OF COBALT, &c. 4*7 



fometimcs irore, fomctimcs left iron. In the pow- 

 der precipitated no mark of the prc fence ofir on 

 appeared, although I fufed it repeatedly by the 

 ad ion of ill*.- blow-pipe, with microcofmic fiilt. 



2. From Solutions of nickel, cobalt, and man- 

 ganefe, v.inc precipitates only hct erogenous fub- 

 fiances, accidentally intermixed with thefe me- 

 tals. Now, as zinc precipitates all metals, ex- 

 cept iron, it follows, that nickel, cobalt, and 

 inanganefe are either particular modifications 

 or fpccies of iron, 'or entirely diftindl from it. 

 Some may, perhaps, fay, that two or three me- 

 tals may be fo combined by nature, that zinc 

 cannot feparate them in a fotutionin \vhichthcy 

 cxift in natural combination, although capable 

 of precipitating each of them from a iblution in 

 which it cxifis by itfelf. To this, if proved by ex- 

 periments I ihi'.ll not refute to agree; but till it 

 be cftabliihcd by experiments, it would be fool- 

 ifh to receive it upon mere fancy and conjec- 

 ture. 



That nickel, cobalt and manganefc arc, as 

 has been thought by fome, modifications or fpe- 

 cies of rron, may be maintained by other argu- 

 ments, bciides thofc above-mentioned. Iron ha^ 

 nmch greater verfatility of nature, and is fufcep- 

 libleof a much greater divcrfity of forms than any 

 other metal. Fufcd iron, cold iron, hot iron, brittle 

 iron, rtccl,&c,liave all diilincl characters: and each 



of 



