Since the introduction of insulin and the work done at the University of Michigan Hospital in 1922, great interest has been taken in the effect of infection on the dosage of insulin. It was soon learned that the occurrence of measles, chickenpox, influenza or any infection in a diabetic patient, particularly in a child, made it necessary to triple or quadruple the dose of insulin. Cowie and Parsons,1 in April, 1923, pointed out the effect of infections on the action of insulin. They also suggested that the efficiency of insulin is decreased in the presence of a hyperglycemia. The effect produced by bacteria and their toxins on insulin is not an unexplored field at present, but it is the purpose of this paper to record data and facts regarding this subject found in experimental studies in this laboratory. As many phases of this problem could not be solved satisfactorily,