28 0. A. Derby — Manganese Ore Deposits 



or, perhaps more probably, a manganese-bearing silicate that 

 has given rise to the garnet in the process of metamorphism. 

 Such a type, if unsheared and decomposed, would give rise to a 

 clay very like that found in the immediate vicinity in the 

 above mentioned pit, and the hypothesis may be ventured that 

 the two represent sheared and unsheared portions of the same 

 rock mass. 



On the opposite side of the hill and a few scores of meters 

 to one side of the ore belt pits had been opened in partially 

 decomposed granite with stringers from a few millimeters up to 

 about a meter in width of graphite, which though somewhat 

 mixed with clayey matter appears to be of good quality. 



At the Barroso locality, which, as already remarked, appears 

 to mark a third ore belt intermediate between that of Piquiry- 

 Sao Goncalo on one side and that of Morro da Mina-Agua. 

 Limpa on the other, two openings have been made several hun- 

 dred meters apart and apparently on independent ore bodies. 

 One exposes a layer about two meters thick and with an incli- 

 nation of about 45° between walls of decomposed schist threaded 

 with stringers of granite. The ore presents for the most part 

 the aspect of the earthy material at Sao Groncalo and like that 

 gives a residue of microscopic garnet but with a greater 

 amount of quartz and of clayey matter. In places it passes to 

 a granular quartz rock thickly sprinkled with macroscopic gar- 

 nets. The enclosing schist presents an appearance suggestive 

 of an original amphibolite and gives a residue of ilmenite only, 

 and apparently represents a sheared eruptive of non-granitic 

 character. In the other and more important opening the ore 

 body is 4 to 6 meters thick, inclined at an angle of about 80° 

 between walls of decomposed schist without granite. The ore, 

 which is of the same general character as that of the first open- 

 ing, though of better appearance is quite impure, giving, accord- 

 ing to Mr. Scott's analysis of an average sample, 28*10 per cent 

 of metallic manganese, 6*00 per cent of iron, 15*80 per cent of 

 siliceous residue and 7*20 per cent of graphite. Much of the 

 ore has the aspect of a decomposed graphitic clay slate charged 

 with secondary manganese oxide, but one of my samples, 

 too friable for a microscopic preparation, resembles, both in 

 aspect and in its residue, the garnet-quartz rock with mica 

 from Morro da Mina and like that has no graphite. Another 

 specimen showing considerable well-crystallized graphite with 

 ilmenite has the garnet enclosed in an earthy siliceous matrix 

 apparently of secondary quartz stained with iron but not with 

 manganese (except in the portions where the garnet is also 

 decomposed), and this appears to represent an original rock 

 composed of garnet with some iron-bearing silicate. The 

 decomposed schist of the walls of this ore body is still quite 



