206 J. C. BRANNER AGGRADED LIMESTONE PLAINS, BAHIA, BRAZIL 



built inward until the stream channels are obliterated. In some places 

 the ingrowing valley margins have inclosed extensive caverns. 



With age the deposits change from soft marly beds to hard and com- 

 pact limestones. 



In the inmiediate valley of the Eio Sao Francisco there are extensive 

 deposits of a similar nature, but apparently older than those forming in 

 the lateral valleys at present. 



From the occurrence of these older deposits of the main valley and the 

 method of their formation it seems probable that the Catinga limestone 

 will be found in all of those parts of the arid and semi-arid portions of 

 Brazil where there are older limestones from which they may be derived. 



They are not to be expected in those parts of the country where rainfall 

 is sufficient to keep the stream channels open, but in regions of strong 

 drainage one would expect a Catinga limestone to be deposited only over 

 the higher portions of the floodplains. 



The formation of the Catinga limestones and their peculiar distribu- 

 tion are believed to be due to climatic conditions in combination with the 

 distribution of extensive older limestone deposits from which the sec- 

 ondary beds have been derived. The conditions referred to relate to the 

 variation in the amount and distribution of the rainfall in the region. 



A former study of the geography and geology of the coast of north- 

 eastern Brazil has led the author to the conclusion that during Miocene 

 Tertiary times the region stood at a much greater elevation than it does 

 at present, and that this elevation was followed by a depression during 

 the Pliocene. It is therefore inferred that precipitation throughout north- 

 eastern Brazil, where the winds set constantly inland, was considerably 

 greater during the period of uplift, and that the greater activity of the 

 streams of that period are attributable indirectly to this elevation. The 

 present semi-arid condition, the cessation or great diminution of the 

 mechanical work of streams, and the choking of their channels and the 

 extensive deposits of Catinga limestone in the secondary valleys are all 

 believed to have been brought about by the Pliocene depression and the 

 consequent decrease of precipitation. These conditions — that is, the 

 former elevation and later depression of the land^have affected a wide 

 area in northeastern Brazil, especially in the interior, including the States 

 of Bahia, Sergipe, Alagoas, Pernambuco, Parahyba, Eio Grande do Norte, 

 Ceara, and possibly still others.^* 



I have no doubt that considerable local differences in precipitation are 

 so produced, but they do not seem to be capable of bringing about the 

 widespread changes that have affected northeastern Brazil as a whole. 



'" Dr. E. L. Voss attributes the aridity of Ceard to the surrounding mountains. See 

 "Die Niederschagsverhaltnisse von Sudamerika." Erganzungsheft, No. 157 zu Peter- 

 mann's MItteilungen, Gotha, 1907. Also Geogr. Jour., July, 1908, p. 77. 



