The hypothesis is advanced that blindness from an early age may lead to a reduced risk of breast cancer through altered patterns of melatonin secretion by the pineal gland. The available experimental evidence in animals and in vitro is consistent with this hypothesis. The hypothesis can be tested in humans by a simple observational study in which the breast cancer risk in blind women is compared with that of all women. The effect of age at onset, duration and degree of blindness could also be assessed, after adjustment for known risk factors for breast cancer. Melatonin might prove to be a natural oncostatic agent of practical value in cancer prevention.