The free radical theory of aging proposes a causal relationship between reactive oxygen species (ROS) and aging. While it is clear that oxidative damage increases with age, its role in the aging process is uncertain. Testing the free radical theory of aging requires experimentally manipulating ROS production or detoxification and examining the resulting effects on lifespan. In this review, we examine the relationship between ROS and aging in the genetic model organism Caenorhabditis elegans, summarizing experiments using long-lived mutants, mutants with altered mitochondrial function, mutants with decreased antioxidant defenses, worms treated with antioxidant compounds, and worms exposed to different environmental conditions. While there is frequently a negative correlation between oxidative damage and lifespan, there are many examples in which they are uncoupled. Neither is resistance to oxidative stress sufficient for a long life nor are all long-lived mutants more resistant to oxidative stress. Similarly, sensitivity to oxidative stress does not necessarily shorten lifespan and is in fact compatible with long life. Overall, the data in C. elegans indicate that oxidative damage can be dissociated from aging in experimental situations.