Davis.] 30 [May 3, 



cial erosion, just as the occurrence of stratified fragmental rocks 

 in great volume gives the best demonstration of the strong ero- 

 sive power of water. But the argument is of difficult and uncer- 

 tain application from our ignorance of the origin, amount, and 

 distribution of the drift, and of the condition of the country 

 before glaciation. To obtain an idea of the latter point, we may 

 compare nonglaciated and glaciated regions of similar age, as the 

 Black Mountains of North Carolina with the White Mountains 

 of New Hampshire, or the Andes of Central Chili with the Alps. 

 It may be fairly inferred that the latter members of the con- 

 trasted pairs were, before the glacial period, as the former still 

 are, covered with a great amount of loose rock and soil lying 

 where it weathered or in the valleys near by, the result of secu- 

 lar disintegration. The passage of ice has rather rapidly scraped 

 such material away from the mountainous or hilly countries, 

 leaving the characteristic, firm, rounded tors, 1 and carrying the 

 detritus off to lower ground ; there it is generally deposited on a 

 smaller area than that from which it came, and so is increased in 

 depth ; the facts that in hilly regions the drift is found mostly in 

 valleys and that valleys are the ordinary lines of travel, serve 

 still farther to exaggerate the impression of its average thickness. 



If New England had as heavy a soil as the Carolinas, there 

 need have been comparatively little erosion of solid rock below 

 the soil to supply the observed amount of drift, but that some 

 solid rock was successfully attached is shown by the unweathered 

 character of many boulders ; these, however, probably came from 

 projecting knobs and ledges, and not from valley bottoms. 



B. 2. Amount of Drift. A number of quantitative estimates 

 of the drift have been attempted, but they are necessarily crude 

 and may be exaggerated ; but even admitting their truth they do 

 not require anything like the amount of erosion claimed by the 

 more advanced glacialists. 



A. Helland (Ueber die glacialen Bildungen der nordeuropaischen 

 Ebene, Deutsch. Geol. Gesell. Zft. xxxi, 1879, 97) finds Lake Wenern 

 in Sweden and the Island of Seeland in the Baltic to be about the 

 same area, but that the amount of drift on the latter is nearly double 



1 This is suggested by J. F. Campbell as a concise substitute for the unwieldy 

 roches tiioutonnees. (Frost and Fire, Phila., 1865, n, 6.) It should certainly have a 

 place along with till, kame, and drumlin. 



