The Historic Towns Trust has been publishing historical atlases of British towns since 1969, with 13 towns and cities covered by the time of the previous atlas, volume III (1989). This book marks the revival of the great Historic Towns Atlas project and increases the number of published towns to 15, with the York volume (also 2015) and volumes on Winchester and Oxford (anticipated in 2017) set to increase the coverage to 18 towns and cities. The appearance and size of this Windsor and Eton ‘book’ mark it as something special: the huge volume comes wrapped in film, one pulls a Velcro tab in order to open the hard portfolio and thus reveal the bundle of maps, historic images and a data CD inside. And, yes, there is a bound book in there as well. The format is large, the quality of printing and production is high. A very clear introduction sets out the surprisingly long history of the project to map this rather unusual double Royal town. The bulk of the work is by David Lewis but there are also significant contributions by Giles Darkes (cartography) and Derek Keene (the Old Windsor chapter).

The text and maps need to be looked at together in order to appreciate the scholarship that has gone into the book (a table at least six feet long is useful here!). Lewis and Keene, in separate chapters, convincingly argue for a pre-Castle history of the (plural) Windsors: an eighth-century high-status Mercian settlement by the Thames at Old Windsor and, nearly five miles upstream, a later Anglo-Saxon settlement on a white cliff promontory overlooking the Thames (New Windsor). Keene's chapter on Old Windsor is a remarkable interrogation of the records of Brian Hope-Taylor's largely unpublished 1950s archaeological excavation. Keene argues that this was not (contra Hope-Taylor) a royal palace, nor a lost minster, but a substantial Mercian settlement, backed by royal or aristocratic investment, situated on the fringes of Mercian territory and ideally placed for trade with London and Kent, and for hunting. The Mercian settlement declined but was revived by Edward the Confessor; the establishment of a post-Conquest castle upstream at New Windsor can thus be seen as a Norman revival of a late Anglo-Saxon revival of a strategic Mercian settlement.

The earliest mention of the castle at (New) Windsor is in Domesday (1086) but a motte and bailey castle may have been built by the time that William I stayed here in 1070. The bulk of the book (and the accompanying maps) is an investigation of the town that grew up outside the walls of this royal castle from the twelfth century, a town which, throughout its history, has been dominated by the massive defended royal residence. The town began as a ‘T’ of burgage plots facing the castle, with that essential new town attribute, a market place, situated between town and castle. Remarkably, the basic shape and extent of the town hardly changed until the nineteenth century. Although the town was certainly dominated by the royal castle, Lewis investigates other urban characteristics including its trade associations and religious institutions. Lewis makes the point that this apparently small town was nevertheless in the ‘top 50’ mercantile towns of England (a fact masked by the separation of Windsor into its separate components in the 1334 royal subsidy). Henry VI's creation of a college of chantry priests over the river in 1441 created the double royal town of Windsor and Eton; the new St George's castle chapel (from 1475) added a further religious dimension to the town. The presence of the royal college, a local saint and a fragment of the True Cross attracted a steady year-round stream of visitors, rather better for trade than the episodic investment in the castle by successive kings.

There are plenty of surprises in the book, such as the fifteenth-century fraternity and guild building, Trinity Hall, elements of which survive today in the fabric of the Three Tuns public house. Another striking chapter sets the new town hall (perhaps designed by Christopher Wren) in the context of a late seventeenth-century urban revival, prompted by Charles II's renewed interest in the castle.

The bundle of fold-out atlas maps are a great success. The short essay by Darkes subtly describes the historical cartographer's careful balancing act between clarity and evidence. The atlas maps include both ‘vertical’ and ‘horizontal’ slices of time: Map 1 is a series of seven sheets at the generous scale of 1:2500 that piece together to form a large single map of Windsor and Eton, based on the late nineteenth-century Ordnance Survey map. A series of unobtrusive colour codes shows the dates of the (then-) surviving buildings, a patchwork of survival from the twelfth century onwards (the vertical slice through time). The horizontal time-slices are illustrated in a series of five clear phased maps (Maps 2 to 6), reconstructing the town(s) between the twelfth and the sixteenth centuries: these maps are the core of the atlas and form the visual counterpart to the century-by-century chapters. There are, perhaps, presentational details of the maps that one might question: this reviewer would prefer slightly more interpretation. During the process of historic map regression that created these maps, some of the straight lines of later roads were (rightly) smoothed to create more historical-looking rural lanes; the same should surely have been done for the Thames itself (the reconstructed twelfth-century Thames has some odd-looking rectilinear edges). And it might have been useful to have more interpretive geological and topographical maps (to illustrate chapter 1), rather than simply to reprint the British Geological Survey sheets. But these are small quibbles: this beautiful and stimulating work heralds the revival of the British Historic Towns Atlas series.