Welsh Journals

antecessores and held royal manors at farm. Unlike Gerbod the Fleming and Hugh of Avranches after him in Cheshire, and Roger of Mont- gomery in Shropshire, William was not given the entire county in chief. Lewis further demonstrates that fitz Osbern was not granted lordships in southern Shropshire, which forms a natural defensive unit with Herefordshire. This is important, for it indicates that the roots of the Marcher lordships of the Middle March do not lie in the tenurial organization of Herefordshire. Sir Goronwy Edwards's thesis, that Marcher powers derived from the Normans' assumption of the regal lordship which came to them when they conquered Welsh commotes, has been challenged and qualified in the past few decades.7 There is no single, simple explanation for the development of the Middle March. One feature, however, which may have helped to shape the develop- ment of the area can be seen in the settlement pattern of Herefordshire. Amongst all the border counties, Herefordshire was sui generis. The county was marked by dispersed settlement; even by the nineteenth century, there were only six nucleated settlements in the eastern half of the county.8 But, the closer one comes to the militarized zone of the March, the more likely it was that settlements would be nucleated for protection. In the rather bald description of historical geographers, the dispersed pattern of settlement in most of the county goes back to the period before 1066, while the nucleated pattern in the west is largely the creation of the Normans.9 This same pattern holds true for south- ern Shropshire. It is in the west that we find the classic Norman triad of castle, monastery, and borough. Smaller in scale than the comtes of Normandy or the large compact groups of manors held by subtenants in Cheshire and Shropshire, these castellaria are described in Domes- day as existing at Wigmore, Clifford, and Ewyas Harold.10 The cluster 7 J. G. Edwards, 'The Normans and the Welsh March', Proceedings of the British Academy, 42 (1956), 155-77; for an important qualification of the Edwards thesis, see R. R. Davies, 'Kings, lords and liberties in the March of Wales, 1066-1272', Transactions of the Royal Historical Society (TRHS), 5th series, 29 (1979), 41-61, esp. 8 Dorothy Sylvester, The Rural Landscape of the Welsh Borderland (1969), pp. 9 Ibid., p. 360. 10 F. M. Stenton, The First Century of English Feudalism (Oxford, 1932; 2nd ed., 1961), pp. 194-6; David Bates, Normandy before 1066 (1982), p.156; John le Patourel, The Norman Empire (Oxford, 1976), pp.308-9.