62 
SIR J. B. LAWES AND PROFESSOR J. H. GILBERT ON THE 
vessels, and free aeration had been kept up, there would have been any loss at all 
from a soil containing, as this soil did, less than O'l per cent, of nitrogen. 
It is further to be observed that the losses with growth were, in No. 6, wdrere the 
one lupin plant died before blooming, and where only 7 of the 20 incarnate clover 
seeds grew, and in No. 7 where only one of the two lupin seeds grew, and it gave 
very small produce. In both cases, therefore, there would, independently of the soil 
itself, be decomposing organic matter, conditions under which, in the experiments of 
Boussingault, and also in those at Rothamsted, there was more or less loss, supposed 
to be as free nitrogen. 
Again, in the experiment above referred to, made to determine whether there was 
any material loss as ammonia, Frank used a very unusually rich soil, containing 
1’1836 per cent, of nitrogen, which after exposure for 180 days to a current of air in 
shallow vessels, only contained 1’0976 per cent. It had lost therefore 0'0860 per 
cent., corresponding to 7’27 per cent, of its total nitrogen, of which only 0‘0004 wms 
as ammonia. In reference to this loss Frank says that if such loss is always going 
on in the soil, we must suppose that there is restoration in some way. But it is to 
be observed that the soil in which he found such loss was not only about 12 times as 
rich as the one used in his other experiments, but probably 8 or 10 times as rich as 
the majority of ordinary arable soils. Hence it is obvious that the amount of loss it 
sustained, cannot be taken as any indication of what happens m actual practice. 
Nor can the conditions of the experiments in the narrow and deep vessels wdthout 
ventilation, be considered comparable with those of ordinary arable surface soils, or 
even of subsoils with fairly good natural, or with artificial drainage. 
That soils do lose nitrogen, not only by the removal of crops, but also by drainage 
of nitric acid, is certain; and if there is no return of nitrogenous manures from 
without, the result is a gradual diminution of the fertility, so far as the nitrogen is 
concerned. But the balaiice of evidence is against the supposition that there is a 
constant and considerable loss by the evolution of free nitrogen, in the case of arable 
soils which are only moderately rich in nitrogenous organic matter, and which are 
fairly drained, either naturally or artificially. 
On this point it may be mentioned that, in those of the field experiments at 
Ptothamsted in which the unusual practice of applying farm-yard manure every year is 
adopted, it is found that there is considerable loss of nitrogen from the soil, beyond 
that known to be removed in the crops, and estimated to be lost in the drainage. 
On the other hand, where no nitrogen has been applied for many years, and the 
amount of nitrogen in the siuTace soil is only about, or little more than, OT per cent., 
the loss of nitrogen by the soil over a long series of years corresponded approximately 
with the amounts removed in the crops, together with those estimated to be lost in 
the drainage. Again, when ammonium-salts are applied, even so late in the season as 
October or November, and drainage takes place soon afterwards, the drainage-waters 
will contain amounts of nitrogen showing a very direct relation to the difierent 
