430 M. F. Strenger on the Conductivity 



a Geissler's thermometer, with which it was possible to read 

 to the 100th of a degree. 



In the beginning of an experiment, F. Weber caused the 

 temperature of the lower surface of the substance to be ex- 

 amined to sink to 0° C, and to remain there by placing the 

 crystal plate and the metal vessel containing the thermo-j unction 

 directly upon a smooth surface of ice. I could not adopt this 

 method of proceeding ; for the weight of the tourmaline slice 

 and the copper cylinder was too small to drive away the water, 

 produced by the melting ice, and to preserve a good contact. 

 I have therefore preferred to leave the tourmaline slice un- 

 changed in its position, and, by a special arrangement, to 

 approach the plane surface of the ice (d) from below, and to 

 press it so tightly as to drive the water away from the neigh- 

 bourhood of the tourmaline plate. 



For this purpose the slice (/) was cemented, with a little 

 wax and rosin, to an ebony plate (e), which was penetrated by 

 a hole of the required size, and this again to a short glass 

 tube (#); the glass tube was then clamped to a retort-stand. 



Vertically under the plate was the shorter arm of a U-tube 

 (A), which communicated with the other by means of a thinner 

 tube provided with a stopcock. In the shorter arm was a 

 test-tube, movable w r ith little friction : upon the upper end of 

 this, directly under the tourmaline itself, a glass plate (k) was 

 cemented. The U-tube was filled with mercury until the glass 

 plate was about 2 centim. under the surface of the crystal ; 

 the stopcock was then closed, and the longer arm filled to the 

 top with mercury. If it was now required to cool the lower 

 surface of the tourmaline plate to 0° C, a piece of ice (about 

 1 centim. thick), possessing a plane surface, was placed upon 

 the glass plate, and the cock opened; the mercury in the longer 

 arm then raised the test-tube, and with it the ice, and pressed 

 the latter firmly against the crystal. 



In order to be able to compare the results of different expe- 

 riments, I kept the temperature of the room as constant as 

 possible ; if the temperature got too high in the course of the 

 afternoon, I stopped the experiment and proceeded with it 

 the next morning. Besides this, sufficient time was allowed, 

 by every measurement, to enable the whole extent of the 

 crystal to acquire the same temperature as the thermoelectric 

 junction (a) and the air of the room. 



Before and after every experiment, I made certain that no 

 electromotive force could come into play other than that caused 

 by the difference of temperature between the copper cylinder 

 and the thermoelectric junction in the oil of turpentine. 

 Lastly, the ice approached the tourmaline so rapidly that, 



