RELATION OF BACTERIA TO LEGUMINOUS PLANTS 183 



again, being apparently too weak to maintain their entrance. This 

 happens before they have formed nodules. If the inoculating material 

 be somewhat stronger, the bacteria may be successful in their attempts 

 at nodule formation. But no nitrogen-assimilation takes place if the 

 cell-nuclei of the bacteroid tissue are able to offer a sufficiently strong 

 resistance. A still stronger inoculating liquid overcomes this resist- 

 ance and nitrogen-assimilation results. This resistance of the plant 

 indicates that the bacteria are rather of a parasitic than of a symbiotic 

 nature. This is strengthened by the fact that a weakening of the 

 plant after the entrance of the bacteria, results in a total absence of 

 nitrogen assimilation, the bacteria acting as pure parasites. On the 

 other hand, if no weakening of the plant takes place, it is able, 

 subsequently, to turn round on its oppressor, and abstract nitrogen 

 from it, that is to say, the plant becomes parasitic on the bacteria, 

 large numbers of which then assume the bacteroid form. We have 

 therefore what may be called a case of alternate parasitism. The 

 bacteria enter as parasites. If the plant be weak they remain such, 

 but if the plant can sustain the attack, a counter-attack is usually 

 successfully made. 



§8. INOCULATION OF THE SOIL WITH BACILLUS 

 EADICICOLA. 



Obviously it will be advantageous to leguminous plants if the soil 

 on which it is proposed to grow them, be well stocked with Bacillus 

 radicicola. But to attain that end it is necessary to do more than 

 merely bestrew the ground with cultures of that organism. We must 

 take into account the conditions affecting the well-being of the bacteria 

 as well as those affecting the well-being of the leguminous plants. 

 In the first place, the soil must not be too acid, neither must it be 

 deficient in lime, potash, or phosphates. Again, the soil must not 

 be too rich in nitrates, for experience has shown that the leguminous 

 plants prefer their nitrogen in this form, if plentiful enough, and 

 consequently in rich soils bestrewing the ground with solutions con- 

 taining Bac. radicicola has not always led to satisfactory results. 

 Further, it is not sufficient to use any variety of Bacillus radicicola. 

 If, for example, the soil on which peas are to be grown be inoculated 

 with a variety taken from bean nodules the nodular production will 

 probably not be great, so a variety must be used that is accustomed 

 to pea plants. In many cases even this has not been found sufficient 



