This study examined the association between three types of physical activities (PA) and depression, and the relationship between PA and later mortality. Previous studies rarely assessed these associations in one single study in randomly selected population samples. Few studies have assessed these relations by adjusting the covariate of friend-relative care. Participants consisted of 624 noninstitutionalized elders (mean age = 77.35) from the Americans' Changing Lives Longitudinal Study. Depression was measured using the Center for Epidemiological Studies Depression Scale. Logistic regression estimated the risk of depression as a function of the three individual PA (gardening, walking, and sport). A separate set of analyses predicted the risk of mortality (six years later) as a function of PA. Each 1-standard-unit increase on the physical inactivity scale significantly predicted adjusted 29%, 30%, and 33% increased risk of depression for gardening, walking, and sport, respectively; and each unit also predicted 48% and 72% increased risk of mortality, for gardening and walking, respectively. Active exercise was not able to predict mortality (p = 0.42). The study concluded that elders with less PA had a higher chance of getting depression. Unlike gardening and walking, rigorous exercise did not significantly predict mortality. This indicated that extra care is needed for elders when they engage in intense exercise. The relationships of social support from friends and relatives with depression and mortality were also discussed. (Contains 3 tables.)