CARNIVOROUS HABITS OF BUTTERFLY LARVJE. 87 



the bead of liquid in place, and probably also serve as a protection to 

 this apparently sensitive organ. The larvae appear to be perfectly at 

 home with the ants, neither molesting the other (Entom., xxxvi., 

 pp. 58-60). 



CHAPTER X. 



CARNIVOROUS HABITS OF BUTTERFLY LARViE. 



The fact that, under certain conditions, in confinement, lepidopterous 

 larvae will live on others of their own or different species, is well known, 

 and the habit of the larva of Thecla w-album to leave its food and feast 

 on the newly-formed pupaa of its own species, has frequently been 

 observed and recorded. That certain butterfly larvae should, however, 

 have a permanent carnivorous diet, is sufficiently unusual for us 

 to devote a short space to the details of one or two of these cases. 



The best known of these species is Feniseca tarquinius, an American 

 insect which Scudder makes a Chrysophanid, and which is undoubtedly 

 a Lycaenid in sens. lat. The larva of this butterfly is purely carnivorous. 

 The eggs are laid in the midst of a group of aphides, or near thereto, 

 and, for protection, are coated during deposition with a thin 

 coagulated albuminous deposit, which, on hardening, covers them 

 like a thin but irregular veil, the egg- stage lasting only three or four 

 days. The larva appears to live entirely on plant-lice, particularly affect- 

 ing the species Schizoneura tessellata (on alder), Pemphigus fraxinifolii 

 (on ash), P. imbricator (on beech), all of which produce much flocculent 

 and saccharine matter ; it has also been fed on aphides from willow 

 and plum, in confinement. The young larva eats a hole through the 

 summit of the egg, and pushes its way under the larger aphides, and 

 forthwith begins to spin for itself a loose web, not close enough to 

 conceal it from view were the aphides away, but sufficient to keep the 

 aphides from walking over the body, and to protect it when a moult 

 is approaching, and the skin sensitive. The web seems to be just about 

 the length of the larval hairs from the body. The aphides may be seen 

 running over it, and often get their legs fast in the meshes, and are apt 

 to be devoured as a consequence ; the larvae appear to pass both the 

 first and second moults beneath this web, but, after this, seek fresh 

 supplies of food, devouring the aphides from the underside, their backs 

 covered with wool from their victims (Butts. New Engl., ii., pp. 1022 et 

 seq). The most remarkable fact, however, connected with this larva, is that 

 the ants, which nurse other Lycaenid larvae, are its sworn enemies, for, 

 by feeding on the aphides that the ants keep, they destroy the source 

 of supply of the ants' sweet food (secreted by the aphides), and are 

 furiously attacked and killed by the latter ; it appears only to be in 

 their later stages, when feeding largely exposed, that the ants are able 

 to successfully deal with them. 



Kershaw gives details (Trans. Ent. Soc. Lond., 1905, pp. 1-4) 

 of the connection between the Chinese Gerydus chinensis and 

 aphides, chiefly those frequenting various species of bamboo, the eggs 

 being laid among a crowd of aphides, and often hidden under 

 a mass of them. They hatch in about four days, the larva at 



