418 THE dlETACEOUS AKD TERTIAllY GEOLO(ir 



shows that out of the total rainfall of 1287 millimetres during the year, ll.:2 milli- 

 metres fell from November to Apiil inclusive. 



Observations made during 18S1 at Sao Bento das Lages in the province of Bahia 

 give a total precipitation of 3984 millimetres for the year, of which 1439 millimetres 

 fell from March to August inclusive.* 



Observations made at Uberaba, province of Minas, from 1880 to 1882 inclusive, 

 show an average annual precipitation of 15G1 millimetres, of which 12G4 millimetres 

 fell from October to March inclusive.f 



Records of the rainfall in Ceara during the twenty-eight years from 1840 to 187G 

 show an average precipitation of l-'^4(i millimetres during one half the year, against 

 143 millimetres during the other half J 



These examples are sufficient to give a correct idea of the distribution in 

 time of the precipitation in Brazil. A rainfall so unevenly distributed cannot fail to 

 do an unusually large amount of erosion. 



On the whole, however, the erosion of these tertiary beds does not appear to be 

 going on so rapidly as formerly, and indeed I am not disposed to believe that the 

 extensive denudation of the tertiary beds which has taken place in Brazil is to be 

 attributed to subaerial erosion alone. It seems probable that the greater part of this 

 work was done at the time of and immediately after the emergence of these beds 

 from beneath the ocean, while the beds were even softer than at present, and before 

 the surface was taken possession of by vegetation. In many places over the tertiary 

 region where the country is thickly clad with vegetation, erosion is practically nil. 

 The great width of many of the valleys and the precipitous faces of the tertiary 

 i-anges that border them suggest that these hills faced the ocean, or were the shores 

 of bays during the time of the land's elevation from beneath the tert'ary seas. 



At many places along the Brazilian coast the ocean is also attacking these terti- 

 ary beds upon the edges, undermining them and spreading the material of which they 

 are composed over the ocean's bottom. The bed of the ocean north of Rio de Jan- 

 eiro is, in most places, covered, not with the glauconite ooze i;sually found along the 

 shores of continents, but for the most part with I'ed mud carried down by the 

 streams, or washed away directly by the waters and undertow from the soft red tei- 

 tiary rocks of the coast.§ The red cliffs, so noticeable to one sailing along the coast 

 in sight of the shore, are all tertiary, and wherever the water breaks against the bases 

 of these cliffs, they arc being rapidly cut away. 



* V. ItevUla dc Kii<jcnhitri<i for Jlay, 1883. 



t V. Hcvisla de Eixjenhiiria, Vol. V. (1883), p. 2.")1. 



J F. E isaio Estalislico dc Pomiieu Oo fiouza IJiazil, p. 10.") (in part). 



t- V, Challenger Reports ; Narrative, Vol. I, Pari I, pp. 315-217. 



