A. Bacot and G. Talbot 
231 
No importance is to be attached to the difference in the absolute number 
of bites as between one preparation and another, owing to the fluctuation in 
the number of hungry female mosquitoes present. We would also point out 
that the individual idiosyncrasies of the two experimenters in respect of their 
attraction for Stegomyia fasciata , may have influenced the results. It will 
be noted in the two parallel tests with No. 2 a, that the percentage of bites 
is very different; in the case of A. B. only 9 % bit, while in the case of G. T. 
39 % bit. The only other test that permits of comparison in this respect is 
in the case of No. 15 a; in the two hours test 33 % bit G. T., while in a three 
hours test only 23 % bit A. B. 
If we analyse the first case, that of 2 a , we notice that, whereas A. B. 
tested it on August 2nd, G. T. tested four days later on August 6th, the same 
cage being used. On August 1st about 16 females were added to the cage and 
these would normally bite on the 3rd and certainly on the 6th. On August 
2nd, about 12 females were added, which would also bite on the 6th. We 
have therefore an addition of about 28 females ready to bite on the 6th. 
It happened, however, that a total of only 69 bit on the second occasion, 
as against 84 on the first. Allowing for the greater attraction, subsequently 
shown to be possessed by G. T. for the mosquito (see Section XII), the 
difference in the total number of bites in the two results must be mainly due 
to some mortality between the two dates, or to some unexplained disinclina¬ 
tion to suck blood on the part of a number of the females. 
If we omit the question of the difference of attractiveness of A. B. and 
G. T., the value of the preparations tested would rank, on the basis of Table III, 
as follows: 2 a , 1, 15 a, and 21 almost on a parity, followed by 9 a , 10 and 5 
in order, but if an allowance be made for individual attractiveness, the 
figures point to 15 a as the most efficient culicifuge. 
The following classification of the preparations is in accordance with the 
percentage of bites in relation to the total number received in test and control. 
In drawing it up no allowance has been made for the greater preference of 
the insects for one of the experimenters over the other, as it was only in the 
later series of trials that this preference was noted and recorded. As shown 
in the discussion of the results of the long period trials this factor may be 
of considerable importance and in judging the results allowance must be 
made for it. 
1. Preliminary Trials. 
0 % . Nos. 1, 2, 10, 11, 15, 15 a, 21 and 22 a. 
3 %. No. 9 (in the first trial no bites were recorded). 
9 % . No. 5 (on result of second trial). 
10 % . No. 3 (on result of second trial). 
19%. No. 22. 
25 % . No. 7 (on result of second trial). 
26 % . No. 4 (on result of second trial). 
30%. No. 12. 
35%. No. 13. 
