640 AQUEOUS EROSION IN PALMA. [Cii. XXIX. 



country. As to the fact that no marine shells have yet been dis- 

 covered in the conglomerate, sufficient search has not yet been made 

 for them to entitle us to found an argument on such negative evidence. 

 At the same time I confess that, having found sea-shells and bryozoa 

 abundantly in certain elevated marine conglomerates in the Grand 

 Canary, before I visited Palma, and being unable to meet "with any in 

 the Barranco de las Angustias, I regarded the old gravel when I was 

 on the spot as of fluviatile origin. Such inferences are always doubt- 

 ful in the absence of more positive data, and the intervention of the 

 sea might perhaps account for some phenomena in the configuration 

 of the Caldera and Barranco more naturally than river action. For 

 example, we have the lofty cliff e, fig. 700, p. 638, already mentioned, 

 and c,f, map, p. 628, extending four or five miles from the Caldera to 

 the sea on the right bank of the Barranco, and no cliff of correspond- 

 ing height or structure on the other bank, where for miles towards 

 the southeast there is the platform r, fig. TOO, p. 638, supporting sev- 

 eral minor volcanic cones. The sea might be supposed to leave just 

 such a cliff as e, after cutting away a portion of the southwestern ex- 

 tremity of the old dome-shaped mountain in the north of Palma, 

 whereas a torrent or river might be expected to leave a cliff of similar 

 structure and nearly equal height on both. As to the fact of the old 

 conglomerate ascending an inclined plane, i, I, 7c, p. 631, from the sea- 

 level to an elevation of about 1500 feet, near the entrance of the Cal- 

 dera, this is by no means conclusive in favor of fluviatile action, 

 although some elevated patches of the same may in truth belong to 

 an old river-bed; but in South America gravel-beds of marine origin 

 have a similar upward slope, when followed inland, and the cause of 

 such an arrangement has been explained in a satisfactory manner by 

 Mr. Darwin.* 



Another argument in favor of marine denudation may be derived 

 from that peculiar feature in the configuration of Palma, before alluded 

 to, called the pass of the Cumbrecito (e, fig. 699, p. 631), forming a 

 notch in the uppermost line of precipices surrounding the Caldera. 

 This break divides the mountain called Alejenado, d, p. 631, from the 

 eastern wall, c, /, and cuts quite through the upper formation ; yet the 

 range of precipice, /, e, on the eastern side of the Caldera is continued 

 uninterruptedly, and retains its full height of 1500 or 2000 feet above 

 its base, to the southward of the Cumbrecito, or from e towards a, 

 map, fig. 695, p. 628. In this prolongation of the cliff for half a mile 

 southward, beds of volcanic matter and dikes are seen, as in the walls 

 of the Caldera. 



The indentation forming the pass of the Cumbrecito, e, p. 631,'hais 

 more the appearance of an old channel, such as a current of water may 

 have excavated, than of a rent or a chasm caused by a fault. In case 

 of a fault the lower formation would not be persistent and uninter- 



* Geolog. Observ., South America, p. 43. 



