1 4 Prof, Buff on theConductivity of Heated Glass for Electricity, 



while an inch above the heated portion of the tube the glass 

 retained its power of insulation. 



The same tube partially filled with mercury was now caused 

 to dip 80 far into a wider one, which also contained mercury, 

 that the hquid metal stood at the same height within and with- 

 out. The object was to increase the number of points of contact 

 of the exterior surface, also of the tube with a good conductor. 

 The mercury in the interior remained, as before, in connexion with 

 the positive conductor of the machine, the mercury outside com- 

 municated with the end of the multiplying helix of a tangent 

 compass, the other end of which was carried to the negative con- 

 ductor, and also connected with the earth. In the liquid within the 

 tube the bulb of a mercury thermometer was immersed. At the 

 commencing temperature (16° C.) the two metallic coatings could 

 be strongly charged similar to a Leyden jar. Nevertheless a por- 

 tion of the electricity seemed to pass through the glass ; for the 

 galvanometer, which was deflected at first by the outflow of the 

 induced electricity, did not retui-n to its position of equilibrium 

 when the machine was turned in the most regular manner pos- 

 sible. The deflection was augmented when the temperature 

 increased, while the phaenomena of charge perceptibly diminished. 

 At 80° C. the needle stood at 3°; at 120°, at 7°; and finally, 

 at 200° it stood at 7°-5. A greater deflection than this could 

 not be obtained by heating the mercury still more. When the 

 glass was excluded from the circuit, and the contact was esta- 

 blished by the metallic connexion of the prime conductor with 

 the galvanometer wire, the deflection of the needle amounted 

 likewise to 7 '5. It may be seen from this, that at temperatures 

 over 100° almost the entire quantity of electricity was permitted 

 to pass through the glass. The force necessary to overcome the 

 resistance of the glass, the tension on the surface of the conductors, 

 was of course not the same at the different temperatures. When 

 the thermometer stood at 100°, by the approximation of a small 

 sphere of metal, sparks 3 millims. long might still be drawn 

 from the conductor. At temperatures over 120° the passage of 

 the sparks completely ceased. At 180° a very sensitive gold-leaf 

 electrometer, unassisted by a condenser, scarcely showed a per- 

 ceptible charge ; and when the mercury approached its boiling- 

 point, the same electrometer, even with the aid of the condenser, 

 could no more be charged. These facts are perfectly in accord- 

 ance with the diminution of the resistance of the glass by the 

 augmentation of the temperature. 



In consequence of the facility with which the glass permitted 

 the passage of the electricity, it was possible to measure the 

 resistance corresponding to each particular stage of temperature. 

 The apparatus made use of for this purpose was that before de- 



