Properties of' Carbon when in Combusti&ii. 345 



copper, by means of the wire which dips into it, forms the nega- 

 tive plate. 



Experiment I. — The apparatus being charged, a wire was in- 

 serted into each of the cups, whose other extremities were so 

 connected by mercury, with a needle apparatus, as to form a 

 continuous wire passing below the magnetized needle, which was 

 delicately suspended by a single fibre of silk, and enclosed in 

 glass to prevent any currents acting on it. 



TheefFect produced on it was sufficient nearly to make it 

 stand at right angles. to its natural position. 

 . Experiment \\. — In this experiment the electricity was made 

 to pass through the space of six inches filled with burning coke, 

 having no metallic communication. 



Having attached to the ends of two copper wires, a rod of 

 iron, about a quarter of an inch in diameter, one end being beat 

 out, so as to present a greater surface to the action of the fur- 

 nace, and the other ends attached to the copper wire being amal- 

 gamated, to form a more perfect contact, they were both placed 

 at the distance of six inches from each other, into a brick furnace, 

 filled with burning coke, and in which there was no metallic gra- 

 ting. 



When the tempierature of the furnace was at a dull red heat, 

 and the circuit completed, part of which circuit was formed by 

 the coke, the needle was very slightly affected, showing that the 

 electricity passed through the six inches of burning coke with 

 some difficulty. When the temperature was raised to a bright 

 r^rf heat, the eftect produced on the needle was much more ap- 

 parent. When it had arrived at the melting point of copper, 

 the effect was; about double that of the preceding, and it went 

 on increasing as the temperature was augmented, till it attain- 

 ed the melting point of iron, when the effect was not much infe- 

 rior to that produced by a complete metallic communication. 



Experiment III. — A solid cylindrical piece of charcoal (Plate 

 V. Fig. 3.) about an inch in diameter, and six inches 



in 



length, being prepared, it was made to form part of the galva- 

 nic circuit, by having copper wires coiled round each end of it, 

 for the space of about an inch and a half, leaving the middle 

 part of it without any metallic communication. 



