We will begin with what we shall call Group A: The butterflies are on the wing in May and early June. They have just come from the chrysalis and continue living for three or four weeks before they deposit eggs, this time being required in order that the eggs may develop in the ovaries of the butterflies. These eggs hatch in about a week and the caterpillars become full grown a month later. They then change to chrysalids in which condition they remain another week, thus requiring five or six weeks for the newly laid eggs to mature into butterflies. Supposing the eggs were laid the first week in June, the butterflies of this second brood would appear about the middle of July. The eggs in the ovaries of some of these butterflies also require several weeks before they are ready to be laid, so that it may be about the tenth of August when this happens. These hatch and mature to chrysalids during the next six weeks, the butterflies of this brood emerging about the middle of September. These in turn lay eggs at once apparently, no extended period being required for their development before they are laid. The eggs hatch during the latter part of September and the young caterpillars feed upon the violet leaves for two or three weeks, moulting perhaps twice and becoming approximately half grown. They now stop feeding and go into a lethargic condition in which they hibernate. Then in spring they awaken and feed again upon the violet leaves for a short time, becoming mature and changing to chrysalids sufficiently early to emerge as butterflies late in May. In this hypothetical group we have a fairly normal condition of a three-brooded butterfly hibernating in the stage of the half-grown larva and requiring some weeks for the development of the eggs in the ovaries of the butterflies in the case of the first two broods but not of the third. In another group, which we may designate as B, the conditions may be similar except that the butterflies lay their eggs very soon after coming from the chrysalis in the case of all three broods. Obviously there would be a tendency here for hastening the earliness of the broods so that the hibernating caterpillars might either become larger or might go into the hibernating condition earlier than those of Group A. In Group C, the variation takes place in the larvae rather than in the butterflies. These may go on in the normal way up to the time the caterpillars of the summer brood become half grown. Then they become lethargic, ceasing to feed and to all appearances going into hibernation. They remain in this condition until the following spring when they come forth from their winter's sleep and feed upon the violets in precisely the same way as the caterpillars of the third brood of Groups A and B.