io6 



" Larva when full grown, appears to be rather large, considering the size 

 of the butterfly and is -f^ of an inch in length ; colour pale green, the body of 

 nearly equal width throughout, the fourth segment rather the widest, the 

 constrictions between the segments hardly visible, the head large, (much larger 

 than in any lycaenid larva known to me), black and shining, hardly hidden 

 beneath the second segment, being quite visible from in front; the second 

 segment marked with a large shining blackish patch which is divided in the 

 dorsal line by a whitish line, with two similar but broader lines on each side ; 

 the third segment in anteriorly similarly marked; there is a double fine dorsal 

 and subdorsal dark green line and a lateral single line; the three posterior 

 segments are marked above much as are the second and third; the twelfth 

 segment bears two prominent blackish pillars, from the upper edge of which 

 spring several strong brisdes. When frightened, the larva protrudes a somewhat 

 long pale green tubercle from each pillar, which bears at its apex a few fine 

 hairs. The pillars and tubercles are larger in this species than in any other 

 known to me except Curetis Thetys Dury. The mouthlike opening in the 

 dorsal line on the posterior edge of the eleventh segment is very conspicuous 

 under a magnifying glass. The whole body is finely shagreened and the lateral 

 edge and anal segment bear a fringe of numerous somewhat stout colourless 

 hairs. The larva in Calcutta feeds on Clerodendron Siphonanthus P. Br. 



Pupa, always found in a spun-up leaf or leaves, is either green or dark 

 brown, of the usual lycaenid shape, smooth and shining, the head rounded, 

 the thorax anteriorly slightly humped and angled at the sides, the abdomen 

 gradually tapering posteriorly. There is much of interest in the habits of the 

 larvae of A. Vulcanus. They are most carefully tended by two somewhat 

 small species of black ants, which Dr. A. Forel of Genoa, has identified for 

 me as Pheidole Quadrispinosa Jerdon and Cremastogaster N. S. (Niceville 

 Forel M. S.). A full dozen of these ants may be seen all at once on the body 

 of a full grown larva, and many others round about, so covering the larva 

 that little else but ants is visible ; the larvae do not seem to mind the ants at 

 all. The larvae pass most of their time in roUed-up leaves (only issuing forth 

 when hungry to eat the surrounding leaves, always returning to their shelters 

 when the meal is over), several in each shelter, four being the greatest number 

 I have seen in any one shelter. Larvae of very different ages are to be found 

 in the same shelter. Some of these nests are formed of the separate leaves 

 spun together with silk, but usually the outer edges of a single leaf are spun 

 together. When about to pupate, the full-grown larva spins a cocoon between 

 two leaves. It is very slight, and both ends are left open ; it is made of white 

 silk, the entire structure being exactly like the nest certain green spiders spin 



