170 RALPH S. TARR 



ning out from the ice-front, and, by removing the smaller fragments, 

 concentrating the larger ones. Between these two theories I have 

 been unable to find facts to decide. 



Decayed bowlders. — Many of the bowlders have been broken open 

 along joint planes since they were deposited by the ice. Although 

 this is the result of postglacial weathering, it is due to the presence 

 of joints partly developed before removal from the ledge. Other 

 bowlders have disintegrated to gravel hills (Fig. 7) since they were 



Fig. 7. — A granite bowlder in process of disintegration on Cape Ann, Mass. 

 Other bowlders are represented only by heaps of gravel. 



brought to their present position, and every gradation from slight 

 disintegration to instances where only a mass of gravel remains to 

 mark the site of a former bowlder, may be found by the score. This 

 disintegration is surely postglacial; but, taken in connection with 

 other facts indicating preglacial decay, it seems certain that such 

 decay must have been made possible by the fact that, before removal 

 from the parent ledge, the rock was already weakened by decay. 

 Nowhere on the Cape is there such difference in the granite in place 

 as to permit marked decay on one surface and slight decay on another, 

 as is true among the bowlders. On the contrary, wherever the thin 

 till-cover has been stripped from the bed-rock, its surface is found 

 to be polished or scratched, proving that postglacial weathering 

 has done little work. 



