Muscle fibre conduction velocity (MFCV) reflects the membrane muscle fibre properties and may be indicative of central control strategies (Andreassen & Arendt-Nielsen, 1987). Recently, the methodological limitations of estimating MFCV in dynamic conditions have been addressed (Farina et al., 2004). The aim of this study was to investigate the effects of force and speed on average MFCV during repetitive dynamic movements. Twelve healthy male participants aged between 19 and 30 years took part in the study. Subjects cycled on an electronically braked ergometer at 45, 60, 90 and 120 revolutions per minute (rpm). For each speed, the subjects performed two exercise intensities, one at the power output corresponding to the lactate threshold (determined previously) and the other at half of this power output. Surface EMG signals were detected from the vastus lateralis and medialis muscles with linear arrays of 8 electrodes (Merletti et al., 2003). Fifteen revolutions were analysed in all cases. MFCV was estimated for each cycle at six window locations corresponding to instantaneous angular velocity from the minimum to the maximum. Data were analysed by a three-way ANOVA (factors: muscle, average speed, and window location) and post-hoc Student-Newman-Keuls test.MFCV at the same average force was affected by the average speed (P < 0.05) and the window location (P < 0.001). MFCV was higher (P < 0.01) at 90 rpm (mean ± S.E.M.; 5.00 ± 0.08 m/s) than at 45 rpm (4.63 ± 0.06 m/s). Moreover, MFCV increased with instantaneous speed over the six locations (P < 0.001; from 3.98 ± 0.13 m/s to 5.09 ± 0.11 m/s). The same results were obtained comparing the speeds of 60 rpm (4.15 ± 0.14 m/s) and 120 rpm (4.93 ± 0.11 m/s). MFCV at 45 rpm was affected by force (P < 0.01; 4.63 ± 0.06 m/s vs 5.14 ± 0.07 m/s at the two force levels) and the window location (P < 0.001). A similar influence of force on MFCV was observed for the other average speeds, except 120 rpm. We have shown, for the first time during repetitive dynamic movements, that the average MFCV increased with force, probably due to the recruitment of high threshold, high conduction velocity motor units (Andreassen & Arendt-Nielsen, 1987). Furthermore, average MFCV depended both on the instantaneous speed and on the average speed of cycling. This result can be interpreted either with a different pool of recruited motor units at different velocities or with a larger average firing rate of the recruited motor units with increasing the speed of the contraction.