448 BKITISH BUTTERFLIES. 



ments are very ill-defined. Subsegmental divisions are not visible 

 unless indicated by a line, sometimes noticeable on the subdorsal area, 

 which divides the segment into two portions. In certain lights this 

 larva has a most singular aspect ; it appears exactly as though an 

 ordinary-shaped cylindrical larva, with two pale oblique stripes in the 

 subdorsal area of the well-marked abdominal segments, has crawled 

 into a translucent velvety envelope, the envelope being more visible 

 at its lateral expansion, where the larva did not quite fill it out. 

 Ventral view : It is only when the larva is turned over on its back 

 that the head and legs can be properly seen. The first thing to strike 

 the eye is the leaden colour of the venter itself, as seen between the 

 legs and the claspers. The head is exceedingly small, being only about 

 2mm. wide, and almost entirely enveloped by the prothorax, which 

 surrounds it like a very much inflated pneumatic tyre. The head is 

 notched on the crown, but not deeply, the clypeus distinct, and the 

 epistoma large. Antennas rather large, ocelli very conspicuous from 

 their blackness, the rest of the head being pearly-grey, the epistoma 

 is, however, browner grey and the jaws light brown. The head is 

 smooth except a few hairs below the mouth. Below the head is a 

 pale ochreous spot, possibly an indication of the chin-gland. The 

 skin, where the head joins the body, is blue-green. The part of the 

 prothorax in which the head is sunk has no long hairs, but is covered 

 with a short pile. Thoracic legs very short and set up on large 

 cushions. The 1st and 2nd abdominal segments have anteriorly two 

 deep pits, one on either side of the medioventral line, and posteriorly 

 two deep transverse furrows. The ventral claspers are also short, and 

 placed on cushions, bat these are not so large as those of the thoracic 

 legs. On the 7th abdominal segment are two pits posteriorly, and a 

 dark triangle on the 8th abdominal ; the 9th is divided by a dark 

 medioventral band with a fine white line of muscle on either side 

 anteriorly, the 10th carries the anal claspers, which, as before stated, 

 are quite hidden by the heavy dorsal flap (Sich). 



Foodplants. — Rum ex hydrolapathum, R. obtusifolius (Sich), R. 

 sanguineus, R. aquaticus (Gillmer), ? Rumex crispus (Nicholson), [Rumex 

 acetosa (Wocke), Polygonum bistorta (Heyne).] Some lame, in the last 

 stadium, came to hand on June 9th, 1906. These fed up fairly well 

 on Rumex hydrolapathum and R. obtusifolius. I certainly believe they 

 preferred the former, but would eat the latter when the former was 

 stale. They would not touch the leaves of Rumex acetosa, with which 

 I also sometimes supplied them (Sich). 



Pupation. — The first of the larvae noticed above spun up for 

 pupation June 16th, 1906. A day previously I had noticed that it was 

 no longer feeding, and I placed it in a cage with earth, stem and 

 leaf of dock, and a piece of pasteboard bent at right-angles and 

 turned over at the top. The whole of it was covered with muslin ; 

 after wandering about for some time, the larva finally settled down on 

 the muslin cover, and spun its platform and girdle. Here it 

 rested till the afternoon of June 18th, when, by 6.30 p.m., it had 

 changed to a pupa. After spinning up, the larva loses some of its 

 flatness, becoming more cylindrical, but, with the exception of turning 

 slightly darker, it does not change from its lively green colour. This 

 seems to suggest that, in freedom, pupation usually takes place on the 

 leaves of the foodplant. The girdle is composed of about six strands, 



