and of Healed Air for Electricity. 179 



them either with a decomposing apparatus or a galvanometer, 

 no evidence could be obtained of the existence of an electrical 

 current, although the copper was rapidly oxidized. 



It is well known that a wire of platina suspended above the 

 flame of an Argand lamp will become heated to bright redness, 

 showing that the air around it has reached at least as high a 

 temperature. It was the conducting power of air heated to 

 redness in this manner that I examined, from the facility of 

 pel-forming experiments upon it. Two platina wires were 

 suspended from insulating supports above the flame of an 

 Argand gas lamp, and connected with the poles of a battery 

 of 20 pairs of plates on Wollaston's construction in vigorous 

 action, but no iodine appeared in the decomposing apparatus. 

 The same negative result was obtained, whether fine platina 

 points approximated as closely as possible to each other, or 

 broad slips of platina foil were employed as poles. 



From this experiment it appears to follow that air simply 

 heated to redness does not conduct the current of a battery of 

 20 pairs of plates, but the singular facts which are now to be 

 described will not admit of so easy an explanation. 



The negative pole of a battery of 25 pairs of plates charged 

 with pump water was connected by metallic contact with the 

 brass tube of an Argand gas lamp, at a distance from the ori- 

 fices through which the gas issued, and a coil of platina wire 

 suspended above but not touching the flame, was attached to 

 the positive pole. When the flame was sufficiently powerful 

 to heat the coil to redness, the current passed freely, although 

 the coil was at least one inch distant from the flame. But 

 when the direction of the current was reversed, the negative 

 pole being connected with the heated coil, and the positive 

 pole with the base of the lamp, the passage of the current 

 could no longer be detected. In the former case the solution 

 of iodide of potassium was decomposed in a few seconds ; in 

 the latter case no decomposition occurred, however long the 

 contact was maintained, yet the direction of the current had 

 alone been changed, the other conditions of the experiment 

 remaining the same. Similar effects were obtained, when a 

 piece of well-burned charcoal was substituted for the platina 

 coil in the heated air. Nor were they different when a bat- 

 tery of 83 pairs of plates with double coppers, charged with a 

 solution of common salt, was used. These experiments were 

 frequently repeated, and every source of error carefully 

 avoided. 



This property of conducting and interrupting the same vol- 

 taic current when flowing in opposite directions is not peculiar 

 to heated air. It also belongs to flames ; but in consequence 



U2 



