Reflections on Cobalt, &c. 369 



Jihgly contain any fubflance, which is known to 

 communicate a different colour to that men- 

 Jiruum in the fame circumjlances ; for there is 

 fcarce any one property of any fubftance that 

 may not be communicable to fome other fub- 

 ftance, as it is only the aggregate of all their 

 properties that fully diftinguifhes fubftances 

 from each other. 



13. He further adds, that nickel cannot 

 be looked upon as a diftincT: femi-metal, be- 

 caufe it cannot be thoroughly purified from 

 iron. What ftrefs fhould be laid on this ar- 

 gument, we fhall prefently confider in treat- 

 ing of manganefe ; but in the mean time we 

 may obferve that it is far from being clear 

 that nickel can never be purified from iron ; 

 for the only ground of this fufpicion is, that 

 the pureft nickel is magnetic ; but this mud 

 be deemed infufficient, unlefs it be taken for 

 granted that magnetifm is a property inhe- 

 rent in iron only, and incommunicable to 

 all other pofiible metallic fubftanees, as we 

 know it is to thofe that are already known ; 

 a pofition which feems to me fully contra- 

 dited in the cafe of nickel; for when it is pu-* 

 rifiedas much as poffible from iron, it becomes 

 more inftead of lefs magnetic, and even ac- 

 quires what iron does not, the properties of 

 a magnet. What hinders us then from al- 

 lowing magnetifm, like the property of giv~ 

 B b 



