Objective Low 25-hydroxyvitamin D (25OHD) levels may be associated with both sarcopenia (the age-related decline in muscle mass and function) and low physical activity (PA). Our objective was to describe prospective associations between 25OHD, muscle parameters, and PA in community-dwelling older adults. Design Prospective, population-based study with a mean follow-up of 2·6 ± 0·4 years. Patients Six hundred and eighty-six community-dwelling older adults (49% women; mean ± SD 62 ± 7 years old). Measurements Appendicular lean mass percentage (%ALM) and body fat assessed by Dual-energy X-ray Absorptiometry, leg strength by dynamometer, leg muscle quality (LMQ), PA assessed by pedometer, self-reported sun exposure by questionnaire, and serum 25OHD measured by radioimmunoassay. Results Participants with 25OHD ≤50 nm had lower mean %ALM, leg strength, LMQ and PA (all P < 0·05). As a continuous function, baseline 25OHD was a positive independent predictor of change in leg strength (β = 5·74 kg, 95% CI 0·65, 10·82) and LMQ (β = 0·49 kg/kg, 95% CI 0·17, 0·82). Also, change in 25OHD was positively predicted by baseline %ALM (β = 2·03 pm/p.a., 95% CI 0·44, 3·62) leg strength (β = 0·30 pm/p.a., 95% CI 0·06, 0·53), LMQ (β = 4·48 pm/p.a., 95% CI 0·36, 8·61) and PA (β = 2·63 pm/p.a., 95% CI 0·35, 4·92) after adjustment for sun exposure and body fat. Conclusions 25OHD may be important for the maintenance of muscle function, and higher skeletal muscle mass and function as well as general PA levels may also be beneficial for 25OHD status, in community-dwelling older adults.