542 THE BUTTERFLIES OF NEW ENGLAND. 



'Sconset road, but most abundantly in the Middle Pasture and especially 

 to the west and southwest of Gibbs Pond. In a half acre patch of 

 golden rod just next the sandy beach at the western end I have found them 

 always present in their season, and easier of capture than in the adjoining 

 pasturing ground. They appear fondest of alighting on the flowers of 

 golden rod and Vernonia, tliough (with the possible exception of asters) 

 they are then the most common flower there. 



Oviposition and food plants. I took one female in Nantucket on 

 September 11, in the act of laying eggs on a composite plant, one of the 

 Asteroideae, Sericocarpus conyzoides ; but all that have been raised 

 have been fed on violet. Specimens enclosed by me on living violets lay 

 indeed a few eggs on it, but these were as nothing compared to the mul- 

 titudes, — some hundreds, — laid on the lace which kept them near the A'iolet ; 

 these eggs were in all cases laid erect upon the outside of the lace, the 

 female thrusting its ovipositor through the mesh (only a millimetre and 

 a half in diameter) curving the tip around and depositing the egg, always 

 on its base. The meshes were felt for as with a snout and many would be 

 tried and pierced before a satisfactory position seemed to be found : the 

 action was repeatedly seen, and the net, sprinkled with eggs on the outside 

 while the parents were within, had a very curious appearance ; this leads 

 me to conclude that in nature the eggs are laid upon the under surface of 

 the leaves by the parent seated on the upper surface. The eggs hatch 

 with us in from twenty-nine to thirty-three days ; some, however, which 

 were laid late in the season remained unhatched, and in the hope of 

 carrying them through the winter in this state were placed in a cold stor- 

 age chamber, l)ut the caterpillars never emerged from them. 



Duration of early stages, etc. Mr. Edwards, who is the only one 

 that has raised this insect from the egg to maturity, found in West Virginia 

 that the duration of the *igg was from twenty-three to twenty-five days 

 (about a week shorter than in the north) and the diflferent stages of the 

 larva from twelve to twenty-three days each, not counting the hibernation, 

 and the chrysalis state seventeen days, making in all, omitting the hiber- 

 nation, one hundred and forty-four days. The caterpillars usually devour 

 the entire ^^g shell as far as the base as soon as they emerge and then at 

 once seek shelter and pass into the winter lethargy. Mr. Edwards found 

 that most of his took refuge at the base of the leaf stalks of the violets on 

 which he placed them, but this was probably because a more suitable 

 place was denied them. 



Life history. The single brood of this butterfly occasionally appears 

 with us as early as June 25, and then becomes tolerably common by the fii-st 

 of July and abundant after the first week in that month ; more commonly 

 however its first appearance is delayed until the first week in July, and 

 its subsequent abundance is correspondingly })Ostponed. Males only can 



