Derby.] J-^ [Feb. 21, 



Tocantins between 3° and 4° of south latitude, the Tapajos between 4° and 

 5°, and the Madeira between 8° and 9°, at Hie rapids of Sao Antonio. The 

 lower Madeira appears to mark approximately the western limit of the 

 ancient nietamorpbic region, because in the next river to tbe westward, 

 the Purvis, the rocks under consideration were not met with by Chandless, 

 in the course of his careful exploration. The parallelism of the course of 

 the lower Madeira with the great surface features of eastern Brazil, where 

 the nietamorpbic rocks are thrown into great folds, trending north easterly, 

 has alread3 r been noted. It seems possible that the Madeira is directed by 

 such a fold, or, what is more probable, by a margin of the metamorphic 

 region, which should there have that direction. It is possible that the 

 Guapore also marks another margin of the same region, which being trans- 

 verse to the folds is independent of their trend. It is certain that in the 

 Guapore region there was a canal between the metamorphic region of 

 Brazil and a similar one in Bolivia, the Chiquitos region of D'Orbigny, 

 comparable with the strait between Brazil and Guiana, now occupied by 

 the Amazonas. 



As in eastern and central Brazil, the metamorphic rocks of the Ama- 

 zonian region can be naturally divided into two very distinct series, of 

 which one, the most ancient, consists of crystalline rocks, including gneiss, 

 gneiss-granite and syenite, and the other, more modern, of altered, but in 

 general non-crystalline rocks, consisting of quartzites, metamorphic schists 

 and crystalline limestones. The older series corresponds in character and 

 geological age with that of the Serra do Mar and Serra do Mantiqueira, in 

 the provinces of Rio de Janeiro and Minas Geraes, which was referred by 

 Prof. Hartt to the Laurentian. This series has been but little studied in 

 the Amazonian region. Casteinau speaks of gray gneiss on the Tocantins 

 above the first rapids, and Chandless met with gneiss in a similar position 

 on the Tapajos. Snr. Ferreira Penna, of Para, informed me that the 

 rapids of theXingii are formed by gneiss and diorite, and showed me speci- 

 mens of the first, consisting of flesh-colored feldspar and quartz with a 

 small proportion of black mica, the rock in hand specimens appearing 

 massive and granitoid. The lower rapids of the Madeira are also formed 

 of gneiss, but I have seen no specimens or descriptions of the rock. On 

 the northern side, gneiss was met with in situ, by Snr. Penna, in the rapids 

 of the Araguary, a small river emptying into the Atlantic, a little to the 

 north of the mouth of the Amazonas, and pebbles of the same rock were 

 met with in the explorations of the Geological Commission, on the rivers 

 Maecurii, Ourua, and Trombetas. I am informed by the engineer, Maj. 

 Coutinho, that gneiss is the prevailing rock on the Bio Branco, except at 

 the mouth, where he found red syenite. I found this last rock also in a 

 zone about half a mile in width, at the second rapid of the river Trombe- 

 tas, and saw pebbles ofthc same on the Maccuru, coming from some point 

 above that reached by our explorations. I could not determine, in the 

 short time at my disposal on the Trombetas, whether the rock is stratified 

 or not, and it is possibly of eruptive origin. The syenite consists princi- 



