CAUSE OF VOLCANIC EXPLOSIONS 83 



has accumulated it distends the film, blowing it up in a 

 dome, till it becomes too thin to resist the pressure, and 

 is then blown to shreds, which are spluttered into the air 

 as scoriae and lapilli. 



The eruption of Stromboli is more or less rhythmical, 

 an explosion taking place at irregular intervals of from 

 30 seconds to 20 minutes. 



This periodicity will readily be seen to follow as a 

 consequence of the process just described. 



The eruption of Yulcano, of so much greater intensity 

 than that of Stromboli, may possibly be due to the steam 

 having to overcome a greater resistance before effecting 

 its escape. 



But the terrible explosions, of which striking examples 

 have been afforded by Krakatoa and Mount Pelee in 

 comparatively recent times, require some more powerful 

 machinery for their explanation. Let us start, as before, 

 with a volcanic pipe filled with molten lava saturated with 

 steam under pressure ; but let us endeavour to devise 

 some other means than the slow 7 movement of con- 

 vection currents to bring about a reduction of pressure 

 on the steam. It would greatly help us in this 

 attempt if we knew what was taking place beneath 

 the base of an active volcano. And, though knowledge of 

 this kind cannot be obtained directly, yet some useful 

 hints may be gathered from a study of ancient volcanic 

 regions. In these, even within the compass of our own 

 islands, we find cases where a once-existing volcano has 

 been completely destroyed by the denuding action of rain 

 and rivers, and perhaps the sea.* Then the ground upon 

 which it stood is exposed to view, and one of the most 

 striking features it exhibits is a multitude of sheets of 

 consolidated lava, some more or less vertical, some 

 approximately horizontal, which proceed from the voJ- 

 * A. Geikie, " The Ancient Volcanos of Great Britain." 



