KINETICS OF WETTING IN WASHOUT OF DUST 291 were limited to voltages above 2 volts. At lower voltages there was a considerable scatter of the data, and apparently the voltage had only a small effect in this region. No complete explanation can be given, but it was suggested by Rayleigh® that uncharged water drops may coalesce in the presence of charged dust in the air. Preliminary results from other studies seem to confirm this explanation.? It was thought that hygroscopic salts might be affected by the humidity of the air. This issue was studied extensively with NaCl crys- tals. These were cleaved out of large single crystals into pieces approximately 12 by 12 by 4mm. They were mounted so as to present a corner, an edge, or a flat surface to the water drop. There were no differences in delay times for these different mountings. Two sets of data were obtained from experiments conducted in a glove box at 3 and 45% relative humidity. There was no difference between the two sets of data (Figs. 4 and 5). The relative humidity was then increased to 60% (Fig. 6). The data then scattered badly at high voltages. A similar phenomenon had previously been observed with aqueous solutions, which gave tV = constant at low voltages and tV’ = constant at high voltages with data between these relations at intermediate voltages.' The scatter was taken as an indication that the NaCl crystal may behave as a saturated solution at this high humidity. This is suggested also by the observations of Orr e¢ al.'° Experiments with a drop of saturated NaCl solution and a water drop (Figs. 7 to 10) and with “dry” NaCl crystals at 75% relative humidity (Fig. 6) confirm this view. The effect wetting of NaCl included. In the effect, but, at 60 of the relative humidity of the air upon the rate of crystals is shown in Fig. 6, in which all the data are range 3 to 45% relative humidity, there is no humidity to 65% relative humidity, the rate of wetting is notice- ably higher. At 75% relative humidity, the crystals are dripping wet and are obviously covered by saturated NaCl solution. The rate of wetting scatters between the curve for the saturated NaCl solution and the values at 60 to 65% relative humidity. Thus the crystal is not uni- formly covered by the saturated solution. Experiments conducted with borosilicate-glass beads gave the results shown in Table 2. There is no trend in the data. However, the glass may not have been conducting enough to be affected by the apTable 2—-DELAYS OBSERVED AT VARIOUS VOLTAGES FOR BOROSILIC ATE-GLASS BEADS, MSEC Treatment None None HNO; Polarity + + 2% 4 0,84 1.29 6 Applied voltage & 10 12 414 1.71 1.34 1.24 1.48 ; #16 18 20 0.86 1.41 1.55 1.18 0.78 1.83 1.26 2.71 1.00 1.0° 0.84 1.24 0.90 1.26 1.30 1.95 1.88 1.75

Select target paragraph3