126 OR VILLE A . DERB Y 



apparent relation with the occurrence of the diamond ; the 

 deposit is conglomeratic, or brecciated,and in both cases the most 

 characteristic elements of the conglomerate represent rocks of 

 an ultra basic type, and in both the diamond is presumed to be 

 an element of the cement rather than of the included pebbles ; 

 perofskite and garnet (pyrope) are characteristic accessories 

 (garnets though frequent in most Brazilian sands and gravels 

 are exceptionally rare in those of the diamond washings ; those 

 of Agua Suja present the rare cubic habit); the diamonds seem 

 to be distributed with a considerable degree of uniformity 

 throughout the mass. 



With these analogies are, however, associated differences 

 that are, in appearance at least, of equal if not greater impor- 

 tance. The Agua Suja deposit is a bed, not a volcanic neck ; in 

 it clastic elements, of a much more varied character than those 

 of Kimberley greatly predominate over those of eruptive origin, 

 which also are more varied in character ; the cement is appar- 

 ently clastic rather than eruptive ; the eruptive elements, exclu- 

 sive of the trap, probably represent basic phases of the nepheline- 

 or augite-syenite type of rocks and not the peridotitic, and there 

 is as yet no direct evidence that they have anything to do with 

 the diamond; the original matrix of the garnets is unknown and 

 there is no evidence that their association with the diamond and 

 with the basic eruptives is direct and not accidental. If, as is 

 quite within the range of possibilities, eruptive necks of the 

 Kimberley type should be discovered in the Agua Suja region, 

 or contemporaneous (?) sedimentary deposits of the Agua Suja 

 type in the Kimberley district, some of these differences would 

 doubtless become analogies, but that of the probable original rock 

 type would still remain and would require an extension of the 

 views at present held regarding the type of eruptive rocks with 

 which the diamond is associated. If, as some hold in regard to 

 the Kimberley occurrence, the diamond is the product of meta 

 morphic action on carbon-bearing rocks and not an element of 

 the eruptive rock itself, this last difference would lose much of 

 its importance. In this case, the Kimberley and Agua Suja 



