STUDIES ON CONTACT INSECTICIDES. 3 
not ascertained because of lack of material; in others its importance 
did not seem to justify the expenditure of time. 
Many factors combine to influence the accuracy of experiments of 
this nature. Some compounds partially annul the spreading and 
wetting effect of the fish-oil soap or capryl alcohol. There is also a 
tendency for the liquid to collect in drops or thick layers on the 
lower edges of the leaves and stems, and, as the water or compound 
evaporates, subject the aphids in those places to a different con- 
centration of the chemical than on other parts of the plants. Further- 
more, young aphids are more susceptible to most chemicals than the 
older ones, so that the ratio of young to mature has a bearing on 
the number which will be killed by a given concentration of a com- 
pound. It has usually been observed also that high humidity or 
low temperature favors toxicity, presumably because of lessened 
evaporation. All of these experiments have been made in a green- 
house where the influence 01 the above factors has been partially 
overcome by repetition and by making the tests only during bright 
weather. No doubt they will give a fairly accurate idea of the 
relative toxicity of the different compounds to Aphis rumicis. 
In Table 1 are assembled the results of the study. The con- 
centrations are given in grams per 100 cubic centimeters of solution 
or emulsion. For comparison, the concentrations in Table 2 are figured 
also in gram molecules per liter. Fish-oil soap was almost invari- 
ably used in the proportion of 0.3 gram of moisture-free soap per 
100 cubic centimeters, and capryl alcohol at 0.2 to 0.3 cubic centi- 
meter per 100 cubic centimeters. Blank experiments showed little 
killing at these concentrations. The classification adopted for the 
compounds is only for convenience. 
