Impacts of the projected increase in atmospheric CO2 on other biogeochemical cycles are uncertain. In a two-year study, Phillips et al. (2001) reported a 16 to 30% decrease in net consumption of atmospheric CH 4 by soils in CO2-enriched plots in a temperate loblolly pine (Pinus taeda) forest. Consumption by upland soils accounts for ∼30 Tg CH4 y-1 and is the only terrestrial sink for atmospheric CH4, which is a greenhouse gas with radiative forcing second only to CO2. However, it is uncertain whether decreased atmospheric CH4 consumption represents a transient or sustained response of forest-soil systems to elevated CO2. This research focused on field observations aimed at investigating the strength and persistence of reduced atmospheric CH4 consumption by temperate forest soils under elevated CO2 at the same study site. It further investigates the causes of this response by CH4 oxidizing and producing communities through field and laboratory experiments. Rates of soil-atmosphere CH4 exchange were repeatedly measured over 3 y from permanently established sampling sites at the Free Air Carbon Dioxide (FACE) site in the Duke Forest, where CO2-enriched plots of a loblolly pine forest are maintained at approximately 200 mL L -1 above ambient concentrations (380 mL L-1), while control plots are exposed to ambient atmospheres. Reduced net atmospheric CH4 consumption persisted in CO2-enriched plots, showing annual declines of 19, 10 and 8% relative to control plots. This study and previous work give a nearly continuous 8 y record of reduced net atmospheric CH4 consumption in CO2-enriched plots that suggests this is likely a sustained negative feedback to increasing atmospheric CO 2. Causitive factors for the observed decrease in net CH4 consumption under elevated CO2 were difficult to identify because of high spatial and temporal variability in microbial activity and limited ability to collect soil samples. However, higher soil moisture and increased incidence and rates of CH4 production in CO2-enriched plots, along with transient inhibition by plant exudates and low overall soil diffusivity, begin to explain reduced rates of CH4 consumption and increased rates of CH4 production that result in long-term reduction in net CH 4 consumption in these soils.