STUDIES ON PERILLUS BIOCULATUS FAB. 87 
to be no change in color of the bugs, and this is just as might be ex- 
pected, since the physiological processes are at a standstill. It also 
shows that temperature of itself can not affect changes in color but 
only influences physiological processes during the active life of the 
insect. However, rapid changes of color have been observed in the 
two or three weeks prior to entering the state of hibernation, and 
again during the period when the bug becomes active in the spring. 
Relation of Temperature to Color Forms 
Rather early in the breeding work, the rearing of a high percentage 
of white color forms was correlated with high temperatures, both in 
the laboratory and in field collections. This was rather forcibly im- 
pressed on the writer during August, 1914, at Batavia, N. Y., while 
breeding for pure lines of color in Perillus. The progeny of pair 
No. 27 (yellow forms) began maturing August 3, and by August ro 
twenty-one adults had emerged, all of which were yellow forms except 
one, which was white. From August 5 to 10, inclusive, the weather 
turned unusually hot, the temperature ranging between 83°F. and 
95°F ., the minimum temperature not going below 76°F. in the labo- 
ratory, and that only early in the morning. Of thirteen specimens 
that matured on August 14 and 15, two males were yellow, while three 
males and eight females were white. The appearance of white forms 
in other breeding jars was-also conspicuous. From August 12 to 21 
inclusive, the temperature remained much cooler, not going above 
78°F. Between August 16 and 21, a total of thirty-nine yellow 
forms and only four white bugs matured in a group of nine breeding 
jars, which also included the progeny of pair No. 27. To the writer 
this was rather convincing evidence that external conditions, notably 
temperature, were having much to do with the color of the maturing 
bugs. However, breeding for pure lines of color was not discontinued 
at that time, but was kept up through the following year. Eventually, 
all progeny of the white parents became yellow or red forms, this 
occurring in the fall generation or in the spring. The progeny of 
red parents all became yellow or white during the hot weeks of July 
or August, and thus the parent type could not be maintained. Again 
in 1921 an effort was made to breed for a pure line of white color 
forms, but all the progeny from three pairs of white bugs became 
either yellow or red during the second or early-fall generation. 
The influence of humidity on color varieties of Perillus has not 
been tested under conditions with absolute control, yet bugs have been 
reared under two extremes of humidity where the temperature was 
kept the same for both lots. This was accomplished by rearing bugs 
