AREAS STUDIED AND THEIR CHARACTERISTICS. 259 



Iral railway (formerly called D. Pedro Segundo) leading from Rio de 

 Janeiro across the Serra do Mar. Some of these cuts are still exposed to 

 view, but owing to the tendency of the soft material to wash and slide 

 the railway company has been obliged to cover many of them with a sort 

 of stone pavement. Some of the most interesting cases with which I am 

 acquainted are those of the tunnels on the railway line where it crosses 

 the Serra to Barra do Pirahy'. All of these tunnels are in granites and 

 gneisses, and aggregate 5,189 meters in length. The rocks are so decom- 

 posed that over 2,000 meters of this distance required to be lined with 

 masonry,* and one of the tunnels is said to have required recutting on 

 account of the sliding of the decomposed rocks. 



Along the railway between Entre Rios and the top of the Mantiqueira 

 ^re many noteworthy cuts in decomposed gneiss. At one point the road 

 was so often and so seriously embarrassed by the slipping of the decom- 

 posed materials that the engineers were finally obliged to construct a 

 tunnel — the Cachoeira tunnel. At the crest of the Serra, where the rail- 

 way passes through the " Gargante de Joao Ayres," the decayed rock of 

 the deep cut had to be kept off the track by enormous walls 22 feet high, 

 though the banks were sloped back as usual to a height of 78 feet, f 



On the Uniao e Industria road, a highway that crosses the mountain 

 and extends from Petropolis to Juiz de Fora in the state of Minas, other 

 and scarcely less striking cuts expose the decomposed rocks in many 

 places. Some of these cuts are as much as 50 feet deep. 



The Sao Paulo railway, from Caxoeira to Sao Paulo, shows similar cuts 

 in decomposed rocks, though not so many of them. The profound de- 

 'Composition of the rocks along this line has beeen the cause of not a few 

 landslides and of at least one serious railway accident. J Where the rail- 

 way ascends and passes over the mountain, from Santos toward Sao 

 Paulo, there are several deep cuts in the decomposed rock. 



In his article upon nephelene rocks in Brazil (Sao Paulo and Minas) 

 Professor Derby does not state the depths to which he found the rocks 

 ■decomposed in the region he describes, but one gets the impression that 

 these depths are very considerable, for the railway cuts and tunnels are 

 mostly in the decayed materials. § Derby states elsewhere || that decom- 

 position has been so general in northern Sao Paulo and southwestern 



* Manoel da Cunha Galvao in Revista de Engenharia, Dec. 10, 1879, pp. 6, 7. Agassiz : Journey in 

 Brazil, p. 528. Estudo deseriptivo das Estradas de ferro do Brazil. Cyro D. R. Pessoa, junior. Rio 

 ■de .Taneiro, 1886, p. 210. 



t Estudo deseriptivo das Estradas de Ferro do Brazil. Pessoa, p. 213. Revista do Inst. Hist, do 

 Brazil, li, pt. ii, p. 202. 



t Revista de Engenharia, ii, no. 2, Feb. 15, 1880. 



§ On nephelene rocks in Brazil. O. A. Derby. Quar. Jour. Geol. Soc, vol. xliii, 1887, pp. 462-470. 



II Contribui^oes para o estudo da geographia physica do valle do Rio Grande. 0. A. Derby. 

 Boletim da Sociedade de Geographia do Rio de Janeiro, i, no. 4, p. 15. 



