6 LYC^NA 11., III. 



till 1878, when many eggs were obtained by confining a female butterfly over a 

 branch in flower. After that, there was no difficLilty in finding both eggs and 

 caterpillars. The former are laid singly, low down on the side of a floweret, and 

 usually well within the flower head. As soon as hatched, the young larva cuts a 

 minute hole, the diameter of its head, into the lower part of the imopened bud, 

 just above the calyx, and feeds upon the filaments of the stamens. After its first 

 moult, it bores into the side of the calyx to get at tlie ovules ; but as the flowers 

 mature and the ovary hardens, the boring is from the top, inside the tube of the 

 calyx, and follows the stalk of the pistil to the ovule. Finally, belated larviB 

 are compelled to gnaw the seed vessel after it has become woody, and in several 

 instances have been found eating the stem below the flower. It is not unusual 

 for the larvaj in confinement to eat of the white involucre of the flower, but I 

 have never observed them to eat of the leaves, even when no other food has been 

 given them. As the eggs laid when the flowers of Cornus are in bud produce 

 larva3 which mature at about the time the flowers fall, it follows that many be- 

 lated larvre must starve. 



I do not know of any other food plant for the winter brood than Cornus, 

 though in confinement the larva? have eaten the tlowers of Begonia, Nasturtium, 

 and Asclepias; also Clover blossoms, but not readily, and females confined over 

 Clover have refused to lay eggs on it. The larva; on Cornus, in their later stages, 

 vary greatly in color and markings, having more or less green, either liglit or 

 dull, with white, brown, and crimson. But in the younger stages they are much 

 the color of the flowers they feed on, and are thus in some degree protected from 

 their numerous enemies, spiders, hemiptera, etc. In confinement, when food is 

 scanty, they will prey on each other, burrowing into the body in the same way 

 they do into a flower. 



Following Violacea, and flying at the same time with the latter half of that 

 generation, comes Pseudarrjlolus, the largest form of the seriej, and dift'ering 

 considerably from Violacea in general appearance. 



In 1877, I observed a female of this form hovering about a stalk of Rattle- 

 weed, Cimicifuga racemosa, which was in bud, and this suggested the confining 

 one of them in a bag over the plant. The immediate result was that many eggs 

 were laid, and thenceforth, in successive years, eggs and larva? have been found 

 in abundance. The Rattle-weed (Fig. 2, Lye. III.) sends up a stalk to the height 

 of five or six feet, branching more or less, and each branch terminates in a spike 

 from eight to twelve inches long, bearing round, greenish- white buds, arranged 

 in rows. The lower buds mature first, and the flowering proceeds slowly from 

 base to top, the whole period lasting at least six weeks. The eggs are usually 

 laid on the buds, and the young larva bores into the side, and gradually eats the 

 contents, till a mere shell is left ; then moves to a fresh bud, and so on. 



