12 BRITISH LEPIDOPTERA. 



male larvae of Notoloplius (Orgyia) leueostigma were found by Eiley to. 

 moult three times, the female larvae four times, whilst Dyar notes that 

 in X. gulosa the male larvae moult three or four times, the female 

 larvae always four, and that whilst male X. antiqua larvae have six 

 stadia the female has seven. This latter observation is surprising in 

 the face of our experience in Britain (ride, post.). 



With regard to the variable number of moults in Notoloplius 

 antiqua, Chapman notes (Ent. Mo. Mag., xxiii, p. 224) that he has 

 observed — 3-moulter males, 4-moulter males and females, and 5- 

 moulter females. Bearing on this he makes the following important 

 observations: "The larvae that moult three times always produce 

 male moths, those that moult five times females, those that moult four 

 times produce both." "There is also another circumstance that is 

 usually associated with this fact, i.e., that the male emerges from the 

 pupa a few days before the female. Now, if we consider only the 4- 

 moulters, consisting of both males and females, this is not so, but if 

 we associate them in this order — (1) 3-moulter males, 4-moulter 

 females, (2) 4-moulter males, 5-moulter females, we find that the usual 

 rule of the males emerging first is observed. This also shows more 

 distinctly that another rule obtains in X. antiqua. It is one that 

 obtains among bees, and would, I think, be found in other insects if 

 observation on the point were made. It is this, that the male, though 

 feeding as larva a shorter time (being a smaller insect), remains a 

 longer time in the pupal state than the female, apparently requiring a 

 longer time to undergo its full development. It would thus appear- 

 that in X. antiqua the female moults once more than the male, a cir- 

 cumstance that I have not seen noticed as occurring in any other 

 species, and that further the moults may vary by one." Riley, as early 

 as 1868, had called attention to the fact that the male larva of the Ameri- 

 can Notoloplms leueostigma underwent one moult less than the female 

 larva, the numbers given (loc. cit., p. 274) by him being three for 3 

 and four for 5 . He further states that " there is a very general ten- 

 dency in individuals to vary from the normal number of moults in the 

 species," and adds that "whenever there is much discrepancy in the 

 size of the sexes, the smaller (usually the male) undergoes a less num- 

 ber of moults, and that the variation in the number of larval moults 

 (except where, as in these cases, it is sexual, and presumably predeter- 

 mined in the egg) is dependent on food- supply rather than hyberna- 

 tion." He thinks that " it may be stated as a very general rule that 

 moulting is correlated with rate of growth and nutrition, those species 

 which have a short larval existence, generously nourished, exuviating 

 least." 



Chapman records (Ent. M<>. Mag., xxxii., pp. 54-57) how, on one 

 occasion, he practically starved some larvae of Triphaena comes that had 

 reached their last instar, by giving them a diet of Arbutus unedo. After 

 putting them back on dock, all but two died, but these undertook an 

 extra (post-ultimate) moult and then died. In a second experiment, 

 the larvae were fed up until they had reached the last instar, and then, 

 by a diminished food supply, they were kept almost at a standstill for 

 five or six weeks, when they were placed on an abundant diet. In due 

 course, several changed normally into pupae, others came to grief, 

 whilst a third section made an attempt to moult again ; two actually 

 succeeded in doing so, others did so partially, whilst the remainder 



