Part VII] Beeson : Life History of Toon Borer. 
39 
value of tlie daily figures. By this means the errors of the actual records, 
due to inequality in daily collection, are eliminated without affecting the 
general value of the curve as determined by date and meteorological 
conditions. 
From the Tables 10—14 and Plate VI, it will be seen that the swarming 
of larvae of H. robusta is fairly general throughout the months of April 
and May. The curves reach their summits between the 6th and 9th 
April and again between the 6th and 9th May, about one month later. 
From the beginning of the fourth week in April until the beginning of 
May, the daily numbers remain low and steady, which points strongly 
to an overlap of the extreme limits of the first and second generations. 
The larvae and pupae collected under sack-bands during this period 
belong, therefore, partly to the latest broods of the first (flower) genera¬ 
tion, and partly to the earliest broods of the second (fruit) generation. 
During the month of April, 77-5 per cent, of the total number of 
banded-trees were attacked, while in May 90*2 per cent, were attacked. 
On the other hand, the average number of larvae per tree during the 
flower generation is 63, while the average number per tree during the fruit 
generation is 33. It is believed that this reduction in the incidence of 
47-6 per cent, is the direct result of the destruction of the larvae, etc., of 
the first generation caught at sack-bands, and is suggested as a possible 
method of control. The normal rate of increase of the first two genera¬ 
tions is very much greater than that of the 3 later generations, owing 
to the greater supply of food ( i.e ., 5 or 6 fruits per head suffice in the 
one case, against 1 or 2 shoots per head in the other). It is even possible 
that there is normally a decrease in the absolute numbers of the insect 
in the last three generations. The result of sack-banding is to start the 
3rd generation at a very much lower incidence than the normal. If x 
is the factor representing the normal rate of increase and y the incidence 
of the 5th generation, then the initial incidence of the 3rd generation is 
reduced by sack-banding from yx 2 to y(A +~) 2 ~roughly -J. Similarly 
in Locality II, where the sack-banding was most successful, the 
incidence is reduced from 39,852 (=74 per tree) to 14,256 (=26*5 per 
tree) i.e., 35-7% ; here the initial incidence of the 3rd generation is 
y © 2 = rou g% It 
1917. In 1917 the swarming of larvae continued up to the 10th 
June. 
Pupae of the Second Generation. 
1914. During May and early June, matured larvae feeding inside 
toon fruits were collected for preliminary work on the pupal stages, and 
the following data recorded. 
[ 134 ] 
