246 THE ENTOMOLOGIST. 



assume the summer form at the ordinary temperature or even at 

 an mcreased temperature, if they belong to the second or the 

 third brood. 



The fact that the summer ids may be induced to become 

 active by a high temperature renders it possible for a numerous 

 second brood of the summer form to occur in a very hot summer, 

 as this, according to my old experiment already referred to, must 

 be the case with V. j^rorsa, it probably is also the case with 

 P. napi. 



But, on the other hand, the fact that the winter ids may be 

 induced to become active by cold provides the possibility that, if 

 cold sets in early in the autumn, the third brood of the year 

 should mostly hybernate, and then produce the spring form in 

 the spring. 



In spite of all these precautionary rules every specimen of a 

 brood does not agree with the time of year, but the exceptions 

 are rare ; I do not know any case where a prorsa has been found 

 in the woods in the spring or a levana in the summer ; still 

 intermediate forms {porima) occur in the summer, and I have 

 already sought to explain them above. With Pleris napi a 

 winter form or an intermediate form appears to have formerly 

 occurred in the summer, but the differences here of the two 

 forms are not so great or so sharp. 



Theoretically the mixed forms can be easily understood from 

 our standpoint : they depend on the fact that both determinants 

 become active at the same time, and that consequently each of 

 them seeks to take part in the colouring of the wings, so that 

 sometimes the one and sometimes the other may preponderate. 

 The same process arises, which according to my proposition 

 takes place on the intermixing of the parents' characteristics in 

 the child,* whereby, indeed, the one parent can make itself felt 

 almost alone, and the other be entirely left out, but all con- 

 ceivable combinations and intermixings of the characteristics of 

 the parents can arise. In this case we know for certain that 

 two separate predispositions come together on fertilization, and 

 that consequently they are both together in the egg which is 

 developing, while with adaptive seasonal dimorphism we only 

 infer this from the phenomena of the change of garb. 



If I now deny the correctness of my earlier conclusion as 

 to the relative age of the alternating seasonal forms, I do not 

 do so with non-cyclic species, as Pieris napi var. bryonies. It 

 it is true, as it appears to be from my earlier experiments, that 

 bryonies cannot be induced by heat to assume the napi-iovrnj the 

 conclusion would be probable, though not necessary, that bryonies 

 is the parent form of iiapi. Now this is not a case of actual 

 reversion, i. e. of that sort which occurs exceptionally , which could 



* Vid. in *' Germ-plasm" the section on " The Struggle of the Ids in Onto- 

 geny" (p. 260, Eng. trans.). 



