THE "SVISCONSIX GLACIATIOX 60 



Q 



eiied main channel of the Grand Conlee S3'stem could contain and all 

 but one of the distributary canyons again were in operation. The level 

 of the Hartline gravel plain was not reached, hoAvever, and Deadmans 

 Draw remained untouched. The evidences for this conclusion are the 

 character of the talus in Deadmans Draw, already outlined, and the 

 gravel deposits in and at the mouth of Dry Coulee. These latter deserve 

 a brief description. 



The gravel terraces of Crab Creek Yalley above the junction of Dry 

 Coulee, at Adrian, are fragmentary remnants in protected places and in 

 general do not have sharp terrace forms. Most of the valley floor is at 

 the floodplain level; but 3 or 4 miles east of Adrian the floodplain is 

 narrowed almost to obliteration by a great gravel fill whose surface is 

 about 100 feet above the valley bottom. The creek here flows in a narrow 

 inner valley, close to the southern wall of the rock-cut main valley. The 

 surface of the gravel fill rises northward across the width of the main 

 valley and continues up Dry Coulee, which in its lower part is likewise 

 nearly filled. Though there is little difference in amount of weathering 

 between this gravel deposit and the Spokane gravel in Crab Creek Valley 

 east of Dry Coulee, it clearly is much j-ounger in terms of erosion. Its 

 dissection has just begun. It is traceable back up Dry Coulee to the 

 three distributary canyons which lead southward out of the uj)per walls 

 of Grand Coulee. In all probability it is a deposit made by the Wis- 

 consin floods before the lower canyon of Grand Coulee was deepened 

 sufficiently to take care of the entire discharge. 



Lower Grand Coulee therefore appears not to have been much deeper 

 at the beginning of the AYisconsin discharge than the floor of these dis- 

 tributary canyons. Grand Falls probably was formed during this epoch, 

 taking origin at the head of Blue Lake, where the coulee begins its course 

 in the tilted flows of the monoclinal flexure. This cataract has receded 

 about 3 miles to that portion knoAvn as Dry Falls, west of Coulee City, 

 and about 5 miles to the less pronounced falls at the head of Deep Lake, 

 about a mile south of Coulee City. They could not have existed farther 

 down the coulee, for there it is eroded on the strike of flows whose dip, 

 on the average, is 45 degrees.^^ 



Previous descriptions of the relation of the Okanogan lobe to Grand 

 Coulee state that the ice-sheet deployed eastward only to the edge of the 

 upper canyon ; but the granite knobs in Grand Coulee above Steamboat 

 Rock are strongly glaciated down at least to the level of the present 



13 O. E. Meinzer, in "The glacial bistoi-y of Columbia Rivei- in the Big Bend region" 

 (Jour. Wash. Acad. Sciences, vol. S, 1918. pp. 411-412), argues that the falls have 

 receded for 17 miles, virtually the full length of the lower coulee. 



