BUREAU OF ENTOMOLOGY AND PLANT QUARANTINE 73 
Insecticide tests in which arsenicals, fluorine compounds, fixed nico- 
tine, and phenothiazine dusts were used were continued on small plots 
arranged in Latin squares. On the basis of the reduction in the num- 
bers of worms per boll the fluorine compounds and fixed nicotine gave 
more promising results than any of the other insecticides that have 
been tested at Presidio. 
ROOT APHIDS 
Of the three species of root aphids injurious to cotton, Trifida'phh 
phaseoli (Pass.) causes the most severe damage. There appears to 
be a correlation between the presence of the aphids and the incidence 
of seedling diseases. In control experiments directed against the 
aphids and the attending ant (Lashis nig&r neoniger Emery), the 
following have been used: (1) Repellents applied to the seedbed be- 
fore planting and around the plants, (2) fumigants applied before 
and after planting, and (3) poisoned baits for the ants. The repel- 
lents and fumigants tested included oil of tansy, oil of sassafras, 
oil of anise, oil of lemon, kerosene, tincture of asafetida, flake naph- 
thalene, clerris, sulfur, tobacco dust, paradichlorobenzene, ethylene 
dichioride, and dichlorethyl ether emulsified with fish-oil soap in 
water. On the plots treated with paradichlorobenzene, the germina- 
tion was very poor and the plants were stunted. The oil of tansy 
treatment delayed germination about 24 hours, whereas the derris has- 
tened germination by 24 hours. No effect on the germination was 
noted as a result of the other treatments. The paradichlorobenzene 
was also the only treatment that controlled the aphids. 
The ant poisons used were varying amounts of thallium sulfate, 
thallium acetate, and tartar emetic in sweetened baits made with 
sugar and honey or honey and water. The poisons, absorbed on 
strips of blotting paper, were exposed on the soil or in perforated 
tins and on sponges in tins. Tartar emetic (2 ounces mixed with 1 
gallon of water, 1 pound of sugar, and 2 ounces of honey) was more 
effective than the thallium poisons, although more species of other 
ants were attracted to the thallium baits because of the higher con- 
centration of sweets used. Counts of the number of active anthills 
were found not to be a good criterion of the aphid control, and no 
definite conclusions were drawn. Cotton following corn interplanted 
with cowpeas or soybeans that serve as late-season hosts for the 
aphids was much more seriously damaged by aphids than cotton 
following cotton, tobacco, or small grain. 
COTTON LEAF WORM 
In 1938 the first report of the cotton leaf worm in the United States 
came from Port Lavaca, Calhoun County, Tex., on May 2. This 
was more than a month earlier than the first report in 1937, and was 
the earliest record in recent years. Leaf worms were reported in 
various parts of southern and eastern Texas during May and June. 
The first leaf worm reported from the vicinity of College Station. 
Tex., was collected on June 18. The worms were found in Madison 
Parish, La., on July G. On July 7 full-grown worms were found in 
Presidio County in the Big Bend scot ion of Texas — a very early date 
for that section. On July 16 the worms were reported from Garvin 
County, Okla., Columbia County. Ark., and Holmes County, Miss., 
and on July 28 from Pima County, Ariz. During August they were 
