63 



used for tbe inoculation; and that tlie plants tlius inocnlatod and bcar- 

 in<;- root tubercles gaiued larjje (|nantities of nitrojjcn from the air, 



Atwater and Woods* made some two hundred trials w itli oats, corn 

 (maize), and peas in purified sand. In all cases mineral salts and in 

 some cases nitrof:feu in tbe form of nitrates, were supplied in the nutritive 

 solutions. In some cases acpieous infusions of the soil in which peas 

 had <;rown were added to the solutions, in others not. Root tubercles 

 appeared upon the peas, but iu)t on the oats or cotn. There was little 

 api)arent relation between the soil infusions and the number of root 

 tubercles on the peas. Some of the plants had larj?e numbers of tuber- 

 cles, otliers did not, and this was true not only of the plants in different 

 pots, but of the diderent i)lants in the same pot, and it was likewise 

 the case whether the soil infusion had been applied or not. As the 

 sand had been carefully washed and ig;nited atahi<j:h temperature, it was 

 belie\e(l that bacteria or their spores were supplied from the air. 

 Neither oats nor maize showed any jjain of atmospheric nitrogen. With 

 the i)eas, whenever the root tubercles were abundant the gain was large. 

 Where there were no root tubercles there was generally a loss of nitro- 

 gen, and the amount of gain varied with the abundance of the root 

 tubercles, both with and without nitrates in the nutritive solutions. 



Lawes ami Gilbert t describe several series of experiments uiuler- 

 taken for the purpose of verifying Hellriegel's results. They experi- 

 mented with peas, red clover, vetches, blue lupines, yellow lupines, and 

 alfalfa. Their plants were cultivated uiuler three different conditions, 

 (I) in washed sand and fed with no nitrogen; (2) in the same, but 

 inoculated with a soil infusion ; (3) in rich garden soil. Their results 

 diftered somewhat with the different species of legumes, but in general 

 conlirmed those of Ilellriegel. The inoculated plants always developed 

 tubercules and fixed nitrogen, while those uninoculated failed to do so 

 witii any regularity. They did not sterilize their soils and therefore 

 found tubercles present in nearly all of their plants. Their results 

 gave no indication as to whether the tubercles are of any value for the 

 fixation of nitrogen by plants growing in rich soils. The conclusions 

 of Lawes and Gilbert are especially interesting since their earlier work 

 led them to deny the power of plants to fix atmosi)heric nitrogen. 



More recently Laurent J has again confirmed the conclusions as to 

 the relation of tubercles to the power of nitrogen fixation. Ilis work 

 is chiefly confiiuMl to the study of the methods of growing the microbes, 

 and his pa])er gives no results of analysis. lie succeeded, however, in 

 growing the plants successfully in water cultures and getting a luxu- 

 riant development of tubercles and a vigorous growth of the plants. 

 A thorough aeration of the water was necessary for this ]mrpose, and 

 the lack of aeration is oliered as the explanation of the failure of other 



"Connecticnt Storrs Station Bulletiu No. 5 and Annual Report for 1889, and Am. 

 Clieni. Jour., loc. cit. 



t I'roc. Roy. Soc, 47 (1890), Xo. 1. 



t Ann. de I'lust. Pasteur, 1891, pp. lO.VUO. 



