GEOGRAPHICAL DISTRIBUTION OF FOREST TREES IN CANADA. 291 



same latitudes. In New Brunswick and Nova Scotia the tree-lines all 

 curve southward, as if to avoid the influence ' of the cold water of the 

 Gulf of St. Lawrence. 



2. The eifect of a different arrangement of land and water at a late 

 geological period is seen principally in the absence of the white cedar 

 from Nova Scotia, and its distance inland on the west side of James 

 Bay. The course of the Banksian pine line round James Bay and north 

 of the St. Lawrence has been caused by a former extension of the sea. 



3. The greater dryness of the region between Red river and the 

 Eocky mountains is one of the principal causes of its treeless and par- 

 tially treeless condition. The former country is called plain, and the latter 

 prairie. The other principal cause is the extremes of heat and cold. Some 

 persons have jumped to the conclusion that the plains and prairies have 

 resulted from repeated fires burning off pre-existing forests in modern 

 times. If this were the case, the tree-lines in their general westward 

 course would all end abruptly at the edge of the burnt space, and some 

 traces of the various species would still be found in the prairie region. 

 But, instead of this, on approaching the prairie country they all curve 

 gradually to the southward in a concentric fashion, so that what was 

 the northern limit of each successive species now becomes in turn its 

 western boundary. Some species reach the Red river valley before 

 they bend south, while others begin their curve at greater and greater 

 distances off. Those affected in this way are the white cedar, black ash, 

 white ash, white pine, red pine, sugar maple, red or soft maple, yellow 

 birch, red oak, hemlock, ironwood, beech, and red cedar. Towards the 

 prairie region these are replaced to some extent by western trees, such as 

 the negundo or Manitoba maple, the green ash, and the cottonwood, which 

 is another indication of climatic change in this region. Such well- 

 marked features of long-established forest distribution are sufficient to 

 prove that the present divisions of prairie and woodland are of ancient 

 date, and not due to the destruction of timber corresponding with the 

 extent of the open country. But if other evidence be required, it 

 may be found in the prairie flora, the absence of hollows, with hillocks 

 accompanying them, left by the bodily upturning of the roots of trees 

 with a large quantity of earth attached, which are always to be found 

 where forests have grown ; the comparative scarcity of water-courses, and 

 the prevalence of regular rings of boulders around the edges of the nearly 

 circular ponds which have resulted from the action of ice in a treeless 

 region. 



4. Extremes of heat and cold have also contributed to prevent the 

 growth of the above-mentioned trees in the plain and prairie country. 



We have already seen that some of our most southern trees may be 

 successfully transplanted far beyond their present natural range ; but the 

 experiments which have been made in the plain and prairie regions seem 

 to show that there only the hardiest trees will flourish at all. If the 

 various species of the forest region lying to the eastward of the prairie 

 region had recently flourished in the now open country, we should expect 

 to find that they would grow again when planted there ; but such is not 

 the case. 



