Arterial PCO2 (PaCO2) has been estimated during exercise with good accuracy in younger individuals by using the Jones equation (PJCO2) (J. Appl. Physiol. 47: 954-960, 1979). The purpose of this project was to determine the utility of estimating PaCO2 from end-tidal PCO2 (PETCO2) or PJCO2 at rest, ventilatory threshold (VTh), and maximal exercise (Max) in older subjects. PETCO2 was determined from respired gases simultaneously (MGA 1100) with arterial blood gases (radial arterial catheter) in 12 older and 11 younger subjects at rest and during exercise. Mean differences were analyzed with paired t-tests, and relationships between the estimated PaCO2 values and the actual values of PaCO2 were determined with correlation coefficients. In the older subjects, PETCO2 was not significantly different from PaCO2 at rest (-1.2 +/- 4.3 Torr), VTh (0.4 +/- 2.5), or Max (-0.8 +/- 2.7), and the two were significantly (P < 0.05) correlated at Vth (r = 0.84) and Max (r = 0.87) but not at rest (r = 0.47). PJCO2 was similar to PaCO2 at rest (-1.0 +/- 3.9) and Vth (-1. 3 +/- 2.3) but significantly lower at Max (-3.0 +/- 2.6), and the two were significantly correlated at Vth (r = 0.86) and Max (r = 0. 80) but not at rest (r = 0.54). PETCO2 was significantly higher than PaCO2 during exercise in the younger subjects but similar to PaCO2 at rest. PJCO2 was similar to PaCO2 at rest and Vth but significantly lower at Max in younger subjects. In conclusion, our data demonstrate that PaCO2 during exercise is better estimated by PETCO2 than by PJCO2 in older subjects, contrary to what is observed in younger subjects. This appears to be related to the finding that PETCO2 does not exceed PaCO2 during exercise in older subjects, as occurs in the younger subjects. However, PaCO2 at rest is best estimated by PJCO2 in both younger and older subjects.