Ammonia losses following urea fertilization of maize and winter wheat were determined in field trials carried out at Fengqiu Experimental Station in the North China Plain in 1998 and 1999. Four experiments were carried out using two simplified micrometeorological integrated horizontal flux methods [IHF(L) and IHF(S)], a chamber method (calibrated Dräger-Tube Method DTM) and the ¹⁵N-balance method using ¹⁵N-labeled urea. The IHF(L) was taken as the reference method. Both IHF methods showed good agreement in one experiment only, while the IHF(S) overestimated as well as underestimated cumulative ammonia losses compared to IHF(L) in the other experiments (deviation ranged from 12.5% to 64% based on cumulative ammonia losses). Regression analysis of the fluxes showed that in particular different sensitivities of the samplers to wind speed accounted for the discrepancies observed. The IHF(L) and the DTM flux curves were very similar in three experiments, while the values obtained with DTM considerably deviated from IHF(L) results in one experiment. A comparison with apparent fertilizer-N losses determined by the ¹⁵N-labeling approach showed that ammonia volatilization was the major pathway of fertilizer-N loss in this study.