A critical review of the literature of retrolental fibroplasia indicates that the cause of this disease is not yet known. Oxygen is certainly a critical factor but it is still not possible to make precise recommendations as to the amount or the duration of therapy that is safe. We have overemphasized the role of oxygen in the past, and as a result of this the false impression has been created that RLF is a disease that can be prevented. This gross oversimplification of a complex disease with multiple causes has resulted in many unjustified malpractice claims. A study of the present epidemic indicates that excessive oxygen administration probably plays a minor role, in contrast to the first epidemic in which prolonged oxygen administration was clearly a major factor. A reasonable working hypothesis is that the developing retina is highly sensitive to any disturbance in its oxygen supply, either hyperoxemic or hypoxemic. The retinal circulation is subject to the same wide fluctuations as the cerebral circulation in newborn infants. The very low-birth-weight, sick premature infant suffers from a number of conditions, many of which can seriously disturb the retinal circulation, resulting in hypoperfusion and ischemia. These factors (immaturity, hyperoxia, hypoxia, blood transfusions, intraventricular hemorrhage, apnea, infection, hypercarbia, hypocarbia, patent ductus arteriosus, prostaglandin synthetase inhibitors, vitamin E deficiency, lactic acidosis, prenatal complications, genetic factors) may all be present in an infant. They may interact to produce various degrees of retinal damage. Nearly all of these factors cannot be prevented or controlled by our present methods of care. Unfortunately, this means that RLF is an extremely difficult disease to prevent, treat, or investigate. A disease of this complexity with multiple causes will require very large numbers of infants in any controlled study of a therapy. Retrolental fibroplasia should not be considered an avoidable iatrogenic disease in very low-birth-weight infants. Its cause in these infants is not known.