Cross-sectional relationships between diet and total serum cholesterol levels were studied in a sample of 428 women from the Framingham Heart Study Cohort, aged 37 to 70 years, from 1957 to 1960. Multiple linear regression was used to control for total calorie intake, systolic blood pressure, physical activity, Metropolitan relative weight, glucose intolerance, and cigarette smoking. There was little evidence for a relationship between total serum cholesterol and dietary fat intake; whereas a marginally significant direct association was found with total fat in postmenopausal women, total and plant fat and cholesterol were inversely associated, and only cholesterol was significant in premenopausal women. A consistent inverse association was observed between total serum cholesterol levels and intake of protein, particularly from plant sources, and a weak inverse association was found with complex carbohydrate intake. Serum cholesterol in women may be influenced by a number of dietary factors and appears to differ according to menopausal status.