506 M. Verdct on the Optical Properties developed in 



m, 015 thick, fig. 4, Plate V*. The apparatus was mounted upon 

 a copper pedestal which rested upon four screws V, V, V", V", 

 so that it could turn around a vertical axis passiug through the 

 middle of the transverse bar AB ; the vernier C, moving upon 

 the surface of the graduated horizontal circle DE, indicated the 

 horizontal displacement of the system. On the upper horizontal 

 faces of the vertical branches two armatures of soft iron were 

 fixed, whose disposition will be easily understood from the figure. 

 Two prismatic pieces F and F' glided along two grooves G and G', 

 and could be fixed in any desired position by means of the screws 

 U and U'. At their extremities these two pieces carried two plates 

 of soft iron HK and H'K', each m> 016 long, m -04 broad, and 

 m, 005 thick. The two edges of these plates were exactly parallel, 

 and, when they were withdrawn from each other to a convenient 

 distance, it was easy to show by the method indicated in the pre- 

 ceding memoir, that the magnetic action was sensibly the same 

 at all points of the rectangular space HK H'K', as well as at a 

 small distance above and below the same; between the same 

 limits the optic action was also invariable. In the middle of the 

 space between the two branches of the electro -magnet was a ver- 

 tical copper bar which reached almost to the level of the upper 

 surfaces of the armatures, and carried at its extremity a hori- 

 zontal plate L, graduated at the circumference. Above this plate 

 was a second plate O, which turned on a vertical axis coincident 

 with the axis of rotation of the electro-magnet, and was pro- 

 vided with a vernier which indicated tenths of a degree. Upon 

 this second plate, and resting on a small ledge R, the transparent 

 substance was placed. A vertical rod fixed to the side of the 

 transverse bar AB carried a pointer S at the same height as the 

 graduation upon the plate L. Two other vertical rods, fixed to 

 the foot of the apparatus and independent of the electro-magnet, 

 carried two blackened screens 15 centimetres in diameter. At 

 the centre of one of these screens was a birefracting prism T, 

 which polarized the incident light ; in the centre of the other 

 screen an aperture 3 millimetres in diameter was pierced, merely 

 to allow a narrow cylindrical pencil of rays to pass through f. 

 The analysing apparatus, which consisted of the small lunette 

 used in my former investigations, was completely independent, 

 and placed at some distance. 



The experiments were conducted as follows : — A ray of solar 

 light, which was reflected by a heliostat and polarized in passing 

 through the birefracting prism T, was made to traverse the trans- 



* The figure represents a view looking down upon the apparatus. The 

 ray of light MN and the direction of magnetic action PQ make an angle 

 of 30°. 



t The apparatus was constructed by M. Ruhmkorff. 



