The relationship between incidence of prostate cancer and intake of dietary fat and foods rich in fat was studied in 25,708 men aged 16-56 years attending a Norwegian health screening in 1977-1983. Linkage to the Cancer Registry of Norway and the Central Bureau of Statistics of Norway ensured a complete follow-up until December 31, 1992. Diet was recorded on a semi-quantitative food-frequency questionnaire at the time of screening, and 72 cases of prostate cancer were identified during follow-up. At the end of follow-up, mean age of the total study sample was 56 years (range 19-68), while mean age at diagnosis of prostate cancer was 60 years (range 47-67). No association was found between energy-adjusted intake of total fat, saturated fat, mono-unsaturated fat or poly-unsaturated fat and the incidence of prostate cancer. Significant positive associations were found for body mass index (BMI) and consumption of hamburgers/meatballs, while no association was found with consumption of frankfurters/sausages and a significant negative association with the weekly number of main meals with meat. A significantly increased risk of prostate cancer was associated with skim milk as compared to whole milk. Milk preference (skim vs. whole) was associated significantly positively with BMI. Our study of a relatively young cohort does not confirm previous case-control and cohort studies suggesting that dietary fat, especially from animal sources, is associated positively with risk of prostate cancer.