On the Food of Fhnifs. 
519 
has been obtainod IVoin the air. The soil cannot possibly have 
yielded it in sufficient quantity to form the v arious azolized pro- 
ducts always present, in very notable proportion^ in such cases : 
we must once more look to tlic atmosphere. 
Again, to use an illustraiion of Boussingault,* let us suppose a 
well-managed farm, so extensive and so well conducted that no 
importation of manure from beyond its limits is ever practised : it 
is easy to see how, by duly apportioning the land between the 
culture of grain and other produce, and the rearing of animals, 
the dung, &c., of the latter may be made fully sufKcient for the 
purposes of the former. Now, every year a large quantity of 
nitrogen, principally in the shape of wheat and other dry grain, 
is sent out of the farm, and exchanged for money and other things 
which may be required, yet at the expiration of a term of years 
the land shall be even in a better state than at first. It is im- 
possible to arrive at any other conclusion than that the excess of 
nitrogen spoken of has been taken up from the air. 
What is here treated as a supposition has been very recently 
demonstrated f to take place in ordinary culture : the nitrogen 
supplied by the manure laid on the land has been positively 
shown to be totally insufficient to furnish the different crops with 
their azotized matter. These researches were conducted by a 
method which in careful hands cannot fail to lead to results infi- 
nitely better and more trustworthy than those got by causing 
plants to grow under unnatural circumstances. We are indeed 
deeply indebted to Boussingault for having applied to the solution 
of questions like the present the exact processes of modern or- 
ganic chemistry. 
A piece of cultivated land was marked out and treated in the 
same manner as the rest. A strict account was kept of every- 
thing relating to it — the quantity of manure j)ut upon it, the 
weight of the different crops raised, and so forth, during the 
whole period of five years occupied by a complete rotation. 
Having these data, by the aid of a sufficient number of exact 
analyses of the manures and products to get a mean result suffi- 
ciently near the truth to be trusted, we can easily understand that 
a good approximation was arrived at towards a knowledge of the 
relative proportions in which the earth and the air contributed to 
the sup])ort of the plants. 
I subjoin one of the tables given by the author, in which some 
of his results are jiresented in a very convenient form, and refer 
my readers to the paper itself, which deserves a careful study. 
* Ann. Chim. et Phys., C7, 14. 
t Boussingault, Ann. Chim. et Phys., Feb. 18il. 
