272 BRITISH BUTTERFLIES. 



they would not touch it. When feeding they never ate the flowers of 

 the Lotus even when very hungry, but only the leaves, and the softer 

 parts of the leaf-stalks. The larvae were very inactive and could only 

 be induced to move by touching the claspers with a moistened artist's 

 brush. By this means they were transferred to fresh supplies of food. 

 They generally bit savagely at the brush, but this was the only means 

 of defence they ever adopted. They never everted the chin-gland. 

 When placed on the fresh food ■ they often ate a portion of a leaflet 

 before spinning a new tent, but they always appeared very anxious to 

 get under cover. The tent is made by drawing the three leaflets of a 

 Lotus leaf together and fastening the edges with very strong silken 

 strands. In the fourth and fifth stadia, and, perhaps also in the 

 earlier stadia, the larva spins a silken platform on which it rests. 

 This platform is not always on the floor of the tent, but sometimes 

 on the ceiling. They always undergo ecdysis in the tent, and, as 

 Sich does not remember to have seen a cast skin (though many cast 

 heads), he thinks the larvae must eat the old skin. In the fifth 

 stadium they are too large to be effectually covered by one Lotus 

 leaf, and they therefore make use of the leaflets of two or more leaves, 

 forming a kind of ball of leaflets. It is astonishing how well 

 these caterpillars are hidden. Their soft green colour, as well as 

 their plump shape, adds much to their powers of concealment when 

 living among the Lotus leaves. It is not always easy to see a larva 

 when one knows that it is on a certain piece of the plant, and one fancies 

 that, if it were of the ordinary cylindrical shape, it would be much more 

 easily seen among the obovate leaflets of the Lotus. The larvae rest in 

 the tent with the head curved round to one side (either side), and, if 

 disturbed, they tuck the head against the 3rd abdominal segment. In 

 feeding, they stretch out of the tent only so far as is necessary to reach 

 the leaves around. They keep the tent beautifully clean, shooting 

 their excrement far away. One larva, noticed doing this, shot the 

 pellet on to a window-pane thirteen inches away with such force that 

 the pellet rebounded some inches. The pellets always seemed to be dry 

 when excreted, and are caught on the teeth of the anal comb preparatory 

 to being shot off. On July 28rd, all but two larva? were in the fourth 

 instar, and three days later the most forward specimen had entered the 

 fifth and final larval stadium, which most of them reached by August 

 1st. Two of them noticed on August 5th, had grown very large and 

 very stout, but were still quite green in colour. Ten days later the 

 larvae began to lose their green tint, getting brownish, and the spiracles 

 became very dark. At the same time they began to shrink in bulk, 

 becoming more wrinkled and only ate very sparingly. A few days later 

 the larvae were returned to Dr. Chapman (Sich). A larva kept by Buckler, 

 left its hybernaculum at this time to make another retreat and then 

 died ; but it does not appear to be usual for the larva to leave the plant 

 to spin a puparium, the hybernaculum generally serving for a puparium 

 in nature. Chapman observes that, for hybernation, the larva likes to go 

 to some distance from its foodplant, and selects a fairly dry place, prob- 

 ably amongst dead leaves, dry grass, or, where available, under stones. 

 Of the larva 1 under observation, some spun together dead leaves, others 

 muslin, some paper, and one or two that escaped were found in odd corners, 

 generally disturbed in discovery, but one had found a bit of dead leaf, 

 and a bit of paper and drew them together. In all cases they appear to 



