RUMICIA PHL^AS. 385 



undersurface of the leaf, eating- the green cuticle there, which is much 

 thicker than that on the uppersurface of the leaf. The third (or fourth) 

 instar appears to be the critical one. In the summer brood, a larva of 

 6mm. will grow to nearly 16mm. in six days, and feed up to pupating 

 stage in four days longer, whilst in the autumnal brood the larvae of 

 this size slacken off, and become more or less sluggish, preparing for 

 hybernation. Hawes observes (Proc. Stk. Lund. Ent. Soc, 1893, p. 140) 

 that, in nature, he has never found larvae feeding on the upperside of 

 the leaf, and that they appear like those of Hamearis (Nemeobius) lucina 

 on Priumla, to attach themselves to the underside, i.e., on the side 

 least exposed to the light and their feathered enemies, where each 

 larva eats out small portions and rests in the cavities so made, from 

 time to time wandering to the other parts of the same leaf during 

 the first week or so of its existence. At this early stage it is difficult 

 to detect with anything like certainty, and the best plan is to gather 

 the riddled leaves with the eggs and eggshells, and keep them in a dry 

 situation -well surrounded with fresh leaves of the foodplant. Scudder 

 further observes that just before pupation the fungiform appendages 

 of the coming pupa appear as white hemispherical papillae dotting the 

 surface of the caterpillar. The different rate at which larvae from the 

 same batch of eggs feed up is very marked in all the broods. Raynor 

 observes (in litt.) that, from eggs laid in early June, 1905, the fastest 

 larvae commenced to pupate on July 20th, that by July 24th there were 

 eight pupae, and, of the remaining larvae, some were fullfed, and others 

 yet quite small. Sabine notes (Ent., xxxvii., p. 285) that, of a large 

 number of larvae reared from August-laid eggs, in 1904, many had 

 pupated by the end of September, and produced imagines in October, 

 whilst, at that time, the greater part of the remainder, consisting of 

 some hundreds, were only half-fed, or little more; so that one supposes 

 most would hybernate, as is usual, as larvae, and not attempt to feed up 

 and produce imagines the same autumn. Cooper also observes that, from 

 eggs laid August 2nd, 1884, some of the larvae fed up rapidly on a 

 growing plant of sorrel, pupated, the imagines appearing between 

 September 19th and 22nd ; at this time, other larvae belonging to this 

 batch were nearly f ullgrown, and others still quite small ; the former 

 would, of course, attempt to complete their life-cycle in the year, the 

 small ones, no doubt, would hybernate. Chapman notes that, "on 

 September 3rd, 1906, a small brood of larvae, from eggs laid about a 

 fortnight previously, differed to such an extent that one is in the pupal 

 state, two or three are girthed up, six or seven nearly f ullgrown, the 

 remainder (two or three dozen) still small, fond of bunching themselves 

 up almost into little balls about 4mm. long, and 2mm. thick ; they 

 still eat, and for the moment evidently are not thinking of hybernating 

 (the last few days has been 90° in the shade, or thereabouts). By 

 September 12th these small ones, however, had made no progress, and 

 seem now to eat nothing. They show a few white points (trumpet- 

 hairs) absent in larvae of Chrysophanus var. rutilus at hybernating stage." 

 Prideaux states that Rumex acetosella is so hardy a plant that, 

 even in severe winters, it probably affords the most suitable resting- 

 place for the over- wintering larvae of R. phlaeas to rest upon, he 

 adds that the larvae hybernate at different ages and feed in mild 

 weather. The earliest and best placed of the hybernating larvae 

 feed up very rapidly in the spring. Hodgkinson says they are already 



