Seasonal generations of short-lived organisms often differ in their morphological, behavioural and life history traits, including body size. These differences may be either due to immediate effects of seasonally variable environment on organisms (responsive plasticity) or rely on presumably adaptive responses of organisms to cues signalizing forthcoming seasonal changes (anticipatory plasticity). When directly developing individuals of insects are larger than their overwintering conspecifics, the betweengeneration differences are typically ascribed to responsive plasticity in larval growth. We tested this hypothesis using the papilionid butterly Iphiclides podalirius as a model species. In laboratory experiments, we demonstrated that seasonal differences in food quality could not explain the observed size difference. Similarly, the size differences are not likely to be explained by the immediate effects of ambient temperature and photoperiod on larval growth. The qualitative pattern of natural size differences between the directly developing and diapausing butterflies could be reproduced in the laboratory as a response to photoperiod, indicating anticipatory character of the response. Directly developing and diapausing individuals followed an identical growth trajectory until the end of the last larval instar, with size differences appearing just a few days before pupation. Taken together, various lines of evidence suggest that between-generation size differences in I. podalirius are not caused by immediate effects of environmental factors on larval growth. Instead, these differences rather represent anticipatory plasticity and are thus likely to have an adaptive explanation. It remains currently unclear, whether the seasonal differences in adult size per se are adaptive, or if they constitute co-product of processes related to the diapause. Our study shows that it may be feasible to distinguish between different types of plasticity on the basis of empirical data even if fitness cannot be directly measured, and contributes to the emerging view about the predominantly adaptive nature of seasonal polyphenisms in insects.