Electricity and Light. 451 



force are more nearly parallel or perpendicular to the plane of 

 polarization throughout the part examined of the electric field. 



30. Second experiment, — All the arrangements are as in the 

 first experiment, except that the optical pieces are brought as 

 close together as the stands will allow. The dielectric is only a 

 few inches from the eye. The light is large and strong, and 

 extends well above and below the level of the terminals. All 

 the observations made in the first experiment are repeated, and 

 with similar results. 



The dark band given by the compensator is now more notice- 

 able, and more regular in its appearance and movements. 



If the liquid contains a few specks of dirt, this experiment 

 succeeds better than the first, sometimes indeed gives a good 

 result when the first fails altogether. The solid particles gene- 

 rally arrange themselves as a chain between the terminals. In 

 the first experiment, the principal effect is obscured and often 

 obliterated by the light from the chain ; while in the second 

 the light is generally well restored above and below the level of 

 the terminals, the place of the chain being marked by a patch of 

 more intense light, which is not affected by the compensator. 

 In this experiment and the preceding, the dependence of inten- 

 sity of optical effect upon intensity of electric action is very ob- 

 servable. When the insulation of the prime conductor is not 

 very good, it requires little irregularity in the working of the 

 machine to make the light clearly glow and fade again with each 

 turn of the plate. 



31. Third and fourth experiments. — The flame is turned 

 broadside-on to the polariscope; and a screen is placed close to 

 the lamp, which reduces the light to a narrow horizontal band : 

 the arrangements are otherwise as in the first and second expe- 

 riments. The results obtained are consistent with those already 

 described, and add little to them. But as I would rather give 

 too much here than too little, I add the final note which was 

 taken at the time of observation. 



The liquid very clean ; the light reduced to a narrow and 

 sharply defined horizontal band, encroached upon at both ends 

 by the terminals ; initial extinction in the polariscope perfect. 

 At 45° one gentle turn of the plate gives a good restoration of 

 the band from terminal to terminal, and the restored band is 

 well extinguished by horizontal compression of the compensator : 

 at 0° and at 90° the same electric force gives no sensible effect 

 whatever. When the electric action is intense, the effects are 

 somewhat similar : at 45° the restored band is very well extin- 

 guished by horizontal compression of the compensator, except at 

 points quite close to the terminals ; at 0° and at 90° the wires are 

 tipped each with a narrow band of light, the intermediate part 



2 G2 



