Previous studies of mortality among white males employed in a Charleston, South Carolina asbestos textile plant using chrysotile demonstrated significant excess mortality due to asbestos-related disease and a steep exposure-response relationship for lung cancer. This cohort was further studied by adding 15 years of follow-up and including mortality among white female and black male workers. Nested case-control analyses were undertaken to further explore possible differences in lung cancer risk by textile operation as well as possible confounding by mineral oil exposures. Preliminary data for white males have been previously published. White males experienced statistically significant excess mortality due to lung cancer (standardized mortality ratio [SMR] = 2.30; confidence interval [CI] = 1.88-2.79), all causes (SMR = 1.48; CI = 1.38-158), all cancers (SMR = 1.50; CI = 1.29-1.72), diabetes mellitus (SMR = 2.05; CI = 1.18-3.33), heart disease (SMR = 1.41; CI = 1.26-1.58), cerebrovascular disease (SMR = 1.50; CI = 1.08-2.02), pneumoconiosis and other respiratory diseases (SMR = 4.10; CI = 3.10-5.31), and accidents (SMR = 1.49; CI = 1.15-1.91). Among white females, statistically significant excesses occurred for lung cancer (SMR = 2.75; CI = 2.06-3.61), all causes (SMR = 1.21; CI = 1.11-1.32), pneumoconiosis and other respiratory diseases (SMR = 2.40; CI = 1.53-3.60), and other respiratory cancers (SMR = 14.98; CI = 4.08-38.7). Among the total cohort of black males, the only statistically significant excess observed was for pneumoconiosis (SMR = 2.19; CI = 1.23-3.62). Based on historical exposure measurements at the plant, there was a positive exposure-response relationship for both lung cancer and pneumoconiosis. Data for the entire cohort demonstrate an increase in the lung cancer relative risk of 2-3% for each fiber/cc-year of cumulative chrysotile exposure. This relationship was more consistent for the white male workers. The excess risk for lung cancer among white males and females appeared to occur at cumulative exposures lower than those for black males. Possible reasons for the lesser lung cancer risk among black males include less smoking and differences in airborne fiber characteristics experienced by black males as a result of plant job placement patterns. The case-control analysis found employment in preparation and carding operations (where most of the black males worked) to be associated with a slightly reduced lung cancer risk, although not statistically significant, whereas spinning and twisting employment was associated with a statistically significant increased lung cancer risk compared to other plant operations.(ABSTRACT TRUNCATED AT 400 WORDS)