It was previously shown that gas exchange could contribute to the rising slope of phase III of the single-breath nitrogen (SB-N2) test. This study was done to quantify this role. We studied eight normal volunteers with a series of SB-N2 derived tests where the RV gas was progressively changed from room air to 90% O2 and 10% N2, by 10% increments in O2 and 10% decreases in N2 concentrations (i.e. room air, 70% N2 30% O2, 60% N2 40% O2, etc.). A similar series of SB-R (single-breath reversed gradients test) derived tests was done. Here the RV contained 100% O2 by previous washout, while the inspired gas was changed by 105 steps from room air to 10% N2 90% O2. We therefore have a situation where dilutional N2 gradients change with the % N2, in either the RV or the inspired gas. However, the alveolar volume loss remains the same for all tests. The mean +/- SD slope of phase III in the SB-N2 series for our eight subjects decreased from 0.87 +/- 0.25 with room air to 0.14 +/- 0.07 with 10% N2 90% O2, while its steepness in the SB-R series decreased from 0.62 +/- 0.23 with the inspired room air to 0.11 +/- 0.06 with the final inspiration being 10% N2 90% O2. From these data, we could calculate that the mean % contribution of gas exchange to the slope of phase III was 10.2%.