Derby — Occurrence of Topaz near Ouro Preto, Brazil. 31 



position to the blue phyllite with which it is associated, 

 differing principally in the presence of manganese and in the 

 greater susceptibility to hydratization of the hematite grains, 

 since in the residues only an insignificant amount of this 

 mineral appears. This would give a muscovite schist normally 

 free from quartz but heavily charged with metallic oxides and 

 with scattered and more or less crushed crystals of feldspar 

 and nodules of feldspar or quartz, or of mixed feldspar and 

 quartz, with or without specular iron, rutile, topaz and mus- 

 covite. These nodules are frequently surrounded by a dark 

 band more heavily charged with the metallic oxides than is the 

 body of the schist and which may be either a primary or a 

 secondary feature, most probably the former since nodules of 

 a similar character and composition also occur. The accessory 

 elements rutile, tourmaline, zircon (?) and florencite are com- 

 mon to both the normal schist and the nodules but are not 

 prominent in either, while topaz, and probably euclase (not 

 identified in any of the samples examined) appear to be con- 

 fined to the latter. The larger undecomposed minerals of the 

 nodules (quartz, topaz, specular iron, rutile and exceptionally 

 euclase) usually show a certain amount of crushing which is in 

 accord with the sheared condition of the rock as a whole and 

 indicates that the nodular structure is a primary rather than a 

 secondary feature. The circumstance noted by Eschwege and 

 others that only very exceptionally do any of the minerals show 

 double terminations probably indicates that the nodules were 

 originally accompanied by drusy cavities that have doubtless 

 disappeared in the shearing. 



In a recent communication in this Journal I endeavored to 

 show from the chemical and mineralogical characters of the 

 schists of this region that many of them are probably sheared 

 eruptives and suggested the hypothesis of decomposition and 

 leaching prior to metamorphism to account for their peculiari- 

 ties. One of the rocks there discussed is identical with the 

 blue phyllite of the Caxambu mine and is from a topaz wash- 

 ing in the immediate vicinity, so that the arguments there 

 presented are applicable to those here considered. The argu- 

 ment for an eruptive origin is somewhat weakened by the 

 occurrence above noted of very similar, though more quartz- 

 ose, rocks with clastic zircons on each side of the mine which 

 perhaps may represent a mixture of eruptive and non-eruptive 

 material. Be this as it may, the argument for decomposition 

 and leaching, giving a concentration of iron oxide and alkalies 

 necessary for the formation of a hematite-muscovite schist, still 

 holds good and in this case the associated topaz schist may be 

 presumed to have been similarly affected. An explanation is- 

 thus afforded for the peculiar .composition of this rock, and 

 with this hypothesis, that of a concentration during the leach- 



