262 SCIENCE PROGRESS. 



from nitrogen compounds, were out of all proportion to the 

 supplies of combined nitrogen from all known sources. 

 This result was only obtained, however, with peas, etc., 

 which had plenty of nodules on their roots ; similar plants 

 devoid of the nodules passed into a condition of starvation 

 similar to that to which gramineae succumbed under the 

 same conditions. 



The answer to the third hypothesis was supplied by 

 experiments in pots, where no subsoil existed to store the 

 presumed nitrogen compounds only available to deep-diving 

 leguminous roots. The gramineous roots had as much 

 chance as those of the leguminosae to explore the medium, 

 but the results were the same — using the same nitrogen-free 

 soil for both crops, the leguminosae in some way obtained 

 nitrogen elsewhere, so long as they had plenty of nodules. 



The fourth hypothesis gave trouble. According to it, 

 as we have seen, the leguminous plant does not itself 

 directly accumulate nitrogen from the air, but it, in some 

 way, favours processes by which the soil accumulates it and 

 renders it available to the roots. 



Traces of combined nitrogen always occur in the air, 

 and can be absorbed by the soil. The dust of the air con- 

 tains combined nitrogen, and rain is always bringing down 

 this and traces of ammonia and nitric acid formed by elec- 

 tric discharges. Schonbein and others have stated (though 

 it has been disputed) that ammoniacal compounds are formed 

 during evaporation from moist surfaces or combustion in 

 confined air, and Berthelot stated that even electric dis- 

 charges of low intensity, such as pass between air and soil, 

 can produce nitric acid by forcing the free nitrogen of the 

 air into combination. 



Berthelot had already stated, moreover, that micro- 

 organisms in the soil can "fix" free nitrogen, and build it 

 into their bodies ; though that view had by no means 

 obtained the acceptance it has since received, being then 

 but feebly supported by experimental evidence. 



The debris of organisms in the soil is continually under- 

 going decompositions with liberation of ammonia, and this 

 ammonia owing to the action of porous bodies and alkalies 



