The aging process is associated with declining exercise abilities and unhealthy changes in the ratio of lean body mass to fat. Exercise, even at reduced levels, can ameliorate the physiological changes associated with aging. Exercise is well-known to protect against many chronic diseases and to extend life expectancy. Despite these benefits, willingness to engage in exercise – and physiological responses to exercise – vary widely in human populations. Unfortunately, long-term exercise is difficult and costly to measure in human cohorts. Furthermore, the physiological effects of aging in humans are confounded with important changes in lifestyle and environment that are often difficult to quantify. We used the common inbred mouse strain C57BL/6J to examine long-term patterns of exercise during aging, as well as the resulting physiological consequences of long-term exercise, in a well-controlled environment. Male (n=30) and female (n=30) mice that were already 1 year old (reflecting approximately 40 year old humans) were randomized to treatments of exercise (access to voluntary running wheels) or no exercise until they were 2 years old (reflecting approximately 70 year old humans). Body composition and metabolic traits were measured prior to, throughout, and at the end of the treatment period. Wheel running levels (distance, speed, duration) were greater in females compared to males and declined with age. Sex-dependent differences in body-composition trajectories were observed during the aging process. Long-term exercise significantly reduced body mass and percent body fat, while maintaining percent lean mass; however, the magnitude of these effects was also sex-dependent. This study provides robust evidence that long-term exercise can be used as a preventative measure against age-related weight gain and changes in body composition that have negative implications on health. Furthermore, we demonstrate that inbred mouse strains can be robustly used to characterize the effects of long-term exercise and factors (e.g. sex, age) modulating these effects. This will facilitate more detailed studies on the relationships between exercise and health in aging populations, including the potential impact of genetic predisposition.