Nitrogen fixation by annual forage legumes and its contribution to succeeding wheat in the Ethiopian highlands

Soil fertility is declining in most agro-ecosystems in sub-Saharan Africa, and incorporation of forage legumes into production systems to utilize the nitrogen fixed by the legumes could alleviate the problem, if efficient nit rogen-fixing legumes are used. The amounts of nitrogen fixed by Lablab, Med icago, Trifolium, and Vicia species and their contribution to the following wheat crop were estimated in field experiments on an Alfisol at Debre Zeit in the Ethiopian highlands. The amounts of nitrogen (N) fixed ranged from 40 kg N ha(-1) for T. steudneri to 215 kg N ha(-1) for L. purpureus. The in crease in grain yields of wheat following the legumes ranged from 16% for T . steudneri to 71% for M. tranculata where no N fertilizer was applied to t he wheat. Additional N fertilizer applied to wheat at 60 kg N ha(-1) had no significant effects on wheat grain or straw DM and N yields. In another ex periment, eight lablab treatments consisting of factorial combinations of t wo cultivars (Rongai and Highworth), two Rhizobium inoculation treatments ( inoculated and uninoculated) and two times of harvest (for hay at 50% flowe ring and for seed at seed maturity), were compared on lablab forage product ion and N yield, and residual effects on two succeeding wheat crops. Inocul ation had no significant effects on nodulation, shoot DM or N yields. Ronga i had significantly higher shoot DM and N yields than Highworth. Lablab har vested at flowering had significantly higher shoot DM and N yields than lab lab harvested at seed maturity. Grain yields of the first wheat crop follow ing the various lablab crops were 93-125% higher than grain yields of the w heat following wheat (continuous wheat) where no N fertilizer was applied. Therefore, lablab is a potential forage crop for incorporation into cereal production systems to improve feed quality and to reduce dependence on N fe rtilizers for cereal production.