The effect of hypoxic breathing on pulmonary O-2 uptake VO2p), leg blood flow (LBF) and O-2 delivery and deoxygenation of the vastus lateralis muscle was examined during constant-load single-leg knee-extension exercise. Seven subjects (24 4 years; mean S.D.) performed two transitions from unloaded to moderate-intensity exercise (21 W) under normoxic and hypoxic (PETO2 = 60 mmHg) conditions. Breath-by-breath VO2p and beat-by-beat femoral artery mean blood velocity (MBV) were measured by mass spectrometer and volume turbine and Doppler ultrasound (VingMed, CFM 750), respectively. Deoxy-(HHb), oxy-, and total haemoglobin/myoglobin were measured continuously by near-infrared spectroscopy (NIRS; Hamamatsu NIRO-300). VO2p data were filtered and averaged to 5 s bins at 20, 40, 60, 120, 180 and 300 s. MBV data were filtered and averaged to 2 s bins (I contraction cycle). LBF was calculated for each contraction cycle and averaged to 5 s bins at 20, 40, 60, 120, 180 and 300 s. VO2p was significantly lower in hypoxia throughout the period of 20, 40, 60 and 120 s of the exercise on-transient. LBF (1 min(-1)) was approximately 35% higher (P >,0.05) in hypoxia during the on-transient and steady-state of KE exercise, resulting in a similar leg O-2 delivery in hypoxia and normoxia. Local muscle deoxygenation (HHb) was similar in hypoxia and normoxia. These results suggest that factors other than O-2 delivery, possibly the diffusion of O-2, were responsible for the lower O-2 uptake during the exercise on-transient in hypoxia.