148 Mr. Knox on the Direction and Mode of 



increase if it passes along the surface, the deflexion of the galvanometer indicating 

 the path. With one pair of elements there was no deflexion of the galvanometer^ 

 with two pair of elements there was a slight deflexion visible through a lens, 

 which increased slightly on immersing the platina wires in the liquid. With 

 four pair of elements, a deflexion of two degrees took place when the platina 

 wires were on the surface of the water ; a deflexion of four degrees when they 

 were immersed to the bottom of the tubes. As the number of alternations in 

 the battery increased, so did proportionably the comparative deflexions of the 

 galvanometer : the experiments proving that water, whether undergoing elec- 

 trolization or not, conveys an electric current through its substance, and not 

 along its surface, and that the decomposition of the water is an effect produced 

 by the passage of the electricity when of sufficient intensity, and not the neces- 

 sary consequence of its passage. 



A similar experiment having been tried with phosphorus melted under spirits 

 of wine, (being a non-conductor,) it was found to Obey the same law with water ; 

 that is, to convey the current through its substance.* 



To determine whether the metals followed the same law, I suspended from 

 the top of the new patent shot tower at Waterloo-bridge a leaden pipe, 170 feet 

 long, and three-fourths of an inch Internal diameter, through which was drawn an 

 insulated copper wire, 180 feet long, one extremity of which being soldered to 

 the inside of the end of the pipe, this end was sealed with fused metal, and to its 

 external surface was soldered a copper wire of the same length as the former ; 

 round the tube, at its orifice, was twisted a copper wire ten feet long. The insu- 

 lated wire being connected with a constant battery of one pair of elements in 

 contact with one pole of an exceedingly delicate galvanometer, (constructed by 

 Mr. E. M. Clarke of the Lowther Arcade, ) the other pole of the galvanometer 

 was brought successively in contact with the extremities of the uninsulated wires. 

 The deflexion was greater when the current passed along the wire connected 

 with the orifice of the tube, (although here the contact was not so good,) than 

 when it passed along that soldered to the sealed extremity. 



Again, the uninsulated wires being connected with separate galvanometers, 



• It was unnecessary to try similar experiments with the analogous bodies, sulphur, selenium, 

 and iodine. 



