ase of light. With organisms of this kind every soil is
probably teeming. A typical example is the organism which is the active
agent in the oxidation of carbonic acid gas, and which has already been
referred to as existing in the soil in such numbers.[59]

_The Second Class of Organisms in the Soil._

The second class of organisms are those which reduce or destroy the soil
constituents. The most important of these, from the agricultural point
of view, are those which effect the liberation of nitrogen from its
compounds. In the putrefaction of organic matter the organisms chiefly
act, it is probable, in the entire absence of atmospheric oxygen; but
it would seem, however, that they may also act in the presence of
oxygen. It is through their agency that the soil may lose some of its
nitrogen in the "free" form. To this class belong the denitrifying
organisms already referred to which reduce the nitrates and nitrites in
the soil.[60]

_Third Class of Organisms._

The third class of organisms are those by whose agency the soil is
enriched. Of this class those fixing the free nitrogen from the air are
the most important. The nature of these organisms is still somewhat
obscure, but that leguminous plants have the power of drawing upon this
source of nitrogen is now a firmly established fact. Further reference
to these interesting organisms may be delayed to another chapter.

The important point to be emphasised is, that for the healthy
development of these organisms, which are so necessary in every fertile
soil, certain conditions must exist. These necessary conditions will be
treated more in detail later on. It is sufficient to notice that they
have to do with the physical properties as well as the chemical
composition of the soil. This furnishes a further reason for the
necessity of having the mechanical condition of a soil satisfactory.

_Recapitulation._

From what we have said, it will be seen that the question of
soil-fertility is a very complicated one, and depends on numerous and
varied conditions; that the properties which constitute fertility, while
seemingly very widely different in their nature, in reality influence
one another to a very great extent; that not merely is the presence in a
soil of the necessary plant constituents necessary to fertility, but
that the possession by the soil of certain physical or mechanical
properties is equally necessary; while, lastly, we have seen that the
presence o