The reproducibility of a 104 item semiquantitative food frequency questionnaire to estimate the intake of energy, fat, fatty acids and cholesterol was assessed in a group of 93 men and women in the Netherlands. The questionnaire was administered by trained interviewers. Subjects were asked to recall the consumption of 104 items during the past month. A second interview was conducted 8 weeks later. The mean difference in nutrient intake between the two assessments was very small, with a maximum of 5% for cholesterol intake, but the variance of individual differences was considerable. Pearson correlation coefficients between two assessments 8 weeks apart ranged from 0.71 for polyunsaturated fatty acids intake (when expressed as percentage of energy intake) up to 0.91 for energy intake. The reproducibility was found to be similar in males and females. Reproducibility was assessed for separate food items as well as for 20 food groups. Items consumed were often highly reproducible and rarely had a poor reproducibility. This food frequency questionnaire is considered to be a suitable tool to estimate and monitor the intake of fat, fatty acids and cholesterol in the Netherlands.