516 CENTRAL VALLEYS. [Ch. XXIX 



cooling slowly under great pressure, like those now incumbent on the 

 impure lignite of S. Jorge. The dip of the latter cannot be clearly deter- 

 mined, since it is exposed to view for too short a distance ; and the same 

 may be said of the leaf-bed, part of which may be traced lower down 

 the ravine. It seems, however, to dip to the north or towards the sea 

 conformably with the general inclination of the basaltic and tufaceous. 

 strata. 



A deep valley, called the Curral (b, fig. 655), surrounded by precipices 

 from 1500 to 2500 feet high, and by peaks of still greater elevation 

 occurs in the middle of Madeira. It has been compared by some to a 

 crater or caldera, for its upper portion is situated in the region where 

 dikes and ejectamenta abound. The Curral, however, extends, without 

 diminishing in depth, to below the region of numerous dikes, and it lays 

 open to view all the beds r, s, fig. 653. Nor do the volcanic masses dip 

 away in all directions from the Curral, as from a central point, or from 

 the hollow axis of a cone. The Curral is in fact one only of three 

 great valleys which radiate from the most mountainous district, a second 

 depression, called the Serra d'Agoa (d, fig. 655), being almost as deep. 

 This cavity is also drained by a river flowing to the south ; while a third 

 valley, namely, that of the Janella, sends its waters to the north. The 

 section alluded to (fig. 655), passing through part of the axis of the 

 island in an E. and W. direction, shows how the Curral and Serra 

 d'Agoa, b and d, are separated by a narrow and lofty ridge, c, part of 



"West 



Section through the central region of Madeira, from East to West. 

 A. Part of the platform, called the Paul da Serra. B. Curral ; a valley, 3000 feet deep. 

 C. Pico Grande. D. The valley of the Serra d'Agoa. 



which is surmounted by the Pico Grande, before mentioned, nearly 5400 

 feet high. There is no essential difference between the shape of these 

 three great valleys and many of those in the Alps and Pyrenees, where 

 the valley-making process can have had no connection with any superfi- 

 cial volcanic action. 



In the Alps, no doubt, as in other lofty chains, the formation of val- 

 leys has been greatly aided by subterranean movements, both gradual 

 and violent, and by the dislocation of rocks. The same may be true of 

 Madeira and of almost every lofty volcanic region ; but, when we reflect 

 that the central heights a and b, fig. 653, are more than 6000 feet above 

 the sea, and that the waters flowing from them, swollen by melted snows, 

 reach the sea by a course of not much more than 6 miles in the case of 

 those draining the Curral, and by nearly as short a route in the Serra 

 d'Agoa, we shall be prepared for almost any amount of denudation effected 

 simply by subaerial erosion. 



