610 THE FIXATION OF NITROGEN BY RHIZOBIUM LEGUMLNOSARUM, 



fix the atmospheric nitrogen. The . quantity of slime is no 

 criterion of the activity of Azotobacter, for in this experiment it 

 was autodigested after formation. 



Other races of Rhizobium were tested to see if similar gains 

 could be detected, the medium being the same as before and the 

 experiment occupying three weeks. 



The medium had the same composition as in the previous 

 experiment, viz., dextrose, sodium phosphate, etc. The Krai 

 races do not form slime from such a combination of nutrients, 

 and it is seen that they are also unable to gain nitrogen. There 

 was, however, a considerable reproduction, probably from the 

 nitrogen contained as impurity in the agar, etc. These races 

 show that in the absence of slime there may be an increase of 

 bacterial cells but there is no fixation of free nitrogen 



Considering the experiment generally, one sees that a small 

 slime product usually means a small or negligible fixation. One 

 thing, however, has to be remembered, and that is that some of 

 these races exhibit more or less autodigestion of their slime, 

 which though not very evident in seven days, becomes consider- 

 able when the experiment extends over a longer time. The 

 general increase in nitrogen-fixation in the presence of Bac. 

 levaniformans shows that Rhizobium derives some advantage 

 from the presence of this bacterium which by itself cannot 

 elaborate atmospheric nitrogen.* I am inclined to think that if 

 this bacterium can assist in the elaboration, it is possible that the 

 cells of the root-nodule may also assist in a similar manner. 



A number of races of Bac. levaniformans from widely different 

 sources, such as Mauritius sugar, New South Wales sugar, Hakea 

 saligna, Blue Lupin, and a race which had been in the laboratory 

 collection for six years, were all tested in combination with 

 several races of Rhizobium. The experiment showed a general 

 gain of nitrogen and an assistance given to Rhizobium by the 

 races of Bac. levaniformans, all of which had a very uniform 

 action. 



* Seep.617. 



