THE SPECTRA OF STARS AND NEUULJ5. 



i 1 



a glow lamp, we find that the spectroscope spreads it out into 

 a band showing the glorious colours of the rainbow in their 

 greatest purity. The colours from red to violet merge into each 

 other by insensible gradations, and we say that the spectrum is 

 n continuous one, because there are no interruptions of any kirn I. 

 All incandescent solid bodies give precisely the same result, and 

 it follows that we cannot distinguish between one luminous 

 substance and another so long as they remain in the solid state. 

 The same is true of incandescent liquids. 



The effects are very different when the substances examined 

 are in the state of luminous gas or vapour. They then emit 

 special kinds of light by which they can be identified, and it 

 does not matter in the least whether they are in our labora- 

 tories or far away in the depths of space, so long as their light 

 reaches our instruments with sufficient intensity. The spectra 

 are no longer continuous, but consist of a number of bright 

 lines of different colours, which are really a succession of images 

 of the narrow slit through which the light is admitted to the 

 spectroscope. Thus, hydrogen is characterized by a line in the 

 red, another in the blue-green, and others in the blue and 

 violet, and since these lines are exhibited by nothing but 

 hydrogen, they serve to indicate the presence of hydrogen 

 wherever it occurs in the luminous state. Similarly, helium 

 signifies its presence by a number of lines, of which one in the 

 yellow is especially conspicuous. Each of the other elements 

 also has its own distinctive family of spectrum lines, some con- 

 sisting of a few members only, but others, such as iron, occur- 

 ring in hundreds. 



Many compounds which can be excited to luminosity without 

 decomposition also exhibit characteristic spectra, which are 

 quite different from those of the elements of which they are 

 composed. 



The luminosity necessary for spectroscopical analysis may be 

 artificially produced in various ways. Gases are usually 

 enclosed in vacuum tubes containing the gases at reduced 

 pressures, and are illuminated by electrical discharges. Sub- 

 stances which are solid at ordinary temperatures may be 

 vaporized and rendered luminous by the oxy-hydrogen flame, 

 the electric arc, the electric spark, and in a variety of other 

 ways which need not now be specified. 



It is most important to study the spectra in as many different 

 ways as possible, because, in opposition to early ideas, it has 

 been found that the same substance may give different spectra 

 when excited in different ways. Thus, at name temperature, or 



