532 Dr. J. H. Vincent : Electrical Experiments 



absolutely empty of liquid mercury. Very small drops could 

 be seen to be present ; they were in rapid movement to and 

 fro along the tube. In spite of this, it is probable that the 

 electric current was carried almost entirely by the vapour, 

 since these drops never became large enough or numerous 

 enough to form a bridge between the ends of the columns. 

 On repeating this experiment with the same tube, similar 

 results were obtained, but the conductivity was even greater 

 than before. Changing the direction of the applied electro- 

 motive force produced no apparent effect. By using a 

 sheet-iron flame-shield for the JBunsen burner, it was found 

 possible to so arrange the height and distance of the flame as 

 to cause the green glow to be emitted for as long as one 

 wished. In this case the tube was shorter, and a little water 

 was put in the bottles to moderate the temperature of the 

 mercury. Using a tube of quartz, mounted as usual (6* 15 cm., 

 5*5 mm., '4 mm.), it was possible to so arrange the flame as 

 to cause the green glow to be given out with great regularity 

 for hours together. 



When examined in a rotating mirror, the luminosity was 

 seen to proceed from positions in the tube the distance apart 

 of which varied in a periodic manner. The appearances 

 showed that the motion of the columns of mercury were of 

 the necklace type. In this case the heat necessary to maintain 

 the oscillations is supplied by the Bunsen burner from without, 

 and in the necklace effect the heat is supplied by the electric 

 current from within. If one continues to observe the glow in 

 the rotating mirror when the flame is turned out, it is seen 

 that the luminosity does not cease at once, but its periodic 

 character is lost very rapidly. 



A very bright persistent green glow can be obtained by 

 placing a Bunsen flame under the quartz tube and exhausting 

 the air from one of the bottles when the mercury column has 

 broken. The luminosity thus produced can be made several 

 centimetres long, and will persist indefinitely. It is caused 

 by the motion of the condensed drops of mercury as they 

 pass towards the cool end of the vapour column. The visible 

 spectrum of this glow shows the following lines : — 



4047 Violet, 

 4358 Blue, 

 5461 Green, 



579o} Yellow ' 



5790 



together with a background of what appears, under the dis- 

 persion employed, to be a continuous spectrum extending 



