61 
vegetation from their inception. This eliminates fires and other second- 
ary influences as causes for their existence, and pushes their history back 
to the origin and exposure of the soil itself. The exposure of the ground 
evidently did not take place immediately after the recession of the glacier, 
but only upon the drainage of the post-Glacial lakes. Furthermore, unless 
it may be considered that grasslands or tundra formerly covered all of 
the upland soils, they had their beginning only after the final stages of 
drainage, when the finest soils, deposited in the deepest parts of the lakes, 
were exposed. From present available knowledge of the region it is impos- 
sible to determine how long ago this was, or what the time relations are 
between the different districts in which prairies occur. The general out- 
line of post-Glacial history indicates that the Salt Plain is much younger 
than the prairies at higher elevations, and lesser extent of timber develop- 
ment on the former shows that the slow process of displacement of the 
openings is far behind that on the latter. The extreme slowness of decay 
and humus formation, as well as of physiographic change, has been empha- 
sized by the writer elsewhere. These things render developmental changes 
in vegetation much slower than they are in more southern regions, so that 
time correlations with post-Pleistocene vegetations in other regions seem 
quite impossible. 
The early stages of the prairies were probably not hydrophytie except 
in local depressions. There was sufficient general slope to set up a drain- 
age system immediately, so that the first plants must have been mesophytes 
or xerophytes, but their specific nature is unknown. There seems no reason 
to think that prairie formerly covered the sandy uplands in other parts of 
the park. The ability of such trees as Canada spruce to live at the margins 
of existing glaciers and to push far out into the sub-arctic plains of northern 
Canada upon similar soils leads one to believe that it could very well have 
been an early vegetation on these hills, preceded only by a lichen mat of 
some sort. High morainic hills near Lane lake, already described, have 
what may be a relic of this early forest. There is almost no humus in the 
sand, and no evidence of an earlier, grassland vegetation. 
So little is known of the climatic history of the area that it is impos- 
sible to form any other than the most general idea of the course of events. 
So far, there is no evidence, botanical or otherwise, of any other climatic 
change since the last ice, than gradual amelioration. The presence of semi- 
open country eastward from the Great Plains region, in post-Clacial time, 
has been explained on the basis of a warm-dry period which no longer 
prevails (28), but it seems unnecessary at present to project this period 
into the north. If such alternating climatic changes did occur, it seems 
clear that they were ineffective in modifying the essential character of 
the present semi-open lands, which must have remained treeless, and which 
were probably covered w r ith grasses or some kind of tundra that did not 
form much peat. 
A close study of the arctic prairie, north of the tree line, should throw 
much light upon the history of all sub-arctic grasslands. 
91963—5 
