TRAVELS IN ICELAND. 



45 



was very uneven; insomuch that we were obliged to go often 

 out of our way., or to run the risk of being precipitated to the 

 bottom. 



PYRAMIDS. 



We here observed two very remarkable particularities; the 

 first is, that the sand has accumulated upon the ice of the 

 mountain in round and black pyramids, in the shape of a sugar- 

 loaf. They are of such a regular figure, that they appear as if 

 the sand had been placed by art. These conical elevations are 

 from four to sixteen feet in height, and are only a few paces 

 distant from each other: while their shape is that of a steep 

 ridge, from which rise others, each being smaller than the one 

 that precedes it. In some places they are not so regular as 

 in others, but particularly where we began to escalade the glacier; 

 because the sand, by being conveyed over the glaciers, mostly 

 falls in them. It is justly supposed, that it is not possible for 

 the sand to remain heaped in this manner, but that these heaps 

 may preserve their form till it is altered by the imbibition of 

 moisture. On pushing our pikes into these pyramids, we as- 

 certained, that their newel, or interior mass, was principally 

 composed of ice. In the first which we examined, the mass 

 that formed its base had given way, or more properly speaking, 

 had been dissolved by the water that ran off from its summit ; for 

 in this, as well as in most of the others, we observed a sort of 

 gutter or trench. 



It is well known, that on the glaciers the greatest quantity of 

 snow falls in winter, and that the winds convey thither the sand 

 and dust from the adjacent mountains, which are generally 

 covered with this substance. Experience has at the same time 

 proved, that the high mountains and particularly the glaciers 

 attract the air towards them, and with it whatever it may en- 

 velope : this sand, which moves about like waves, accumulates 

 by falling in the lowest places upon the heaps of snow, and 

 to four or five feet in height on the ice of the mountain. In 

 spring the snow melts by the action of the rain or the sun ; and 

 the thawed water meeting in its course with a mass of snow or 

 sand, increases in the parts where such mass is most abundant, 

 till it form a lake. The great masses being, on the contrary, 

 more compact from their weight, resist these thawed waters, 

 and retain the sand which covers them, so that they suffer but 

 little from the influence of the air. It should also be observed, 

 that the snow which forms them, entirely absorbs the water, 

 which in the night freezes again ; for though in summer the air 

 be very clear during the day, and the sun very brilliant, it freezes 



