THE ECONOMIC ASPECTS OF SOIL EROSION 
371 
age channel of the area. This underground water, except in the 
rare places where caverns abound, moves very slowly and lias 
no erosive effect. What material it takes away — a relatively 
small amount— is removed in solution. Penetrating to the bed- 
rock, this water, charged with carbonic dioxide and other com- 
pounds which add to its decay-bringing effects, attacks the bed- 
rocks, breaks them up, and with the assistance of the roots of the 
sturdier plants brings the hard stones into the state of soil. In 
this way the natural waste derived from the solutions effected 
by the underground water, the cutting along the streams, and the 
slight wearing of the general surface by water action is compen- 
sated for by the steadfast reproduction of the soil at its base. 
In the state of nature the rate of degradation of the earth’s 
surface over a region such as the Mississippi valley appears by 
the studies of Humphries and Abbot to be not far from one foot 
in five thousand years. At this rate of erosion, we may from the 
field evidence presume that the underground decay will keep 
somewhat ahead of the wearing actions, and so the soil rather 
gain than lose in depth. Under complete tillage, such as is now 
applied, the rate of downwearing will probably become as great 
as that which exists in the valley of the Po, where the surface 
descends at the rate of about one foot in a thousand }’’ears. Under 
these conditions we may be sure that the underground replace- 
ment of the soil cannot compensate for the w'earing, and that 
consequent!}’- the fertile layer will gradually disappear, as it has 
done over large areas in the Old World and is now doing in other 
fields of this country. 
Before proceeding to questions of a distinctly economic nature 
— those which concern the steps which should be taken to arrest 
the wasting of our soils — it will be well for us to consider the pro- 
cesses and rates of erosion on two of the many varieties of soils 
which plentifully exist in this country, namely, those of our 
glaciated districts and those found in the alluvial plains beside 
the true rivers. The first of these classes constitute about one- 
third of the possible agricultural and forestlands of this country ; 
the second is of much smaller aggregate area, but on account of 
its e.xceeding fertility is of almost equal tillage value. 
In glaciated districts exi»erience shows that the risks of de- 
structive erosion are relatively small. This is owing to the fact 
that the drift covering, which in its superficial modification con- 
stitutes the soil of those regions, is almost always com|)osed of 
debris so dee[> and so loosely aggregated that the greater [>art 
