OF THE SKRGIPE-ALAGOAS HASIN OF BRAZIL. 415 



have no difficulty in knowiiioj what is to be expected. It is much more widespread 

 through the western than the eastern portion of the province of Sergipe, while in 

 Alagoas it is i-epresented by a succession of mutilated terraces. In the western part 

 of Sergipe the dry, thirsty soil of this formation prevents the growth of important 

 forests, and gives to it the name of ngreste in distinction from the wooded cretaceous 

 soil of the eastern portion. 



Xear the coast, along the Siio Francisco river and the other larger streams of the 

 region, there are scarcely any i-emnants of the tertiary to be seen, but as one recedes 

 from the valleys, outliers of the tertiai'}' become more and more common until it forms 

 continuous taboleiros, deeply gashed at their margins and sometimes abutting uncon- 

 formably against the paleozoic or cretaceous rocks of the serras. On the sketch map 

 on page 'ASH the outliers of tertiary in the vicinity of Aracaju and Maroim are located 

 approximately. The tertiary hills are, as a rule, characterized by having Hat tops. It 

 not infrequently happens, however, that small outliers have been so extensively eroded 

 that tlie upper and most resisting bed of clay has been washed away and the lower 

 beds have broken down rapidly, leaving a more rounded and less characteristic topog- 

 raphy. 



Toward the northern end of the basin the tertiary beds approach nearer and 

 nearer the coast, until, north of Maceib, they form a chain of hills and bluH's which 

 continue with occasional interruptions to near Kio Grande do N^orte. 



T/ie Oiirjiiial Extent of the Tertiary. — The original extent of the tertiary beds 

 is indicated partly by the horizontality and the approximate uniformity in thickness 

 and character of the strata, and, secondly, by the distribution of the existing remnants. 



A satisfactory statement of the original limitations is not possible with our 

 present knowledge, because we are ignorant of and have no means of determining 

 the amount of erosion that has taken place from the up[)er surface of these beds. 

 It is evident, however, fiom the continuity of the lithological characters of preserved 

 beds where exposed over wide areas, ihat the strata were originally continuous over 

 the whole of the lower parts of the Sergipe- Alagoas basin, and over many of its con- 

 siderable elevations as well. 



It is easy thus to calculate a part of the erosion that has taken place, but in 

 view of the impossibility of knowing the oiiginal thickness of these beds, we cannot 

 make a trustworthy estimate of the whole. It is a noteworthy fact, however, that 

 in the Sergij)e-AIaguas basin the remnants of the tertiary beds do not extend 

 beyond the limits of the cretaceous basin into that ul the crystalline rocks. They 

 end uni'onforniahly against the beds of the Serras d'ltaliaiana and M n;i').i, and only 



