NO and NO emissions from an Andisol maize field were studied. The experimental treatments were incorporation of urea into the plough layer at 250 kg N ha(-1) by two applications (U1250), band application of urea at a depth of 8 cm at 75 kg N ha(-1) plus incorporation of urea into the plough layer at 75 kg N ha(-1) (UB150), band application of polyolefin-coated urea at a depth of 5 cm at 150 kg N ha(-1) (CB150), and a control (without N application). N2O fluxes from U1250 and UB 150 peaked following the incorporation of supplementary fertilizer, and declined to the background level after that, while the NO flux from CB 150 was relatively low but remained at constant level until shortly after harvest. Accordingly, the total N2O emissions during the whole cultivation period from the three treatments were not significantly different. The fertilizer-derived N2O-N losses from U1250, UB150 and CB150 were 0.15%, 0.27% and 0.28% of the applied N, respectively. However, it was suggested that, due to the low plant N recovery, U1250 had a significantly larger potential for indirect N2O emission than the other treatments. On the other hand, NO emissions from U1250 and UB 150 were 12 times higher than that from CB 150, and the fertilizer-derived NO-N losses from the three treatments were 0.16%, 0.27% and 0.026% of the applied N, respectively. Significant NO fluxes were detected only when urea-N fertilizer was surface-applied and incorporated into plough-layer soil.