Oxygen isotopic analysis of nitrous oxide (N₂O) is increasingly used to study its sources and production pathways. However, O exchange between ¹⁸O-enriched H₂O and (de)nitrification intermediates may disturb the ¹⁸O isotopic signal of N₂O and complicate data interpretation in such methodology (Kool et al., Rap. Comm. Mass Spec. submitted). We set up a series of experiments to identify and quantify such O exchange in soil. For this purpose, we employed a novel ¹⁵N and ¹⁸O labeling approach: we incubated soils with different combinations of ¹⁵N labeled NH₄⁺ and NO₃^-, and ¹⁸O labeled H₂O and ¹⁸O labeled NO₃^-. O exchange appeared to be present and significant in all three soils studied. The extent of O exchange was different for the three soils, but in all soils appeared to stabilize at an equilibrium. We conclude that though the extent of O exchange differs between soil types, it is significant in most soil ecosystems. It will therefore interfere with interpretation of ¹⁸O-N₂O analyses for N₂O source determination. O stable isotopes can provide a useful tool in studying N transformations, but analyses should yet be used with caution while the presence and extent of O exchange are still unclear.