This article is in part a response to a plea recently presented in this journal that the long-neglected druids be more fully acknowledged in our narratives of the Later Iron Age. It is also an attempt to give voice to an important, but frequently marginalized, example of discrepant indigenous experience in Rome's Western provinces.

In his discussion of the function of the druids in Later Iron Age Gaul and Britain, Creighton has argued that the influence of this social group was in decline by the mid-first century B.C. The druids' fortunes waned, he suggests, because their traditional role as mediators between the human and divine worlds came into conflict with emergent new power structures in the first century B.C.

In recent years, enough objects of a religious or votive nature have been found to suggest that there was a temple in one particular area of the Roman settlement of Great Walsingham with Wighton, in north central Norfolk. These include some important figurines, amongst them three quite different Mercury figures, two goats, one of them very well modelled, and three cocks. There is a portrait bust of Minerva and another of a three-horned deity, masks of Cupid, and representations of satyrs. Nearby, in separate contexts, were discovered a figure of Minerva and a mount in the form of a head of Jupiter. The votive material includes three rings with dedicatory inscriptions—to Mercury, and probably to Toutates, and to the Matres Transmarini; there is also a bronze letter ‘V’ used in a dedication, possibly even the first letter of V.S.L.M. There are six model objects—a tiny enamelled stand, two wheels from different moulds, a mattock head, an axe, and a spear; there is also a lead coin-mould, perhaps for making ‘temple money’ and the handle of a trulla. The finds imply polytheistic beliefs. However, the recovery of so many representations of the god Mercury suggests that here, as elsewhere in the Roman North-West, he was the most popular deity; the finds relating to Cupid and satyrs suggest the existence also of a cult to Bacchus.

The so-called ‘Child's Grave’ from Colchester is a unique assemblage and particularly famous for a large and diverse group of rare Central Gaulish pipeclay figurines (PL. X). The burial also contained a wide range of other Roman objects, namely ten unguent flasks, twelve pottery vessels, a clay picture lamp, two glass vessels, a bronze vessel, numerous bone and iron fragments derived from a funerary couch, and thirty-four coins.

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 village of Piercebridge in County Durham lies on the northern bank of the River Tees and on the route of the Roman Dere Street (FIGS 1 and 2). The village lies at c. 59 m aOD, sited on the river gravels of the Tees as it cuts through the surrounding Boulder clay. A Roman fort at Piercebridge has been known since at least Gibson's second edition of Camden's Britannia in 1722 and Roman Piercebridge is perhaps best known for the series of small excavations undertaken on that fort between 1934 and 1964. On this basis the fort could be identified as being essentially of later Roman date. Inhumation burials certainly or probably of Roman date have been recorded on a number of occasions.

Since 1856 when Sir James Simpson first asked the question ‘was the Roman army provided with medical officers?’ innumerable articles and books have been written on the subject. This paper does not intend to repeat the many points already raised concerning medicine in the Roman Empire but merely aims to assess the evidence for health care in the military zone of North Britain in the first four centuries of our era. Evidence from elsewhere in Britain and the Empire will be used only to clarify details, the geographical area covered being confined to the region north of Catterick and south of the Antonine Wall.

It has long been accepted that the Roman soldier was as likely to be engaged in craft or building activities as he was in specifically military ones such as fighting or drill practice. Vegetius listed the specialists that could be expected in a legion. They included builders of various types, people who could make vehicles, armourers, and weaponsmiths. The soldiers in auxiliary units also needed similar skills as the lists dating to the beginning of the second century at Vindolanda make clear. These gave details of the numbers of men engaged in different activities, and noted builders, shoemakers, and men working at some type of kiln or furnace. The archaeological evidence shows that the range of craftsmen was a large one. Potting industries were set up when the local potters did not provide the suite of vessels the soldiers required. Metalsmiths turned their hands to enamelled vessels as well as utilitarian items with a more obviously military function such as harness-fittings and studs. This paper will explore another possible facet of military craftsmanship, that of the manufacture and working of glass which is suggested by the recovery of glass-melting pots and partially reacted glass batch material at 16–22 Coppergate, York.

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.

During the late 1950s it was recognised that a succession of five rectangular, defensive enclosures, each astride the Watling Street ran from Uxacona (Redhill, Salop, SJ 72 10) to Tripontium (Cave's Inn Farm, Churchover, Warwicks. SP 53 79). The intervening enclosures were at Pennocrucium (Water Eaton, Staffs. SJ 90 10), Letocetum (Wall, Staffs. SK 09 06), and Manduessedum (Mancetter, Warwicks. and Witherley, Leics. SP 32 96)(FIG. 1). At the 1967 Roman Frontiers Conference, Webster referred to these as a chain of burgi, gave some details of their size and construction, and suggested a link with the restoration of order by Theodosius following the barbarian conspiracy of 367–368.

The report is focused on excavations in a suite of four rooms in the villa dating to the late third or early fourth century, associated with a plunge-pool excavated in the nineteenth century; the excavations were undertaken in advance of consolidation work. The rooms overlay a second- century bath-house also known from earlier work. The most important find was a hoard of fourth-century counterfeit coins which had been deposited in one of the rooms after it had gone out of use.

Pollen analysis has long been recognised as of key importance to archaeology for environmental reconstruction, but its application has concentrated on prehistory. This is, perhaps, due to a combination of two factors. First, there has been a greater interest by pollen analysts in questions relating to prehistoric than historic environments. Second, there are fewer suitable sequences available for later periods, especially from peat deposits where the upper layers have often been disturbed or removed by peat cutting. There is, however, an increasing amount of pollen-analytical data relevant to the first millennium A.D., and the purpose of this paper is to collate this information in order to illustrate the nature of the environment of Roman Britain.