Irrigated rice fields in Uruguay are temporarily established on soils used as cattle pastures. Typically, 4 years of cattle pasture are alternated with 2 years of irrigated rice cultivation. Thus, oxic upland conditions are rotated with seasonally anoxic wetland conditions. Only the latter conditions are suitable for the production of CH4 from anaerobic degradation of organic matter. We studied soil from a permanent pasture as well as soils from different years of the pasture-rice rotation hypothesizing that activity and structure of the bacterial and archaeal communities involved in production of CH4 change systematically with the duration of either oxic or anoxic conditions. Soil samples were taken from drained fields, air-dried and used for the experiments. Indeed, methanogenic archaeal gene copy numbers (16S rRNA,mcrA) were lower in soil from the permanent pasture than from the pasture-rice alternation fields, but within the latter, there was no significant difference. Methane production started to accumulate after 16 days and 7 days of anoxic incubation in soil from the permanent pasture and the pasture-rice alternation fields respectively. Then, CH4 production rates were slightly higher in the soils used for pasture than for rice production. Analysis of δ13C in CH4, CO2 and acetate in the presence and absence of methyl fluoride, an inhibitor of aceticlastic methanogenesis, indicated that CH4 was mainly (58–75%) produced from acetate, except in the permanent pasture soil (42%). Terminal restriction fragment length polymorphism (T-RFLP) of archaeal 16S rRNA genes showed no difference among the soils from the pasture-rice alternation fields withMethanocellaceae and Methanosarcinaceaeas the main groups of methanogens, but in the permanent pasture soil,Methanocellaceaewere relatively less abundant. T-RFLP analysis of bacterial 16S rRNA genes allowed the distinction of permanent pasture and fields from the pasture-rice rotation, but nevertheless with a high similarity. Pyrosequencing of bacterial 16S rRNA genes generally revealedFirmicutesas the dominant bacterial phylum, followed byProteobacteria, Acidobacteria and Actinobacteria. We conclude that a stable methanogenic microbial community established once pastures have been turned into management by pasture-rice alternation despite the fact that 2 years of wetland conditions were followed by 4 years of upland conditions that were not suitable for CH4 production.