Accurate measurements of methane emission rates from livestock in their undisturbed natural environments are required to assess their impacts on radiative forcing (i.e. enhanced greenhouse effect) and the environment. Here we compare results from two non-intrusive techniques for the measurement of methane emissions from cattle. The cows were kept in an outdoor feeding strip that allowed them to follow natural behavioural patterns, but contained them within a well-defined space. In the first technique, nitrous oxide was released as a tracer at the upwind edge of the feeding strip, and the downwind concentrations of nitrous oxide and methane were measured simultaneously using Fourier transform infrared (FTIR) spectroscopy. Average methane emission per cow was calculated each half-hour on three separate days from the correlation between the two gases. The second technique was the well-established integrated horizontal flux (IHF) or 1-D mass-balance method, in which we used the measured vertical profiles of methane concentration and wind speed downwind of the cows to determine the total methane emission. Comparing the IHF results to the known release rate of nitrous oxide further allowed us to test the IHF technique independently. We found agreement within 10% for all comparisons on all days. The daily methane emission rate averaged over all tracer and IHF measurements was 342 g CH4/head.day. This is within the range of previous measurements for mature lactating dairy cattle (200-430 g CH4/head. day), but higher than expected for yearling cattle. The high CH4 emissions are accompanied by high CO2 emissions also determined from the FTIR measurements. The bias is most likely due to the measurements being made during and after supplementary feeding of the cattle.