Skip to main content

The original rampart of period II was composed of a layer of soil and stones, 20 ft. wide, and only about 1 ft. or less in height. On the back 9 ft. of this layer larger stones were piled, but the front portion of the rampart must have been built mainly of soil. This front part was held in position by two rows of posts, one a true line of revetment, the other in the midst of the rampart. They were set in holes, 7 in. in diameter, and about 12 in. deep in the rock, and were regularly spaced, 64 in. apart. These post-holes occurred in front of the rampart on both sides as it approaches the entrance, and a line of them was also found returning at a right-angle to the actual gate. The arrangements of the gate had been much obscured by later readjustments, but it seems that there was a double gate, possibly with a bridge above it. The main uprights of this gate rested in pairs of larger round post-holes, cut into the rock. In period III the front part of the rampart at least was remodelled. The posts were replaced with larger ones, set in holes of 30 in. diameter at the top, 19 in. at the bottom. In two cases post-holes of this type had partly cut away examples of the earlier, smaller kind. The rampart was of soil and stones as before, but this time it was laced with horizontal as well as vertical timbers, and impressions of two of the former were noticed. This rampart had been burnt, whether deliberately or acci- dentally, it is difficult to say. There were large patches of rampart, which were entirely red in colour. Detached pieces of vitrified material were found in quantity, as in 1937, and in one place it could be said that the top surface of the rampart, as it was found, was actually vitrified. For the most part, however, it is only the lowest and least burnt part of the rampart which remains in situ. The rest seems to have been thrown down in ancient times, and some of the material was re-used later for building purposes. No burnt rampart was found in the fairway of the entrance, but this may be due to later destruction and denudation. The gate in this period was in the same position as before, but a new system of post-holes was dug. This consisted of large oblong holes. They were very well cut, vertically on the downhill side and at an angle on the uphill side. Three of these with adjacent rectangular slots, apparently for struts, occurred at the actual gate it is unlikely that there was a bridge at this time. Two other oblong holes occurred down the fairway, at the ends of the rampart revetments. There was no trace of any guard-house in either period, but further work will be devoted to this point next season. B. H. ST. J. O'N. A Roman GAME-BoARD FROM HOLT, Denbighshire. — In the Holt Collection at the National Museum of Wales, Cardiff, there is a fairly well-preserved Roman game-board of buff ware (Fig. 1). Mr. \V. F. Grimes thus describes it (1" Cymmrodor, XLI, 1930, p. 131) "the arrangement consists of lines of twelve roughly-incised conventionalised ivy leaves, on opposite sides, with a geometrical pattern in the middle. Down the centre of the board runs an incised line carrying twelve pairs of scrolls corresponding to the ivy leaves, with the geometrical compass-drawn pattern in the centre. The whole is enclosed by a raised moulded border, probably to prevent the pieces leaving the board." Mr. Grimes tells me that it may be dated to the first half of the second century A.D. This board must certainly have been used for the game known as xii scripta, the "game of the twelve points" this was of the backgammon type, played with the aid of dice, and was extremely popular among the Romans. There are extant