1879. J lfr° TDerby. 



masses of the green mineral above mentioned. There was also observed 

 in the cachoeiras, two exposures of crystalline rocks which appear to form 

 dykes, but this character was not well determined. One of these is fine- 

 grained and dark-colored, the other consists of light-red feldspar, with 

 grains of quartz. 



"We found on the Trombetas a series very similar to a part of that of the 

 Tapajos. It is exposed in the third cachoeira, called Quebra-potes, and also 

 in the lower course of the river Cachorro, which empties into the Trom- 

 betas just above that cachoeira. The rock varies in color, some beds being 

 dark red, others purplish, and like that of the Tapajos it is marked by 

 green spots. The mass is amorphous, feklspathic, sometimes with small 

 grains of glassy quartz, and it may be classified as fe.site or eurite. The 

 stratification is very distinct, and the lamination, wave and ripple-marks 

 are as clearly shown as in any modern sandstone. The beds of felsite rest 

 on those of the syenite already described, which is also marked by green 

 spots, and dip 20° N. E., the strike being N. 30° W. Resting unconform- 

 ably on this scries are beds of sandstone, containing Upper Silurian fossils. 



This last observation is important, proving as it does that the metamor- 

 phism of the rocks and the dislocation of the beds must have taken piace 

 during the Lower Silurian or Archean. I am convinced that this conclu- 

 sion can be extended to the whole metamorphic region. The similarity on 

 lithological characters of the rocks of the Trombetas and those of the 

 Tapajos is such, that it can scarcely be doubted that the formations in the 

 two localities are identical. The difference in strike, from N. N. W. on 

 the Trombetas, to N. N. E. on the Tapajos, can readily be admitted in a 

 single system of upheaval, which can include also the disturbed rocks of 

 the Tocantins. where the strike is N. or N. N. W. It should be observed 

 that, while the compact cptartziles of the Tocantins resemble the rocks of 

 the Tapajos and Trombetas, the rest of the Tocantins series, consisting of 

 granular quartzites, talcose schists, and crystalline limestones, recall the 

 rocks of the rivers Araguay, and upper Tocantins, and of the mountains of 

 Goyaz and Minas Geraes. 



It has long since been observed that the metamorphic rocks of Brazil, 

 Guiana and Venezuela have in general a north-easterly strike ; later obser- 

 vations, however, have shown that the strike is often variable, frequently 

 taking a north westerly direction. It seems probable, therefore, that the 

 epoch of metamorphism and upheaval of the ancient rocks was the same 

 in eastern Brazil and Guiana as in the Amazonian region, that is to say, it 

 was anterior to the Upper Silurian. 



The evidence in respect to the epoch of metamorphism and upheaval 

 afforded by other regions of Brazil is very scanty, but, as far as it 

 goes, it sustains this generalization, although it must be confessed it is 

 as yet insufficient to entirely confirm it. In the provinces of Bahia and 

 Sergipe there is a series of beds of undetermined age, but which ap- 

 pears to be either Devonian or Carboniferous. These beds have been 

 disturbed without being metamorphosed, and they rest unconformably 



