1903.] on Some Itcccnt Investigations on Electrical Conduction. 299 



mainly in the narrow portion d d, and this alone, with the branch «, 

 was kept hot. 



It was found that at a full red heat, the resistance of the liquid 

 mercury was about doubled. The resistance of the saturated vapour 

 was taken with the same apparatus, the narrow part d d being in this 

 case filled with the vapour instead of the liquid ; it was still ten 



million times as great as that of the liquid. But the vapour did 

 conduct very appreciably ; and a current easily measurable with the 

 galvanometer could be sent through it with a single battery cell. 

 I think that in all probability, if we could trace the charge up to the 

 critical temperature, we should find that the saturated vapour would 

 approach in its electrical behaviour to the liquid metal. 



