BEVDOBIXINM. 71 



eleventh segment that are often present in lycsenid larvae, and I presume such would 

 be useless, as ants do not attend these larvae as far as I have seen. The larvae feed on 

 the fruit of Randia dumeiorum, Lamk. The above description has been taken from 

 larvae obtained in Sikkim at about 2,500 feet ; in January on the 20th of the month the 

 fruit contained larvae full grown, or nearly so and pupae, (de Niceville.) 



The larva has the same curious instinct as V. i-socrates, Fabricius, and needs it more, 

 for the Ghela fruit withers at once where attached, and would inevitably fall before its 

 tenant had reached the pupa state if not artificially supported. I have found only one 

 larva in each fruit, and have sometimes noticed ants going in and out of the hole made 

 by it, for what purpose I cannot say. The stony hardness of the fruit turns the edge 

 of one's penknife, and of one's curiosity too. This butterfly has the habit of taking its 

 station, during the hottest hours of the day, on a particular leaf, from which it darts 

 out in pursuit of every other butterfly that passes by. (Aitken.) 



Having lately reared several larvae of Virachola perse, Hewitson, for the purpose 

 of investigating the action of the ants attendant upon them, I now give an account of 

 my observations, which, although very incomplete, may serve as a stepping-stone to 

 further researches, if not as an explanation of the several rather conflicting accounts 

 upon the subject. In the first batch of larvae I obtained on the Fagoo Tea Estate, 

 British Bhutan, at 2,500 feet elevation, in June, 1895, the larvae were about half- 

 grown, and feeding on the interior of the fruit of wild pomegranates. In every case 

 one larva occupied a fruit to itself, with one exception only, in which the fruit was 

 inhabited by a half-grown larva, and was bored near the apex by a very small larva. 

 The small larva, however, soon left the larger in full possession, and sought a fruit for 

 itself. Some of these half-grown larvae were attended by a black ant of slow 

 movements with extremely flattened head and abdomen. As the hole made by the 

 larva in the fruit was the same size as its anal scutate segment, and that segment only 

 was exposed, only two ants at most were found attendant upon the larva. The 

 excrement of the larva, which would otherwise have filled up the hole, was presumably 

 removed by the ants in order to allow themselves entrance. Although I never 

 happened to observe this operation, still it is probable that it was so, as I occasionally 

 found the hole filled with excrement, the attendant ants being on the outside of the 

 fruit, and soon after found the passage cleared and the ants busied on the exposed 

 segment of the larva. Of course, it is quite possible that the larva itself removed the 

 stoppage by backing, as it must have done where no attendant ants were found. In the 

 earlier stages the larvae seem in a particularly unsettled state, residing in the interior 

 of one fruit for a few days only, and then beginning on another. They are occasionally 

 found on the outside of the fruit for a few hours, but I am of opinion that they do not 

 remain outside for any great length of time, except in any case of injury, when they 

 prefer to come outside to die. After this first batch of larvae had been in captivity for 



