THE GREGARIOUS HABIT IN BUTTERFLY LARV.E. 15 



its structure, as a winter residence, but then invariably leave it on the 

 approach of spring, and part company, though often being still found 

 in close proximity. Others leave it at the hybernating season, to seek, 

 each for itself, its own hiding-place." Scudder then adds that "this 

 alteration of habit from companionship to solitariness is a natural 

 incident due to growth," a statement that may be true of Laertias 

 philenor, and some other cases, but cannot, one supposes, apply 

 to all instances. He says : " Up to the end of the third moult 

 the size of the caterpillar has not increased enough to make it a 

 conspicuous object, but, by the time the third moult is passed, the 

 caterpillar is half -grown, and, during this stage and the next, its 

 size becomes an important element in its security, and this alone 

 is sufficient to account for the fact that mature caterpillars of 

 butterflies are rarely found in company. It is at this stage, 

 too, that, in many instances, the winter season overtakes the cater- 

 pillar, and it hybernates, and since, in the spring, it revives when the 

 plants have put forth but tender leaves, impossible to nourish more 

 than one, or at most two, such ravenous beasts as now come out of 

 their winter-quarters, such a change of habit would seem to be 

 compulsory. Possibly the change in habit which generally takes 

 place at this middle period of caterpillar life, even when winter does 

 not intervene, is an inheritance from a common ancestor, whose 

 habits were fixed by the necessity of hybernation at this age." 



We have already referred to the gregarious habits of the larvae of 

 Melitaea aurinia, M. cinxia, and M. atJialia, and we may here note 

 that the American Melitaeids have practically identical habits ; thus the 

 oviposition of Euphydryas phaeton, as described by Emery, Edwards, 

 and Scudder (Butts, of New England, i., p. 696), might do almost 

 equally well for that of M. aurinia, and the marshy habitat is similar 

 to many the latter loves ; for Edwards says (Can. Ent., xvi., p. 131) 

 that the foodplant grows always in swampy places, often half under water, 

 and the webs in winter are beaten down by snow and rain, but the 

 inhabitants get through safely. Emery states that a batch of eggs he 

 discovered upon a leaf of Chelone glabra, consisted of three layers of 

 irregular outline, the first 8*5mm. long and 5'5mm. broad, this was 

 the floor of the second, which was rather smaller, and this in turn the 

 floor of the topmost layer, 6mm. long and 3*5mm. broad. The total 

 number of eggs w 7 as supposed to be about 200, and Edwards has 

 estimated similar clusters as containing from 100 to 400 eggs, and, in 

 one, instance there were five layers. The newly-hatched larvae prepare 

 a small web upon the under surface of a leaf, the web thin, and 

 covering a space only sufficiently large for feeding. They feed in 

 rows, those of each row simultaneously moving the head and anterior 

 part of the body from side to side, frequently wandering uneasily and 

 rapidly from the web, but always returning again. After a day or two, 

 however, the young larvae no longer feed on an open web, but bend 

 the leaf upon itself into a knot, or construct a covering web, usually 

 on the topmost leaves of the stem, and feed on the green leaves 

 enclosed, whilst, as they are consumed, the web is extended down the 

 stem, covering fresh leaves. As the larvae grow they enlarge the web, 

 all working for the common good, and, especially as each moult 

 approaches, all wanderers come home, the web is made tight, and into 

 it they retire and pass the moult, which over, the web is extended 



