346 Mr. D. Forbes* s Researches in 



tinum capsule until all the iron and titanium had been ex- 

 tracted; the whole was then poured into a large excess of cold 

 water and filtered from the insoluble siliceous matter, which was 

 determined as usual. 



The filtrate was now supersaturated with ammonia, which pre- 

 cipitated ail the titanic acid, oxide of iron, &c. ; without filtra- 

 tion the whole was then subjected to the action of a stream of 

 pure sulphurous acid gas, which reduced the sesquioxide of iron 

 to the state of protoxide and dissolved it, but ultimately also 

 took up the titanic acid and produced a clear solution. This 

 solution, after boiling for some time, deposited the whole of the 

 titanic acid as a white precipitate, which, after ignition, was of 

 a faint yellow colour from still retaining a trace of oxide of iron. 

 From the filtrate the iron, manganese, alumina, lime, and mag- 

 nesia were now determined as usual, and the results obtained 

 were as follows : — Quantity of mineral employed 24*82 grs., in- 

 soluble residue obtained 7'67 grs., titanic acid 5*43 grs., sesqui- 

 oxide of iron 11*16 grs., manganoso-manganic oxide 0*18 gr., 

 alumina 0*56 gr., carbonate of lime 0*76 gr., pyrophosphate of 

 magnesia 0*13 gr. 



Before the blowpipe this mineral is infusible, and unaltered by 

 the oxidating-flame; in a strong reducing-flame the sharp edges 

 become somewhat rouuded off, but no further change is percep- 

 tible. In the oxidating-flame, when treated with borate or phos- 

 phate, it merely shows the usual reactions of oxide of iron ; but 

 when the phosphate bead is treated for some time in a good redu- 

 cing-flame, the glass on cooling has a brownish-red colour. 

 When this glass is further treated with tin on charcoal in there- 

 ducing-fiame, it acquires the reddish-violet colour characteristic of 

 titanium. By fusion with bisulphate of potash, as before men- 

 tioned, the titanic acid may be separated and examined. 



The percentage composition of the mineral, calculated from the 

 above data, will be as under : — 



Titanic acid 5*43 21-87 



Iron . . . 7*811 , n . QQ 31*471 /n . Q o 



Oxygen as loss 2*58 j ' 10 39 10*46/ 41 93 



Protoxide of manganese . 0*16 0*67 



Alumina 0*56 2*25 



Lime 0*42 1*71 



Magnesia 0*19 0*67 



Insoluble siliceous matter . 7*67 30-90 



24-82 100*00 



Deducting from these results the amount of the silicates, we find 

 for the pure titanoferrite the following composition : — 



