Transjmrent Bodies hy the Action of Magnetism. 427 



Thus the two chromates of potash and chromic acid have a 

 negative magneto-rotary power, and the absolute vahie of this 

 power increases with the quantity of chromic acid. The nega- 

 tive power of the neutral chromate is feeble, but quite unmis- 

 takeable ; that of the bichromate is considerably stronger, while 

 that of chromic acid is comparable with the power of the proto- 

 salts of iron. 



The experiment with the neutral chromate of potash is espe- 

 cially worthy of attention. We know, in fact, that while chromic 

 acid and the bichromate of potash are magnetic, the neutral 

 chromate is diamagnetic*, and it is rather surprising that the 

 magneto-rotary power of a diamagnetic body should be negative. 

 We shall meet with other analogous examples in the sequel. 



The five metals whose compounds I have just considered have 

 long been considered as magnetic by all physicists. This is not 

 the case with those which follow, which have only been recog- 

 nized as magnetic since the investigations of Faraday, and some 

 of which indeed I have now examined for the first time. It has 

 thei-efore been essential for me to vei'ify the magnetic character of 

 these bodies before studying the magneto-rotary power of their 

 compounds. In this verification I have followed the rule laid down 

 by Faraday, which consists in regarding as doubtful the magnet- 

 ism of every metal whose apparent magnetism is only feeble, and 

 which does not give rise to any magnetic compound, especially to 

 no magnetic oxide. The quantity of a very magnetic metal (iron, 

 nickel, or cobalt), the presence of which it is necessary to admit 

 in order to account for the magnetism of a feebly magnetic body, 

 is so small as to escape all analytical test, and against which no 

 process of purification is a sufficient guarantee. The case is 

 quite difi^erent with a salt or an oxide ; for if we wish to explain 

 the magnetism of such a body by the accidental presence of a 

 salt, or oxide of iron, or some analogous compound, we should 

 be obliged to assume the presence of a quantity sufficiently great 

 to be easily detected by analysis, because the magnetism of such 

 a salt or oxide is incomparably less than that of the correspond- 

 ing metal. Hence it is that we see some metals, apparently 

 feebly magnetic, give rise, on oxidation or solution, to diamag- 

 netic compounds. 



Titanium. — Mr. Faraday classed titanium among magnetic 

 bodies. I have verified his observations with some pieces of pure 

 titanium, with which I was furnished by J\I. Deville. The mag- 

 netism of titanium appeared to me to be greater than that of 

 pure chromium, and too great to be attributed to impurities 

 which liad escaped analysis. Amongst the compounds of this 



* See meiuoir of FiinuUiy refcnuil to btfoie. 



2 F2 



