730 CONTRIBUTIONS TO OUR KNOWLEDGE OF SOIL-FERTILITY, xiv. 



dilute bouillon in the presence of several disinfectants. A similar 

 stimulation may also be exercised upon the bacteria that survive 

 the partial sterilisation of soils. As there are so many reasons 

 to account for the increase in soils, it was believed that, if the 

 disinfectant did act as a stimulant, the action could be more 

 clearly shown in porcelain-filtered extracts of soil, treated directly 

 with the disinfectant, and seeded with soil-bacteria. 



Extracts of soil were accordingly prepared by shaking a 

 quantity of soil with an equal weight of water, and filtering the 

 extract thi-ough w^atei' and then porcelain. A volume of the 

 sterile filtrate was treated with one-fifth volume of chloroform- 

 water of two strengths, and seeded with a suspension of Bac. 

 prodiyiosns or of soil-bacteria. After an incubation of 22° for 

 20 hours, the bacteria were counted by the plate-method. 

 ExPEKiMKNTS 1. and ii. 



'•«" was ordinary soil-extract; ■"/>" was the extract of the residue fronj "a"* 

 The experiments show that traces of chlorofoi-m, varying from 



13 to 220 parts per million of soil-extract, stimulate the growth 



of Bac. prodigiosus. 



Chloroform rapidly deteriorates in aqueous solution, as is well 



known, but although a portion would disappear during 24 hours' 



incubation, enough has been left to exert a stimulative action. 



