670 CONTRIBUTIONS TO OUK KNOWLEDGK OF SOIL-FERTILITY, vi., 



The experiment generally confirms the previous ones, and shows 

 that the removal of some of the protozoa, by filtering the soil- 

 extract through cotton wool, lias little influence upon the multiplica- 

 tion of the bacteria in the soil, beyond what is to be expected from 

 the behaviour of tlie microbes in the extracts. 



The curves of the last two experiments do not show the sharp 

 I'ise noted in the experiments with protozoa, etc., in the earlier part 

 of this paper. There is little doubt that the rapid rise was occa- 

 sioned by the destruction of the soil-toxins by heat, and, in a con- 

 firmatory experiment, the soil was heated to show that such was 

 actually the case. The same alluvial soil was used as in experiment 

 iv. a, and the incubation temperature was the same, viz. 28". It 

 was lieated at 60" to 70" for lialf-an-hour, but otherwise the con- 

 ditions were the same. A fiftli test was included to show the effect 

 of chloi'oform. 



Experiment vi. — The Use of Heated Soil. 



The numbers are so near to one another, especially with Nos. 2 

 and 3, which received the un filtered and filtered soil-extracts, that 

 the inactivity of the soil-protozoa is well demonstrated. The results 

 with heated soils have confirmed those obtained in the previous 

 experiments with un heated soils. 



The effect of heat alone, and of chloroform alone, is very 

 marked, as can be seen by comparing the curves of the control tests 

 in Experiments iv., a, and vi. These tests were made upon the same 

 soil, containing the same amount of moisture, and were incubated 



