BEHAVIOR OF LEGUME BACTERIA 311 



counts show that calcium carbonate in all proportions hour 

 fitted the growth of Bacillus radicicola from alfalfa. The great- 

 est number occurred in the soil treated with two tons of lime- 

 stone per acre. 



Here again, the red clover bacteria failed to show any gain 

 in numbers in the presence of a basic substance. Large amounts 

 of calcium carbonate retarded the growth of legume organisms 

 from clover. Apparently red clover does best in neutral Miami 

 silt loam soil. 



Effect of Reaction on Growth and Nodule Formation of High- 

 er Plants. — In order to test the effect of reaction on the plant 

 and the formation of nodules it was necessary to grow the plants 

 with a pure culture of Bacillus radicicola. Large test tubes 

 were used. These contained soft filter paper pulp plus mannit 

 solution of known reaction. The same range of reaction was 

 used as in the experiments with solution cultures. 



Figure A of Plate VIII shows the results with red clover. The 

 plants developed best in the presence of 0.5 per cent of normal 

 acid and even 1.0 per cent of the acid does not seriously injure 

 growth. The nodules on the roots were carefully counted. The 

 0.5 per cent of acid also seemed most favorable for nodule for- 

 mation. But, judging from nodule formation in the presence of 

 alkali, it would seem that the latter was less injurious to the 

 bacteria than to the higher plant. 



Figure B of Plate VIII shows the same experiment carried 

 out with alfalfa. An alkaline reaction seems to be favored by 

 this plant. Not only the best growth, but also the greatest 

 number of nodules were found in an alkaline medium. The al- 

 falfa plant and bacteria seem to favor alkali in about the same 

 concentration that the red clover favors acid. 



These experiments were repeated in soft mannit agar and in 

 sterilized soil. In all cases the results confirmed those cited 

 above. The filter paper pulp furnished the best medium for 

 the formation of nodules. 



Further experimentation is needed to test the effects of dif- 

 ferent acids or to represent more closely the conditions which 

 actually exist in an acid soil. 



