1. Increased blood or plasma viscosity has been observed in almost all conditions associated with accelerated atherosclerosis. Cognizant of the enlarging body of evidence implicating increased viscosity in atherogenesis, we hypothesize that the effects of low-density lipoprotein and high-density lipoprotein on blood viscosity correlate with their association with risk of atherosclerosis. 2. Blood viscometry was performed on samples from 28 healthy, non-fasting adult volunteers using a capillary viscometer. Data were correlated with haematocrit, fibrinogen, serum viscosity, total cholesterol, high-density lipoprotein-cholesterol, triglycerides and calculated low-density lipoprotein-cholesterol. 3. Low-density lipoprotein-cholesterol was more strongly correlated with blood viscosity than was total cholesterol (r = 0.4149, P = 0.0281, compared with r = 0.2790, P = 0.1505). High-density lipoprotein-cholesterol levels were inversely associated with blood viscosity (r = - 0.4018, P = 0.0341). 4. To confirm these effects, viscometry was performed on erythrocytes, suspended in saline, which had been incubated in plasma of various low-density lipoprotein/high-density lipoprotein ratios. Viscosity correlated directly with low-density lipoprotein/high-density lipoprotein ratio (n = 23, r = 0.8561, P < 0.01). 5. Low-density lipoprotein receptor occupancy data suggests that these effects on viscosity are mediated by erythrocyte aggregation. 6. These results demonstrate that the effects of low-density lipoprotein and high-density lipoprotein on blood viscosity in healthy subjects correlate with their association with risk of atherosclerosis. These effects on viscosity may play a role in atherogenesis by modulating the dwell or residence time of atherogenic particles in the vicinity of the endothelium.