NATIVE WOODY PLANTS OF THE UNITED STATES 3 
Many workers are of the opinion that a plant with a large root 
system is necessarily the best one to hold soil in place. This idea is 
so universally held that it may not be amiss to review briefly just 
what takes place when a soil is eroded, especially by water. In doing 
this we may come to understand more clearly how erosion can be pre- 
vented by restoring vegetation. 
In exceptional and rare instances, washing may take place below 
the surface, but under ordinary conditions the washing away of soil 
occurs at the surface. It is the movement of soil particles downhill in 
water or into the air as dust that we wish to prevent. Since such 
movements start in the very uppermost layer of the soil, our efforts 
must of necessity be directed to holding topsoil in position. Roots of 
woody plants do not, in themselves, offer much help in the top inch or 
so of soil; they penetrate deeper and may be said to be holding the 
soil in place below but not necessarily at the surface. 
Observations in the field show that soil easily washes away from 
roots and that the better the surface of the soil is covered, the less 
the soil washes. Experimental evidence has confirmed such’ observa- 
tions. Kramer and Weaver (346)? conducted a series of tests on 
many kinds of plants, mostly grains and other herbs. In their experi- 
ments they discovered that it was not the soil-binding effect of roots 
that produced the best protection, but the plant cover, which pre- 
vented most of the water from coming in direct contact with the soil. 
When the plant cover was intact its effectiveness in controlling erosion 
exceeded that of underground parts alone many times, and com- 
paratively little relation was found between the quantity of under- 
ground parts and resistance to erosion. They found also that cover 
need not be living to be effective; any kind of cover protects the soil. 
In the course of their experiments they noticed accidentally that a 
single elm leaf protected the soil below it until a column over 3 inches 
high, capped by the leaf, remained after the surrounding soil had 
been washed away. Undercutting eventually toppled the column. 
These experiments would appear to justify the use of a mulch in 
planting on eroded lands. 
It is clear that since erosion is greatest at the surface the roots 
of woody plants have comparatively little to do with its prevention. 
As a matter of fact, when water washes the covering soil off and 
begins cascading over roots, they may only aggravate soil washing. 
The surface layer of the soil can best be held in place by some sort 
of protective cover. This may consist of close-growing plants, litter, 
or a combination of the two. An efiective protection for soil is 
afforded by a mat of grasses, which are among the best of all plants 
for erosion control. An equally good cover consists of an established 
forest and the duff produced by it. A third type might be made up of 
close-growing, thicket-forming shrubs and the litter produced by them 
or of a mat of entangled vines. Run-off and removal of soil are 
reduced to a minimum under such covers as these. It is almost un- 
necessary to mention that grass sod can be formed more quickly than 
can a thicket of shrubbery and that the establishment of a forest 
and forest litter takes longer than either of these. 
2 Italic numbers in parentheses refer to Bibliography, p. 293. 
