The severity and increase of the Imperata cylindrica constraint as a weed, the decline of the traditional fallow systems as a means of soil fertility management and the lack of inorganic fertilizer appear to have created opportunities for adoption of mucuna (Mucuna pruriens) technology by smallholder farmers in some areas in the derived savanna of West Africa. What is not known, however, is the extent to which the establishment and N contribution of mucuna in these areas depend on symbiotic properties such as effective nodulation and mycorrhizal infection. Short term surveys carried out in 34 farmer's arable fields located in four different sites in the derived savanna, southern Benin, West Africa, together with results of greenhouse and field experiments showed that mycorrhizal infection rate of mucuma ranged from 2 to 31% and correlated positively with nodulation and shoot dry matter production. Nodulation occurred in 79% of the fields with numbers of nodules ranging from 0 to 135 plant−1. Mucuna responded both to inoculation and N fertilizer in degraded soils but growth response depended on the rhizobia strains and mucuna varieties. Mucuna accumulated in 12 weeks about 313 kg N ha−1 as either a sole crop or 166 kg N ha−1 when mixed/intercropped with maize, respectively. Across all cropping systems it derived an average of 70% of its N from atmospheric N2 (estimates made by the 15N isotope dilution method), representing 167 kg N ha−1 per 12 weeks in the field. Mucuna interplanted with maize obtained a greater proportion of its nitrogen (74%) from fixation than did mucuna grown alone (66%) suggesting that competition for soil N influences the proportion of nitrogen fixed by mucuna. The total amount of N2 fixed per hectare was, however, reduced significantly by intercropping mucuna with maize. A preceding mucuna crop provided a maize yield equivalent to 120 kg N kg ha−1 of inorganic N fertilizer.