OBJECTIVES: To examine associations between short/medium-term variations in black smoke air pollution and mortality in the population of Glasgow and the adjacent towns of Renfrew and Paisley over a 25-year period at different time lags (0-30 days). METHODS: Generalised linear (Poisson) models were used to investigate the relationship between lagged black smoke concentrations and daily mortality, with allowance for confounding by cold temperature, between 1974 and 1998. RESULTS: When a range of lag periods were investigated significant associations were noted between temperature-adjusted black smoke exposure and all-cause mortality at lag periods of 13-18 and 19-24 days, and respiratory mortality at lag periods of 1-6, 7-12, and 13-18 days. Significant associations between cardiovascular mortality and temperature-adjusted black smoke were not observed. After adjusting for the effects of temperature a 10 μgm(-3) increase in black smoke concentration on a given day was associated with a 0.9% [95% Confidence Interval (CI): 0.3-1.5%] increase in all cause mortality and a 3.1% [95% CI: 1.4-4.9%] increase in respiratory mortality over the ensuing 30-day period. In contrast for a 10 μgm(-3) increase in black smoke concentration over 0-3 day lag period, the temperature adjusted exposure mortality associations were substantially lower (0.2% [95% CI: -0.0-0.4%] and 0.3% [95% CI: -0.2-0.8%] increases for all-cause and respiratory mortality respectively). CONCLUSIONS: This study has provided evidence of association between black smoke exposure and mortality at longer lag periods than have been investigated in the majority of time series analyses.