102 
SIR J. B. LAWBS AND PROFESSOR J. H. GILBERT ON THE 
amount supposed, even with the ordinary periodic return, and without such return 
gradual exhaustion would be very marked. Indeed, it is well known that there is no 
more exhausting practice than the annual removal of hay without return of manure; 
so that, in point of fact, restoration in anything like the degree supposed certainly 
does not take place. Next to the removal of ha}^ the consumption of grass for the 
production of milk is the most, but still very much less, nitrogen-exhausting; whilst 
if the grass be consumed by store or fattening animals, the loss is very much less still; 
indeed it is very small. 
Obviously, however, it is more important to consider, what is the probable average 
loss of nitrogen over a given area by the removal of crops generally, and not by that 
of grass alone. Moreover, in making such an estimate it is not the total nitrogen of 
the crops that has to be reckoned; but, taking into account the return by manure, only 
tire amount eventually lost to the soil. With the great variation according to circum¬ 
stances, it is of course very difficult to estimate this at all accurately; but we may 
state that two independent modes of estimate lead to the conclusion that, for Great 
Britain for example, the average annual loss of nitrogen is more probably under than 
over 20 lbs. per acre (= 22*4 kilog. per hectare). In fact, the loss by cropping, under 
the usual conditions of more or less full periodical return by manure, is by no means 
so great as is generally assumed in discussions of this subject. 
The loss of nitrates by drainage may, however, in some cases be considerable. 
There may also, under some circumstances, be loss from the soil by the evolution of 
free nitrogen. Such loss may take place in the manure heap, or in soils very heavily 
manured, as in market gardening, for example. But in ordinary agriculture such 
excessive manuring seldom takes place; and the soil is generally much poorer in 
nitrogen than in the cases of the experiments which have been quoted as showing 
great loss from rich soils. Loss may also take place when the soil is deficiently 
aerated ; but here again the conditions of the experiments cited, in which considerable 
loss by evolution of free nitrogen was observed, are not the usual conditions of soils 
in actual practice. Indeed, the balance of evidence is against the supposition that 
there is a constant and considerable loss by the evolution of free nitrogen from arable 
soils which are only moderately rich in organic nitrogen, and which are fairly drained, 
either naturally or artificially. Some illustrations bearing upon this point will be 
found at pages 62-3. 
Again, M. Berthelot thinks it probable, though not absolutely established, that 
there is loss of nitrogen from the plant itself during growth. Long ago, we ourselves 
supposed that there was such loss; but careful consideration of the evidence relating 
to the subject has led us to conclude that it is not proved, and to believe that it 
probably does not take place. It may be observed that when in his vegetation 
experiments M. Boussingault found a loss of nitrogen, there was coincidently some 
decaying vegetable matter, such as fallen leaves ; and in somewhat parallel experiments 
at Rothamsted, no loss of nitrogen was found as a coincident of growth, and in the 
