The experiment investigated the nitrous oxide production under different C/N ratios during denitrification, taking nitrate and nitrite as electron acceptor respectively. Ethanol was selected as carbon source. The C/N ratios were 0, 1.2, 2.4, 3.5, 5.0 and 20 when nitrate was taken as electron acceptor and C/N ratios 0, 1.8, 2.4, 3.0, 4.3, 5.2, 6.6, 20.6 when electron acceptor was nitrite. The results indicated that: the optimum C/N ratio was 3.0 taking nitrite as electron acceptor and the N2O production was 0.044 mg x L(-1); the optimum C/N ratio was 5.0 taking nitrate as electron acceptor and the N2O production was 0.135 mg x L(-1) which was 3 times higher than that of nitrite as electron acceptor. Though the electron acceptor changed, the trend of N2O production was similar: when carbon source was badly insufficient, the production of N2O and denitrification rate were both quite small; the N2O production increased with the increasing of the quantity of carbon source; when the carbon source was excessive, the N2O production sharply raised. Consequently, compared to complete nitrification and denitrification, short-cut nitrification and denitrification could save 40% carbon source. Moreover, controlling C/N = 3 could reduce the production of N2O in short-cut nitrification.