THE CENTRAL REGION OP NORTH AMERICA. 619 



In these sheltered hollows, and on the shady sides of the higher 

 peaks, are masses of perennial snow, which have no doubt kept 

 up the direct succession from the time when great neves filled the 

 heads of the valleys and the mountains around them were com- 

 pletely snow-clad, and are only waiting some change in the climatic 

 conditions, to advance again down the old valleys and occupy the 

 places they formerly filled. 



State of the Interior 'Region of the Continent previous to the Glacial 



Period. 



Having briefly stated the main phenomena of the Glacial period 

 in the central region of North America, it may be well to recapitu- 

 late and to give some of the conclusions to which I have been led by 

 their study. 



Before the onset of glacial conditions we find the continent stand- 

 ing at least at its present elevation, with its complete system of 

 drainage from the larger river-valleys to many of their less impor- 

 tant tributaries already outlined. Subaerial action must before this 

 time have been in operation for a vast period, all the great features 

 of the western plains having been already marked out, and the re- 

 moval of a truly enormous mass of the soft and nearly horizontal 

 Tertiary and Cretaceous rocks effected. That some very con- 

 siderable changes in the direction of the drainage of the country 

 in preglacial and in modern times took place, however, is probable. 

 An examination of the Lake-of-the- Woods region and a comparison 

 of levels render it almost certain that the waters of the area now 

 drained by its tributary streams then found their outlet southward 

 and westward, towards the present valley of the Red River, and 

 that only after the blocking up of the southern region with the de- 

 posits of the drift did the waters flow over the preexisting breach 

 in the northern rim of the lake, and descend over the surface of the 

 Laurentian to Lake Winnipeg. The Winnipeg River does not show 

 any of the characters of a true river-valley, but consists of eroded 

 and glaciated rock-hollows, from one to another of which the stream 

 falls. There is also some evidence to show that the Red River 

 itself, agreeing with the general structure of the country, flowed 

 southwards ; and if so, the Saskatchewan, too, would probably with 

 it join the former representative of the Mississippi. 



This subject, however, requires a more detailed discussion than 

 can be granted it in this place. 



Mode of Glaciation and Forination of the Drift Deposits. 



To the precise manner in which the Glacial period was initiated, 

 the area now in question gives no clue ; but I have not found, either 

 in the Laurentian region, or over the area of the plains, or in the 

 !Rocky Mountains, any evidence necessitating the supposition of a 

 great northern ice-cap or its southward progress. 



The great drift ridge of the Missouri Coteau at first sight resem- 



