1919] 



ALLEN — AZOTOIACTER CHROOCOCCUM 13 



Agar was prepared by the addition of 1.5 per cent Bacto 1 

 agar to the above. Soil extract agar was prepared by sub- 

 stitution of 1 : 2 soil extract in place of distilled water. A 

 pinch of CaCC>3 was added to each agar tube or culture vessel. 

 In platings special care was taken to get the CaC03 well into 

 suspension before pouring, as growths on the plates were bet- 

 ter when this precaution was taken. 



Although, as stated above, the Beijerinck medium is prob- 

 ably superior to Ashby's, we chose the latter, as it was of 

 known composition, whereas the use of tap water introduced 

 unknown factors. Beijerinck 's solution was used therefore 

 only in a few cases for purposes of comparison. 



Our experimental work is best considered under three sepa- 

 rate heads: (1) the preliminary or orientation work which 

 is concerned chiefly with the checking of important results; 

 (2) the improvement of experimental methods; and (3) the 

 final experiments which are designed to throw light on the 

 reason for improved growth in cultures of Azotobacter me- 

 chanically agitated. 



ORIENTATION EXPERIMENTS 



It seemed to us that it wis well worth while to attempt to 

 duplicate certain phases of Remy and Rosing 's work. At the 

 outset we experienced some difficulty in preparing a stable 

 colloidal ferric oxide solution according to the directions of 

 Eemy and Rosing. This we attributed to the presence of sul- 

 phate ion. After a few empirical experiments, a colloidal 

 ferric oxide solution was prepared as follows: One gm. Kahl- 

 baum's pure ("zur Analyst;") FeCl3 and 10 gms. of saccha- 

 rose were dissolved in approximately 700 cc. of distilled wa- 

 ter, 0.34 gm. Na2C03 added, the solution heated till per- 

 fectly clear, cooled, and made to 1 liter. This solution was then 

 strongly reddish in color, and showed no tendency to precipi- 

 tate when heated to boiling, although a small portion of it 

 did precipitate when added to Ashby's solution and auto- 

 claved. Assuming that the FeCb. 6 H 2 was pure, the solu- 

 tion contained .295 gm. Fe 2 ()3 per liter. This iron-sugar solu- 



1 Prepared by Digestive Fermer ts Co., Detroit, Mich. 



