Comparative demonstrations of three methods (flux sampling, external tracer ratio, and internal tracer ratio), were mounted in four real farm situations. A flux sampling method was demonstrated at a commercial dairy cow house (slurry-based), at a commercial piggery (straw-based), at a full-scale above-ground cylindrical slurry store (dairy cow slurry) and a full-scale earth-bank lagoon (pig slurry). An external tracer ratio method was demonstrated, in parallel with the flux sampling method, at the dairy cow house and at the above-ground slurry store. An internal tracer ratio method was demonstrated at the dairy cow house only. At the dairy cow house, the corrected emission rates from the flux sampling method and from the external tracer ratio method agreed to within the estimated experimental range, while the emission rate from the internal tracer ratio method was significantly lower. The overall conclusion of the study is that all three methods can have a useful role, the choice of which to deploy depending on the particular measurements needed in each case. The paper includes a table (No. 7) which indicates which is the recommended method when each of various considerations has top priority. (C) 2004 Elsevier Ltd. All rights reserved.