56 ASSOCIATION OF ECONOMIC ENTOMOLOGISTS. 



expected in a season, so even those who were disposed to profit by 

 their failures knew not which way to turn. Further investigation 

 revealed the fact that a somewhat similar condition prevailed through- 

 out the entire intermountain region. 



In view of this condition of excessive worminess in apples and the 

 general uncertainty that prevailed with reference to so many of the 

 factors involved, it was decided to take up but one factor at a time 

 and try to work that out by exact methods, so that whatever was dis- 

 covered would be the result of known factors and could be explained 

 by the known variation in the factors according to the needs of the 

 experiment. In this way it has been possible to get quite definite 

 results on several factors in the course of the four years' work, and 

 the most important ones of these are summarized below. 



The poisons were first investigated and were found to be first class 

 in every respect. A few samples of Paris green had been found in 

 the West that contained some free arsenic, but that would rather in- 

 crease than decrease their killing power, so the failure of the spraying 

 could not be laid at that door. 



NUMBER OF BROODS. 



In the work on this insect in Colorado, during the time the writer 

 was an assistant at the State agricultural experiment station, the fact 

 that there were two definite generations in that region and that these 

 generations occurred at fairly definite times was thoroughly estab- 

 lished. These tests were carefully repeated for northern Utah con- 

 ditions, and a life-history chart showing the times of appearance of 

 all stages of the two generations was published in Bulletin 87 of the 

 Utah Agricultural Experiment Station. 



From these investigations it was found that there would be a period 

 of a few clays in which it would be possible to separate the few worms 

 of the two broods then occurring in the apples. This period, as was 

 shown by reference to the chart above mentioned, occurred during the 

 first few days in August in a normal season. As on the accurate 

 separation of the damage done by each brood depended much of the 

 value of the other data obtained, this damage was carefully checked 

 up each year, and in no case was there more than 1 per cent of the 

 " total wormy " in doubt, and often almost no doubtful cases 

 appeared. 



In practical work methods must come before results; but in this 

 summary it has been thought best to give results first, and then the 

 full significance of the methods that gave them can be more fully 

 appreciated. 



