72 



BEITISH BUTTERFLIES. 



may choose leaves, flowers, or seeds, respectively as their pabulum. 

 On the other hand, there are many exceptions. Celastrina argiolus 

 appears throughout its vast range to choose shrubby Leguminosae, but 

 it is equally at home on a host of plants belonging to other orders, of 

 which holly, ivy, dogwood and buckthorn may be specially noted, and 

 these frequently appear, in some districts, to wholly replace the leguminous 

 foodplants, from which this species has possibly spread, and here it may 

 be well to draw attention to the similarity of the "food" selection made 

 by Callophrys ruibi and this species, full lists of which may be obtained 

 respectively from our account of each species in this volume. The 

 restriction of the larva of Polyommatus donzelii and Aricia eumedon to 

 Geranium, whilst that of Aricia astrarche will also accept plants of this 

 order, although, in nature, it seems usually to choose ELelianihemum 

 vulgaris; may also be observed. The larvae of the true Lycaenids also 

 are aberrant in their food-habits, e.g., Lycaena arion chooses thyme, 

 L. euphemus and L. areas are said to select Sanguisorba, L. alcon, Gentiana 

 pneumonanthe, whilst Scolitantides is also aberrant, the larva of 

 S. orion choosing Sedum, and S. baton, Thymus. Like so many 

 purely alpine species, e.g., certain Coliads, in which the family larval 

 food-habit also tends to Leguminosae, the larva of Polyommatus optilete 

 chooses Vaccinium. On the whole, though, our knowledge of the 

 food-habit of the family is very infinitesimal. Among the exotic 

 groups, of which the larval food -habits are known, considerable 

 specialisation has been noted. Thus the peculiarly Australian group 

 Ogyris appears to be confined, in all its species, to Loranthus 

 (mistletoe), and the larvae are all night-feeders. [See also Journ. As. 

 Soc. Beng., lxix., pt. 2, pp. 198 et seq.] 



The most remarkable modification of food-habit in Lycaenid larva is 

 undoubtedly the development of cannibalistic or wholly carnivorous 

 habits. Thus Thwaites remarks (Moore's hep. Ceylon, i., p. 70) that "it 

 is difficult to realise that the larvae of some species of these lovely Lycaenids, 

 such as Ambly podia, etc., are carnivorous or even cannibalistic in their 

 habits, and do not hesitate to eat their own brethren of the same 

 brood, when any of the latter are commencing their change into the 

 inactive chrysalis state, with their consequent inability to protect 

 themselves from their voracious kindred, who devour them with 

 avidity." Scudder remarks, that the same carnivorous tendency has 

 been observed in several Nearctic species, and it is certainly common 

 among the European Strymonids (see antea p. 65). Chapman says 

 that Langia telicanus, even though supplied with plenty of food, 

 cannot safely be kept in any number together, all those that are ready 

 to pupate earlier than the others being eaten by their later fellows ; 

 as soon as they have spun their silken pads and girth they are in 

 danger, both larvae and pupa? being found with a larger or smaller 

 hole in the skin and the interior gone ; now and then the interior is not 

 quite cleared out, the cannibal being apparently fullfed before his supply 

 of unholy pabulum has been finished. It would thus happen that only 

 two or three pupae resulted out of a dozen Or more larvae, the fate of the 

 others being testified by their remains. So soon, however, as the pupa 

 has become mature and hard it is safe. Edwards notes (Butts. Nth. 

 Amor., ii., Lye. p. 6) that, in confinement, when food is scanty, the 

 larvae of Celastrina pseudargiolus will prey on each other, burrowing 

 into the body the same way as they do into a flower. Frohawk, 

 referring to the cannibalistic habits of L. arion (Entom., xxxvi., p. 58) 



