In the study of Romano-British villas the reconstruction of a site's economy is treated as fully as the evidence will allow. One approach to this has been to attempt to calculate the size and productivity of an estate from the excavated remains of the principal dwelling and other buildings in its close vicinity. It is now recognised that such guesses are of little value, and for some the response has been to excavate larger and larger areas so that a more complete picture can be offered. However, our present emphasis on the productive economy of villas reflects the pre-occupations of our own day and age. In this there is a danger that we will be tempted to neglect what villas can tell us about consumption and expenditure, and in particular to transfer our own patterns of private spending back into the first and second centuries A.D. This paper attempts to focus on some aspects of expenditure and consumption, and of other non-productive activity, that are perhaps now being overlooked in our present climate of thinking.

Hadrian's Wall studies owe a great debt to those who raised the flag of revolution in 1848. They forced John Collingwood Bruce to abandon his planned holiday in Rome and visit Hadrian's Wall instead. Although he is known to have visited the Wall as a boy, a student, and a young man, now, at the age of forty-two, Bruce commenced the study by which his name is principally remembered. Bruce's Handbook to the Roman Wall remains the primary academic guide to Hadrian's Wall. Four successive editors, and now myself, the sixth including Bruce, have revised — or are revising — the Handbook in ten editions since Bruce's death: thirteen editions so far. This discussion flows from the work of editing the new edition of the Handbook, which is due to be published in 2004, and of considering the contributions of the successive editors.

Compared with many other branches of Romano-British archaeology, the study of buildings is in its infancy. This paper concentrates on one aspect of building study, the principles of design upon which plans were based; and it is further confined to military buildings. These were selected partly to limit the material under consideration to a manageable group and partly because the methods used to erect them were more likely to have been standardised to some degree. Military architecture should not, however, be seen as a selfcontained discipline: it can only really be understood as a manifestation of architectural methods and philosophy in a wider world.

Excavations and observations on adjacent sites in the heart of London's ancient harbour produced exceptional evidence of waterfront activity from the mid-first into the fifth centuries. To the south of the natural pre-urban river bank were found substantial traces of structures thought to represent a pier base for a timber bridge across the Thames and an open-work landing-stage. The latter was replaced in the late first century by an infilled timber-faced quay traced for over 70 m east–west. Two storage buildings, each 25 m × 6 m, were laid out over the northern half of the quay area. The line of the waterfront had been advanced beyond the southern limit of excavation by the early second century, and again by the early third century, and these major modifications are reflected in the subsequent development and layout of the buildings and associated features recorded in the excavations.

The well-preserved timber structures are described and illustrated, the development of the storage buildings is discussed in detail, and the first century river level is assessed.

In the course of examining the charioteer mosaic discovered at Rudston (Yorks.) in 1971 it struck me that certain aspects of the iconography had been either misunderstood or over looked in the first publications. In particular the idea that the bust of a Season holding a rake must be identified as Winter seems to rest on a misconception. It is much more likely that the Season in question, both here and in other similar representations, is Autumn.

Archaeological knowledge of Canterbury's remote past obtained by excavation is of comparatively recent acquisition, being in large measure the result of work carried out since the second world war. In 1868 the streets were trenched for the installation of drains, and the city engineer of the time, James Pilbrow, F.S.A., wrote a brief account of his discoveries for Archaeologia, xliii (1871), illustrated with a plan of the city and one or two diagrams. Unfortunately the opportunity was not taken to note the Roman street plan. Nevertheless important details of the building here to be discussed were recorded under the south end of St. Margaret's Street and the west end of Watling Street.

The building of some 75 miles of massive curtain-wall, incorporating forts, milecastles, turrets and bridges, together with a further chain of forts, milefortlets, and turrets 32 miles long on the Cumbrian coast, represents an impressive civil-engineering undertaking by any standards. The fact that this was accomplished in a relatively short space of time on the remote northern fringe of the Roman Empire makes the achievement even more remarkable. Unfortunately, there is a lack of firm evidence for many aspects of the building of the Wall, so that it is impossible to assess precisely the scale of effort and organisational complexity involved in this operation. Many details of its original structure remain a matter for conjecture, since it has nowhere retained anything like its original height, and large sections of its length lie beneath buildings or roads, or have been destroyed by quarrying, cultivation, or the robbing of stone for construction. Alterations during the three centuries of Roman activity on the Wall have, in many cases, obliterated evidence of the earliest structural features. The actual source of much of the original building material is uncertain, so that transport routes and distances must remain a matter for debate. There is little archaeological record of Roman transport methods, and depictions of vehicles, such as those on Trajan's Column, owe more to artistic licence than to realistic detail, so that here again methods can only be conjectured. Literary record of happenings in Britain around A.D. 120 is sparse and precise dating of events, such as Hadrian's presumed visit to Britain, rests largely on a concensus of opinion based on circumstantial evidence.

The existence of chains of signalling-posts in some Roman provinces has been accepted as a normal feature of military control, but recently Donaldson has raised some common-sense questions about the efficacy of such systems and the technicalities of their operation. As he points out, over anything but short distances, visual signals of any kind whether by flags, or lights, or any other means, can be disrupted by inclement weather and are susceptible to faulty transmission and misinterpretation, so that only the most uncomplicated messages can be sent by these methods. Consequently, long distance communication by relays between chains of signalling-towers was probably entirely out of the question. It may be significant that in the section ‘on sending and receiving messages’ in Frontinus' Stratagems there is no mention whatsoever of communications by visual signals and no mention of signalling between towers, despite the fact that Frontinus was fully aware of the towers built by Domitian in Germany, described in a different passage. All the messages which Frontinus includes in his list were sent either verbally or by letter. One series of messages was sent by pigeon.

Three aspects of Cassius Dio's account of the Claudian invasion of Britain are discussed. First, the convention, by which ancient historians routinely introduced their detailed narrative of military campaigns with a summary, allows the two first battles in the sequence to be eliminated as separate events. Secondly, the kings, Togodumnus in Cassius Dio and Cogidumnus in Tacitus' Agricola, are taken to be the same individual, who after defeat was reconciled to be Claudius' client-ruler. Thirdly, the argument is restated that the invasion took place through the harbours of West Sussex on a route suggested by the description given by Dio and by the evidence of place-names, known from the Geography of Ptolemy and the Antonine Itinerary.

Religious beliefs current in the Imperial army are attested by numerous inscriptions. The Roman soldier was probably no more superstitious than his civilian counterpart but the dangers to which he was exposed rendered him in especial need of divine assistance, and so, in Britain at any rate, a large percentage of the extant dedications to gods and goddesses were erected by men from the legions or the auxiliaries. Other related aspects of military thought have received less attention however. Particularly worthy of examination in this connection is the attitude of the officer class within the army to the great heroes of the past. Even the slightest knowledge of Ancient epic must have been enough to ensure that the officers in both legions and auxiliaries (and almost certainly other ranks in the legions, as well) were acquainted with such events as the Trojan War and the foundation of Rome. It would have been natural for these men to have seen themselves as the inheritors of a glorious tradition established by heroes who had overcome all difficulties.

This paper is concerned with the nature of the contacts between late Roman Britain and the seafaring peoples of the continental North Sea coast. Evidence for Germanic piracy during this period is extremely slight, with the consequence that notions about its character are poorly defined. However, this paper argues that there is a basic similarity between these barbarian attacks and those of the late eighth- and early ninth-century Vikings against England, Ireland and northern France. The Vikings are much better evidenced, both in terms of written sources and the archaeological record: this makes it possible to offer a model for the nature, scale and consequences of Germanic piracy in late Roman Britain.

This paper provides description and context for some of the discoveries made by the Royal Commission on the Ancient and Historical Monuments of Wales during aerial reconnaissance in the drought conditions of the summer of 2018. New discoveries include two marching camps, three auxiliary forts and a remarkable series of stone buildings outside the fort at Pen y Gaer. The photographs also clarify the plan of several known villas as well as identifying some potential villa sites and enclosure systems of probable Romano-British date in south-eastern, south-western and north-western Wales. The recognition of a new road alignment south of Carmarthen is suggestive of another coastal fort at or near Kidwelly.

In the past few decades, the spread of the knowledge of ironmaking technology from its origins in the Near East throughout the Old World has been the subject of considerable study. Archaeologists and metallurgists have collaborated in the investigation of technological material, notably furnace remains, slags, and iron artefacts. Slowly a coherent picture is beginning to emerge, but many questions remain unanswered.

Romanization is a term that people currently love to hate. Few these days approve of the view that Romanization was about bringing an inherently superior Roman civilization to the natives. Many would go further and attempt to replace the term altogether — the most successful of these attempts being Woolf's focus on what it meant to become Roman. The heart of Woolf's argument is that being Roman was not a static quality that people embraced or rejected, but rather a set of cultural forms always in the process of ‘becoming’ and that all within the Empire participated in its creation. ‘Gauls were not “assimilated” to a pre-existing social order, but participated in the creation of a new one’. Roman culture was an entity created anew in different parts of the Empire at different times, through the actions of all groups, and it was not just native peoples who were being Romanized through the expansion of Empire, but also the Romans themselves. The view of mutual creation is one we find attractive. However, there is one element of the argument that we would like to qualify and that is that the process is seen as primarily happening through the élites and the new urban centres created within the Empire — ‘A symbolic system did exist in the Roman cultural system, but it was not located in any one place or region but rather in the set of manners, tastes, sensibilities and ideals that were the common property of an aristocracy that was increasingly dispersed across the empire’.

The circus at Colchester was discovered late in 2004. By the end of 2007, a succession of targeted investigations helped reveal its basic plan and provided some limited dating evidence for its period of use. The building proved to have been of average length for a provincial circus (457 m) but, at about 71.5–74 m across, it was relatively narrow. The stands appeared to have been raised a section at a time. The seating was carried on an earth bank revetted by two stone walls. The circus possessed only eight starting-gates. Excavation of the near meta yielded some remarkable evidence about the size and nature of the cones, the supply of pressurised water to the barrier, and the tactics adopted by the charioteers as they swung around the metae. The circus was built in stone probably in the second century A.D. The building ceased to be used if not c. A.D. 275, then sometime later in the Roman period.