The axiom ‘an army marches on its stomach’ applies to all armies of all times. The task of providing a constant and sufficient supply of food was by no means the least important part of the work involved in the day to day running of the Roman army. In time of war the troops would forage from enemy countryside, requisition supplies from defeated tribes or towns, and receive them from allies. According to Josephus the legionaries carried with them as part of their equipment sickles to reap the crops and also rations for three days. A scene on Trajan's Column depicts the legionaries carrying their kit on a stake; this consisted in part of a string-bag for forage, a metal cooking-pot and a mess-tin, examples of which have been discovered in most parts of the Empire. When an army was not on active service, the arrangements to supply food for men and animals were extensive and complex.

Recent theoretical debate combined with a greater attention to the nature of individual deposits in the archaeological record has focused on the nature of ritual behaviour, its material correlates and its significance as social practice. In southern Britain a particular focus for this debate has been presented by the study of ‘special deposits’ and the structured (ritual) deposition of archaeological deposits from the Iron Age. Cunliffe's review and interpretation of an extensive range of evidence from the hillfort of Danebury, Hampshire, can be set in the context of Hill's penetrating regional study of ritual in the Iron Age of Wessex. For the former, ‘special deposits’ or ‘structured deposition’ have been seen as the evidence of propitiatory rites, particularly to ensure fertility; for the latter, they are seen as a defining reflection of the ‘mediation, transformation and classification of people, animals, spaces and things’.

Severus…. was in the habit of saying that he had gained a large additional territory and made it a bulwark for Syria. But the facts themselves show that it is a source of continual wars for us, and of great expenses. For it provides very little revenue and involves very great expenditure; and having extended our frontiers to the neighbours of the Medes and Parthians, we are constantly so to speak at war in their defence.’ So writes Cassius Dio about the extension of the Eastern frontier in the 190s and the creation of the provinces of Mesopotamia and Osrhoene. The significance of the passage however, extends beyond the question of the Eastern frontier itself at that moment. Written by an ex-consul, and former assessor of Severus, it reveals two types of justification for conquest uttered by the Emperor himself - one straightforwardly imperialistic, the other strategic; and a critique of this from two points of view, the balance of income and expenditure, and the wider strategic commitments incurred. Whether Dio had formulated such views already in Severus’ reign we cannot know; this section of his History will have been written at the earliest towards 220, and probably later. If he had, we have no reason to think that he expressed them to Severus. If he did, it can only have been after the event, for his own narrative at this point makes clear that the new province of Mesopotamia was entrusted to an eques, and an ‘honour’ (the status of colonia) given to Nisibis, either after the campaign of 195, or (less probably) after that of 198, in neither of which Dio himself took part. None the less, the fact that the passage retails both the authentic views of an Emperor and a critique of them by a consularis may encourage us to ask some general questions: how, by whom and within what conceptual frameworks were the foreign and frontier ‘policies’ of the Empire formulated?

For the last twenty years or so, archaeologists of Roman Britain, among other provinces, have been seeking ways of moving beyond the concept of ‘Romanisation’ as a framework for thinking about Roman imperialism. Many of the ideas proposed have been drawn from two related bodies of thought which have emerged as ways of understanding the contemporary world: postcolonialism and globalisation theory. While achieving significant success in transforming interpretations of the Roman world, applications of these approaches present some fresh problems of theoretical and practical coherence. These in turn point to important issues to do with the role of theory in Roman archaeology, issues which have rarely been tackled head-on but which present obstacles to interdisciplinary dialogue. The aim of this paper is to evaluate and compare the perspectives of postcolonial and globalisation theories, assess their strengths and weaknesses, and suggest some possibilities for linking the insights of these and other approaches to define a more holistic agenda for Roman archaeology.

The large tracts of estuarine alluvium which border the Severn Estuary (FIG. I) and inner Bristol Channel have never been the subject of systematic archaeological enquiry. Now largely reclaimed and known as levels, the greatest of these tracts are the Somerset Levels, on the English side, and the Wentlooge and Caldicot levels between Cardiff and R. Wye on the Welsh shore. Smaller areas of reclaimed wetland range upriver as far as Gloucester. The manner and extent to which these wetlands entered into the economy particularly in the Roman period has hitherto been assessed only in the form of speculations or inferences from circumstantial evidence, in contrast to the firmness of our understanding, based on buildings as well as artefacts, concerning military activities, settlement, and daily life on the surrounding slopes and hills beyond tidal influence.

Approximately twenty temple excavations have yielded significant assemblages of animal bones. All come from Romano-Celtic temples in southern Britain, with the exception of four shrines for eastern cults. This paper picks out major characteristics of the assemblages and draws some general conclusions about the nature of the ritual activity that led to their deposition. At temples such as Uley or Hayling, sacrifices were probably an important part of the rituals, and the animals carefully selected. At other temples, animals had a lesser role, with little evidence of selection. At healing shrines, such as Bath and Lydney, animal sacrifices are not clearly attested, and would probably have taken place away from the areas used for healing humans. In contrast to the Romano-Celtic temples, animal remains at the shrines of eastern cults have very different characteristics: individual deposits can be linked to specific rituals within the cult buildings, and have many similarities to the continental evidence

The archaeobotanical record of Britain in the Roman period is reviewed. The data are plotted against area of the country, phase of occupation, type of site, and mode of preservation. Lacunae in the dataset are identified and research priorities formulated. More data are needed, especially from South-Western and North-Western England, Wales and Scotland, from major towns (especially from waterlogged deposits), from rural sites with waterlogged preservation (all parts of the country), and from burials and temple/shrine sites. Matters of concern are the identification of a downward trend in the average number of samples analysed from the 1990s onwards, and poor access to unpublished archaeobotanical reports (grey literature). Possible solutions to redress these are offered.

Fish bone assemblages from 109 sites were analysed for evidence of Roman influence on fish consumption. Temporal divisions within the period were not informative, but sites were divided by region. Secondary evidence, including amphorae and fishing tackle, was also considered. Eel was most common overall but some species were regionally important, e.g. salmon. Towns and villas showed the greatest range of fish, from freshwater and inland marine fisheries and also imported salted fish and fish-sauce. Native sites continued to show restricted fish consumption from very local sources, while Roman sites suggested an increase in variety and some evidence for status, which may result from cultural change in culinary practices.

An extensive Romano-British site, on the bank of the River Arrow at Alcester, Warwickshire, was excavated under the direction of Miss C. Mahany in 1964–65. The settlement contained, as well as habitation, what appeared to be an industrial area. In this area, with the remains of buildings, a pit was uncovered containing a muddy matrix in which was embedded a great number of scraps of leather. This so called ‘leather pit’ has been provisionally dated by Miss Mahany as belonging to the Antonine period, i.e. middle to late second century, or possibly a little later, on archaeological evidence. It was the infilling of this pit from which the insect fauna described here was recovered. The sample, collected by Professor F. W. Shotton, Drs. G. R. Coope and C. H. S. Sands and the author, was dug directly from the pit after the surface layers had been removed. The material was put into clean polythene bags and taken immediately back to the laboratory for processing. As the site of collection was well away from present-day buildings, the possibility of contamination by modern insects would seem to be remote.

The coins which form the basis for this discussion, summarized in Tables I to IV (facing p. 252), have been published in more detail in the Numismatic Chronicle (Southern France 1964, northern Italy 1971, northern France 1972) and in Britannia iii, 269-76, where full details of the nature, location and peculiarities of the groups may be found, together with a subjective estimate of each group's archaeological reliability. For ease of reference the full names of the groups, and the periods into which the coinage has been divided is repeated:

Poseidonios (c. 135—51 B.C.) has for long been regarded as the most important and influential source on the Celts writing before Caesar, and even as the most important ancient source on that subject without qualification. In order, however, to evaluate his contribution it is necessary to obtain an accurate reconstruction of his text, since no manuscript of his original work exists. Professor Tierney's monograph on Poseidonios's Celtic ethnography has for a long time been the fullest guide in English to this subject. It is valuable both as an introduction to Greek ethnography on the Celts and as a collection and translation of the main texts necessary for a reconstruction of Poseidonios's work. But it suffers from two major faults, which will be examined below. The first is the weakness inherent in the method of textual reconstruction by means of comparing texts thought to have drawn on the lost author; the second is the un-historical approach adopted by Tierney to his subject, which prevents him both from offering a realistic assessment of what Poseidonios could have known about the Celts, and from accepting Caesar as a source independent of Poseidonios. These two points of criticism of Tierney's article will be considered under separate headings.

It is over 30 years since Paul Buckland first presented a series of arguments concerning beetle (Coleoptera) grain pests: their origin, the timing of their introduction to Britain, and their implications for agricultural production during the Roman occupation. Here we return to the topic in the light of new data from a range of archaeological deposits, including civilian and military sites dating from the earliest period of Roman occupation. Infestation rates and, potentially, grain loss may have been high throughout Roman Britain, though many infestations may have been in equine feed. Beetle grain pests are not recorded in Britain prior to the Roman invasion, and it appears that they were absent, or extremely rare, in the early medieval period and up to the Norman Conquest. This pattern of occurrence is reviewed and it is suggested that ecological theory offers an explanation which is in accord with supposed socio-economic changes and trade. The role of grain pests is considered in the economic modelling of Romano-British agriculture.

The good survival of ceramics in the archaeological record has ensured that pottery has played a major role in the study of ancient industry and trade. How far Roman pottery can be used as a proxy for understanding the organisation of other industries is not a debate we wish to enter here, but it is clear that there remain important areas to be researched before the usefulness of pottery is exhausted. There have been significant advances in understanding over the last twenty-five years, with research concentrating more on characterisation, and the nature and scale of distributions, than on typology and chronology. The benefits of adopting a quantitative approach have been well rehearsed, although not always accepted. In this way, such research has made a major contribution to the ongoing debate about the extent to which pottery production and supply, and hence the economy as a whole, were controlled by market as opposed to institutional interests such as the state and, in particular, the Roman army.

So said Eric Birley in 1952. Since then, two major works by Callies and Baatz have appeared with relevance to numeri, but there has been no general article in English incorporating the discussion in these and previous publications. Speidel's article on ethnic troops filled the gap to a certain extent but he concentrated mainly on Moorish units.

Central to this paper is the meaning of the actions that lead to iron objects being found in archaeological contexts of later prehistoric and Roman date. It is argued that the placing of iron objects within the physical landscape reflects upon the changing nature of society at this time. In Iron Age studies, many deposits in rivers, bogs and in the pits, ditches and post-holes of settlements are now interpreted as ‘special’ material buried for significant reasons, through acts that are often called ‘structured deposition’. This approach has had a deep influence on the excavation and post-excavation of Iron Age settlements and is now coming to influence the study of the deposition of artefacts on Roman sites. This paper develops the idea that much of the later prehistoric and Roman ironwork found on settlements and elsewhere was deliberately deposited for what might loosely be called ‘ritual’ or ‘religious’ motives; for much of this period the proportion of the artefacts lost accidentally was possibly quite small. Artefacts in other materials also require comparable study, but, while work that integrates the examination of items made from different materials on individual sites is important, this paper focuses upon iron due to its potential significance as a highly symbolic medium.

The discovery of infant burials on excavated domestic sites in Roman Britain is fairly common but in the past these burials have often been dismissed as a product of unceremonious disposal. There is a growing literature which considers the phenomenon, but it has been dominated by debates around the suggestion that these burials provide evidence for infanticide, with a focus on the osteological evidence for and against this hypothesis. There has been less systematic consideration of the archaeological context of such burials. In this paper we examine the excavated evidence of two large groups of such burials from sites in East Yorkshire which demonstrate that the burial of neonatal infants followed a careful age-specific funerary rite. We suggest that this conclusion further undermines the widespread assumption that infants were disposed of without ceremony and as a result of infanticide.

Of the problems facing historians of Roman Britain few are so pressing, or so difficult to solve, as those associated with the Roman occupations of Scotland. The difficulties stem from the sparsity of contemporary or near-contemporary literary accounts, but they are exacerbated by the apparent lack of dated building inscriptions firmly assignable to the second Antonine period. Nevertheless, any assessment of Antonine Scotland must clearly start from the ancient historians, take into account all the available inscriptions and add the evidence provided by coins. The structural evidence for the forts has to be reviewed against this background. However, in recent years we have come to realise more and more that all these categories of evidence taken together are not enough to allow precise conclusions on some vital points. All this applies primarily to the Antonine period: the Flavian occupation indeed has its problems, but on the whole they are relatively minor ones, and do not involve widely different chronological interpretations.

This paper discusses the results of the first regional and bioarchaeological analysis of health in late Iron Age and Roman Britain. This tested the hypothesis that cultural and environmental changes in Dorset would result in changes to demography, stature, dental health and infectious disease. The study observed change to all health variables, supporting environmental and archaeological evidence for the introduction of urban centres, changes in living conditions, greater population movement, and development of the agricultural economy. Importantly, the study demonstrated that these responses did not reflect changes observed in other areas of Britain or Gaul.

This article reports a study of the distribution of Romano-British coarse wares in southern England. The term coarse wares, as used here, includes all wares except colour-coated or painted fabrics, samian and mortaria. These coarse wares were examined in the hope that a detailed study would provide some information about local marketing processes, or at least indicate what was the maximum amount of information that could be obtained from existing pottery-distributions by the methods employed. An area was chosen, the south and west, where a large number of site collections have been recovered. Pottery from as many sites as possible in museums and private collections was examined with the aim of finding exceptionally distinctive types whose distributions could be reliably plotted and whose percentages in each assemblage counted.