10 



THE ROYAL SOCIETY OF CANADA 



that period and registers a very severe cold spell. As the soil was 

 frozen twenty- two inches before this moisture increase started, the 

 frost cannot be held responsible for it. The decrease in bacteria on 

 the five and ten inch levels runs parallel with the lowering of the 

 temperature. The cold wave starting the second week in January 

 and reaching its lowest point in the second week of February. 



Nowhere in the diagram do we find any evidence of a considerable 

 bacterial increase in the frozen soil, and in the same way any evidence 

 is missing that bacteria are brought up from lower levels by the frost 

 action. 



Conclusions. 



1. Severe frost stopped bacterial development entirely, causing 

 a decrease in the numbers of bacteria. 



2. Milder conditions, but still below the freezing point, allowed 

 a slight increase in the number of bacteria. 



3. A bacterial increase was noticed under moderate frost con- 

 ditions in December, due to the presence of fermentable raw material 

 in the soil. 



Experiment II. 



Table lia. 

 Sod Covered Field Eight Years Old. 

 Average number of bacteria (in millions) per gram of soil at different depths. 



^ Different place of sampling in same field. 



