EFFECT OF REINFECTION 



277 



soils is obtained by studying the effect of temperature on 

 bacteria] numbers. Untreated soils were maintained at 10, 

 20, 30 C., etc., in a well moistened aerated condition, and 

 periodical counts were made of the numbers of bacteria per 

 gram. Rise in temperature rarely caused any increase in 

 bacterial numbers ; sometimes it had no action, often it caused 



TABLE XCVI. Effect of introducing Untreated Soil into Partially 



Sterilised Soil. 



a fall. But after the soil was partially sterilised the bacterial 

 numbers showed the normal increase with increasing tempera- 

 tures. Similar results were obtained by varying the amount 

 of moisture but keeping the temperature constant (20 C.). 

 The bacterial numbers in untreated soil behaved erratically, 

 and tended rather to fall than to rise when the conditions were 

 made more favourable to trophic life; on the other hand, in 

 partially sterilised soil, the bacterial numbers steadily increased 

 with increasing moisture content. Again, when untreated soils 

 are stored in the laboratory or glass-house under varying 

 conditions of temperature and of moisture content the bacterial 

 numbers fluctuate erratically ; when partially sterilised soils 

 are thus stored the fluctuations are regular. 



7. When the curves obtained in (6) are examined, it becomes 

 evident that the limiting factor in the untreated soils is not the 

 lack of anything * but the presence of something active. 



8. This factor, as already shown, is put out of action by 



* The soils included fertile loams well supplied with organic matter, 

 calcium carbonate, phosphates, etc. 



