in Mercury by Deoxidation. 493 



centim. wide, 0*6 centim. deep, and 10 centims. long. In 

 the middle of it was placed a globule of mercury, a b (tig. 5); 



rig. 5. 



PLAN 



Ur 



-SAT (& ^ o\ -w r 



y i 



and on each side of the mercury the spaces, w and v/ } were about 

 half filled with water. Two wires, + and — , dipped into the 

 water near the ends, and were connected with a battery. The 

 object in view was to oxidize one portion of the surface of the 

 mercury globule whilst another portion of it was being deoxi- 

 dized. When the battery was connected so that the current 

 passed through the trough from + to — , it is evident that 

 the surface a of the mercury was deoxidized, whilst at the 

 same time the opposite surface b of the globule was oxidized. 

 The result of this double action it was easy to foresee. The 

 surface a becoming deoxidized, contracted, whilst the surface 

 b, becoming more oxidized, expanded. 



The upper surface of the mercury globule was thereby 

 thrown forwards, in the direction shown by the arrows, car- 

 rying with it more oxidized surface, which was promptly 

 reduced on its arrival at a. The whole mercury globule 

 was set in rapid circulation, the end b supplying the oxi- 

 dized and expanding skin as fast as the end a could reduce 

 and contract it. When the water was slightly acidulated, 

 this passage of the oxidized surface became so vehement that, 

 the globule being retained in its place, the water from iv f was 

 carried mechanically over and heaped up on the side id. But 

 when the mercury globule was free to move, it marched 

 bodily forwards from iv to w' (opposite to the arrows) rather 

 quickly towards the negative wire of the battery — that is, in 

 the same direction in which the electricity is assumed to flow. 



At the time the upper surface of the mercury globule was 

 speeding along, the under surface was also observed to circu- 

 late in the same direction ; but on account of friction offered 

 by the trough, its motion there appeared to be slower. The 

 interior circulation of the globule appeared to bo approxi- 

 mately that shown by the arrows in fig. 6 — b being the ex- 



