31^8 THE BUTTP:RFLIES of new ENGLAND. 



thirty or forty. Yet they do not seem to be strictly gregarious, although 

 so very abundant in certain })laces as to bring the caterpillars into compar- 

 atively close contact. Not many eggs are laid on a single leaf, and though 

 usually, at least while young, more than one caterpillar is found on a leaf 

 (rarely more than four or five), these are never found feeding in rows side 

 by side, as in the allied genera whose larvae are gregarious for a part or the 

 whole of their lives. In his later writings Mr. Edwards has more cor- 

 rectly said that "the young larvae gather into a loose colony." 



From the very start, in feeding, holes are eaten through the leaf, and 

 the caterpillar "during the first stages feeds about the margin of this hole." 

 During its whole life it rarely seeks any other concealment than to live on 

 the under surface of the leaf, but in one case I discovered them on hop 

 making nests precisely similar in every respect to those made by P. comma. 

 It moves rapidly when young, and spins a thread very carefully for at 

 least the first half of its life and to some extent until maturity, and it is 

 very tenacious of it. It has the curious habit when resting after a meal 

 of turning the front part of its body around abruptly, in which case the 

 jaws come opposite the first pair of prolegs, and the head is held angu- 

 larly, so that the coronal spine of that side of the head nearest the side of 

 the body is uppermost. 



Pupation. The chrysalis is often suspended from the leaf or stem of the 

 plant on which the larva has been feeding. Mr. Angus once found one on 

 a leaf of the Kentucky coffee tree (Gymnocladus). Mr. Gosse found one 

 specimen "attached by the tail to a growing stalk of grass and of course 

 hanging parallel to it." It is more frequently attached to the under sur- 

 face of palings, etc. Judging from the dates given in Harris's Corres- 

 pondence, the chrysalis state lasts in the north from eleven to seventeen 

 or even twenty-six days. Gosse (in Canada) says eleven days, Edwards 

 (in West Virginia), seven to eleven days. Braun in Bangor had them 

 hang twenty days at the end of July. From an experience Miss Murtfeldt 

 had in rearing this insect, she concluded (Psyche iv : 184) that the chrysa- 

 lids with gilded spots on the back Avei'e those of the female, while those of 

 the male were not thus ornamented and were darker. But this was merely 

 an accidental occurrence. 



Life history. This insect is double brooded, the butterfly hibernating 

 and appearing again early in May. It disappears by the middle of June 

 or earlier, having laid its eggs late in May and early in June. These 

 hatch in from four to ten or eleven days, the larvae attain their growth 

 rapidly, and after passing generally from eleven to fourteen days in the 

 chrysalis state emerge as butterflies in July, some early specimens appear- 

 ing during the last days of June. They continue on the wing until nearly 

 the middle of August, laying their eggs late in July and throughout 

 August, and the butterflies of the second brood appear toA\ard tlic end of 



