Mr. Tate oti the Celtic Town at Greaves Ash. 301 



which are carefully formed of large blocks, is on the west 

 leading down to the lower Forts ; another is on the north-east, 

 leading probably to the spring of water in the ravine ; and 

 there are four more openings, through the south rampier, on 

 to a fine grassy hill commanding an extensive view over the 

 valley of the Breamish. Along the base of the hill runs the 

 outermost defensive rampart, Plate 4, G. 



Within this Fort are distinct remains of 15 hut-circles ; 

 and there are besides some large enclosures and obscure 

 traces of other circular foundations. Seven of the huts have 

 been partially cleared ; and the results prove their characters 

 to be the same as those of the lower Forts. Most of them are 

 flagged and paved ; the masonry exposed is of a ruder descrip- 

 tion, but more of it has been preserved, for in huts, h and e, 

 Plate 4, the walls are from 2^ feet to 3 feet high above the 

 floors. A cutting was made through the deeply hollowed 

 enclosure — h, Plate 4 — without disclosing anything import- 

 ant ; but on clearing the hut — h, Plate 4 — which opens into 

 it at a higher level of 5 feet, rude steps were found, ascend- 

 ing from the enclosure into this hut. 



A discovery of some interest was made in hut i, Plate 4, 

 which also is at a higher level than the enclosure to the 

 south-west. It is flagged like the other huts with flattish 

 porphyry stones, tolerably well fitted to each other, so as to 

 form a rude yet not unsuitable floor. Every stone in the 

 ramparts, walls, and floors is of the porphyry of the district, 

 excepting one, and that was found in this hut used as a flag 

 in the floor. Curiously enough this stone is part of a well 

 formed Quern made of sandstone, such as occurs at Titling- 

 ton, Eglingham, and other moorlands of Northumberland. 



Water, — Though there are no wells or springs within the 

 town itself, water was abundant and accessible. Linhope 

 burn was not far distant from the lower forts, and the way 

 thither was strongly defended. The Upper Fort had near to 

 it a streamlet or sike, running in the ravine on the east, 

 and protected by the outermost defensive rampart, which 

 extends at least as far as this streamlet. A spring rises out 

 of the hill side a little northward of the Western Fort, and 

 another issues out of the declivity, near to the south-east 

 corner of the outer rampart, through which there has been a 

 narrow opening to give ready access to this spring. 



Roads. — The principal road — Plate 4, E — winds dowTi 



