466 THE PHOTOTROPISM OF THE MOURNING-CLOAK BUTTERFLY, 



on, I followed two to their hiding-places. One alighted on the trunk of a fallen tree 

 and, without expanding its wings, crept immediately into a large crack in the bark. 

 The second settled on a stone fence and crept into a hole between some loose stones. 

 The period during which this occurred was marked not so much by a diminution of 

 light as by a rapid fall in temperature. I can judge of this only by the coolness I 

 felt myself, for I was unfortunately without a thermometer, but the rapid cooling 

 of the air was unmistakable. I therefore believe that the most important factor 

 in inducing a retreat of these butterflies is a decrease in temperature rather than in 

 light, and that the butterflies remain under cover at night because of coolness rather 

 than because of lack of light. 



That this explanation is probably correct may be seen from the following experi- 

 ment. A cage made of fine wire gauze and containing half a dozen mourning-cloak 

 butterflies was placed opposite a west window in a cool (7° C.) basement room. The 

 day was overcast so that only diffuse daylight entered the window. In about ten 

 minutes all the butterflies had taken up resting positions with their wings closed, and 

 they remained in these positions even when the cage was shaken. After they had 

 been in this quiescent state over an hour, the cage was transferred to a warm (21° C.) 

 basement room and placed opposite a west window whose light relations were almost 

 exactly like those in the first room. Within ten minutes all the butterflies were 

 actively fluttering about, though the only significant change in their surroundings 

 had been that of temperature. On transferring the cage to its original position in 

 the cool room the insects in a few minutes again became quiescent. I therefore 

 believe that the states of repose and of activity are subject to control through tem- 

 perature. This problem, however, is one for further experimentation. 



The retreat of the butterflies as coolness comes on must not be confused with 

 their sudden disappearance, often observed even at midday. If, while the woods 

 are warm, the sun is suddenly overcast and remains so for some time, the butterflies 

 disappear. They, however, have not crept into holes and other hiding-places, but 

 have simply folded their wings, and in this inconspicuous condition they remain quiet. 

 Soon after the sun comes out they are on the wing again. This reaction seems to 

 depend on sudden light diminution, and I believe it to be essentially different from 

 the retiring of the animals for the night or for the winter hibernation. 



