8 IXSECTS AFFECTING VEGETABLES. 
live larvae were counted on the underside of one leaf. These beds 
were examined daily, and on September 7 it was concluded that the 
experiment was a failure, as the larva?, except on the leaves which had 
been thoroughly sprayed on the underside, were unharmed. It was 
noted that on leaves which had been but partially sprayed on the 
underside the larvae avoided the poisoned portion but continued to 
feed on the part which had been missed by the spray. The failure of 
this experiment was due to the habit the larvae have of feeding on the 
underside of the leaves, without eating through the upper epi- 
dermis. The mixture used in this experiment burned the partly con- 
sumed leaves slightly but caused no extensive injury. 
Experiment No. 2. — Paris green. 2 pounds, and whale-oil soap, 8 
pounds, in 100 gallons of Avater. September 7, three beds of table 
beets, each 10 yards in length and 1 yard wide, were sprayed with 
about 3 gallons of this mixture. The same apparatus was used as in 
experiment No. 1. A particular effort was made to wet the under sur- 
face of the leaves thoroughly. The mixture adhered perfectly and an 
extremely even distribution of the poison was obtained. The weather 
was hot and clear. 
The following day nearly all the larvae were dead and the few 
individuals which remained alive appeared very sick. September 10, 
the beets were free of larvae and the foliage showed no burning by 
the spray mixture. Two weeks later the sprayed plants were still 
free of larvae, while beets in adjoining (check) beds were badly in- 
fested. This experiment plainly demonstrates that if the spray is 
applied properly, i. e., to the underside of the leaves, it can be de- 
pended on to control this pest. 
In addition to the experiments with Paris green, a few preliminary 
and somewhat inconclusive tests were made with arsenate of lead, 
lead chromate, and lime-sulphur solution. For the sake of con- 
venience these will be classed as experiments. 
Experiment No. 3. — Arsenate of lead, 1^ ounces, and molasses, 
1 quart, in 1| gallons of Avater. This mixture was experimented 
Avith as a poisoned bait for the melon fly (Dacus cuewrbitm Coq.), 
and was not originally intended for use against the Hawaiian beet 
ATebworm. September 7, a patch of Amaranthus, badly infested by 
laiwae varying in size from those recently hatched to others nearly 
mature, was sprayed AAith some of this mixture. The spray was 
applied very thoroughly to the upper and loAA T er surfaces of the 
leaves with the same apparatus as used in experiment No. 1. The 
folio Aving day the laiwae Avere unharmed. The leaA T es showed a good 
even coat of the arsenate. The sprayed plants were kept under daily 
observation until September 13, at which date they Avere destroyed. 
On the 9th and 10th of September it Avas noted that the larvae Avere 
groAving nicely and appeared to relish the poison-coated leaves. 
