Three atmospheric research aircraft were used to explore the atmospheric boundary layer during FIFE: the National Research Council of Canada Twin Otter, the National Center for Atmospheric Research (NCAR) King Air, and the University of Wyoming King Air. The aircraft were used to measure the mean and turbulent structure of the boundary layer and its variation with height, time, and space. These measurements are important to FIFE because they are being used to scale up point surface observations to landscape scales and because they can be used to relate satellite radiance measurements to boundary layer processes. Because the aircraft were used in coordinated flight patterns to investigate changes within and between intensive field campaigns, wing-to-wing intercomparisons were made so that measurements from one aircraft could be related to another. Intercomparisons were flown on 4 days in 1987 and 3 days in 1989. The eddy correlation measurements of the mixed layer fluxes of moisture and sensible heat were of particular interest to FIFE. Sensible heat fluxes agreed within 15 W m-2 and moisture fluxes agreed within 21 W m-2. Mean wind component differences were within 1.0 m s-1, air temperature within 0.3 ¿C, and mixing ratio within 2 g kg-1. Standard deviations showed similar good agreement, with mean differences generally less than 0.1 m s-1 for the wind components and 0.03 ¿C for potential temperature. Intercomparisons between the NCAR King Air and the Twin Otter showed better agreement in 1989 than in 1987. Overall, the results suggest that data from the FIFE boundary layer aircraft will need little correction to account for instrument biases and spurious fluctuations. Researchers should consider these intercomparison results before using combined sets of boundary layer aircraft data. ¿ American Geophysical Union 1992