638 / HARLEN BRETZ 



ated, high-gradient rivers. Across this monodine between Columbia 

 River and Coulee City, the canyon is 30 miles long and averages 

 all of 2 miles in width and 800 to 900 feet in depth. In the 

 making, at least 10 cubic miles of basalt were excavated and 

 removed. Though a later flood of glacial waters used this route,^ 

 it did but little to deepen it.^ By far the greater part of the erosion 

 of Upper Grand Coulee was performed by the earlier glacial stream. 

 It is very probable that this immense task was performed by the 

 stoping of cascades and cataracts which retreated entirely through 

 the monoclinal uplift to the deeper valley of Columbia River and 

 thus left the ^reat notch. ^ 



It also seems probable that when the retreat of the ice sheet 

 began, the plateau west of Grand Coulee was abandoned last. 

 Earlier clearing of the Spokane and Columbia valleys to the east 

 allowed all the drainage of the ice sheet to use the Grand Coulee 

 route, which was then the lowest of all. Grand Coulee's greatest 

 flood and probably its greatest erosion thus came after the other 

 scabland routes had gone dry. 



VOLUME OF THE GLACIAL STREAMS 



If the channeled scablands of the Columbia Plateau are the 

 erosive results of glacial waters, certain statements as to the 

 volume of the streams can be made. Measurements are possible 

 if remnants of the preglacial floor of the main valley exist in places 

 where the valley brimmed over with the glacial flood to produce 

 distributary courses. Should it appear that the amount of canyon- 



' During the Wisconsin glaciation. 



2 Grand Falls, below Coulee City, consists of Dry Falls, Deep Lake Falls and 

 a smaller unnamed falls a mile east of Deep Lake. The lip of the smaller falls is 125 

 feet higher than the floor of the channel leading to Dry Falls. Yet all were made by the 

 same glacial stream and only Dry Falls and Deep Lake Falls were used and modified 

 by the later discharge. 



Furthermore, the Wisconsin ice did not cross Spokane River or Columbia River 

 east of Grand Coulee and its waters were free to use the lowest of the ten earlier routes. 

 Only Grand Coulee was so used, showing that it had been eroded by the earlier dis- 

 charge to a depth not far short of that which it now has. 



3 This inference has no physiographic evidence in Grand Coulee to substantiate it, 

 but is based on the known procedure of the glacial' streams in similar situations, e.g., 

 Lower Palouse Canyon, Devils Canyon, Frenchman Springs, and The Potholes. 



