S84f 



Analysis of some fossil?. 



Greenland gar- 

 net. 



The Greenland garnet 



silex 



. 



43 



aluinine . 



- 



15-50 



magnesia 



- 



8-50 



lime 



- 



1-75 



oxide of iron 



- 



29-50 



oxide of manganese 



0-50 



Capillary py- 

 rites native 

 nickel. 

 Succinic acid 

 separates iron 

 from manga- 

 nese. 



Benzoic acid 

 supposed to do 

 the same. 



Oxalic acid 

 preferable. 



Titanium pre- 

 cipitated by 

 tannin, not by 

 gallic acid. 



98-75 

 The capillary pyrites is not a sulphuret of iron, but na- 

 tive nickel, with a little cobalt and arsenic. 



Several years ago, you know, in my examination of am- 

 ber, and its acid, which 1 have not yet published, I found, 

 that the succinate of iron was insoluble in water; that of 

 manganese, on the contrary, very soluble; and that on this 

 I founded an easy method of separating these two metals, 

 which has since been much employed by Klaproth, Vau- 

 quelin, and others. The basis of this process being thedif-i 

 ferent, or I may say, inverse solubility of two saline com- 

 binations, it was obvious, that other acids might produce Ji 

 similar result. Accordingly Mr. Berzelius mentioned the 

 benzoic acid, Avhich however I cannot think very suitable, 

 as Mr. TrommsdoriF asserts, that the benzoate of iron is 

 very soluble. Mr. John has lately employed with the same 

 view the oxalic acid, and acidulous oxalate of potash ; and 

 Mr. Simon, an eminent chemist of Berlin, confirms their 

 utility in a paper he has just written on the analysis of some 

 fossils, the colophonite, scapolite, &c. ; and he observes 

 they are preferable to the succinic acid, because the oxalate 

 of iron is less bulky than the succinate. 



The same chemist has examined the phenomena exhibited 

 by the gallic acid and tannin with titanium. This metal is 

 not precipitated from its solutions except by tannin, or sub- 

 stances containing it. 



