90 Dr. J. Kerr's Electro-optic 



lvzer Q is then laid so as to suit the observer's eye, and is 

 turned into the position of perfect extinction ; and this being 

 done, the two Nicols are left untouched, if possible, till the 

 end of the experiment. The cell A, with conducting wires led 

 from its outer balls to prime conductor and earth respectively, 

 is now put in position under direction from the observer, 

 who sits at the polariscope, and restores the light by the use 

 of the hand compensator. When A is well placed, the line 

 LM is perpendicular to the plate-faces of the cell, and the 

 object restored by the hand compensator is a fine streak of 

 light, passing midway between the balls in the cell, and pro- 

 jecting well above and below them. When this has been done 

 once for all, the cell is left unmoved throughout the experi- 

 ment. The fixed compensator B, containing either one plate 

 or two, is nowplaced carefully, so that the line LM passes through 

 the centre of each plate in a direction perpendicular to its faces. 

 The last piece C is not required always, but only when the 

 purity of initial extinction in the polariscope has been in any 

 degree lost by the introduction of A or B ; it is a piece of com- 

 mon plate, the more irregular in temper the better, about J of 

 an inch thick, fixed in a movable stand ; its modes of application 

 and action are precisely the same as those of the larger neu- 

 tralizing plate described in my first paper. When things have 

 been thus arranged, the cell is charged with clean liquid (2), 

 the observer sits at the polariscope, and, the initial extinction 

 being still unimpaired, the electric machine is set in motion. 



Definite Chemical Compounds. 



8. Carbon Disulphide as a Dielectric. — My first trial of 

 the new cell was w T ith this liquid, which is much the 

 best dielectric yet discovered. The arrangements and proce- 

 dure are exactly as described in the last two articles; the cell 

 is charged with perfectly clean carbon disulphide, and the 

 initial extinction is perfect. A small movement of the machine, 

 one turn or less, gives a very fine restoration of the light in 

 the polariscope. As the potential rises, the light increases 

 steadily till it is quite brilliant ; but if a spark be taken upon 

 the knuckle from the prime conductor at any point in this 

 process, the phenomenon vanishes instantly. 



0. Character of the Optical Effect. — The preceding experi- 

 ment is repeated, with the addition of the hand compensator 

 (4). The compensating slip is put in position with its length 

 horizontal, and the initial extinction is found to be unimpaired ; 

 the light is then restored steadily by electric action, and the 

 slip is successively stretched and compressed with forces in- 



