The long-term consequences of accidental chlorine gas exposure have been investigated, mainly in the community setting, among persons exposed as a result of a nearby chlorine spill. This circumstance is not analogous to the more frequent chlorine or chlorine dioxide gas overexposures that occur commonly in pulpmills over a background of a low level of gas exposure. To investigate the respiratory health consequences of these accidental exposures ("chlorine gassing") in the industrial setting, we carried out a cross-sectional respiratory health survey among workers at a British Columbia coastal pulpmill and a nearby rail maintenance yard. A greater proportion of pulpmill workers were unavailable for study because of illness (10.5% versus 2.4% in the railyard, p less than 0.01). Procedures involved simple spirometry, respiratory symptom assessment, and measurement of average levels of air contaminants. Average chlorine levels in the pulpmill were below 1 ppm; however, 60% of the pulpmill workers reported one or more accidental "chlorine gassing" incidents. Pulpmill workers who reported being "gassed" were significantly more likely to report wheezing on occasion than were other pulpmill workers and railyard workers (rate for these three groups: nonsmokers: 8, 2, 1%; ex-smokers: 17, 11, 7%; current smokers: 42, 21, 19%; p less than 0.05). No significant lung function differences were found between the overall pulpmill group and the railyard workers; however, nonsmoking and formerly smoking pulpmill workers who reported being "gassed" had significantly lower average midmaximal flow rate and FEV1/FVC ratio than did their counterparts in the remainder of the pulpmill population (p less than 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)