Geology of the Lakes and the Valley of the Mississippi. 205 



like an endless saw, it is difficult to think that it did not make the 

 groove in which we find it. If this groove were originally but a 

 valley of denudation, why are its sides perpendicular even at the 

 brink, and why is the original inclination of its slope broken by a 

 cataract now? In the opinion of Professor Rogers and many oth- 

 ers, an inland sea, vastly more immense than the present fresh water 

 lakes, sent a current along the course of the Niagara river, tearing 

 up the exposed portion of the land, and imperfectly excavating the 

 rough and unshapen trough below the falls. The traces of an over- 

 whelming current are doubtless every where visible ; and it is rea- 

 sonable to suppose that, seeking the lowest part of the barrier, it 

 would gradually narrow and confine its action at that point, at least 

 sufficiently to mark out the course of the subsequently diminished 

 stream. But we are unable to imagine how a wide spread torrent 

 could have spent its entire action on a strip six hundred yards in 

 breadth, giving to the sides of the gutter made by it, the character 

 and appearance of perpendicular walls. No such walls are found in 

 the water gaps of the Alleghany mountains. Granting tliat even a 

 narrow diluvial current ran along the slope, till it reached the termi- 

 nation of the plain, what reason have we to think there was not, ori- 

 ginally, an overfall there ? It is a postulate of the argument, that 

 the bottom of the trough at the close of the diluvial abrasion, de- 

 scended gradually to the foot of the escarpment. Admitting it for 

 the moment to have been so, it is not easy to see why a cataract 

 should have been formed in it since. According to the same writer, 

 the process of excavation now going on, will eventually reproduce an 

 inclined trough, differing from the supposed diluvial one, but in its 

 greater length and lesser inclination. It is more reasonable to con- 

 clude that the present one is altogether of postdiluvian origin, and 

 that it has been formed by the present stream conducted to the point 

 of embouchure at Queenstown, as it is now conducted to the cata- 

 ract, by an excavation of the gravel which it retired into and deep- 

 ened at the subsidence of the waters. But the eventual existence 

 of an unbroken line of descent, can be admitted but on the very con- 

 cessible ground of its corresponding to the dip of the strata ; for 

 while any portion of the lias remains to be worn away, an over fall 

 will be a necessary consequence of its structure. It is reduced less 

 by direct detrition than by the absorption and expansion of moisture 

 from frost ; a process of disintegration common to argillaceous rocks, 

 and strikingly exhibited in the sandstone steps and lintels in Pitts- 



