AGRIADES THETIS. 361 



ing plants at the end of June, 1909, were full-fed (at Reigate) by 

 August 15th and brought indoors, and that by the 23rd all those found 

 had pupated. Four other larvae, from eggs laid at the same time, 

 were taken (July 6th, 1909) to Switzerland and carried about (to 

 Evolene, Arolla, Trelechamp, etc.), and with them a small larva, 

 several days older (brought in on food gathered for the hatched larvae); 

 this larva pupated a day or two only ahead of those left at Reigate, but 

 the four younger larvae were, by August 23rd, only about 4mm. -5mm. 

 long ; the earliest of these pupated on September 22nd, and three had 

 not pupated by October 3rd, although now apparently well-grown in 

 the last instar. They finally pupated on October 6th, 8th, and 16th 

 respectively. Buckler says that those that hatch in September feed 

 exceedingly slowly, stretched out along the midrib of a leaflet on sunny 

 days, and are only l*5mm. in length in December, continuing to feed, 

 however, so that, in January, they are 2mm. long, make larger blotches 

 in the leaves, and also begin to nibble their edges ; by the middle of 

 March they have increased in length and still more markedly in stout- 

 ness and, by this time, eat the leaflets from the side, so that, by the 

 end of the month, they are fully 3mm. long, and eat a whole leaflet at 

 a meal; through April, they grow rapidly, especially in warm weather, 

 and eat voraciously, the most forward being full-fed by the end of the 

 month, and the stragglers in May. Chapman says that he has not 

 succeeded in determining the number of larval instars by watching 

 an individual larva, several of which were isolated for the purpose, 

 owing to the close resemblance of the living larva in its successive 

 stages ; in several cases a scrap of cast skin was found showing that a 

 moult had taken place, and the rest of the skin had been eaten ; in 

 other cases, where no trace of cast skin was found, it was felt to be 

 tolerably certain that a moult had taken place, the larva having been 

 apparently resting with that object, and making a total growth unlikely 

 to occur during one instar; by the bye, a very large proportion of the 

 total growth occurs in the last instar. The third stage appears to be the 

 hybernating one, although larvae in the second stage are noted as late 

 as November 14th, 1904, whilst others, at the same time, were in the 

 third stage, and, in addition, a larva was noted in early February as 

 apparently only just moulted into third instar, so that one might 

 suppose that the second was the real stage of lethargy, that the hyber- 

 nation is (as indicated by Buckler) (supra) incomplete, and that the change 

 from the second to the third instar may take place during the occurrence 

 of a few warm days any time very late in the year or in the very early part 

 of the next year. Chapman further notes, however, that he has cast skins 

 of moult from the third to the fourth instar as early as March 5th, and a 

 third instar larva preserved on March 19th, and one of fourth instar 

 of same date, so that there is considerable variation in the dates of 

 reaching and of leaving the 3rd instar. The diameters of the head of 

 the larva at each instar forms the best index and proof of the instar in 

 which any particular larva is — first instar — 0'25mm. ; second instar — 

 0-36mm.; third instar — 0'50mm.; fourth instar — 0"74mm.; fifth instar 

 — l-2mm. We note the almost full-grown larva (September 21st, 1909) 

 as being exceedingly sluggish, and, when resting among its foodplant, 

 usually under one or two leaflets, it is somewhat difficult to detect, the 

 ground colour of the body, the yellow markings, etc., comparing very 

 effectively with the sheen of the light on the edge of the leaves, 



