ABSTRACT.The effect of exercise, either long or short term, on levels of 2,3‐DPG and P50are at present controversial. The present study compared the effects of shortterm maximal exercise (average time 5 min 35 sec, maximal group) on levels of 2,3‐DPG and P50to those after a 10 mile road race (average time 51 min 36 sec, endurance group). The resting levels of 2,3‐DPG of both groups of athletes were significantly higher (P < 0.01) than the mean for a group of sedentary males simultaneously determined. The resting P50value‐for endurance group athletes was also significantly elevated. After short‐term maximal exercise (maximal group), a mean 15% increase in 2,3‐DPG with an increase ofin vitroP50(7.4;37C)for 27.2 ± 0.5 to 30.0 ± 0.9 mmHg was observed. On the other hand after prolonged endurance exercise (endurance group), a small but significant (P < 0.05) decrease in 2,3‐DPG and a decrease ofin vitroP50(7.4;37C)from an elevated resting level of 28.9 ± 1.5 mmHg to the normally determined sedentary value (26.7 ± 1.5 mmHg) resulted. Different stimuli are involved in short‐term anaerobic maximal exercise compared to prolonged aerobic exercise and a model is presented to account for the alterations in 2,3‐DPG and oxy‐Hb affinity in the two types of exercise.