The aim of this study was to compare the aetiology of neuromuscular fatigue following maximal sprints of different distances. It was hypothesized that increasing the distance would modify the type of peripheral and induce central fatigue. 11 subjects performed 100-, 200- and 400-m sprints on a motorized instrumented treadmill. Neuromuscular function, evaluated before (Pre), 30 s after (Post), and 5 and 30 min after the sprints (Post5 and Post30), consisted in determining maximal voluntary knee extensors torque (MVC), maximal voluntary activation of the knee extensors (%AL), maximal compound muscle action potential amplitude and duration on vastus lateralis, single twitch (Tw), and low- (Db10) and high-frequency torque. Compared with peak values, running speed decreased by 8%, (P < 0.01), 20% (P < 0.001) and 39% (P < 0.001) at the end of the 100-, 200- and 400-m sprints, respectively. MVC was not altered following 100 and 200 m, but decreased by 14% (P < 0.001) after the 400 m, was still depreciated Post5 (-11%, P < 0.01) and went back to initial values Post30. A decrease in %AL (-6.0%, P < 0.01) was observed Post5 for the 400 m. Tw, Db10 and low-to-high doublets ratio decreased Post-sprints and were not recovered Post30 after all sprints. Single maximal sprints of 100-400 m did not alter sarcolemmal excitability but induced progressive and substantial low-frequency fatigue and a slight reduction in neural drive with increasing sprint duration. Despite altered single or paired stimulations, MVC strength loss was detected only after the 400 m.