This study was designed to investigate whether the mortality of British pharmaceutical industry workers showed any evidence of occupationally related patterns associated with particular sections of the industry with special emphasis on cancers and respiratory diseases. The data were extracted from census schedules by the Office of Population Censuses and Surveys using occupational categories recorded by pharmaceutical workers at the 1961 and 1971 censuses of England and Wales. Subsequent mortality was ascertained through the National Health Service Central Registry. Both cohorts, followed up to the end of 1981, showed a "healthy worker effect," with overall SMRs ranging from 0.77 to 1.04, though comparison with the employed populations of the longitudinal study reduced this effect by 25%. Evidence of inconsistent excess mortality was noted in small numbers for suicide (SMR 0.91 to 2.79) and cancer of the large intestine (SMR 1.09 to 2.38). There was no clear association between the type of industry and the causes of death investigated. Although the census based study has several methodological shortcomings, there was no evidence, on the basis of this study, to suggest any excess mortality risk from employment in the pharmaceutical industry.