364 RAMBLES OF A GEOLOGIST. 



labouring, to the general detriment of the pile, in detaching 

 one from the walls above, he should set himself to seek one 

 here. The blocks, uninjured by the hammer, exhibit, in most 

 cases, the angular character of the original fragments better 

 than those forcibly detached from the mass, and preserve in 

 fine keeping those hollower interstices which were but par- 

 tially filled with the molten matter, and which, when shat- 

 tered by a blow, break through and lose their character. 



One may spend an hour very agreeably on the green sum- 

 mit of Knock Farril. And at almost all seasons of the year 

 a green summit it is, greener considerably than any other 

 hill-top in this part of the country. The more succulent 

 grasses spring up rich and strong within the walls, here and 

 there roughened by tufts of nettles, tall and rank, and some- 

 what perilous of approach, witnesses, say the botanists, that 

 man had once a dwelling in the immediate neighbourhood. 

 The green luxuriance which characterizes so many of the 

 more ancient fortalices of Scotland seems satisfactorily ac- 

 counted for by Dr Fleming, in his " Zoology of the Bass.' 

 " The summits and sides of those hills which were occupied 

 by our ancestors as hill-forts" says the naturalist, " usually 

 exhibit a far richer herbage than corresponding heights in the 

 neighbourhood with the mineral soil derived from the same 

 source. It is to be kept in view, that these positions of 

 strength were at the same time occupied as hill-folds, into 

 which, during the threatened or actual invasion of the district 

 by a hostile tribe, the cattle were driven, especially during 

 the night, as to places of safety, and sent out to pasture in 

 the neighbourhood during the day. And the droppings qf 

 these collected herds would, as takes place in analogous cases 

 at present, speedily improve the soil to such an extent as to 

 induce a permanent fertility." The further instance adduced 

 by the Doctor, in showing through what protracted periods 

 causes transitory in themselves may remain palpably influen- 



