Treatment of liquid swine manure (LSM) offers opportunities to improve manure nutrient management. However, N2O fluxes and cumulative emissions resulting from application of treated LSM are not well documented. Nitrous oxide emissions were monitored following band-incorporation of 100 kg N ha(-1) of either mineral fertilizer, raw LSM, or four pretreated LSMs (anaerobic digestion; anaerobic digestion + flocculation: filtration; decantation) at the four-leaf stage of corn (Zea mays L.). In a clay soil, a larger proportion of applied N was lost as N2O with the mineral fertilizer (average of 6.6%) than with LSMs (3.1-5.0%), whereas in a loam soil, the proportion of applied N lost as N2O was lower with the mineral fertilizer (average of 0.4%) than with LSMs (1.2-2.4%). Emissions were related to soil NO3 intensity in the clay soil, whereas they were related to water-extractable organic C in the loam soil. This suggests that N2O production was N limited in the clay soil and C limited in the loam soil, and would explain the interaction found between N sources and soil type. The large N2O emission coefficients measured in many treatments, and the contradicting responses among N sources depending on soil type, indicate that (i) the Intergovernmental Panel on Climate Change (IPCC) default value (1%) may seriously underestimate N2O emissions from fine-textured soils where fertilizer N and manure are band-incorporated, and (ii) site-specific factors, such as drainage conditions and soil properties (e.g., texture, organic matter content), have a differential influence on emissions depending on N source.