244 THE ENTOMOLOGIST. 



as the summer form ; the 34 which hybernated may, on the 

 other hand, have been predisposed to delayed development 

 (Latenz), and to the winter form. 



Thus also a varying percentage of the second brood of the 

 year may be induced by heat to assume the summer form, and 

 to develop immediately ; some will emerge in this form, even at 

 the usual intermediate autumn temperature, but the greater 

 number will incline to latency and to the winter form from the 

 beginning in so strong a way, that they are no longer to be 

 altered from it ; but more accurate proof is still wanted on this 

 point. 



In my work of 1874, on seasonal dimorphism, I believed that 

 I could decide, from the metabolism of a species in its different 

 broods, which of the seasonal forms was the older and which the 

 younger. From the observation, which is generally correct, that 

 the brood of the summer generation in its principal part cannot 

 be diverted from the levana-foYm by temperature, I then con- 

 cluded, that this is the primary and that ijrorsa is the secondary 

 form, since I regarded the artificial alteration as reversion to 

 the parent form. Such, so I thought, could arise only with the 

 younger and not with the older form. On the same ground, 

 with Pieris napi^ I considered the winter form the older. Although 

 I still have the same view as to the relative age of these forms, 

 yet I should now no longer rely on this conclusion, as, quite 

 apart from the fact that with V. levana some specimens of the 

 third brood emerge as the prorsa-ioxm even in the late autumn, 

 with a more mature insight into the processes of heredity, I 

 cannot to-day a7iy longer generally employ the conceptiori of 

 reversion in reference to seasonal dimorphism^ as was explained 

 above. Not* that I consider it exactly incorrect to speak of 

 reversion here, as a phyletically older form actually arises in 

 this case ; but it seems to me more convenient and ^ more 

 suitable for explaining the processes to limit the conception of 

 reversion to those cases of recrudescence of an earlier living 

 form, which do not follow in a regular cycle, that is to say, in a 

 normal way. According to my proposition every reversion 

 depends on the fact that a certain number of unaltered ancestral 

 determinants are transmitted in the germ-plasm of a species, 

 which under especially favourable circumstances (see *Keim- 

 plasma,' p. 392) occasionally become active, and can impress the 

 resulting individual with ancestral characters. With adaptive 

 seasonal dimorphism such an idea of ancestral ids in the germ- 

 plasm is to be assumed, but not as a small vestige, rather in the 

 same number as the modern ids of the other seasonal form. Its 

 germ-plasm is conceived of as consisting of an equal number of 

 winter ids and of summer ids, which usually alternate with each 

 other in the course of ontogeny. This is not quite the same as 

 the exceptional occurrence here and there of reversion to an 



