TOXIC GASES AS A CONTROL OF THE PEACH-TREE BORER. 9 
ever, and no definite statement can be made as to the exact time 
required for the gas to kill all the larvae. This, of course, depends 
upon the concentration of the soil vapor and the condition of the 
insect. These points are discussed more fully in a later paragraph. 
Occasionally larvae subjected to the fumigation apparently live for 
weeks in a comatose condition, their bodies shrinking to one-half 
or one-third normal size before death. There is no means of know- 
ing whether this is due to an unequal distribution of the gas or to 
the greater vitality of the individual so affected. As a rule, after 
the application of an effective dose in late summer and early fall 
most of the larvae are killed within a period of about two weeks. 
Use as a Control Measure. 
time of application. 
The extent to which an application of p-dichlorobenzene may act 
as a control measure for the peach-tree borer depends greatly on the 
season of the year in which it is made. In the central latitudes, 
where most of the experimental work was done, the hatching period 
of the insect extends more or less over at least three months — July, 
August, and September. In exceptional seasons, and perhaps to 
some extent every season, it may be extended from one to two weeks 
in either direction from these limits. 
It has been found that a single application of an effective dose 
made in the early fall gives a very fair degree of control. Ap- 
parently the ideal time would be from two to three weeks before the 
end of the hatching period. Applied at that time it kills all the 
larvae except a few of the more perfectly protected that have already 
entered the tree, and provides an immunity from the attacks of 
those which hatch later and appear during the progress of the 
fumigation. Also the soil temperatures at this time are sufficient 
to vaporize most of the material before winter, a point discussed 
in a later paragraph in connection with injury. The seasonal 
fluctuations in the period of egg deposition prevent the determina- 
tion of the time of application to any very great degree of exactness. 
The period for effective application, however, apparently has no 
arbitrary time limits. Table II gives the summarized results of 
about 15 experiments carried on in West Virginia, Virginia, and 
Maryland in 1916, 1917, and 1918. 
125552°— 19 2 
