We undertook a population-based case-control study to investigate early life risk factors for prostate cancer. Information on dietary habits during childhood and adolescence, childhood environment, pubertal development, and physical activity was collected by face-to-face interviews with 256 (74.6%) of all eligible cases and 252 (76.6%) of all selected controls, frequency matched by age. All potential controls were screened for prostate cancer with negative findings. Odds ratios with 95% confidence intervals were estimated by logistic regression. Analyses of localized (T0-2' M0) and more advanced cancers were made separately. In general, there was no clear association between diet and prostate cancer risk. An increased risk associated with childhood living in more densely populated, compared with rural, areas was found (odds ratio = 2.1; 95% confidence interval = 1.3-3.5); this effect was most apparent for localized cancers (odds ratio = 3.2' 95% confidence interval = 1.7-6.2). There was no substantial association between adult height or body mass index and prostate cancer, but exercise appeared negatively associated with risk (P value for trend, 0.13). We conclude that our study provides some indications that exposures early in life are important in the etiology of prostate cancer.