G. F. Wright — Continuity of the Glacial Period, 171 



I. C. White has specially called attention.* In the second 

 place, I have been led to reflect much for some years past upon 

 another cause which may produce great irregularities in the 

 depth of the rock bottom of river gorges, namely, the effect 

 of the plunging force of a cataract in hollowing out the gorge 

 at its foot to a depth much below the general level. The 

 extent of this process will depend partly upon the character of 

 the underlying rock, and partly upon the conditions which 

 determine the extent of the plunge of the water over the 

 cataract. It is well known that the plunge of the water over 

 the Niagara escarpment has scooped out a channel to a depth 

 of 200 feet or more below the level of the stream north of the 

 falls. While studying the gorge of Snake River above the 

 Lower Shoshone Falls in Idaho three years ago, I was struck 

 with a similar phenomenon. The water is but a few feet deep 

 over the crest of the falls; but, less than half a mile above the 

 falls, the depth becomes 100 feet or more, and so continues 

 for some distance. The most natural explanation of this 

 would seem to be that during the recession of the upper falls 

 there were conditions which favored deep erosion much more 

 at certain places than at others. These conditions may have 

 been those which should determine that at certain places the 

 fall was mainly over rapids, and at other places in a perpen- 

 dicular plunge. The varying hardness of the strata under- 

 neath may also have combined to produce the result. It has 

 seemed to me that some such combination of causes as this 

 may account for many, if not all, the facts which are appealed 

 to in the upper Ohio Yalley in proof of the former northward 

 flow of its drainage. 



But the settlement of this point is immaterial to the ques- 

 tion in hand. The extent of the preglacial erosion now dem- 

 onstrated to have taken place in the Allegheny Yalley at 

 Warren is so great that it carries with it, by natural inference, 

 a similar extent of preglacial erosion all down the Ohio and in 

 its tributaries. The erosion at Clark's Run, in the valley of 

 the Beaver, which I have already described, is in accord with 

 the facts concerning the Allegheny near Warren. In examin- 

 ing the Allegheny River from Warren down to Pittsburgh, 

 although I did not find any other facts which are so unequivo- 

 cal in their significance as these are ; yet I did find much 

 which confirms the view, and so far as I can see there is noth- 

 ing which can be adduced as strongly favoring the other view. 



The 200-foot rock shelf reappears at frequent intervals all 

 the way down the Allegheny to its junction with the Ohio, 

 and down the Ohio as far as Wheeling certainly. It varies 



* This Journal, for 1884, p. 149. 



