The photographs on which this account is based are part of the Collection of Aerial Photographs of the University of Cambridge.
Inventory of Caernarvonshire (Royal Comm. Ancient Mons.), vol. I (east) 1956, vol. 11 (central) 1960. Earlier Inventories covering the counties Montgomery, Flint, Radnor, Denbigh, Carmarthen, Merioneth, Pembroke and Anglesey were published between 1911 and 1937.
The references are to the National Grid.
‘The henge monuments of Great Britain”, in Excavations at Dorchester, Oxon, Atkinson et al., 1951, 81-107.
‘Blue-stone’ is a convenient term for the igneous rock used in the building of Stonehenge, but it hides the fact that at least four petrologically distinct varieties of rock are in question. Outcrops of all four varieties occur in a restricted area between Cam Meini (PLATE XXXVI (a)) and Foel Trigarn (PLATE XXXVI
Rare examples of prehistoric stone-axes in preselite show that the rock was also known and quarried for this purpose.
For an alternative explanation cf. Atkinson, Stonehenge, 1956, 173-175.
J.R.S., XLVIII, 1953, 85-87; XLV, 1955, 88; XLVIII, 1958, 95-97; LI, 1961, forthcoming.
Blaen-cwm-Bach, Twyn y Briddallt, Pen-y-coedcae and Pen-y-Gwrhyd were identified by officers of the Royal Commission on Ancient Monuments in Wales, Arosfa-Gareg by Mr J. F. Jones.
J.R.S., XLVIII, 1958, p. 94, fig. 8.
Tacitus, Annals, book XII, chapters 33 and 35.
This valley, larger than that of the Severn between Llandinam and Llanidloes, may well have been eroded by the Clywedog or even by the Severn itself: the river-system here has a complicated geological history and diversions have occurred.
The Tarannon valley, though wider than the Severn, has a flat, marshy floor, unsuited for mounting an attack.
The Breidden Hills and Long Mountain on the borders of Shropshire and Montgomery have often been linked with this battleground. But these hills lie in the territory of the Cornovii not the Ordovices, and there would be no need to force a river-crossing since the ridge can easily be approached from the east, while the Severn valley to west cuts off escape.
Arch. Camb., 93, 1938, 192-211; Nash-Williams, The Roman Frontier in Wales, 1954, 35-37.
Ordnance Maps, 25-in. scale, Carmarthenshire, sheet XXVII, 7 (1906 edition).
Trial-trenches dug by Mr Putnam in 1960 showed the rampart to be of turf.
By Thomas Price. Professor E. B. Birley first identified the camps as practice-works.
Recently recognized by Mr G. D. B. Jones, Bull. Bd. Celtic Studies, 17, 1958, 309-315. The practice-camp was found by the writer. J.R.S., 48, 1958, 96.
For details of the camps on Gelligaer Common, Mynydd Carn-Gôch and Stafford Common see J.R.S., LI, 1961, forthcoming.
Who also identified the camp south-west of Caer Gai.
Nash-Williams, , The Roman Frontier, 1954, 83–87
, figs, 41-42 and footnote 4 on p. 87.