4 ANCIENT FORTIFICATIONS in 



thofe hills, of which the walls, remaining in fome places of fe- 

 veral feet in height, were evidently compacted together by the 

 vitrification of the ftones of which they were built ; and he of- 

 fered fome ingenious conjectures with regard to the means em- 

 ployed in forming fuch extraordinary ftructures, and the pur- 

 pofes for which they might have been reared. 



This account, which Mr Williams himfelf candidly owned, 

 was by many people treated as a ficlion, excited, however, the 

 curiofity of feveral travellers to vifit and examine fome of thofe 

 hills which he had mentioned. In the fame year, 1 777, Dr James 

 Anderson of Monkfhill, tranfmitted to the Society of Anti- 

 quaries of London, a very elaborate account of fome ancient 

 monuments and fortifications in the Highlands of Scotland, 

 contained in two letters, which are publifhed in the 5th and 

 6th volumes of the Archaologla. In thefe he treats, at confider- 

 able length, of the vitrified forts, and particularly of that upon 

 the hill of Knockfarril in Rofs-fhire ', and, agreeing with Mr 

 Williams in the general idea, that, in rearing thofe ftruclures, 

 the builders had employed fire for the purpofe of cementing 

 the materials, he differs from him a little as to the manner in 

 which he fuppofes the fire to have been applied to the mound 

 or rampart. 



It is curious to remark, how the fame appearances, to dif- 

 ferent obfervers, lead to the moft oppofite opinions and conclu- 

 fions. The two gentlemen above mentioned feem not to have 

 entertained the fmalleft doubt, that the vitrified materials on the 

 tops of thofe hills, were the veftiges of works of art, and the 

 remains of ftruclures reared for the purpofes of fecurity and 

 defence. The Biihop of Derry, when on a tour to the north 

 of Scotland, vifited the hill of Craig-Phadrick near Invernefs, 

 and expreffed his opinion, that the mounds of vitrified matter 

 were not the remains of any artificial work, but the traces of 

 an ancient volcano. In the Philofophical Tranfa&ions of the 

 Royal Society of London for 1777, Part II. No. 20. is an ac- 

 count 



