Conductivity in Tourmaline Crystals. 23 



crystallographic axis. The crystal was heated to 100° Centi- 

 grade in a steam-bath. A 5 -microfarad condenser was charged 

 through the crystal with 10 or 12 Daniell's cells for a minute ; 

 and then the condenser was discharged through a sensitive 

 astatic Thomson galvanometer of 7000 ohms resistance. The 

 conductivity of the tourmaline was so slight that the motion 

 of the spot of light was only about two inches. The limit 

 of swing was accurately observed, and then the operation 

 repeated with the tourmaline (electrically) reversed. This 

 was done again and again alternately in opposite directions. 



When the temperature of the tourmaline was rising, a 

 distinct difference was perceived between the to-and-fro dis- 

 charge. Also when it was falling there was a difference, in 

 the other direction. (These effects are of course due to that 

 which is ordinarily termed electromotive force in the warming 

 or cooling crystal.) But with a perfectly steady temperature, 

 which was only attained after some hours, not the slightest 

 difference could be perceived. But before being satisfied 

 with this imperfect and negative result, we should wish to 

 use a battery of very much higher electromotive force (say 

 1000 cells), and, if possible, heat the tourmaline above 100°, 

 so as to increase its conductivity, which at 100° is slight, 



Note added , May 1879. 



Since the above was communicated to the British Associa- 

 tion, a large number of attempts have been made to detect a 

 different conductivity for electricity from a to /3 to that from 

 /3 to « when the temperature of the crystal was uniform. 

 Numerous leakage methods have been employed, the most 

 powerful having been one with a quadrant electrometer and a 

 dry pile. Another slice, larger in area and only a millimetre 

 thick, has been also used ; and the crystals have been tried 

 when cold, when at 100°, and when heated up to 300° in an 

 air-bath ; but although a small difference was many times ob- 

 served, indicating a better conductivity in one direction than 

 in the opposite, yet it always decreased, and usually vanished 

 when excessive care was taken. It was very difficult to be 

 quite sure that the temperature of the crystal when hot was 

 not slightly changing ; and the electromotive force due to such 

 change of temperature was perceptible by the methods em- 

 ployed when its direct effect on a quadrant electrometer was 

 inappreciable ; and when the crystal is cold its conductivity 

 is exceedingly small. It is intended, however, to try the best 

 method once more, and more carefully ; and also it is intended 

 to see if no difference can be perceived between the specific 



