The purpose of this study was to compare two methods of CO2 rebreathing-equilibrium method (Collier, 1965) and extrapolation method (Defares, 1958) -for determining the mixed venous CO2 pressure (PvCO2) during exercise. Following a sitting rest for 5 min, five healthy males performed upright bicycle ergometer exercises with work rate of 100 and 150W for 5 min. During both at rest and in exercise, PvCO2 was measured twice to evaluate the reproducibility of both methods. The equilibrium method was performed according to that by Jones & Campbell (1982) . In the extrapolation method, instead of using the graphical analysis described by Klauaen (1965), PvCO2 was directly calculated by applying the modified exponential equation (Defares, 1958) . In the duplicate tests, coefficient of variation was lower while correlation coefficient was higher, with increasing work rate in both methods. It was noted that these tendencies were more pronounced in the extrapolation method than in the equilibrium method. It was concluded that in contrast to the results of Auchincloss et al. (1980), the extrapolation method was more reproducible than the equilibrium method.