Muscle ATP, creatine phosphate and lactate, and blood pH and lactate were measured in 7 male sprinters before and after running 40, 60, 80 and 100 m at maximal speed. The sprinters were divided into two groups, group 1 being sprinters who achieved a higher maximal speed (10.07 +/- 0.13 m X s-1) than group 2 (9.75 +/- 0.10 m X s-1), and who also maintained the speed for a longer time. The breakdown of high-energy phosphate stores was significantly greater for group 1 than for group 2 for all distances other than 100 m; the breakdown of creatine phosphate for group 1 was almost the same for 40 m as for 100 m. Muscle and blood lactate began to accumulate during the 40 m exercise. The accumulation of blood lactate was linear (0.55 +/- 0.02 mmol X s-1 X l-1) for all distances, and there were no differences between the groups. With 100 m sprints the end-levels of blood and muscle lactate were not high enough and the change in blood pH was not great enough for one to accept that lactate accumulation is responsible for the decrease in running speed over this distance. We concluded that in short-term maximal exercise, performance depends on the capacity for using high-energy phosphates at the beginning of the exercise, and the decrease in running speed begins when the high-energy phosphate stores are depleted and most of the energy must then be produced by glycolysis.