Approached on a bright sunny day from the south side, the angle of slope within the crescent of a barchan is most deceptive. Nine out of ten persons will confidently state that the face of the dune cannot be less than 60 or 70 degrees from the horizontal, though, as a matter of fact, it never exceeds by more than two or three degrees a third of a right angle. Unless held together by vegetation or compacted by moisture, it cannot exceed the angle of rest of dry blown sand, which is about 32 degrees. In normal weather this face of a barchan is a straight, unrippled, even slope of loose sand from top to bottom; when a light or gentle breeze is blowing from the north, the dune can be seen ‘smoking’ at the top of the crest, the sand being blown up the much more gentle and markedly rippled northerly slope, and dropped over the crest on to the steep, smooth, southerly face. The bulk of this sand in a light breeze falls on the upper part of the slope, equilibrium being adjusted from time to time by a downward sliding of the superficial layers. A cross-section of the dune, from north to south, would indeed show it to be composed of well-stratified layers of sand, dipping steadily southwards at a little over 30 degrees. In this district I have never observed southerly winds hold for a sufficient length of time to do more than slightly modify the general form of a barchan, though during their continuance the progress of the dune is arrested, the outline of the crest becomes considerably altered, ripple-marks appear on the steep southerly face, and an apron of sand of greater or lesser dimensions forms at its foot.