Response of common bean lines to inoculation: comparison between the Rhizobium tropici CIAT899 and the native Rhizobium etli 12a3 and their persistence in Tunisian soils

Since Phaseolus vulgaris (L) is poorly nodulated in all regions of Tunisia where this crop is grown, the response of common-bean lines CocoT and Flamingo to inoculation with reference Rhizobium tropici CIAT 899 or native rhizobia, namely Sinorhizobium fredii 1a6, Rhizobium etli 12a3, and Rhizobium gallicum 8a3, was studied in a field station. Since R. etli 12a3 was found to be the most effective native rhizobium, it was subsequently compared with R. tropici CIAT 899 in a broader study in two stations over 3 years. A significant interaction between bean and rhizobia was observed for nodule number, shoot dry weight, grain yield, and contents of nitrogen and chlorophyll. The native rhizobia was more efficient than CIAT899 for Flamingo, though not for CocoT. The Enzyme-linked immunosorbent assay technique was used with polyclonal antibody to assess the occupancy in nodule and persistence in soil of the inoculated rhizobia. For both stations the nodule occupancy was 100% during the first year for each rhizobium, but during the next 2 years, between 7 and 15% of nodules were formed by the rhizobia inoculated in the neighboring plot. It is concluded that the first-year inoculation is sufficient to maintain an adequate rate of nodulation during three growth cycles, and that the native R etli can be recommended for the common-bean inoculation in similar soils of Tunisia.