The aim of the present study was to investigate the influence of pedalling rate and power output in cycling on the neuromuscular activity of the knee extensor muscles. Ten subjects took part in 15 randomised trials, which consisted of three levels of power outputs (60%, 80% and 100% maximal aerobic power) and five cadences (70%, 85%, 100%, 115% and 130% of the freely chosen cadence, FCC). Root mean square (rms) was utilized to quantify electromyographic activity of the vastus lateralis (VL), vastus medialis (VM) and rectus femoris (RF) muscles. The mean (SD) FCC did not change with power output, ranging from 85.0 (11.9) to 88.0 (11.1) rpm. A significant power effect ( P<0.01) for the rms of VL, VM and RF muscles was observed. Results showed no significant cadence effect on neuromuscular activity of the VL and VM muscles, while the rms of the RF muscle was significantly greater ( P<0.05) at 70% FCC when compared to other cadences. In conclusion, the neuromuscular activity of the knee extensor muscles was not significantly influenced by cadence manipulations. Thus, minimisation of the neuromuscular activity of these muscles would not seem to lead to the choice of a cadence in cycling.