280 G. F. WRIGHT POSTGLACIAL EROSION AND OXIDATION 



period necessary to effect the observed results to 12,500 years, which 

 probably is not far from correct. 



There are, however, some other modifying causes on both sides which 

 *must be considered. In the cut-off around the reservoir the shortening 

 of the course has slightly increased the gradient of the stream through 

 that district. This would evidently increase its efficiency. But, on the 

 other hand, the original gradient (about 12 feet to the mile) with which 

 the stream began its work would have been the same as that in the cut-off 

 at the present time. Again, the retarding influence of the forests would 

 not begin until some time subsequent to the beginning of the erosion. 

 Probably, also, the precipitation in that early period was much larger 

 than now, thus increasing the early rate of erosion. These things may 

 therefore be permitted to balance each other so nearly that the with- 

 drawal of the ice from the northern part of Ohio is to be measured by 

 thousands of years rather than by tens of thousands, fully sustaining the 

 general impression which is made upon the observer almost everywhere 

 in the entire section of country. 



As this region is 300 miles south and west of the Mohawk and Saint 

 Lawrence valleys, it is evident that the opening of the drainage lines, 

 which would permit the Niagara River to begin its erosion of the gorge 

 above Lewiston, must have been considerably later — j^robably two or 

 three thousand years later. This we should infer from the size of the 

 various shorelines or lake ridges which occur south of Lake Erie, and 

 from our general impression of the rapidity with which the ice-front 

 retreated. At any' rate it is impossible to extend the age of Plum Creek 

 sufficiently to be consistent with a date of thirty or forty thousand years 

 for the beginning of the erosion of Niagara. There must be some error 

 in the data from which those calculations have been made which give 

 tens of thousands of years to the age of the Niagara gorge. 



A similar conclusion follows from my observations upon the lateral 

 erosion at the mouth of the Niagara gorge, detailed in the Popular 

 Science Monthly for June, 1899, and the American Geologist, volume 29, 

 pages 140-143, and summarized in the fifth edition of the Ice Age in 

 North America, pages 548-552. 



Stream Erosion South of the Saint Lawrence-Mississippi 



Watershed 



In considering the effects of stream erosion in front of the continental 

 ice-sheet in the channels which opened freely to the south, we have to 

 bear in mind the enormous floods of water set free by the melting ice. 



