ACTION OF THE ATMOSPHEEE 159 



nnder ordinary temperatures lias been a matter of experiment 

 W. H. Bartlett^ has shown that the average rate of expansion 

 for granite amounts to .000004825 inch per foot for each de- 

 gree Fahrenheit ; for marble .000005668 inch, and for sandstone 

 .000009532 inch. Adie, in a series of similar experiments, found 

 the rate of expansion for granite to be .00000438, and for white 

 marble .00000613 inch.^ Slight as these movements may seem, 

 they are sufficient to in time produce a decided weakening and 

 afford a starting-point for other physical and chemical agencies 

 The writer well remembers the peculiar impressions produced 

 during one of his earlier trips into the comparatively arid 

 regions of Montana, at finding the slopes and valley bottoms 

 strewn with small, beautifully fresh, concave and convex 

 chips of a dense, coal-black, andesitic rock that occupied 

 the crest of one of the higher hills. So fresh were the frac- 

 tures, so free were they from oxidation or other signs of de- 

 composition, it was at first felt that they must be of human 

 origin, that they were chips flaked off by aboriginal workmen in 

 making stone implements, and some time was wasted in seeking 

 for the more complete results of their handiwork. It, however, 

 did not take long to convince him that the flakes were far too 

 abundant and too widely spread to have originated in any such 

 manner, while the finding, on the top of the hill, of the coal- 

 black r;ck, broken into larger colnn^iar blocks, each with its 

 angles rendered more obtuse or even fluted by the springing off 

 of just such flakes, -this, coupled with the knowledge that 

 during the day, exposed under a cloudless sky, the rocks became 

 so highly heated as to be uncomfortable to the touch, whilst at 

 night the temperature sank nearly to the freezing-point, sufficed 

 to teach, as it must have taught the most obtuse, that the ordi- 

 nary daily temperature variations were amply sufficient to ac- 

 count for the phenomenon. In cold climates, and particularly 

 where glaciation has prevailed, the results of such flaking are 

 sometimes strikingly manifest. Explorers in Northern Labrador 

 brought back to the National Museum sheets of coarse red granite 

 of remarkably uniform thickness (about 5 mm.) found over areas 

 of many acres, still perfectly fresh and showing glacial striae or 

 scouring on their upper surfaces. 6. K. Gilbert, of the United 



1 Am. Jour, of Science, Vol. XXII, 1832, p. 136. 

 ^TraBS. Eoyal Soc. of Edinburgh, Yol. XIII, p. 366. 



