100 Dr. J. Kerr's Electro-optic 



short, not more than an eighth of an inch, when those from the 

 prime conductor were about an inch. The sparks were ac- 

 companied by an almost continuous effect in the polariscope, 

 which improved as the knuckle approached the ball. At con- 

 tact of ball and knuckle the optical effect was continuous, and 

 certainly at its greatest intensity. In these two liquids, and 

 in all the other insulators that gave any distinct effect, the 

 phenomenon was apparently of the same kind throuo-h its 

 whole progress, and very unlike that obtained from nitro- 

 benzol. 



27. Stannic Chloride (SnCl 4 ). — I was sorry to find, on trial, 

 that my method is quite inapplicable in the case of this in- 

 teresting compound, one of the best insulators known anion cr 

 liquids. Still the trial ought to be described, as it gave me a 

 good glimpse of what I believe to be a new fact. The liquid was 

 let into the cell as rapidly as possible, through filtering-paper 

 and funnel. During this brief exposure to the air, which 

 was unfortunately rather moist at the time, the liquid gave 

 off a dense white cloud of suffocating fumes. The liquid, as 

 it rose in the cell, was fairly transparent; but above it there 

 lay a deep shell of dense white froth. The charging of the 

 cell was completed in several seconds. The most of the froth 

 had run over the mouth of the vertical boring, and was rapidly 

 wiped away : but some of it had evidently dissolved in the 

 liquid, giving it a uniformly milky or misty appearance, 

 deadening the transparency without seriously impairing it. 

 As all the pieces had been already put in final position for the 

 electro-optic experiment (8, 9), the second ^Ticol was now 

 turned at once to good extinction, and the machine was set in 

 motion. At the first turn of the plate there was a vivid resto- 

 ration in the polariscope, stronger a great deal (such at least 

 was my impression at the time) than any thing of the kind 

 that I had yet seen, even in the case of CS 2 . The hand com- 

 pensator was immediately applied in the usual way (4), but 

 without a trace of the ordinary effect : the light restored by 

 electric force was not sensibly weakened either by horizontal 

 tension or by horizontal compression. This result was so un- 

 expected, and, indeed, so extraordinary, that I had to spend a 

 little time in making sure there was no mistake. Trying 

 other means, I soon found that the effect of electric force was 

 neutralized, either perfectly or very nearly, by a definite ro- 

 tation of the second JSTicol through a small angle ; and up to 

 this point I am very confident of the facts. 



By this time the liquid had evidently deteriorated, not as an 

 insulator, but as an optical medium ; it had become in appear- 

 ance faintly discontinuous, partly speckled and partly reticu- 



