Results of laboratory experiments show that 14C-labeled methane added to soil was consumed faster than atmospheric 12C methane. This implies a source of methane, presumably through methanogenesis, in a soil that is a net consumer of atmospheric methane. The soil was well-drained forest soil from Ispra, Italy. An undisturbed sample was taken with a steel corer and incubated under oxic conditions in a jar. Headspace samples were taken at time intervals and analyzed for total methane by gas chromatography and analyzed for 14C methane by liquid scintillation counting. Fluxes calculated from the decreasing headspace mixing ratios were, for example, -6.5 and -7.1 &mgr;molm-2hr-1 for 12C methane and 14C methane, respectively. A simple model is considered which reproduces reasonably well the observed mixing ratios as function of time. ¿ 1998 American Geophysical Union