of Minas Geraes, Brazil. 19 



Piquiry and Sao Goncalo, situated some 3 or 4 kilometers apart 

 on the same ridge, with a number of known intermediate out- 

 crops and probably prolonged southward by others that have 

 been prospected but not worked, 



2d. An eastern belt, marked at the extremes, so far as known, 

 by the Morro da Mina (Mine Hill) and Agua Limpa (Clear 

 Water) deposits ; and 



3d. A central belt, represented by the abandoned Barroso 

 workings close to the railroad a few kilometers south of the 

 town. 



At all of the localities above mentioned perfectly sound gran- 

 ite occurs within a few hundred meters of the ore bodies and 

 at Piquiry the wall or walls are constituted by a residual clay 

 resulting from the complete decomposition in situ of a typical 

 granitic rock that gives an abundant and characteristic residue 

 of sharp-edged zircons. At Sao Goncalo, where only the 

 foot wall is exposed to view, this and some intercallated layers 

 near it are of completely decomposed schists whose original 

 character will be discussed farther on. At Barroso both foot 

 and hanging wall are of decomposed schists that are suspected 

 to have been originally amphibolic, while at Agua Limpa a 

 sound amphibolic schist of peculiar character occurs, within a 

 few meters of the ore body. At Morro da Mina, aside from 

 dubious decomposed schists apparently similar to those at Sao 

 Goncalo, a peculiar residual clay occurs that will be discussed 

 below. A peculiar feature is the occurrence of well-defined 

 layers of graphitic schist in the ore of Agua Limpa, Morro da 

 Mina and Sao Goncalo, while at all of the localities parts of the 

 ore body are more or less graphitic. 



The Piquiry ore body presents the appearance of a mass of 

 secondary material, or gossan, resulting from the alteration of 

 a vertical dike or vein, some ten or dozen meters wide. The 

 margins are sharply cut against a tough clay which is undoubt- 

 edly decomposed granite, presumably identical with the sound 

 granite occurring a few hundred meters distant from the mine. 

 In appearance both foot and hanging wall are presented, but 

 Mr. Scott thinks that the opening is on an elbow of the ore 

 body, so that in reality only one wall is seen. The ore is a hard 

 spongy black oxide apparent^ consisting for the most part of 

 psilomelane but with an admixture of other oxides that fre- 

 quently occur in beautiful crystallizations in the spongy cavi- 

 ties. STo cargo analyses are at hand, but Mr. Scott gives for a 

 sample of ore prepared for shipment : manganese (metallic) 

 51*40 per cent, iron 2*00 per cent, siliceous residue 5*02 per 

 cent, phosphorus 0*13 per cent. 



In the midst of the merchantable ore occur inconstant bands 

 and patches of hard siliceous material with the appearance of a 



