Prof. Buff on the Conductivity of Heated Glass for Electricity. 15 



scribed j the closed end of the glass tube chosen for the experi- 

 ments was almost perfectly spherical ; the height of the interior 

 column of mercuiy after the introduction of the thermometer 

 was carefully measured, and before commencing the experiments, 

 the surfaces of the metal within and without the tube were brought 

 exactly to the same level. By means of a cork which embraced 

 the small tube, and fitted into the exterior one, the necessary 

 position was easily secured. 



The heating was effected by means of an Argand lamp, above 

 the glass chimney of which the apparatus was placed and well 

 insulated. The heat generated by this lamp, the wick of which 

 could be regulated with the greatest nicety by a suitable screw, 

 was sufficiently great to elevate the mercury to its boiling tem- 

 perature, while at the same time any lower temperature was more 

 easily obtained than with the flame of the spirit-lamp. 



The mercury was pure, dry, and besides this boiled in the 

 apparatus, for the purpose of establishing the most intimate con- 

 tact possible with the surface of the glass. The connexions, as 

 before, were established by platinum wires which dipped into the 

 mercury. 



This apparatus formed at the same time, and in connexion 

 with a very delicate tangent compass, the circuit of a single ele- 

 ment of Daniell. The circuit closed by the tangent compass 

 alone produced a deflection of 79°. 



From some preliminary experiments it was observed, that 

 during the passage of the current through the glass, the needle 

 of the galvanometer assumed only a transitory deflection, which 

 soon began slowly to approach zero, just as occurs when a liquid 

 between two platinum plates is introduced into a circuit of small 

 electromotive force. In the latter case the effect is a consequence 

 of polarization, that is, a consequence of the action of an op- 

 posing electric force which is gradually developed. The glass 

 apparatus also had evidently assumed a charge which acted in 

 opposition to the electromotive force of the circuit ; when 

 the voltaic element was excluded, and the coatings connected 

 directly by a wire, a transitory deflection in the opposite direc- 

 tion to the original one was obtained. In order to avoid as much 

 as possible these disturbing influences, we must limit ourselves 

 to the measurement of the first impulsion of the needle. But 

 as, during even this period, the opposing force, though in a very 

 small degree, had begun to develope itself, the precaution was 

 taken to conduct the current first from the interior of the glass 

 to the exterior, and afterwards in the opposite direction, and to 

 take the mean of the two deflections thus obtained. The differ- 

 ence amounted in some cases only to a few tenths of a degree ; 

 but when the deflections were large, amounted to 1°*5 and even 



