76o 
Journal of Agricultural Research voi. xxiv, No. 9 
organisms appear to be very close to Ph 5.9 to 6.0. It seemed of interest, 
therefore, to ascertain the effect of varying the hydrogen-ion concentra¬ 
tion of laboratory media upon the activity of a number of strains of 
Azotobacter. 
METHODS EMPLOYED 
Crude cultures of Azotobacter were prepared from a number of soils 
by inoculating a mannite or dextrose cultural solution and incubating 
until a characteristic Azotobacter film developed. Portions of these 
films were then streaked upon mannite or dextrose agar and the streaking 
process repeated from isolated colonies until all colonies developing 
were similar and yielded, when stained, only typical Azotobacter cells. 
No effort was made to identify the various cultures. 
The medium employed had the following composition: Monobasic 
potassium phosphate, 0.2 or 5.ogm.; magnesium sulphate, 0.2 gm.; sodium 
chlorid, 0.5 gm.; ferric chlorid, 2 drops of a 10 per cent solution; dextrose 
or mannite, 20.0 gm.; distilled water 1,000 cc. Flasks of this medium, 
slightly alkaline to brom-thymol-blue, were inoculated from an agar streak 
culture and aerated vigorously at 28° C. until heavily clouded (two to 
four days). The cultures were again stained and examined, and all flasks 
showing morphological forms other than typical Azotobacter were dis¬ 
carded. One to five per cent of such a culture was used as an inoculum. 
The quantity of sterile media necessary to set up an experiment was 
prepared and inoculated. Fifty cubig centimeter quantities were 
measured aseptically into carefully washed sterile 300 cc. Erlenmeyer 
flasks. The quantity of NaOH or HCl necessary to adjust the 50 cc. 
of media to approximately the desired reaction was determined and this 
quantity, previously sterilized, was carefully added to four flasks. The 
hydrogen-ion concentration of the contents of one of these flasks was im¬ 
mediately determined and recorded as the initial reaction. Hydrogen-ion 
determinations were made colorimetrically, a control being occasionally 
run electrometrically. A small quantity of formaldehyde was added to 
a second flask to act as a control with which to compare growth and 
nitrogen fixation. The remaining two cultures were incubated at 28® C. 
for two weeks. The quantity of acid or alkali required to produce a given 
change in reaction varied only slightly from time to time. 
In the earlier experiments, where only 0.02 per cent KH2PO4 was 
used, the buffer index was very low and necessitated the use of very 
dilute (N/80) NaOH in adjusting the reaction. Where 0.5 per cent 
KH3PO4 was used (experiments 16 to 20 inclusive) the buffer index 
was much higher, requiring approximately ten times as much acid or 
alkali to produce a given change in reaction, and N/io NaOH was used 
to correct the reaction. 
During incubation the cultures were examined at frequent intervals 
and the presence or absence of growth recorded. The formaldehyde 
produced a slight change in the physical appearance of the medium in 
the control flasks and rendered it somewhat difficult to detect very 
slight changes in the appearance of cultures. Where it was impossible 
to determine definitely whether growth had taken place it has been 
indicated in Tables I and IV with a question mark. 
After two weeks’ incubation a final examination for growth was made, 
the hydrogen-ion concentration of the cultures recorded, and the total 
nitrogen present determined by the Kjeldahl method. The quantity of 
