STUDIES ON CONTACT INSECTICIDES. 13 
The present experiments indicate that trimetlrylamine hydrochloride 
is much more effective than methylamine hydrochloride, the reverse 
of what Tattersfield and Roberts found with the corresponding 
bases when used as fumigants. 
GENERAL DISCUSSION. 
In the present state of knowledge of insecticides, it is doubtful 
whether one may safely draw any far-reaching conclusions as to the 
relation between the physical properties of compounds and their 
toxicity. Any large and varied series of toxic compounds will show 
differences in toxicity which can not be correlated with molecular 
weight, volatility, solubility, ionization, permeability, or other 
properties. 
Neither can these differences always be harmonized with differ- 
ences in chemical structure. Many powerful poisons like nicotine 
and coniine contain nitrogen; others like "pyrethron," 5 which, in 
crude form (contaminated with plant extractives), was found to be 
slightly more toxic to Aphis rumicis than nicotine, contain no nitro- 
gen. The saturation of a cyclic compound with hydrogen may either 
increase or reduce toxicity, as has been pointed out in the case of 
piperidine and hex ahydronico tine. The introduction of chlorine 
into benzene compounds seems usually to increase toxic activity, 
but its effectiveness varies greatly with the nature of the other groups 
present. 
Other instances could be mentioned, but these will suffice to illus- 
trate the point. Indeed, the more these facts are studied, the more 
it seems probable that pharmacological action in insects is, in many 
respects, fundamentally like that in higher animals, and that the state- 
ment of Cushny (5, p. 20) can be accepted when he says: "From the 
present confusion the only legitimate conclusion seems to be that the 
activity of drugs depends on a large variety of factors and that 
pharmacological action can not be brought under any- one law, either 
chemical or physical." 
In this investigation the writers have chosen chemical structure as 
the best basis for comparison; that is, so far as obtainable, those 
compounds have been selected which are chemically allied to others of 
known toxicity. While this method has brought many failures, it 
has led to the disco very of apparently new compounds of high toxicity 
which may properly be the subject of another paper. 
CONCLUSIONS. 
This is a report of a laboratory study on the effect of a number of 
organic compounds as contact insecticides for Aphis rumicis L. living 
on nasturtium plants. 
Pyridine, alpha picoline, and commercial pyridine containing the 
higher homologs of pyridine were of little value as contact insecticides. 
The alkaloids, with the exception of nicotine, were of low toxicity. 
Nicotine, however, was the most toxic compound investigated (ex- 
cluding "pyrethron"). 
5 A series of tests were made with crude "pyrethron," but these were not extensive enough to in- 
clude in the tables. 
