We examined the effects on serum sex steroids, lipids and markers of oxidative stress of supplementing the diets of healthy male volunteers with scones made with soya flour. A randomized placebo controlled cross-over trial. University Hospital of Wales. Twenty volunteers recruited by advertisement. Male volunteers ate three scones a day in addition to their normal diet for a period of 6 weeks. The scones were made with either wheat or soya flour (containing 120 mg/day of isoflavones). Blood was analysed for sex steroids (testosterone, dihydro-testosterone, oestradiol, oestrone, sex hormone binding globulin, albumin and the concentration of non-protein bound sex steroids were calculated), lipid profile (total cholesterol, high density lipoprotein cholesterol and triglycerides) and measures of oxidative stress (hydroperoxides, susceptibility of LDL to oxidation with copper and myeloperoxidase). The volunteers' mean age was 35.6 (s.d. 11.2) y. Total serum testosterone fell in volunteers taking the soya scones (19.3-18.2 nmol/l; 95% CI 1.01, 1.12; P=0.03). No significant changes were seen in the concentrations of the other serum sex steroids, albumin or sex hormone binding globulin throughout the study. Significant improvements in two of the three markers of oxidative stress were seen in volunteers taking soya scones. Lag time for myeloperoxidase rose from 55.0 to 68.0 min (95% CI -16.0, -3.5; P=0.009) and the presence of hydroperoxides decreased from 2.69 to 2.34 micro mol/l (95% CI 0.12, 0.71; P=0.009). There were no changes seen in serum triglycerides or cholesterol. We have shown that soya supplements reduce serum testosterone and improve markers of oxidative stress. These findings provide a putative mechanism by which soya supplements could protect against prostatic disease and atherosclerosis. Further dietary studies with clinical end points are warranted.