Cartimandua first appears in the historical record in a.d. 51, in which year she handed Caratacus to the Romans. She was already ruler of the Brigantes: how long she had ruled is not known, but it is quite possible that she was in power even before Claudius' invasion in a.d. 43. The Brigantes seem to have been a federation: how close-knit this federation was and how far and firmly Cartimandua's authority extended over it must be matters of some doubt. Her history after a.d. 51 has often been narrated by some of the greatest students of Roman Britain: it comes as something of a surprise, therefore, to discover that this latter part of her career also raises major difficulties. For knowledge of this portion of her career, we depend almost entirely on two passages of Tacitus: Ann. xii 40 and Hist. iii 45, written first of course. The central difficulty is simple enough: how are these two passages to be related?

The recognition by R. E. M. Wheeler in 1920 that the massive Norman keep at Colchester encapsulated the podium of the Temple of Claudius stimulated Dr P. G. Laver, his collaborator in an earlier study of the colonia, to excavate in and around the keep during the 1920's and 30's. Some of the results were published by the late Rex Hull in Roman Colchester (1958). At the time of Hull's death in 1976, the writer was preparing the drawings to accompany a paper which included a summary report by Hull on the excavations undertaken within and south of the keep by Laver, assisted by E. J. Rudsdale, in 1931–3. Subsequently much additional information about these excavations came to light in Colchester Museum. This raised a number of important questions, which led in 1977 to one of Laver's trenches in the sub-crypt being re-opened and extended, and the cutting of a trench in front of the keep. Mrs B. R. K. Niblett (née Dunnett) also made available the results of her excavations on the west side of the Temple precinct in 1964 and 1969. The evidence from these sources relevant to the history of the site in the Roman period is presented here; and in conjunction with previously published material has been used as the basis of a re-appraisal of its development, summarized in TABLE I. A similar study of the post-Roman history of the site has recently appeared. Much of the discussion is necessarily speculative, being intended primarily to advance hypotheses which may be tested in future excavations and by further analysis of the surviving structures.

This note draws attention to the outer defensive works enclosing a number of Roman forts and fortresses in Britain. Most of these have been individually noted by St. Joseph in his periodic reviews of aerial reconnaissance in Roman Britain, but while some more general remarks were made by Wilson, there has been no detailed discussion of their form and possible functions.

Urban defences in Roman Britain and the dates of their erection have been discussed by various writers during the last three decades, and the subject has received renewed attention in three recent publications. The last of these has put forward views concerning both date and motivation which a wider consideration of the evidence makes hard to accept. In general, however, the outline of events is clear. Although certain cities received earthworks in the first century or masonry defences from early in the second, the great majority remained open until, in the second half of the second century, a very large number of earth-work circuits was erected in what, all are agreed, must be either a concerted programme or a chain reaction. This embraced not only the cities (the civitas capitals) but also a great variety of minor centres. Beginning quite early in the third century, or even at its outset, and continuing for 70 or 80 years, a long series of towns was receiving walls in masonry. Where earth ramparts already existed the front was normally cut back to receive the wall; elsewhere fresh banks were piled behind the new work. Free-standing walls appeared only in the fourth century and are found at very few towns, for most were already defended. Another development of the same period was the addition of external towers to existing walls: most of these towers seem to date from the third quarter of the fourth century, but some were earlier, beginning, perhaps with Verulamium, c. 265–70.

The Cawthorn encampments in North Yorkshire (FIG. I) have found wide publication as ‘practice forts’ with attendant construction camps following their investigation by the late Sir Ian Richmond in the 1920s. The fundamental reason for his interpretation of Cawthorn as the site of military manoeuvres rather than as a normal fort site was its alleged unsuitability for permanent occupation coupled with apparently unfinished defences. However, it has recently been postulated that Camp D, the most regularly laid-out of the four, represents an auxiliary fort. This suggestion does not contravene the evidence; indeed, Camps A and B (the secondary enlargement of A) need no more be connected with manoeuvres than Camp D. One of the major objections to their being regular forts is based upon the presence of ballistaria. Richmond used these as proof that legionaries rather than auxiliaries were involved at Cawthorn; certainly, it seems that the auxilia were not equipped with artillery in the first and second centuries a.d., SO that artillery-platforms at this date should indicate the presence of legionaries. The conclusion that the encampments are not regular forts but practice-works is reinforced by the fact that none of the ballistaria is situated at a strongly defensive position. Thus, the application of Richmond's hypothesis to Camps A and B appears to be sound. Nevertheless, it represents the only evidence that the Romans ever built forts specifically for practice. Although practice camps are abundant both in the archaeological record and in the literary sources, nowhere are practice forts attested.

It is just over one hundred years since Friedrich Hultsch published the second and enlarged edition of his Metrologie, so that it would seem appropriate, in view of the current discussion on Roman town-planning, to proffer some observations on one of the best-known, yet to some extent little researched, Roman towns in Germany.

Several recent articles in this journal have highlighted the fact that much effort has been expended trying to understand the intricacies of Roman planning, both in the civil and in the military context. They have also reiterated the point that one may not take it for granted that a given metric measurement may be simply divided by (say) o · 296 in order to provide a measurement in terms of so many ‘Roman feet’. The imperial and metric weights and measures are the regulated quantities of a modern era obsessed with standardization; one cannot expect such a rigorous uniformity when dealing with Roman standards in the north-west provinces, often based, as they were, on a wooden measurement-rod. However, when confronted with an excavated building or, on a grander scale, a whole town, it is the task of the excavator to examine whether or not a systematic plan on the part of the Roman architect or surveyor could have existed. More often than not there will have been one. The problem then is to identify not only whether the standard foot or the Drusian foot was used, but also whether differences within the length of these measurements themselves are discernible. Only by accurately documenting a great number of buildings on an equally great number of sites can one hope to make good the limited amount of evidence which the ancient authorities have left us.

Face pots and head pots which, while never common, were clearly much more widely used in Roman Britain than the number of surviving whole vessels might lead one to believe, are among the most attractive and least documented products of the Romano-British pottery industries. Face pots, with their crude, barbaric, rather comic-looking features stuck incongruously on a well-made Roman jar, are quite unlike any other type of Roman pottery, where free-hand figurative decoration is practically unknown. Head pots, with their contrived hair styles and naturalistic features, are more classical-looking and more acceptably ‘Roman’, and yet they seem to be a purely insular development, found only in the remote province of Britain and with no obvious close counterparts anywhere else in the Roman Empire except perhaps in North Africa. The purpose of this article is to outline the development of these two types of vessels in Britain and to make a preliminary attempt to define the regional groups that evolved, with a brief reference to the earlier and parallel developments on the Continent, and finally to discuss, in the light of the meagre evidence available, what might have been their function and possible significance.

In the past our recognition of Gaulish pottery in Britain has been largely confined to products of Central, Southern and East Gaul. It is significant that this pottery is, almost without exception, oxidized and/or colour-coated; the study of grey wares has traditionally lagged behind that of the more classical slipped table wares, and this has resulted in a somewhat uneven picture of Roman trading patterns. Coarse grey pottery, with its more utilitarian function, has not proved easy to define and date on stylistic grounds, and ‘miscellaneous grey wares’ have often been dismissed as the products of unknown local kilns, or, as form and fabric analysis have improved, described as unprovenanced imports. In the last ten years we have become increasingly aware of the complexities of the coarseware industry, and more receptive to the possibilities of long distance and inter-provincial trade. Gillam, Farrar and Webster have drawn attention to the evidence for large scale transportation of coarsewares by sea and Fulford has demonstrated that the products of larger British coarseware industries, Alice Holt/Farnham ware, and Black-Burnished ware 1 and 2, are found in small but significant quantities in Northern France. This paper describes a greyware type which was imported from Northern France into Britain. It stems from research on the late second-/early third-century pottery from New Fresh Wharf, London (B.R.), and the coarseware industry of Gaul (P.T.) and accordingly incorporates two sections reflecting these different aspects, with a concluding discussion by both authors on the British distribution.

In December 1976 the construction of a new private road and car park at The Hall, Welwyn, exposed traces of Roman building material. These were observed by Tony Rook, who led the rescue team of the Welwyn Archaeological Society on the site. Initial excavation during the Christmas break was limited to the removal of the hardcore already on the new road and the plotting of the feature so exposed. Later, excavation was permitted throughout 1977 in two bays of the proposed car park. The corner of a very large robber-trench was investigated, and, with the information from the road, supplemented with trial-trenching, the plan of a mausoleum and its enclosure ditch was recovered. One length of the enclosure ditch contained inhumation burials of Roman date. A report on the skeletons by C. B. Denston is appended.

A large number of hoards of silver coins are known from late fourth-century Britain, and every year still more come to light. These hoards usually contain more or less the same sorts of coin and it may be helpful to give a summary of the main issues, as they are not presented very clearly in the standard reference book for the period, RIC IX.

In the received history of Britain in the fourth century, a special place is occupied by the Saxons. It is asserted that the defensive system of the Saxon Shore existed to prevent them from landing; and it would seem that, unlike any of the barbarian tribes with whom Rome maintained uneasy relations along the Rhine and the Danube, unlike even the Goths, the possibility of their peaceful settlement within the boundaries of the most accessible Roman province could not be entertained. Their attacks, apparently, were so unpredictable, and their nature so savage, that in this case alone the policy of the Roman authorities was one of total exclusion. In 367, moreover, after forming an alliance with the tribes of the north and the west, they overwhelmed the Roman defences and ravaged Britain; and this date has been accorded the status of a turning-point in fourth-century history. But afterwards the Saxon Shore forts resumed their function of keeping the Saxons out. Finally, the Roman garrison was withdrawn and the Shore forts abandoned; and this permits historians to assign an early fifth-century date to the beginning of the Anglo-Saxon settlement.

Tacitus is the only ancient writer to give any information about the Agri Decumates, or indeed even to mention them. Thus, all that we know of them is confined to the period c.a.d. 70-98. In the Germania, dated firmly to the latter year, Tacitus tells us that the occupants of the Decumates agri could hardly count as Germans. They were rather immigrants of fairly recent origin drawn from the least substantial groups among the Gauls, who, through want, had taken up possession of this risky territory. More recently, however, a defensive line had been drawn, garrisons sent in and the area had become a pocket of imperial security and a part of the province (of Upper Germany).

The context numbering for 1983 follows on from the 1982 excavations. Feature numbers quoted and shown on plans and sections are those used on site. However, for the convenience of this report the actual buildings are renumbered (1–11) and either retain the number given in the previous report (Vol. xiv), or follow on.

Following the reporting of the excavations at this site in the previous issue of Britannia (Vol. xiv), adjacent plots of land were purchased by the developers and incorporated into the new housing estate, the enlarged area comprising a broad swathe of land south of Nos. 108-114 Wood Lane End (FIG. I ) and, in particular, part of the garden of No. 104 which covered much of Building I. Most of the additional land lay outside the protected area but scheduled monument consent was required for construction over the garden of No. 104. This was granted by the DOE provided that the Roman building known to lie there remained undisturbed within the gardens of the new houses. However, during top-soil clearance the south-west side of Building I was accidentally exposed and it became apparent that its plan was not that previously assumed on the evidence of trial-trenching and geophysical survey carried out in 1966-7 and 1982 respectively. In addition, following the stripping of top-soil elsewhere, traces of other Roman buildings and features were discovered. Although development had already begun, the architect and builders were most co-operative in allowing excavation to take place; the excavations were carried out by the DOE's Central Excavation Unit between August and October 1983.

The summer of 1983 was remarkable in certain parts of North Britain for the severity of the drought which developed during the months of July and August, all the more so because it was unexpected. The weather had been far from settled, even in early June, and among those engaged in aerial reconnaissance praesumpta illius anni quies. By the end of the season, however, flying in central and eastern Scotland, south of the Highland Line, had met with unparalleled success in terms of the quality and quantity of archaeological evidence recorded from the air. Considerable interest attaches to the identification of no fewer than three Roman forts, all of which appear to be of Flavian date (see below, p. 275), but one of them, the newly discovered work at Doune, appears to possess a special significance which justifies the presentation of this early report.

Over the past decade most historians and archaeologists have come to accept that the landscape of Roman Britain was an extensively farmed and settled one. Fieldwork has consistently revealed evidence of Romano-British settlement in areas of heavy soils once thought to have been first extensively cleared and colonized only in the Anglo-Saxon and medieval periods. This is particularly true of the areas of southern and eastern England which have an ‘Ancient Countryside’. In the medieval period these lacked a strong tradition of open-field agriculture and had a settlement pattern which included large numbers of hamlets and isolated farmsteads. Both these features have often been directly related to the chronology of settlement in these areas, on the assumption that early and pre-Saxon settlement was limited to restricted areas of more favourable soils. The dispersed pattern of settlement and the complex pattern of small open-fields and irregular hedged closes are normally interpreted as the result, therefore, of a gradual expansion of settlement in the later Saxon and medieval periods at the expense of formerly uncleared woodland and waste. Compared with the Midland counties, where settlement was more strongly nucleated in the medieval period, such areas were relatively well-wooded at the time of Domesday and often still carry extensive areas of Ancient Woodland. In such areas place-names indicative of Saxon woodland clearance, especially those containing the O.E. element leak (modernley, leigh etc: usually in place names translated ‘a clearing in a wood’) are common, and this has been used to support the conventional model. Various aspects of the appearance of such landscapes, for example the irregular pattern of fields and woods, are also often thought to reflect their origin in late and piecemeal colonization.