The mitochondrial theory of ageing was tested. Isolated mitochondria from the quadriceps muscle from normal, healthy, young (age 20+ years, n=12) and elderly (70+ years, n=11) humans were studied in respiratory experiments and the data expressed as activities of the muscle. In each group, the subjects exhibited a variation of physical activity but, on average, the groups were representative for their age with maximum O(2) consumption rate of 50+/-9 and 34+/-13 ml min(-1) kg(-1) (mean+/-SD), respectively. Thirteen different activities were assayed. alpha-Glycerophosphate oxidation was lower in the 70+ group (38%, P~0.001), as was the respiratory capacity for fatty acids (19%, P~0.03). The remaining eleven activities, including those of the central bioenergetic reactions, were not lower in the 70+ group. Pyruvate and alpha-ketoglutarate dehydrogenase activities (i.e. the tricarboxylic acid cycle turnover) and the respiratory chain activity could all account for ~14 mmol O(2) min(-1) kg(-1) muscle (37 degrees C). The capacity for aerobic ATP synthesis was ~35 mmol ATP min(-1) kg(-1). The mitochondrial capacities were far in excess of whole-body performance. They were related to physical activity, but not to age. The mitochondrial theory of ageing, which attributes the age-related decline of muscle performance to decreased mitochondrial function, is incompatible with these results.