76 BRITISH BUTTERFLIES. 



the structure of the caruncles of the 8th, and the honey-gland of the 

 7th, abdominal segments, as well as the secretion of a fluid by the 

 latter, were first fully detailed by Gruenee [Ann. Soc. Ent. France, 4th 

 ser., vii., pp. 665-668), as they occur in larvae of Lampides boeiicus (see 

 postea, pp. 348-350). But it was not until 1877 that Edwards (Can. 

 Ent., x., pp. 1 et seq.) first discovered the true meaning of the con- 

 nection between ants and Lycaenid larvae. The species in which it 

 was first noted was Celastrina pseudargiolus, the North American form 

 of our common C. argiolus. He noticed that- ants frequented the 

 flower-spikes of Cimicifuga racemosa, on which the larvae of Celastrina 

 pseudargiolus were feeding, and observed "an ant running up and 

 down the back of one of the larvae, drumming and gesticulating with 

 its antennae, and was surprised to find that the larva, though feeding, 

 did not seem in the least disturbed at the treatment, neither with- 

 drawing its head from the bud nor wincing in the body. . . . A further 

 search showed other ants, and sometimes several of them, busy about 

 other larvae, running from one to another, on different parts of the 

 spike, and always repeating the same drumming motions, stopping 

 often to lick the surface, as it seemed, and the presence of ants 

 became a sure indication of larvae." . . . The ants attended most 

 diligently to the last two or three segments, and especially to the back 

 of the 11th (7th abdominal)," etc. But, though Edwards suspected 

 that a saccharine fluid was secreted, and that it was this that the ants 

 sought, he could not demonstrate it, so he sent larvae to Hagen and Mack, 

 who discovered the three organs previously described, and similar ones 

 in larvae of Plebeius argus and Agriades corydon, and, later, in Everes 

 comyntas. Hagen also sent to Edwards Guenee's account of the 

 structures, and his observations on the secretion of the honey-gland, 

 as observed in Lampides boeticus (see postea, pp. 348-350), and 

 Edwards saw, in this account — that, "from the middle of the 

 opening (on the 7th abdominal) comes forth, at the will of the 

 larva, a sort of transparent hemispherical vesicle, wmich gives 

 escape to a fluid sufficiently abundant to form a good-sized drop, 

 which reproduces itself when it is absorbed " — the explanation of all 

 his puzzles, and concluded that " the larvae of several species of 

 Lycaenids have one or more special secreting organs, and that one 

 species, at least, is regularly attended by ants for the sake of the 

 excreted fluid," etc. In 1878 he completed the observation and 

 noted (Can. Ent., x., pp. 131-134) the operation in detail: "The ants 

 run over the body, caressing the larva with their antennae, plainly with 

 the object of persuading it to emit a drop of the fluid from segment 11 (7th 

 abdominal). Most of this caressing is done about the posterior segments, 

 and, while the ants are so employed, or rather while they are absent 

 from the last segments, the tubes of segment 12 (8th abdominal) are 

 almost certainly expanded to their full extent and so remain, with no 

 retracting or throbbing, until the ants come tumbling along in great 

 excitement, and put either foot or antennae directly on, or close by, 

 the tubes, when these are instantly withdrawn. The ants pay no 

 heed to the tubes, do not put their mouths to them, or to the openings 

 from which they spring, nor do they manipulate that segment. They 

 seek for nothing and expect nothing from it ; but they do at once turn 

 to segment 11, caress the back of that segment, put their mouths to 

 the opening, and exhibit an eager desire and expectancy. By holding 

 a lens steadily on segment 11, a movement at the back of this segment 



