Alan Wilson has examined the effects of Roman occupation between the Walls for over 20 years, through a series of regional studies of Strathclyde, Galloway and Dumfriesshire published in the Glasgow Archaeological Journal and the Dumfriesshire and Galloway Society's Transactions. In this volume his focus moves eastwards to an area traditionally equated broadly with the territory of the Selgovae, perhaps best known for Roman Newstead and the nearby indigenous hillfort on Eildon Hill North, and the ambiguous nature of the relationship between the two.

The stated aims (2) are to ‘provide a perspective on native culture, lifestyle and economics, the impact of Rome on the native peoples of the Central Scottish Borders and the effect of Roman withdrawal’, based primarily upon an examination of a selection of artefacts (Roman, indigenous and potentially of either origin) recovered from Roman and indigenous sites, including previously unpublished material. This work then follows in the footsteps of James Curle, who excavated at Newstead a century ago and was the first to publish an inventory of Roman artefacts from non-Roman sites in Scotland. Conversely, it precedes the final publication of the Newstead Research Project investigations of the years around 1990 at Newstead fort and Iron Age sites in the surrounding landscape, the results of which are likely to have much to add to this topic.

W.'s general approach will be familiar to those aware of his earlier work. A substantial review (I estimate 45,000 words) is followed by an inventory of the finds with summary tables, supported by a location map, a selection of reproduced plans of many of the better-known Roman and Iron Age sites of the region, a selection of artefacts drawn by Marion O'Neill and Alan Braby, and an extensive bibliography.

This is a thorough study underpinned by a prodigious amount of research. The review is wide-ranging, placing the Roman presence within a view of the region spanning much of the first millennia b.c. and a.d., a sensibly broad range given the continuing difficulties of dating accurately many of the Iron Age site types discussed (e.g. linear boundaries and field-systems, cord rig). The evidence — archaeological, artefactual, environmental, epigraphic and literary — is examined through themes including political geography, socio-political structure, economic structure, Celtic art and Celtic religion. While this approach brings undoubted practical benefits, this compartmentalisation of the evidence for indigenous society was arguably provided at the expense of presenting a holistic view of its character before, during and after the Roman occupation. The latter part of the review consists of a chronological overview of the Roman occupations and aftermath, within which the regular juxtaposition of historical narrative and specific information about artefact findspots and sites highlights the difficulties of marrying archaeological and historical evidence within a coherent interpretative framework.

The interpretations of previous researchers, not infrequently contradictory, are presented throughout; sometimes to such an extent that this reviewer was uncertain what W. himself believes. Perhaps as a result, the approach to understanding what the material can tell us about the relationship between Romans and indigenous peoples tends to follow existing, familiar models. Nevertheless, there are certainly controversial statements, as for example (46), ‘The larger native settlements such as Eildon Hill North were presumably the equivalent of towns in southern Roman Britain but the majority of the native population in the Scottish Borders may only have frequented them from economic necessity preferring country life to a more urbanised way of living.’ There is presently no good archaeological evidence to support indigenous Borders settlements as urbanised, and I am unconvinced that indigenous inhabitants thought in terms of rural/urban or economic necessities.

The artefacts are comprehensively catalogued and provided with parallels, but the section could have been tabulated to make consultation easier. It would also have been helpful had W. provided information about what elements/proportions of site assemblages are not published here. Nevertheless, the Inventory should form a useful resource and stimulus for future researchers, as the author intended, but the supporting summary tables referred to in the review will be of more use for the general reader. Overall this volume provides a detailed reference account of what we know and what we think we know about the subject, and allows us the space to consider where our researches might take us next.