THE MECHANICAL ASPECT OF ERUrTIONS. 129 



ever essential the presence of water may be to the total process of erup- 

 tivity, something more is obviously needed, and this additional want Ts 

 apparently well satisfied by a local rise of temperature in the rocks to be 

 erupted. For it cannot be insisted upon too strenuously that from a 

 dynamical standpoint the problem to be explained is the passage of lava- 

 forming materials from a dormant to an energetic condition. And when 

 we resolve this very general statement into a more special and definite one, 

 we find that it means the passage of solid materials into the liquid condi- 

 tion and (as will be indicated further on) a decrease of density. Whatever 

 may be the ulterior cause of volcanicity, a rise of temperature in the 

 erupting masses seems to be an indispensable condition, and in assuming 

 it we are apparently doing nothing more than taking the most obvious facts 

 and giving them the plainest and simplest interpretation. 



III. The third general consideration has reference to the mechanics 

 of eruptions. The fact that lavas are generated at the depth of several 

 miles below the surface being given, how do they reach the surface 1 A 

 study of the geological relations of eruptive masses furnishes a decisive 

 answer to this question. The power of lava to penetrate and burrow into 

 solid rock would never have been credited or even suspected had we not 

 the proof of it in the rock exposures. The opening of fissures and the 

 rise of lava into the gaps is one of the commonest and most intelligible 

 methods. All volcanic areas are traversed by dikes, and near the centers 

 of eruption they are exceedingly numerous. But what is most suggestive 

 is the fact that many lavas, after rising part-way to the surface, suddenly 

 tear open the strata and diffuse themselves between the beds, forming sub- 

 terranean lakes at levels far above their original source. These intrusive 

 lavas are exceedingly common, so much so, that they appear to have con- 

 stituted in all ages a notable proportion of volcanic movements. 



But when a vent is established through which lavas can find escape, 

 we have still to consider the propelling force which urges them onwards or 

 upwards. A veiy common view, long entertained by many geologists, is 

 that the escape of lavas is analogous to what takes place when a bottle of 

 warm champagne is suddenly uncorked. So comprehensible and plausible 

 is this explanation that its wide acceptance is not surprising. In some 

 9 II p 



