A pot experiment was carried out to investigate the effect of antecedent water regime and rice straw application time on methane (CH4) emission from rice cultivation. The results showed that water regimes of the previous crops or fallow seasons and timing of rice straw application significantly influenced CH4 emissions during the wetland rice-cropping season. The mean CH4 flux of treatment with continuously flooded fallow (FFE) in the previous crop season was 6.06 and 5.15 times higher than that of treatment with dry fallow in 1996 and 1997, respectively. Applying rice straw just before soil was flooded for rice growth (DFL) significantly enhanced CH4 flux by 385.96% in 1996 and 1017.46% in 1997 compared with rice straw application just before the previous crop season (DFE). Antecedent water regime and rice straw application time also markedly affected temporal variation patterns of CH4 fluxes and soil redox potential (Eh) during the period of time from soil being flooded to rice harvest. Considerable CH4 was emitted into the atmosphere from flooding to the early stage of the rice-growing season, and 4 emission maxima occurred several days after flooding and in the early, mid, and late stages of the rice-growing period for treatments DFL and FFE. However, for treatment DFE, there was almost no CH4 emission until the middle of the rice-growing season, and CH4 fluxes in the mid and late stages of the rice-growing period were also very low. Soil Eh values of treatments FFE and DFL were compatible for methanogenic bacteria almost every day after flooding. However, it took 65 days of flooding before the Eh of treatment DFE dropped to the same level. Differences in the temporal variation patterns of soil Eh were the main reason why variation patterns of CH4 fluxes were different.