ances being washed out of the soil--a property
which, as we shall afterwards see, is possessed also by the clay
contained in greater or less quantity in most soils. On the other hand,
the air and moisture which penetrate the soil cause its decomposition,
and the carbonic acid so produced attacks the undecomposed minerals
existing in it, and liberate the valuable substances they contain.

In considering the composition of a soil, it is important to bear in
mind that it is a substance of great complexity, not merely because it
contains a large number of chemical elements, but also because it is
made up of a mixture of several minerals in a more or less decomposed
state. The most cursory examination shows that it almost invariably
contains sand and scales of mica, and other substances can often be
detected in it. Now it has been already observed that the minerals of
which soils are composed, differ to a remarkable extent in the facility
with which they undergo decomposition, and the bearing of this fact on
its fertility is a matter of the highest importance, for it has been
found that the mere presence of an abundant supply of all the essential
constituents of plants is not always sufficient to constitute a fertile
soil. Two soils, for instance, may be found on analysis to have exactly
the same composition, although in practice one proves barren and the
other fertile. The cause of this difference lies in the particular
state of combination in which the elements are contained in them, and
unless this be such that the plant is capable of absorbing them, it is
immaterial in what quantity they are present, for they are thus locked
up from use, and condemn the soil to hopeless infertility.

It is admitted that unless the substances be present in a state in which
they can be dissolved, the plant is incapable of absorbing them; but it
is a matter of doubt whether it is necessary that they be actually
dissolved in the water which permeates the soil, or whether the plant is
capable of exercising a directly solvent action. The latter view is the
most probable, but at the same time it cannot be doubted, that if they
are presented to the plant in solution, they will be absorbed in that
state in preference to any other. Hence it has been considered important
in the analysis of a soil, not to rest content with the determination of
the quantity of each element it contains, but to obtain some indication
of the state of combination in whic