ld be applied almost always as a
top-dressing.

_Loss of Nitric Acid._

The risks of the loss of nitrogen in the form of nitric acid, although
not so great as they are in the case of ammonia, are still considerable.
As nitric acid is not a base but an acid, what is to be avoided in
mixing nitrates is bringing them in contact with any other manure which
contains another free and stronger acid--as, for example,
superphosphate. The free acid present in superphosphate has the tendency
to drive out the nitric acid from the nitrate and usurp its place. The
risk of loss of expulsion in the above cases is always augmented by the
rise of temperature which invariably accompanies chemical action of any
kind; and although the loss of nitrogen, in the form of nitric acid,
caused by mixing superphosphate and nitrate of soda, might, under
ordinary circumstances, amount to very little, yet, if the mixture were
to be allowed to stand any time, and the temperature of the mass to be
heightened, the loss which would undoubtedly then ensue would be
considerable.

The nitrogen salt which it is safe to mix with superphosphate is
sulphate of ammonia.

_Reversion of Phosphates._

But, as has already been mentioned, there is another loss which may
result from the mixing of manures. This is the deterioration of the
value of an ingredient by reason of change of chemical condition. This
is a source of loss that was little suspected a number of years ago, but
it is now well known that superphosphate of lime, under certain
conditions, is changed from its soluble to an insoluble form. We have
already referred to the reversion of phosphate in the chapter on the
Manufacture of Superphosphates.[247] It was there pointed out that
reversion is often caused by the presence of iron and alumina or
undissolved phosphate, and that the risk of reversion is therefore very
much less in a well-made article, made from pure raw material, than in
one made from a raw phosphate containing much iron and alumina.
Superphosphates containing a large percentage of insoluble phosphates
ought not to be kept too long before being used as a manure, otherwise
much of the labour and expense involved in their manufacture will be
lost by the reversion of their soluble phosphate. Further, it is highly
inadvisable to mix superphosphates with basic slag, which contains a
large percentage of both iron and free lime. Lastly, if it is desired to
mix superphosphate with insolubl