The data from the boundary-layer experiment carried out in November and December 1985 in Adelie Land, Antarctica, are analyzed. This area is famous for strong surface winds, with highly constant wind direction, which have been explained by the baroclinicity associated with surface radiational cooling over sloping terrain. However, during daytime in summer at D47, the net radiation was found to be positive, and the Richardson number as negative, indicating that the surface boundary layer was unstable. Unexpectedly, the wind directional constancy remained high. It was caused by the highly constant wind direction in the free atmosphere, which could be due to the large horizontal temperature gradient existing between the Antarctic ice sheet and the ice-free ocean along the coast of Adelie Land. A low level wind maximum was found, which was strong at night and weak during daytime. During nighttime, it was due to the slope induced baroclinicity, and during daytime, to the meso-scale baroclinicity caused by the large horizontal temperature gradient.