32 
it, arborescent growth would ultimately prevail on account of the peren- 
nial character of this kind of vegetable life and its power to shade out 
the lower vegetation. In a large part of the world this victory is seen 
to be attained in a few years, or at least in a lifetime. In other parts 
it may take geological ages to establish the arborescent growth against 
the lower vegetation and against unfavorable climatic conditions. Such 
are mainly the interiors of large continents and those localities which, 
for cosmic and orographic reasons, have a climate unfavorable to vege- 
tation in general. This unfavorableness, as a rule, is mainly to be found 
in moisture conditions—uot necessarily deficient rainfall, but an unfavor- 
able balance between the factors of conservation and of dissipation of 
moisture. In such localities the progress of the forest growth contend. 
ing for supremacy must be a gradual advance from the more favored 
border land, but the extension of its area, if not interfered with by man 
and beast, though slow, is as certain as in the more favored localities, 
where it proceeds rapidly. 
It is conceivable, then, that while admitting the unfavorable rainfall 
and moisture conditions in parts of this region as a potent cause in 
making forest extension difficult, this extension would yet have taken 
place if fire and the tramp and browsing of buffaloes had not prevented 
it; or, since the scattered tree growth found on this area suggests that 
forest growth once existed, it would now exist if fires had not destroyed 
much of it, thus disturbing the conditions which were favorable to the 
conservation of the scanty moisture, reforestation being prevented by 
continued fires and countless herds of browsing and soil compacting 
buffalo. 
Where, during the months of vegetation, May to August, the relative 
humidity of the air sinks as low as 50 per cent and the rainfall is less 
than 2 inches, it may be questioned whether tree growth could main- 
tain itself before the conditions of the surrounding country have been 
in a measure modified; or unless we may be able to find and introduce 
species not native, which, like some of the desert trees, the tamarisk, 
_ for instance, can exist for several years without rainfall. 
There is a very peculiar and intimate connection between vegetation 
(especially forest vegetation) and climatic conditions. Most naturalists 
will tell you that vegetation depends upon climate. So it does; but 
there is also areaction of vegetation upon climatie conditions, and this 
truth is strikingly expressed by the poet, who speaks of “Africa’s arid 
sands, where nothing grows because it does not rain, and where it does 
not rain because there nothing grows.” This sounds like a paradox, 
and yet it is true; there is such a relation between the forest cover 
which shades the ground, and the changed temperature and water con- 
ditions of soil and air under its cover, as to favor its own development, 
so that we can say “the forest creates its own favorable conditions of 
growth.” But it is not the single tree that has such an effect, nor a few 
scattered trees; it is by effective shading of the ground that the 
