'498 STRUCTUEE AND ORIGIN OF THE [Cn. XXIX 



The section (fig. 646) is at right angles to the preceding, and cuts 

 through the cone in the direction of the great Barranco, or from north- 

 east to southwest. 



The lowest of the two slanting lines, m, i, descending from the Caldera 

 to the sea along the bottom of the Barranco, represents the present bed 

 of the torrent ; the upper line, k, I, the height at which beds of gravel, 

 elevated high above the present river-channel, are visible in detached 

 patches, shown by dotted spaces at l\ and to the southwest of it, on the 

 same slope. These, and the continuous stratified gravel and conglomer- 

 ate lower down at / and i, are newer than all the volcanic rocks seen in 

 this section. 



The upper volcanic formation, to be described in the sequel, is traversed 

 by numerous dikes, ^vhich could not be expressed on this small scale. 

 The vertical lines in the lower formation represent a few of the perpen- 

 dicular dikes which abound there. Countless others, inclined and tor- 

 tuous, are found penetrating the same rocks. The five outliers of some- 

 what pyramidal shape, at the bottom of the Caldera (on each side of m), 

 agree in structure and composition with the upper formation, and may 

 have subsided into their present position, if the Caldera was caused by 

 engulfment, or may have slid down in the form of land-slips, if the cavity 

 be attributed chiefly to aqueous erosion. 



In the description above given of the section (fig. 646), the cliffs which 

 wall in the Caldera are spoken of as consisting of two formations. Of these 

 the uppermost alone gives rise to vertical precipices, from the base of 

 which the lower descends in steep slopes, which, although they have the 

 external aspect of taluses, are not in fact made up of broken materials, or 

 of ruins detached from the higher rocks, but consist of rocks in place. 

 Both formations are of volcanic origin, but they differ in composition and 

 structure. In the upper, the beds consist of agglomerate, scoriae, lapilli, 

 and lava, chiefly basaltic, the whole dipping outwards, as if from the axis 

 of the original cone, at right angles varying from 10 to 28 degrees. The 

 solid lavas do not constitute more than a fourth of the entire mass, and 

 are divided into beds of very variable thickness, some scoriaceous and 

 vesicular, others more compact, and even in some cases rudely columnar. 

 All these more stony masses are seen to thin out and come to an end 

 wherever they can be traced horizontally for a distance of half or a quar- 

 ter of a mile, and usually sooner. Coarse breccias or agglomerates pre- 

 dominate in the lower part, as if the commencement of the second series 

 of rocks marked an era of violent gaseous explosions. Single beds of this 

 a2:oTe2"ate of ano*ular stones and scoricC attain a thickness of from 200 to 

 300 feet. They are united together by a paste of volcanic dust or spongi- 

 form scorise. 



At one point on the right side of the great Barranco, near its exit 

 from the Caldera, we observed in the boundary pi-ecipice a lofty column 

 of amorphous and scoriaceous rock in which the red or rust-colored 

 scorice are as twisted and ropy as any to be seen on the slopes of 

 Vesuvius ; seeming to imply that there was here an ancient vent or 



