45i 



planting is shown by the following extract from the field notes of Mr. 

 Sanders, the responsible assistant in charge of this work. 



April 23, raining hard. 



April 24, rained steadily almost all day. 



April 27, raining more than half the time since 11 a. m. yester- 

 day ; now cold with some snow. Ground saturated before these rains. 



May 1, showers today. 



May 6, rainy on the 4th and 5th, the soil now too wet for plowing. 



May 8, rained rather hard last night. 



May 14, rained very hard late the preceding night, the ground 

 still too wet to work. 



May 16, rained during the night. 



May 25 hard rain during the night ; too wet to plow on the fol- 

 lowing day. 



May 28, rained during the night and all the morning. Also rained 

 on nights of 27th, 28th, and 29th, and rained very hard nights of June 

 6 and 7. 



June 12, rained in the afternoon. 



June 15, rains have prevented work in the fields since the 10th. 



It will be noticed that the ground was planted on two days after 

 an interval of seven days since the last-mentioned rain, but that a hard 

 rain followed immediately after the planting was finished, and that 

 rain fell also on the second, third, and fourth days thereafter. 



The effect of the treatment of this field was determined in five 

 different ways: (a) 500 hills were examined in each plot on each of 

 five dates between June 6 and July 10, and 1,000 hills in each plot 

 were examined June 16, record being made of the number of hills out 

 of each 500 found infested by ants; (b) 10 hills infested by ants were 

 dug up in each plot, including, of course, the checks, on three dates 

 from Tune 17 to July 5, and the number of ants and root-lice found in 

 tach of these hills was determined by counting; (c) the stalks and the 

 vacant hills in 500 hills of each plot were counted June 17; (d) at 

 husking time, in fall, the fertile and barren stalks in 1,000 hills of each 

 plot, and the ears borne by these hills, were counted, the ears being dis- 

 tinguished as small, medium, and large; and finally (e) the stalks and 

 ears in 2,000 hills of each plot were counted, and the yield of ears for 

 each 2,000 hills was weighed, 200 hills being taken for the purpose 

 from each of ten rows at the center of the 20-row strip. 



As the last test was the most significant, its results are first given 

 in the following table. 



