THE CENTRAL REGION OP NORTH AMERICA. 611 



same relative position on the drift plateau ; but I believe that nearly 

 the whole thickness of the horizontal deposit belongs to the great lake 

 of a later period. Stones of any kind are very seldom found on this 

 prairie ; they are so rare, indeed, that those which I have seen during 

 all my excursions over it probably do not exceed twenty in number. 

 They have no doubt been brought to their present position by the 

 shore-ice of the lake itself, and are similar to those associated with 

 the drift- deposits of its bounding escarpments. 



Ascending the front of the western escarpment, it is found, as 

 might almost have been foreseen, to be terraced ; and on leaving the 

 alluvial flat, boulders are again found abundantly, both strewing the 

 terraces and the summit of the "mountain" or second prairie- 

 steppe. The terraces not only occur on the front of this escarp- 

 ment, but extend westward along the banks of the great valley of 

 Pembina River, which at the time of their formation must have been 

 an inlet of the lake, and is therefore probably of preglacial age. 



Second Prairie-Plateau. 



The surface of the second plateau or steppe of the plain (PL XXXII., 

 c) appears to be almost everywhere very thickly covered with drift 

 deposits ; and the undulations and slight irregularities of its contour 

 seem in the main due to the arrangement of these surface-materials, 

 which, though no doubt somewhat modified by subsequent denudation, 

 do not seem to have suffered much. Over large areas no systems of 

 " coulees" or stream-valleys are to be found ; and the generally un- 

 dulated surface must be due to original inequality of deposition, 

 though a certain quantity of material has no doubt been removed 

 from the rounded hillocks into the intervening basin-like swamps and 

 hollows. Such an arrangement not only implies the porous nature 

 of the subsoil, but is in accordance with the comparatively very 

 small rainfall of the region, and would tend to show that at no time 

 since its emergence has the precipitation been great. It was ob- 

 served that in many places boulders and gravel are equally abundant 

 on the crests of the gentle ridges and hillocks and in the hollows, 

 while they are comparatively seldom seen on the intervening slopes. 

 A similar observation has been made by Prof. Bell in a part of the 

 second steppe considerably further north, and would tend to show 

 slight erosion of the surface by marine currents subsequent to the 

 deposition of the heavier materials. 



The drift material is found generally to consist in great part of 

 local debris derived from the immediately underlying soft formations ; 

 but this is always mixed with a considerable quantity of far- trans- 

 ported material, which is generally most abundant in the upper 

 layers. Large erratics are in some localities very plentifully strewn 

 over the plains, but they seem to be almost always superficial. They 

 are generally of Laurentian rocks ; but whitish and yellowish lime- 

 stone, derived from the Silurian flanking the western base of the 

 Laurentian region, is abundant. A bank in Long-River Valley 

 shows in an interesting section, about 30 feet of drift, resting on 



