Globally identifying mitigation options for the emission of reactive N gases from agricultural soils is a research priority. We investigated the effect of urea size and placement depth on sources and emissions of N gases from a Cambisol cropped to spring wheat (Triticum aestivum L.). In Exp. 1, wheat received either prilled urea (PU) mixed within the soil, urea super granule (USG; diam. 10.1 mm) point-placed at a soil-depth of 7.5 cm, or no N fertilizer. In Exp. 2, wheat received either USG (diam. 10.2 mm) point-placed at 2.5-, 5.0-, and 7.5-cm soil depths, or no N fertilizer. In both experiments, maximum peaks for nitrous oxide (N2O) fluxes and nitrification were delayed by 2 to 3 wk in the USG compared with the PU treatment. The added N-15-urea lost as N-15-N2O over 116 d was only 0.01% for both PU and USG treatments in Exp. 1. This loss for USGs was higher in Exp. 2 (0.02-0.15%) measured over 70 d, mainly related to higher moisture-induced denitrification. Temporal N2O fluxes were significantly related to changes in soil NO3--N, water-filled pore space and NH4+-N (R-2 = 0.50, P < 0.05). However, the previous predictive model of Khalil et al. (2006) could best estimate its cumulative fluxes over time. The relative losses of ammonia (0.07-1.17%) and nitrogen oxides (0.19-1.54%) measured in Exp. 2 over 43 d decreased with increasing depths of USG placement. The USG point-placed at the 5.0- and 7.5-cm depths decreased the pooled gaseous N losses by 35 and 77%, respectively, over the shallower placement. The N-15 results imply that soil N could be the major source of N2O emissions (79-97%). Field studies are suggested to validate our findings that the deeper placement of USG can decrease N emissions under arable cropping.