20 DiitKCTou's Kki'out of the 



Taken as a v/hole these experiments indicate the following: 



1. Vigorous trees are probably less liable to injury by crude 

 petroleum than weak ones. 



2. Peach and plum trees are more sensitive to crude petro- 

 leum than apples, cherries or pears. 



3. There is less danger of injury if trees are sprayed in early 

 spring than during the fall or winter. 



4. The 25 per ct. emulsion of crude petroleum and water can- 

 not be depended upon to kill the hibernating scales in the 

 latitude of western New York while the 40 per ct. has proven 

 efficient. 



5. Much pains should be taken to avoid over-drenching the 

 trees. Only enough of the emulsion should be applied to wet 

 the bark evenly and thoroughly. 



Washes. — The resin-lime mixture and government whitewash 

 did not adhere to the trees well and apparently had but little 

 effect on the scales. 



Fumigation. — The fumigation experiments in western New 

 York with hydrocyanic acid gas were also divided into two 

 series. Series I included the experiments to determine the 

 effect of the gas upon bud sticks for budding purposes, and 

 Series II the strength of the gas required to kill the hibernating 

 scales. In both series the gas was used at strengths varying 

 from .18 to .3 gram of cyanide of potassium per cubic foot of 

 air space. The exposure of the buds to the gas varied from one- 

 half hour to one hour. 



The experiments with buds, while not entirely satisfactory 

 owing to the somewhat unfavorable conditions surrounding the 

 treated buds, gave sufficiently uniform results to indicate clearly 

 that the gas is harmless except in the case of the peaches, which 

 were evidently injured slightly by the gas at .3 gram of cyanide. 

 There was but little difference in the percentage of treated buds 

 that set and the checks. In all 4,483 buds were treated, 78 per 

 ct. of which set. The checks numbered 4,804, of which 85.5 per 

 ct> set, making but a slight diiference in their favor. This differ- 



