Avian vision models and field experiments determine the survival value of peppered moth camouflage

Animal defensive coloration has long provided many important examples of evolution and adaptation. Of these, industrial melanism in the peppered moth is the classic textbook example of evolution in action, whereby dark and pale morphs suffer differential predation in polluted and unpolluted woodland based on their camouflage. Despite extensive work, a striking gap remains in that no study has ever objectively quantified their camouflage or related this directly to predation risk. Here we use image analysis and avian vision models to show that pale individuals more closely match lichen backgrounds than dark morphs. Artificial predation experiments in unpolluted woodland show 21% higher survival rates of pale than melanic individuals. Overall, we provide the strongest direct evidence to date that peppered moth morph frequencies stem from differential camouflage and avian predation, providing key support for this iconic example of natural selection. Olivia Walton and Martin Stevens revisit the classic example of the peppered moth, objectively quantifying moth camouflage and predation risk. With bird vision models, pale individuals more closely match lichen backgrounds, and survive better, providing support for this iconic example of natural selection.