The impacts of planted transgenic rice varieties on bacterial communities in paddy soils were monitored using both cultivation and molecular methods. The rice field plot consisted of eighteen subplots planted with two genetically modified (GM) rice and four non-GM rice plants in three replicates. Analysis with denaturing gradient gel electrophoresis (DGGE) of PCR-amplified 16S rRNA genes revealed that the bacterial community structures were quite similar to each other in a given month, suggesting that there were no significant differences in bacterial communities between GM and non-GM rice soils. The bacterial community structures appeared to be generally stable with the seasons, as shown by a slight variation of microbial population levels and DGGE banding patterns over the year. Comparison analysis of 16S rDNA clone libraries constructed from soil bacterial DNA showed that there were no significant differences between GM and non-GM soil libraries but revealed seasonal differences of phyla distribution between August and December. The composition profile of phospholipid fatty acids (PLFA) between GM and non-GM soils also was not significantly different to each other. When soil DNAs were analyzed with PCR by using primers for the bar gene, which was introduced into GM rice, positive DNA bands were found in October and December soils. However, no bar gene sequence was detected in PCR analysis with DNAs extracted from both cultured and uncultured soil bacterial fractions. The result of this study suggested that, in spite of seasonal variations of bacterial communities and persistence of the bar gene, the bacterial communities of the experimental rice field were not significantly affected by cultivation of GM rice varieties.