The detection of an insulin antagonist in the serum of insulin resistant subjects has been attempted on a number of occasions. In certain instances, the excessive quantities of insulin used in the neutralization experiments in animals may explain the failure to obtain evidence of an insulin antagonist. In the few successful experiments which have been done, little information of a quantitative nature has been obtained. Glen and Eaton (1) and Marble, Fernald and Smith (2) were able to demonstrate decreased responsiveness to insulin in rabbits injected with serum from insulin resistant patients. This change in the experimental animal was still demonstrable a week after injection of the serum. In the same year, Banting et al. (3) described experiments in which the incidence of hypoglycemic convulsions was reduced in mice injected with mixtures of serum and varying amounts of insulin. The serum was obtained from a patient with schizophrenia who developed insulin resistance during insulin shock therapy. This case is especially interesting because diabetes was not present, suggesting that the resistance was directed against prepared insulin only. When the same serum was retested a week later in mice maintained on a high carbohydrate diet, practically no insulin neutralization could be demonstrated. Lerman (4), in studying serum of 5 patients with insulin resistance, obtained evidence for the presence of an insulin antagonist in 2. One of these serums was provided by the writer, and was obtained from a diabetic patient, A. M., the subject of this and previous reports (5,6). A high degree of allergy to insulin, coupled with marked insulin resistance, combined to make the giving of insulin to this patient both disagreeable and ineffective. A further important feature was the presence of a relatively mild degree of diabetes, making possible prolonged periods during which insulin therapy could be omitted without immediate danger to the patient's life. A number of hospital admissions provided opportunity to show that: a. At the end of a period of 5 months or more without insulin therapy, small doses of insulin would give rise to urticaria, constriction in the chest, and, on one occasion, collapse. b. After desensitization with graded doses of insulin, a period of 3 to 5 days followed during which the injection of relatively small doses of insulin would be followed by a fall in blood sugar. c. Within about 11 to 15 days of starting desensitization and treatment with insulin, a solid resistance to insulin developed. These changes were correlated with the absence, and then the appearance, of insulin-neutralizing activity in the patient's serum. The allergic reactions appeared to be mediated by a mechanism quite independent of that giving rise to resistance (5).The method previously used for the demonstration of insulin-neutralizing activity of the patient's serum (6) consisted in observing the incidence of hypoglycemic symptoms in mice injected with a mixture of serum and insulin. This method had a number of disadvantages whic...