128 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY. 



mangrove swamps or tide marshes. To be specific, however, we shall 

 mention what are regarded as more or less satisfactory evidences of this 

 depression. 



Open hays. — In all probability those geographers who have expressed 

 the opinion that the Brazilian coast is one of recent depression have 

 based their opinions upon the peculiar forms of the bays of Rio de 

 Janeiro and Bahia. These bays not only have many of the peculiarities 

 of harbors formed by depressions, but may even be accepted as the very 

 types of such harbors. 



Before the depression took place Bahia was a hilly, almost a moun- 

 tainous region. Its streams were rapid, its valleys steep-sided and 

 v-shaped ; its hills were on the whole even-topped, but erosion had 

 worked down among them until many of these were left as isolated 

 peaks above the lowering valleys about them. Most of the streams 

 flowed approximately where they now flow, but those now entering the 

 bay united in the bottom of a broad valley and flowed out through what 

 is now the mouth of the bay into the sea. All that is now known of 

 the date of this erosion is that it was post-Eocene. 



When the great depression came, the Bahia valley sank and the waters 

 of the ocean backed up into it and made the bay of Bahia very much as 

 we now see it. The scour of the tides has kept the entrance to the bay 

 open, but the upper ends or indentations of the bay have been silted up 

 with the aid of the encroaching mangroves. The tops of some of the 

 hills were left sticking above the surface as islands. Some of these hills 

 were cut off" by the waves, and others were left as islands, and are now 

 known as Mare, Ilha dos Trades, Madre Deus, Ilha das Fontes, etc. 

 Itaparica and Ilha Santo Amaro became islands at the same time. 



The Bay of Rio de Janeiro has had a similar histor3^ Its basin was 

 carved out by subaerial erosion when the land stood at a higher eleva- 

 tion. It sank and water flowed over and filled the valley leaving it a 

 broad but not very deep bay. Some of the oi'iginal islands of the bay 

 had only shallow waters between them and the mainland, and these shal- 

 lows were eventually built up by the shore waste thrown upon them 

 until the islands were joined to the land. The Pao d'Assucar case cited 

 on page 124 is an illustration of this process. 



The Bay of Santos, about 36.5 kilometres along the coast southwest of 

 Rio de Janeiro, is an excellent illustration of an opeir island-filled bay 

 smaller than that of Rio de Janeiro formed by a depression of the land. 

 This bay is now almost entirely choked up by the silts washed into it 

 from the surrounding mountains and thrown back into it from the coast. 



