Several empirical studies have suggested that neighborhood characteristics influence health, with most studies having focused on neighborhood deprivation or aspects of the physical environment, such as services and amenities. However, such physical characteristics are not the only features of neighborhoods that potentially affect health. Neighborhoods also matter because of the nature of their social organization. This study examined social capital as a potential neighborhood characteristic influencing health. Using a cross-sectional study design which linked counts of death for persons 45-64 years by race and sex to neighborhood indicators of social capital and poverty for 342 Chicago neighborhoods in the USA, we tested the ecological association between neighborhood-level social capital and mortality rates, taking advantage of the community survey data collected as part of the Project on Human Development in Chicago Neighborhoods. We estimated a hierarchical generalized linear model to examine the association of race and sex specific mortality rates to social capital. Overall, neighborhood social capital-as measured by reciprocity, trust, and civic participation-was associated with lower neighborhood death rates, after adjustment for neighborhood material deprivation. Specifically, higher levels of neighborhood social capital were associated with lower neighborhood death rates for total mortality as well as death from heart disease and "other" causes for White men and women and, to a less consistent extent, for Blacks. However, there was no association between social capital and cancer mortality. Although, the findings from this study extend the state-level findings linking social capital to health to the level of neighborhoods, much work remains to be carried out before social capital can be widely applied to improve population health, including establishing standards of measurement, and exploring the potential "downsides" of social capital.