Although many studies report that exposure to air pollution harms health, few have examined associations between pollution sources and health outcomes. We hypothesized that pollution originating in different locations has different associations with heart rate variability (HRV) among 497 men from the Normative Aging Study in Boston, Massachusetts. We identified the paths that air masses traveled ('back-trajectories') before arriving in Boston on the days the men were examined. Next, we classified these trajectories into six clusters. We examined whether the association of measured air pollutants with HRV (standard deviation of normal-to-normal intervals, high-frequency power (HF) and low-frequency power (LF), and LF/HF ratio) differed by cluster. We also examined whether the clusters alone (not considering air pollution measurements) showed different associations with HRV. The effects of black carbon (BC) on all HRV measures were strongest on days with southwest trajectories. Subjects who were examined on days where air parcels came from west had the strongest associations with ozone. All particle pollutants (particulate matter <2.5 microm in aerodynamic diameter (PM2.5), BC, and sulfates) were associated with increased LF/HF ratio on days with relatively short trajectories, which are related to local, slow-moving air masses. We also observed significant increases in LF/HF in days where air came from the northwest and west, compared to north trajectory days. Health effects associated with exposure to air pollution can be evaluated using pollutant concentrations as well as aspects of the pollution mixture captured by identifying locations where air masses originate. Independent effects of both these indicators of pollution exposure were seen on cardiac autonomic function.