liV U. GREIG-SMITH. 493 



is known that othor soil-bacteria have also the power, and pos- 

 sibly chief among these is Rhizohium leyuminosarum, which, as T 

 have shown,* ought to find in soil a medium well adapted to its 

 growth and activities. It is also known that, in artificial culture, 

 it may be very little less active than Azotobacter in its charac- 

 teristic power.! 



I am not aware that any definite experiments have been made 

 to determine the actual numbers of Azotobacter or of Rhizobium 

 in soil, beyond those of Lohnis; and it appears that tliis is an 

 important matter, because the chief role can only be credited to 

 Azotobacter, if it is found that this organism is at least half as 

 numerous as Rhizobium. Lohnis t found but a small niunber of 

 nitrogen-fixing microbes in a gram of soil, and it may be on this 

 account that so little attention has been given to the possibility 

 of organisms other than Azotobacter being chiefly responsible for 

 the fixation that occurs in soil. 



If it can be shown that there are as many as two millions and 

 a half of nitrogen-fixing bacteria other than Azotobacter in a grain 

 of soil, our ideas with regard to the active agent must undergo a 

 complete change. This is what I have found in certain soils. 



The research originated in the desire to obtain a selective 

 medium for demonstrating the number of Azotobacter in soil, but 

 it immediately became apparent that the most numerous nitrogen- 

 gatlierer wfni R?iizobium. Indeed, in all the work with agar-media 

 herein described, Azotobacter was only twice found. I am aware 

 that the enumeration of Rhizobia in soil will, if the method be 

 speedy and convenient, only givt; another factor, although an 

 important one, in arriving at the value of a soil. 



Bearing in mind the selective media used in other domains of 

 bacteriology for isolating certain species, I tried various nutrients 

 in combination, and found the best to be a medimn containiiig 

 1('\ ulose, asparagin, and citrate. To improve this, I tried some 



* These Proceedings, 1906, pp.283, 286. 



fibid. 1906, pp.608, 618. 



X Lohnis, Ceutralbl. f. Bakt. 2te Abt. xii.(1904), 459. 



