INFLUENCE OF THE ABSOLUTE REACTION OF A SOIL 
UPON ITS A20T0BACTER FLORA AND NITROGEN 
FIXING ABILITY' 
By P. L. Gainey 
Professor of Bacteriologyy Kansas Agricultural Experiment Station 
INTRODUCTION 
In a preliminary report the writer {2)^ called attention to the appar¬ 
ent close correlation existing between the absolute reaction of a soil 
extract and the presence of Azotobacter in the soil. The data presented 
in this report showed that when the hydrogen-ion concentration of the 
soil extract exceeded iXio“® the soil, with very few exceptions, failed 
to initiate the growth of Azotobacter when introduced into a suitable 
mannite culture solution. On the other hand, when the hydrogen-ion 
concentration of the soil extract was less than i X lO”® similar cultures 
almost always developed typical Azotobacter films. The data, though 
meager, indicated that the maximum hydrogen-ion concentration en¬ 
dured by Azotobacter in soils was near that represented by a Ph of 6. 
In the preliminary report only 90 soils were examined and the hydro¬ 
gen-ion concentration of the soil extract was determined by the Clark 
and Lubs colorimetric method as modified for soils by Gillespie (7). 
Some investigators place little credence in the colorimetric method for 
determining hydrogen-ion concentrations, particularly in a medium as 
complex as a soil extract. Gillespie, however, found only slight differ¬ 
ences in the hydrogen-ion concentration of soil extracts determined by 
this method and in suspensions of the same soils determined electro- 
metrically. Since the publication of the preliminary report 418 soils 
from widely varying localities and conditions have been subjected to 
similar examinations, the resulting data being the basis of this paper. 
The hydrogen-ion concentration of these soils has been determined 
colorimetrically upon an extract and electrometrically upon a suspension 
of the soil. 
The soils of Series I were collected within a few miles of the station 
either by the writer or one of his assistants. These soils represent 
practically all types and conditions of soil found in this immediate 
vicinity. Those of Series II were also collected near the station and in 
many instances at or near the point where soils of Series I were taken. 
These collections were made 2 years later to serve as a control on the 
first examinations. Where soils are duplicates of former samples, it 
has been indicated in Table VII by recording the former numbers. The 
soils of Series III were collected, as indicated in Table IX, from different 
counties in the State of Kansas by the county agents. Sterile con¬ 
tainers were sent out and requests were made to collect soils representa¬ 
tive of conditions in the area covered. The directions called for remov¬ 
ing the inch or two of surface soil and collecting from at least four points 
within a short distance of each other in order to obtain as far as possible 
a representative sample of the immediate area. In Series IV ^e soils 
1 Accepted for publication Jan. 22, 1923. Contribution No. 51, Department of Bacteriology, Kansas 
Agricultural Experiment Station. 
* Reference is made by number (italic) to “literature cited,” p. 937-938- 
Journal of Agricultural Research, 
Washington, D. C. 
aey 
(907) 
Vd. XXIV, No. ,i 
Jtme 16,1933 
Key No. K£^.-36 
43325—23 - 3 
