Appendix. \ 39 



also diminished. All these changes are in the direction 

 of TelamonideS) or the winter form. 



That the effect of cold is not simply to precipitate the 

 appearance of the winter form, causing the butterfly to 

 emerge from the chrysalis in the summer in which it be- 

 gan its larval existence instead of the succeeding year, is 

 evident from the fact that the butterflies come forth with 

 the shape of Marcellus, although the markings may be of 

 Telamonides or Walshii. And almost always some of 

 the chrysalides, after having been iced, go over the 

 winter, and then produce Telamonides, as do the hiber- 

 nating pupae in their natural state. The cold appears to 

 have no effect on these individual chrysalides. 2 



With every experiment, however similar the condi- 

 tions seem to be, and are intended to be, there is a 

 difference in results ; and further experiments perhaps 

 many will be required before the cause of this is under- 

 stood. For example, in 1878, the first butterfly emerged 

 on the fourteenth day after removal from ice, the period 

 being exactly what it is (at its longest) in the species in 

 nature. Others emerged at 19 96 days. In 1879, the 

 emergence began on the ninth day, and by the twelfth day 

 all had come out, except three belated individuals, which 

 came out at twenty, forty, and fifty-four days. In the 

 last experiment, either the cold had not fully suspended 

 the changes which the insect undergoes in the chrysalis, 

 or its action was to hasten them after the chrysalides were 

 taken from the ice. In the first experiment, apparently 

 the changes were absolutely suspended as long as the 

 cold remained. 



It might be expected that the application of heat 

 to the hibernating chrysalides would precipitate the 

 appearance of the summer form, or change the mark- 



2 This is a striking illustration pf the diversity of individual 

 constitution so frequently insisted on by Dr. Weismann in the 

 foregoing portion of this work. 



