June, '02] ENTOMOLOGICAL NEWS, I95 



number of days the bands were on since being last removed, we shall 

 have a proper set of figures to show the rise and fall of the broods and the 

 results will be as follows : In 1897 the first brood gradually increased in 

 numbers until a maximum of 181 larvae per day was attained in the period 

 closing July i6th ; then there was a rapid falling off to a minimum of 41 

 per day in the period closing August 9th. A gradual increase than began 

 which gave a second maximum of igr per day in the period closing Sept. 

 2d, and then a decline, at first slow and then rapid, to a minimum of 52 

 per day in the period closing Oct. 19th. As the last period was a long 

 one — 15 days — it is probable that the larvae had almost ceased to come to 

 the bands on the date of their removal. 



Figures could hardly have been made to order that would better serve 

 to indicate just two broods of this insect. The maximum of the first 

 brood July i6th, and of the second brood Sept. 2d, compare remarkably 

 well with our records for northern Colorado, and gives 48 days as the 

 average time for the complete round of development. In my paper 

 above referred to I gave 49 days, or seven weeks, as the average time in 

 Colorado. 



The records given by Mr. Simpson for 1898 are similar, except that the 

 broods are both later, probably on account of a later spring, and the time 

 elapsing between the two maxima is a little greater. 



Both records are strong proofs of just two broods and yet the writer 

 says " From these records, supplemented by observation, I can say defi- 

 nitely that there are three broods in the vicuiity of Boise and the greater 

 part of the Snake River Valley." I can only account for such a state- 

 ment on the supposition that Mr. Simpson had a pre-formed opinion that 

 he must make out, at least, three broods, and he did not have the courage 

 to announce less. Pre-formed opinions are dangerous things when 

 carrying on an experiment. 



The data given by Mr. Cockerell in Bulletin 25, pp. 48-50, of the N. M. 

 Experiment Station, 1898, indicate that there may be more than two 

 broods of the codling moth in that State, but the evidence is not at all 

 conclusive, particularly when we notice that on Sep. 3rd there were 53 

 larva? taken, from which only three moths hatched. At Fort Collins we 

 have hatched late moths of the second brood as late as Sept. i6th. 



Is it not possible, yes probable, that many errors have been made and 

 are being made by entomologists who announce an insect as varying in 

 its number of generations in a locality or in different localities ? It is cer- 

 tain that such announcements are repeatedly made in entomological 

 literature without accompanying data to support them. Too often the 

 the time required for the life cycle is estimated in days, and then the 

 duration of the insect's occurrence in the year is estimated in days, and 

 the latter number divided by the former to determine the number of 

 broods. Such an estimate I believe will usually be erroneous if the num- 

 ber of broods is more than one or two. The codling moth requires, on an 

 average, about seven weeks to pass through its transformations. This 



