NODULATION AND ESTIMATION OF SYMBIOTIC NITROGEN-FIXATION BY HERBACEOUS AND SHRUB LEGUMES IN GUINEA-SAVANNA IN NIGERIA

Twelve herbaceous and shrub legume species were grown in pot and field experiments in five sites representing three agroecological zones in moist savanna in Nigeria. The objectives were to: (1) assess natural n odulation of the legumes and characterize their indigenous rhizobia, ( 2) determine their need for rhizobia inoculation and (3) estimate the amount of N-2 fixed by each of these legumes. At 4 weeks after plantin g (WAP), Crotolaria verucosa was not nodulated at any of the sites whi le Centrosema pascuorum had the highest number of nodules in all sites . At 8 WAP, all legumes were nodulated, with Mucuna pruriens having th e least number of nodules and Stylosanthes hamata the highest. The num ber of nodules, however, was inversely correlated to the mass of nodul es. Significant differences in nodulation of the legume species grown in the field also occurred between and within sites. Mucuna pruriens a nd Lablab purpureus produced more shoot and nodule biomass than the ot her legumes in all sites. Growth of most of these legumes responded to fertilizer application, except for C. verrucosa and Aeschynomene hist rix. Except for C. verrucosa, average proportion of N-2 fixed was abou t 80% and this was reduced by about 20% with N fertilizer application. The majority of rhizobia isolates (60%) were slow growing, belonging to the Bradyrhizobia spp. group. Selected rhizobia isolates evaluated on Cajanus cajan, C. pascuorum, M. pruriens and Psophocarpus palustris varied from ineffective to highly effective in Leonard jar conditions . However, only growth of M. pruriens responded to inoculation in pott ed soils, whereas it was lower than that obtained with N fertilizer ap plication. This indicated the need to screen more rhizobia in order to improve N-2 fixation and growth of legume species such as M. pruriens when it is introduced in soils deficient in N.