ox OUn KNOWLEDGE Ok' SrECTBCM AXALTSI:?. 277 



sun, the relative intensities of the lines is often the same as that we 

 observe in our laboratory experiments, and where it is not, it does not 

 show such changes as would be produced by a mere increase of the 

 absorbing layer. In liquids, of course, and some vapours which have 

 large absorbing powers, an effect of the thickness of the absorbing layer 

 can be traced. 



If the temperature of a radiating gas is increased, the absorbing 

 power for each ray remaining the same, the radiation will vai'y in the 

 same proportion as e ; that is, as the radiation from a perfectly black body. 

 It follows that the more refrangible rays will relatively gain more- than 

 the less refrangible rays, but it must be boi'ne in mind that the absolute 

 intensity of any given line can never decrease, unless the quantity of 

 luminous matter decreases. 



II. Widening of Lines. 



In his celebrated paper, ' Optiska Undersokningar ' ' (1853,) Angstrom 

 gives two drawings of the hydrogen spectrum. In neither of them are 

 the lines sharp ; but in one of them especially they are drawn out into 

 broad bands. The property of widening their lines under certain circum- 

 stances has since been found to be common to all bodies, though some of 

 them possess it to a much larger extent than others. Hydrogen and 

 sodium are the best known instances of elements which widen their lines 

 considerably, though in one of the spectra of oxygen, the lines broaden 

 even more easily. Wiillner ^ has given a detailed description of the 

 behaviour of the hydrogen lines under different pressures, both with the 

 condensed and uncondensed discharge. The same author has also given 

 us information as to the widening of some of the oxygen lines in the same 

 paper, and his observations were confirmed and extended by the author 

 of this report.^ Ciamician '' has described the widening of the lines ol 

 mercury, sodium, and some other gases. The papers of Lockyer, and of 

 Liveing and Dewar will also be found to contain many observations on. 

 the widening of lines. We shall refer presently to some of their most 

 important experiments on the subject. 



It is a fact which is often, though by no means generally, true, that if 

 a spectrum widens its lines easily, the widening begins with the most 

 refrangible lines. This was first noticed by Plticker and Hittorf in the 

 case of hydrogen. They express themselves as follows : ' 



' Hydrogen shows in the most striking way the expansion of its 

 spectral lines and their gradual transformation into a continuous spectrum. 

 By increasing the power of the coil, Hy (coincident with solar line near G) 

 first expands, then H/3 (solar F). Let the aperture of the slit be so regu- 

 lated that the double sodium line will separate into two lines, nearly 

 touching ; then the angular breadth of H/3 becoming two or three 

 minutes,*' the breadth of Hy is about double. Ha remains almost un- 

 changed after Hy has passed into an undetermined hazy band, and H/3 

 extended its decreasing light on its two sides.' A fourth hydrogen line, 

 more refrangible than the others, was discovered by Angstrom, and as 

 Goldstein "^ has remarked, this line is the first to widen, thus followine 



' Translated, Pogg. Ann. xciv. p. 141 (1855). 



^ Pogg. Ann. cxxxvii. p. 339 (1869). ' PMl. Trans, clxx. p. 37(1879). 



♦ Wien. Per. (2) Ixxviii. p. 867 (1878). ^ PMl. Trans, civ. p. 21 (1S66). 



" The large Steinheil spectroscope was used. ' Berl. Per. p. 593 (1874). 



