The importance of rheumatic fever and its after-effects as a cause of mortality in childhood has been stressed more and more in discussions from all parts of the world, particularly the United States. A statistical survey has recently been issued by the Children's Bureau 1 to shed more light on the differences in mortality from rheumatic fever and heart diseases in the United States, its sections and the individual states. In order to diminish chance fluctuations, which necessarily become large in individual states when the total is split into the respective age, race and sex groups, three year averages of the mortality data, 1939-1941, have been taken and centered on the population of the census of 1940. Exact data on population in the individual states by age, race and sex are available only for the years of census enumerations. However, for the United States as a whole additional data are