0023-6152

AA00697606

1434379

査読無

Root occupation of rhizobia inoculated on seed coat was known to be low because of their low competitiveness against indigenous rhizobia. This study was carried out to clarify the effect of inoculation methods and inoculum density of Bradyrhizobium japonicum USDA 110 on the production of soybean. Five experimental plots with three replications corresponded to: no inoculation (NI), seed coating inoculation at 10^5 cells seed–1 (SI5) and 10^7 cells seed^-1 (SI7), root zone inoculation at 1.7×10^3 cells g^1 dry soil (PI7) and 1.7×10^5 cells g^-1 dry soil (PI9). PI plots were plowed after treatment application. Our results indicated a significant higher occupation of serotype USDA 110 in SI5, SI7, and PI9 plots, compared to the other treatments. Their yield (g m^–2) was significantly increased. There was no yield increase above 105 cells seed^–1, and this density was considered most effective for seed inoculation. Furthermore, results of PI9 treatment indicated that inoculation of 1.7×10^5 cells g^–1 dry soil using BM2 (neutralized peat moss) was effective to compete with indigenous rhizobia for nodulation. This study revealed the significant yield increase in SI5 and PI9 treatments with lower inoculation concentrations of 10^5 cells seed^–1 and 1.7×10^5 cells g^–1 dry soil, respectively. Increased inoculum density above these levels did not increase seed yield. We concluded that it was possible to increase soybean yield by considering proper inoculum densities of efficient rhizobia and inoculation methods.