208 Derhij — Notes on Certain Schists of the Gold and 



IV — Black schist found loose in a small topaz washing near the 

 Caxambu topaz mine, Ouro Preto, and which possibly repre- 

 sents the original matrix of the topaz occurring in nests in a 

 clay similar to what would be given by the decomposition of 

 this rock ; analyst, Hussak. 



V. — Unctuous greenish schist from Boa Vista near Ouro Preto ; 

 anal}' st, Gorceix. 



VI. — Fibrous schist containing at times crystals of pyrophillite 

 and altered pyrite; Boa Vista near Ouro Preto; analyst, 

 Gorceix. 



VII. — Whitish unctuous scaly rock from Boa Vista near Ouro 

 Preto; analyst, Gorceix. 



VIII. — Compact violet schist in form of dike in limestone at Gan- 

 derela, north of Ouro Preto ; analyst, Gorceix. 



From macroscopic examination, No. I is evidently a chloritic 

 rock with enormous crj'stals of cyanite up to 5 cm or more in 

 length. The portion insoluble in sulphuric acid (43*41 per 

 cent) calculated, according to Hillebrand's suggestion, as 

 normal cyanite, mnscovite, quartz and rutile, gives : cyanite 

 16*46 per cent, muscovite 8*46 per cent, quartz 13*66 per cent, 

 rutile 4*73 per cent, with a doubtful excess of 0*09 per cent 

 referred to zircon, which could not be found in the heavy 

 residue, while a doubtful trace of zircon oxide was noted in the 

 hydrochloric acid solution indicating that this element prob- 

 ably belongs to a soluble constituent. The soluble portion 

 (56*66 per cent) reduced to 100 agrees very closely with Eaton's 

 analysis of corundophilite from Chester, Mass., the most 

 important differences being an excess of 1*88 per cent Si0 2 , 

 1*13 per cent A1 2 3 , and a deficiency of 4*15 per cent of iron 

 oxides. It is interesting to note that this Chester type of 

 chlorite is also associated with an aluminous mineral, corundum, 

 and that the somewhat similar type with the same association 

 from the Culsagee mine, N. C, carries Mn, Ni and Co like 

 the soluble portion of the Brazilian rock. Another element, 

 monazite, revealed in minute quantities by washing, is probably 

 indicated by the phosphoric acid of the analysis. 



In the paper above cited an argument for the eruptive origin 

 of this and other schists of the region was drawn from the 

 occurrence of autigenetic monazite, which was considered as an 

 original element that had passed unchanged through the process 

 of metamorphism. This argument must now be withdrawn 

 since, as will be shown in the following paper, evidence has 

 recently come to hand that monazite may be formed by second- 

 ary processes ; one of the proofs being furnished by the rock in 

 question, which shows rutile included in monazite in the same 

 manner as in the associated chlorite and cyanite. The prob- 

 able origin of this rock will be discussed farther on. 



