February, '09] JOURNAL OF ECONOMIC ENTOMOLOGY 27 



and experience for a sufficient review of these matters of recent his- 

 tory, I would like to use my present brief and rare opportunity and 

 pri\'ilege in an endeavor to forecast the immediate future, and, judg- 

 ing from what we have seen and what we now see in progress, to de- 

 duce the probable next steps in the development of method in our 

 work. 



Economic entomology is an extremely complex subject, not only by 

 reason of the number of factors which it must include, but especially 

 because of the variability of many of these factors, and our inability 

 to predict the course of events with certainty in our field. We study 

 the present and the past in a practical way in order that we may pre- 

 dict the future. We observe, generalize, experiment and verify in 

 order that we may be able to say to the farmer or the fruit grower, "Do 

 thus and so in any given case, and this or that desired result will fol- 

 low;'' but we can rarely express our conclusions safely in so definite 

 a form. Often the best we can fairly say is that if the weather should 

 be wet. or dry, or neither one nor the other, as the case may be; or if 

 it has been very wet. or very dry. for the last two or three or four 

 years ; or if the winter has been, or is to be, open or severe ; if the crop 

 in question has been preceded by some other kind of crop, or by one of 

 the same kind ; and if the insect situation was thus and so last year and 

 the year before : if, furthermore, the land is light or heaw. high or 

 low. well drained or wet ; if it has had this or the other management 

 or treatment during the last year or two ; and if several other variable 

 elements of the problem vary to such or such a degree, in this or the 

 other direction — then if the operation X be performed, the result will 

 prohahly be T. but with what degree of probability it is impossible for 

 us to say. Agriculture is itself one of the more uncertain callings, 

 and the farmer every j^ear bets the cost of his crop on the chances of 

 his harvest ; but the entomology of agriculture is more uncertain still, 

 for insects liable to infest a crop are affected, directly or indirectly, 

 obversely or inversely, by everything which affects the crop itself, and 

 by several other things beside. How may we approximate certainty 

 of prediction in this variable tangle of uncertainties within uncertain- 

 ties in the midst of which we have to work? It is only by long-con- 

 tinued observation, by comprehensive survey of all related matters, 

 by repeated and varied experiment, and by the use of statistical meth- 

 ods such as will teach us the range of variation and the character of 

 the average in any given case. By an intelligent use of counts and 

 estimates and averages we can often approximate certainty, where 

 without them our uncertainty would be complete. We can say that in 

 about such a per cent of so many trials you will get your desired 



