24 Proceedings of the Royal Irish Academy. 



chasm, and then curves regularly round to meet it. This is evident 

 proof that it curved round the mound, which, for at least 15 feet, 

 has disappeared with the falling cliff since the earthworks were dug. 

 A shallow fosse outside a stone wall, though rare, is not unknown. It 

 occurs at Caherconree, a mountain cliff-fort, and is even mentioned as 

 at the fort in the legend of Cuchullin, in " Bricriu's feast " — " out over the 

 stone wall (cathair) till he fell into the mire of the fosse at the palace-gate."' 

 We also find it at Doonaunroe, in Clare, and at certain Cork and Kerry- 

 ring forts, like Staigue, Cahernanackree, and Cahermoygilliar. The mound 

 is of the usual type, about 3 feet high and 9 feet thick on top. It is of light, 

 rather loose soil, with a layer of stones under its surface, and above the 

 present level of the field. The fosse is cut into the rock to the " north-east " 

 of the entrance. It is straight, 6 to 7 feet under the field-level, and is 6 feet 

 wide at the bottom, and 22 feet at the field-level. It commences at 28 feet 

 from the entrance wall, and runs for about 54 feet to the break of the gully ; 

 as we saw, it probably ran for about 15 feet farther when entire. The 

 inner mound is 21 feet thick, and is nearly covered by the dry-stone rampart 

 and the talus of fallen stones to either side. 



It seems that the original works consisted of the natural stream-bed, and 

 scarps to the left (south-west) of the entrance and the fosse (or fosses), with 

 the earth-mounds to the right. The rampart was probably rebuilt several 

 times. The outer entrance seems different from any recorded in the primitive 

 forts, not only in Ireland, but elsewhere in the British Isles, and on the 

 continent of Europe ; although external defences of the gateway, sometimes 

 of considerable complexity, arp not unknown. 



As shown by Otway (and the remains tally generally with his plan) the 

 enclosure was shield-shape, running to a point, with the entrance in the 

 right flank. It is about 46 feet long, the wall 5 to 6 feet thick. To the 

 right, about 25 feet out from the wall, is a nearly levelled circular hut, 13 feet 

 across, between the entrance and the head of the great fosse. This and 

 another hut (of which I saw no trace, and which Otway does not mark) to the 

 left are shown, flanking the entrance, in the Ordnance Survey Map of 1840. 

 Wakeman gives the gateway as 3 feet 8 inches wide, which closely agrees 

 with that of Dun Fiachra (3 feet 9 inches), but no trace of this is visible 

 through the debris, and a new wall built across the gap. 



The rampart is straight, resting for the " southern " part on the upturned 

 ends of the strata along the fault ; a low ledge about 9 feet high, apparently 

 a natural scarp, turns back at the south-west end. The " northern " part 

 rests on the mound inside the fosse. The whole is 118 feet long, 65 feet 



' Fled Biicrend (ed. G. Hendei-son, Irish Texts Soc), p. lO.i. 



