THE CENTRAL REGION OF NORTH AMERICA. 613 



greenish, epidotic, and horiiblendic altered rocks. It is interesting 

 to observe that the proportional importance of the Laurentian and 

 Huronian, thus ascertained for the drift, is nearly that of their 

 areas where they have been mapped. The proportions in the drift 

 are respectively three to one. Prof. Bell, of the Geological Survey, 

 has stated the proportion by area of Laurentian and Huronian in the 

 region north-west of Lake Superior as two to one, leaving a slight 

 preponderance of the former over the latter in the drift, as compared 

 with the areas in the metamorphic axis, which arises no doubt 

 from the greater prominence of the harder Laurentian rocks. The 

 limestone is that of the flanks of the Laurentian axis; and its great 

 abundance is an interesting feature, and one tending to prove that 

 this rock must in preglacial times have lapped far up on the Lauren- 

 tian. These three classes are derived from the north-east or east. 

 The fourth or Quartzite drift is a general name which I have applied 

 to that coming from the Rocky Mountains, which, although not 

 entirely composed of quartzite, is characterized by the great abun- 

 dance of that material, and has a peculiar and distinctive appearance. 

 This drift was met with abundantly in many places further west ; 

 but it was only in August last that I was able to trace it to its 

 origin in the mountains. It occurs, as will be noticed, very spa- 

 ringly on this second prairie-level, and is not found over its whole 

 area. The first clearly recognizable fragments were met with near 

 the 101st meridian, 580 miles from the Rocky Mountains, and over 

 200 from the nearest part of the Laurentian region. 



On the surface of this prairie-level there occur some remarkable 

 elevated regions, which seem to be entirely composed of accumulated 

 drift materials (see Map, PL XXXII. ). The most prominent of these 

 are included under the names of Turtle Mountain, Moose Mountain, 

 and the Touchwood Hills . Though quite un connected, these elevations 

 follow in a general way a contour-line of the surface, and form a 

 range roughly parallel to the Coteau, to which in their appearance and 

 material they also bear the closest likeness. Of these elevations the 

 only one which I have personally examined is that known as Turtle 

 Mountain, which is bisected by the forty-ninth parallel and forms 

 the most southern of the series. It is a region of broken hilly 

 ground, which may be about 20 miles square, and is for the most 

 part thickly wooded — a circumstance which renders it a specially 

 prominent feature when viewed across the prairie. Its extreme 

 height is not more than 500 feet above the prairie at its base ; and 

 its general elevation is a little more than 2000 feet above the sea, 

 or nearly the same as that of the surface of the Coteau. On ap- 

 proaching it from the east the already gently-swelling plain becomes 

 more markedly undulating, small basin -like swamps and ponds 

 are more frequent, and its junction with the region of the " Moun- 

 tain " would be undefinable but for the limiting border of the woods. 

 The western end of the mountain is more abrupt towards the plain, 

 and is much diversified with ridges, between which lie swamps and 

 lakes, which show a general tendency to arrangement in north-and- 

 south lines. Towards the eastern end there are somewhat extensive 



