ON THE GENESIS OF THE DIAMOND I 35 



diamonds themselves of this region are easily distinguished from 

 those of the rolled deposits by their rugose faces, sharp angles, 

 and superficial greenish blue coloration." 



On the occasion of my visit in 1 88 1 only fallen masses that 

 appear to represent the bodies 3 and 2 of the above description 

 were to be seen, the operations of the mine being at that time 

 suspended. One of these masses was so evidently a section of 

 a vein that the conclusion was announced 1 that at Sao Joao da 

 Chapada the diamond is a vein mineral accompanying quartz and 

 an argillaceous rock of indeterminable character in the series of 

 metamorphic schists. This conclusion was subsequently fully 

 confirmed by Gorceix as the result of prolonged prospecting 

 operations, 2 and thus one mode of occurrence of the diamond 

 in the matrix was clearly established which, in appearance at 

 least, is widely different from that at Gao Mogol and at Kim- 

 berley. 



In the various papers by Gorceix and myself the schistose 

 series of the gold and diamond region of Minas Geraes, in which 

 the diamond occurs at Sao Joao da Chapada, is assumed to con- 

 sist essentially of metamorphosed elastics, though no direct 

 proof of this assumption is given. As regards the diamond- 

 bearing layers, they are called veins, but no definite opinion 

 regarding their mode of origin is expressed. For the question 

 of the genesis of the diamond both these points are important. 



All writers on the geology of this region are agreed that the 

 characteristic formation is a great series of phyllites, quartz 

 schists (itacolumite), iron-mica schists (itabirite) , and limestone, 

 and that this series constitutes a geological unit. This last point 

 is assumed rather than proven, since there may be a break in the 

 succession which has thus far escaped observation, just as that 

 above indicated between the upper and the lower itacolumite was 

 overlooked, or disregarded, by the older writers. For the pur- 

 poses of the present discussion, however, it is of little conse- 



^m. Jour. Sci., 1882, Vol. XXIII, p. 97. 



2 Comptes Rendus, No. 25, 1881. Am. Jour. Sci., 1882, Vol. XXIII, p. 97, and 

 Vol. XXIV, p. 34. 



