838 THE JOURNAL OE GEOLOGY. 



The topographic relations of the great sheet of drift make it 

 clear that the agencies which produced it must have been 

 measurably independent of topography over the greater part of 

 the drift area, but forces which were dependent upon topography 

 carried stratified drift into valleys far beyond the borders of the 

 area which is generally drift-covered. 



Topographic termi?uis of the drift. Remembering that most of 

 the drift was transported in a general southerly direction, it is 

 a not insignificant fact that the line marking its southern 

 boundary is largely independent of present topography. The 

 drift does not rise to a given altitude and then fail ; neither 

 does it descend to a certain level below which it does not occur. 

 In general, its southern boundary may be said to be lower than 

 the larger part of the area which it covers to the north. In view 

 of the direction in which it was transported, its distribution with 

 relation to present topography is such as to indicate that in the 

 process of distribution it was frequently stopped on a downward 

 slope. The only escape from this conclusion would seem to lie 

 in the assumption that the land surface has been extensively 

 deformed since the drift was deposited, an assumption for which 

 we have no warrant. In detail, the terminus of the drift is now 

 on level lowland, now on level highland, now on a surface sloping 

 toward the direction whence the drift came, and now on a surface 

 sloping in the Opposite direction. At some points, the terminus 

 is found upon hilltops, while at others closely adjacent it is in the 

 bottoms of valleys. The margin of the drift is therefore far from 

 being horizontal. If it were allowable to suppose that the drift- 

 covered area to the northeast had been elevated since the drift 

 was formed, or that the driftless territory to the south had been 

 depressed since that event, we might suppose that some more 

 definite relationship than now appears formerly existed between 

 altitude and topography on the one hand, and drift distribution 

 on the other. But no general northward elevation or southward 

 sinking will account for the topographic irregularity of the 

 border of the drift. The character of this irregularity is such 

 that, taken in connection with its surroundings, it is clearly not 



