In the current study, we compared muscle morphology in three advanced aging cohorts that differed in physical function, including a unique cohort of lifelong endurance athletes. Biopsies from the vastus lateralis muscle of seven lifelong endurance athletes (EAs) aged 82-92 yr, and 19 subjects from the Uppsala Longitudinal Study of Adult Men (ULSAM) aged 87-91 yr were analyzed. ULSAM subjects were divided into high- (n = 9, HF) and low- (n = 10, LF) function groups based on strength and physical function tests. The analysis included general morphology, fiber type and cross-sectional area, capillarization, deficient cytochrome c oxidase (COX) activity, number of myonuclei and satellite cells, and markers of regeneration and denervation. Fibers with central nuclei and/or nuclear clumps were observed in all groups. EA differed from LF and HF by having a higher proportion of type I fibers, 52% more capillaries in relation to fiber area, fewer COX-negative fibers, and less variation in fiber sizes (all P < 0.05). There were no differences between the groups in the number of myonuclei and satellite cells per fiber, and no significant differences between LF and HF (P > 0.05). In conclusion, signs of aging were evident in the muscle morphology of all groups, but neither endurance training status nor physical function influenced signs of regeneration and denervation processes. Lifelong endurance training, but not higher physical function, was associated with higher muscle oxidative capacity, even beyond the age of 80.NEW & NOTEWORTHY Here we show that lifelong endurance training, but not physical function, is associated with higher muscle oxidative capacity, even beyond the age of 80 yr. Neither endurance training status nor physical function was significantly associated with satellite cells or markers of regeneration and denervation in muscle biopsies from these very old men.