484 M. Verdet on the Optical Properties developed in 



in order that the light might always pass through the same 

 thickness; as long as I did not bring it in contact with one or 

 the other armature, the tint of passage did not suffer any modi- 

 fication. The experiment succeeded as well with the other sub- 

 stances which I have used in my researches. 



Thus, in my apparatus, the space comprised between the two 

 terminal armatures was so constituted, that a piece of transparent 

 substance placed in any point of this space, excepting in the 

 vicinity of the edges, was everywhere affected in the same manner. 

 Hence the optical properties of the fragment were the same in 

 all points of its mass, and consequently they were the same 

 (excepting the magnitude of the rotation) as those of an infinitely 

 small element. At the same time this space was what Faraday 

 terms a magnetic field of equal intensity. In fact, by employing 

 a method which will be indicated further on, I found that the 

 resultant of the action which the electro-magnet would exert on 

 a molecule of magnetic fluid situated in any point of this space, 

 was sensibly constant in magnitude and direction ; as long as the 

 molecule was not very near the limits, e. g. in a series of expe- 

 riments where the distances of the terminal faces of the arma- 

 tures were successively 50, 60, and 90 millims., I measured the 

 resultant before defined at the centre C of the intermediate space, 

 fig. 1, and found the numbers 134-12, 116-33, and 86-17. I 

 measured it at the point D situated on the axis, and 15 millims. 

 distant from one of the armatures, and found the numbers 

 133-87, 116-75, and 86-00 : I measured it at a point E 25 mil- 

 lims. from the axis and in the plane, and found the numbers 

 133-5, 116-00, and 85-5. 



The differences of the values corresponding to the same position 

 of the armatures are sufficiently small to allow the conclusion, 

 that, in the interior of a region whose dimensions do not exceed 

 those of the transparent bodies submitted to experiment, the 

 resultant of the actions which would be exercised on a mole- 

 cule of free magnetic fluid does not vary one-hundredth part of 

 its value, which fraction about represents the degree of accuracy 

 attainable in these experiments. 



For the sake of brevity, the resultant before defined will be 

 hereafter called the magnetic action in a. given point. 



This simultaneous constancy of the magnetic action and of the 

 optical properties would naturally lead to a simple conjecture, 

 which experiment has entirely confirmed, and whose confirmation 

 removes the second difficulty which I before noticed. We might, 

 in fact, ask if the magnitude of the rotatory power developed in 

 a transparent substance did not depend solely upon the magni- 

 tude of the magnetic action considered in the space occupied by 

 the substance. In fact, these two quantities are simultaneously 



