At the onset of skeletal muscle repetitive contractions, there is a significant delay in the time to achieve oxidative phosphorylation steady state. The purpose of the present study was to examine the factors that limit oxidative phosphorylation at the onset of contractions. NAD(P)H was measured in real time during two contractile periods (2 min each) separated by 5 min of rest in intact single muscle fibres (n = 7) isolated from Xenopus laevis. The fibres were then loaded with the dye tetramethylrhodamine methyl ester perchlorate (TMRM) to evaluate the kinetics of the mitochondrial membrane potential (Δψ (m)) during two further successive contractile periods. At the onset of contractions in the first period, NAD(P)H exhibited a time delay (14.1 ± 1.3 s) before decreasing toward a steady state. In contrast, Δψ(m) decreased immediately after the first contraction and started to be reestablished after 10.7 ± 0.9 s, with restoration to the pre-stimulation values after approximately 32 s. In the second contractile period (5 min after the first), NAD(P)H decreased immediately (i.e. no time delay) after the first contraction and had a significantly shorter time constant compared to the first contractile bout (3.3 ± 0.3 vs. 5.0 ± 0.2 s, P < 0.05). During the second bout, Δψ(m) remained unchanged from pre-stimulation values. These results suggest: (1) that at the onset of contractions, oxidative phosphorylation is primarily limited by the activity of the electron transport chain complexes rather than by a limited level of substrates; and (2) when the muscle is 'primed' by previous contractile activity, the faster enhancement of the cellular respiratory rate is due to intrinsic factors within the myofibre.