Welsh Journals

work, The Shrewsbury Drapers and the Welsh Wool Trade in the XVI and XVII Centuries. DAVID WALKER Swansea THE ROMAN FRONTIER IN WALES. By V. E. Nash-Williams. Second revised edition by Michael G. Jarrett. University of Wales Press, Cardiff, 1969. Pp. 206, 14 pis., 90 figs. 84s. It was inevitable with the rapid growth of knowledge that the masterly survey by the late Keeper of Archaeology in the National Museum of Wales, published in 1954, should become out-dated. This second edition was aptly timed to coincide with a visit of the Twelfth Roman Frontiers Congress, and both Dr. Jarrett and the University Press are to be congratulated in getting it out in time despite difficulties. It is a totally revised edition with an enlarged format. The general arrangement of chapters remains the same but there is a much-needed expansion of the historical chapter. Although the total number of forts remains the same at thirty-eight, there has been much reshuffling with new ones like Whitchurch, Walltown, Beulah, Caerphilly, Coed-y-caerau, Discoed, the Leintwardine complex, Pen Llystyn and Trawscoed. Some of the sites originally included as forts are now relegated to a section on fortlets and include Cae Gaer, Carno, Llanfair Caereinion, Bryn Glas and Caer Gybi. The older identifications now rejected are Prestatyn (which still remains a problem), Varae (yet to be identified) and Tremadoc (where the bath-house is now considered non-military and omitted); Castle Flemish is also now considered to be civil in character and Cwmbrwyn is rightly regarded as a fortified farmstead. Loughor is also out, but with the new discoveries of 1969 it will have to be put back again in the next edition. All the plans have been re-drawn and metric scales added. In many cases there is a great deal more information, thanks to the work of Dr. Jarrett himself and Dr. G. D. B. Jones, as at Carmarthen, for example. The overall effect of this new information, here so ably summarized, is that of growing complexity. Nash-Williams was a very cautious scholar and rarely allowed himself the privilege of speculation. This has given the impression that the network of forts standing out as red squares on his map was rigidly imposed by the Romans and remained the pattern of control only with minor modifications. Dr. Grace Simpson had given us a foretaste of the truth in her Britons and the Roman Army, in which she critically reviewed the pottery evidence, clearly demonstrating a great deal of change. We now see that it is even more complex. Most forts have several periods of occupation but not necessarily on exactly the same site, as is now seen at Leintwardine and Caersws. This new edition should provide the platform from which an attack can be launched on a selected number of problems. It is clearly no longer