HASEMAN, GEOGRAPHICAL DISTRIBUTION IN SOUTH AMERICA 41 



the Brejo de Varedao to Bio Tocantins, on account of a waterfall in the 

 lower course of Eio Nova, known as Cachoeira da Yelha. 21 



The Cachoeira da Velha is about 200 feet wide. It had, when I saw 

 it, a small island with two trees. These trees are evidence that the chan- 

 nel was originally a few feet to the north of its present site, where there 

 is still evidence of an old side channel which had no perpendicular falls. 

 During half-flood season, I estimated the highest point of the falls at 24 

 feet. The north side was 12 feet and the south side was a rolling mass of 

 water about six feet high. The south side has nearly all of the water,, 

 because there is an abrupt bend in the river about 500 feet above the falls. 

 These falls are not any higher than the Theothono Falls of Eio Madeira,, 

 which have been passed by nearly all of the Amazonian fishes. It must, 

 however, be remembered that Bio Madeira has a volume almost as large 

 as the Mississippi and that the Theothono Falls are only about 100 meters 

 above sea level, while the Cachoeira da Velha has a small volume of water 

 and is at least 300 meters above the sea level. 



The rocks of the waterfall are the same general type of ( ?) pre-Cam- 

 brian found in various waterfalls of the Piano Alto region. I picked up a 

 piece of flint which I believe corresponds to the Jacuipe flint of Serra de 

 Jacobina. To the northeast of this region, the iron, manganese and dia- 

 mond-bearing gravels were observed near Paranagua, Piauhy. This indi- 

 cates a similar structure as the Jacobina series of Bahia. 



The dissected faces of the Piano Alto are far away from the south side 

 of the Cachoeira da Velha and the river does not flow off of a mountain,, 

 as sketched by Froh. On the north side of the Cachoeira da Velha, there 

 are several mesas of the dissected Piano Alto between Eio do Somno and 

 Eio Nova. 22 



BARRIERS TO AQUATIC MIGRATION AND CONDITIONS OF ENVIRONMENT 



Under the above heading, I will not consider physiological barriers, i. e. r 

 if a fish is found in the upper course of a large river like Eio Paraguay, 

 which is navigable by small ocean steamers, and is not found in the lower 

 course of the same basin (Eio Parana), we may look on this restricted 

 distribution as being due at least in part to a physiological barrier. 



Under the subject of barriers to aquatic migration, I will discuss only 

 one of the geological barriers, namely, waterfalls. Too much' emphasis 

 cannot be placed on the location, size and character of the numerous 



n It is so named because an old fish-like woman is supposed to have drowned a fleeing- 

 had Indian at this point about 100 years ago. The natives are afraid to go to this 

 waterfall. 



22 Canoes descend from Porto Franco on Rio do Somno to Para during the rainy season. 



