Does social capital, resources embedded in social networks, influence health? My dissertation examines whether social capital directly impacts depression, and how it interplays with other established structural risk factors linked to depression. I analyze unique data from the thematic research project "Social Capital: Its Origins and Consequences," collected in 2004-5 in the United States. I measure social capital through one recently developed network instrument, the position generator. I use structural equation modeling to test the direct, mediating, and moderating effects of social capital on depressive symptoms. I also use the instrumental variable method to verify the causal order in the relationship between social capital and depression. Results show that social capital is associated with the level of depression in four ways. Social capital is associated with lower levels of depressive symptoms net of other variables. Part of the effect of social capital on depressive symptoms is indirect through subjective social status. Social capital mediates the associations of age, gender, being black (versus being white), marital status, education, occupation, annual family income, and social integration with depression. Social capital also interacts with gender, being black (versus being white), education, annual family income, and social integration. This research indicates that social capital is an important social antecedent of disease and illness.