followed by an increased production of ammonia ; there is a delay of 

 two or three weeks to which our present knowledge affords little 

 clue. 



The loss of nitrogen is partly due to a definite evolution of 

 gaseous nitrogen. This does not occur in entire absence of air or in 

 complete access of air, but only under intermediate conditions of 

 aeration. This result is of interest as showing that the evolution of 

 nitrogen is due neither to a simple oxidation nor to a simple reduction 

 but to some more complex action. The application of the discovery 

 to manure heaps has already been mentioned and is further discussed 

 on p. 28 ; its application to the particular soil problem under consider- 

 ation is, however, less easy. 



III. THE STUDY OF THE ORGANISMS OF THE SOIL. 



Mr. Appleyard has shown that the numbers of bacteria are pro- 

 foundly affected by the soil temperature in late autumn, winter, spring 

 and early summer, but during summer and early autumn soil moisture 

 is more important, and rainfall still more so. The effect of rainfall has 

 been studied by Mr. Richards, who finds that rain always brings down 

 oxygen in solution, so that dissolved oxygen is maintained in direct 

 contact with the plant roots and the soil organisms. Experiments are 

 now in hand to study the effect of this renewal. 



But these seasonal factors do not account for the whole of the 

 variations in bacterial numbers ; some other factor is clearly at work. 



Miss Crump has therefore developed the dilution method of counting 

 soil protozoa and has improved it considerably ; she has made system- 

 atic counts of bacteria and protozoa in two of the field soils at intervals 

 of about ten days during over two years, and has plotted the results in 

 a series of interesting curves. The results show beyond doubt the 

 existence of a living protozoan fauna in the soil, multiplying and dying, 

 and fluctuating considerably in numbers ; the amoebae vary in numbers 

 according to the soil conditions from a few hundreds up to 50,000 per 

 gram ; the flagellates vary up to 100,000 per gram, while there are 

 also numbers of thecamoebae which are now under investigation. The 

 bacteria do not exceed, and rarely even approach 50,000,000 per gram ; 

 their number is usually 10,000,000 to 20,000,000 ; and as the protozoa 

 are much larger than the bacteria, it is evident that the total mass of 

 protozoa is comparable with that of the bacteria. 



Further, during the summer and autumn it is found that the number 

 of baci.eria present is closely connected with those of the protozoa ; 

 when one is high the other is low, and vice versa. This, of course, is 

 what was expected on the view already put forward, that the protozoa 

 are detrimental to the soil bacteria. A new possibility is also opened 

 up, however ; the bacteria may be detrimental to the protozoa. 



The protozoan fauna is very interesting and its mode of life urgently 

 needs working out. Miss Crump has made considerable progress with 

 these difficult problems ; the results promise well. 



The possibility of the production of toxins in the soil by bacterial 

 action has been studied by Drs. Hutchinson and Thaysen, but no 

 evidence whatsoever could be found justifying the belief that they 

 are present. 



