REVIEWS LANDSCAPES OF BRITAIN: THE WELSH MARCHES. By Roy Millward and Adrian Robinson. Macmillan, 1971. Pp. 204. £ 3.50. The new series of Landscapes of Britain deserves a very warm welcome. Each volume covers a particular area, The West Midlands, South East England, The South West Peninsula, and the volume under review, The Welsh Marches. An introductory essay deals with the physical and human setting of the Marches, and provides a study of the historical geography of the whole area, roughly from Wrexham to Chepstow, down to the end of the middle ages. The second part of the book consists of twelve regional studies analysing the peculiar characteristics of such widely differing areas as the Vale of Llangollen, the Forest of Wyre, the castle-towns of the middle March, and the lower Wye Valley. This second part has also been published as four separate booklets, each of about thirty pages, so that those who wish to use the series for local field work may buy at compara- tively low cost the section of immediate importance for their studies. The text is very readable, and a careful balance has been maintained between physical geography and historical geography. The maps are clear and easy to use, and the book is profusely illustrated. Particularly good use is made of aerial photographs. They have a keen sense of history, and can handle a wide range of archaeological and architectural evidence, and they deal with medieval records with skill. The authors are concerned primarily with the English lands of the border. Their purpose is to show the evolution of the landscape, and readers of the WELSH HISTORY REVIEW will probably find the greatest interest in the way in which man has responded to his environment and, in the process, moulded the landscape, whether by building the earth- works of Iron Age hill-forts, or by planning new towns in the twelfth and thirteenth centuries. Cefnllys and New Radnor were planned as new boroughs, but like Grosmont or Richard's Castle they failed and became, not thriving towns, but sleepy villages. By contrast, such places as Kington, Montgomery, Bridgnorth and pre-eminently Ludlow were successful plantations. Ludlow is essentially a medieval marcher borough. Great Malvern, on the other hand, was the product of the eighteenth and nineteenth centuries as a spa where invalids could take the waters, or try Dr. Wilson's famous hydropathic cure. The authors have an eye for detail, but they never forget the basic control which 'the rocks' exert over environment and development. The Breidden hills, for instance, may be 'one of those distinctive small areas which lend character and diversity to the border landscape', but 'from a farming standpoint, the Breiddens represent an island of poverty in a sea of plenty'. Historians may be quick to see the import of place-name and settlement, of castle or town, but may still have much to learn about the structural basis of the land- scape, and the relationship between geological formation and human