338 
THE ECONOMIC ASPECTS OF SOIL EROSION 
A brief comparison of the effect of a heavy rainfall on a newly 
tilled surface hare of vegetation and on a like area which is 
protected hy the natural covering of living and dead plants 
will show tlie peculiar influence of the vegetable shield on the 
history of soils. On wood and grass lands the rainfall has prac- 
tically no erosive action whatever. In the forests the mat of 
decayed vegetation is in most cases able to take in three or four 
inches of water, which it yields up so slowly as to distribute the 
flow over weeks and in such a manner that it removes not a hit 
of the soil. On the meadows the outgoing of the water is more 
rapid ; it may, indeed, pass to the i)ermanent streams quite as 
ra})idly as from the plowed ground, but it is kept from contact 
with the soil hy tlie closely set and entangled stems through 
which it cannot break, even when gathered into considerable 
streams. Unless field mice or moles have made burrows leading 
uj) and down the slopes and thereb}'’ providing a way in which 
the water is able to work below the grass, a rainfall of two inches 
an hour, a rate which may be called torrential, ma}" be carried 
from a large field of well grassed land having a slope of twelve 
feet in a hundred without notably eroding the soil. 
If I were an extreme selectionist I should probably not hesi- 
tate to attribute to their own agency, as developed by survival 
of the fittest, the admirable system by which the plants pre- 
serve the soil on wliich they depend from the rapid degradation 
to which it would be subjected but for this defense. The pro- 
tective w'ork which is here accomplished is indeed more perfect 
than elsewhere. It may be conceived that the plants have pros- 
pered in proportion to the efficiency of the shield which they 
afford to the soil on which their life depends. Interesting as is 
this question, it lies apart from our inquiry, and we must turn 
our attention to the further histoiy of the rainwater. 
{To be coniinued.) 
