Silchester was a key town in the Roman province and its exploration both mirrors the development of the study of Roman Britain and provides a lens through which successive generations have understood the province. It is all the more curious therefore that since George Boon's synthesis of the evidence, which was published in its second and fuller edition in 1974, there has been no new full-scale reappraisal of the site. This is despite the key and extensive excavations undertaken by Mike Fulford and his team. The present volume changes this, providing a thorough and up-to-date re-evaluation of the site. As such John Creighton has provided the key new work of reference for Silchester.

This is achieved through the unifying thread of a comprehensive geophysical survey which covered the walled area and an extensive area beyond (a total of 217 ha of gradiometry, complemented by more focused areas of electrical resistance and ground-penetrating radar survey, as well as all the available LiDAR data). However, the product is far more than an account of those surveys, as it uses them to draw together the results of all previous work to provide a comprehensive discussion and careful reappraisal. This includes work done before the well-known campaign of excavations sponsored by the Society of Antiquaries (1890–1908), as well as since. This very thorough analysis of past work includes a comprehensive replotting of the buildings found by the Antiquaries and recorded on their great plan that has so-often been reproduced over the past century. The core of this study is contained within the mapping section (Part II), which is organised around an atlas that deals with the evidence from each part of the site comprehensively. This has been achieved in an extremely effective manner with the thoughtful and carefully designed integration of illustrations (showing past evidence, the geophysics results and their interpretation set against a topographic background). It is thus easy to use and forms a fundamental work of reference. Furthermore in an innovative addition, the digital archive (doi: 10.5284/1038434) held by the Archaeology Data Service contains, among other things, a ‘zoomable’ layered map of all the results in .pdf format, enabling the average reader fully to explore the evidence for themselves. As such the core of this volume and its integral digital archive set a new standard for the publication of the new generation of surveys of Roman urban sites.

This book takes the study of Silchester far beyond the presentation of a comprehensive and critical edition of the evidence, for in the second half of the volume C. provides a series of essays that offer new interpretations and fresh perspectives on some broader issues. This discussion is very well founded in a broad and up-to-date understanding of the archaeology of Britain and the north-west Roman provinces. While not offering a comprehensive account like that provided by George Boon, the discussion is immensely important and impressive. Part III focuses on the defences and surrounding earthworks, providing a thorough new understanding of their development. Part IV moves on to consider a broader range of themes. Among the most important are C.’s discussion of the town's origins, military involvement and the cemeteries. The re-evaluation of the role of the Roman military is particularly refreshing, moving debate away from the familiar ground of fort/no fort, to offer a more nuanced perspective. We may also note his tentative but convincing suggestion that there was a circus just to the north-west of the town (427–9).

Some of C.’s innovative use of the geophysical survey evidence to contribute to broader debates is also methodologically significant. First, he has been able to identify various different cemetery areas. Those seen as high-status Iron Age burial enclosures are important for our understanding of the development of the site, but are not methodologically significant. By contrast, his success in mapping the extent of large inhumation cemeteries and potential cremation pyre areas around the town is very important as few other surveys have achieved this. Second, his use of magnetic signals associated with burning to map areas of industrial activity provides a new approach to the use of extensive urban survey data which may add a new dimension to our understanding of the economic dynamics of urban sites.

It will be clear from my comments that this is an immensely important book that makes a key contribution to the study of Roman Britain. C.’s ideas offer much scope for discussion, but it is clear that this volume places all future debate on new and secure foundations.