430 M. H. Becquerel's Experimental Investigations 



IV. Relation between the Magnetic Rotatory Powers of Gases 

 and their Indices of Refraction. 

 I have demonstrated that a remarkable relation exists be- 

 tween the rotations of the planes of polarization of light tra- 

 versing non-magnetic solid and liquid bodies and the indices 

 of refraction of these bodies for the same luminous rays. It 

 was of the greatest importance to prove whether such a relation 

 still existed in the gaseous state; and I took the opportunity of 

 saying at the beginning of this memoir, that I was guided by 

 these considerations in the construction of the apparatus em- 

 ployed in the researches. The experiments just described 

 have justified my expectations. 



When the magnetic rotations obtained for the different 

 gases are compared with the indices of refraction of the latter 

 for the yellow light of soda, it is seen that the rotations increase 

 regularly with the indices of refraction. In the table which 

 follows, I have placed opposite to the magnetic rotations of 

 the gases their indices of refraction: — 



Magnetic Rotations of Gases compared with liquid Carbon 

 Bisulphide. 



G-as. 



1. 



Magnetic 



rotatory 



powers, 



E. 



2. 



Magnetic 

 rotations. 



3. 



Indices of 



refraction, 



n. 



4. 



5. 

 E 



n\ri 2 -\y 



Oxygen 



0-000146 

 0-000159 

 0-000161 

 0-000302 

 0-000393 

 0-000730 

 0-000802 



0-918 

 1-000 

 1012 

 1-900 

 2-471 

 4-591 

 5044 



1-0002706 



(Mascart) 



10002936 



(Mascart) 



1-0002977 



(Mascart) 



1-0004544 



(Mascart) 



1-0005159 



(Mascart) 



1-0006650 



(Dulong) 



1-0006780 



(Dulong) 



0-850 

 1-000 

 1-027 

 2-393 

 3-086 

 5-130 

 5-327 



0-269 

 0-277 

 0-274 

 0-332 

 0-381 

 0-548 

 0590 



Air 



Nitrogen 



Carbonic acid . . . 

 Nitrous oxide ... 

 Sulphurous acid. . . 

 Olefiant gas 



In order to make a comparison between the numbers in the 



given smaller numbers than those previously published ; and the results at 

 which they arrived do not agree any better with mine. The number that 

 they attribute to air (0-0001 27) is identical with that which I had pre- 

 viously published in a work on atmospheric polarization {Annates de 

 Cliimie et de Physique, t. xix.). I observed at the time that this num- 

 ber only approximated within two or three tenths of its value ; it results 

 from the present memoir that it was too little by 0-00003. 



