Plasticity is the ability of a genotype to produce multiple phenotypes in response to environmental stimuli. Adaptive plasticity is expected to evolve when different phenotypes are optimal in different environments. In order to understand the evolution of plasticity we need to understand costs and benefits of trait expression across environments. In Hyles lineata, the white-lined Sphinx moth, melanin pigmentation is a plastic trait. Maximal melanization is induced in response to seasonal cues, including low temperatures and short photoperiods. Melanin is expected to be beneficial in cold environments due to its ability to absorb radiant heat. In this study we test the hypothesis that melanin pigmentation plasticity is an adaptive response to temperature variation. We induced melanic versus non-melanic larvae of H. lineata and raised them in cold (21°C) versus warm (33°C) environments under a source of radiant heat. We measured growth rate and development time in the 5th instar. We found that in cold environments melanic larvae grow significantly faster than non-melanic larvae and have a longer development period, resulting in larger sizes. Furthermore, mortality was higher for non-melanic morphs in the cold environment. In the warm environment growth rates, development periods, and mortality of the two morphs were similar. These results suggest that melanic individuals have an advantage over non-melanic individuals in cold environments but not warm environments, consistent with the hypothesis that melanin plasticity is an adaptation to temperature variation. This study contributes to our understanding of the evolution of phenotypic plasticity, a major goal in this field.