QUATERNARY AGE. — GLACIAL PERIOD. 537 



bowlders from the Cordilleras, scores of miles to the east of the moun- 

 tains, as well as to the west on Chiloe, observed by Darwin, re- 

 quires the same explanation. 



The- absence of glacier action, over a large part of the region from 

 Virginia to Georgia and Alabama, is shown by the great depth 

 of decomposed rock, covering in situ the crystalline rocks in many 

 places ; at the north, all such soft superficial material was scraped off 

 and carried away by the glacier. 



It hence appears that the glacier theory is alone capable, as first 

 shown by Agassiz, of explaining all the facts. 



The surface of the glacier in North America must have been of 

 unblemished whiteness ; for, from New England to the Rocky Moun- 

 tains, there was not a peak above its surface, excepting the White 

 Mountains, and these probably had their cap of snow. Hence there 

 were, among the depositions, no true lateral moraines, although every- 

 where under-glacier moraines, or linear ranges of stones and gravel. 



3. Probable head and lower limit of the Glacier of Eastern North 

 America : Terminal Moraines. — The direction of the scratches, and 

 the extent of the country they cover, appear to show that the head or 

 zipper part of the ice-mass, over New England, New York, and the 

 Canadian region from Labrador to and beyond Lake Huron, was on 

 the water-shed between the St. Lawrence River and Hudson Bay. 

 The lower limit of the New England portion probably coincided with 

 the outline of the deep-water slope, about eighty miles south of Long 

 Island, and St. George's Shoal, between Cape Cod and its continuation 

 northeastward in Sable Island Shoal, just outside of Nova Scotia. In 

 this part, therefore, the depositions from the melting extremity, the 

 terminal moraines, would have been made in the shallow waters of the 

 ocean's border, and have increased its shallowness. St. George's Bank 

 and Sable Island Shoal may be mainly terminal moraines. Over the 

 continent, to the west, there must have been true terminal moraines 

 formed. But they were mostly obliterated by the floods of the suc- 

 ceeding period. 



The highest ice-surface must have been somewhere in British America, in order that 

 the ice might have moved across the St. Lawrence valley, climbed and passed the 

 mountains near the northern New England boundary, and then, without any essential 

 change of course, have traversed all New England to the ocean on the southeast. The 

 direction of the scratches, in Maine, New Hampshire, and Vermont, and in Canada on 

 the north, in eastern New York, western New York, and the country about Lake 

 Huron, varies from west-northwest on the east to northeast on the west; and the 

 scratches, thus converging, point to the watershed between the St. Lawrence and Hud- 

 son Bay, as the place of origin. The southwestward course of the scratches, about the 

 western limit of this great region, is continued, still farther west, over the Maumee Val- 

 ley, through northwestern Ohio. 



The height of the upper surface of the glacier, at the "White Mountains, as the facts 

 show, was at least 0,000 feet above the sea. According to calculations, the details of 



