In January 1948, morbidity and mortality studies were initiated to determine the prevalence of multiple sclerosis in the United States and Canada and to investigate the influence of geography, climate, race and other factors on the distribution and outcome of cases of the disease. In the course of these studies, an error resulting from the use of certain vague terms in reporting causes of death was discovered in the annually published mortality statistics on multiple sclerosis. The purpose of this report is to show how this defect may significantly affect the results of other reported studies on multiple sclerosis and to illustrate the general need for exercise of greater care in stating the causes of death on death certificates Certificates of death are classified by vital statistics offices according to the International List of Causes of Death.1 Between 1938 and 1948, multiple or disseminated sclerosis was classified in the