414 M. Verdet on the Optical Propei-ties developed in 



The thickness of the liquids employed varied from 10 to 50 



niillims. . 



The temperature of the laboratory where the experiments were 

 performed remained always between 12° and 18° C. 



The rotation of the plane of polarization produced by a given 

 thickness of a transparent substance subjected to a magnetic 

 action of given strength, is a physical constant as characteristic 

 of the substance as its refractive index or dispersive power. It 

 is therefore advisable that some name be given to this constant: 

 the term "magneto-rotary power" seems to me to be appli- 

 cable. I shall make constant use of this expression in the 

 sequel. Representing the magneto-rotary power of distilled 

 water by unity, the rotary powers of other substances will be 

 represented by their relations to this unit. I may add, in order 

 to give an idea of the magnitude of the phajuomena observed, 

 that when my electro-magnet was excited by a recently fitted up 

 battery of twenty Bunsen's elements, a thickness of 40 miUims. 

 of distilled water, placed between the armatures, which were 

 50 millinis. apart, produced a rotation of about 1° 5'. 



My first task was to test experimentally certain laws which 



the electro-magnet, the following two series of azimuths of the passage- 

 tint were observed : — 



Mean 



Whence 45' 30" is derived as the rotation due to the action of the glass 

 plates. The agreement among the individual observations is sufficient to 

 guarantee the accuracv of this result. I did not consider it essential always 

 to determine each azimuth eight times ; in general, four determinations 

 seemed to me sufficient. ■ 



In my former researches, in operating upon bisulphide of carbon, this 

 correction was omitted ; for, having in fact no other ])urpose beyond the 

 determination of how the action of the sulphide of carbon varied with the 

 magnitude and dircctiou of the magnetic force, it was a matter of indif- 

 ference to me that the action actually measured was the sum of two distinct 

 actions subject to the same law of vai-iation. Moreover, by the an-angc- 

 ment of the apparatus, the correction in point would fall within the class 

 of observation errors ; for, by reason of the great size of the armatures and 

 their distance from the glass plates, the magnetic force exerted ujion these 

 plates was much less than in my later experiments, where a great magnetic 

 energy was concentrated in small armatures, brought as closely as possible 

 to the bodies under exaiaiuutiun. 



