88 ANNUAL REPORTS OF DEPARTMENT OF AGRICULTURE, 1 
A method of processing quassia for the preparation of qnassin was devel- 
oped, and tin- latter material was shown to consist of two Isomers which have 
been separated and purified. Chemical studies on the cockroach plant <//'//>- 
lophyton cimicidum) lead, among other things, to the Isolation from ii of b rare 
cyclic alcohol known as quebrachite. 
CHEMICAL INVESTIGATIONS TO DEVELOP SYNTHETIC ORGANIC ENSBCTICIDEfl 
About IO0 synthetic organic compounds were made to test their possible 
Insecticidal usefulness, such tests being made by other divisions. 
Of the compounds tested against mosquito larvae, the following were most 
toxic, being effective .-it 5 parts per million or less: Phenothioxindibenzothio- 
phene, thioxanthydrol, dibenzofuran, 2-chlorodibenzofuran, 2-aminodibenzofuran, 
2-methylanthraquinone, p-chlorodiphenyl, 2, •''». 4, <"» tetrachlorophenol, n-phenyl-1- 
naphthylamine, phenanthrene, xanthene, p-bromonitrobenzene, and p-dinitro- 
benzene. In tests against the codling moth, the first three of these compounds 
among the most effective, as were also 4, ft-dinitro-o-cresol methyl ether, 
thiocoumarin, bexachlorophenol, diphenyl disulphide, p-nitrosodimethylaniline, 
thioxanthone, 4, 6-dinitro-o-cresol acetate, and the three nitroiodobenzen 
which i he para compound Is the best . 
Following the discovery that certain azo compounds are toxic to mosquito 
larvae, many such preparations were made, but none of them look very promis- 
ing against ot her Insects. 
Phenothiazine received considerable attention, as it Is still the most promising 
organic substitute for load arsenate in spraying apples. This compound Is now 
being made by at least three companies and high-grade material is available. 
A tendency for it to decompose has been noted, bul the actio;, is - ( . slight a- to 
he unimportant. Certain derivatives of phenothiazine were also prepared, hut 
none showed especial promise. 
CHEMICAL INVESTIGATIONS ON THE REMOVAL OF SPRAY RESIDl I 
The work on spray residues was active at three of the Division's laboratories, 
namely, those at fakima, Wash.. Vincennes, Ind., and Washington, I». C. The 
fruit-washing experiments conducted cooperatively with the Bureau of riant 
Industry al Yakima and Washington required the making of 600 analyses for 
lead at the former place and over 900 al the latter. At both places it was 
demonstrated that 1 percent of a low-viscosity mineral oil materially aids the 
regular acid wash in removing lead arsenate residues. 
Last year's work concerning the great variation to he found among the indi- 
vidual apples <»f a given lot were repeated this year, except that the washed 
ami unwashed apples were taken from the same lot. One hundred unwashed 
fruit- gave results of from 0.135 to 0.401 grain of lead per pound, with an 
average of 0.256 and a standard deviation ot 0.056. A like number of washed 
fruits showed a range of from 0.014 to 0.074 grain per pound, with an average 
Ot 0.0392 and a standard deviation of 0.0143. These result- indicate that if in 
washing experiments we wish to say with a certainty of 20 t<» l that two lots 
of apples really differ when our analyses give figures differing from each 
other i.\ in percent we must obtain these figures from about W unwashed or 
ion washed fruits. At Vincennee it was determined that a deposit of 20 micro- 
grams of a--.»>. per square centimeter is necessary for control o\ codling moth 
larvae 
A i: pge number <>t experiments on the removal of fluorine residue were made, 
and the earlier finding that hydrochloric acid is still the best wash was 
< onflrmed. 
ition of i Icoti e residues on apples was continued, and a maxi- 
mum of 0.026 -lain per pound found on some plots. An acid wash re 
• 0.004, hui was not \cr\ effective in removing the bentonite that accom- 
panied t lie Dicot inc. 
Numerous analyses of miscellaneous residues were made for other divisions, 
Including s| determinations of sulphur on orange leaves sprayed with various 
ii, i inn- products; nicotine on :; samples >'( grapes and on v "f mush- 
rooms, the latter grown in soil drenched with nicotine solutions; barium Quo- 
silicate mi i samples of peaches; lead and arsenic on 15 Bumples ^\' peaches 
: and arsenic in 'Ji samples of v,.iis. 
\ omewhat related piece of work Involved the determination of copper in 
v,| o ampl< of wood from trees Injected with copper sulphate for control of 
bark beetles, and of /.inc. arsenic, and mereurj in 300 samples each from trees 
