248 THE ENTOMOLOGIST. 



lias already pointed to G. c-aureum from Japan as an example 

 of this. According to Edwards, in G. inter rogationis the summer 

 brood in Virginia has a muddy yellow under side with a compli- 

 cated pattern of marking, but the autumn and spring butterflies 

 have the brownish red of a dry leaf, and as a principal marking 

 the bent midrib of a leaf. 



So it may be supposed that with our Pieridae with protective 

 under sides, so far as they possess a seasonal dimorphism, they 

 owe it to the adaptation to the finely pinnate leaves, especially 

 to the foliage of various plants (Cruciferse, Umbelliferse) on which 

 the butterfly is wont to rest. It would be interesting to observe 

 one of these species, e.g. Anthocharis belia-ansonia, in its habits 

 of life, in order to unravel the question, whether a greater simi- 

 larity to the resting plant of each season does not produce the 

 variations in the green and white pattern of the under side, 

 which are not altogether insignificant. 



Such cases would only be explained by a double process of 

 selection, which in the first brood rejects those specimens which 

 during the summer guarantee less protection to the resting 

 butterfly ; and in the second, those which guarantee less pro- 

 tection in the autumn or spring. This process of selection must 

 always have been double ; moreover, if a single-brooded species, 

 which already possesses adaptive colouring, is induced by a 

 warmer climate to interpolate a second brood, then in this case 

 the necessity of the adaptation of the second brood of butterflies 

 to the altered environment of the later summer will first, indeed, 

 have chosen out only the individuals of this second brood itself, 

 but very soon the hereditary transmission of the newly assumed 

 characters of the second brood will have made itself felt on the 

 first brood, and rendered necessary a rejection of the individuals 

 possessing them. A double adaptation can only be realised and 

 retained by continual sifting of both broods, and, as it appears 

 to me at least, this has only become possible under the further 

 hypothesis that several ids* (anlagen) of the wings and of the 

 whole insect are contained in the germ-plasm of each individual, 

 from which some can be reared in this direction, and others in 

 that, while at the same time they are adapted to different 

 excitants which give them scope for action, — heat, cold, &c. 



In cases of pronounced double adaptation of a protective 

 nature, we may safely infer such an origin of seasonal dimor- 

 phism ; but with butterflies seasonal dimorphism is not confined 

 to the under side, but accompanying these differences, which are 

 certainly protective, there are also some on the upper side. With 

 Vanessa levana-prorsa this is the other way about, i.e. the upper 

 side is more strongly variable than the protective under side. 

 On the latter there is in levana a large washed-out lilac spot, 



'■= Cf. note on p. 205, where ' anlagen ' is used as synonymous with ' deter- 

 minant.' — Translator. 



