788 MB,. W. BAELOW OjST THE HOEIZOI^TAL 



penetrate, preserving wonderfully straight courses, even across frac- 

 tured and irregTiIar strata, often for miles together. 



An observation recorded by lli\ Dutton relative to the situation 

 of some volcanic vents seems to be confirmatory of the view sub- 

 mitted above, that the presence of precipices or steep declivities has 

 a weakening effect on the masses of rock which they bound, producing 

 in these masses faulting and Assuring that greatly facilitate their 

 degradation. Thus he tells us that basaltic vents occur very often 

 upon the brink of cliffs of erosion, and never (within his observation) 

 at the base of one ; often upon the top of the wall of a canon and 

 never within the canon itself, though the stream of lava often runs 

 into the canon ; and he instances ten large cones standing upon the 

 very brink of the Grrand Canon which have sent their lavas do^vn 

 into it. And he also mentions, away from the Canon, a considerable 

 number of craters upon the various cliffs near the Hurricane Ledge, 

 and far to the north-eastward half a dozen upon the crests of the 

 White Cliffs. He states that out of rather more than three hundred 

 basaltic cones of this region, he has noted thirty-three, or nearly 

 eleven per cent., occupying such positions *. 



The fact of no vents being opened at the bases of the cliffs is quite 

 in harmony with my views, for if the spread of the rock underlying 

 the cliffs is producing a thrust against the crust lying near their 

 bases, as I have argued it is, the tendency will be for this lateral 

 pressure to keep fissures closed. 



If, however, the underlying mass of lava is of great extent, and 

 the ground-surface beyond it much lower than the ground-surface 

 above it, so that the spreading movement of the lava is general and 

 considerable, the local effect just traced may be partially lost in a 

 more general one; the two walls of a canon may move bodily 

 further apart and produce a fissure within it, the site of the canon 

 being a line of weakness. 



Even in this case, if the movement takes place gradually and 

 slowly, it is possible that the local effect just referred to would keep 

 the bottom of the newly forming vent closed, and prevent the 

 extravasation of lava within the canon. 



An interesting case of a volcanic eruption on the verge of the 

 Grand Canon of the Colorado recorded by Mr. Dutton may be referred 

 to in support of my views. On the south side of the canon a lateral 

 gorge or amphitheatre is excavated in the chasm-wall, very nearly 

 as deep as the main abyss. At the summit of the wall of the inner 

 chasm, just at the angle which it makes with this lateral gorge, a 

 ruined basaltic crater stands upon the very brink, the dyke through 

 which the lava came up and several neighbouring dykes being seen 

 projecting from the face of the wall of the lateral gorge throughout 

 a depth of half a mile. The strike of all these dykes is parallel to the 

 river, showing a prohahle connexion between the position of the river 

 and the formation of the dykes. The presence of some remnants of 

 tufa-beds several hundred feet down indicates that the subsidiary 



* ' Geology of the High Plateaus of Utah ' (Dutton), note, p. 203. 



