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



(jiosua were added, together with varying amounts of water. 

 In 20 hours the following increases were obtained. 



EXPERIMKNT i. 



20 gr. of soil with 2 c.c. of suspension and addition of 



10 bacteria became 



No water ^ 10% added. 



3 c.c. of water.. .. = 25 % added 



8 c.c. of water =50% added 



18 c.c. of water =100%added. 



29 

 40 

 55 

 75 



The soil-bacteria were not influenced by the water, as approxi- 

 mately the same number of white colonies developed in all tests 

 in 20 hours. 



An attempt to solve the matter was made in another way. 

 An extract of soil was prepared by using 200 c.c. for 500 gr. 

 The extract, however, was toxic; 1,000 bacteria became re- 

 duced overnight to 35, wliile, in a water-control, they 

 increased to 4,050. 



Dilution of the soil-extract alters the relation between the 

 toxins and the nutrients, and, in place of having a toxic 

 action, the diluted solution becomes more or less nutritious. 

 This is seen in Experiments iv., and xxiii. 



In considering the action of the volatile disinfectants, it is 

 necessary to know if they have any action upon the soil-bac- 

 teriotoxins, either destroying or dissolving, and subsequently 

 translating them to other parts of the soil, as in the case of agricere. 



The Action of Fat-solvent a upon the Soil Bacteriotoxin. — 

 In order to determine if the bacteriotoxin were soluble in fat- 

 solvents, ten experiments were made with chloroform and with 

 ether. The general outcome was that neither of these disin- 

 fectants had any solvent action. When the chloroform-extract 

 of soil was evaporated in a current of air, and the waxy resi- 

 due taken up with warm water, a slight toxic action was 

 noted ; but if the residue was treated with warm saline, the 

 solution was found to be slightly more nutritive than a saline 

 control. Traces of chloroform have no pronounced bacteri- 



