432 M. H. Becquerel's Experimental Investigations 



not, however, exceed the limits of those which are presented by 

 solid and liquid bodies ; and this study must be extended to a 

 much greater number of gases before we are able to decide 

 whether these variations are not simply characteristics of the 

 physical and chemical constitution of gases, as we have dis- 

 covered in regard to other substances. 



Some interesting remarks may be made on this subject. 

 The numbers given above enable us to compare the magnetic 

 rotatory powers of gaseous sulphurous acid with that which 

 it possesses in a liquid state. M. de la Rive found for liquid 

 sulphurous acid a magnetic rotatory power variable with the 

 temperature, and which at about 12° is nearly 0*382. But 

 the index of refraction of this body has not been accurately 

 measured. Faraday merely says that the index is the same 

 as that of water ; on this hypothesis we should find that the 



ratio 9/ 9 — ^ x is 0*277 — that is to say, exactly the half of 

 n\n 2 — 1) J ' 



the number found with the gaseous body. Is this statement 



a mere conjecture, or is it the expression of a more general 



fact?* 



An analogous result is observed on comparing with the 



number obtained for oxygen the numbers given by certain 



highly oxygenated liquids, such as nitric and sulphuric acids, 



&c. 9/ » — ^-r is about 0*11 for these bodies, while it is 0*27 

 n\n z — l) 



with gaseous oxygen. Finally, we may connect with these 

 facts an observation relative to various salts (chloride of so- 

 dium and chloride of potassium) endowed with a positive 

 rotatory power. Their magnetic rotatory power is less in a 

 crystalline condition than in solution. 



These various remarks would then tend to show that the 

 positive rotatory power of a body is so much the greater in 

 relation to its index of refraction as the particles of the bodies 

 are further apart, and that it increases on passing from a liquid 

 to a gaseous state. 



This question requires a special study, which could not form 

 part of the present work, but to which I intend to return at 

 some future time. If we are not absolutely authorized to 

 assert that the function ?i 2 (n 2 — 1) plays the same role for the 

 magnetic rotations of gases as for other substances, this re- 



* In a recent article (Journal de Physique, August 1880) M. E. Bichat 

 has studied sulphurous acid, both liquid and gaseous. He found for the 

 magnetic rotation of the liquid a number very near to that of M. de la 

 Rive ; for the index of refraction of the same body he gives the number 

 1-34- but the magnetic rotatory power that he attributes to the gas dif- 

 fers very much from that deduced from my experiments. 



