1. Twelve years' experience with 539 cases of erythroblastosis fetalis has been reviewed. 2. It is not possible to show beyond the statistical limits of chance that any change in mortality of erythroblastosis fetalis in relation to therapy in live born infants has occurred in this time. 3. It is probable that the early induction of labor has caused an increased risk to live born babies during part of this period, when erythroblastosis fetalis has been complicated by immaturity, particularly before 38 weeks' gestation. 4. Caution is urged in assaying the value of any therapeutic measure, the use of which in erythroblastosis fetalis may result in a selected body of clinical material.