38 O. A. Derby— Occurrence of the Diamond in Brazil. 



tion in the river just below would naturally form a back-water 

 with eddies at this point. The small stream mentioned may 

 have facilitated the process by cutting a narrow channel through 

 the upper quartzite, and thus admitting the river to undermine 

 it by the excavation of the softer bed below. The canon was 

 filled with fallen blocks, and with sand and gravel. The latter, 

 having been introduced since a former working a number of 

 years ago, was unproductive. It was on the supposition that 

 the former cleaning out had been incomplete that work was 

 undertaken again, but as the event proved only unimportant 

 remnants of the original gravel were met with. 



The gravel of the Jequetinhonha is very varied. Professor 

 Gorceix gives a partial list of 28 minerals, which he has recog- 

 nized among the finer and heav the miners 

 mi under the name of diasuoud formation, and which 

 they consider as indicative of the presence of the gem, and 

 notes the abundance of quartz, the oxides of iron and of titan- 

 ium, tourmalines, and the presence of chlorophosphates (Comp- 

 tes Bend us, No. 25, 1881). The coarser material of the gravel is 

 evidently derived from the rocks of the two series through 

 which the river flows at the mines and has not been transported 

 very far. The finer material including many rare minerals 

 with the diamond may have come from greater distances, either 

 from the main river, or its tributaries, but there is a strong 

 tion in favor of the view that this also has been 

 derived from one or the other or both of the two series. Hav- 

 ing followed the river for nearly the whole of its course be- 

 tween the Acaba-Mundo mine and its source, and having 

 crossed the upper valley along several lines, I have never seen 

 any evidence of the existence of any other series from which 

 the gravel can be supposed to be derived. It should be men- 

 tioned that at many points along the river the upper series 

 presents a conglomerate character. 



Of the services <lo compo some of the most extensive are 

 situated on the high rid'_m Vtu-,^, Ihamautina and the Jeque- 

 tinhonha, known as the Heights of Curralinho. Of these the 

 most interesting are those of Born Successo and Boa Vista 

 which almost overlook that of Acaba-Mundo. The heights of 

 (.'urniimho are capped by a heavy mass of coarse bowlder con- 

 glomerate belonging to the upper quartzite series. Bom Suc- 

 cess.) and Boa Vista are situated on opposite -ides of a small 

 valley at the northern extremity of the ''onghmmrate ridge ami 

 the material , shed is tin dis teu'rated rouu l.unerate, parts of 

 which have been redistributed by water while other parts have 

 never been disturbed. 



A similar mode of occurrence is seen on the high ridge to- 

 the southwestward of Diamantina at Guinda and Sopa. These 



