204 J. C. BRANXER AGGRADED LIMESTONE PLAINS, BAHIA, BRAZIL 



when they spread over this flat region the lime was precipitated wherever 

 the shallow waters were warmed sufficiently. In periods of floods the 

 water spreads out over enormous areas, and when these floods subside the 

 mere draining off of the shallow waters requires a long time, for its move- 

 ment is greatly retarded by vegetation and by ground friction. The lime 

 is precipitated more promptly in the shallower waters — that is, on the 

 higher portions of the flooded areas — owing both to the warming of the 

 waters and to their disturbance. 



Two additional bits of evidence of the former greater volume of the 

 Sao Francisco deserve mention. One is the distribution of the river gravels 

 over the ancient fioodplain far beyond the reach of the highest waters of 

 recent times ; the other is the water-worn condition of the bedrocks. 



On the fazenda Itumerim, some 30 kilometers south of Eio Sao Fran- 

 cisco, and within 200 meters of the base of the Serra do Mulato, are some 

 bosslike outcrops of compact and very hard, fine grained granites that 

 rise 2 or 3 meters above the general level of the plain. These bosses are 

 somewhat exfoliated from exposure to the sun, but at many places over 

 their surface, and especially near their bases, are well preserved remnants 

 of water wearing. These worn surfaces are not pits or depressions, such 

 as are often made by aborigines in grinding food, but they are uneven 

 and rounded, pecked and polished like the channel of a stream cut in the 

 solid rock. These bosses, moreover, stand on an open plain, where they 

 are quite out of the reach of any possible wearing by local streams. From 

 these bosses all the way to the Eio Sao Francisco the surface of the plain 

 is strewn with water-worn and subangular boulders, here more abundant 

 and there thinning out until they almost entirely disappear. 



The plain between the river and the base of the Serra do Mulato is not 

 alluvial, but a nearly flat floor of granites and very old crystalline rocks, 

 with here and there steep-sided hills of quartzite. The evidence of a 

 broad, water-worn and abandoned fioodplain might, if taken alone, be re- 

 garded as evidence of the later lowering of its channel by the river. And 

 some lowering has taken place, but the abandoned fioodplain, considered 

 in connection with the distribution of the limestones, seems rather to be a 

 part of the history of the drainage when the entire region had a much 

 larger rainfall. 



Age of the Catinga Limestone 



From what has been given in regard to the method of formation and 

 distribution of the Catinga limestone, it is evident that it has been and is 

 being formed by a process that has long been in operation and is in oper- 

 ation today. I am therefore unable from field observations alone to refer 

 the rock to any particular geologic horizon. 



