368 CELESTIAL CHEMISTRY. 



thallium salts make it g'reen, strontium compounds tinge it red 

 and those of potassium g'ive it a violet colour. 



If we examine by means of a spectroscope a flame coloured 

 in any of these ways, we shall naturally expect to see the 

 spectrum of sodium, thallium, strontium, or potassium, as the 

 case may be. Please observe that these are spectra of glowing^ 

 gas — not of glowing" liquids or solids — and such a spectrum of 

 glowing gas we find to present a complete contrast to the 

 spectrum of the sun's rays, or of a red hot poker, or of some other 

 glowing" solid or liquid. While a glowing solid or liquid gives 

 a more or less continuous spectrum, i.e., a coloured or bright 

 spectrum crossed, in the case of the sunbeam, by dark lines, 

 the g"lowing" gas or vapour or the metals sodium, potassium, 

 etc., produces a dark spectrum crossed by bright or coloured 

 lines. So we have, as you see, two classes of spectra, viz. : 

 continuous spectra and bright line spectra, the former yielded 

 by glowing" solids or liquids and the latter by glowing gases. 



For a moment let us put on one side the bright line spectra 

 and g"o back to the continuous spectra of glowing solids. I 

 remarked that the spectrum of a red hot poker, or of a fish 

 tail luminous gas burner, or of incandescent lime, such as is 

 used in a lime-light lantern, is absolutely continuous; that is 

 to say, it is not crossed by any dark lines like those in the 

 spectrum of a sunbeam. But we can put dark lines into the 

 spectrum of the lime light or of the red hot poker. 



Imagine that we are looking through a spectroscope at a 

 piece of white-hot lime. What do we see? An absolutely 

 continuous spectrum, of course; one without any dark lines 

 whatever. Now remove the white-hot lime, and put in its 

 place the flame coloured yellow by sodium vapour and view the 

 spectrum again. Now it is all dark except for the bright yellow 

 sodium line at the point on the spectrum usually marked D. 

 Now comes the important step. Place the white-hot lime in 

 such a position that the yellow sodium flame intervenes 

 between it and the spectroscope, and look throug'h the telescope 

 again. The result is remarkable: you have now a spectrum 

 all but continuous; that is to say, it would be quite continuous 

 but for the fact that it is crossed by just one single dark line, 

 and that dark line occupies exactly the position which the single 

 yellow line of sodium occupied before. The sodium vapour 

 has, in fact, absorbed from the white lime-light the particular 

 light which it would otherwise have given out itself, and which 

 would have fallen on the place occupied by the dark line. We 

 may say that the intervening" sodium vapour has thrown a line 

 of shadow across the lime-light. 



What are we to infer from this phenomenon ? We have 

 imagined ourselves looking through the spectroscope at an 

 intensely heated white-hot solid body, and between us and that 

 white-hot solid there intervenes a glowing" vapour which casts 

 its shadow across the spectrum of the solid body behind it : 

 can we conclude that all dark lines in what would otherwise 

 have been a continuous spectrum have been caused in this 

 way? Take another case. The spectrum of the vapour of 

 the metal Indium consists of two bright lines, one blue and 



