Transparent Bodies by the Action of Magnetism. 429 



magnetic action, and both in aqueous solution evinced a nega- 

 tive magneto-rotary power. Tlie sulphate gave a tolerably 

 transparent reddish solution, whose action on polarized light 

 was markedly less than that of water. However, as I did not 

 analyze this solution, I am only able to regard the negative cha- 

 racter of the action which I attribute to it on polarized light as 

 simply probable. With the chloride no doubt can exist. A 

 concentrated solution of this salt, placed between the poles of an 

 electro-magnet, exerts an action on polarized light almost equal 

 in absolute value to that of water, and contrary in direction. 

 The perfect limpidity of the solution makes the verification of 

 these phsenomena vei-y easy. 



Uranium. — Mr. Faraday allows some doubt to rest upon the 

 position which uranium should occupy among the magnetic or 

 diamagnetic metals. He found, in fact, that the protoxide was 

 slightly magnetic, and the peroxide non-magnetic. The red and 

 the black oxides of uranium, prepared by heating at a higher or 

 lower temperature crystals of the nitrate purified by repeated 

 crystallizations, are magnetic. This experiment is rendered quite 

 conclusive from the fact that the nitrate of uranium itself is dia- 

 magnetic. The nitrate of uranium is the only compound of 

 uranium whose magneto-rotatory power I have determined ; and 

 the result I obtained is remarkable. An aqueous solution of 

 this body, under the influence of magnetism, exerts a less action 

 on polarized light than the water which it contains, which leads 

 us to regard the magneto-rotary power of the nitrate as negative. 

 This conclusion is confirmed by the study of its solutions, which 

 may be formed with alcohol and aether. The nitrate of uranium 

 therefore furnishes us with a third example, to be placed by the 

 side of the neutral chromate of potash and the bichloride of tita- 

 nium*. The absolute value which my experiments show to 

 belong to the negative action of the nitrate of uranium is, how- 

 ever, very small. 



Lanthanium. — As far as I am aware, the magnetism of lantha- 

 nium has not hitherto been examined. Some perfectly pure 

 carbonate of lanthanium, furnished me by M. Deville, was 

 strongly magnetic. This is suificicnt to show that lanthanium, 

 like its analogue cerium, is to be classed among the magnetic 

 metals. On treating the carbonate with pure hydrochloric acid, 

 I obtained a solution whose magneto-rotary power was less than 



were only assured that the salts did not contain the slightest trace of any 

 other metal besides cerium. 



• The existence of three diamagnetic compounds whose magneto-rotary 

 power is nc'<rativu, clearly demonstrates what I before asserted, namely, 

 that no simple relation exists between the diamagnetic capacity of bodies 

 and their magneto-rotary power. 



