XL 



Entomologie (19 12). He also observed, that the larvae were made use of by 

 the ants as weaving-spools. This will undoubtedly be the case with various 

 other species of Lycaenidae in Java. According to an article in the Deut- 

 sche Entomologische National- Bibliothek of Sept. is"" 19 10 this subject has 

 been dealt with by H. Viehmeyer in The Philippine Journal of Science V, 

 No. I. Section D. pp. 69 — 72 and 73 — 77, where he enumerates a number of 

 Austro-Malayan larvae, of which the following are also found in Java : — Miletus 

 Symethus Cram, Lycaenae Puspa Horsf., Ubaldus Cram., Celeno Cram., 

 Pandava Horsf., Cnejus F., Boeticus L., Lycaenesthes Bengalensis Moore, 

 LiPHYRA Brassolis Wcstw., Aphnaeus Vulcanus R, and Deudorix Orseis 

 Hew. I have found a good many of these in Java, but, with one exception, 

 I never noticed ants in their neighbourhood. It is of course quite possible 

 that, with the same species, the circumstances might be different in another 

 place to what they are in Java, for instance the particular species of ants might 

 not occur in Java. In the above mentioned journal H. Viehmeyer is also said 

 to have described and illustrated the chrysalis of an Arhopala species, which 

 was found in great quantities in the earthnest of the ant Camponotus Quadri- 

 sectus Smith, and which secrete a sweet fluid just in the same part of the 

 body as the myrmicophile larvae. 



A short time ago, a very remarkable thing was observed in connection 

 with this. Colonel H. I. W. Barrow, in British India saw the butterfly 

 Allotinus Horsfieldi Moore, one very common in Java also, refreshing itself 

 with the sweet fluid of one of the Aphididae, just as the ants do, while it 

 stroked the insect with its long fore-legs, as the ants do with their antennae. 

 An illustration of this curious fact is found in the Fatma of Brit. India Butt. II 

 edited by Bingham, from which it is seen, that the Lycaenidae do not only 

 fulfil the passive role of the Aphididae towards the ants, but, like the ants, 

 take an active part with regard to these insects. In connection with this it is 

 interesting to note, that in the various Allotinus species, as with all Gerydinae, 

 with certain modifications, the fore-legs are formed in a particular way, which 

 will certainly make them peculiarly suitable for this stroking process and this 

 raises the question of whether all the Gerydinae have this habit, and if the 

 transformation of their fore-legs should be regarded as an adoptation for this 

 purpose. 



Thus we see that the Lycaenidae present a case of a passive symbiosis 

 with the ants, and an active one with the Aphididae, both of a friendly nature. 

 But they also act antagonistically towards the ants and towards the Aphididae. 

 The gigantic species Liphyra Brassolis Westw. which inhabits British India 

 and many islands of the East Indian Archipelago and Australia, is found, though 



