90 Washington — Pyroxenite and Hornhlendiie in Brazil. 



formation of feldspar in these melts under the conditions 

 obtaining. It may be recalled, however, that the ariegites of 

 Lacroix are much higher in alumina and lime, so that their 

 chemical composition is more favorable to the formation of 

 calcic feldspars, as is also shown by the greater abundance of 

 the anorthite molecule in their norms. Further investigations 

 along these lines are to be undertaken later on. 



The Manganese and Copper Content. 



It has already been noted that the percentage of manganese 

 shown in my analyses is exceptionally high for igneous rocks. 

 There are, it is true, some other instances of similarly high 

 figures for manganese such as the mijakite of Petersen,* but 

 there is little doubt that in such cases the high figures for 

 manganese are due to incomplete precipitation of alumina in 

 the sodium acetate method and the weighing of the unprecipi- 

 tated A1 2 3 as MnO. In the Bahian rocks, manganese was 

 determined by the colorimetric method, which is far less sub- 

 ject to sources of error than, and is as accurate as, the gravi- 

 metric methods for small amounts of these constituents. 



The high manganese content shown by these rocks is espe- 

 cially interesting in view of the fact that the crystalline zone 

 of Bahia abounds in deposits of manganese ores. The largest 

 and best known of these are those near Nazareth, about 250 

 km. east of Maracas, as well as those near Vilha Nova da 

 Beinah, some 300 km. to the north. At both of these (which 

 have been visited by me), there are extensive deposits of psilo- 

 melane of excellent quality. There are also smaller deposits at 

 Cachoeira, some 200 km. to the east, and elsewhere in the 

 state, but always in the metamorphic and crystalline belt. The 

 high manganese of the rocks studied suggests the source of 

 these valuable deposits. 



The distinctly high amounts of copper found in the analy- 

 ses explain the considerable staining of the somewhat altered 

 rock near the contact by secondary copper minerals. While 

 prominent on account of their bright colors, the amount of 

 copper present is not sufficient to warrant its extraction, quite 

 apart from the small size of the mass and other considerations. 



Geophysical Laboratory, 



Carnegie Institution of Washington, 

 Washington, D. C, April 25, 1914. 



* Jb. Hamb. Wiss. Anst, viii, 1891, p. 511. 



