916 rilK BUTTERFLIES OF NEW ENGLA^riX 



A caterpillar feeding on a {)lant flourisliing well in water wished to 

 change to chrysalis and early on August 25 took up position on the upper 

 surface of a leaf. The next day, noticing it had kept the same position 

 but seemed a little less visible than before, I observed that it had woven 

 together two overhanging leaves with a web of silk, so as entirely to con- 

 ceal it from view above. On August 27 A. m. it had spun a loose band 

 across its body and noticing that the overhanging leaves seemed now still 

 a little nearer I discovered that the tip of one was attached by a few very 

 slight long threads at considerable distance from each other to the leaf on 

 which it itself rested. The head of the larva is completely concealed 

 while it is restintf for the chanifc to occur. 



Life history. It is triple brooded ; the first butterHies appear from 

 the sixth to the tenth of May and the brood usually becomes abundant in 

 five or six days and continues plenty throughout the month ; by the mid- 

 dle of June, sometimes by the end of the first week it has disappeared. 

 The effss are doubtless laid toward the end of Mav and eai-lv in June ; 

 they hatch in a few days, the caterpillar grows rapidly and the chrysalis 

 continues but a short time,* for the second generation usually makes its 

 appearance between the sixth and ninth of July, occasionally as early as 

 the fourth and sometimes as late as the twelfth or fifteenth ; it is abundant 

 in the latter half of the month but afterwards becomes rare, although 

 fresh specimens may often be obtained the last of July and even very early in 

 August and a few specimens continue on the wing until the third genera- 

 tion makes its appearance. The eggs of tliis brood are laid principally in 

 the latter half of July, those obtained by me being laid July 14 to 19 and 

 31 ; the caterpillars, maturing rapidly, attain their growth during the 

 second week in August and after nine to eleven days spent in the chrysa- 

 lis (Harris) the third generation appears, about the nineteenth or twentieth 

 of the month, thougli sometimes as early as the fifteenth, and remains on 

 the wing until at least the end of the third week of September ; this 

 brood must lay its eggs early in September, giving the caterpillar time to 

 attain its full growth before winter and to hibernate in this condition 

 (which analogy with its Eui'opean and Californian representatives renders 

 wholly probable), remaining unchanged until about a fortnight before the 

 appearance of the butterfly in the spring. The European species, how- 

 ever, is only double l)roodcd and its history may differ from ours in other 

 respects. 



In the north, as among the White Mountains, our butterfly, too, can have 

 but two broods, since specimens found just before the middle of July — the 

 remnants of the first brood — were all worn. The number of broods in the 

 southern states is unknown, but is very likely more than three, for the first 

 generation appears as early as April 12th (Chapman) and the second about 



* In the south but eight ihiys, according to Abbot. 



