694 CONTRIBUTIONS TO OUR KNOWLEDGE OF SOIL-FERTILITY, iv., 



incubated at 28° for three or four days. The numbers repre- 

 sent the colonies that grew upon plates seeded with the sus- 

 pension from 00001 gr. of soil. With a short incubation 

 period, one cannot expect to get true comparative results 

 when there is a difference in the treatment of the soil -tests. 

 The above result is probably exceptional, and chanced to 

 agree with the tests made with Bac. prodigiosus. A soil which 

 has been pasteurised and treated with chloroform should con- 

 tain fewer bacteria than a soil which has only been pas- 

 teurised ; the double treatment should further reduce the 

 number of bacteria. Thus, at the time when the soil is 

 moistened, there should be a greater number of micro- 

 organisms in the pasteurised than in the pasteurised and 

 chloroformed soil. This will give the former a favourable 

 start, and a considerable time, weeks perhaps instead of 

 days, will elapse before the effect of the soil-treatment can 

 really be seen. For this reason, the experiments with Bac. 

 'prodigiosus are to be preferred, as indicating the true response 

 of the soils to the various methods of treatment. 



After noting that Russell and Golding had said that the 

 phagocyctic protozoa are completely destroyed at 60°, the air- 

 dried soils were heated at 65°-68° for 10 minutes, and tested as 

 in the previous experiment. 



Further sets of experiments were made with the idea of 

 testing the various solvents, and the results confirmed those 

 previously obtained. 



