saltation events Co. and total dust flux C. yielded similar, but poorer fits: 7.58 x 1073PSoqU2 Cs (8) (r = .902, F = 61, p < 0.01), and 1.88 x 1072PSg QU? (9) WV aQ (r = .960, F = 164.1, p < 0.01); where s sum of all five saltation compartment dust collections sum of ground creep and saltation compartment collections. t The distribution of dust flux with height above the ground surface is shown in Table 2. A good fit between cumulative fractional dust flux over all time intervals, and midheight (Z) of each collector compartment of the Bagnold sampler was obtained: F = 76.45Z72-21 (10) (r = .999, F = 29117, p < 0.01) fraction of dust flux at height Z for saltation and soil creep processes combined midheight of collecting surface, cm. fit was obtained when saltation dust flux alone was considered: 28.43z71-79 (1) (r = .999, F = 1488, p < 0.01) total saltation flux from all five collecting compartments. were also obtained when collector fractions were compared: 1.75, (12) (r = .992, F = 61.26, p < 0.01), 2.23C. (13) (r = .987, F = 36, p < 0.01), and 1.23C. (14) (r = .958, F = 11.2, p < 0.01). 691