14 BULLETIN 731, U. S. DEPARTMENT OF AGRICULTURE. 
TIME OF DAY. 
Another question of importance is the time of day for application. 
It is, of course, well known that much more effective poisoning with 
dry dust can be conducted while the dew is on the plant, as the poison 
not only clings to the plant better but has much less tendency to 
drift away from the cotton. A number of tests have been conducted 
in an attempt to secure some information on this score but were so 
hampered by the light infestation of 1917 that definite conclusions 
are not warranted. It is evident that more effective poisoning 
usually can be done from about 4 p. m. until about 9 a. m. than at 
other times, although fairly successful results have been secured from 
applications made throughout the day. It will probably be found 
that it is advisable to poison as much as possible during the evening, 
night, and early morning, and to plan to poison during the day only 
in case of emergency. 
MACHINERY FOR APPLYING THE POISON. 
The machinery for the application must, of course, vary according 
to the requirements of different conditions. A very satisfactory hand 
gun of the type shown in operation in figure 9 was already on the 
market and was largely utilized in the experiments described. This 
gun, however, will cover only about 4 or 5 acres a day, and it was, of 
course, necessary to develop machinery adapted to larger areas. For 
this purpose a power machine has been developed somewhat of the 
type shown in figure 10. These have been increased in efficiency 
until now nearly 200 acres per day can be covered by a single machine. 
In addition, efforts are being made to develop an intermediate type 
of machine which can be sold comparatively cheaply and which will 
be adapted to the man planting 50 to 100 acres of cotton, and cover 
20 to 30 acres per day. 
AMOUNT OF POISON PER APPLICATION. 
The amount of poison required per application has depended so far 
more on the requirements of the machinery utilized than on the 
amount necessary for thoroughly dusting the cotton. As a general 
rule experimental applications have averaged about 5 pounds per 
acre, but it 1s apparent that this amount is excessive, and with 
further improvement in the machinery it will be possible to accomplish 
an effective poisoning with a much smaller amount. 
NUMBER OF APPLICATIONS. 
The number of applications necessary undoubtedly will vary. 
This must depend entirely on the conditions prevailing within the 
-particular cut under consideration. In most of the experiments 
conducted so far from three to five applications were made but, as 
has been shown, the effectiveness of these was considerably reduced 
by the fact that they were on such small plats. In the only case in 
which experiments on a very large scale were conducted, the effect 
