A study of air pollution and daily hospital admissions for respiratory causes was conducted in Toronto, Ontario. Fine aerosol (da < 2.5 microns) samples were collected daily at a central city site during July and August 1986, 1987, and 1988 and were subsequently extracted and analyzed for daily particulate phase aerosol strong acidity (H+) and sulfates (SO4 =). Daily counts of respiratory admissions to 22 acute care hospitals and daily meteorological and environmental data (e.g. ozone [O3], total suspended particulate matter [TSP], and thoracic particle mass [PM10] were also obtained. Regression analyses indicated that only the O3, H+, and SO4 = associations with respiratory and asthma admissions remained consistently significant after controlling for temperature. Even after excluding days with maximum 1-hr O3 > 120 ppb, O3 was still strongly significant. In the various model specifications considered, the relative particle metric strengths of association with admissions were generally H+ > SO4 = > FP > PM10 > TSP, indicating that particle size and composition are of central importance in defining the adverse human health effects of particulate matter. On average, summertime haze air pollution was associated with 24% of all respiratory admissions (21% with O3, 3% with H+). On peak pollution days, however, aerosol acidity yielded the highest relative risk estimates (e.g., RR = 1.5 at 391 nmole/m3 H+), and summertime haze was associated with roughly half of all respiratory admissions.