We propose the hypothesis that a vegetarian diet reduces the risk of developing diabetes. Findings that have generated this hypothesis are from a population of 25,698 adult White Seventh-day Adventists identified in 1960. During 21 years of follow-up, the risk of diabetes as an underlying cause of death in Adventists was approximately one-half the risk for all US Whites. Within the male Adventist population, vegetarians had a substantially lower risk than non-vegetarians of diabetes as an underlying or contributing cause of death. Within both the male and female Adventist populations, the prevalence of self-reported diabetes also was lower in vegetarians than in non-vegetarians. The associations observed between diabetes and meat consumption were apparently not due to confounding by over- or under-weight, other selected dietary factors, or physical activity. All of the associations between meat consumption and diabetes were stronger in males than in females.