ROTATING-DISK-SAMPLER MEASUREMENT OF AEROSOLS 231 DISCUSSION The two consecutive time-averaged distributions differ about twofold in magnitude in the range r < 0.025 u and also show a shift in the maximum with radius. In both runs the maximums are well defined although the behavior of the distribution function below r = 0.005 pis not known, For r > 0.025 yp, the distribution function showslittle change with time. The very small particle end of the spectrum seems more susceptible to variation than the part above r = 0.025 u. This variation may be the result of fluctuations in the strength of local sources of very finely dispersed matter, Such particles can diffuse rapidly from their point of introduction into the atmosphere, On the other hand, the possibility of an artifact in the sampling cannot be ruled out as an explanation of the variation between the two runs, The results of the two runs are shown in Fig. 6, with the distribu- tion functions calculated from data presented by Junge’ and Cartwright!” for the small particle range, Junge calculated his data points from ion- mobility measurements. The Zugspitze curve shows a maximum at about the same radius observed in this study. The original data of | I I Pott | | r Tf _ 10° — | — -— I ot 10° _ =! o +— ° = = S by | = = 1o¢bK— o — = T a 6 Oo — _] Oo 4 A eo QO 8 a en T I _!| | — “| °° . A O | —— THIS INVESTIGATION (BALTIMORE ) 10° -— x JUNGE (FRANKFURT ) | O)-JUNGE (ZUGSPITZE) L$ CARTWRIGHT (SHEFFIELD) | A wot 10-3 CARTWRIGHT (BUXTON ) tp 10° A “] 0 —| _| —_| 44 tt re 197! Fig. 6 —Comparisonofthe particle size distribution data obtained in the small size vange in this study with the resulis of other investigators. A consider- able variation occurs in the shape ofthe distribu- tion from place to place and time to time. This is in contrast with the be- havior of the upper end of the distribution (r >0.02 w, which appears to be more - stable.