『Abstract
　Anaerobic incubations of upland and wetland temperate forest soils from the same watershed were conducted under different moisture and temperature conditions. Rates of nitrous oxide (N2O) production by denitrification of nitrate (NO3^-) and the stable isotopic composition of the N2O (δ¹⁵N, δ¹⁸O) were measured. In all soils, N2O production increased with elevated temperature and soil moisture. At each temperature and moisture level, the rate of N2O production in the wetland soil was greater than in the upland soil. The ¹⁵N isotope effect (ε)(product-substrate) ranged from -20‰ to -29‰. These results are consistent with other published estimates of ¹⁵N fractionation from both single species culture experiments and soil incubation studies from different ecosystems.
　A series of incubations were conducted with ¹⁸O-enriched water (H2O) to determine if significant oxygen exchange (O-exchange) occurred between H2O and N2O precursors during denitrification. The exchange of H2O-O with nitrite (NO2^-) and/or nitric oxide (NO) oxygen has been documented in single organism culture studies but has not been demonstrated in soils prior to this study. The fraction of N2O-O derived from H2O-O was confined to a strikingly narrow range that differed between soil types. H2O-O incorporation into N2O produced from upland and wetland soils was 86％ to 94％ and 64％ to 70％, respectively. Neither the temperature, soil moisture, nor the rate of N2O production influenced the magnitude of O-exchange. With the exception of one treatment, the net ¹⁸O isotope effect (εnet)(product-substrate) ranged from +37‰ to +43‰.
　Most previous studies that have reported ¹⁸O isotope effects for denitrification of NO3^- to N2O have failed to account for the effect of oxygen exchange with H2O. When high amounts of O-exchange occur after fractionation during reductive O-loss, the ¹⁸O-enrichment is effectively lost or diminished and δ¹⁸O-N2O values will be largely dictated by δ¹⁸O-H2O values and subsequent fractionation. The process and extent of O-exchange, combined with the magnitude of oxygen isotope fractionation at each reduction step, appear to be the dominant controls on the observed oxygen isotope effect. In these experiments, significant oxygen isotope fractionation was observed to occur after the majority of water O-exchange. Due to the importance of O-exchange, the net oxygen isotope effect for N2O production in soils can only be determined using ¹⁸O-H2O addition experiments with δ¹⁸O-H2O close to natural abundance.
　The results of this study support the continued use of δ¹⁵N-N2O analysis to fingerprint N2O produced from the denitrification of NO3^-. The utilization of ¹⁸O/¹⁶O ratios of N2O to study N2O production pathways in soil environments is complicated by oxygen exchange with water, which is not usually quantified in field studies. The oxygen isotope fractionation observed in this study was confined to a narrow range, and there was a clear difference in water O-exchange between soil types regardless of temperature, soil moisture, and N2O production rate. This suggests that ¹⁸O/¹⁶O ratios of N2O may be useful in characterizing the actively denitrifying microbial community.』

1. Introduction
2. Materials and methods
　2.1. Soil collection and processing
　2.2. Geochemical characterization of soils
　2.3. Soil moisture content
　2.4. Preparation and measurement of ¹⁸O-H2O of the incubation waters
　2.5. Incubation design and protocol
　2.6. N2O concentration and mass spectrometric analysis
3. Results
　3.1. Soil parameters
　3.2. Net N2O production rates
　3.3. Stable isotope abundances of the emitted N2O
4. Discussion
5. Summary and implications
Acknowledgments
References