The air-surface exchange of mercury (Hg) was measured, using a dynamic polycarbonate flux chamber, for soils with low or "background" Hg concentrations (<0.1 mg/kg) at eleven locations across the contiguous United States. Sampling locations included agricultural, desert, grassland, mixed and pine forest ecosystems (n=1326 soil flux measurements at 46 individual sites). An overall soil Hg flux of 0.9+/-0.2 ng/m2/h for these background soils was obtained by averaging the means for the different locations. Soil Hg fluxes were significantly lower in dark conditions than in the light for all but the grassland sites. Mean inlet air Hg concentrations were 1.0+/-0.1 ng/m3 in the dark and 1.3+/-0.2 ng/m3 in the light. Soil temperature inside and outside of the chamber, air temperature, relative humidity, and irradiance were measured concurrently with soil Hg flux. Soil-air Hg exchange was weakly predicted by environmental variables (R2 from 0.07 to 0.52). For a single location, flux was better correlated with soil moisture than other measured environmental parameters, suggesting that soil moisture might be an important driver for Hg emissions from background soils. In addition, based on data collected we suggest some quality control measures for use of Tekran 2537A analyzers when measuring low mercury fluxes. Using basic scaling procedures, we roughly estimate that natural emissions from soils in the contiguous U.S. release approximately 100 Mg/yr of Hg to the atmosphere.