80 Intelligence and Miscellaneous Articles. 



purpose the exciter was placed so that the spark, by means of a 

 suitable optical system, illuminated a vertical slit. A fixed mirror 

 sent back the light proceediug from this slit upon the rotatiug 

 mirror, and thence into the telescope. The movable mirror being 

 at rest, the vertical images of the two slits were seen in the field 

 distinctly. By regulating the position of the fixed mirror, these 

 two slits, which had the same breadth, were brought to be each 

 precisely in the prolongation of the other. 



During the rotation of the mirror, at the moment when the bat- 

 tery is discharged, each of the images dilates in the horizontal direc- 

 tion. Two systems of bauds alternately luminous aud dark are 

 thus seen one above the other : the one is due to the bght of the 

 spark ; the other proceeds from the polarizing-apparatus. 



The experiment shows that the brilliant bands of one of the systems 

 forms exactly the prolongation of the luminous bands of the other, and 

 that it is the same with the dark bands. If the mirror be rotated 

 more and more rapidly, the breadth of the bands increases, but the 

 correspondence of the two systems still remains perfect. Therefore, 

 with the close approximation which our apparatus permits us to 

 obtain, we may conclude that the two phenomena, electrical and 

 optical, are simultaneous. 



In order to measure that approximation, we slightly displaced 

 the fixed mirror, so as to destroy the correspondence of the two 

 images. It is clear that the displacement produced the same effect 

 as any delay that might have existed between the two orders of 

 phenomena. We thus secured that a delay of 7 u l ou of a second 

 should be quite appreciable ; we can therefore affirm that the delay, 

 if it exists, is less than 3 * v of a second. 



The experiments were made with heavy flint glass and bisulphide 

 of carbon successively as the transparent body. M. Yillari, by 

 causing a cylinder of flint glass to rotate between the poles of an 

 electromagnet*, ascertained that at a sufficient velocity the pheno- 

 menon of rotatory polarization ceases to exist ; hence he concluded 

 that, to produce magnetization of the flint glass, a time comprised 

 between 0-001244 and 0-00241 of a second is required. Now the 

 sensitiveness of our method permitted us to appreciate a displace- 

 ment corresponding to a forty -fourth part of that time. 



An unpublished experiment of MM. P. Curie and Ledeboer 

 accords with our conclusions. Substituting for the copper disk of 

 Toucault's apparatus a glass disk, and causing it to rotate at the 

 rate of a hundred turns per second, they observed no diminution 

 in the rotation of the plane of polarization. It seems, then, that 

 another explanation of the very interesting experiment of M. 

 Yillari must be sought. 



"We are moreover in accord with him upon this point, that the 

 rotation of the plane of polarization ceases at the same instant as 

 the electric action. — Comptes Rendus de VAcademie des Sciences, 

 June 12, 1882, t. xciv. pp. 1590-1592. 



* Fogg. Ann, cxlix. p. 324(1873). 



