Apolipoprotein E (ApoE) genotype was determined in sixty-five subjects who had taken part in a 4-week randomized crossover trial to compare the effect of six mugs of black tea per day v. placebo on blood lipids and blood coagulation factors. Four ApoE genotype variants (seven E2/E3, forty-five E3/E3, twelve E3/E4 and one E4/E4) were found. ApoE allele frequency was within the range typical for Caucasian populations (ApoE-E2 5·4 %; ApoE-E3 83·8 %; ApoE-E4 10·8 %). Individuals bearing at least one E4 allele had substantially higher levels of serum total cholesterol, LDL cholesterol and triacylglycerols. Mean plasminogen activator inhibitor (PAI-1) activity was higher in ApoE-E4 allele-bearing individuals (E3/E4 + E4/E4, 11·89 (SE 1·27) U/ml; E3/E3, 9·19 (SE 0·80) U/ml; E2/E3, 7·21 (SE 1·04) U/ml, P values of E4-group v. E3 and E2 being respectively 0·093 and 0·030). These unexpected findings imply that elevated PAI-1 activity may be a hitherto unrecognized additional factor involved in the increased cardiovascular disease risk associated with apoE-E4 allele. The interactions between tea drinking and genotype were also examined. In the E3/E3 homozygotes, HDL-cholesterol was significantly reduced in the tea period (mean placebo 1·54 mmol/l v. mean tea 1·50 mmol/l, P = 0·027). In the E2/E3 group, triacylglycerol concentration was significantly reduced (mean placebo 1·18 mmol/l v. mean tea 1·09 mmol/l, P = 0·039). Tea also caused a significant decrease of PAI-1 activity in the subjects with E2/E3 genotype (mean placebo 7·21 U/ml v. mean tea 5.88 U/ml, P = 0·007). In the other two genotype groups, there was no significant effect of tea. The results indicate that tea drinking has a beneficial effect on some cardiovascular disease risk-associated factors, especially in E2 allele-bearing individuals. Dietary intervention may be particularly effective in population groups with certain genetic characteristics.