310 M. H. Becquerel's Experimental Investigations 



ratus was constructed sufficiently powerful to demonstrate the 

 feeble rotations which the hypothetical calculations had led 

 me to expect. 



The results of the preliminary observations were published 

 at the end of March 1879. The numbers obtained at that 

 time were the first that had been given on the measurement 

 of these phenomena, or which had established the relative 

 magnitude of the magnetic rotatory powers in liquid and solid 

 bodies and those in a gaseous condition. 



Since that time, Bichat and Kundt and Rontgen have pub- 

 lished various communications relative to the magnetic rota- 

 tory power of certain bodies in a gaseous state. 



About the same time Lippich obtained a magnetic rotation 

 of several seconds of arc with a column of 0*50 metre of air 

 subjected to a strong magnetic action. Bichat more particu- 

 larly confined himself to the determination of the relation 

 between the magnetic rotatory power of carbon disulphide at 

 various temperatures in the liquid and gaseous condition. 

 His results agree with my own. 



Kundt and Rontgen have studied the magnetic rotatory 

 power of certain gases from a point of view quite different 

 from mine. They compressed the gases to 250 atmospheres, 

 while I more particularly studied them under the ordinary 

 temperature and pressure. The German authors have obtained 

 under considerable pressure magnetic rotations of several 

 degrees; but the imperfection of their apparatus did not allow 

 them to take advantage of the extent of the rotations in order 

 to obtain very accurate measurements. They limited them- 

 selves to publishing numbers relative to five gases — hydrogen, 

 oxygen, air, oxide of carbon, and marsh-gas ; and they were 

 unable to discover any relation between the index of refraction 

 of the gases and the observed rotations. 



In the present memoir I attempt to show that the existence 

 of a relation between the index of refraction of bodies and 

 their magnetic rotatory power in a solid, liquid, and gaseous 

 state cannot be called in question. I would further record 

 the fact that I have demonstrated the existence of a perturba- 

 tion in the phenomena of atmospheric polarization, due, appa- 

 rently, to a slight rotation of the plane of polarization of the 

 luminous rays passing through the atmosphere, under the in- 

 fluence of terrestrial magnetism. 



Method of Experiment. 



The study of the magnetic rotatory power of gases implies 

 an accurate knowledge of their temperature and pressure at 

 the moment of the experiment. The same volume of the dif- 



