ROOT HABIT IN THE FAR NORTH 225 



the cases of many cultivated plants and Cannon' has worked them 

 out in great detail for some desert species. It is further possible 

 that not only do plants differ in their root habits, as they do in 

 their top habits, but also that they differ in the rigidity with which 

 the habit is maintained under varying external conditions. If 

 the root habits of some trees are rigid it would not be expected ' 

 that those characterized by deep roots could exist in these north- 

 ern regions, wholly apart from any direct effect of low tempera- 

 tures upon their physiological processes. It is further possible 

 that trees with sufficiently flexible root habits may be found in 

 these thin-soil regions even though they may, under more favor- 

 able circumstances, penetrate to considerable depths. 



The root habits of several trees were investigated chiefly ^ 

 during the summer of 1916 in a strip of country lying between 

 latitude 55°N. and 56°N. and longitude 96°W and 98°W. in the* 

 province of Manitoba. This strip covers a portion of the region 

 of deep, uniform clay soil in the area formerly occupied by Lake 

 Agassiz and a portion of the shallow-soil region beyond the limits 

 of this glacial lake. For comparison, the root systems of the 

 same species were also investigated in a deep (several hundred 

 feet) sandy soil in the neighborhood of the Brule river near the 

 south shore of Lake Superior in Douglas County, Wisconsin. 

 The characteristics of these root systems, in so far as they bear 

 upon the points at issue, are exhibited in the accompanying figures. 

 It should be noticed that these figures are dimensioned and not 

 drawn to scale. To present a clearer idea of the general char- 

 acter of the systems the very small roots of some of the trees are 

 not indicated in every place they occur; instead each system as a 

 whole (or such a portion in the case of an extensive system that 

 an idea of its character may be obtained) has been drawn and a 

 few roots only are represented in detail. Each drawing is that 

 of a portion of the root system of a particular individual, which 

 was obtained by dissection from the soil mass with a pocket 

 knife. In the drawings, principal roots are shown as in a plane 



1 Cannon, W. A., The root habits of desert plants. Carnegie Institution of 

 Washington Pub. 131. 1911. 



