This paper reports a net snow-to-air Hg transfer from the remote and temperate Experimental Lakes area (Ontario, Canada). More than 40% loss of total Hg concentration was observed in surface snow within 24 hours of its deposition on the ground. Stratigraphic profiles of total Hg in various snowpacks demonstrated a systematic decrease in total Hg concentration with snow depth pointing to snow-to-air Hg transfer and not snow-to-ground transfer. These results confirm observations made in a suburban area (Sainte-Foy, Quebec, Canada) receiving higher atmospheric deposition of Hg perhaps differently associated chemically. The occurrence of this phenomenon is therefore extended geographically to include snow from pristine regions. It is hypothesized that the loss of Hg is caused by a sunlight-initiated Hg(II) reduction in snow and subsequent gas transfer of Hg0 to the atmosphere. Polychromatic action spectra demonstrated that Hg(II) reduction in snow was mostly mediated by UV-B irradiation and not visible, or UV-A wavelengths. In addition to Hg(II) reduction in snow, we observed Hg0 oxidation in snow samples spiked with Cl-. Hg0 oxidation could limit the potential for Hg loss from snowpacks from coastal polar or subpolar regions (where snow often contains high chloride levels) by competing with Hg(II) reduction and slowing the snow-to-air Hg transfer. However, in regions under minimal marine influences, watershed budgets of Hg should consider the possibility of Hg loss from snow with time of deposition on the ground. Also, snow core studies should consider historic sunlight irradiation if extrapolation from snow cores is desired to estimate past ambient Hg levels.