The reproducibility and validity of a food frequency questionnaire designed to measure intakes of total fat, saturated and polyunsaturated fats, vitamins A, C, and E, selenium, and dietary fiber were tested from March to October 1984 among 297 Finnish men aged 55-69 years. The questionnaire asked about consumption of 44 food items. In the reproducibility study, 107 subjects filled in the questionnaire three times, at three-month intervals. Intraclass correlations varied from 0.52 for vitamin A to 0.85 for polyunsaturated fat. In the validity study, 190 subjects kept food consumption records for 12 two-day periods distributed evenly over a period of six months and filled in the questionnaire both before and after this period. Correlations between the nutrient intake values from the food records and those from the food frequency questionnaires ranged from 0.33 for selenium to 0.68 for polyunsaturated fat. On the average, 40-45% of the subjects in the lowest and highest quintiles based on food records were in the same respective quintiles when assessed by the food frequency questionnaire, and 70-75% were in the two lowest and two highest questionnaire quintiles, respectively. The food frequency questionnaire and a quantitative food use questionnaire tested in the same study were compared. Use of these two instruments in large-scale epidemiologic studies is discussed.