Tea has consistently been shown to inhibit the occurrence of tumors in experimental animals. The evidence for such a beneficial effect in humans, however, is limited. The authors examined the association between non-herbal tea consumption and cancer incidence in a prospective cohort study of 35, 369 postmenopausal Iowa women. In this cohort, information on the frequency of tea drinking and other dietary and lifestyle factors was collected by mailed survey in 1986. After 8 years of follow-up, 2,936 incident non-skin cancer cases were ascertained in this cohort through the State Health Registry of Iowa. Proportional hazards regressions were used to derive adjusted relative risks and 95% confidence intervals for the association between tea consumption and cancer incidence. After controlling for confounding factors, the authors found that regular tea consumption was related to a slight, but not statistically significant, reduced incidence of all cancers combined. Inverse associations with increasing frequency of tea drinking were seen for cancers of the digestive tract (p for trend, 0.04) and the urinary tract (p for trend, 0.02). For women who reported drinking >2 cups (474 ml) of tea per day, compared with those who never or occasionally drank tea, the relative risk for digestive tract cancers was 0.68 (95% confidence interval (Cl) 0.47–0.98) and for urinary tract cancers, 0.40 (95% Cl 0.16–0.98). Similar inverse associations were seen for specific digestive and urinary tract cancers, although site-specific analyses were not statistically significant. No appreciable association of tea drinking was found with melanoma, non-Hodgkin's lymphoma, or cancers of the pancreas, lung, breast, uterine corpus, or ovary. This study suggests that tea, one of the most popular beverages consumed worldwide, may protect against some cancers in postmenopausal women. Am J Epidemiol 1996; 144: 175–82.