54 BULLETIN 1147, U. S. DEPARTMENT OF AGRICULTURE. 
Of all the arsenicals tested, acid lead arsenate and zinc arsenite 
were the most adhesive and Paris green the least adhesive on potato 
foliage. The use of lime with arsenicals applied to potato foliage did 
not increase their adhesiveness. 
The suspension properties of the powdered arsenicals are of value 
in differentiating "light" from "heavy" powders, as determined by 
their apparent densities. 
The physical properties of the commercial powdered arsenicals 
could not be satisfactorily determined by sieving, as they are generally 
amorphous and pack in the sieve on shaking. Arsenious oxid samples 
sometimes contain or consist of relatively coarse crystals, so that 
sieving may provide valuable data. 
Microscopic examination gave little information concerning the 
desirable physical properties of the amorphous or seemingly amor- 
phous powdered arsenicals. Differences in size of crystals present in 
the arsenious oxid samples were detected under the microscope. 
The toxicity findings are based on the use of equivalent quantities 
of arsenious and arsenic oxids. Higher percentages of toxicity were 
found for acid lead arsenate than for basic lead arsenate. The differ- 
ent samples of calcium arsenate tested varied widely in toxicity. 
When lime or Bordeaux mixture was added to arsenicals, the toxici- 
ties were reduced. The average toxicity of the three samples of 
Paris green and that of one zinc arsenite tested was slightly more 
than that of an acid lead arsenate and a sodium arsenate. Of the 
four samples of arsenites, the Paris green samples gave the highest 
values, zinc arsenite being much less toxic. Based on equivalent 
metallic arsenic percentages, the Paris green samples gave values no 
higher than that of the acid lead arsenate tested. Several new arse- 
nates tested did not show as high toxicities as did acid lead arsenate. 
Of the various bases tested, lead oxid showed some insecticidal value, 
while the oxids of zinc, magnesium, and copper showed little and 
lime no value. Arsenic acid, acid lead arsenate, and one sample of 
calcium arsenate gave high and practically equal toxicities. Arse- 
nious oxid (white arsenic) gave lower and variable results. The per- 
centages of water-soluble arsenic in the original arsenicals had little 
or no influence on the toxicity, except in the case of those arsenicals 
which were entirely or largely water soluble. These had high per- 
centages of toxicity. 
The determination of reaction in terms of the pH value of water 
extracts from the bodies of various insects fed all of the different 
arsenicals, and also from the bodies of control insects, showed uni- 
formly a slight acidity. These results indicate that the arsenic 
compounds fed did not affect the pH values as determined on dead 
insects. 
The minimum dosage of metallic arsenic required to kill a honeybee 
is approximately 0.0005 milligram, while 0.0273 milligram (or 54 
times as much) is required to kill a full-grown silkworm. Honey- 
bees, confined in cases, void none of the arsenic eaten, whereas silk- 
worms void 90 per cent of the amount ingested. Thus, in reality 
about 6 times, rather than 54 times, as much arsenic is fatal to a 
silkworm as is required to kill a honeybee under the somewhat 
unnatural living conditions. 
