As is frequently the case with the BAR series, this book started life as a doctoral thesis. When it was published, the reviewer was in the process of checking the short summary of shipping that was being co-ordinated for the chapter on transport in the new Roman Frontiers in Wales and the Marches (RCAHMW, 2010). Not unnaturally, it was expected that revision would have to be made as a consequence. This was not the case. The question is: why not?

The title is poorly chosen: it should be Maritime and Riverine Transport in the West of Roman Britain. A book on the landscape ought to pay more attention to the difficult question of the changes in the coastline that have occurred since Roman times. In his chapter on changes in the coastal and riverine landscape, Jones gives us under three pages on coastal change, none of which adds anything new to our knowledge of changing tidal levels in Roman Britain. Isostatic change, the rise in sea level driven by ice melting after the last glaciation, is briefly introduced, but the complementary phenomenon of eustatic change, the readjustment of the land to the loss of its ice cover, appears only as ‘the land surface is also changing in different directions, and in different parts of the country’ (8). Given that the south of the country is rising and the north sinking about an axis that runs approximately through Anglesey and Liverpool Bay, this important concept, which needs to underpin any discussion on how sea level change is likely to have affected different areas along the length of the west coast of Britain, is totally missing. Nor is there any consideration of how local geology and topography will have affected the relative positions of the Roman and modern coasts.

However, on the assumption that J. did in fact set out to write a book on sea and river transport, the reviewer will concentrate on assessing how well he has achieved this. The introduction sets out the themes to be covered in subsequent chapters. Following Changes in the Coastal and Riverine Landscape (ch. 1), these are: Ships, Coastal Vessels, Boats and Local Craft of the Period (ch. 2); Supply and Provisioning of the Roman Army (ch. 3); Roman Sea Power and the Classis Britannica (ch. 4); Naval/Military Operations in the West of Britain (ch. 5); Development of Rivers, Ports and Landing-places (ch. 6); Cargoes within the Province (ch. 7); Imports and Exports (ch. 8); Cargo-carrying Capacities (ch. 9); Waterways and Roads (ch. 10); The Size of the Population (ch. 11); Demand, Supply and the Volume of Shipping (ch. 12). With this programme, the book's origin as a doctoral thesis, and the fact that the author is an experienced sailor, this reviewer expected a comprehensive coverage of all aspects of water transport and their relation to the military and economic history of Roman Britain, examined in new ways and informed by the understanding that comes from a writer with a full practical knowledge of his subject. However, disillusion soon set in. The focus on published material that characterised the first chapter is replicated again and again — and there is nothing that could not be found within the walls of a university library. There is no evidence that J. made any attempt to tap the extensive resources of grey literature, local short-run and antiquarian publications, and local topographic knowledge available in any of the English county or Welsh regional Historic Environment Records. Selection and arrangement of material, and the use to which it is put, are also poorly thought out. In the section on archaeological evidence of harbours (51ff.) Wroxeter, where no harbour has been located, is included whilst Heronbridge, where river dock facilities have been found, is omitted.

The biggest disappointment is that J. makes very little use of his experience in sailing these coasts to throw new light on our understanding of the archaeology. He seems to be entirely constrained by limitations of what he has read, and has not been able to give it the extra value that his background should have equipped him to provide. While he is undoubtedly correct in pointing out that ‘it is unwise directly to relate modern craft and weather conditions to those of the Roman period’ (1), this ought not to have prevented him from digesting the library material and showing us what it means from the point of view of the practical sailor within the context of different models of wind, current, shoreline and craft performance. There are a couple of useful and illuminating pages in ch. 10, where information is provided on such matters as passage times around Land's End or the distance out at sea from which a coast is visible, but the overall impression is one of missed opportunity.

It is unfortunate that these shortcomings are compounded in too many places by careless mistakes. I noted in passing, from my own part of J.'s study area, the misinformation that the Roman name for Abergavenny is Burrium (Burrium is Usk — correctly on the previous page — and Abergavenny is almost certainly Gobannium); that the river at Neath is called the Nid (it is in fact the Neath); and that the charred bread wheat excavated at Loughor was non-local and must have come from the West Country (the excavation report states categorically that its origin cannot be determined and it could be either local or an import). The organisation of the gazetteer is confused, not helped by the fact that some of the subheadings have run into the text above. Some forts are grouped under foundation date, and others by their length of occupation, and with the entries for towns on earlier fort sites unhelpfully given separate entries. Where the book could potentially be useful, as a compendium of previously published material, it is undermined by sloppy referencing. Publication dates for authors too often fail to match with dates given in the bibliography, and there is a complete lack of references in the gazetteer. It is not clear how much time elapsed between when it was submitted as a thesis and when it was published: what look like updates in the bibliography that have failed to make it into the text, as for example the definitive publication of the Barland's Farm boat, suggest that this may have been some years, but that updating was only half-hearted.

On the plus side the writing is literate and properly punctuated, virtues that one cannot unfortunately take for granted these days. In short, the book promises much, but fails to deliver and we still wait for a comprehensive treatment of navigation in Roman Britain.