82 
The Food of our Agricultural Crops. 
experiment of sand containing no nitrogen, but supplied with 
minerals, and with a very minute quantity of nitrogen taken 
from a fertile soil, where leguminous plants have increased con- 
siderably in nitrogen. Interesting as these experiments are, 
they fail to explain the behaviour of leguminous crops in an 
ordinary agricultural soil. My garden soil is far higher in 
fertility than any which a farmer is likely to cultivate, unless 
he happens to farm one of the black soils of Russia ; and he could 
not afford to cultivate a sand free from all vegetable matter and 
nitrogen, such as was used in the pot experiments. From the 
opposite results obtained in the garden soil and in the sand, are 
we to assume that the power of leguminous plants to obtain 
nitrogen from the atmosphere is dependent upon, and in propor- 
tion to, their inability to obtain it in sufficient quantity from 
the soil ; and that it is only as soils become more and more ex- 
hausted of their fertility that this restorative process comes into 
force ? 
It is exceedingly difficult to account for the large crops of 
clover grown upon the unmanured bean-land. The amount of 
nitrogen taken up by the crop was very great, but it must not 
be forgotten that the quantity of mineral matter taken up was 
very much larger, as we may consider that the proportion in 
the clover-plant is about four parts of minerals to one of 
nitrogen. These minerals must have come from the soil, and 
must have accumulated there, in some form available for the 
clover, during the period when beans were grown upon the land. 
Whether at the same time some compounds of nitrogen suitable 
for the clover were also formed we have no evidence to show. 
An attempt to grow red clover last year where it was grown 
four years previously failed upon all the rotation experiments. 
We did succeed in growing it in 188G, after we had grown a 
large crop in 1882 ; but in the latter year there was a good 
deal of disease in the crop upon the highly-manured land^ but 
none on the land which received only mineral manures. 
From Avhat I have said, it must be apparent that our know- 
ledge at present of this important subject does not go beyond 
the facts that leguminous plants take nitrogen in large quanti- 
ties from the soil, and that, under special conditions, they can 
obtain it from the atmosphere through the agency of micro- 
organisms ; but we have no evidence to show that they do so 
when grown in the ordinary course of agriculture. The most 
probable means by which a solution of this problem can be 
arrived at, is to grow leguminous crops upon ordinary agricul- 
tural land with a liberal supply of mineral manures ; in the 
course of time, the stock of nitrogen remaining in the soil must 
