FAMILY HABITS IN BUTTERFLY LARViE THE LYCJENIDS. 77 



will soon be apparent, and, suddenly, there protrudes a dull green, fleshy, 

 mammilloid organ, from the top of which comes a tiny drop of clear 

 green fluid. This the ants drink greedily, two or three of them 

 perhaps standing about it, and they lick off the last trace of it, stroking 

 the segment meantime. As the drop disappears, this organ sinks in 

 at the apex and is so withdrawn. . . . The intervals between the 

 appearance of the globule varied with the condition of the larva ; if 

 exhausted by the long-continued solicitings, some minutes would 

 elapse, and the tubes meanwhile remained concealed, but a fresh larva 

 required little or no urging, and one globule followed another rapidly, 

 sometimes even without a retracting of the organ. Six emissions have 

 been counted in 75 seconds. The larva did not always await the approach 

 to the 11th segment, but gave out the drop unsought, and as soon as 

 it was aware of the presence of the ant. Now and then the drop 

 was preceded by a bubble several times larger than itself. . . . When 

 a fresh larva, taken from the house, was placed on the stem, there was 

 an immense excitement among them, and a rush for the last segment, 

 as soon as the ants discovered it. The larva forthwith relieved itself 

 by the excretion of the fluid, and the tubes stood out with domes 

 expanded between the times of secreting ; if a fresh larva were placed on 

 a stem on which were no ants, there was no excitement in the larva, 

 no appearance of the tubes, and no movement in the 11th segment, 

 . . . but, if ants were now transferred to the stem, the moment the 

 caressings began the larva changed its behaviour. From what I have 

 seen I am led to believe that these tubes are merely signals to the 

 ants, and that, when the latter discover them expanded, they know 

 that a refection is ready, and rush to the orifice on the 11th segment." 



Thus, a century after it had been noted that Lycaenid larvae were 

 attended by ants, and some years after the honey-gland and its secretion 

 had been discovered by Guenee, Edwards was able to show the con- 

 nection between the two and the symbiotic relationship existing 

 between the ants and larvae. Other observers were put on the right 

 track, and Saunders noticed (Can. Ent., x., p. 14) the connection 

 between ants and the larvae of Plebeius scudderii, whilst Edwards 

 himself described in detail the connection between the larva of Plebeius 

 melissa and its attendant ants (Papilio, iv., pp. 92-3). 



We are not at all sure when the connection between ants 

 and Lycaenid larvae was first noticed in Asia. In 1881, Moore 

 quoted (hep. Ceylon, i., p. 70) Thwaites as saying that the 

 larvae of some Lycaenid species, e.g., Amblypodia, were canni- 

 balistic in their habits, but a protection is found "for these helpless 

 individuals, in the instinct of an ant, Formica smaragdina, which, 

 finding a substance most palatable to it, secreted naturally from a 

 defined glandular spot upon the bodies of these helpless larvae, takes 

 possession of them as 'cows,' surrounding each separate one . . . 

 protecting them zealously and attacking most fiercely any living thing- 

 intruding upon them." This was followed up by Doherty in 1886 

 (Journ. As. Soc. Beng., Iv., pt. 2, p. 112), and by de Niceville and Wylly, 

 in 1888 (Journ. Bomb. Nat. Hist. Soc, iii., p. 164) and supplemented 

 by de Niceville, in 1890 (Butts. India, etc., pp. 7-9), and in 1901 

 (Journ. As. Soc. Beng., lxix., pt. 2, pp. 187-192). The remarks of the 

 two former we have already noted at length in the preceding volume 

 (A at. Hist. Brit. Butts., L, pp. 33-34). De Niceville writes (Butts. Ind.) : 



