36 



differ with species. Assuming this hypothesis to be true we can 

 readily determine the date of the appearance of an insect in the 

 spring by keeping record of the accumuhited temperature and its 

 departure from the normal with the aid of the formula worked out 

 for that insect. By considering the degrees of temperature per day 

 as heat units the desired date can be readily computed by mathe- 

 matical formula'; or if the temperatures be platted the determina- 

 tion may be made more readily with the aid of a planimeter. 



If this hypothesis be true, we can readily see that a marked excess 

 of temperature for a week or two after the daily mean had passed 

 C)S° F., at which tiuie the first weevils would conmience to appear, fol- 

 lowed by a slight deficiency in temperature subsequently antil the 

 mean daily temperature had reached 78° F.. would result in the date of 

 maximum appearance occurring Ix'fore the normal rather than after 

 it, and vice versa, in a case with the opposite conditions. These con- 

 ditions are nmch more readily appi-eciable by the study of tem- 

 perature platted in curves. 



That the rainfall is also a factor governing the time of emergence 

 is ])robable, but it will be largely reflected in the temperature. 

 Undoubtedly the propei* combination of the de|)arture from normal 

 of temperature and rainfall reduced to a formula in which both were 

 included in a single '"'■ unit of weather " would give us the exact 

 method of computation. 



I offer this hypothesis merely tentatively. It may be old for 

 aught I know, although I have never seen it applied to insects. 

 However, in fish hatcheries the time of hatching of the eggs is deter- 



1904 curve from tlie normal after passing tbe point of first emergence (H8° F. ) is 

 at D. One montli prior to that would lie A" on the normal curve and Z on the 

 1904 curve. The normal temperature accumulated hetween this date and the 

 normal date of maximum emergence (where the normal curve crosses 78° F.) at 

 B would 1)6 the area A"BC'. Then the date of maximum emergence in 1904 

 would be that date on which temperature had been accumulated equivalent 

 to that represented by A"BC', which would be determined by an area 

 A"ZXY, in which the position of the line XY must be determined by computa- 

 tion, with matliematical formula^; and upon establishing its position so that 

 it confines an area in A"ZXY e(jual to A"BC'. the point X will be the date of 

 maximum emergence for 1904. which in 1904 was X'Y', or May 12. This may 

 be readily done by the aid of a planimeter. 



The curves given are of interest in that prior to the presentation of this 

 palter the date of maximum emergence in 1904 had been determined by the 

 above method as being between May 9 and 12, according to slight variation 

 from above in method of detei-mining. Since then, in .January. 1905, the 

 author has received Farmers' Bulletin 211, in which Mr. W. D. Hunter shows, 

 on page 21, that the maximum emergence of the weevils in 1904 was on May 

 11, which fact was previously unknown to the writer. The close approxima- 

 tion seems to lend support to the probability of the hypothesis in general, as 

 it has similarly proven correct in at least three other instances. 



