1849.] LYELL ON THE STRUCTURE OF VOLCANOS. 233 



that a mass of such melted rock may consolidate on a slope of no less 

 than 50° or even 60°, and be continuous for 300 or 400 feet. '' Such 

 masses are narrow," he adds, *' but if the source had been more gene- 

 rous, it is not difficult to see that they would have acquired a greater 

 breadth, and by a succession of ejections upon each cooled layer, even 

 a considerable thickness might have been attained'''." 



The same author has also shown, that in the cinder-cones of the 

 Sandwich Islands the strata have an original inclination of between 

 35"" and 40°t, while in the tufa-cones formed near the sea, they have 

 a slope of about 30°. 



No one who reads the work alluded to will be of opinion, that the 

 laws governing the formation and consolidation of sheets of basaltic or 

 other kinds of lava have as yet been fully ascertained, or that the 

 original inclination which they may have when flowing down the flanks 

 of a volcanic mountain has been definitively determined by the emi- 

 nent French geologist who has collected together so much valuable 

 information on the subject. 



There is another class of facts, however, brought to light by Mr. 

 Dana's investigations, which bear directly on the rectangular junctions 

 of dikes and streams of lava to which I have called attention in refer- 

 ence to the Val del Bove. He has shown, that, while copious streams of 

 lava have been recently known to pour out from Mounts Loa and Kea 

 from openings 13,000 feet above the level of the sea, there have been 

 other contemporaneous fissures, produced at various elevations on the 

 flanks of the same dome, out of which lava has streamed, unaccom- 

 panied by the ejection of any scoriae. It appears that the lava is so 

 liquid, that the entangled gases escape very freely from it, without 

 casting up to great heights in the air liquid jets of the molten rock, 

 to which volcanic dust and cinders owe their origin. Now as these 

 rents are described as running in various directions, it is quite clear that 

 currents of lava descending from higher points must, as often as they 

 pass over them, give rise to junctions resembling those in the Val del 

 Bove, though not strictly at right angles. Still it is quite necessary 

 in the case of Etna, where we have to account for enormous masses 

 of interpolated scoriae, and where there has been so much viscidity in 

 the lava, to derive the beds of fragmentary matter, as I before sug- 

 gested, from a higher and more powerful and permanent central vent, 

 for they could never have proceeded from the lateral openings or dikes 

 without disturbing that uniformity and parallelism of the strata, on the 

 existence of which M. de Beaumont has so emphatically insisted. It 

 is not a little satisfactory to me to discover that Mr. Dana, with whose 

 opinions I was previously unacquainted, has been led by his extensive 

 examination of the volcanos of the Pacific Islands to reject Von Buch's 

 theory of elevation-craters, although he has not alluded to the denu- 

 ding action of the sea as affording an explanation of the large dimen- 

 sions of many of the so-called cavities, such as Santorin, and the others 

 on which I have dwelt in the preceding pages. 



* Dana, Geol. of Amer. Explor. Expecl. p. 359, note. f Ibid. p. 35 1. 



