GLACIO-AQUEOUS ACTION IN NORTH AMERICA. 169 



would break up and round blocks of the formation, and strew 

 them along the surface, carrying of course the lightest the farthest. 

 But if this were the only mode of their dispersion, we should 

 expect that the gi-eater part of them would be found crowded 

 together in the first deep valley in a southeasterly direction from 

 the original rock. But m fact we do not usually find them there, 

 though it must be acknowledged that the northern slopes of hills 

 are frequently more sti-ewed over with them than the southern 

 side. This we should expect, if many of them were conveyed 

 by huge icebergs, whose lower extremity, striking against these 

 hills, would drop many blocks. Upon the whole, the phenomena 

 of the dispersion of bowlders are best explained, by supposing a 

 part of them floated along the surface of a current, attached to 

 icebergs, and a part of them urged forward, pell meU, by ice, 

 grating along the surface. The latter effect might to a certain 

 extent result from the mere expansion of a mass of ice without 

 aqueous currents. Bi^t that currents of water were concerned, 

 where bowlders have been carried four hundred, or five hundred, 

 or even one hundred miles, cannot be doubted ; and to this extent 

 at least have they been transported in the central parts of our 

 country. Besides, the uniformity in the direction taken by our 

 bowlders, agrees far better with the idea of currents of water, than 

 any other agency. 



The removal of bowlders from lower to higher levels is another 

 fact of great importance in the history of our drift. It is an ex- 

 tremely difficult point to ascertain precisely how much they have 

 been raised from their original situation, because it may be that 

 the original rock, now scattered in bowlders over lofty ridges, may 

 once have constituted ridges much more elevated than at present. 

 AVhen, however, the transported rock is considerably newer than 

 that over which it is scattered, the presumption is, that the former 

 always occupied the lowest level. Let us take for example the 

 bowlders of quartz and Silurian rocks found strewed over the 

 primary strata of the Green and Hoosac Mountains, and the 

 Highland chain extending through New York and New Jersey. 

 Excepting a few limited peaks of these rocks, as the hill of quartz 

 called Bald Mountain in North Adams, and the sandstone raoun- 

 12 



