EXPERIMENTS ON INSECTS. 315 



but others of thoni requiring two, three, and even 

 four years.* IMcinockcn k(.'[)t several pupae of the 

 emperor-moth [Sulunita pdvunia) tlnough the winter 

 in a room heated daily by a stove, and others in a cold 

 chamber. Some of both these appeared in iNIarch, 

 and others, though evidently healthy, had not ap- 

 peared in July.t 



It is certain, however, that this is not the natural 

 order of things, even in this species ; for we have 

 reared several broods of the species respecting which 

 the preceding iacts are recorded, without having 

 observed them. In a large brood of the small egger 

 {Eringaster lanestris) i\ve or six of the cocoons did 

 not produce insects, and we consequently anticipated 

 their appearance next year, but we have now kept 

 them live years without any change, and therefore 

 conclude they are dead. J The inference deduced, 

 however, from the facts observed, is very plausible, 

 namely, that it is intended by Providence to preserve 

 the species : for were all the individuals of a brood to 

 appear in (he same season, it might happen to be so 

 ungenial, particularly in the early months of spring, 

 as to destroy them before they could deposit their 

 eggs; whereas, by their appearing in different seasons, 

 some of them have the chance of coming forth in 

 mild weather. § Yet, perhaps, this may be an unne- 

 cessary though ingenious refining upon a final cause : 

 for even in the most ungenial spring weather, there 

 always occur some fine days, and, further, a brood of 

 insects does not all appear on one day, but more usually 

 on many successive days, as may be seen by the experi- 

 ments of Reaumur recorded at the beginning of this 



• * Scriba, Journ. i, vol. iii, p. 222 ; and Haworth, Lepidopt. 

 Brit., vol. i, p. 125. 



t Xaliirf., vol. viii, p. 14:1 t J. R. 



§ Kkby and Spence, vol. iii, p. 267. 



