xxvii.] MESSRS. LIVEING AND DEWAR'S RESULTS 



Coal Gas, which contains oxidisable gases but no oxygen, 

 should act effectively in bringing out reversals, in fact almost as 

 effectively as hydrogen. 



Hydrogen, passed into crucibles in just sufficient quantity 

 to burn at the mouth of the tube, should prevent oxidation with- 

 out diminishing the temperature, and consequently many lines 

 due to finer molecules should be reversed. If the current of 

 hydrogen passed in is too rapid, the cooling effect would to some 

 extent destroy the finer molecules of the metallic vapour and 

 hence they would give no reversals. Hydrogen also, since it 

 prevents the oxidation of the metal and the formation of solid 

 oxidised products which absorb generally, must brighten the 

 continuous spectrum and cause many bright lines in it to 

 become fainter by contrast. Hydrogen should be the most 

 effective of all the agents used for bringing out reversals, since it 

 acts as a shield without producing solid oxidised products. 



Abstract of Messrs. Liveing and Dewar' s Results. 



Calcium as oxide in an arc of thirty Grove cells in a carbon 

 crucible, gives reversals of 4226 and K, but under these conditions 

 H is never reversed. When hydrogen is passed into a Siemens' 

 arc in a lime crucible, the lines 4434, 4425, and 4454 are reversed. 1 

 Messrs. Liveing and Dewar do not state whether H and K were 

 visible when hydrogen was passed in. 



Dark bands frequently appear with ill-defined edges in the positions 

 of the bright green and orange bands. 



Aluminium added to the lime reverses 4226, H and K, and the 

 two lines of aluminium between them. Gradually the dark line 

 disappears from H and afterwards from K, while the aluminium lines 

 remain reversed for some time. 2 



A mixture of lime and potassium carbonates (said by Messrs. 

 Liveing and Dewar to produce a strong current of vapour in the 

 tube) in the arc of thirty Grove cells in a carbon crucible, reverses 

 4425, 4434, and 4454, and the line 4095, a doubtful calcium line. 



1 Proc. Roy. Soc. vol. xxviii. p. 472. 2 Ibid. p. 367. 



