1. A randomized controlled trial of cross-over design was set up to examine the effect of alcohol on blood lipids and certain haematological variates relevant to ischaemic heart disease.
2. One hundred subjects drank some alcohol for 4 weeks (mean intake 18.4 g/d) and abstained totally for 4 weeks, the order of these periods being randomized.
3. Alcohol appeared to produce a rise of 7% in serum high-density-lipoprotein(HDL) cholesterol, probably due to a rise in the HDL2 subfraction.
4. No significant change was detected in plasma fibrinogen or the other haematoiogical indices.
5. These results are consistent with the hypothesis that a moderate intake of alcohol confers some protection against heart disease.