July 8, 1920] 



NATURE 



587 



of white light. These bands are clearest when the 

 incident light is completely polarised, as you now 

 see them projected on the screen. But they can 

 still be seen when the polarisation is but slight. 

 I will illustrate this by removing the polarising 

 Nicol which I have been using, and substituting 

 a single glass plate, through which the incident 

 light passes. If I incline this plate so as to 

 polarise a small fraction of the light, you see the 

 bands, faint but sufficiently distinct. In examin- 

 ing the light of the night sky, a photographic 

 plate is substituted for the paper screen I have 

 been using to-night, and the apparatus is designed 

 for the utmost economy of light. With two hours' 

 exposure a definite image of the sky was obtained, 

 with the stars superposed upon it. The Savart 

 bands could be seen, but they were very faint 

 compared with what would have been observed 

 with an equally good image of the daylight sky. 

 The part of the sky examined was near the pole, 

 and therefore nearly at right angles to the sun. 

 If, as seemed possible, the night sky derived its 

 light from an attenuated atmosphere so high as 

 to be outside the earth's shadow, we should expect 

 it to show the same polarisation as the day sky. 

 Since it does not do so, we must attribute the 

 light at night to some different origin. 



I was fortunate in being able to interest Prof. 

 Hale in this matter while he was on a visit to 

 England, and as a result Mr. Babcock repeated 

 the observations in a modified form at the Mount 

 Wilson Observatory in California. The traces of 

 polarisation which he obtained in that clear atmo- 

 sphere were even less than what I got in England. 



Ozone, and the Limit of the Solar Spectrum. 



Although, as we have seen, the idea that the 

 blue colour of the sky is due to any action of ozone 

 cannot be admitted, yet there are points of great 

 optical interest connected with the presence of this 

 gas in the atmosphere. We may now turn to the 

 consideration of some of these. 



It is of course well known that when the solar 

 spectrum is formed by a prism of quartz or by a 

 grating, the spectrum can be observed to extend 

 beyond its visible limit in the violet into the region 

 called ultra-violet. When, however, we examine 

 the spectrum of an electric arc (and for this 

 purpose an iron arc is particularly suitable), the 

 extension is observed to be very much greater 

 than in the solar spectrum. This is not because 

 the sun does not emit any rays of the kind in 

 question, but because the earth's atmosphere will 

 not allow them to pass through so as to reach us 

 at the earth's surface. There are many reasons 

 for feeling sure that this is the true explanation, 

 but one of the simplest will here suffice. When 

 the sun is near the horizon, so that the rays pass 

 obliquely through the earth's atmosphere, and 

 consequently have to traverse a thicker absorbing 

 layer, the extent of the ultra-violet spectrum is 

 found to be even less than when the sun is high 

 and less air is traversed by the rays. This suffi- 

 ciently proves the point. 



It has long been suspected that ozone in the 

 NO. 264^, VOL. I05I 



atmosphere is the effective cause of this absorption 

 of the ultra-violet rays. The most important con- 

 stituents of air, oxygen, and nitrogen do not 

 appreciably absorb at the point where the solar 

 spectrum ends, nor do the constituents of second- 

 ary importance, carbonic acid, water-vapour, and 

 argon. We must therefore look to some rare 

 constituent of air which is very opaque to this 

 region of the spectrum. Ozone possesses this 

 opacity, as I shall now show you. So far as I 

 know it has not been attempted to show this 

 before to an audience, but I think you will be able 

 to see it without difficulty. As a source of light an 

 iron arc is used, and the lenses and prism employed 

 in forming the spectrum are of quartz. I allow 

 the spectrum to fall on a piece of paper, and you 

 see the usual succession of colours, red, yellow, 

 green, blue, and violet, forming a comparatively 

 narrow rainbow-like band. Beyond the violet all 

 appears dark, the eye being insensitive to the 

 ultra-violet rays. If now I substitute for the 

 paper a screen of barium platinocyanide ~ (of the 

 kind used in X-ray work), we see an immense 

 extension of the spectrum beyond the violet. The 

 screen has the property of transforming the ultra- 

 violet rays, which the eye cannot detect, into 

 green rays which are readily visible. Thus beyond 

 the violet region we see green, which is, of course, 

 in no way to be confused with the original green 

 which was present in the source, and appears in 

 its normal position in the spectrum, on the other 

 side of the blue-violet. I interpose a thin sheet 

 of ordinary glass, and the greater part of this 

 extension of the spectrum which we get on the 

 fluorescent screen disappears. What I want 

 specially to show you, however, is that a thin 

 layer of ozone, much too thin to have any per- 

 ceptible colour, will have the same effect. There 

 is a glass tube, about 6 in. long and | in. in 

 diameter, situated between the quartz lantern con- 

 denser and the slit, when the beam is parallel, 

 and the walls of the tube are projected as two 

 thin transverse lines on the slit, dividing the spec- 

 trum into thin horizontal strips, one over the 

 other. The light constituting the middle strip 

 has traversed the tube, but the light constituting 

 the upper and lower strip has traversed the open 

 air above and below the tube. A stream of oxygen 

 passes through a Siemens ozone generator and 

 enters the middle of the observation tube, stream- 

 ing out at the two ends. While the ozone gener- 

 ator is not excited, the middle strip of the spec- 

 trum is similar to the comparison strips above and 

 below. If the induction coil is turned on so that 

 ozone passes into the tube, you see that in a few 

 seconds the greater part of the ultra-violet spec- 

 trum fades out from the middle strip, which con- 

 trasts sharply with the uoper and lower ones. 

 When the coil is turned off, the ozone is rapidly 

 blown out by unozonised oxygen, and the original 

 state of things restored. 



It must be remembered that the ozone used in 

 this experiment is extremely dilute, probably only 

 a fraction of i per cent, of the oxygen in the tube. 



'^ Kindly lent by Messrs. Watson. 



