mating behavior cannot be excluded at this point. Detailed behavioral experiments are necessary to settle this issue.
Possible selection in the previous range margins
Although the modified phenotypes found in Fukaura were reproduced by physiological treatment, such a revelation of plastic phenotypes cannot solely explain the Fukaura outbreak, because we failed to induce the modifications sufficiently in the temperature simulation experiment. In fact, most modified individuals were found in September and October when temperature is high in Fukaura, indicating that the modified phenotypes induced in winter should be genetically fixed to achieve the outbreak in summer. One possibility is that natural selection promoted cold resistance that increases survival in the previous northern range margins, and this physiological trait was linked or associated with the color-pattern changes as a side effect. The modified forms may be naturally selected against in the previous range margins if disadvantageous behaviorally, but this selection may be weakened by habitat changes or a founder effect in Fukaura, leading to the outbreak of the modified forms. Similar evolutionary history has already been postulated in speciation of Vanessa butterflies [25-28]. This is an important point to understand what happened in Fukaura. Therefore, the plausible association between cold resistance and color-pattern changes should be experimentally demonstrated in the future.
It is important to note that the outbreak occurred at the third year of arrival at Fukaura, probably within about ten generations. We performed artificial selection for the outward phenotype for ten generations, and yet we were unable to obtain the severe outward type despite obtaining the outward type with relatively small degree of modifications. More generations may be necessary to achieve the severe outward individuals that are comparable to the Fukaura individuals. The rapid increase of the modified individuals from 2000 to 2002 immediately after the colonization of Fukaura cannot be achieved without a certain degree of previous genetic changes.
Possible founder effect behind the Fukaura outbreak
In summer, it is generally hot in Japan including the range margin area, so that butterflies can fly north. But in winter, most of their offspring cannot withstand cold temperature. This cycle may select for cold-tolerant individuals at range margins. We think that in Japan Z. maha has expanded its range to the north in this way. Thus, in the populations of the previous range margins (i.e., Hachiryu, Akita, Honjo, and Kisakata, but not Fukaura), the cold resistant characters had been genetically accumulated already at least