122 SOILS AND FERTILIZERS 



were grown in sterilized soils, in prepared pumice stone, 

 and in soils with a limited quantity of nitrogen beyond 

 that contained in the seed. Different kinds of plants 

 were experimented with. The work was carried on with 

 the utmost care and with apparatus so constructed as to 

 eliminate all disturbing factors. The results in the 

 aggregate clearly show that plants, when acting in a 

 sterile medium, are unable to make use of the free 

 nitrogen of the air for the production of organic 

 matter. 39 



130. Atwater's Experiments. Atwater carried on 

 similar experiments in this country. 41 His results in- 

 dicate that when seeds germinate they lose a small part 

 of their nitrogen, and when legumes are grown in a 

 sterile soil, but are subsequently exposed to the air, a 

 fixation of nitrogen may occur. He ascribed this gain to 

 micro-organisms or other agencies. 



131. Field and Laboratory Tests. --By a five years' 

 rotation of clover and other leguminous plants, Lawes 

 and Gilbert found a soil gained from 200 to 400 

 pounds of nitrogen per acre, in addition to that removed 

 in the crop, while land which produced wheat contin- 

 uously, gradually lost nitrogen. The amount in the 

 subsoil remained nearly the same. These facts plainly 

 indicated that crops like clover have the power of gain- 

 ing nitrogen from unknown sources. The results of 

 prominent German agriculturists led to the same con- 



