The reason for the increased incidence and severity of infections in diabetic patients has never been satisfactorily explained. This investigation was designed to study the circulating antibody response in alloxan diabetic, insulin treated diabetic and normal CF-1 mice injected with bovine serum albumin. Only those animals treated with alloxan who had elevated serum glucose levels (250 mg./100 ml. or higher) were included in the study, together with a group of normal animals. Animals were bled from the orbital sinus and the serum analyzed for antigen binding capacity of BSA, glucose concentration and serum proteins. BSA was iodinated with I-131 and the antigen binding capacity of each serum sample was determined as micrograms of BSA nitrogen bound by 1 ml. of undiluted serum. Our studies demonstrate that there is no significant difference in antibody response of alloxan treated mice, alloxan treated mice given insulin and normal mice when immunized with BSA under the conditions of the experiments. If the alloxan treated mouse may be considered a laboratory model of diabetes mellitus in man, the results described herein would be consistent with those studies of diabetics which have shown no primary defect in their immune systems.