Boundary-layer observations in GATE helped to promote the mixed-layer concept, which had been proposed in the context of the trade-wind boundary layer Malkus nearly 20 years before. Observations showed flux profiles to be similar to what was observed earlier over land, in BOMEX, and in the Caribbean to the north of Puerto Rico. Surface temperatures and surface fluxes, like those in ATEX, showed little diurnal trend in fair weather. What was new? GATE aircraft profiles showed near-constant wind profiles with height, reflecting the predicted behavior seen in the then recent first large eddy simulations. The buoyancy flux was found to have consistent behavior, with values at cloud base (rough top of the mixed layer) roughly -0.1 times the surface value. In contrast, the humidity-flux profile varied with cloud cover: in clear skies, the humidity flux at the top of the mixed layer was near zero, but in the presence of more cumuli, the flux at the top of the layer exceeded evaporation at the surface, leading to drying. Sensible heat flux divergence was not adequate to explain the mixed-layer temperature evolution: radiation had to be accounted for. Finally, grid patterns by one of the GATE aircraft revealed mesoscale bands whose horizontal dimension was several times the depth of the mixed layer. It is thought that these structures result from tropospheric gravity waves interacting with the mixed layer. Tethered balloon and radiosonde data revealed variability in the height of the mixed layer of the order of 20 to 25 percent mixed layer depth (about 500 m). Perhaps most significantly, the GATE boundary layer was one of the best sampled, with up to three gust-probe aircraft documenting the same mixed layer over a long period of time. Supplementing these data were tethered balloon, surface data, acoustic-sounding data, and radiosonde data from nearby ships