588 REPORT— 1881. 



y. The pinkish white mineral. 

 After allowing for 36'5 per cent, of admixed chrome iron oxide. 



Silica 32-6 



Magnesia ■ . . . . 28-7 



Alumina with a little ferric oxide ...... 27-4 



Water (loss on ignition) 11 -8 



1000 



This analysis was made by Mr. Kawakita, another of the instructors, in the 

 unavoidable absence of Mr. Haga. 



The tabular statement of the composition of the ore shows that it is in almost 

 exact agreement with that of pure ferrous chromite, as regards the ratio of iron 

 oxide to chromium oxide. No such pure ore has hitherto been recorded as occur- 

 ring, I beheve, unless possibly in the papers of Christomanos, which I have only 

 seen in abstracts. The insolubility of the iron in acids, and its non-magnetic 

 character, prove that it is really combined as chromite. There remains the mag- 

 nesia to account for. This cannot be in the free state, for not only is the occurrence 

 of free magnesia in nature unknown and improbable, but in this mineral the mag- 

 nesia, like the iron oxide, is for the most part insoluble in acid. From the abstracts 

 (' Jom-n. Chem. Soc.,' xxxii.) of his papers, I learn that Christomanos in one paper 

 asserts that pure chromite is always FeCroO^, and all other constituents of the ore 

 part of the matrix, and in another paper admits the existence of compounds of 

 ferrous oxide with chromic oxide in three proportions, other than that of the ordi- 

 nary spinel type. I see nothing for the case of the Oita ore but to accept the view 

 that the magnesia is a chemical part of it, the proportion of which, after deducting 

 soluble magnesia, is almost exactly that required by the formula, MgO, 2FeO, 2Cr303, 

 corresponding to the 3FeO, 2Cr203 of Christomanos. 



J, , Calculated for 



l<ouna MgO, 2FeO, 2Cr^03 

 Almost insoluble — 



Chromic oxide . . . . . 59-30 59-30 



Ferrous oxide 27-90 2800 



Magnesia 7-93 7-78 



95-13 9508 



Soluble and silica — 



[Above magnes. ferr. chromite (about) 0-57] 



Hydrous magnes. alum, silicate . . 3-62 



Magnetite 0-29 



9904 



The visible presence of the white mineral, and the magnetic character of a very 

 minute part of the powdered mineral, serve to confirm the propriety of the above 

 arrangement. 



The estimation of chromium and iron was made by the usual volumetric methods. 

 To bring the ore into a soluble condition, Dittmar's method was employed, this 

 method being very satisfactory, and, in Mr. Haga's hands, much easier than 

 described in Dingl. 



Fused borax, 2 parts, and dry potassium sodium carbonate, 3 parts, were fused 

 together in a platinum crucible till effervescence ceased, and the whole cooled. 

 About 1 part of very finely-powdered, ore was then placed on the top, and the 

 crucible covered and slowly heated to fusion. When the powder had become 

 moistened with the melted flux, the lid was removed and the crucible kept at a red 

 heat by a Bunsen flame. The liqiud mass was occasionally stirred with a small 

 platinum spatula. When particles of ore were no longer visible in the mass adhering 

 to the spatula, the heat was maintained for only a few minutes longer, and the 

 crucible then quickly cooled, to favour the detachment of the mass. One hour was 

 suflacient for one gram of ore, and half-an-hour for half a gram. In this time, 

 indeed, all the chromium may not have become chromate, but it will be all soluble 



I 



