There is sufficient evidence for an excess occurrence of stomach and lung cancer among rubber workers. However, evidence for causal associations with specific exposures is still limited. A cohort of 8,933 male German rubber workers was followed for mortality from January 1, 1981 through December 31, 1991. Work histories were reconstructed using routinely documented cost center codes. For each cost center code, calendar time- and plant-specific levels of exposure to nitrosamines, asbestos, talc (low, medium (m), and high (h)), and carbon black (two levels) were estimated by industrial hygienists. Rate ratios (RR) and 95% confidence intervals (CI) were calculated using Cox proportional hazards models, with the lowest exposure level used as the reference category. Exposure was lagged 10 years to account for latency. Exposure-response relations between exposure to asbestos and lung cancer mortality (RRm = 1.3, 95% CI: 0.9, 1.9; RRh = 2.0, 95% CI: 0.9, 4.1) and between exposure to dust (talc and asbestos combined) and stomach cancer mortality (RRm = 1.8, 95% CI: 0.9, 3.8; RRh = 2.7, 95% CI: 1.0, 7.1) were observed. Exposure to nitrosamines was not associated with mortality from stomach or lung cancer. These results suggest that the increased mortality from lung and stomach cancer among rubber workers is associated with exposure to asbestos and dust, respectively. Am J Epidemiol 2000;152:297–306.