250 
Smith, and of course would not be expected with the occurrence of a 
single brood, but conforms with the experience of nearly all who have 
made any positive observations on the subject. In substantiation of 
this fact, I quote from the table given on page 173 of the Kansas report 
cited. It represents the results of early and late picking from the 
same trees. About one-half of the fruit of each tree was gathered on 
the first date and the balance on the later date: 
Variety. 
Jonathan 
Yellow bellflower. 
Do 
Wagner 
Date of 
gathering. 
Aug. 8, 9. 
do . . . 
do ... 
do ... 
Per cent ! Date of I Per cent 
wormy, gathering, j wormy. 
Sept. 15 
do . . 
... do .. 
....do . 
* This tree had received two applications of Paris green and the preceding tree one of Climax 
Insect Poison. 
The great increase in the percentage of unsound fruit shown in the 
above record was almost altogether late in August and in September, 
and by larvae recently hatched or less than half grown. 
All these records practically coincide in recording the transformation 
of the first brood of larvae into moths during the latter half of July 
and throughout August, and the occurrence of a second brood of larvae 
from August to November. The normal and usual occurrence, there- 
fore, can not be doubed of two well-defined broods in the portions of 
the United States directly referred to, with a third in California and 
probably in the South. Of these two broods the second is by far the 
more destructive. 
The records of the State of Maine, however, indicate that normally 
one annual brood is the rule, with a partial or supplemental second 
brood, the abundance of the latter depending on the season. The most 
thorough and careful observations on this point are those of Mr. 0. A. 
Atkins (Agriculture of Maine, 1883, pp. 356-363), whose work is a model 
ol painstaking and systematic investigation and deserves wider publi. 
cation. It covers the entire life-cycle of the insect, and in the matter 
of the number of broods establishes beyond a peradventure that the 
great majority of the larva3 — the immediate progeny of moths from 
hibernating larvae — remain unchanged in the cocoons from the 1st of 
August to from June 15 to August of the following year, transforming 
chiefly between June 15 and July 15. In September, however, a very 
few moths came out from the August cocoons of the same year, and in 
all 50 moths were bred. Mr. Atherton, in discussing this paper, states 
that one brood is the rule in Maine, but that last season (an unusual 
year, 1882) there were two broods, apples taken from a tree in Novem- 
ber containing small worms. Prof. F. L. Harvey states that his obser- 
vations indicate two broods, having also observed half grown larvae 
in November (Ann. Eept. Maine Agl. Coll., 1888, p. 174); and Prof. 
William M. Munson states that "it was observed that a large percen- 
