362 BRITISH BUTTERFLIES. 



although, of course, the leaves are not even tinged with yellow. When 

 resting, the body is somewhat humped, the thoracic segments being 

 bent over, and the head hardly or not at all exposed, the anterior part 

 of the body in this position wider than the posterior, although, when 

 stretched, and moving slowly; the anterior is the narrower. The 

 conspicuous black spiracles just above the yellow section of the sub- 

 spiracular flange (the prothoracic as conspicuous as the others), the 

 dense coating of black hairs, the conspicuousness of the black hair- 

 bases, and the dark mediodorsal line (dark green on the meso- and meta- 

 thorax and 1st abdominal, slaty-blackish beyond to the end of the 

 dorsal groove, and then again dark green to the anus) appear very marked 

 in the larvae under observation ; the projecting green lateral flange, at 

 its most prominent part edged with yellow above (broken slightly at 

 segmental incisions) and continued round the anus and, more faintly, 

 round the prothorax, together with the bright yellow segmental marks 

 on the dorsal ridges of either side from the mesothorax (faint on pro- 

 thorax) to the 6th abdominal, and the median space widening from the 

 back forwards, are also very characteristic and strongly marked. The 

 larval movements are extremely slow, and, if disturbed, the caterpillar 

 falls to the ground, the head completely retracted within the prothorax 

 and the true legs well under the projecting prothoracic hood ; the body 

 is, at this time, slightly curved ventrally at the anal end, and the larva 

 rests, immovably, for a long time on its side or back without any attempt 

 to stir ; after a time, however, it just pushes its head level with the front 

 of the prothoracic hood, and moves it slowly from side to side, then 

 throws back the thoracic segments and turns them so that the head 

 and legs face the ground, and so that they are twisted almost at right 

 angles to the rest of the body. It then slowly stretches the head 

 downwards, until the body is quite twisted, and the true legs touch the 

 ground, after which the abdominal segments are turned over, and the 

 prolegs brought to the ground ; even then the larva moves forward so 

 slowly that it is almost impossible to detect any muscular movement 

 as it progresses. 



Symbiosis of ants and larvae of Agriades thetis (bellargus). — 

 Of seven full-grown larvae of A. thetis (bellargus), found at Folkestone 

 on August 11th and 12th, 1906, two had ants upon them when found, 

 and, in one instance, an ant was observed to be busy stroking the 

 gland with its antennae in the endeavour to excite the flow of the 

 liquid secreted by that organ, with the desirable qualities of which 

 it was evidently familiar. The ants found attendant upon these were 

 workers of Lasius niger, whereas those found on larvae of A. coridon at 

 Reigate, in June, were of a different species, viz., Lasius flavus. Efforts 

 to obtain a demonstration of the function of the honey-gland by means 

 of these black ants, some of which were brought with the larvae from 

 Folkestone, failed, whilst complete success attended the first attempt 

 when an example of L. flavus, taken from a nest introduced some time 

 ago into the garden for the purpose of experimenting with larvae of 

 Lycaena arion, was made the medium. Possibly />. flavus, which is 

 common on the be llargus- affected hill-slopes at Folkestone, is more 

 generally associated with these Lycaenid larvae than are other species of 

 ants, and the instinct for " nursing " more highly-developed in it than is 

 the case with the workers of Lasius niger. That ants are not essential 

 to the well-being of the larvae of this species appears certain, for a 



