6 BULLETIN 1201, U. S. DEPARTMENT OF AGRICULTURE. 
and leaves (Nos. 33a, 5, and /, each used as a dust) were efficient but 
slow against roaches, grasshoppers, flies, and bees (Table 1) , but ineffi- 
cient against Aphis spp. A and B (Table 2), and had little effect on 
tent caterpillars. Used as a stomach poison, these powders were effi- 
cient against grasshoppers, silkworms, and flies (Table 1), but had no 
effect on large web worms. 
The water extracts (highly concentrated) from the leaves and bulbs 
of ainianthium, each sprayed upon the insects and their food, had 
considerable effect on roaches, potato-beetle larva?, and silkworms, 
but none on webworms, potato aphids, rose aphids, and Aphis spp. 
A and B (No. 35fa, Table 3). The alcoholic and benzene extracts 
(Nos. 504 and 513, Table 4) from the bulbs, used with soap, were 
inefficient against four species of aphids (Aphis spp. A and Z>, Mac- 
I'osiphum spp. A, and M. UriodendH) . The alcoholic extract, how- 
ever, was efficient against silkworms. 
At Vienna, Va., three apple trees, each bearing a nest of tent cater- 
pillars from one-half to three-fourths grown, were selected for pre- 
liminary field tests. One tree was sprayed with a 10 per cent solution 
of a water extract from the leaves of amianthium; another tree with 
a 10 per cent solution of a water extract from the bulbs; and the 
third tree served as a control. A week later the caterpillars on the 
sprayed trees appeared shrunken and apparently had not eaten since 
the trees had been sprayed. The caterpillars on the control tree and 
others near by were almost full grown. 
INSECT POWDER OR PYRETHRUM. 
The following results were obtained by using a commercial insect 
powder, here called pyrethrum, and probably derived from Chrysan- 
themum cinerariae folium. Most of the details will be given when 
comparing these results with those obtained by using certain other 
plants (see pp. 10 to 21). The powders (Nos. 103 and 503, Table 2), 
used as dusts, were found efficient against five species of aphids (Aphis 
spp. A and B. Macrosiphum spp. A and B y and Macro^iphovrVa 
sanbomi), grasshoppers, silkworms, flies, potato-beetle larvae (Table 
1), and tent caterpillars; used as a decoction, not filtered (No. 103a, 
Table 3), it was efficient against Aphis spp. A and Z?, but used as a 
hot -water extract (filtered, No. 103b), it had no effect on these in- 
sects; used as a fumigant (No. 103), it was efficient against Macro- 
siphwm sp. C and Myzus persicae, silkworms, webworms, and a lady- 
beetle tested. The hot-water extract (filtered) and a distillate 
wore efficient against silkworms, but the cold-water extract (filtered) 
was inefficient. 
The alcoholic and benzene extracts, when sufficiently strong and 
used with soap or kerosene emulsion, were found efficient against 
aphids (Tables 4, 5, and 6). The alcoholic extract, used with soap, 
was efficient against small webworms (first instar) and half -grown 
sawfly larvae, but only about 50 per cent of the larva? and none of 
the adult potato beetles tested were killed within seven days. 
" CUBE." 
In 1920, while collecting fishes in Peru for Indiana University ,- 
Dr. W '. R. Allen procured a supply of the dried roots of ''cube" 
(see footnote on p. 34) ; some of these were used as a fish poison and 
