igo6.] Inoculation of Leguminous Plants. 657 



nodules, and consequent increased growth, on plants growing in 

 soils deficient in combined nitrogen and altogether free from 

 the organisms producing nodules on the roots of the crop under 

 experiment ? 



If these conditions did actually exist, as shown by the blank 

 pots growing plants free from nodules and starved for want of 

 nitrogen (for the fact must not be lost sight of that leguminous 

 plants can grow to perfection without nodules and entirely at the 

 expense of the combined nitrogen in the soil), it still has to be 

 shown that the soil was in such a condition that the nodule 

 organism when added could exist until such time as it would be 

 able to enter the root hairs of the young plant. This could be 

 shown (1) by positive results ; (2) by the inoculation of the soil 

 with the washings of a fertile soil in which the nodule required 

 was always produced ; or (3) by inoculation of a pot with the 

 crushed healthy nodule of another plant similar to the one 

 under experiment ; a method which a large number of experi- 

 ments has proved never to fail when the soil is suitable for the 

 existence of the organism. 



It is very difficult, if not impossible, to find a natural soil in 

 which no nodules are produced on the roots of a leguminous 

 plant, for if the organism adapted for the plant in question 

 happens to be absent, others of the same species, as Buhlert has 

 shown (Centralblatt f. Bakt, Zweite Abteilung, Vol. IX., 1902, 

 pp. 148-226) will slowly adapt themselves to the plant in 

 question. Hence the necessity of sterilising the soils before 

 use. Here another difficulty arises, for an ordinary soil which 

 contains some 2,000,000 organisms per gram in a well-balanced 

 condition, is completely altered by sterilisation, after which, as 

 some of the experiments showed, it will grow a luxuriant, 

 unbalanced, bacterial flora, which cannot fail to be without 

 direct or indirect action on plant and inoculated organism 

 alike. For the above reasons sterilised quartz sand was also 

 taken for many of the experiments ; but here again it may be 

 urged that insufficient organic food was supplied for the needs 

 of the organism, or that the necessary manurial treatment was 

 not suitable for its requirements. 



Granted that all these conditions in the reported experiments 

 were at least as suitable for the production of nodules as those 

 which occur in nature, there still remains the question of the 



G G G 



