The prospective association of social relationships and activities reported during a round of interviews and medical examinations in 1967-1969 with mortality over the succeeding nine to 12 years was examined for a cohort of 2754 adult (aged 35-69 years as of 1967-1969) men and women in the Tecumseh Community Health Study. After adjustments for age and a variety of risk factors for mortality, men reporting a higher levels of social relationships and activities in 1967-1969 were significantly less likely to die during the follow-up period. Trends for women were similar, but generally nonsignificant once age and other risk factors were controlled. These results were invariant across age, occupational, and health status groups. No association was observed between mortality and satisfaction with social relationships or activities. How and why social relationships and activities predict mortality are discussed and identified as important foci for future research.