Nitrous oxide, a greenhouse gas, is mainly emitted from soils during the denitrification process. Nitrogen stable-isotope investigations can help to characterise the N(2)O source and N(2)O production mechanisms. The stable-isotope approach is increasingly used with (15)N natural abundance or relatively low (15)N enrichment levels and requires a good knowledge of the isotopic fractionation effect inherent to this biological mechanism. This paper reports the measurement of the net and instantaneous isotopic fractionation factor (alpha(i)(s/p)) during the denitrification of NO(3)(-) to N(2)O over a range of (15)N substrate enrichments (0.37 to 1.00 atom% (15)N). At natural abundance level, the isotopic fractionation effect reported falls well within the range of data previously observed. For (15)N-enriched substrate, the value of alpha(i)(s/p) was not constant and decreased from 1.024 to 1.013, as a direct function of the isotopic enrichment of the labelled nitrate added. However, for enrichment greater than 0.6 atom% (15)N, the value of alpha(i)(s/p) seems to be independent of substrate isotopic enrichment. These results suggest that for isotopic experiments applied to N(2)O emissions, the use of low (15)N-enriched tracers around 1.00 atom% (15)N is valid. At this enrichment level, the isotopic effect appears negligible in comparison with the enrichment of the substrate. Copyright (C) 2007 John Wiley & Sons, Ltd.