We investigated the association between air pollution and hospital admissions for chronic obstructive pulmonary disease and pneumonia among the elderly in Minneapolis-St. Paul, MN, and Birmingham, AL, over the period January 1, 1986, to December 31, 1991. Pollutants included in our analyses were PM10 (particulate matter less than 10 microns in aerodynamic diameter), SO2, NO2, O3, and CO in Minneapolis-St. Paul, and PM10, O3, and CO in Birmingham. After adjusting for temperature, day of week, season, and temporal trends, we found little evidence of association between air pollution and hospital admissions for respiratory causes in Birmingham. In contrast, we found that air pollution was associated with hospital admissions for respiratory causes in Minneapolis-St. Paul. Among the individual pollutants, O3 was most strongly associated with admissions (estimated increase in hospital admissions associated with a 15-parts-per-billion increase in O3 on the previous day = 5.15%; 95% confidence interval = 2.36-7.94%), and this association was robust in the sense that it was little affected by the simultaneous consideration of other pollutants. PM10, SO2, and NO2 were also associated with hospital admissions, although none could be singled out as being more important than the others.