THE WISCONSIN GLAC'IATION 605 



the granite hills in the coulee close to the eastern wall do not show glacial 

 smoothing, though all others do. This is interpreted to mean that the 

 escaping waters kept a passage open along the eastern edge of the ice, 

 which thus failed to close the conlee completely. All the water at this 

 time flowed on the east side of Steamboat Eock, and perhaps the eastern 

 wall of the canyon here was eroded back to expose the nnglaciated granite 

 hills and to make the greater width w^hich the canyon possesses between 

 Steamboat Eock and the Colnmbia.^* 



A widespread submergence of the lower Columbia Valley is known to 

 have occurred during the Wisconsin glaciation.^^ It is recorded by berg- 

 floated erratic boulders, some of great size, scattered widely in the C^o- 

 lumbia Valley below the present altitude of about 1,250 feet above tide. 

 The submergence was due to a lowering of the entire region relative to 

 sealevel. The ponded waters rose sufficiently high to spread over consid- 

 erable areas of the plateau, and glaciated boulders now are found where 

 glacial ice or glacial streams could not possibly have transported them. 

 Most of these boulders, and all of the large ones, are of granite. They 

 are strikingly abundant in some parts of the Quincy basin, a distribution 

 which points to Grand Coulee as the route by which they reached the 

 basin. After one has seen the dozens of granite knobs in upper Grand 

 Coulee, heavily glaciated on the northwest and a|)parently much plucked 

 on the southeast, the conviction grows that most of these large granite 

 fragments were quarried in the head of the coulee when the Okanogan 

 lobe was at its maximum deployment. The ground moraine of the 

 Okanogan lobe has but a small percentage of granite boulders compared 

 with basalt boulders, and a still smaller percentage of large granites com- 

 pared with large basalts. It appears, therefore, that some special condi- 

 tion, such as that outlined above, must have existed to reverse the ratio 

 among the berg-floated boulders. 



The upper limit of these erratics earlier reported was 1,283 feet above 

 tide, and in the eastern part of the Quincy basin none have since l3een 

 found above that altitude ; but on Babcock Eidge, near Trinidad, boulders 

 of gneiss, granite, quartzite, schist, slate, and argillite have recently been 

 found as high as 1,350 feet above tide. On the hills northeast of Trin- 

 idad a "nest" of twelve granite fragments from an inch to 16 inches in 

 diameter and one quartzite |)ebble have been found — all within a radiu^ 

 of 15 feet and at an altitude of 1,400 feet above tide. These seem clea rlv 



" K. Oestreich ("Die Grande Coulee."" Transcontinental Exciu'sion of 1912. America?! 

 Geographic Society, 1915, pp. 259-274) has suggested, and J. T. Pardee ("Glaciation in 

 the Cordilleran region," Science, vol. 56, December 15, 1922, pp. 686-687) has asserted. 

 that an ice-stream traversed Grand Coulee. 



^^ J. H. Bretz : The late Pleistocene submergence in the CoIuml)ia Valley of Oregon 

 and Washington. .Tour. Geol., vol. 27, 1919, pp. 489-506. 



