On the Chemistry of the Hottest Stars. 



153 



of determining exactly the difference between the flame and arc 

 spectra. 



The general resnlt of this line of work was to show that there was 

 a step similar in kind to that from arc to spark between the flame 

 and arc. To take iron as a case in point, a few lines only constitute 

 the spectrum of the flame. The number is enormously increased on 

 passing to the arc, but none of the flame lines are dropped. 



A very important fact to be borne in mind is that the intensities 

 of the lines common to the flame, arc, and spark differ greatly at the 

 three temperatures, so that the three stages of flame, arc, and spark 

 lines are generally recognised. 



II. Application of the Early Results to the Investigation of 

 Stellar Spectra. 



General Remarks. 



In attempting to assign origins to the lines seen in the spectrum 

 of the sun or stars, it is clear, from what has been stated, that we 

 shall not get a perfect matching with terrestrial lines, either in regard 

 to the intensities or lengths of lines, unless the terrestrial lines in 

 question are those seen at a temperature closely approximating to 

 that of the particular part of the sun or star in which the vapour 

 which produces the absorption exists. 



In the atmosphere of the stars we have probably the closest 

 approximation open to our observation of that equally heated space 

 condition to which I have previously referred. 



The question arises whether we have any means, independently of 

 the line phenomena we are now discussing, of determining whether 

 the effective absorbing region of the stellar atmospheres are at the 

 same temperature or whether they vary. And iu the latter case, 

 can we go further and arrange them approximately in the order of 

 temperature ? 



First Indications of Relative Stellar Temperatures. 



On this point I wrote as follows in 1892* : " An erroneous idea 

 with regard to the indications of the temperature of the stars has 

 been held by those who have not considered the matter specially. It 

 has been imagined that the presence of the series of hydrogen lines 

 in the ultra-violet was of itself sufiicient evidence of a very high tem- 

 perature. The experiments of Cornu,f however, have shown that 

 the complete series of lines can be seen with an ordinary spark with- 

 out jar. Hence the high temperature of such a star as Sirius is not 



* ' Phil. Trans.,' 1893, vol. 184, p. 721. 



f 1 Journ. de Physique,' 1886, series 2, vol. 5, p. 341. 



