CELESTIAL CHEMISTRY. 367 



a shower of such drops the result is the series of prismatic 

 images which we call the rainbow. 



The coloured band seen by Newton was really a series of 

 overlapping circles — and this is most important — so many 

 differently coloured circular images, in fact, of the hole in his 

 shutter, but to-day, instead of a rotuid hole, a str'aight slit is the 

 usual form of opening through which the light is allowed to 

 fall on the prism. 



For the purpose of examining the spectrum more closely 

 an instrument termed the spectroscope is employed. It consists 

 of a central dark chamber, containing a prism, towards which 

 three tubes are directed. One of these tubes is provided with 

 a small upright slit through which the light passes and falls 

 on the prism by which it is refracted through a telescope 

 towards the eye of the observer. The third tube is fitted with 

 a scale of numbers by means of which the observer is able to 

 identify any particular portion of the spectrum. 



The light of the sun is not the only source from which a 

 spectrum may be obtained. Any glowing" solid — and in fact 

 also any glowing liquid, like white-hot< molten iron — if viewed 

 through a spectroscope, will show a spectrum. A red hot 

 poker, for instance, will serve quite well as a source of light; 

 so will the gas flame from a fish tail burner, consisting, as it 

 does, of highly heated carbon particles. But — and this too is a 

 most important point — the spectrum produced by a glowing 

 red-hot solid or molten liquid is absolutely continuous from 

 end to end while that of the sun is traversed by thousands of 

 dark lines, in other words, many of the rays of light that come 

 from the glowing poker are wanting in the sunlight. Upon 

 these dark lines in the sun's spectrum is founded the w'onderful 

 knowledge that we have gained respecting the chemical com- 

 position of stars so far distant that their light rays which to- 

 day enter the tubes of our spectroscopes set out on their 

 journeys many thousands of years ago : indeed we may date 

 the birth of the whole system of the New Astronomy from the 

 time when the cause of the dark lines in the sun's spectrum, — 

 called after P^raunhofer, their discoverer — was first satisfactorily 

 explained. 



Now how do these dark lines come to be there ? First let 

 me say that each dark line is a dark image — if I may call it so — of 

 the slit, and the remarkable thing is that in all spectroscopes, 

 and at all times, they are shown in exactly the same relative 

 position in the sun's spectrum. But surely the sun is just as 

 truly a glowing body as a red hot poker or a luminous gas flame 

 — why then this difference and what do these dark lines in the 

 sun's spectrum indicate ? We shall try to answer these inquiries 

 by and by : meanwhile let us take up another thread of our 

 subject. A gas flame, in order to produce a spectrum, must be 

 glowing or luminous, that is to say, it must emit rays of light : 

 and if we supply a gas burner with air in such a way that the 

 flame does not glow we shall fail to find any spectrum. But 

 now introduce certain substances into this colourless gas flame, 

 and distinctive colours will at once be imparted to it. Sodium 

 • compounds, for instance, colour the colourless flame vellow, 



