The cohort consisted of persons found on the payroll of one of the Lódź cotton plants in 1964-1993 who were employed in the plant for at least 10 years. Death risk by causes was analysed using standardised mortality ratios (SMRs) calculated by the person-years method. The general population of Poland was used as the reference. In all, 7892 people were observed. As of December 31, 1995, the follow-up was completed for 7545 people (2852 men and 4693 women), i.e. the availability of the cohort was 95.6%. A total of 2069 deaths were recorded; the information on the cause of death was available for 97% of the subjects. In the male cohort, the level of the general mortality was the same as in the general population (SMR = 99). However, there was a significant increase in the number of deaths from diseases of the digestive system (SMR = 142) and larynx cancer (SMR = 188). The analysis of the results by production departments revealed in the weaving department significantly higher mortality from atherosclerosis (SMR = 141), peritoneal carcinoma (SMR = 1057) and melanoma (SMR = 677); and in the spinning department the increased risk of the hypertensive disease (SMR = 239), atherosclerosis (SMR = 175), and Hodgkin's disease (SMR = 768). Mortality in the female cohort was lower than that in the general population (SMR = 88). None of the disease groups or tumour sites caused statistically significant excess deaths either in the total cohort or in subcohorts selected according to departments. Special attention was paid to the chemical processing departments where chemicals used could contribute to the increased risk of death from cancer. Our analysis did not reveal any significant increase either in the total cohort of the workers employed in those departments or in the cohorts analyzed by duration of employment. Our results confirm the lower risk of lung cancer in the analysed group as compared with that in the general population. The numerous, but statistically insignificant increases in the incidence of malignant tumours at some specific sites detected in the subcohorts, distinguished according to the duration of employment or department, confirm the reported findings on the incidence of oral cavity, nose, throat, and larynx tumours among people exposed to harmful agents in the cotton industry.