634 CONTRIBUTIONS TO OUR KNOWLEDGE OP SOIL-FERTILITY, xiii., 



crease during 20 hours at 22° in the extract by the numbers in 

 the control-test. A soil which gives an extract that is neither 

 nutritive nor toxic, shows a growth ratio, E/C = l, a nutritive 

 soil is greater than 1, and a toxic soil is less. 



The numbers in the tables are compared with 100 bacteria 

 added at the start or originally present For the actual numbers, 

 the following are taken from experiments c and d. 



There are two points in the experiments that are not quite 

 clear. These are, (1) the jump of the soil-bacteria in the | toxic 

 soil (Expt. c) from 100 to 232 upon incubating at 28° for 20 

 hours, and the fall of the soil-bacteria in the nutritive soil 

 (Expt.d); (2) the diminution of £ac. j)rodigiosus in a soil which 

 gives a nutritive extract (Expt. d). 



