Ammonia losses from surface-applied nitrogen fertilizers are important for agronomic and environmental reasons. Some of the simple methods developed to measure such losses use passive-flux samplers, which do not require power or sophisticated instrumentation. The passive-flux method described by Schjoerring et al in 1992 consists of measuring ammonia losses from a circular plot by mounting samplers at different heights in four masts located at ninety-degree angles on the periphery of the plot. The simplicity of this method has allowed researchers to carry out many measurements of ammonia losses during the past 15 years. In spite of its simplicity, however, the method requires a large number of samplers. A modified method proposed by Wood et al. in 2000 consists of a single, rotating mast located at the center of the circular plot, with a single passive-flux sampler at each height. We carried out two field studies to compare the method with masts on the periphery to the method with a central rotating mast. In the first study, three circular plots, 40-m in diameter were fertilized with 67 kg N ha^-1 as urea-ammonium nitrate solution. Four masts were mounted on the periphery of the plots and one mast was mounted at the center of each plot. Each mast had passive flux samplers at 0.4. 0.75, 1.5, 2.25, and 3 m. Samplers were replaced at 4, 8, 12, and 26 d after application. In the second study, 210 kg N ha^-1 as urea was applied to three, 30-m circular plots and ammonia losses were measured for 9 days. Average ammonia losses measured with the periphery masts were 68% of those measured with the central-mast method. Ammonia bypass in the passive-flux samplers of the periphery masts may explain the differences observed.