21 



turn to another side of the subject. It has been found by experience 

 that a continual growth of the same crop in a field year after year 

 ensures a gradual failure of the crop, of course assuming that the 

 crop is cut and carried every year This arises from the fact that 

 the particular foods required for that plant are not formed in the 

 earth sufficiently quickly to supply a full amount year after year. 

 But if different crops are sown in rotation, as they do not all require 

 exactly the same proportion of the same compounds, much better 

 crops can be maintained. But in addition to this, certain necessary 

 compounds can be added to the soil as manures, to supply afresh 

 the deficient ingredients. Several manures are used that are 

 specially noted for the amount of N they supply to the soil, either 

 as nitrates or ammonia. These manures are, many of them, very 

 expensive. Some scientific men, working at the chemistry of 

 agriculture, found that when certain leguminous plants were grown 

 in soil, the soil ultimately became much richer in nitrates. Others 

 working at the farmer's side of the question, found that a crop of 

 leguminous plants in the rotation of crops added very much to the 

 success of the latter crops. Many botanists had long known that 

 the Leguminosae, or plants of the pea and bean tribe, usually had 

 little swelliDgs or nodules like little tubercules on their roots, and 

 these nodules were found to be highly nitrogenous. Kecently all 

 these isolated facts, the special property of the chemist, the farmer 

 and the botanist, have bean welded together in a most interesting 

 manner by the bacteriologist who makes a special study of bacteria 

 and all microscopic organisms. It is now known that these nodules 

 are little exuberant growths of the roots containing colonies of 

 bacilh. The irritation of the bacilh causes the extra growth in the 

 root thus forming a nodule. It has been proved by experiment 

 that if the plants are grown in sterilised sand or auy medium con- 

 taining no bacilh, the nodules do not form, but if the bacilli are 

 sown in the soil the nodules form, showing that the nodules are a 

 consequence of the presence of the bacilli. The bacilh exist in all 

 ordinary soils, but if there is a large proportion of nitrates already 

 in the soil, then the tubercules are very poorly developed, as the 

 plant does not need them ; if there is a great want of nitrates then 

 the tubercles are developed to their utmost degree. It is not known 

 yet what exactly happens to change free nitrogen into a form which 

 the plant can make use of, but it is certain now that the bacilli in 

 some way make the free N of the air which exists in the soil avail- 

 able for the plant, which no other agency is at present known to be 

 capable of doing. It seems to have been proved by some good 

 observers that different bacilli are found in different plants, and 

 that the same bacilli will not grow in the roots of different plants. 

 But [ should think that this only applies to different genera or to 

 species of very different habit. It does not seem probable that 



