We compared the composition of phospholipid fatty acids (PLFA) to assess the microbial community structure in the soil and rhizosphere community of non-transgenic watermelons and transgenic watermelons in Miryang farmlands in Korea during the spring and summer of 2005. The PLFA data were seasonally examined for the number of PLFA to determine whether there is any difference in the microbial community in soils from two types of watermelons, non-transgenic and transgenic. We identified 78 PLFAs from the rhizosphere samples of the two types of watermelons. We found eight different PLFAs for the type of plants and sixteen PLFAs for the interaction of plant type and season. The PLFA data were analyzed by analysis of variance separated by plant type (P<0.0085), season (P<0.0154), and the plant type${ imes}$season interaction (P<0.1595). Non-parametric multidimensional scaling (NMS showed a small apparent difference but multi-response permutation procedures (MRPP) confirmed that there was no difference in microbial community structure for soils of both plant types. Conclusively, there was no significant adverse effect of transgenic watermelon on bacterial and fungal relative abundance as measured by PLFA. We could reject our hypothesis that there might be an adverse effect from transgenic watermelon with our statistical results. Therefore, we can suggest the use of this PLFA methodology to examine the adverse effects of transgenic plants on the soil microbial community.