224 PHENOMENA CONNECTED WITH CLOUDY CONDENSATION. 



the vertical bands of ditfractioii colors are distinctly seen on each side 

 of the bright light. If now we keep the amount of dust in the air con- 

 stant during the experiments, we shall find that, on opening the stop- 

 cock to the vacuum receiver very slowly, we will get the u.sual cloudy 

 condensation, and that the diffraction colors will be quite distinct. But 

 if we repeat the experiment and this time open the stop-cock very sud- 

 denly, so as to cause a rapid expansion, the colors will be found to be 

 very much improved, being far more brilliant. This is due partly to 

 the greater number of particles engaged in producing the effect, but 

 chiefly to tlie much more equal size of tlie particles when they are sud- 

 denly developed tluin when slowly grown. 



It is found that we must not have too many particles present, or the 

 diffraction colors will not be good; their size does not seem to be great 

 enough to ])rodiice the phenomena. If, for instance, in place of using 

 the air of the room, we take into the flask air coming from a small 

 flame, the color phenomena in the flask all change; when there were 

 few particles the light transmitted directly through them has so little 

 color it is not noticed, while the diffraction colors are tine; but with 

 many particles the direct light becomes colored, while the diffraction 

 colors are softened and have lost much of their brilliancy. When the 

 particles are sufllciently numerous to cause the directly transmitted 

 light to be of a thin blue, the dift'raction color next the blue light is 

 nearly the complementary yellow, and this yellow light extends to near 

 the limits of the flask. If more particles be added, the color of the 

 transmitted light becomes deeper blue, but it is difticult now to say 

 what the diffraction colors are. The convection currents in the flask 

 now make themselves visible; the air on each side of the blue direct 

 light is suffused with a variety of colors, not now in regular vertical 

 bands, but irregularly distributed and in movement through the flask. 



Cause of the color. — These experiments show that the color produced 

 by the small drops of water depends on the size of the drops, and the 

 depth of color in their number. But it is not so easy to follow the man 

 uer in which the drops iDroduce the color. If we take the siniplest case, 

 we can easily see how part at least of the cohn^ is produced. In the 

 steam jet condensing dense, and coloring the transmitted light yellow, 

 part of the effect is no doubt due to some of the particles in that form 

 of condensation being so small that they reflect and scatter the shorter 

 waves of light, while they allow the longer ones to pass through. The 

 color in this case is partly caused in the same way as the yellow pro- 

 tduced by small particles suspended in liquids, as in Brack's experiment 

 with mastic, or as when silver chloride is formed from a solution of the 

 nitrate. The light reflected by the liquids in these experiments is of a 

 bluish tint, complementary to the yellow light transmitted by them, 

 and this blue light is polarized. It has been found that when the steam 

 jet is of a good yellow by transmitted light, it reflects a good deal of st/ 



