Experiments on various Liquids. 255 



This dielectric reminded me of nitrobenzol ; but the action 

 was much less abrupt, and therefore more easily characterized. 



Azobenzol, C l2 H N 12 , fuses at 65° C. into a clean and trans- 

 parent liquid of a deep red colour. Tried in the fusion-cell 

 as a conductor, with increasing charges of the jar, it gave a 

 series of fine and increasingly brilliant effects, all purely posi- 

 tive. 



DipheRylamine, C 12 H N n , fuses at 54° C. into a transparent, 

 colourless, and very volatile oil. Tried in the fusion-cell as a 

 conductor, with moderate charges of the jar, it gave a pretty 

 strong effect which was purely positive. With strong charges, 

 the regular effect was masked in some degree by movements 

 in the liquid. 



Dimethylaniline, C 8 H n N, is purely positive. Tried in the 

 plate cell as a conductor, without the jar, it gave a good 

 restoration, which was too quick to be characterized. With 

 small charges of the jar, the effect was very strong and purely 

 positive, strengthened by tension, and extinguished by com- 

 pression. 



Conine gave no trace of effect, even with the strongest dis- 

 charges of the coil. My sample was almost opaque, and cer- 

 tainly far from pure. 



Compounds containing CI or Br. 



25. Benzyl chloride, C 7 H 7 CI, is clearly positive, but acts 

 only as a conductor. Tried with the jar, it gave no effect 

 with 5 turns, a good effect with 10 or more turns; strengthened 

 always by tension, and sensibly weakened by compression. 

 My sample was not very clean. 



Chlorobenzol, C 7 H 6 Cl 2 , is purely positive. Acted well, 

 though feebly, as a nonconductor, giving an effect which was 

 strengthened by tension and extinguished by compression. 



Ethylene chloride, C 2 H 4 Cl 2 , is uncertain. It did give a 

 faint and apparently regular effect with strong discharges \ 

 but the disturbance (as in the formic ethers) was excessive, 

 and made the determination of the sign impossible. The re- 

 sults obtained with the next liquid make it probable, if not 

 certain, that the failure in the present case was due to im- 

 purity of the sample. 



Ethylene bromide, C 2 H 4 Br 2 , is purely and strongly positive. 

 Tried as a nonconductor (spark §■ inch), it gave a fine steady 

 restoration, which was strengthened by tension, and extin- 

 guished perfectly by compression. As a conductor, with in- 

 creasing charges of the jar, it gave a series of fine effects, 

 which rose through a large range of intensity, always purely 

 positive. 



