Drainage Features of the Upper Ohio Basin. 279 



the sum total of work done is the least that will apparently 

 satisfy the phenomena, though more of the work is thrown 

 within the limits of the glacial period than in the following 

 hypotheses. 



(1.) This hypothesis supposes that the Allegheny valley was 

 cut down before the glacial period only to the observed base 

 of the high-level sections of gravel, making such allowances 

 •for the incompleteness of observations as may be required. 



(2.) The first known ice incursion, which reached the ap- 

 proximate limits at which thin drift and scattered bowlders 

 are now found, gave origin to the high terrace gravels through 

 its outflowing waters. 



The amount of these gravels, even under this hypothesis, 

 which minimizes them, seems extraordinary, when the thin- 

 ness of the adjacent old drift is considered. This drift, how- 

 ever, has undoubtedly suffered large wastage as have the 

 gravels themselves. 



(3.) After the deposition of the gravels, the ice retired be- 

 yond the basin. So far as this problem is concerned, it mat- 

 ters little whether farther or not. Erosion followed, trenching 

 the gravels not only, but the rock beneath to about the pres- 

 ent depth. 



(4.) Another ice invasion followed forming the adjacent 

 terminal moraines and. pouring its overwash gravels down 

 through the trench previously cut, forming the lower gravel 

 system. 



(5.) Since the retreat of the ice beyond the limits of the 

 basin, these gravels have been trenched by the present streams.. 



In this hypothesis, the amount of gravel deposition is min- 

 imized. The amount of interglacial erosion is made a max- 

 imum. 



Hypothesis II — -This hypothesis supposes that (1) the river 

 trenches of the region were cut to essentially their present 

 depths before the glacial period, and (2) that, during the pres- 

 ence of the ice, the valleys were filled to the heights of the 

 uppermost gravels — say 300 feet in round numbers. (3) Sub- 

 sequently, these were trenched into nearly the present form. 

 (4) Afterwards, the later incursion of the ice formed the ad- 

 jacent terminal moraines and poured the overwash down the 

 valleys forming the lower terrace planes. (5) These have since 

 suffered trenching as postulated in the previous hypothesis. 



This differs from the preceding hypothesis in very greatly 

 increasing the deposit of the first glacial epoch, which we 

 observed, even under the preceding hypothesis, was surpris- 

 ingly large in comparison with the associated drift. It post- 

 ulates the same volume of interglacial erosion, but this takes 

 place in glacial gravels, instead of part gravels and part rock. 



Am. Jour. Sci.— Third Series, Vol. XLVII, No. 280.— April, 1894. 

 19 



