146 OR VILLE A . DERB Y 



rocks is very meager, and further developments must be awaited 

 to determine whether or not this resemblance has any special 

 significance. Another point in common that may prove to be 

 of greater significance than at first sight appears is the occur- 

 rence mentioned by Carvill Lewis 1 of tourmaline and disthene 

 which seem to be formed by metamorphic action about inclosed 

 fragments of schist. It is true that disthene is not a specially 

 characteristic mineral at Sao Joao da Chapada, but it is 

 extremely widespread and abundant in the schist series in which 

 the mine is excavated, and of the aluminous silicates, is the 

 most constant and characteristic of the associates of the dia- 

 mond in the Brazilian alluvial washings. The significance, if 

 any, of its persistent association with the diamond (now veri- 

 fied at Kimberley) is that of a mineral characteristic of the 

 metamorphism (by contact or otherwise) of argillaceous rocks. 

 In order to bring the Kimberley and Brazilian mode of 

 occurrence into line as phases of a single mode of genesis, it 

 seems necessary to put aside the idea that the recent interesting 

 experiments on the artificial production of the diamond afford a 

 solution of its terrestrial origin, and that the Kimberley type of 

 rock and mode of occurrence are essential features. Presum- 

 ably also the genesis must be sought in the rocks affected by 

 the eruptive masses rather than in those masses themselves. 

 There are still many obscure points about both places, and until 

 these are cleared up no satisfactory comparison can be made. 

 At Sao Joao da Chapada there is little prospect of working being 

 resumed so that no additional light is to be expected from 

 there, but at Kimberley the workings may ultimately reach a 

 depth that will give a complete solution of the problem for that 

 place and mode of occurrence. When this occurs, if it is veri- 

 fied that the ultra basic type of eruptive rock, brecciated struc- 

 ture, and slot-like mode of occurrence are necessary features, 

 the Brazilian occurrences must be put into another category. 



Orville A. Derby. 



Sao Paulo, November 29, 1897. 



1 Papers and notes on the genesis of the diamond. 



