Abstract: A pot experiment in greenhouse was carried out showing that mean CH₄ flux of flooding treatment was 5.58 (1996) and 4.16 (1997) times higher than that of dry treatment. Soil regime in winter cropping also markedly affected seasonal variation patterns of soil Eh and CH₄ fluxes. Soil Eh of the flooding treatment was in the active range of methanogenic bacteria during the whole rice growing season, so CH₄ emission occurred throughout the entire period (except for the days when the soil was exposed for drying). For the soil Eh of the dry treatment, however, it took 42 days after rice transplanting to drop down to negative value and CH₄ emission was very low during the period.