44 
Nebraska. Cockerell concludes there are three full generations near 
Mesilla Park, N. Mex. Aldrich in 1900 stated that there were three 
generations in Idaho, and in 1903 concluded after a series of breeding- 
experiments that there was a partial third generation at Lewiston. 
At various times writers have made assertions that in the warmer 
sections of the United States a partial fourth brood was produced. 
In carefully sifting all these statements the writer tinds man}^ points 
which throw doubt upon and render them of but little value, principally 
because definite dates and localities are not given. The date and exact 
localities are often t)f as much importance to future workers, and per- 
haps of more importance, than the o])servation itself. 
METHODS BY AVIIICH THE NUMBER OF GENERATIONS MAY BE 
DETERMINED. 
From the nature of the case the determination of the number of 
generations of the codling moth is a most diflicult problem to solve 
accurately. The methods used must be scrutinized carefully, and all 
possible sources of error must be taken into consideration or elimi- 
nated. Xhe correctness of a conclusion can be assured only by exact- 
ness in methods and by corro])orative evidence secured l)y different 
methods. Observations made in orchard examinations have constituted 
one of the methods largel}^ followed. Although ol)servations are of 
great value when used in connection with other methods, they often 
lead to erroneous conclusions when used alone, as it is possible to 
obtain evidence of the condition of an orchard onl}- from the study of 
a veiy small portion of it during a very short period. Past conditions 
are often unknown, and conclusions obtained are largely based upon pre- 
conceived ideas. If a large number of insects can be bred throughout 
the season, much valuable data can be secured and the prol)lem solved 
beyond an}^ doubt. As 3"et we have no records of breeding experiments 
carried on throughout the season with the necessary accurate data. 
The writer has attempted many times to breed the insects throughout 
the season, but has alwa3\s failed, usually on account of some unfore- 
seen diflicult}" which caused the experiment to end. However, it is 
believed that with proper care and experience this breeding can be 
successfully done. Breeding the insect and harmonizing the results of 
the breeding b}^ observations in the orchards has been the method 
most used in working upon this question. By breeding the insect 
through parts of its generations valual)le data have been secured, 
which, if pieced together and corroborated by other methods, are 
almost as valuable as if the insect had been bred throughout the season. 
Man}" entomologists have neglected to increase the value of their 
breeding experiments by keeping the insects under conditions of tem= 
perature and moisture diflerent from those prevailing in the orchard 
and keeping no record or a very fragmentary record of the tempera- 
