has emerged from greenfield sites like Calleva which have very largely escaped modern interference. Since the final season of the Society of Antiquaries’ excavation of Silchester in 1909 when attention focused on the ditches surrounding the town wall, only limited investigations have been made of the extramural territory of Calleva. Both Cotton and Boon excavated parts of the Outer Earthwork on the western side, but the evaluation of the defensive sequence was their principal objective, rather than an understanding of the extramural terrritory per se. That has had to wait for the systematic study of the aerial photography and material collected by field-walking which has given us a first glimpse of the organization and use of the extramural territory, and which has now been complemented by the publication of all the aerial photography from Silchester. The two types of survey have shown that the main axis of extramural activity is east-west along the roads leading to London and Cirencester/Bath. However, there is almost no aerial photography to illuminate our understanding of the north-south routes to Dorchester-on-Thames and Chichester/Winchester, although surface collection has revealed spreads of material extending over three hundred metres south of the South Gate through to trie fourth century.6 To the north, however, extensive semi-permanent pasture has hindered non-intrusive investigation and only the evidence of major structures such as the road to Dorchester, and elements of lanes and streets, have been evident from the aerial photography.

Woodchester, a small village near Stroud in Gloucestershire, is famous for the Orpheus mosaic which underlies its former church-yard. This mosaic is the largest in Britain and appropriately enough is known as the great pavement. It is however only occasionally uncovered, and, indeed, has been seen on only three occasions since the war, in 1951, 1963 and 1973. In 1973 it was comprehensively studied by Dr David Smith, it was drawn by Mr David Neal, and a full photographic record was made by the Royal Commission on Historic Monuments (PL. XIV). In that year it was visited by over 140,000 people during the seven weeks it was open, and plans were mooted, and indeed still are being mooted, to place it on permanent exhibition. So far these have been thwarted by a lack of money and by the fact that the former churchyard has only recently gone out of use.

The Roman Legionary Museum at Caerleon, since 1930 a branch gallery of the National Museum of Wales, stands within the legionary fortress of Isca above a plot of the still little-explored scamnum tribunorum in the dextral angle of the via principalis and via praetoria (FIG. 1). With confirmation in 1982 of proposals for the complete rebuilding of the museum to an enlarged plan, the National Museum undertook responsibility for the archaeological excavation of the affected area (some 290 m2) which, in part, was scheduled under the Ancient Monuments Acts. In several episodes, between October 1983 and March 1985, all was systematically reduced to the level of the natural ground surface (at a depth of some 2–3 m below the present surface). Pressure of time and funds required that, in the later stages of this work, much of the basal (mainly Flavian) stratigraphy was removed by machine to reveal the fullest possible plan of the primary timber buildings of which so little is yet known at Isca (PLS IV A and B).

Of the senior Roman officials who took part in the administration of Britain perhaps the most intriguing is Severus' procurator, M. Oclatinius Adventus. His service in the province is recorded for us on two inscriptions erected in the governorship of L. Alfenus Senecio between 205 and 207. The first was found at Chesters on Hadrian's Wall:

This article continues the series on air reconnaissance in Roman Britain published by Professor J.K. St Joseph in the Journal of Roman Studies between 1951 and 1977. It is at once an affirmation of the undiminished value of this mode of archaeological prospection and an acknowledgement of the notable contribution made by that means to the study of Roman Britain by Professor St Joseph over a period of thirty-five years.

In November 1977, a large relief sculpture of Romandate was discovered during drainage work on farmland in the parish of Brant Broughton and Stragglethorpe in Lincolnshire. The sculpture had been unearthed by a mechanical excavator in the process of recutting the sides of a field-boundary dyke. Although it was not possible to examine the sculpture in situ, it had apparently been found lying face downwards in a shallow scoop in the subsoil below a residual east-west plough ridge of the broad ridge-and-furrow cultivation which until recently had covered the field. Apart from a fine follis of Licinius I (A.D. 308–324), the sculpture was not found in association with any other Roman material. Part of a worn and battered Roman altar and heavy concentrations of Romano-British pottery and building stone were later found after ploughing some 200 m to the north-west, and fieldwork has located a series of such concentrations in the surrounding fields. Air photographs of the immediate area show an enclosurecomplex together with a pit-alignment nearby. Two kilometres to the west aerial photography has pin-pointed enclosures of probable Romano-British date in the parish of Beckingham and occasional finds of Roman material have been recorded over the years in Brant Broughton.

The Claudio-Neronian kilns at Greenhouse Farm are one of the few extensively excavated pottery production sites of this period in Cambridgeshire. The analysis presented in this paper attempts to do justice to this fact as well as touching on some broader themes; divided in two parts, the first discusses the site, kilns and pottery production, the second tackles the broader issue of early (i.e. pre-Flavian) pottery production in the region. Apart from the military kilns associated with Longthorpe in Peterborough, the only other comparable site of this date in the county is that at War Ditches, Cherry Hinton, which lies only 3.5 km south of Greenhouse Farm and at which several phases of salvage work were undertaken as the site was gradually quarried away early in the last century. On a smaller scale, recent excavations at Blackhorse Lane, Swavesey have found two kilns of a similar date, while another first-century kiln was found in the 1950s at Water Newton.

This paper seeks to explore the contribution to our understanding of the Roman coastal forts between the Wash and the north Kent coast which can be provided by a study of the surviving building materials which can be attributed to them. Although some of our sites have little or no masonry surviving above ground level, we argue that the materials to which we draw attention were derived from their defensive wall circuits which were constructed between the late second and the late third century. While certain of our sites eventually belonged to the Late Roman Saxon Shore of the Notitia Dignitatum, others were not thus designated. All our sites have long histories but, since the materials in question probably derive from the initial, stone-building phase of these forts, it is likely that our conclusions will relate to, and inform our interpretation of, their primary function. Our survey includes the forts at Brancaster, Burgh Castle, Bradwell, Reculver, and Richborough, all of which can be assigned to the Saxon Shore system with reasonable confidence, as well as the fort, rather than defended civil port at Caister-on-Sea. We also consider evidence of materials which we argue were derived from Walton Castle, which has been identified by some as the Portus Adurni of the litus saxonicum listed in the Notitia Dignitatum, and which was lost to the sea by the nineteenth century.

Since 1983 the Museum of London's Department of Greater London Archaeology has completed nine rescue excavations (HGS. I and 2) and numerous watching briefs in advance of redevelopment within the area of the Roman cemetery to the east of the City of London. The excavations have uncovered 104 cremations and 575 inhumations, dating from the first to the fifth centuries A.D. This article is intended to make known some of the initial results in advance of full publication.

The cemetery lay on a gravel terrace 600–700 m north of the River Thames. The geology of the area consists of sands and gravels, locally overlain by brickearth. The cemetery was typical of urban Roman cemeteries, sited beyond the city limits and adjacent to a road from the city. It extended at least 600 m to the east of the city wall and covered an area of over 16.5 ha.

This article contains full editions with commentaries of the second and final instalment of the approximately 37 ink writing-tablets from Vindolanda discovered in the excavation seasons of 2001, 2002 and 2003. The editions are numbered continuously from 870, following the sequence in Tab. Vindol. IV.1, and are grouped in the following categories: Military Document, Letters, Descripta.

The earthwork at Garden Hill, Hartjield, East Sussex, until then unrecognised, was identified in 1968 by Mr C. F. Tebbutt, who found early Romano-British material in a trial excavation. Five seasons of excavation (1972–76) by the Garden Hill Excavation Group have established the broad outline of the site's history. A scatter of worked flints indicates slight occupation in the Neolithic/Bronze Age period. Attributed to the late pre-Roman Iron Age are a round house and part of what may be another. A hill-fort, with stonerevetted and palisaded defences, was built, possibly against the Roman invasion, but soon fell into disuse and was followed by Romano-British occupation. This included a rectangular timber building, roasting- and smeltingfurnaces and a forging-hearth of the first century; a rectangular building with two verandahs, using timber uprights set on padstones and in post-holes, and a four-post structure on the same alignment, both probably first-century; a timber building set on a stone platform and attached stone bath-building, of the second century; and undated post-hole and timber-slot systems (not fully excavated) representing fences and other timber structures. It is possible that Garden Hill was the base from which local iron-smelting sites were operated in the first and second centuries.

From the moment of its first publication in 1903 the career inscription of C. Velius Rufus has been the subject of a great deal of attention and scholarly debate. Why this should have been so is clear enough: the text records the career of a man who, beginning as a centurion (probably not in the ranks), not only rose to the coveted primipilate but went on to important independent field commands and an administrative career which culminated in major procuratorial governorships. More important still for historians of the period, the career included special commands which shed vital new light on the otherwise often obscure warfare of the reign of Domitian. The interpretation and chronology of the career as it appears in standard textbook treatments of the period may be said to be essentially that of Ritterling as modified and refined by a succession of eminent scholars, notably Syme, Pflaum and Dobson. This mainstream interpretation was summarized most recently by the last of these in 1978. From the outset, however, and in parallel with the development of this standard view, there have been a number of dissenters: von Domaszewski disagreed with Ritterling on a crucial point and, since then, there have been challenges to various aspects from, for example, McElderry, Hanslik, Saxer and Visy. Indeed, in the same year as Dobson presented a refined restatement of the orthodox interpretation, Visy coincidentally, offered a major re-examination. These earlier challenges, however, tend often to disagree with one another and, just as many to-day will be unconvinced by Visy's proposed revisions so, in most cases, his forerunners have found little support. This should not, however, obscure an important point, namely that there has been persistent dissatisfaction with the standard view from the outset. Nor should the implausibility or weakness of the proposed alternatives detract from the very serious nature of some of the objections raised.

This paper examines the landscape context of the Bartlow Hills, a group of large Romano-British barrows that were excavated in the 1840s but have been largely neglected since. GIS is employed to test whether it was possible to view the mounds from nearby roads, barrows, and villas. Existing research on provincial barrows, and especially their landscape context, and some recent relevant applications of GIS are reviewed. We argue that barrows are active and symbolically charged statements about power and identity. The most striking pattern to emerge from the GIS analysis is a focus on display to a local rather than a transient audience.

The present paper originally formed part of a Forschungsbericht on all the main branches of interior decoration in Roman Britain prepared for Aufstieg und Niedergang der römischen Welt n, 12, 4 in 1981-82. Since fifteen years have elapsed without any sign of ANRW, 11, 12, 4 seeing the light of day, I have decided to publish a new version, revised and updated to cover the half-century 1945-1995. The main difference is that the section on wall-paintings, on which I have written extensively elsewhere, is now omitted. What remains is primarily a review of research into Romano-British mosaic pavements, with brief comments on other media (opus sectile, wall-mosaic, and stuccowork) appended in a concluding section. The nucleus of this review is the Bibliography, arranged partly by themes and partly by locations, with items numbered serially (Section iv); reference will be made to it, where appropriate, in footnotes.

The research undertaken on the Scheldt Valley Commercial Activity Zone was the frst large-scale excavation in the Valenciennes region (northern France), permitting an insight into the organization of a large territory. It has yielded a great wealth of data on land use and settlement and it forms a key point of reference for any future research within the region for each of the studied periods. The article discusses the characteristics of the occupation of the territory, allowing the development of new models for the occupation, structure, and management of the land and the institution of settlements and their related material culture. It argues for a Roman-indigenous land cadaster, until now not observed for this part of Roman France. The different enclosures and their internal organization are described (buildings, wells, ponds, an artisanal zone, a road system, and a burial zone), showing the persistence of numerous protohistoric sites into Roman times, which is explained by the continuity of the agricultural system and a voluntary integration into a Roman administration. The Onnaing-Scheldt Valley demonstrates the importance of large-scale excavations for the understanding of land use.

An initial, superficial occupation (Period I, c. A.D. 43) of the site probably reflects a Roman military base of the Conquest period. The first formal layout (Period II, c. 44–50) included a main east-west street, north of which lay a metalled area and two wooden structures separated by a north-south road, again almost certainly part of a military base. A large scale Claudian redevelopment (Period III, c. 50–60) relaid both the east-west road and the metalled area. In the angle formed by these lay at least three large timber-framed buildings, on an east-west axis and probably consisting of ranges of rooms, at least one frontal verandah and two semi-basements; two of these buildings were divided from the third by a minor east-west road. The framed walls were infilled with wattle, daub or brick, over wooden sleeper-plates supported on trenched piles. The floors were of clay, some roofs tiled and many walls painted, one four times. Water was piped along the principal road. The layout, still reminiscent of military work, probably reflects the initiation of the civilian town east of the Walbrook which set the permanent street-pattern for the eastern half of the city. All the buildings on the site were destroyed by the Boudiccan fire of A.D. 60.

Since the first discovery of its remains in 1790, or rather since Samuel Lysons published his exceptionally fine reconstruction-drawing of it in 1802, the Temple of Sulis Minerva at Bath has remained outstanding among the monuments of Roman Britain that are known to have been built and decorated in the classical style. It has, however, proved difficult to establish its date. There are two available sources for this, in the absence of any inscription: the archaeological evidence, and the ornament of the temple's architecture. Cunliffe has summarized what little can be said of the former: although the precinct wall appears to post-date the wall of the reservoir of the baths, the temple itself may well have been built before its precinct was enclosed; since it is likely that the baths and the temple were planned together, the late first-century date indicated for the baths renders probable a first- or early second-century date for the other buildings. The arguments depend on circumstantial evidence. Unless excavations in the future produce more closely-datable archaeological material, it would seem that we are not on very strong ground here.

Most writers have seen the Car Dyke as a navigable Roman canal, connecting the fenlands of Cambridgeshire and Lincolnshire with the major Roman markets of Lincoln and York. Its course is shown on the Ordnance Survey Map of Roman Britain as running from Waterbeach on the Cam to Washingborough, a few miles below Lincoln, on the Witham, and many, following Stukeley, have seen it as supplying the northern markets with the agricultural produce of the Roman fenland. The importance of cereals in this trade has indeed been questioned, but the general theory has persisted.

In order to assess the validity of the theory, and to establish the purpose of the work and the date and method of its construction, the northern half of the Dyke, from the Witham to Bourne, a distance of about 36 miles (57-6 km), has been studied over a period of some 10 years. This paper summarizes the findings to date. It is, of course, realized that all the conclusions reached here do not necessarily apply equally to the southern end.