The purpose of this study was to examine the effects on lipoprotein risk markers for CHD of oestradiol given alone and in combination with the androgenic progestogen, norethisterone. Eighty postmenopausal women were randomly allocated to receive oestradiol (2 mg/day) alone or with continuous norethisterone (1 mg/day). Serum lipoprotein levels, including lipoprotein(a), were monitored during 12 months on treatment in all the women, and in a sub-set of 32 patients cholesterol was measured in the two major density subfractions of LDL. Oestradiol caused a transient rise in triglycerides, a small decrease in LDL cholesterol (significant only at 3 and 6 months, P < 0.05) and a consistent significant increase in HDL cholesterol (16%, P < 0.01). There was a downward trend in lipoprotein(a) levels which did not achieve statistical significance. The combined preparation caused significant, sustained decreases in triglycerides (31%, P < 0.01), total cholesterol (15%, P < 0.001), VLDL (42%, P < 0.01), LDL (9%, P < 0.05) and HDL (11%, P < 0.001). Lipoprotein(a) was also reduced (39%, P < 0.05). In the sub-set of patients in which LDL subfractions were measured, the reduction in LDL induced by oestradiol monotherapy was significant only at the 3-month visit (6%, P < 0.05). This was due to a decrease in the 'light' (1.025 < d < 1.044 g/ml) subfraction (10%, P < 0.05) and resulted in an apparent shift in subfraction distribution towards the 'heavy' (1.044 < d < 1.060 g/ml) subfraction, although there was no absolute increase in the latter. None of these changes was statistically significant at 12 months. Oestradiol/norethisterone caused sustained decreases in both 'light' (15%, P < 0.05) and 'heavy' (29%, P < 0.05) subfractions, with no significant change in the relative amounts. The changes in 'light' and 'heavy' LDL in this group were highly correlated with changes in triglyceride levels (r = -0.57, P < 0.05 and r = 0.82, P < 0.01 respectively). Therefore, at the end of 1 year's treatment with unopposed oestradiol the only statistically significant change was an increase in HDL cholesterol. Addition of norethisterone to the preparation reversed this potentially beneficial change, but favourably influenced triglycerides, VLDL, LDL subfraction profile and lipoprotein(a), which may counteract the adverse effect on HDL.