1884.] On a New Method of Generating Electricity. 



213 



heated hydrogen may give rise to electrical currents under a variety of 

 circumstances. 



Small cells were made by nearly covering short wires or rods of 

 metal with melted glass. The glass was then covered with platinum 

 foil and the] two metals were connected by wires with the galvano- 

 meter. On heating a cell of this construction in an oxidising flame an 

 electrical current was almost invariably produced, due to the with- 

 drawal of hydrogen from the inner wire or rod. When the current 

 diminished in force a reducing flame, containing free hydrogen, was 

 applied to the cell. This immediately caused an energetic reverse 

 current, accompanied by the re-absorption of hydrogen by the inner 

 metal. Then with an oxidising flame the original effect could be pro- 

 duced. These results were obtained with wires of platinum, nickel, 

 iron, and copper. In working with cells of this description made with 

 iron rod, it was found that a current of electricity of long duration 

 could be produced by the oxidising flame. This result appears to be 

 due to the continuous absorption of hydrogen liberated from aqueous 

 vapour by that portion of the iron which was not covered by glass. 



When the entire surface of the iron was covered by glass (with the 

 exception of the conducting wire, which was away from the heat) then 

 the deflection of the galvanometer gradually came to zero when the 

 cell was heated in an oxidising flame. 



Recently experiments have been made with a differently arranged 

 apparatus, as follows : — ■ 



A platinum tube, b\ inches long and 1 inch diameter, was set 

 upright in a Fletcher's gas furnace and nearly filled with a fusible 

 glass composed of the diborates of lime and magnesia. This apparatus 

 being heated to bright redness, a plate of platinum, 2 inches long and 

 0*6 inch wide, suspended by a platinum wire, was immersed in the 

 fluid glass. The platinum tube and the plate being connected with 

 the galvanometer, the phenomena of alternate electric currents could 

 be produced with great facility by altering the nature of the flame in 

 the furnace. When the platinum tube was surrounded with a visible 

 pale flame there was an electrical current from the tube to the plate 

 until the plate was apparently saturated with hydrogen. When more 

 air was supplied to the furnace, so as to cause more perfect combus- 

 tion, . the needle of the galvanometer was violently reversed. The 

 deflection produced on the galvanometer by either the " normal " or 

 "reverse" current was at first 18° or 20°, and it fell to nearly zero 

 within ten or fifteen minutes. 



These effects could of course be repeated as often as required. 



It appears quite plain that the hydrogen in flames has a powerful 

 molecular or atomic action. 



If glass be fused in a large platinum crucible heated by flame as in 

 a Fletcher's furnace, bubbles of hydrogen may be observed forming 



