﻿330 Mr. D. Forbes's Researches in 



1*604 per cent. The oxides of manganese and iron were then 

 separated by carbonate of barytes and determined as'usual, the 

 sesquioxide of iron obtained amounting to 4*82 grs., and the 

 manganosomanganic oxide to 0*27 gr., equivalent to 1*25 per 

 cent, protoxide of manganese in the mineral. The solution, after 

 separating the alumina and oxides of iron and manganese, was 

 now precipitated by oxalate of ammonia, and the resulting oxa- 

 late of lime converted into carbonate by gently heating to red- 

 ness with the subsequent employment of carbonate of ammonia. 

 The amount of carbonate obtained was 7*46 grs., equal to 4*17 

 grs. lime, or 20*87 per cent, lime in the mineral. From the fil- 

 trate the magnesia was separated by addition of ammonia and 

 then phosphate of ammonia, and the precipitate converted into 

 pyrophosphate of magnesia by ignition ; it weighed 2*04 grs., 

 equal to 0*73 magnesia, or 3*67 per cent. 



As it was found that the iron contained in the mineral was 

 present in the state of both protoxide and sesquioxide, it 

 became necessary to determine their respective amounts. For 

 this purpose a separate portion of the mineral was employed and 

 treated as follows : — A portion of the mineral finely powdered 

 was mixed with four times its weight of the carbonates of soda 

 and potash, and placed at the bottom of a platinum bottle (beaten 

 out of a single piece without soldering) about 3^ inches deep and 

 1^ inch in diameter ; a small porcelain tube like the stem of a to- 

 bacco-pipe was then suspended halfway down the interior of the 

 platinum bottle, and through it a current of dry carbonic acid gas 

 was kept passing, provided by a suitable apparatus; the mixture 

 of the mineral and carbonates was then heated until fully decom- 

 posed in this atmosphere of carbonic acid gas ; and after allowing 

 it to cool, the fused mass was decomposed by the addition of sul- 

 phuric acid until all the iron was brought into complete solution, 

 whilst during the whole of the operation the stream of carbonic 

 acid gas was kept passing continuously through the apparatus, 

 so as effectually to prevent oxidation. The contents of the bottle 

 were now rapidly poured out into a beaker containing water 

 which had been previously boiled to expel all air, and the bottle 

 itself well rinsed with boiling water into the same solution, which 

 was at once titrated by a normal solution of permanganate of 

 potash in order to determine the amount of protoxide of iron 

 present. 



Two separate determinations were made in this manner, and 

 the amount of sesquioxide of iron corresponding to the quantity 

 of protoxide indicated was then subtracted from the total amount 

 obtained in the course of the analysis, as before described. 



The results of the analysis thus executed, when tabulated 

 showed the composition of the mineral to be as follows : — 



