Abstract The hypothesis that phenotypic plasticity is adaptive has rarely been tested explicitly. To support this hypothesis, it is necessary to demonstrate that the phenotype induced in each environment experienced by an organism has a high relative fitness in that environment. In plants, phytochrome-mediated responses to the reduced ratio of red: far-red light (R:FR) characteristic of dense vegetation have frequently been assumed to be adaptive. Such “shade avoidance” responses include stem elongation, suppression of branching, reallocation of biomass, and accelerated flowering. Phytochrome-mediated responses to neighbors provide an ideal model system for studying the evolution of adaptive plasticity because the ecological context is clear and the physiological and genetic mechanisms are well studied at the molecular level. Moreover, plant phenotypes are easily manipulated by alteration of the R:FR signal. The shade avoidance hypothesis for adaptive plasticity is supported by studies of mutant and transgenic plants in which normal photomorphogenic responses are disabled, and by field experiments employing phenotypic manipulation in natural populations. Suppression of phytochrome-mediated stem elongation at high density results in decreased fitness, as does constitutive expression of the elongated phenotype at low density. There is also evidence for ecotypic variation in shade avoidance responses. In a common garden, plants from woodland populations of Impatiens capensis are less responsive to R:FR than plants from a nearby open site, suggesting local adaptive differentiation for plastic response to light quality.