Objectives. The possible etiologic relevance of occupational factors such as cadmium, cutting oils, diesel fuel and fumes, herbicides, polycyclic aromatic hydrocarbons (PAH), polychlorinated biphenyls, soot, tar, mineral oil, and solvents to prostate cancer was studied. Methods. A case-referent study design was used to recruit 192 subjects with histologicaliy confirmed prostate cancer and 210 referents who had prostate cancer histologically excluded either in one of two urologic practices (Hamburg and Frankfurt) or in the urological policlinic of the Frankfurt University. Data were gathered with a self-administered questionnaire and analyzed using logistic regression to control for age, region, and cigarette smoking. A job-exposure matrix was used for assigning exposure. For the calculation of dose-years, the duration of contact with specific substances was weighted by the intensity and probability of exposure according to a job-exposure matrix. Results. The analysis of dose-years yielded a statistically significant association between occupational exposure to diesel fuel or fumes and prostate cancer (odds ratio 3.7, 95% confidence interval 1.4-9.81 for subjects exposed to more than 25 dose-years in a comparison with subjects never exposed). For the other substances, no statistically significant differences in exposure were found between the cases and referents. When only jobs with a high exposure probability were used to classify the participants as exposed, only exposure to PAH was significantly associated with prostate cancer. Conclusions. In keeping with results from other studies, this study provides further evidence that exposure to diesel fuel or fumes - possibly mediated through PAH - may be associated with the development of prostate cancer.