THE ACQUISITION OF ATMOSPHERIC NITROGEN. \2"J 



and Laurent* as well as by Koch and Kossowitschf and others. 

 This fact of assimilation of algae, long overlooked, is of the 

 greatest importance in harmonizing the results of various inves- 

 tigators in studying non-leguminous plants where the amount 

 of nitrogen claimed to be assimilated is always small. 



SOIL INOCULATION. 



Some of the most valuable work in the inoculation of soils 

 with tubercle bacilli is that of Nobbe| and others at Tharand, 

 Saxony. In these experiments peas, lupines, beans, common 

 locust (Robinia pscudacacia), honey locust (Gleditschia triacan- 

 thos) and Laburnum (Cytisus Laburnum,) were used. Both pure 

 cultures of bacteria, prepared from the tubercles of each species, 

 and extracts of soil in which each of the above mentioned plants 

 had previously grown were employed. 



It was found that the extracts of different soils are quite dif- 

 ferent in their action on different plants. Nearly all the plants 

 inoculated produced tubercles, but in varying numbers, and the 

 tubercles were confined almost exclusively to those roots near 

 the surface. 



One very interesting fact in this connection is that where the 

 inoculation of Robinia was successful the amount of dry matter 

 produced and the percentage of nitrogen in the same, were 

 larger than when the plants received a dressing of nitrogenous 

 fertilizers instead. The results obtained from these experiments 

 will be referred to more in detail hereafter. 



Lawes and Gilbert, in 1888, carried on extensive experiments 

 with peas, beans, vetches, lupines, white and red clover, sainfoin 

 and lucern. Plants were grown in sterilized soil; also in rich 

 garden soil to which a watery extract from soils which had pre- 

 viously grown each of the various crops under study, was 

 applied. Favorable results were obtained and the next year the 

 work was repeated. It was found that without the "microbe- 

 seeding," nodules were not formed and there was no gain of 

 nitrogen ; but when the microbes were added, there was nodule 

 formation and, co-incidently, considerable gain of nitrogen. 



*Compt. rend. 115; 732 (1S92). 



fBot. Ztg. 51; 342 (1893). 



JLandw. Vers. Stat. 39, pp. 327-359. 



