Journal of Agricultural Research 
Vol. XXIV. No. a 
l88 
If these examinations failed to reveal organisms morphologically 
similar to Azotobacter they were regarded as absent. Owing to the 
above-mentioned complex conditions it is quite possible that Azotobacter 
were sometines reported present when in reality they were absent and 
vice versa. The end to be gained did not seem to justify the large amount 
of time that would be necessary to isolate and identify Azotobacter from 
the various soils. It is believed that if Azotobacter are not present in a 
soil in sufficient numbers and vigor to develop unmistakable evidence of 
their presence by the methods just described, for practical purposes 
they may as well be absent. 
RESULTS 
Several hundred samples of soil from Kansas and other States have 
been examined by the methods described above. The following is a 
comparison of the average quantity of nitrogen fixed by 200 soils. 
All samples. 5.87 mgm. 
Presence of CaCOa. 7.10 mgm. 
Absence of CaCOa. 4.60 mgm. 
Azotobacter film formed. 7.70 mgm. 
No Azotobacter film formed. 4.10 mgm. 
There were only two samples that failed to show some nitrogen fixation, 
and both of these were in media containing no CaCOg. 
When calcium carbonate was added to the medium an Azotobacter 
film was formed from 117 samples, or 58 per cent of the soils. The average 
quantity of nitrogen fixed in tiiese was 8.1 mgm. The average quantity 
of nitrogen fixed in the 83 samples having no Azotobacter film was 
5.7 mgm. 
When no calcium carbonate was added to the medium an Azotobacter 
film was formed from 75 samples, or 38 per cent. These had fixed on 
the average 7.1 mgm. of nitrogen. One hundred and twenty-four 
samples, or 62 per cent, produced no Azotobacter film, and the average 
nitrogai fixed for these was 3.1 mgm. 
Twenty-seven samples, or 14 per cent of all soils examined, fixed more 
nitrogen in the samples to which no CaCOg was added, while 173 samples, 
or 86 per cent, fixed larger quantities of nitrogen in those samples receiv¬ 
ing an addition of CaCOg. The microscope revealed Azotobacter in 
cultures from 130 samples, or 65 per cent of all. No Azotobacter were 
observed in cultures from 70 samples, or 35 per cent of all. Some nitrogen 
fixation took place in practically all samples inoculated regardless of the 
source of the soil. 
There were 12 samples containing Azotobacter or organisms resem¬ 
bling Azotobacter that failed to form an Azotobacter film. The average 
nitrogen fixed by these 12 soils where CaCOg was added was 6.2 mgm. 
The average in the absence of CaCOg was 3.1 mgm. This is 0.5 mgm. 
higher than the average fixed by those giving no film when CaCOg was 
added and exactly the same as those giving no film in the absence of 
CaCOg. It is highly probable, ther^ore, that some soils contain 
Azotobacter but are incapable of initiating the growth of an Azotabacter 
film in a mannite culture solution. 
Practically all soils that failed to produce Azotobacter films formed 
more or less heavy films of fungi in the medium containing CaCOg. 
As a rule, no such films were formed in the medium containing no 
CaCOg. \^ether or not these fungi are associated with the increased 
