Aug. 1909.] 



Soil Bacteriology. 



219 



less, the growth on the agar plates was apparently all of one 

 species. 



The influence of moisture is perhaps best shown by the re- 

 sults obtained between May 20 and June 2. During this time 

 the soil was more or less saturated. It will be noticed that the 

 number of bacteria diminished quite markedly during this 

 period, and rose again on June 2 nearly to its former height, 

 followed by a decrease on June 9. This was after the per cent, 

 of moisture in the sandy soil had fallen from 27.9 to 25.2 per 

 cent., and from 28.9 to 26.8 in the silt. 



A better idea of the influence of deep plowing can be ob- 

 tained by arranging the averages of tables from 1 to 6 in one 

 table and the averages from tables 7 to 12 in another and com- 

 paring the results : 



TABLE XIII. Sandy loam. 



TABLE XIV. Silt loam. 



The average in tables XIII and XIV does not give the total 

 number of bacteria in a surface foot of soil, but shows the 

 ratio of the number of bacteria. For instance, in table XIII 

 (sandy soil) the sum of the averages for the unstirred check 

 plot was 47.28 million. The bacterial contents of 3A, 4A and 



