87 



THE PUPAL PERIOD. 



An individual that pupated July 29 at 4 p. m. transformed to adult 

 at about the same hour August 6, just eight days later. A second 

 that transformed to pupa at the same time did not issue till 1 p. m. the 

 following day. A third example found to have pupated early on the 

 morning of July 28 did not issue as a moth till 3.15 p. m. of August 7. 

 These three rearings alone show that there may be considerable indi- 

 vidual variation in even so small a matter. The two pupae were kept 

 side by side and under precisely the same conditions. 



The temperature was exceedingly hot, the thermometer ranging- from 

 85° to 92° F. 



Other experiments at different temperatures gave from nine to twenty 

 seven days as the period of the pupa state, as follows: 



Pupa formed. 



Moth 

 issued. 



Aug. 3 

 June 9 



J ul v 25 



May 30 



August 29 Sept. 8 



April 29 Hay 15 



March 1 Mar. 28 



Period. 



days . 



id days 



10 days 

 15 (lavs 

 27 (lavs 



Temperature 

 average. 



83^ 

 Seasonable. 



Seasonable. 

 Cool. 



THE LARVAL PERIOD. 



The duration of the larval period is more difficult of observation 

 than is that of the egg or the pupa. Time, labor, and patience may be 

 saved by simply calculating' it by subtracting the egg and pupal periods 

 from the entire cycle periods. 



This would give us, by estimate from experiments Nos. 4 and 5 (thirty- 

 eight and thirty-nine days), a larval period of twenty four or twenty- 

 five days as a known minimum. 



Mr. Johnson's article on the flour moth is the most complete work yet 

 published on any species of insect affecting stored grain. His results 

 are beyond question, and although he has refrained from unfavorable 

 comment, in self-defense I must pass criticism. 



Admitting a difference in atmospheric conditions between the Dis- 

 trict of Columbia and central Illinois, where Mr. Johnson's shortest 

 period was obtained, it should not be more than a week, as affecting the 

 entire life cycle of an insect. Again, the months of May and June, in 

 which this period was obtained, can scarcely be considered midsummer. 

 Finally, if it could have been foreseen that our hot spell of this year 

 Mas to occur m August, and observations conducted accordingly, shirt- 

 ing an experiment about three or four weeks later than was done. L 

 think it will be conceded that this period would not have occupied more 

 than the estimated live weeks. 



