Female Golden Syrian hamsters, F-344 rats, Swiss CD-1 mice, and B6C3F1 hybrid mice were exposed 6 hr/day, 5 days/week to carcinogenic levels of vinyl chloride (VC) for 6, 12, 18, or 24 months (rats and hamsters only). Other groups of rodents were held for 6 or 12 months and then exposed for 6 or 12 months. At the end of the study the incidence of VC-induced neoplasms was compared in each of the groups to assess the effects of duration of exposure and age at the start of exposure on carcinogenicity of VC. In rats, with early initial exposure, hemangiosarcomas, hepatocellular carcinomas, and mammary gland carcinomas occurred with increasing incidence with longer exposure duration. Rats held for 6 months before exposure developed VC-related neoplasms, while rats held 12 months before the start of exposure failed to show a significantly increased incidence of these neoplasms. In hamsters, hemangiosarcomas, mammary gland carcinomas, gastric adenocarcinomas, and skin carcinomas resulted from VC exposure. The highest incidence of malignant neoplasms occurred in hamsters exposed for the first 12 months, whereas exposure begun after 12 months of age did not cause neoplasms. In both strains of mice, VC exposure during the first 6 months of the experiment induced a high incidence of hemangiosarcomas and mammary gland carcinomas. Swiss mice also developed lung carcinomas after only 6 months of exposure. In all three rodent species an initial 12 month exposure to VC was adequate to detect its carcinogenic potential, but the shortened survival of VC exposed mice and hamsters precluded a meaningful comparison with longer periods of exposure. Exposures were most effective when started early in life.