Andrew Cunningham 



60 



protozoa will be excluded. It appears, therefore, that saponin is of 

 little value for this purpose and its use has been abandoned. In the 

 work described by P. T. Miiller the action of the saponin was quite 

 satisfactory. But it must be noted that water was employed as the 

 medium, not a nutrient solution. 



Recourse was next had to the simple method of inoculation of the 

 solutions with bacteria alone and with bacteria + protozoa. 50 c.c. 

 quantities of 1 % bloodmeal solution (filtered) + -05 % K2HPO4 were 

 employed. One flask was inoculated with bacteria + protozoa from 

 a culture of protozoa from soil, the other received as nearly as possible 

 an equal inoculation from the same culture of bacteria alone. The 

 method of inoculation was the single drop method already referred to. 

 Table 14 shows the numbers of bacteria and protozoa developing in 

 the solution. 



Table 14. 



It will be observed that in both experiments the solutions to be 

 compared started with practically equal inoculations of bacteria and 

 that the subsequent depression in the bacterial numbers is marked and 

 runs more or less parallel with the numbers of active protozoa present. 

 In Experiment B, after 20 days in addition to the 1600 cihates given 

 at least -50,000 cysts were counted. This accounts for the very rapid 

 fall in the number of bacteria between the tenth and twentieth days. 

 The results after 30 days indicate very clearlj' the destructive power 

 of the protozoa. In A, flagellates only were j^resent ; in B ciliates only, 

 and as was to be expected the results show that the latter are the more 

 active in the kilhng off of the bacteria. 



Journ. of Agric. Sci. \ai 6 



