During the past 50 years, an abundant literature has been directed to various aspects of autoimmune hemolytic disease. Recognition and measurement of changes in erythrocyte life span,1 improved methods for demonstration of erythrocyte autoantibody,2 and advances in knowledge of bile pigment metabolism 3 have facilitated diagnosis and understanding of this disorder. Fortunately, most clinical and hematologic manifestations of autoimmune hemolytic disease are correctible by treatment with the corticosteroids,4,5 even though evidences of autosensitization of erythrocytes may persist. On the other hand, the existence of inordinately severe cases which prove resistant to therapy have provided further opportunity to observe the natural course of the disease. We have studied a patient in whom continuous production of autoantibody in high titer resulted in constant anemia, reticulocytosis, and icterus, despite intensive treatment with corticosteroids. These exaggeratedly abnormal laboratory values lent themselves particularly well to a study of the effects of "autosensitization"