51 



This plowing was done in November, however, after there had 

 been several hard frosts, which have the effect to drive the ants deeper 

 into the ground to escape the cold. 



No accurate weather record was kept at the Bloomington experi- 

 mental field, but the observers' notes show that three "rains" (one very 

 heavy) and seven "showers" (four of them "slight" or " very slight") 

 fell during the forty-nine days from July 5 to August 22. The longest 

 interval without rain was from July 30, when there was a "shower," 

 to August 10, when there was a "slight shower." The "rains" fell on 

 July 5 and 26, and August 13. 



As this field was plowed for us too late to mature a crop, the ef- 

 fects of the several treatments could be learned only by an inspection 

 of the growing corn. The principal observations were made by dig- 

 ging up ninety hills (practically one complete row) in each plot on 

 seven different dates from July 28 to August 18 — making a total of 

 3,150 hills examined — and noting for each the number of hills in- 

 fested by ants and by aphids respectively, and the total number of 

 ants and aphids in the infested hills. 



Bloomington Experiment, 1909. Infestation of Plots 1-5, 6 Days 

 AFTER Planting. 90 Hills from each Plot 



Assuming that the check plot was a fair index to the general in- 

 festation of the field, and that the ninety hills dug in this plot July 28, 

 six days after planting, were a fair sample of the 2,100 hills in the 

 whole plot, and making a like assumption concerning the ninety hills 

 dug this day in each of the other plots, we find that disking three 

 times with a 20-inch disk with intervals of three days between the 

 successive disking, and harrowing once the next day after the last 

 disking, had reduced the number of hills infested by ants from 51 out 

 of 90 in the check plot down to 29 out of 90 in Plot 2, a decrease of 

 43 per cent ; that the total number of ants in the infested hills had been 

 diminished by 18 per cent ; and that the number of hills infested by root- 



