DEEP-SEATED CRACKING. 2(59 



viscous flow and would not be allowed to accumulate ; and even if solid 

 but viscous, they might not accumulate to any great extent ; still, they 

 would be of importance as producing a mild state of tension and slow 

 flow upon which a quickly added stress can more eff'ectively act ; but if 

 the great continental elevations are due to unequal cooling, the cold sea 

 bottoms cooling fastest,^ there must be considerable power of accumula- 

 tion of stress. 



The stress diff'erences due to mountainous and continental elevations, 

 which Darwin finds t are from 500 to 600 kilos per square centimeter 

 above those which granite can stand. Their steady application would 

 doubtless be productive of flow, but if owing to the contraction just 

 spoken of the upper compressed crust were left for awhile partly self- 

 supporting and then broke, coming down with sudden stress on the earth 

 below. + there would be a strong tendency for the crack to extend down 

 farther. 



Finally we have the tidal stresses, of comparatively rapid application. 

 The resistance of the earth to them points to its ability to resist, for 

 times measured by fractions of a day, very large tangential stresses. § 



I think that we may conclude, then, that the earth may crack for an 

 indefinite distance down, some ways beyond the depth at which empty 

 caves would become impossible. 



Explanaiorij of volcanic Phenomena. — If this be granted, then the escape 

 of gas from such cracks || would admirably explain volcanic action. The 

 enormous escape of gas at the outset,*f the eruption beginning with ashes, 

 which are followed by lava later ; the known connection of volcanoes 

 w^ith great fissure lines ; their general occurrence on the margin of con- 

 tinents,'^* where special internal tangential stresses must be produced, are 

 all quite natural. Such an origin would also make clear some things 

 hard to explain in any other way, as, for instance, the apparent indifl*er- 

 ence, spoken of by Button, as to whether a volcanic^ eruption occurs on 

 valley or crest. So, too, on the idea that the lava is brought up by the 

 gas from a solid earth, it is easy to see why the lava should stand at 

 different levels in adjacent vents ; also, a change of barometric pressure 

 would naturally effect the outflow of gas, and hence influence the 

 activity .ft Finally, it is easy to see that in the case of a long-continued 



*Faye : Comptes Rendiis, August 19, 1889. 



t Thomson and Tait : Natural Philosophy, p. 425. 



J I do not mean that there is any cavity formed between them, but a temporary relief of com- 

 pression such as is indicated by the less density of rocks under continents. See Preston : Am. 

 Jour. Sci., vol. x.x.xiv, 18S8. p. 305. Faye : Comptes Rendus, 18S6, vol. cii, pp. G27, 1222. 



§ Becker: Am. Jour. Sci., vol. 40, 1893, p. 139. 



II There is a little difference in t!i9 working of a siplion of soda water and champagne. 



^Geikie: Op. oit.. p. 193. 



**Geikie : Op. cit., pp. 206, 259, 



tt Geikie : Op. cit., p. 205. 



