Great interest has been manifested in various problems bearing on the etiology of diabetes mellitus, especially during the last few years. This is due largely to the menace of the disease as revealed by the increase in its incidence as well as in its mortality rate, at least until the time of the introduction of insulin.1
The observations of Emerson and Larimore2 have definitely proved that there has been a progressive and actual increase in the death rate from diabetes in New York, as well as in other larger cities, between the years 1860 and 1923. They direct attention to the fact that an increase in the death rate from a certain disease may be considered as pointing to an actual increase in its incidence, especially during a period in which clinical diagnosis as well as treatment have progressed to such an extent as to make possible early recognition of