Welsh Journals

the bank east of the Small Enclosure (E-F, Fig. 7) was found a layer about 4 in. thick of occupation material, containing some crushed and burnt bone, and much charcoal.1 On the west this was bounded by a shallow curved gutter connecting two pits, possibly post-holes but containing no packing stones. Outside the area protected by the bank, the layer could not be traced. The remains are probably of a hut of about 20 ft. diameter, perhaps with a wooden wall. IV. THE ENTRANCES (a) The entrance to the Large Enclosure was the only one completely excavated (Fig. 8 and Pl. xm (b). Structural features are projected on to Section E-F.). The entrance passage was 17 ft. long and 8 ft. 6 in. wide at its narrowest point, between walls of large orthostats forming the ends of the rampart. The end of the eastern rampart was however curved so that the width of the entrance at its outer end was 12 ft. 6 in. and at its inner end 11 ft. Slight traces of paving lay almost immediately below the turf and 4-6 in. above the subsoil, but the floor of the entrance passage had been much eroded. The ends of the rampart had a plinth of large, flat blocks set in a founda- tion trench dug into the subsoil this plinth was considerably ruined and varied in width but was well preserved at the south-west corner where it projected 1 ft. 3 in. beyond the face of the wall. At the outer corners of the gate it continued round the corners and was returned for some distance along the base of the outer face of the rampart; the ruin was too great on the west side to determine the extent of this return, but on the east it extended for 17 ft. The purpose of this plinth was no doubt to give additional support to the great thrust of the thickened rampart founded on sloping ground on either side of the gateway the top of the plinth was practically flush with the surface of the subsoil so that when the gate was in use the plinth was not visible above the surface. The construction of the plinth was such that portions could be removed to allow for the erection of new posts without disturbing the remainder further, it was very badly ruined in places, so that some holes may have been destroyed. Four well-defined squarish post-holes with packing stones, however, were located in two pairs at the sides of the entrance passage. No other post-hole was found 1 Generally well preserved, and composed mainly of oak with some hazel. Thanks are due to Dr. K. B. Blackburn for this identification,