It is generally accepted that diabetes mellitus is a familial or hereditary disorder of metabolism. In fact, Pincus and White1 have been able statistically to show that diabetes is inherited as a mendelian recessive characteristic. Experimental proof of this fact has been brought forward by Cammidge.2 He was able, by mating strains of mice having a high fasting blood sugar with those having normal blood sugars, to transmit the hyperglycemia to succeeding generations as a recessive quality. Cammidge suggested that in human beings the same blood sugar conditions might occur. Joslin and his co-workers3 have shown that new cases of diabetes mellitus are continually being encountered in the families of diabetic children. They have shown a progressive incidence in new cases in those families from year to year.
Because of the familial tendency of diabetes, a study was started in 1928 with the purpose of detecting early