m\ 
- 44 - 
The proef station (70) in 1S31 reported that M. per si cae was 
unusually abundant on tobacco, and large amounts of derris extract 
were used to combat it. Neoton at 0,1 percent was more effective and 
cheaper than the 1:45 akar toeba (derris) of the proef station; however, 
it was necessary uo use a.' small proportion of soap to prevent the separa- 
tion of oil from the Neoton, which otherwise leads to unequal distribution 
of the spray. In 19 33 the same station (7l) reported comparative experi- 
ments made at two place.? with Neoton and akar toeba, to see whether these 
insecticides had any influence on the quality of tobacco to which they 
were applied. The Neoton was used at 0.1 percent, together with some 
soap; the akar toeba, at a dilution of 1:40 in a 0.1 percent soap solution. 
Neither treatment had any marked effect on quality. From the insecticidal 
standpoint, the akar toeba extract was not inferior and there is no reason 
for replacing it with iTooton. Derrothan as a dust was effective in 
laboratory tests out Derrothan emulsion burned the tobacco plants. During 
the 4 months January-April, inclusive, 1932 the akar toeba factory at 
Sikambing, Sumatra, distributed 708,000 liters of aqueous extract of derris 
Davidson (63) in 1930 killed 98.2 percent of these aphids in a 
greenhouse with rotenone suspended in water at a concentration of 1:100,000 
and 94.3 percent at a concentration of 1:200,000. A dust containing 2 
parts of rotenone and 98 parts of diatomaceous earth killed 76.6 percent. 
This aphid was i ..ised as a test insect by Davidson and Jones (65) in 
1931 in studying the. loss of toxicity suffered by rotenone in certain 
solvents and in aqueous suspension. A freshly prepared suspension of 
rotenone made by adding an acetone solution of it to water at 1:50,000 
killed more than 95 percent. 
The green peach aphid was used by Davidson as a test insect and 
referred to by Shepard ( 241 ) in 1931 (see Brev ic oryn e brassicae L., 
page 18 . 
Tattersf :'i eld (258) in 1932 discussed laboratory methods for evalu- 
ating insecticides. Results of tests with two samples of derris root 
against M. p ersi cae are stated as follows: 
The two samples had the seme rotenone content, and 
over a definite range of concentrations (up to 0.05 per- 
cent;, expressed in terms of root, the results for 
equivalent concentrations do not differ by more than the ' • . . 
experimental error, although sample F is always slightly 
less effective. The results are normal up to a certain 
concentration (0.05 percent), and then, above that strength, 
both samples are shown to be less effective than below it. 
In the case of root B, the results are scarcely significant, 
but in the case of root F definitely significant. Obviously 
two factors are working in opposite directions. 
In discussing how far the factor of time should be taken into 
account in judging toxic effects quantitatively, Tattersf ield wrote: 
1 '' i ii mmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm 
