168 Electro-optic Experiments on various Liquids. 



but faint restoration with five turns of the plate, strong "with 

 ten turns °, and the effect was purely positive by both tests. 



Propyl bromide, as a conductor, but without the jar, gave a 

 good effect which was positive by one test, clearly strengthened 

 by tension parallel to lines of force, but not sensibly affected 

 by compression. With the jar, one turn of the plate gave a 

 strong effect, which was purely positive by both tests. 



Isopropyl bromide acted more strongly, giving a clear effect 

 as a nonconductor. Tried as a conductor, with the jar, it 

 gave increasingly brilliant restorations with two, five, ten, 

 or more turns ; and the effect from weak to strongest was 

 purely positive, strengthened by tension, and extinguished by 

 compression. 



Butyl bromide, not a very clean sample, was hardly so strong 

 as the last, giving a mere trace of effect as a nonconductor. 

 Tried as a conductor, with the jar, it gave a good restoration 

 with one turn, strong with three turns ; and the effect was 

 purely positive by both tests. 



Amyl bromide, tried as a nonconductor, was quite inactive 

 at first, giving no trace of effect in the polariscope, no sen- 

 sible spark from the prime conductor ; but Avhen the machine 

 had worked for a little time, the optical effect came out 

 clearly, and was found to be purely positive. Tried then as 

 a conductor, with the jar, it gave a splendid restoration with 

 one turn of the plate. When the jar was detached, well 

 charged, and again connected for discharge through the cell, 

 the light flamed up in the polariscope with great brilliancy, 

 and faded away gradually, remaining visible for sixty or more 

 seconds. The increase of insulating and electro-optic powers 

 of the dielectric in the course of this experiment, and in a 

 few minutes of time, was very remarkable. 



Chlorides, C n H 2)l+1 Cl. 



12. These are purely positive, and stronger than the bro- 

 mides ; and they show a great increase of power as the expe- 

 riment proceeds. 



Propyl chloride, tried in the plate cell as a nonconductor, 

 gave at first no trace of effect in the polariscope, no sen- 

 sible spark from the prime conductor. Tried then as a 

 conductor, it gave a clear and strong restoration at each dis- 

 charge. When the jar was added, one, two, three, or more 

 turns gave increasingly brilliant effects, which were found to 

 be purely positive by both tests. When the observations had 

 been carried on for some time, the liquid was tried again as 

 a nonconductor. The prime conductor gave now a spark 

 fully a quarter of an inch long ; and the polariscope gave a 



