736 CONTRIBUTIONS TO OUR KNOWLEDGE OF SOIL-FERTILITY, vii.-xi., 



inch of rain, which fell upon the surface-area, measured approxi- 

 mately 160 c.c. Each portion, therefore, held 57, 41, and 44 c.c. 

 respectively, or 142 c.c. in all, so that the soil had been leached by 

 the 160 c.c. of rain. 



The experiment was repeated in a modified form, a week later, 

 and during this time, no rain had fallen. A saline solution, con- 

 taining 0-2 % of potassium sulphate, was used in making the ex- 

 tracts, as the flocculation of the clay-particles, by water, retarded 

 the filtration of the soil-suspensions. Instead of heating the ex- 

 tracts at 60°, they were boiled in flasks fitted with aerial conden- 

 sers. In this, and the previous experiment, 400 grams of moist 

 soil were shaken with 400 c.c. of tap-water or saline for an hour. 



Experiment xiii. 



The soil is decidedly, though feebly, toxic, and, according to 

 expectation, the toxins were destroyed by heat, allowing the nutri- 

 ents to produce an increase of bacteria. The subsoil is also toxic, 

 inasmuch as the unboiled extract produced fewer bacteria than the 

 saline control. The action of heat, upon the subsoil-extract, is in 

 contrast with the soil-extract, but is confirmatory of the previous 

 experiment. Thus there appear to be two toxins in soil, one ther- 

 molabile, the other thermostable, unless it is that the latter is a 

 product of the action of heat upon some soluble and filterable soil- 

 constituent. Prolonged or excessive heat develops thermostable 

 toxins in the soil itself, and the thermostability of the extracts of 

 the subsoil, and, in some cases, of the soil, may be due to a similar 

 phenomenon. As it is, we have to deal with a complex action. 



