Nine asthmatic subjects exercised at low, moderate, and high work rates on a cycle ergometer while breathing filtered, humidified air with or without 0.5 ppm of sulfur dioxide (SO2) in a double-blind study. Subjects first performed these experiments breathing through a mouthpiece while wearing a noseclip (oral breathing) and then repeated the experiments breathing through a facemask that separated and permitted independent measurement of oral and nasal air flow (oronasal breathing). We determined specific airway resistance before and after exercise by body plethysmography. Inhaled by mouthpiece, 0.5 ppm So2 caused bronchoconstriction at moderate and high but not at low work rates. There was a dose-response relationship between the work rate performed and the degree of bronchoconstriction induced. Inhaled oronasally, 0.5 ppm SO2 caused bronchoconstriction only at the high work rate. These findings demonstrate that So2-induced bronchoconstriction is dependent on the work rate of exercise during exposure, that oronasal breathing is only partially effective in preventing the bronchoconstriction observed with oral breathing, and that oronasal breathing is less effective in preventing bronchoconstriction with high than with moderate exercise at this concentration of SO2.