Transparent Bodies by the action of Magnetism. 483 



tioned, the resultant would remain almost constant in magnitude 

 and direction. It is for experiment to show what case satisfies 

 this condition. 



The accuracy of these observations is evidently in no respect 

 subordinate to the particular law of magnetic action, which is 

 the law of the square of the distance. Consequently, without 

 knowing the law according to which the optical properties of 

 transparent substances develope themselves under the influence 

 of magnetism, since we know that this law involves a variation 

 with the distance, we may presume that, in placing a transparent 

 substance between two armatures similar to those already men- 

 tioned, compensations of the same kind will be established for 

 the different points of the substance, and the several infinitely 

 small plates of which we can conceive it formed, will all acquire 

 optical properties sensibly identical. Experiment completely 

 confirms this supposition. 



I have succeeded, in fact, without difficulty in fulfilling the 

 conditions in question by furnishing one of Ruhmkorff's electro- 

 magnets with suitable armatures. This electro-magnet was 

 formed of two cylinders of soft iron, AB, A'B', m- 20 long 

 and m, 075 in diameter, with a narrow canal bored in the 

 direction of their axis to allow passage to the light ; each cylinder 

 was surrounded with about 250 metres of copper wire of 2'5 

 millims. diameter, and joined by the pieces of iron P and P' 

 (Plate V. fig. 1)*; these pieces are made to slide at pleasure 

 along the piece of iron US, and can be fixed in any position by 

 means of the screws V, V. On the two extremities B and A' 

 of the horizontal branches I screwed two cylinders of soft iron, 

 P, F', m -05 in height, and m, 14 in diameter, bored along 

 their axes by a narrow canal ; and I found that when the 

 distance between the terminal faces of these armatures was 

 neither too great nor too small, when it amounted, for instance, 

 to 50 or 90 millims., a transparent substance placed in the in- 

 termediate space acquired the same optical properties, whatever 

 was its situation, provided it was not extremely near one or the 

 other of the armatures. In fact, allowing a pencil of solar light 

 to pass through the apparatus, I placed across the course of the 

 pencil a parallelopiped. of heavy glass, and having excited the 

 magnetism of the electro-magnet, turned the birefracting prism 

 which served as analyser until I perceived the violet tint known 

 to physicists under the name of the sensible tint or tint of passage ; 

 I then displaced the parallelopiped of heavy glass parallel to itself, 



* Fig. 1 being an elevation of the apparatus, the two cylinders of soft 

 iron are of course concealed by the copper wire which covers them ; the 

 internal canal is merely indicated by the lines, and the axis of the canal by 

 »n interrupted dotted line MN. 



212 



