314 BOTANICAL GAZETTE [OCTOBER 
organism was materially lowered, and to this condition may be 
referred in large measure the decrease in addition of nitrogen. 
A rich garden soil was obtained in the spring of 1915. It was 
known to have a vigorous Azotobacter flora and was examined to 
determine the effect of short periods of drying on its fixing power. 
Tests were made at various intervals up to 30 days. The results are 
recorded in table IT. 
TABLE II 
THE EFFECT OF SHORT PERIODS OF DRYING ON NITROGEN 
FIXING POWER OF GARDEN SOIL 
nitrogen added by ro gm. 
Percentage of Date of tests oti in os Ge As bie Vik * medium 
Orr oa: June 7 | 16.4 
ee See ae yer 17.0 
| 
2s craigs aie Mga 14.8 
DP aR ote ag 18 10.1 
ee ee ee ace July 7 | 10.1 
{ 
The gradual decrease in the moisture content of the soil was not 
reflected in any notable falling off in its nitrogen fixing power until 
the test made after 10 days’ drying. After 30 days, the moisture 
content remaining stationary, there was no further diminution. 
Up to this time the nitrogen fixing flora had retained practically 
60.per cent of its original efficiency for fixing nitrogen. 
The nitrogen fixing flora significantly decreases in its activity 
under the influence of drying, as the foregoing tests indicate. This 
deterioration is manifested as early as two weeks after removing the 
soil from the field and exposing it to the ordinary laboratory con- 
ditions of drying. After 15 months, however, a number of soils 
retained an ability to fix nitrogen that was somewhat surprising, 
which indicates that some species at least have considerable resist- 
ance to drying. A number of qualitative tests for Azotobacter in 
soils kept in storage for 15 months indicate that this organism is 
more-easily attenuated than some other species. 
The striking failure of some soils to fix nitrogen lends itself to 
speculations as to the underlying causes for such a condition. It 
appears, however, that the nitrogen fixing flora is so intimately 
