SEASONAL DIMORPHISM OF LEPIDOPTERA. 183 



me to prove this. Pupae, which had been kept for ten weeks at 

 4° C* were half of them then brought into a temperature of 

 13° C, in which they emerged after 34-36 days as the pure 

 golden spring form; the other half of the iced pupae were 

 brought into a high temperature (32° C.)) where they emerged in 

 5-6 days, and, indeed, as black powdered forms approachiug the 

 form eleus. The latter experiment, to be sure, was only made 

 with five or six specimens, and in the first about half the pupae 

 died, or emerged crippled ; but yet the result is so definite, that 

 it may well be regarded as conclusive. Nevertheless, I should 

 have liked to have repeated it once more with a larger number of 

 individuals, if I could have obtained material for it. The question 

 of the critical period for the influence of temperature appears to 

 me the more important, since with other species I have found 

 just the opposite, viz. that the beginning of the pupal period 

 determines the form of the dimorphic butterfly, and since it 

 appears to me possible, on theoretic grounds, that this may be 

 different in direct and in adaptive seasonal dimorphism. 



It is conceivable, on direct alteration of climate, that the 

 effective temperature must set to work, when the colours of the 

 wings are beginning to form, as how should their alteration be 

 otherwise brought about, if not by changes of the chemical pro- 

 cesses, which underlie the production of colour? The deter- 

 minants of the scales will therefore be influenced in this way 

 at the moment, in which they become active ; they undergo 

 various small modifications by different temperatures, which lead 

 to an alteration in the course of the colour-chemistry. But 

 whether with many species, which appear to be altered directly 

 by the climate, quite other factors do not co-operate to influence 

 the colour-chemistry, is another question, whose solution is 

 certainly not possible at the present time. I might, indeed, 

 consider it probable from the result of Experiment VIII., in 

 which hybernating pupae of a large number of the most various 

 species of Lepidoptera were kept at 30° C, and thus induced to 

 emerge earlier. None of them exhibit anything special in 

 marking or colouring, although here certainly an increased 

 temperature was operating just at the time when the formation 

 of colour is going on. This points to the fact, that with the 

 various considerable alterations, which Dorfmeister, Merrifield, 

 Standfuss, E. Fischer, and many others have produced by cold 

 or a higher temperature in many butterflies, it is not the 

 chemical processes on the formation of colour itself, but rather 

 the various predispositions to the colour-patterns of their 

 ancestors still contained in the germ-plasm, or later on in the 

 rudiments of the wings of the species in question, which might 



'^ A reference to Merrifield's paper shows that this should be 0*1° C. 

 (Translator). 



