110 
less due in part to the fact that the upper surface of the leaf is more 
exposed to the spray than the under, and possibly also in part to the 
inferior vitality of the scales on the more exposed surface of the 
leaf. That there is really such a difference is shown by the differ- 
ence in ratios of dead to living on the two leaf surfaces before the 
trees were sprayed. The general average of all the counts is 3.6 
per cent, of dead on the upper surface of the leaf and 2 per cent, on 
the lower. 
In the tree sprayed July 3 the effect on the scales on the upper 
and the lower parts of the top of the tree was brought into compar- 
ison by examining the scales on 25 leaves from each. Twenty-two 
per cent, of the scales on the upper leaves had been killed and 35 per 
cent, of those on the lower, the ratio of killed being more than half 
as great again for the lower leaves as for the upper. This is doubt- 
less due, at least in part, to the greater and more prolonged effect 
produced on the lower leaves by the drip from the upper part of the 
top. 
We may infer from these experiments that two sprayings with 
a 10 per cent, kerosene emulsion, one applied at the middle and the 
other at the end of the hatching period, separated, that is, by an in- 
terval of about ten days, will produce the maximum effect on the scale, 
and that they may be expected to destroy 80 per cent, or more of the 
insects then alive. It .should be added that no harm to the tree was 
done by any of these treatments except a slight burning at the edges 
of the leaves of the tree which was twice sprayed. This appearance 
of injury was perhaps due to humid weather which followed the 
first treatment, delaying the evaporation of the kerosene. Two trees 
were treated July 11 with a 10 per cent, kerosene emulsion, at a 
time and place when rain was falling with a result to diminish by 
half the killing effect of the spray. 
The cost of die 10 per cent, emulsion used in these experiments 
was 4.3 cents per gallon, and the trees were large enough to require 
three or four gallons each for a single treatment. 
The whale-oil soap solution applied at the end of the hatching 
period had about two thirds the effect of the corresponding kerosene 
treatment, as shown by a comparison of the results of the data 
already given for the latter with those derived from an examination 
of 77,000 scales on 75 leaves taken from the tree treated with the 
former insecticide. 
Winter Insecticide Measures. 
Winter spraying, when the trees are bare, has the advantage 
that stronger insecticides may be used, and that less than half as 
