Geol.— Vol. I.] TURNER— ORIGIN OF YOSEMITE VALLEY. 277 



very improbable that a river, having established a channel 

 in Tertiary time, should be diverted in the early Pleistocene 

 into another drainage, and at a somewhat later time revert 

 to its original course. 



To get a clear conception of the explanation of the for- 

 mation of Yosemite Valley here presented, let us go back 

 to the time when the drainage of the Merced was being 

 initiated. The great granitic mass of the Sierra Nevada 

 was, in all probability, at one time covered by other rocks, 

 since all the evidence extant indicates that acid igneous 

 rocks never become thoroughly crystalline except under 

 pressure. In other words, granitic rocks have never formed 

 at the surface of the earth. This original cover may be 

 presumed to have been composed of pre-Cretaceous sedi- 

 ments and associated lavas, and remnants of it are to be 

 recocrnized in the small areas of these rocks which form the 

 summits of many of the high points. Let us suppose that 

 erosion has gone far enough to have removed the larger 

 part of this cover, exposing the underlying granitic rocks 

 in nearly all the area under discussion. 



Like all mountain ranges, the Sierra Nevada has been 

 subjected to earth stresses. The original folding of the 

 pre-Cretaceous sediments, the slaty structure, and the 

 schistosity of the rocks are clearly all due to these stresses. 

 In addition to the slaty and schistose structures, the older 

 rocks of the range are intersected by systems of joints, 

 which may also be due to compressive stresses. These 

 joints comprise two vertical sets, two or more diagonal sets, 

 and a horizontal set. As a matter of fact, it very seldom 

 happens that all of these systems of joints are represented 

 in the same rock mass. There is often a combination of 

 one set of vertical joints with the horizontal joints, but in 

 nearly all cases some one system is most strongly developed 

 at a given locality. On Plate XXXII a set of vertical joints 

 striking northeast and southwest is shown : and on Plate 

 XXXIII a diagonal set may be seen. 



When the streams of this early period began to form 

 channels it is clear that the water would follow the jointed 



( 2 ) November 7, 1900. 



