84 PHYSICAL GEOGRAPHY. 



naturally combined with the clay in such proportions as to 

 produce a mass of extreme hardness, which, together with the 

 toughness imparted by the clay, renders its excavation almost 

 as difficult as that of rock. 



The sand of the unaltered drift is seldom separated from 

 its other materials in any degree of purity, but it is not unfre- 

 quently the case that it exists in excess of the others ; and in 

 some cases small local accumulations or pockets of it are 

 found, having a considerable degree of purity. The sand is 

 quite variable as to fineness, and it sometimes approaches or 

 grades into gravel. 



The graml of the drift is derived largely from rocks that 

 are more or less purely silicious, but occasionally they are 

 found to be of granitic composition. 



In southwestern Minnesota, the Sioux quartzite is often 

 found in the form of beds of considerable thickness, composed 

 of a mass of silicious pebbles, varying in size from coarse 

 grains of sand to three or four inches in diameter. Such beds 

 are not so hard and compact as the ordinary quartzite is, but 

 the pebbles often separate with considerable facility; although 

 this, as well as the quartzite, has undergone metamorphism. 

 In western and northwestern Iowa also, the Cretaceous sand- 

 stones are found to contain beds of silicious pebbles scarcely 

 cemented together at all. The pebbles of both these forma- 

 tions are silicious and never granitic. 



From these sources, and also from the beds of rivers that 

 existed upon the surface at the commencement of the Glacial 

 epoch, it is believed a large part of the pebbles of the drift 

 were derived. These of course were already formed as 

 pebbles at the commencement of the Drift epoch, some having 

 been worn in Azoic and some in Cretaceous seas; while 

 others received their form in the beds of pre-glacial rivers. It 

 is not denied that a part of the drift pebbles may have 

 received their rounded form by glacial attrition, or other 

 causes subsequent to the glaciers; but there can be no doubt 

 that a large part of the drift gravel of Iowa existed as gravel 

 before the Glacial epoch. Gravel constitutes only a small 



