1. Transport of lactate, H+ and fluid across muscle sarcolemma was studied in contracting muscles under varying blood acid-base conditions. 2. Subjects performed two-legged submaximal knee-extensor exercise for 29-35 min consisting of warming up for 5 min followed by 10 min of leg exercise (L1), leg and arm exercise for 6-10 min (L2 + A) and leg exercise for 10 min (L3). The experimental protocol was performed on two occasions; inspiring air (normoxia, N) or breathing 14% O2 in N2 (hypoxia, H). Leg blood flow was measured and femoral arterial and venous blood was sampled before and during each phase of exercise. 3. Arterial blood lactate concentration increased progressively during exercise to 5.9 +/- 0.8 (N) and 8.2 +/- 0.8 mmol l-1 (H) (P < 0.05) after 5.5 min of L2 + A. Arterial blood pH was higher (P < 0.05) in H than in N, whereas arterial blood HCO3- concentrations were the same. Leg lactate release was higher in H than in N (3.1 +/- 0.7 vs. 2.0 mmol l-1 (P < 0.05) during L1. In L2 + A a net uptake of lactate was observed in both N and H. The concentration of lactate in the red blood cells increased during exercise to 2.3 +/- 0.4 (N) and 4.3 +/- 0.7 mmol l-1 (H) (P < 0.05) after 5.5 min of L2 + A, but no red blood cell femoral arterial-venous lactate difference was observed. 4. Net proton release, estimated from actual base excess (ABE) adjusted for changes in reduced haemoglobin, was significant (P < 0.05) both at rest and during each phase of exercise. Furthermore, the difference between net proton and lactate release was positive throughout exercise and of similar magnitude in N and H. 5. The present data suggest that (1) H+ exchange in muscle during submaximal exercise can to a large extent occur through mechanisms other than via coupling to lactate; (2) muscle transport of H+ is insensitive to changes in blood pH in the range of 0.02-0.08 pH units; (3) transport of lactate across the membrane of red blood cells appears to be of minor importance for lactate release from active muscles.