vii.] HOW IS NITROGEN FIXATION EFFECTED? 207 



however, very much smaller than are fixed by a 

 leguminous plant on whose root the nodules are well 

 developed. To fix the nitrogen, some expenditure of 

 energy is required, and this is derived from the 

 combustion of carbohydrate supplied to the bacteria by 

 the higher plant; indeed it has been observed that the 

 nitrogen fixation and general growth of the Leguminosae 

 is stimulated by a supply of sugar or other carbohydrate 

 to the soil. The organism, Pseudomouas radicicola, 

 appears to be capable of considerable modifications ; 

 in the nodules it forms rather large rod or Y-shaped 

 organisms, but if an active subculture be obtained by 

 inoculation from a nodule into a non-nitrogenous 

 medium as described above, excessively minute rod- 

 shaped organisms appear, generally in rapid motion. It 

 is in this minute unspecialised form that they exist free 

 in the soil, and it has been shown that they infect the 

 leguminous host by getting through the thin walls of 

 the root-hairs. The characteristic Y forms have also been 

 obtained in artificial cultivations by introducing certain 

 substances into the medium. Much investigation has 

 also been applied to the question of whether there 

 is only one kind of bacterium living in symbiosis 

 with all the Leguminosae, or whether there is not a 

 definite race appropriate to each species of leguminous 

 plant, with which it alone can bring about nitrogen 

 fixation to the full extent. The earliest investigations 

 had already shown that lupins and serradella did not 

 develop nodules when infected with an ordinary 

 garden soil, but only when an extract was added 

 from a sandy soil on which these plants had been 

 previously grown ; and Nobbe brought further evidence 

 to show that, though there is very widely distri- 

 buted in the soil an organism which will cause 

 some nodule formation and fixation of nitrogen, 



