Thermal plasticity in protective wing pigmentation is modulated by genotype and food availability in an insect model of seasonal polyphenism

Abstract Phenotypic variation in natural populations results from complex interactions between organisms and their changing environments. The environment shapes both phenotypic frequencies (during adaptation) and organismal phenotypes (through phenotypic plasticity). Developmental plasticity, in particular, refers to the phenomenon whereby an organism's phenotype depends on the environmental conditions during development. It can match phenotype to ecological conditions and help organisms to cope with environmental heterogeneity, including differences between alternating seasons. Experimental studies of developmental plasticity often focus on the impact of individual environmental cues and do not take explicit account of genetic variation. In contrast, natural environments are complex, comprising multiple variables with combined effects that are poorly understood and may vary among genotypes. We investigated the effects of multifactorial environments on the development of the seasonally plastic eyespots of Bicyclus anynana butterflies. Eyespot size depends on developmental temperature and is involved in alternative seasonal strategies for predator avoidance. In nature, both temperature and food availability undergo seasonal fluctuations. However, our understanding of how thermal plasticity in eyespot size varies in response to food availability and across genotypes remains limited. To address this, we investigated the combined effects of temperature ( T ; two levels: 20°C and 27°C) and food availability ( N ; two levels: control and limited) during development. We examined their impact on wing and eyespot size in adult males and females from multiple genotypes ( G ; 28 families). We found evidence of thermal and nutritional plasticity and temperature‐by‐nutrition interactions (significant T × N ) on the size of eyespots in both sexes. Food limitation resulted in relatively smaller eyespots and tempered the effects of temperature. Additionally, we found differences among families for thermal plasticity (significant G × T effects), but not for nutritional plasticity (non‐significant G × N effects) nor for the combined effects of temperature and food limitation (non‐significant G × T × N effects). Our results reveal the context dependence of thermal plasticity, with the slope of thermal reaction norms varying across genotypes and across nutritional environments. We discuss these results in the light of the ecological significance of pigmentation and the value of considering thermal plasticity in studies of the biological impact of climate change. Read the free Plain Language Summary for this article on the Journal blog.