Does social capital, resources embedded in social relationships, influence health? This research examines whether social capital impacts depressive symptoms and overall perceived health status over and above the effects of social support. Our analyses use unique data from the Taiwan Social Change Survey collected in 1997, and measures social capital and social support through two network instruments (the position generator and the name generator). Results replicate the effects of social support, as measured through the name generator, on both outcomes. Results show that social capital, as measured through the position generator, has direct effects on both outcomes net of social support, while social support is a stronger predictor than social capital. This research indicates that social capital contributes to health beyond and distinct from the contribution of social support, and it suggests that social capital and social support are two independent relationship-based causes of disease which require different instruments of measurement. (Contains 6 tables and 15 notes.)