Foods and beverages rich in phenolic compounds, especially flavonoids, have often been associated with decreased risk of developing several diseases. However, it remains unclear whether this protective effect is attributable to the phenols or to other agents in the diet. Alleged health-promoting effects of flavonoids are usually attributed to their powerful antioxidant activities, but evidence for in vivo antioxidant effects of flavonoids is confusing and equivocal. This may be because maximal plasma concentrations, even after extensive flavonoid intake, may be low (insufficient to exert significant systemic antioxidant effects) and because flavonoid metabolites tend to have decreased antioxidant activity. Reports of substantial increases in plasma total antioxidant activity after flavonoid intake must be interpreted with caution; findings may be attributable to changes in urate concentrations. However, phenols might exert direct effects within the gastrointestinal tract, because of the high concentrations present. These effects could include binding of prooxidant iron, scavenging of reactive nitrogen, chlorine, and oxygen species, and perhaps inhibition of cyclooxygenases and lipoxygenases. Our measurements of flavonoids and other phenols in human fecal water are consistent with this concept. We argue that tocopherols and tocotrienols may also exert direct beneficial effects in the gastrointestinal tract and that their return to the gastrointestinal tract by the liver through the bile may be physiologically advantageous.