Reading the publication date of this work, and its BAR edition number, it seems amazing that a detailed exploration of this topic has evaded us for so long. It was in 1978 that Henry Cleere (Roman harbours and landing-places south of Hadrian's Wall, in J. du Platt Taylor and H. Cleere (eds.), Roman shipping and trade; Britain and the Rhine provinces, CBA 24) drew attention to a modest number of recognizable landing-places on Britannia's coastline. This was prompted by the course of certain Roman roads, some of which were aligned on virtually anonymous coastal destinations. Since that time, the siting and function of the forts of the ‘Saxon shore’ have attracted interesting debate on landing-places (A. Pearson, 2002, The Roman shore forts: coastal defences of Southern Britain. Oxford). We have also seen a general and well-considered review of Roman Britain and the Roman Navy (Mason, 2003. Oxford). Both studies include some mapped reconstructions of past coastal and riverine configurations. While a general review of Britannia's coastal villas and maritime villas (D. J. Tomalin, 2006, ‘Coastal villas, maritime villas; a perspective from Southern Britain’, Journal of Maritime Archaeology 1, 29–85) drew particular attention to the navigational aspects of the bays and estuaries of the Romano-British coast, where the siting of certain of villas appeared to be largely led by maritime considerations. From Henry Cleere's 45 perceived landing-places, the number was now advanced by 124.

In this current work Dr Jones carries this topic much further. Here, at last, is a well-synthesized appraisal of both the maritime and riverine routes that once conveyed all manner of goods to and from Britannia. This study takes us beyond the mariner's primary concerns with safe anchorage and tides, and introduces the practical considerations of stowage capacities, cargo weights, the sourcing of commodities, patterns of demand, and the economics of shipping to the disparate markets of the Romano-British and Gallo-Roman world.

In his introduction to this work, the author sets out his approach and his credentials. In the first instance, this study is something of an offshore view, taken by someone who is certainly familiar with sailing conditions and navigational considerations. It is, in short, a fresh approach, and in the author's own words it ‘diverges from the well-beaten track of Romano-British history as, for example, it argues that ‘the perils of Land's End’ and ‘the dangers of the west coast route’ are much exaggerated, and there is considerable evidence that voyages under sail round the coasts of Britain, and in particular the south-west coast, are not the difficult undertaking suggested by some writers’.

So here, at the start, we can expect some new thinking, and we are not disappointed. ‘Sea-keeping’ and ‘passage-making’ are topics that the author promptly introduces while he also explains that vessel types, and their origins and methods of construction, should not detain us. In this study it is sufficient to know that craft like those found at Blackfriars and St Peter Port, Guernsey, as well as the described ships of the Veneti ‘have much in common, in terms of design and construction, with the medieval cog’. Once we acknowledge this simple analogy, we can join the author in weighing written accounts of medieval shipping against our woefully weak contemporary record of Romano-British seafaring.

In Chs 5–7, the book gives thorough attention to the military aspects of coastal and riverine sailing. Literary sources have been well trawled, and they include a reminder that ‘Agricola started his fifth campaign with a sea passage, and in a series of successful actions subdued nations hitherto unknown’. Here we see a naval force successfully acting in its own right, without essential support from ground troops. Where a goodly number of islands, headlands and rivers on the west coast of Scotland have acquired Roman names, the author suggests that ‘Roman sailors were well aware of the opportunity for navigation in these northern waters’. Elsewhere in his text, the author draws attention to the anomalous status of the Dumnonian peninsula, where the well-ordered civil settlement of southern Britain gives way to something that archaeologists have always struggled to define. In this instance, Dr Jones draws our attention to the Cornish coastal forts at Nanstallon, Lostwithiel and Calstock, while inferring that the population of the south-west peninsular, and the handling of its vital mineral products, were best managed by simple military supervision at the harbour-side.

As we progress into Dr Jones’ book it is important to remind ourselves that the word ‘shipping’ is not in the title. This is an assumption easily made. Essentially, this book is a study of routes and landing-places, markets and hinterlands. Once these are viewed within their geographic framework, we immediately find that there are obstacles set in both the past and the present.

For the military planners and commercial opportunists of the Romano-British period, problems were often posed by difficult navigational access to ports and by the varying usefulness of navigable rivers. Rivers, after all, offer no more than opportune routes to places that may, or may not, be rewarding to visit. Once its bridge had been installed, London was a strategic choice that was bound to succeed. Gloucester and Chester both owe their siting to rapids that force a transshipment of cargoes to and from upstream craft. In the territory of the potentially troublesome Brigantes, we find the garrison at York sited at a point that might be conveniently provisioned via a riverine link with the sea.

For archaeologists dealing with today's settings of Romano-British maritime sites, there are inherent uncertainties concerning the scale and pace of ongoing coastal processes, sea-level change, and behavioural shifts in our rivers. Examples of these changes, cited by the author, are evocative enough. The silting and loss of the Wantsum Channel; the reclamation of the Wash wetlands; the sea's destruction of Walton Castle; the accretion of Romney Marsh; the engulfing of Londinium's waterfront; all offer images that emphasize the chronological chasm between then and now.

Where rivers could channel cargoes to and from Britannia's interior, the author helpfully lists 18 principal routes that were demonstrably favoured. Of these, the Thames, Severn and, presumably, the Trent offered the longest lines of communication. Medieval records of Thames navigation present some surprises when we learn that, without the locks to which we are now accustomed, grain-laden craft could still begin an unimpeded downstream journey from a point 35 km upstream of Oxford. By the 15th century, however, upstream navigation had retreated to Reading. In figure 9.8 we are shown a map of the riverine routes of medieval England and Wales, with an indication of their relative use. By this period, a number of Roman towns and forts had become county towns. Here, one wonders just how much of this later riverine traffic had been predetermined by choices of settlement and development, made in the first century AD. A further point to ponder is how much of the archaeological evidence of riverine landing-places may now lie in the path of the navigational dredger?

In ch. 3 we meet with an array or riverine and sea-going craft. On the Oceanus Britannicus and the Oceanus Germanicus, it seems that high-bow craft, similar to the Blackfriars/St Peter Port vessels, offered the best stowage capacity, with an optimum tonnage, perhaps not exceeding 72 tons. The author uses a helpful formula to calculate relative quantities of wine barrels, grain sacks, ragstone or amphoras that might be carried in craft of this kind. Some notable uncertainties arise where freeboard safety practices are unknown. When all currently known factors are taken into account, it is suggested that an average merchant ship in British waters may have had a cargo-carrying capacity of some 60 tons and a hold capacity c.33 cubic metres.

In ch. 8 we come to the heart of Dr Jones’ subject. Here is an appraisal of specific harbours, ports and landing-places. Barry, Caerleon, Chepstow, Chester, Dover, Gloucester, Heronbridge, Ilchester, Lincoln, Littlecote, London and Plymouth Sound are all accorded particular attention. This is certainly a curious heterogeneity.

Of all of Britannia's noted coastal and riverine rural buildings, Littlecote, and its attendant stream, in the heart of the Wiltshire chalklands, must by one of the least convincing contenders. Ilchester's credentials hang on an uncertain claim to a canal, yet we find no mention of the neighbouring area of Langport, where a cluster of villas attend the former riverine margins of the Somerset Parrett. Here the villa of Low Ham, with its ship images in a mosaic scene of the voyage of Aeneas, seems a surprising omission. We should certainly consider whether its estate reached the bank of the Parrett.

Another omission is the chain of villas attending the navigable Medway, an estuary later noted for its prolific production of wooden ships. The Solent's array of coastal and maritime villas also awaits another day. This includes the well-known villa at Brading and the great natural haven that formerly attended this residence. A mosaic panel in its principal room exhibits an explicit maritime scene, proclaiming the owner's seafaring interests.

Given Southampton's reputation as a Saxon and medieval port, it would have been good to see some discussion of Romano-British shipping in the environs of Southampton Water. Here, the Vespasianic lead pigs from the bed of the Itchen and the submerged timbers of a Roman waterside revetment are worthy of attention. The identity of the enigmatic Magnus Portus cited by Ptolemy, has also escaped discussion. Now that anchorage strews of fragmented amphoras and other Roman ceramics have been recovered from the floor of the Eastern and Western Solent, the nature of this ‘Great Port’ and its role in the maritime landscape of Britannia call for re-appraisal. Ptolemy's other named Romano-British anchorage, generally translated as ‘Safe-haven Bay’, and attributed to Bridlington Bay, also deserves attention.

While virtually all reviewers will raise a point or two of this kind, these perceived omissions are best seen as a matter of academic preference, and they should not detract from the sound scholarship of this book. If any criticism is to be made, it must be applied to matters outside the control of the author. In the presentation of the book, the relatively small font of the contents list and its printing on a verso page is something of an excessive paring. The chapter titles introduced in small type on verso pages are equally eccentric and there is an annoying absence of chapter numbers from the page headers. A few odd spellings like ‘Carrera’ (Carrara) marble and ‘Hadrianatic’ (Hadrianic) are, perhaps, artefacts of typesetting. On the positive side, the generous provision of sub-headings is most helpful. For future reference, one wonders whether readers might have been still further assisted if these, and the inventory entries, had been numbered as well. This inventory briefly describes 198 coastal and riverine landing-places that beg further archaeological attention.

Is a pity to find that the publishers have impaired the author's work by the poor quality of figure reproduction we have become accustomed to enduring in BAR. In figure 2.5, the map of Roman roads in Britain is almost unintelligible and the final fuzzy map, showing the relative use of roads noted in the Antonine Itinerary, simply leaves us rubbing our eyes. The reader will also find the appendix maps unnumbered, a potential impediment to future citation. Despite these niggles, it must be said that this book can never lapse into a shelved volume, usefully read just once. Dr Jones’ coverage of his subject is thorough, wide-ranging and perceptive. Above all, it offers counterpoise to any misapprehension we may have acquired while wading through so many standard works in which the riverine and estuarine dimensions of Roman Britain have been virtually ignored. The author deserves our thanks for a discerning work that is sure to become very well thumbed.