Atmospheric biological nitrogen fixation (BNF) by cowpea (Vigna unguiculata) and groundnut (Arachis hypogea) was evaluated using a 2-year (2000–2001) experiment with different fertilizer treatments. The 15N isotopic dilution method with a nonfixing cowpea as test reference crop was used. The effects of the two legumes on soil N availability and succeeding sorghum (Sorghum bicolor) yields were measured. Groundnut was found to fix 8 to 23 kg N ha-1 and the percentage of N derived from the atmosphere varied from 27 to 34%. Cowpea fixed 50 to 115 kg N ha?1 and the percentage of N derived from the atmosphere varied from 52 to 56%. Compared to mineral NPK fertilizer alone, legumes fixed more N from the atmosphere when dolomite or manure was associated with mineral fertilizers. Compared to soluble phosphate, phosphate rock increased BNF by cowpea. Significant correlation (p<0.05, R 2=0.94) was observed between total N yields of legumes and total N derived from the atmosphere. Compared to monocropping of sorghum, the soils of cowpea–sorghum and groundnut–sorghum rotations increased soil mineral N from 15 and 22 kg N ha?1, respectively. Cowpea–sorghum and groundnut–sorghum rotations doubled N uptake and increased succeeding sorghum yields by 290 and 310%, respectively. Results suggested that, despite their ability to fix atmospheric nitrogen, N containing fertilizers (NPK) are recommended for the two legumes. The applications of NPK associated with dolomite or cattle manure or NK fertilizer associated with phosphate rock were the better recommendations that improved BNF, legumes, and succeeding sorghum yields.