A rapid infrared CO2 analyzer has been used to evaluate a bag rebreathing method for the determination of mixed venous CO2 tension. Using small bag volumes and an initial bag CO2 tension averaging 11 mm Hg above effective alveolar tension, a plateau of CO2 concentration was obtained during normal spontaneous breathing in 18 seconds or less in 16 out of 18 trials during cardiac catheterization. The plateau so obtained was compared with the CO2 tension of a mixed venous sample drawn approximately 1 minute before the rebreathing procedure. The mean difference between the two was +0.1 mm Hg with a standard deviation of 3.1 mm Hg. This accuracy is not sufficient for the determination of cardiac output. Recirculation of blood within the rebreathing period did not introduce apparent errors. The mixed venous CO2 tension by the rebreathing method was determined with greater accuracy than was the effective alveolar CO2 tension from the peak end-tidal CO2 tensions. Abnormalities of distribution of the blood and gas in the lung interfered less with the determination of mixed venous tension by this method than with the determination of alveolar CO2 tension. Thus it is possible to utilize the mixed venous CO2 tension as obtained from rebreathing plateaus to evaluate the ventilatory status of subjects with moderate abnormalities of distribution of gas and blood in the lungs. Submitted on November 14, 1955