A. Bacot and Gr. Talbot 
227 
1. The arm, from wrist to elbow, was coated as evenly as possible with 
the preparation. One gramme was found to be sufficient for the purpose in 
most cases, and was used as the standard quantity; in a few instances the 
consistency of the preparation allowed of an economy, so that a few tests 
were carried out, using -75 grm. In the case of the liquid preparations, 
Nos. 18 and 19, larger quantities were necessary, in order to completely coat 
the desired area of skin. 
The hand, during the exposure of the arm in the cage, both in tests and 
controls, was covered by a rubber glove, it being necessary to restrict the 
bites to the forearm, owing to the fact that on many parts of the hand the 
skin reaction is so slight that it is not possible to enumerate bites with 
accuracy. 
In the first series of experiments the arm, after treatment, was covered by 
the sleeve for 15 minutes to allow of some volatilization, before exposure in 
the cage for 10 minutes. 
In the earlier trials with preparations 1-9 the control exposure of the 
untreated arm for a similar period, followed immediately after the test. It 
was found, however, with some preparations, that very few, if any, attacks 
were made by the insects under these circumstances. This presumably was 
due to the persistence of the vapour in the cage preventing the mosquitoes 
from detecting the presence of food. In all subsequent tests, therefore, a 
longer or shorter interval, usually of from 2-4 hours, was allowed between 
the test and the control. 
2. In the second series of experiments the procedure was the same, except 
that the arm, after treatment, was exposed to the air for periods of from 2-5 
hours before insertion in the cage. 
It is to be noted that tests made in the laboratory under the conditions 
described are really more severe than would be the case were the preparations 
used in the ordinary way out of doors. Under ordinary circumstances the 
insects would have a greater range of choice, for even if all the population of 
a district were similarly protected, there would still be the chance of feeding 
on other animals; in the laboratory the choice was restricted to a preference 
for the treated or untreated arm of A. B. or G. T. 
VIII. EXPERIMENTS, SERIES I. 
Prior to conducting the first series of trials a preliminary census was made 
in order to ascertain the number of feeding females in the respective cages. 
On May 21st, 1918, an unprotected arm was exposed in each cage; the number 
of bites recorded was: Cage A, 41 bites; cage B, 43 bites. 
The preparations were tested according to the method already explained. 
The results are given in Table I. 
