268 SCIENCE PROGRESS. 



ing the nitrogen molecule into the sphere of action of some 

 carbon compound that the riven molecules of the gas will 

 enter into combination with the latter before they have a 

 chance of again combining among themselves into the 

 ordinary inert form. At the same time the objection 

 suggests itself that if it is all a matter of an active cell 

 over-supplied with carbo-hydrates, why cannot any vegetable 

 cell do the trick? It seems probable that further research 

 with anaerobic organisms and fermentation may help us to 

 answer this question. 



On the other hand, we ought to be very careful before 

 accepting rough agricultural experiments as too conclusive 

 in these matters. This is not in deprecation of the empirical 

 trials so valuable in all pioneer work, and I would emphasise 

 the fact that we are in need of very much more experimental 

 work of the nature of Caron's (10), especially summarised as 

 cautiously as he has done it. 



Schultz-Lupitz's(32)experiments have clearly proved that 

 the growth of lupins on poor sandy soils does improve them, 

 and in the course of years brings up the nitrogen contents of 

 these soils to a very valuable amount, and Caron, struck 

 with the general consensus of results indicating that bacteria 

 in the soil have great effects on processes concerned in 

 vegetation, set himself the task of examinino- the soils to see 

 how far the artificial infection of soil by various bacteria 

 improves the crops and affects the quantity of nitrogen. 



Shortly summarised, Caron shows pretty clearly that 

 oats, rye, wheat, etc., yield better results when there are 

 plenty of soil bacteria at work among their root-hairs than 

 when the latter are in a soil not thus infected. Caron him- 

 self believes that the bacteria he used fixed free nitrogen, 

 though he admits that the beneficial effects may be due to 

 the bacteria rendering compounds available which would 

 otherwise have remained less so. 



I think his experiments give less support than he believes 

 to the first hypothesis, and that he does not clearly show all the 

 possibilities, but there can be no doubt that his results open 

 up a splendid field of inquiry, which ought to be followed 

 up as vigorously as possible, by means of exactly con- 



