I explore the relationship between active-duty military service and self-reported health measured at age 40. Based on selectivity, veterans of active-duty service might be expected to have better health than civilians. Using data taken from the NLSY-79, I show that this is not the case. Although veterans of reserve-duty service, and nonveterans who passed the military’s physical exam for entrance into the military report better physical health, active-duty veterans do not. The lower than expected self-reported health of active-duty veterans cannot be explained by differences on confounding variables such as income, education, and marital status. In addition, the lower physical health of these veterans cannot be explained by differences in health-related behaviors such as excessive alcohol use, cigarette smoking, and body mass index.