48 
of western Ontario there are two g-enorations, the second })eino- inva- 
riably the more destructive. The observations of Atkins. Harvey, 
and Munson ao-ree with those of Harris. Slingerhmd says in 1898 
that his observations indicate that in New York a large number of 
the larvje of the rirst generation develop into moths, the percJMitage 
transforming depending upon the weather conditions of the season. 
In 1894 Smith found ))y a series of observations that the larvte col- 
lected in midsummer did not transform further that y<'ar. but liii)er- 
nated. Later, in 1897, he states that near New Brunswick, X. J., there 
is positively only a single annual generation, and. further, that south 
of Burlington C>ounty there is at least a partial second g(^nerution. 
In addition to the observations already given of conditions (juite simi- 
lar to these in New Jersey, we find that Trimble in 1865 carried out a 
very careful and accurate series of experiments upon the life history 
of the codling moth at Newark. He found that on August 1<> there 
were three pupa^ among the insects under observation, and that on 
August 2() many moths had emerged: on August 23 he found that one 
in five of the larvae had transformed. Sanderson tinds that there is 
one generation and a partial second generation in Delaw^irc^ He 
states that of the larvw found July 31 about 29 transformed and 5 
remained as larvte. Taking these numerous observations and the data 
given in regard to them into consideration, we must conclude that 
Doctor Smith's observations are too few in numl)er and do not justify 
the assertion that there is but one generation of the codling moth at 
New Brunswick. Man}^ observers in widely different sections of the 
United States have found two generations clearly defined. Le Baron 
states that '* in the latitude of Chicago a great majority of the moths 
of this brood (first) emerge the last two weeks in July.*' Riley, after 
many years of close observation, states that the insect is "invariably 
two ))rooded in Missouri." Popenoe and ^larlatt found two genera- 
tions in Kansas. Gillette indicates two generations in Iowa. Walton 
b}' breeding experiments discovered two generations in the same 
State. From a series of observations extending over several years, 
checked hy l)reeding experiments, Cordlev concludes that there are 
two generations at Corvallis, Greg. Koebele says there are two gen- 
erations in the Santa ('ruz ^lountains of California, and the insect 
proba])ly does not ditter in its liai)its tliroughout tlu» State. Based 
upon one of the most extensive studies of this ([uestion that has ever 
been made, Gillette arrived at the conclusion that then^ nvo two 
generations in Colorado. Cooley says that in 19o*i tluMi* wen^ two 
generations at ^Missoula. ^lont. Forbes indicates a third generation 
in Illinois, based upon the fact that very young larv;v were found on 
October 4. Coquillett states that his notes indicate that the insect 
has three generations in California. Washburn says there are three 
to four generations at Corvallis, Oreg. Card gives two to four in 
