344 Dr. J. Kerr on a new Relation between 



is then left untouched, as well as the other optical pieces, till 

 the end of the experiment. The electric action being still kept 

 up, the distance of the spark-terminals is suddenly increased up 

 to 5 or 6 inches ; the light soon reappears and is allowed time 

 to come out distinctly. When the primary circuit is now 

 broken, the light fades away to extinction, and afterwards re- 

 appears ; but before and after this passing extinction the light 

 has contrary characters, as tested by the compensator, being 

 extinguished in the former case by tension parallel to the lines 

 of force, and in the latter case by compression. 



17. From the last three experiments, or simply from the first 

 of them, interpreted by the illustrative optical experiment (12), 

 we infer that the dielectrization of plate glass is equivalent opti- 

 cally to a compression of the glass along the lines of electric 

 force. Dielectrified glass acts upon transmitted light as a ne- 

 gative uniaxal with its axis parallel to the lines of force. 



18. Eighth experiment. — All the preliminary arrangements 

 are as in the fifth experiment (14). A strong electric action 

 (spark of 6 inches) is kept up without ceasing for 20 minutes, 

 and all the optical pieces are left untouched for an hour, the in- 

 duction-terminals being connected with each other through the 

 secondary coil from first to last. One thing very noticeable 

 under such conditions is, the length of time which the light takes 

 to fade away to extinction. In the present experiment there is 

 a distinct, though faint, effect in the polariscope, even as long as 

 30 or 40 minutes after the external electric action has ceased. The 

 effect has a constant character from first to last : the light re- 

 stored by electric action is always extinguished perfectly by a 

 right degree of tension of the compensator in a direction parallel 

 to the lines of force. With the standard compensator (13), 

 there is not a very great effort required for extinction at any 

 time during the whole hour of observation. 



19. Ninth experiment. — The analyzer is mounted in such a 

 way that it may be moved in different directions at right 

 angles to the beam, without rotation; the other arrangements 

 are as in the fifth experiment. When a good effect has been 

 obtained through the centre of the electric field, the second 

 Nicol is moved about so as to receive the light through different 

 parts of the dielectric. 



Keeping first to the perpendicular bisector of the line joining 

 the terminals, the intensity of the optical effect diminishes as 

 the distance from the centre of the field increases : and this is 

 particularly noticeable at first, while the effect is rising ; but the 

 action has always the same character, is always neutralized by 

 horizontal tension of the compensator. As far as my means 

 would allow, I have assured myself of the fact that, as soon as 



