The purpose of this study was to determine the relationship between blood gases and acid-base measurements in arterial, arterialized venous, and venous blood measured simultaneously during short-term maximal exercise. Ten well-trained male cyclists performed a graded maximal exercise test on a cycle ergometer to determine the power output corresponding to their peak oxygen consumption (test I), and a short-term maximal test on a cycle ergometer at peak power output (test 11). During test 11 arterial, arterialized venous and venous blood were sampled simultaneously for determination of partial pressures of oxygen and carbon dioxide, pH, bicarbonate (HCO3−), base excess (BE), and lactate (La). Samples were taken at rest, the end of 1 min of exercise (1 ME), at the end of exercise (EE), and at 2 min of recovery (REC). During test II, subjects maintained a peak power output of 370.6 (62.1) W [mean (SD)] for 4.5, SD 1.6 min. Except at rest venous and arterialized venous measurements tended to be the same at all sampling intervals, but differed significantly from measurements in arterial blood (Pr=0.83 [regression equation: BEa=(0.84 BEav)−0.51]. Arterial La was significantly higher than venous La at 1 ME (2.8, 0.7 vs 0.8, 0.3mmol · 1−1) and higher than both venous and arterialized venous La at EE. At EE La concentration was 9.2, SD 2.0, 4.6, SD 0.4, and 5.1, SD 0.1 mmol · 1−1 in arterial, venous, and arterialized venous blood respectively. It is concluded that except for base excess, blood gases and acid base measurements in venous and arterialized venous blood do not accurately reflect values found in arterial blood during short-term maximal exercise. We suggest that these differences may be due in part to clearance by inactive muscle near the sampling site or vasoconstriction at the inactive sampling site.