To test the hypothesis that red wine, by virtue of its relatively high concentration of polyphenols, is more protective against atherosclerosis and coronary heart disease (CHD) than white wine, and that grape juice enriched in one of these, trans-resveratrol, may share some of these properties, studies were performed on 24 healthy males aged 26-45 years. Each consumed the following beverages for periods of 4 weeks: red wine, white wine, commercial grape juice and the same grape juice enriched with trans-resveratrol. Apart from the last beverage, 2 weeks abstinence was maintained before commencing the schedule. Blood was taken at the beginning and end of each schedule to determine plasma thromboxane B2 (TxB2) concentration and the IC50 (concentration required for 50% aggregation) for ADP and thrombin-induced platelet aggregation. White wine (P < 0.05) but not red wine increased the IC50 for ADP. Both wines increased the IC50 for thrombin (P < 0.02 and P < 0.001, respectively) and also lowered plasma TxB2 concentrations (P < 0.01 and P < 0.025, respectively). Neither grape juice altered ADP-induced aggregation or TxB2 concentrations, but the commercial juice lowered the IC50 for thrombin (P < 0.001) whereas the resveratrol-enriched juice caused a dramatic increase (P < 0.001). In vitro experiments demonstrated that the aggregation of fresh washed human platelets by ADP and thrombin was moderately reduced by both grape juices, strongly by red wine and not at all by white wine. The synthesis of TxB2 by platelets from labelled arachidonate was stimulated by commercial grape juice, slightly enhanced by resveratrol-enriched juice and strongly inhibited by red wine with white wine having little effect. Platelets from subjects consuming the commercial juice had a higher ratio of cyclo-oxygenase to lipoxygenase product formation and those consuming the resveratrol-enriched juice a lower ratio than during the control period. We conclude that trans-resveratrol can be absorbed from grape juice in biologically active quantities and in amounts that are likely to cause reduction in the risk of atherosclerosis. The failure of red wines (which have a 20-fold excess of polyphenols over white wines) to show any advantage suggests that, in vivo, ethanol is the dominant anti-aggregatory component in these beverages which are more potent than grape juices in preventing platelet aggregation in humans.