Tea and tea compounds have been shown to inhibit carcinogenic processes in experimental animals, raising the possibility that tea drinking may lower cancer risk in humans. However, epidemiologic studies have produced inconsistent evidence on the relation between tea drinking and cancer risk. Ecological data show considerable international variation in tea consumption but relatively small differences in cancer rates. Results from case-control and cohort studies also are inconclusive. Nevertheless, high consumption of tea has been linked to a reduced risk of digestive tract cancers in a number of epidemiologic studies. A lack of detailed information on duration and amount of tea drinking, a narrow range of tea intake in some study populations, inadequate control for confounding, and potential biases in recall and reporting of tea drinking patterns in case-control studies may have contributed to the diverse findings. Further research is needed before definitive conclusions on tea's impact upon cancer risk in humans can be reached.