314 Mr Shaw, On experiments with mercury electrodes. [Dec. 8, 



mean position by magnets, and which caused the solution of so 

 much mercury, I next endeavoured to employ an electrolyte freed 

 from air. In order to attain this an apparatus was constructed as 

 in Fig. 2. A cylinder of glass A, with rounded ends, had sealed into 

 the upper end a bent glass tube with airtight glass tap. Into the 

 lower end a funnel with a long stem C was sealed. The vessel 

 was filled with a boiling 2 p.c. solution of Sulphuric acid, made 

 with well-boiled water, and the lower end of the stem C, connected 

 by India-rubber tubing with a small glass tube E, fixed to the 

 frame of the apparatus, and the tube E again by India-rubber 

 tubing with the small globe G, fixed at the end of a lever, which 

 was movable about a screw fixed in the frame of the apparatus. 

 When the fluid had cooled, mercury was made to flow up the 

 tube G, and partly fill the vessel A outside the funnel. The 

 mercury inside and outside the funnel respectively was connected 

 with the galvanometer by means of platinum wires sealed through 

 the cylinder at E' and the glass tube at E. When the quantity of 

 mercury was adjusted to its proper height in the funnel, leaving an 

 approximately vacuous space above the electrolyte, the level could 

 be altered, by moving the lever L. The vacuum thus obtained 

 remained exceedingly good for several weeks, maintaining the fluid 

 almost perfectly freed from air. 



With this apparatus an electromotive force of "7 Daniell was the 

 smallest which produced any appreciable permanent current. The 

 effects too of raising and lowering the globe were very large when 

 only small electromotive forces were employed ; for higher electro- 

 motive forces they were as under: 



The Meidinger cell was left in circuit for the night, and a small 

 permanent current was passing. The following morning bubbles of 

 gas were seen between the sides of the glass cylinder A and the 

 mercury which it contained. The electromotive force was further 

 increased by the addition of a Daniell's element with a variable 

 shunt and the bubbles between the mercury and glass gradually in- 

 creased in size, the throw of the needle remaining approximately 

 the same. With e.m.F. of about 1"1 Daniell there was a slow effer- 



