We tested whether the leftward shift of the oxygen dissociation curve of hemoglobin with hyperpnea delays the oxygen uptake (O₂) response to the onset of exercise. Six male subjects performed cycle ergometer exercise at a work rate corresponding to 80% of the ventilatory threshold (VT) O₂ of each individual after 3 min of 20-W cycling under eupnea [control (Con) trial]. A hyperpnea procedure (minute ventilation = 60 l/min) was undertaken for 2 min before and during 80% VT exercise in hypocapnia (Hypo) and normocapnia (Normo) trials. In the Normo trial, the inspired CO₂ fraction was 3% to prevent hypocapnia. The subjects completed two repetitions of each trial. To determine the kinetic variables of O₂ and heart rate (HR) at the onset of exercise, a nonlinear least-squares fitting was applied to the data averaged from two repetitions by a monoexponential model. The end-tidal CO₂ partial pressure before the onset of exercise was significantly lower in the Hypo trial than in the Con and Normo trials (22 ± 1 vs. 38 ± 3 and 36 ± 1 mmHg, respectively, P < 0.05). The time constant of O₂ and HR was significantly longer in the Normo trial (28 ± 7 and 39 ± 18 s, respectively) than in the Con trial (21 ± 7, 34 ± 16 s, respectively, P < 0.05). The O₂ time constant of the Hypo trial (37 ± 12 s) was significantly longer than that of the Normo trial, although no significant difference in the HR time constant was seen (Hypo, 41 ± 28 s). These findings suggested that respiratory alkalosis delayed the kinetics of oxygen diffusion in active muscle as a result of the leftward shift of the oxygen dissociation curve of hemoglobin. This supports an important role for hemoglobin-O₂ offloading in setting the O₂ kinetics at exercise onset.