26 0. A. Derby — Manganese Ore Deposits 



so far as they can be made out, it is probable that the rock con- 

 sists essentially of polianite and spessartine with small amounts 

 of some lime and magnesia silicates, apatite and possibly ihnen- 

 ite, though a portion of the water is doubtless combined with 

 the manganese, which is in part visibly altered, and the iron 

 may also be, in part, combined with the manganese oxide. 

 The presence of a small proportion of nickel and cobalt is an 

 interesting feature. 



The clay in which the tunnel terminates shows no structural 

 features except scattered patches of soft earthy manganese and 

 iron oxides suggestive of segregations of original bisilicates in 

 an essentially feldspathic rock. The slime obtained by washing 

 consists largely of minute micaceous flakes, and the residue of 

 a small amount of minute quartz grains with heavier dirty 

 white earthy grains that give a titanium reaction and are pro- 

 bably leucoxene representing original ilmenite. A specimen 

 taken close to the contact with the ore gives also a small 

 amount of well-formed brown tourmalines, but this does not 

 appear to be generally distributed throughout the clay mass. 

 From the above characteristics it seems almost certain that this 

 clay represents the decomposition product of an original mas- 

 sive eruptive rock that was predominantly feldspathic but with 

 segregations of a manganese-bearing bisilicate element and with 

 ilmenite as the only accessory. If, as seems most plausible, 

 the small residue of quartz be regarded as secondary, the origi- 

 nal rock was probably a gabbro or some closely related type. 



At the point of the hill and apparently in prolongation of 

 the smaller dike-like outcrop above mentioned as occurring at 

 the top, mining operations had been commenced, in part in 

 loose secondary ore, in part in ore in situ. The latter has an 

 impure earthy appearance like that above described from Sao 

 Goncalo and such as might be expected to result from the 

 alteration of the hard garnetiferous ore of the tunnel and of the 

 said dike-like ore body. The siliceous residue is understood to 

 run from 16 to 20 per cent and the contents in metallic man- 

 ganese from 37 per cent to 45 per cent. A small pit showed 

 two ore bodies of this type in situ, each being about three 

 meters thick, and separated by a foliated layer of granular 

 quartz rock about one meter thick. This quartzose layer resem- 

 bles closely the itacolumite of the Ouro Preto region, but, unlike 

 it, is without mica and gives no clastic residue, the only heavy 

 element being transparent hematite that evidently comes from 

 altered pyrite. Underneath the lower ore body comes lami- 

 nated clays, in part graphitic, that are evidently derived from 

 decomposed schists but are too incoherent to show distinctly 

 the original structure. These give on sliming a residue of 

 coarse quartz with nodules of earthy iron and manganese oxides 



