APPENDIX. 479 



perienced in the same parallel in the interior of the 



continent. 



The phenomenon of the isothermal lines sinking on 

 the western coast of Hudson's Bay, instead of rising 

 as they do on the eastern coast of Europe, has been 

 variously accounted for. Dr. Brewster assumes two 

 northern poles of cold, and places one of them on the 

 meridian of 92°, which is the longitude of Churchill ; 

 but we think that the peculiarities of the climate of this 

 part of the country may be greatly owing to the con- 

 figuration of the land. The coast to the northward is 

 deeply indented by gulfs and sounds, and fringed by 

 numerous islands, among which the drift ice is detained 

 until late in the season. This melting depresses the 

 summer heat ; while the ice-covered sea has little or no 

 effect in tempering the cold during the winter. The 

 subsoil north of latitude 56° is perpetually frozen, the 

 thaw on the coast not penetrating above three feet, and 

 at Bear Lake, in latitude 64°, not more than twenty 

 inches. The frozen substratum does not of itself 

 destroy vegetation ; for forests flourish on the surface at 

 a distance from the coast, and the brief though warm 

 summer gives birth to a handsome flora, matures several 

 pleasant fruits, and produces many carices and grasses. 



The direction of the northern termination of the 

 woods shows the gradual ascent of the isothceral lines 

 (or lines of equal summer heat) as they recede from 

 Hudson's Bay. On the coast near Churchill the 

 woods cease about the 60th parallel ; but at the distance 

 of fifty or sixty miles from the sea their boundary 

 rises rapidly to the northward, and then takes a nearly 

 straight W.N.W. course until it reaches Great Bear 

 Lake, in latitude 65°. The most northerly tree is the 

 white spruce ; but the canoe birch, which is deciduous, 



