998 
Soils.” I shall consider this subject under 
several heads as follows, setting aside purely 
physical and chemical phenomena and limiting 
myself to the biologic aspects of the question. 
The origin of nitrates in soil nitrification 
and nitratation. 
The destruction of nitrates in soil denitri- 
fication. 
Methods. 
The actual importance of nitrates. 
Nitrification consists in the conversion of 
ammoniacal nitrogen into nitrites then ni- 
trates, processes shown by the classic work of 
Winogradsky to be dependent on two separate 
groups of bacteria, the nitrite and the nitrate 
bacteria. Wainogradsky’s work has been abun- 
dantly confirmed and is not to be questioned. 
And so far as we know yet all nitrification re- 
sults from this dual activity, the several re- 
ports that have been made of direct nitrata- 
tion of organic nitrogen or of ammonia by bac- 
teria lacking proper confirmation. 
Ammonia is thus essential to nitrate forma- 
tion and ammonification of organic nitro- 
genous substances is an essential preliminary 
step. Ammonifying bacteria are of numerous 
species, indeed some years ago in my labora- 
tory a long search for soil bacteria that could 
not ammonify ended in failure. Ammonifi- 
cation in the light of present knowledge ap- 
pears to be an absolutely essential process in 
the circulation of nitrogen, but the need is 
amply met and in no soil that I know of, and 
we have ourselves examined many hundreds, 
is there any actual deficiency in ammonifying 
power. 
Recently the Rothamsted Station has at- 
tributed low yields to low ammonifying power 
due to consumption of ammonifying bacteria 
by predatory protozoa. This condition may 
exist in exceptional cases, but that it is in any 
wise general is not probable. 
Nitrifying organisms have generally also 
been assumed to be present practically every- 
where and in ample numbers. The results of 
a Bacterial Soil Survey conducted in my own 
laboratories and about to be published show 
that this assumption is not warranted. In a 
large per cent. of the soils tested the N.E. 
SCIENCE 
[N.S. Vou. XXXV. No. 913 
was very low. It appears, therefore, that 
while a deficiency in ammonification is not 
to be feared there may be instances, per- 
haps many of them, where the amount of 
nitrification falls below that of the nitrogen 
equivalent of a good crop. Why nitrification 
is vigorous in some soils and very poor in 
others is not known. In some instances it 
may be referred to unfavorable acidity, mois- 
ture, etc. Many cases are due to causes as yet 
unknown. That abundance of organic matter 
does not inhibit nitrification was clearly 
demonstrated in our own experiments, in 
which vigorous nitrification occurred in pure 
cow manure. 
On the other hand, an exceptional case of 
injury from too great nitrification has re- 
cently been reported by Sackett. 
Dentrification is the destruction of nitrates. 
It is brought about by many species of bac- 
teria and may result in reduction of nitrates, 
to ammonia, various oxides or even to free 
nitrogen. This is unquestionably a detrimental 
process if it proceeds below the ammonia stage. 
The conditions necessary to denitrification are 
usually stated to be the proper organisms, 
moisture and organic matter. The organism 
is conceded to be commonly, almost univer- 
sally present. Yet the fertilizer formule of 
the chemist and agricultural teacher usually 
eall for an admixture of nitrates with dried 
blood, cottonseed meal, ete., thus surely fur- 
nishing ideal conditions for denitrification. 
Theory here opposes practise and as yet no 
decisive experiments have shown which is cor- 
rect. 
Methods.—I can not refrain in passing from 
referring to the absolute necessity in soil bac- 
teriological work of making the tests in soils 
and not in solutions. Winogradsky found that 
in solutions organic matter inhibited nitrifi- 
cation. From this he and others have general- 
ized that it does so in soil, a conclusion that is 
far from the truth, as our experiments have 
conclusively shown. 
THE ACTUAL IMPORTANCE OF NITRATES 
In all of the foregoing discussion we have 
assumed that nitrates are the necessary or at 
