Isometric quadriceps strength was remeasured in 14 healthy survivors of a group of elderly people first studied 8 years previously. There were 4 men (median age 81 years, range 79 to 84) and 10 women (82 years, range 79 to 89). They were selected for their health, not their levels of physical activity. Nevertheless, they were active when first studied and, with 1 exception, had maintained or increased their levels of physical activity. Isometric quadriceps strength was well preserved; the median change in the strength of the stronger quadriceps was only -0.3% per annum (95% confidence interval = -1.4 to +0.8).