In order to evaluate the accuracy of our method of P50determination, we measured P50and 2,3-DPG on different days on several members of our laboratory staff and were surprised to find significant variation. Initially, we indicted our methodology but subsequent data suggest that significant P50variability occurs, not correlated to alterations in 2,3-DPG or red cell pH. The largest variability on different days, 8.5 torr, occurred in a laboratory staff member studied 7 days apart.The P50determinations were made by fixing Po2by tonometry and determining oxyhemoglobin saturation by Van Slyke analysis.In contrast to our in vivo data, serial studies over 5–20 days performed upon 3 samples of stored CPD bank blood demonstrated the expected progressive decline in both 2,3-DPG and P50