Nitrogen fixation by trees in relation to soil nitrogen economy

The N-2-fixing potential (NFP) (i.e. the amount of fixed N-2 in a constraint-free environment) of N-2-fixing trees (NFTs) varies with the genotype. The NFP can be higher than 30-50 g N-2 fixed tree(-1) year(-1) in the most active species, be they leguminous trees such as Albizia lebbeck, Gliricidia sepium and Leucaena leucocephala, or actinorhizal trees such as Casuarina equisetifolia. The actual amount of nitrogen fixed (ANF) (i.e. the amount of N-2 fixed in the field) is lower than the NFP or even nil because of various constraints, especially drought, nutrient deficiencies, excess of available N and pathogenic nematodes. As tree litters are mineralized, the amount of available N in the soil increases with time, this process leading to the cessation of N-2 fixation in aging plantations. When the mineralization rate is slowed down or inhibited, N-2 fixation can continue. NFTs improve the N status of soils, but the transfer of fixed N to associated plants is not always ensured. Three main approaches are appropriate to increase N-2 fixation: clonal selection of trees combined with vegetative propagation, inoculation with effective rhizobium or Frankia strains, and proper fertilization (especially P). In the absence of major environmental constraints, a positive response to inoculation is expected only when specific (non-promiscuous) NFTs are grown in sites where the density of compatible rhizobia is low or nil. The potentialities of NFTs are far from being fully exploited. Further investigations are proposed and the economics of NET management is briefly discussed.