THE EFFECT OF BACTERIAL NUMBERS ON THE 
NODULATION OF VIRGINIA SOY BEANS 1 
By Alfred T. Perkins, 2 New Jersey Agricultural Experiment Station 
Since the discovery that legume 
nodules are caused by bacteria, many 
questions have arisen as to the factors 
which effect the maximum nodulation 
obtainable. One little-considered fac¬ 
tor which affects nodulation is the 
number of Bacillus radicicola organisms 
present in the substrata. It is but 
logical to suppose that the greater the 
number of infecting organisms present 
the greater will be the number of 
nodules produced. The proper num¬ 
ber of organisms to be used for inocu¬ 
lation is a question since most limed 
soils favor the reproduction of this 
organism and, consequently, a scant 
inoculation will serve to assure the 
presence of large numbers of infecting 
organisms. It is maintained by many 
investigators that after legumes have 
become nodulated to a certain extent 
they are immune to additional infec¬ 
tion. If this is so, increasing numbers 
of bacteria should increase nodulation 
only up to a certain point. The 
accompanying data show that nodula¬ 
tion will be limited if the number of 
infecting organisms is limited and 
that there seems to be an immunity 
set up in the plant after it has become 
inoculated to a certain degree. 
The nodulation data were obtained 
from plants grown in washed sand. 
The sand was fertilized by saturating 
it with a mixture composed of 0.08 
per cent of magnesium sulphate, 0.15 
mono basic potassium phosphate, 0.50 
per cent calcium carbonate and 0.05 
per cent ferrous sulphate dissolved 
in distilled water. Subsequently, the 
substrata were maintained 75 per cent 
saturated by daily waterings with 
distilled water. The crop was har¬ 
vested after four weeks’ growth. In 
making the inoculations an emulsion 
from a one-week-old culture of the 
organism was standardized by count¬ 
ing with a haemcytometer and diluted 
to the required degree. The amount 
of water that a given quantity of seed 
would absorb was determined and the 
dilutions were made so that this volume 
of emulsion would contain the desired 
number of organisms. Laboratory 
tests indicated that the fertilizing mix¬ 
ture used would not permit the repro¬ 
duction of the organism in the absence 
of an available carbohydrate but that 
it contained nothing toxic to the 
bacteria. Therefore it is seen that the 
substrata used would keep the num¬ 
bers of infecting organisms practically 
constant. The plants were grown in 
tumblers containing one pound of sand. 
Twenty seeds were planted in each 
tumbler and the seedlings were thinned 
out to the ten best. The experiment 
was conducted in triplicate. 
Table I indicates that maximum 
nodulation may be obtained when the 
number of infecting organisms applied 
is between ten and one hundred per 
seed. Table II shows that maximum 
nodulation may be obtained when the 
number of infecting organisms applied 
is between twenty-five and fifty. 
Table I.— Effect upon nodulation of in¬ 
creasing the number of bacteria 
Tumbler No. 
Number 
of or¬ 
ganisms 
applied 
per seed 
Plants 
inocu¬ 
lated 
Number 
of 
nodules 
per 10 
plants 
1A_ 
0 
Per cent 
0 
0 
IB_ 
0 
0 
0 
2C_ 
0 
0 
0 
2 A_ 
1 
0 
0 
2B_ 
1 
0 
0 
3C_ 
1 
0 
0 
1A_ 
10 
40 
12 
3B_ 
10 
80 
24 
3C_ 
10 
70 
25 
4 A_ 
100 
100 
63 
4B_ 
100 
100 
73 
4C_ 
100 
100 
59 
5 A_ 
1,000 
100 
71 
5B_ 
1,000 
100 
65 
5C__ 
1,000 
100 
80 
6 A_ 
10,000 
100 
72 
6B_ 
10,000 
100 
78 
6C_ 
10,000 
100 
84 
1 Received for publication June 11, 1924. Paper No. 173 of the Journal Series, New Jersey Agricultural 
Experiment Stations, Department of Soil Chemistry and Bacteriology. Part of a thesis submitted to 
the faculty of Rutgers College in partial fulfillment of the requirements for the degree of Doctor of Philoso¬ 
phy- 
2 The author wishes to express his appreciation to Dr. J. G. Lipman for helpful suggestions in planning 
this work. 
Journal of Agricultural Research, 
Washington, D. C. 
( 95 ) 
Vol. XXX, No. 1 
Jan. 1, 1925 
Key No. N. J.-II 
