252 Dr. J. Ken-. Experiments on a [Mar. 15, 



III. " Experiments on a Fundamental Question in Electro- 

 Optics : Reduction of Relative Retardations to Absolute." 

 By JOHN KERR, LL.D., F.R.S. Received March 9, 1894. 



To prepare the way, I begin by recalling these well-known facts : 

 that when light passes through an electrostatically strained medium 

 in a direction perpendicular to the line of electric force, it undergoes 

 a uni-axal double refraction, the optic axis coinciding with the line 

 of force ; that with reference to this action, dielectrics are divisible 

 into two classes, the positive* and the negative,f which are optically 

 related to each other in the same way as the positive class of crystals 

 to the negative; that the intensity of the action, or the quantity of 

 optical effect per unit thickness of the dielectric, is measured by the 

 product CF 2 , where C is a constant which is characteristic of the 

 medium, and P is the value of the resultant electric force : that the 

 effects are generally observed and examined still as they were dis- 

 covered first, by simple experiments with a pair of Nicol's prisms and 

 a slip of strained glass or other phase-difference compensator. 



In every such experiment the effect specified by the compensator 

 is a difference of phases, or a relative retardation ; and we may there- 

 fore view it as a resultant effect that is to say, as the resultant, 

 or the difference, of electrically-generated absolute retardations of 

 two component lights, whose planes of polarisation are parallel and 

 perpendicular to the line of electric force. What, then, are the values 

 of these two absolute retardations in any given case ? What are the 

 two absolute components of any electrically-generated relative retarda~ 

 tion? Such is the question here proposed for solution by ex- 

 periment. 



As long ago as 1882, and several years following, I was much 

 occupied at intervals with this interesting question. In the summer 

 of 1885, in some experiments with the dielectric CSj, I obtained 

 results as decisive as could be desired. Other dielectrics, both solid 

 and liquid, were tried afterwards, but only with partial success, the 

 experimental difficulties being, in some cases, too much for my 

 methods and time. To these cases I shall make no further reference, 

 as I will keep to the one line of experiment, and to those experiments 

 in particular in which the indications were quite regular and unmis- 

 takeable. With these limitations, the inductions extend to four liquid 

 dielectrics, two positive and two negative ; and all the experiments 

 point clearly in one direction. 



General Result. It appears that the proper and immediate optical 



* Carbon diaulphide, the hydrocarbons, Ac. 

 t Amjl oxide, the heavy oils, Ac. 



