84 INTRODUCTION. 



caterpillar enters upon that state in July, produces 

 the butterfly in thirteen days ; but when the chry- 

 salis is formed in the end of autumn, the perfect in- 

 sect is not evolved till the succeeding June. Suck 

 variations were conjectured by Reaumur to depend 

 on the temperature to which the chrysalides are ex- 

 posed ; and he proved this to be the fact by a series of 

 very simple and conclusive experiments. By placing 

 a variety of chrysalides in an atmosphere artificially 

 heated, he succeeded in bringing out several broods 

 of butterflies in the very middle of winter, which, if 

 left to natural influences, would not have appeared 

 till the ensuing summer. He found that when the 

 temperature was rather high, the chrysalides made 

 as much progress to maturity in five or six days as 

 they would have done in ordinary circumstances in 

 an equal number of weeks. Having thus proved 

 the influence of heat in hastening the exclusion of 

 these insects, he next tried the effect of cold in re- 

 tarding it ; and the result was equally satisfactory. 

 He preserved several pupae from heat, by keeping 

 them during summer in an icehouse, in consequence 

 of which the butterflies were not disclosed till a year 

 after their ordinary and natural time. * 



When the butterfly is fully matured, it extricates 

 itself from the puparium, by bursting that portion of 

 it which covers the thorax, an operation which is ea- 

 sily accomplished, as the membrane has by that time 

 become weak and friable. On its first exclusion, it, 

 • Reaumur, ii. 10. 



