234 z 
The average mean daily temperatures and the length of the periods 
were determined for each observation. In all the groups of eggs, larve, 
and pup which began their periods on the same date there was a con- 
siderable variation in the recorded length of the period of individuals, 
due, no doubt, partly to the fact that individuals differ somewhat in 
respect to the amount of time required for them to develop even under 
identical conditions as to heat, light, moisture, and environment; and 
partly to errors of observation due for the most part, especially in case 
of eggs and pupe, to the use of the day as the unit of time, on account 
of which the recorded period may, if the observations are made at the 
same hour each day, be almost a day longer or shorter than the true 
period, and, if the observations are not made at the same hour, the 
recorded period may be more than a day longer or shorter than the true 
period. 
To avoid these variations the observations beginning on the same 
dates were thrown into groups, and the average period and average mean 
daily temperature were determined for each group. These smaller groups 
were then thrown into larger groups, using the mean daily temperature 
as the basis, all those passing through the period at an average mean daily 
temperature of 60 and 61 degrees or fraction thereof forming one group, 
and those passing through the period at temperatures of 62 and 63 
degrees or fraction thereof forming another group, etc. The mean daily 
temperatures and the average length of the periods of these groups were 
then computed. All averages are weighted and all periods were averaged 
harmonically. 
The detailed tables which it was necessary to prepare for the purpose 
of making the computations are voluminous and would not be of special 
interest. Only summaries made as above indicated will be given. 
The reciprocal curve was used, with the mean daily temperature for 
the horizontal axis and the reciprocals of the periods for the vertical axis, 
to help locate approximately the zero of development and the degree of 
maximum rate of development; but this had to be supplemented by other 
methods on account of the displacement of points at the upper and lower 
ends of the line because of temperatures above the degree of the maxi- 
mum rate of development and temperatures below the theoretical zero. 
RELATION OF TEMPERATURE TO DEVELOPMENT OF THE EGG 
The eggs used for observation were obtained from pairs of moths 
isolated on branches of trees in small wire cages. These cages were 
examined morning and evening. The day was used as the unit of time. 
Eggs found in the cages in the morning were recorded as having been 
laid the preceding day and those found in the afternoon were recorded 
as having been laid the same day. It is possible, therefore, that in some 
cases the recorded length of the period may be either almost a day longer 
or shorter than the actual period, but the averages of a large number of 
observations should approximate the true period very closely. 
