Welsh Journals

entrance, however, appears to be marked out as of more importance by the existence of a slightly sunken way, flanked by traces of a low bank on the west side, running out from it northwards for 60 yards, downwards in the direction of a rivulet upon which the ravine to the east of the site opens (Plate VIA). The outermost bank on the west of the site joins the northern end of the bank flanking this sunken approach, and there are remains of two cross-banks to the east of it, which appear to defend a small outer enclosure in the angle formed by the north-east defences of the 'Gaer' and the ravine. Although the description already given will have suggested that the 'Gaer' is tactically sited and defended, it must be emphasized that its position is by no means a commanding one. While the ground falls away appreciably on the north and west towards the bottom of the Llynfi Valley just above its junction with the Wye at Glasbury, the plateau on which the 'Gaer' stands is less than 100ft. above the bottom of the main valley at Glasbury, and is commanded by higher ground on the opposite side of the ravine less than 100 yards away. The approach from the south is almost level. What appeared before excavation, at any rate, to be the main entrance, with its outwork and 'barbican', points down slope towards the rivulet which flows into the Llynfi nearby and the general area of the Wye valley floor. None the less the 'Gaer' clearly resembles, in a number of respects, the hill-forts of proved or probable Early Iron Age date which are so numerous in the Marches of England, Wales and the western regions of Britain generally. Above all, it resembles the very numerous class of small embanked enclosures in South Wales, which have multiple banks and complicated entrances recalling those of large hill-forts. A site particularly apt for superficial comparison is the fortified settlement on Mynydd Bychan, Llyswomey (Glam.), which I excavated in 1949-5 o.4 This earthwork, though of even smaller compass than the Aberllynfi 'Gaer', had three ramparts on at least one side, and an entrance approached by a narrow passage, flanked on the south by a wall and ditch, and on the north by a wall joining the ends of three ramparts. This 'barbican' type of entrance is, of course, characteristic of large hill-forts, as is the bridge linking the ends of the parapet walks in front of the entrance, the post-holes for which were found at Mynydd Bychan.5 A comparable arrangement, on a much larger scale, is likely to have existed at the large hill-fort on the hill above Llanbleddian (Glam.) not far from Mynydd Bychan.6 Here the entrance on the western side not only lies at the end of a long passage, flanked by multiple banks on the south, but has outside it, on the north of the passage, a small 4 Arch. Camb., 1949, pp. 85ff; 1950, pp. 14fT. 6 Arch, Journ., 1948, pp. 56ft. 6 Arch. Camb., 1936, pp. 2off.