me Bm 
action on the young larvae, The following materials were used: Cal- 
Cium orsenate with 5 percent light oil, copper arsenite (1 part to 
1 part flour and 5 parts lime), copper carbonate (with 50 percent flour), 
Bordeaux mixture (dust), sodium fluosilicate (with 50 percent flour), 
sodium fluoaluminate (with 50 percent flour), sulphur (pure and with 
5O percent sodiun fluosilicate), derria (4,8 percent rotenone) ,’ 
rotenone (4.72 percent), nicotine sulphate (5 percent nicotine with 
lime), free nicotine (5 percent with lime), flour, and lime. Most of 
the insecticides ‘were used in all classes of tests except the-nicotines, 
which were tested only against the moths in the field. In all tests 
heavy applications were used, the object being to determine whether 
control could be obteined ani, so far as possible,.tests were made under 
optinum conditions, Considerable-time is required for boll exoninations 
to determine the presence of small larvae and so only 30 to 40 bolls 
were used per test.: Some of the tests were in duplicate, and‘in all 
experiments comparable checks were used. 
In the laboratory tests for larval mortality, green bolls which 
had deen grown unter centres ond were worm free were used. These were 
cut from the plants with attached stems, which were inserted through 
cardboard covers over jars of water, The bracts were removed and 10 
eges about reacy to hatch, the fertility of which was known,. were in- 
serted beneath the involucre. The bolls were then thoroughly dusted 
and were examined 4-to 5. days later. 
The field tests were confined to small groups of plants, All-of 
the bolls of suitable size were tagged and the infestations "stepped up" 
in half of them by placing 10 eres under the involucre and the other 
half left with the natural infestntion. The bolls were collected after 
10 days and examined for the number of worms which had entered. 
In all of the tests using bolls, the effectiveness of the insecti. 
Cicdes was rated in terms of reduction of number of larvae per boll, as 
compared with the checks, When the treatment resulted in a: reduction 
it was considered: significant only when the difference divided by its 
probable error was as large as, or larger than, 4.0. In the petri dish 
tests to determine the contact action on young larvae, a small cork was 
placed in the center of each dish, which was then put ina closed room 
and dust blown to the ceiling. As scon as a thin layer of dust had 
settled the cork was removed and 10 to 40 larvae were placed in the 
undusted srea and allowed to crawl over the dust film. The results 
were measured by tne effect on the larvae and by the time within which 
they died. In addition to the tests with larvae the insecticides were 
also tested on the moths and oviposition. The effect of dusting bolls 
on oviposition was tried by placing 5 pairs of moths in breeding cages 
with dusted and undusted bolls and determining the total number of eggs 
laid end the percentage laid on the bolls. 
