Glenn: Insect Development. 319 



Tlio usual method for determining the threshold of development 

 is to use the reciprocals of the periods and the average mean daily 

 temperatures as the coordinates and plot the one against the other, 

 then draw a line which best fits the points, and the point where it 

 crosses the temperature axis is the threshold of development. This 

 method serves quite well when the temperatures are constant, but 

 the more widely the temperature varies during the daily variations 

 the less accurate are the results. The accompanying figure is really ■ 

 four figures placed on the same sheet, and represents the points 

 in the various positions assumed by them at the different stages 

 in the solution of the problem. 



The circles represent the position of points resulting from plotting 

 the reciprocals of the periods against the average mean daily 

 temperatures. These points do not fall in a straight line, but at the 

 lower end they are too far to the left, because the average mean 

 daily temperatures are the averages of temperature readings below 

 the threshold of development (ineffective temperatures) as well as 

 readings above this point, and the points at the upper end are too far 

 to the right, because the average mean daily temperatures here are 

 averages of readings above the degree of the maximum rate of 

 development (retarding temperatures) as well as readings below 

 this point. 



A line that would fit these points would be too flat. Now the 

 points that are least affected by these ineffective and these retard- 

 ing temperatures are those between the average mean daily tempera- 

 tures of 68 and 72. A line drawn through these two points cuts 

 the temperature axis at 51.92 degrees. This is so nearly 52 de- 

 gress that, to avoid fractions, 52 was assumed to be the approximate 

 threshold of development. 



The average daily degrees above 52 were then computed with 

 results as shown in column 7. The points marked by dots were 

 plotted by using the reciprocals of the periods and the average 

 daily temperatures above 52 degrees. It will be seen that at the 

 lower end of the line the points come into line with the two points 

 used to establish the line, but at the upper end of the figure they 

 are in the same position as the points located by plotting reciprocals 

 against average mean daily temperatures. 



It was necessary now to ascertain the degree of the maximum rate 

 of development. The only way to determine this point is by trial. 



It was assumed at first that temperatures above the degree of the 

 maximum rate of development were ineffective and did not ac- 



