14rt 



of drift follows in a general way that of the rock surface whereon 

 it lies ; and that the component layers of this drift conform in 

 their inclination, in a general way, both to the slopes of this core 

 of rock, and to the inclinations of the surface of the mound itself. 

 And this is true whether the inequalities of the surface are on the 

 small scale or on the large. We all know too, how, in a general way, 

 modern streams follow approximately the same courses as their pre- 

 glacial representatives. I dwelt at considerable length upon the 

 origin of these curious phenomena in the papers mentioned at the 

 head of this article. The surface of the drift mounds very 

 frequently represents in a general way the contour of the rock 

 surface beneath it, just as the surface of a field of corn represents 

 in a general way the larger inequalities of the surface whereon the 

 corn is rooted. And what is more remarkable still is, that this 

 curious relation between the external form of the drift mounds to 

 their internal structure, and to the form of the surface beneath, is 

 not by any means confined to mounds of Till, but is observable 

 almost as frequently in well-stratified deposits, such as the mounds 

 of sand and gravel. In eskers the fact of the correspondence 

 between their external form and their structure has long been 

 known, and, one may add, is not so difficult to account for on any 

 of the theories of their origin commonly received. But how are 

 we to account for such relations as those referred to, when the 

 origin of the more clayey drifts and the Till are being discussed ? 

 Certainly neither the theory that the Till represents a moraine 

 profonde, nor that theory that refers its formation to the action of 

 floating ice, in any form, is adequate to account for the facts. In 

 considering the arguments in support of either of these theories, 

 one cannot help being as much struck with their inadequacy to 

 explain such phenomena as either these or that remarkable trans- 

 portal of erratics uphill far above their parent sources, which is so 

 common a phenomenon in nearly all highly glaciated districts. 



All these difficulties (and many others that we need not consider 

 now) vanish, if we reject these older views, and regard all the 

 forms of drift found in this area as so many modifications of sedi- 

 ment, once dispersed through the mass of the ice, and liberated, 



