482 On the Rotation of the Plane of 



The employment of spars and solai' light permits of re- 

 moving the electro-magnets to a great distance from the 

 ihernio-electric pile. With respect to the arrangement of 

 the prisms, the law of Malus shows all the advantages which 

 it presents. In fact, let us take for unity the deviation wliich 

 the solar ray transmitted through the principal parallel sec- 

 tions would produce. The deviation, when the prisms form 



an ancle of 45°, will be cos^45°=— -. If the current is set in 



action, and it produces a rotation of the plane of polarization 

 equal to 8, the deviation will be, according to the direction of 

 the current, cos^(45°— 8) or cos^(45°-fS), and we shall then 

 have, for the difference of the effects observed when the cur- 

 rent is made to pass in a contrary direction, 



cos2(45°-8)-cos2(45° + 8) = sin28. 

 On placing the principal sections at 90^^', the difference of 

 the deviations would be only 



cos2(90°-8) - cos2 90''= sin^ 8, 

 or 



cos2(90° + 8) - cos290''= sin^S. 



Now sin^8 is considerably less than sin 2^. If, for example, 

 we suppose 8=8°, sin 28 is equal to more than fourteen times 

 sin 28. 



The eye, it is true, appreciates readily the transition from 

 darkness to light, but not so the difference in brightness of 

 two luminous images. This is not the case with the ihermo- 

 scopic apparatus. There is therefore, when heat is concerned, 

 a great advantage in proceeding as above directed. 



The following are the details of the experiment: the solar 

 ray, reflected by a heliostat, traverses at first a doubly re- 

 fracting achromatic prism. The extraordinary bundle was 

 intercepted : the ordinary bundle traverses the electro-magnet 

 of M. RuhmkorfF'sapparatus,and aflint-glass ofSSmillimetres 

 in thickness between the poles of the electro-magnet. It after- 

 wards encounters, at about S'^'SO, the second prism of spar, 

 bifurcates again, and gives two images, one of which may be 

 received on the thermo-electric pile placed at four metres from 

 the electro- magnet. The galvanometer was still a little further 

 removed from this disturbing force. It was ascertained, by 

 direct and repeated experiments, that on establishing the cur- 

 rent there were no phsenomena of induction, and that the 

 electro-magnets had no appreciable action on the magnetic 

 needle which, under their influence, remained at zero in a state 

 of perfect rest. In order to understand this, it must be borne 

 in mind that the two opposite poles are very close together, 



