ms present in tubercles or root
excrescences found on the roots of leguminous plants.[82] Not merely has
this been placed beyond doubt, but attempts have been made to isolate
and study the bacteria effecting this fixation. From Nobbe's exceedingly
interesting experiments, recently carried out, it would seem that the
different kinds of leguminous plants have different bacteria. Thus the
bacteria in the tubercle on the pea seems to be of a different order
from the bacteria in the tubercles of the lupin, and so on. This
discovery is of great importance, it need scarcely be pointed out, as it
throws much light on the principles of the rotation of crops.

_Influence of Manures in increasing Soil-nitrogen._

It may be doubted, however, if under any other conditions there is a
positive gain of soil-nitrogen. In other cases the amount in the soil is
only _maintained_ under liberal manuring. In connection with this point
a very striking fact has been observed with regard to the effect of
continuous large applications of farmyard manure. It has been found at
Rothamsted that in such a case, after a while, the manure does not seem
to increase the soil-nitrogen, although where the nitrogen goes to
remains a mystery. In the case of the application of artificial manures,
there does not seem to be almost any appreciable gain to the
soil-nitrogen. The soil-nitrogen is only increased by means of the
residue of crops. In this way, of course, by increasing the amount of
this crop-residue, artificial manures may be said indirectly to increase
the soil-nitrogen.[83]

SOURCES OF LOSS OF NITROGEN.

We now come to consider the sources of loss. The chief source, of
course, is that by drainage. Land under cultivation will suffer very
much more from this source of loss than in a state of nature. Our modern
system of husbandry, involving as it does thorough drainage, can
scarcely fail to very considerably increase this source of loss.

_Loss of Nitrates by Drainage._

The form in which nitrogen is lost in this way is as nitrates. It is a
somewhat striking fact, and one worthy of note, that of the three
important manurial ingredients--nitrogen, phosphoric acid, and potash,
the first of these, in its final and most valuable form, is alone
incapable of being fixed by the soil, and thus retained from loss by
drainage.

As nitrates are constantly being formed in the soil, the loss to its
total nitrogen must be considerable. It is due to th