588 J. H. BRETZ GLACIAL DRAINAGE OX COLUMBIA PLATEAU 



ville route is 53 feet above the surface of Colville Lake in the adjacent 

 scabland, and Ealston is not 100 feet above the floor of the Cow Creek 

 channel. Along both routes are deposits of stream gravel.^ Yet no scour 

 of the floor by glacial waters, nor steepened valley walls, nor scattered 

 erratic boulders or cobbles were seen." It is possible that these routes 

 were used for l3ut a short time early in the glacial flooding, or it may be 

 that the gravels are older than the Spokane giaciation. It may be, also, 

 that the gravels in channels near Winona and Lacrosse, south of Rock 

 Lake, are of pre-Spokane age. 



The scablands of the Palouse drainage, with channeled basalt, deposits 

 of stratified gravel, and isolated linear groups of Palouse Hills, their 

 marginal slopes steepened notably, bear abundant evidence of a great 

 flood of glacial waters from the north. This flood was born of the Spo- 

 kane ice-sheet. Its gradient was high, averaging, perhaps, 25 feet to the 

 mile, and it swept more than 400 square miles of the region clean of the 

 weaker material constituting the Palouse Hills. The hills which have 

 disappeared averaged 200 feet in height, and in some places the glacial 

 torrents eroded 100 to 200 feet into the basalt. This flood originated at 

 several places along the ice-front. Great river channels exist among the 

 remaining hills in the flood-swept region. The area overridden by the 

 ice itself has lost every trace of Palouse Hills.^^ 



CRAB CREEK DRAIXAGE 



General statement.— -l^y at least ten different routes, glacial waters 

 from the Spokane ice-sheet west of Cheney and Medical Lake converged 

 to Crab Creek. Another discharge way, still farther west, that of Moses 

 Coulee, found its own Avay to the Columbia above the entrance of Crab 

 Creek drainage. The glacial streams tributary to Crab Creek, named 

 from east to west, were Rock Creek (Lincoln County), the headwaters 



* M. M. Leighton : The road building sands and gravels of Washington. Wash. Geol. 

 Survey. Bull. 22, 1911). pp. 104-lO.j. Leighton clearly recognizes that the Spokane re- 

 gion had been glaciated (p. 246). that numerous glacial drainage courses exist on the 

 plateau (p. 84 ). and that the scablands are due to erosion of the sedimentary material 

 by escaping glacial waters (p. 279). 



'■* Campbell, op. cit.. also notes the absence of the granite boulders along the Northern 

 Pacific Kailroad, which follows the Ritzville route, west of the lower end of Colville 

 Lake. 



*" Explanation should liere be made of certain glacial deposits which are older than 

 r' e Spokane drift. In an abandoned clay pit near the Cheney Normal School is a very 

 old clayey till first rei)orted by Frank Leverett (Bull. Geol. Soc. Am., vol. 28, 1916, p. 

 14:i). It contains striated quartzite boulders and cobbles. One such boulder also has 

 tine chatter-marks on it. Granite is present, but is crumbling or etched to such an ex- 

 tent that no ice-marked surfaces remain. This till also is exposed along roadsides south 

 and east of Granite Lake, and striated erratics have been found back in the mature 

 vaMc.vs of tills liill group. It seems probable that the old till underlies much of these 



