For differentialis, a late hatching species, the curve is rather flat, 
indicating that as a rule there is very little difference when the heat comes, 
just so long as it is continuous and there is enough of it to effect a hatch. 
Both minimum and maximum points on the graph are below the zero line. 
The curve for femurrubrum shows a maximum in the first half and a minimum 
in the second half of the period of the curve, both both are below the zero line. 
The greatest effect of a l-degree change in temperature comes as a negative 
effect in the second half. This means that the greater the heat at that time, 
May 16-20, the earlier the development of the infestations. 
Tables for Predicting the Hatch of the End Infestation 
Tables 7a and b to lla and b, inclusive, are for the purpose of predicting 
when the development of the end infestation will take place. This is measured 
by the period when O to 100 percent of the grasshoppers making up the infestation 
will have hatched. 
In column 1 of the "a" part of the tables are shown the inclusive dates of 
the 5-day intervals. These dates are to be used in connection with current 
weather data in column 3. Colum 2 shows the average daily maximum temperatures 
for the intervals for all the data used in the analysis of a species, and these 
figures are treated as constants. In columm 3 are the current average daily 
maximum temperatures for the intervals in column 1, computed from data obtained 
from the weather station in the area. The deviation of these figures from those 
in column 2 is recorded in column 4, with strict attention paid to the proper 
sign. These deviations in degrees are then multiplied by the figures in column 
5, which represent days and show the effect of 1 degree. Again positive and 
negative signs must be considered carefully. The results of these multiplications 
are then totaled in column 6 for the pertinent interval. Signs mst again be 
considered. 
The result, which is in days, is added to, if positive, or is subtracted 
from, if negative, the dates shown in the "b" part of the table. This then 
becomes the predicted hatching pericd when O to 100 percent of the end infesta- 
tion will have hatched. As the season progresses and the weather data for more 
5-day intervals are included, this prediction will change with the computations 
in columns 4, 5, and 6 if the current weather shown in column 3 continues to 
differ with the average figures given in column 2. Thus any marked difference 
or change in weather will show up at once in the prediction. The unfolding 
effect of the spring temperature pattern on the seasonal development of an infes- 
tation can be computed with a reasonable amount of confidence up to any time as 
the season progresses. 
How Prediction Tables Are Used 
Two examples, one of early and the other of late hatching singuinipes, have 
been selected to illustrate how ‘the prediction tables are used. Both examples 
are actual cases of the hatching periods of sanguinipes. The case of the early 
hatching was at Circle, Mont., in 1939, when 10 to 90 percent of the hatch took 
place between May 2 and 16, inclusive. The case of the late hatching was at 
Carson, N. Dak., in 1945, when 10 to 90 percent of the hatch did not take place 
until July 2 to 23, inclusive. The average maximum temperatures for the 5-day 
= ON 
