508 M. Verdet on the Optical Properties developed in 



action, there is a certain rotation of the plane of polarization 

 produced ; when these two rotations are perpendicular to one 

 another, the rotation is zero ; when the angle enclosed by both 

 directions varies from 0° to 90°, the rotation decreases in a con- 

 tinuous manner. This general fact once established, all that 

 remains is to determine, by exact measurements, the law of the 

 decrease of the rotations. The apparatus and optical methods 

 described in my preceding memoir remain, therefore, entirely 

 applicable. For reasons stated in another place, I have always 

 preferred the use of white light and the observation of the tint 

 of passage, to the use of homogeneous light. 



The measurement of the phenomena have again led me to a 

 very simple law. The rotation of the plane of polarization is 

 proportional to the cosine of the angle enclosed between the direction 

 of the luminous rays and that of magnetic action; consequently 

 it is proportional to the component of magnetic action parallel 

 to the direction of the luminous rays. 



I have experimented on the same substances as in my former 

 researches, i. e. on heavy glass, flint, and bisulphide of carbon. 

 The bisulphide of carbon was always enclosed in the same small 

 vessels described in the first part of this memoir ; the specimen of 

 flint was the same as that I had already used, but besides the 

 two specimens of heavy glass, of which I have before given the 

 dimensions, I was enabled, through the kindness of Mr. Faraday, 

 to add a third of somewhat smaller size. 



In the following tables I shall merely record the results of 

 two experiments, both of which were made by the method of 

 observing the tint of passage ; in these tables « represents the 

 angle enclosed by the rays of light and the direction of magnetic 

 action, R and R' the complete* rotations corresponding, respect- 

 ively, to positions on the right and on the left of the luminous 

 rays ; M the mean of these rotations ; Q the ratio of this mean, 

 expressed in minutes, to the cosine of the angle a. ; M. l the calcu- 

 lated values of M, supposing the law of the cosine to be exact, 

 and adopting, for the ratio of the rotation to the cosine of the 

 angle a, the mean of the values of Q ; and A the difference 

 between M and M'f. 



* I repeat that by this term, the difference between the two azimuths of 

 the tint of passage, corresponding to two opposite directions of the electric 

 current, is intended. As in my former researches, each of these azimuths 

 was determined from four separate observations made at intervals of five 

 or six minutes at most. 



f In order to avoid the effect of variations in the electric current, I have 

 adopted the method of alternatives ; the rotation corresponding to x=z0 

 being always taken as a term of comparison ; the values in the table are 

 those which would have been observed had the rotation corresponding to 

 «=0 remained constant. I have calculated them by means of a proportion. 



