assumed by Insulators at High Temperatures. 199 



were polarized. The same experiment was made with frictional 

 electricity. Here also a polarization immediately ensued, which 

 was similar to the one observed by Henrici* when passing the 

 electricity of a machine through water. In order to determine 

 whether to ascribe the counter currents which were formed to 

 an electrolytic action, or to a deportment similar to that shown 

 by imperfect conductors, a piece of water-glass, after the current 

 of a battery had been passed through it for some time, was 

 broken loose from its supports. Before, the glass had a weak 

 alkaline reaction j this reaction was now no longer manifest at 

 the end of the bar which had rested on the positive plate, but at 

 the other it was strong. This substance, therefore, had deported 

 itself, even before fusion, exactly as an electrolyte would have 

 done. 



After this I proceeded to experiment upon common glass : 

 the form generally employed was that of a thin tube, into which 

 two platinum wires were fused so as not to touch one another, 

 in order to avoid every external influence. When a gentle heat 

 was required, the sand-bath was used; when a strong one, the 

 immediate action of a flame. The sand-bath could be well em- 

 ployed in all experiments; for the sand, even when strongly 

 heated, conducted so badly, that it was only able to discharge a 

 charged electroscope very gradually. The several kinds of glass 

 which were employed began to conduct the galvanic current 

 when between 200° and 220° C. (Becquerelf found this to occur 

 only at 300°), when the wires became strongly polarized. In 

 order to find whether an electrolytic action was also present in 

 this case, the experiment was made in the same manner as with 

 the water-glass ; a small glass bar was fused fast to two strips of 

 platinum, and broken off after it had been exposed for two hours 

 to the action of a current from a battery of six or ten elements ; 

 the end which had lain on the negative plate was rubbed smooth 

 and moistened with water : in two out of seven experiments a 

 weak basic reaction was detected. As this experiment was 

 not decisive, the glass bar, which had been thus fused to the two 

 plates and exposed to the action of the current, was broken off 

 and replaced by a fresh one after the battery had been withdrawn 

 from the circuit. A polarization-current was always obtained 

 when this bar was heated, thus proving that chemically-opposite 

 changed substances had remained behind on the plates. Fric- 

 tional electricity also presented the same phsenomena of polar- 

 ization, in a much less degree it is true, but still always in the 

 proper direction. In these experiments, however, the action of 

 thermo-electric currents, which are produced by unequally heat- 



* Poggendorif's Annalen, vol. xlvi. p. 585^ 

 t Comptes Rendus, vol. xxxviii. p. 905. 



