Intact skeletal muscle fibres from adult mammals exhibit neither spontaneous nor stimulated Ca²⁺ sparks. Mechanical or chemical skinning procedures have been reported to unmask sparks. The present study investigates the mechanisms that determine the development of Ca²⁺ spark activity in permeabilized fibres dissected from muscles with different metabolic capacity. Spontaneous Ca²⁺ sparks were detected with fluo‐3 and single photon confocal microscopy; mitochondrial redox potential was evaluated from mitochondrial NADH signals recorded with two‐photon confocal microscopy, and Ca²⁺ load of the sarcoplasmic reticulum (SR) was estimated from the amplitude of caffeine‐induced Ca²⁺ transients recorded with fura‐2 and digital photometry. In three fibre types studied, there was a time lag between permeabilization and spark development. Under all experimental conditions, the delay was the longest in slow‐twitch oxidative fibres, intermediate in fast‐twitch glycolytic–oxidative fibres, and the shortest in fast‐twitch glycolytic cells. The temporal evolution of Ca²⁺ spark frequencies was bell‐shaped, and the maximal spark frequency was reached slowly in mitochondria‐rich oxidative cells but quickly in mitochondria‐poor glycolytic fibres. The development of spontaneous Ca²⁺ sparks did not correlate with the SR Ca²⁺ content of the fibre, but did correlate with the redox potential of their mitochondria. Treatment of fibres with scavengers of reactive oxygen species (ROS), such as superoxide dismutase (SOD) and catalase, dramatically and reversibly reduced the spark frequency and also delayed their appearance. In contrast, incubation of fibres with 50 μm H₂O₂ sped up the development of Ca²⁺ sparks and increased their frequency. These results indicate that the appearance of Ca²⁺ sparks in permeabilized skeletal muscle cells depends on the fibre's oxidative strength and that misbalance between mitochondrial ROS production and the fibre's ability to fight oxidative stress is likely to be responsible for unmasking Ca²⁺ sparks in skinned preparations. They also suggest that under physiological and pathophysiological conditions the appearance of Ca²⁺ sparks may be, at least in part, limited by the fine‐tuned equilibrium between mitochondrial ROS production and cellular ROS scavenging mechanisms.