Roman Piercebridge lies either side of the River Tees where three Roman bridges are recorded. The second one was associated with the original course of Dere Street and traces of an earlier one have been found upstream. The third was constructed downstream when the road was re-routed further east. The second and third ones were discussed by Fitzpatrick in this journal in 1999, and the volume under review publishes the evidence for the earliest one for the first time. It was to investigate these bridges that Bob Middlemass and Rolfe Mitchinson started diving in the river in the mid-1980s. Over the next 30 years they amassed a remarkable assemblage of Roman material from the riverbed including nearly 1,300 coins, over 1,600 small finds, 60 kg pottery and small assemblages of vessel glass and animal bone. This book publishes all this material and explores whether it should be regarded as the result of rubbish disposal or votive offerings.

Part I places the material in context. It starts with an overview of riverine assemblages across the empire and how they have been regarded. There are useful considerations here of how boundaries in scholarship have impeded our understanding of them. These boundaries are not just those between prehistorians and Romanists, but also between different national preferences for rubbish or ritual explanations and between specialisms with some categories of material attracting much more attention than others. This chapter will undoubtedly become the starting point of any future consideration of river deposition across the empire.

The Piercebridge context is then explored. Extensive excavation between 1969 and 1981 (published by Cool and Mason in 2008) revealed occupation starting in the first and continuing into the sixth century, with widespread evidence of intense military involvement starting in the later second century and continuing into the third. The material recovered during those excavations becomes a baseline against which the river assemblage can be compared. A section by Christopher Green considers the river and the mechanisms by which the assemblage could have remained in position, given it is prone to flash flooding. Detailed recording of the finds on the riverbed was not undertaken by the divers, but the records are sufficient to show that deposition was concentrated in the area of the earlier bridges.

Part II discusses all the material by functional category. Catalogue entries are available online in the Portable Antiquities Database. The discussions in the book are succinct, well informed and well illustrated. They will deservedly provide a happy hunting ground for finds specialists. Throughout this part comparisons between the riverine and land assemblages are made, and these clearly bring out differences between the two. For all categories of material late Roman finds, which are strongly present in the land assemblage, are much rarer in the river finds. A case is made for river deposition being stronger in the first to second centuries than it was on land. This is certainly evident for the coins (fig. 10.2) but is less convincing for the small finds (e.g. the brooches, fig. 4.3) and the pottery. An interesting aspect of the coin assemblage is the degree of mutilation and defacement seen in the river assemblage, a phenomenon possibly also seen in the jewellery and metal vessels. This is an aspect which cannot be meaningfully compared to the land assemblage as, 40 years ago when those reports were written, such aspects were not routinely recorded or discussed.

Part III draws all the patterns together to show how different the river assemblage is both to that from the land and from neighbouring sites such as Catterick and Binchester. Here the fact that the total assemblage has been studied, and not just selected ‘interesting’ aspects of it, shows its strength. In part of this discussion the authors are joined by Kris Lockyear to produce elegant and clearly explained correspondence analyses to show these differences.

This is a most interesting book, both from the methodological point of view and for what it tells us about Piercebridge and questions it raises. Why did the river cease to be the focus of votive offerings at some point in the third century? Occupation continued and by then was concentrated very close to area of deposition. The fortunate combination of the extensive excavation on land, the devotion of the divers over three decades and the existence of the Portable Antiquities Scheme to record their finds may never happen again, but this book shows the potential of river finds. The authors promise us they are preparing another one looking more widely at such deposition in Roman Britain, and this is something we can all look forward to.