72 



28 $ wliile the sixth and last was sown October 12, and did not come 

 up until about the 20th. These plats were examined by me on Octo- 

 ber 17 ; the first three and the last sown were very poor, the fourth 

 and fifth promising a fair yield. A field adjoining, sown on the same 

 day as plat 5, did not suffer from the fly and produced nearly an aver- 

 age yield of 20 bushels per acre. 



The results of these meager experiments have, as a rule, proven 

 correct in the fields of the farmers. I have not only observed this 

 myself, but it has become well known in the locality that wheat sown 

 before September 15 and after the 30th of the same month seldom pro- 

 duces a good crop, while that sown between the loth and the 25th is 

 the most likely to escape the attack of the Hessian fly, and, as a general 

 thing winters, as well as that sown earlier. 



In summing up the results of this entire system of experiments, it 

 seems that while no exact date can be laid down for the appearing of 

 the fall brood of fly in any precise locality, there is, notwithstanding, 

 a gradual delay in its appearance as we go from the north southward. 

 In other words, there is here a characteristic element in the life history 

 of the species which may be utilized by the farmer to his advantage. 

 Fruit-growers, I believe, estimate that in spring the season advances 

 northward at the rate of about 12 miles per day. This would be a 

 trifle less than 6 days per degree of latitude. If farmers in extreme 

 northern Indiana and southern Michigan can sow their wheat with 

 safety about the 12th to the 15th of September (and we have demon- 

 strated that the fall brood emerges largely prior to the 15th), and 

 farmers in extreme southern Indiana must delay sowing until after the 

 first days of October, there must be a general system of retardation, 

 which, if understood, may be used to advantage throughout the inter- 

 vening territory. 



Starting in southern Michigan on the 12th to 15th and passing 4 

 degrees south to the vicinity of Evansville, Inch, we should expect 

 about the same condition of the Hessian fly during the first week of 

 October. That is, if we pass the danger line about the second week 

 of September in southern Michigan, we should expect to encounter it 

 again in southern Indiana in the first or second week of October. A 

 considerable correspondence and my own experiments indicate that 

 this is usually true. It is not to be supposed, however, that it is 

 possible for me to give precise dates for given localities, as there is 

 another element which is likely to figure in these calculations, viz, ele- 

 vation. It has been stated upon reliable authority that " an elevation 

 of 350 feet is equal to 1 degree of cold in the mean annual tempera- 

 ture, or 60 miles on the surface northward." * While we can hardly 

 expect this to influence comparatively level countries like the State of 

 Indiana at least to any marked degree, extensive areas of high table- 



* Draper's Intellectual Development of Europe, Harper Bros., New York, revised 

 edition, vol. 1, p. 29. 



