Nitrogen fertilizer is most widely used, of which one third is lost through emission of greenhouse gasses and leaching. Moreover, fertilizer application is expensive and difficult to afford by resource poor farmers. Thus, to overcome these challenges, alternative genetic resources to supply plant-available nitrogen have become absolutely essential. Nitrogen is unique among the other essential elements because N2 from the atmosphere can be fixed by biological nitrogen fixation (BNF), a process carried out by prokaryotes that possess the enzyme nitrogenase. Cultivated rice (Oryza sativa) and most other cereals appear to lack the genetic predisposition to support high rates of nitrogen fixation. Interestingly, an aquatic wild species of Oryza was recently found by us to support high BNF rates lying in the range of many legumes. The exploitation of this capacity coupled to application of modern genomic and breeding techniques can lead to improved cultivated plants that use nutrients more efficiently. In the attempt to initiate transfer of the high BNF potential in wild rice germplasm by wide hybridization into O. sativa, interspecific hybrids were generated at IRRI and screened by us for nitrogenase (nifH) mRNA expression. Moreover, for the first time a large EST-library of O. longistaminata roots, active in N2-fixation, was generated and yielded 72 179 uni-EST sequences with an average length of 263 bp. This is an important genomic resource for a species for which the genome has not been sequenced yet. In microcosms of O. longistaminata, root-associated N2-fixation measured by nifH-mRNA expression is drastically reduced by ammonium-nitrate fertilizer application. An SSH library was constructed from mRNA of these roots in order to identify genes expressed differentially in roots supported by N2-fixation or combined nitrogen. These are important first steps towards identifying plant genes putatively involved in the high BNF potential of wild rice in order to elucidate the mechanisms of the process.