Bright continuous light has been implicated in the pathogenesis of retinopathy of prematurity. To investigate the influence of light on the incidence and severity of retinopathy of prematurity, we enrolled 127 preterm infants (birth weight < or = 1,500 g; gestational age < or = 32 weeks) in a controlled clinical study. Randomization was done separately for three birth-weight groups (< 1,000 g; 1,000 to 1,249 g; 1,250 to 1,500 g). The babies' eyes were patched all day and night from birth to a gestational age of 35 weeks. The infants in the control group were exposed to cycled lighting conditions (that is, reduced light level during the night). Of 62 infants with patched eyes, 26 (42%) developed retinopathy of prematurity. In the control group, 25 of 65 infants (39%) showed retinopathy of prematurity (P = .596). There were also no statistically significant differences in the incidences of retinopathy of prematurity in the birth-weight subgroups or in the severity of retinopathy of prematurity. Thus, patching of the eyes from birth to 35 weeks of postconceptional age does not decrease the risk of retinopathy of prematurity in preterm infants when compared to a control group exposed to cycled lighting conditions.