40 Geology, as illustrated by Chemistry and Fhysics. 



only be sought in the increasing heat of the earth towards 

 the centre. The temperature of springs ranges from a few 

 degrees above 32° Fahr. to the boiling point ; consequently at 

 a certain distance below the surface there must be a tempe- 

 rature of 212° Fahr. 



So long as the few springs, which have a very high tem- 

 perature, such as those of Carlsbad, Aix-la-Chapelle, Wies- 

 baden, &c, alone attracted the attention of physicists, it was 

 not difficult to consider them as consequences of local con- 

 ditions, as for example the burning beds of coal or iron 

 pyrites. But since it has been shewn that an excess of 

 temperature of only one degree demands explanation fully 

 as much as the boiling temperature of the Iceland springs, 

 it has become evident that it is only in rare cases that the 

 phenomenon of hot springs depends upon local causes. 



Volcanoes are phenomena which indicate the existence of 

 a temperature in the interior of the earth very much higher 

 than that of the hottest springs. 



Davy's important discovery of the metals of the alkalies 

 and alkaline earths led him to put forward a not very pro- 

 bable explanation of the high temperature which is possessed 

 by the lava streams and ignited stones thrown out during 

 an eruption. As these metals, even at ordinary tempera- 

 tures, decompose water with considerable evolution of heat, 

 it was possible, assuming the existence of these substances 

 in the metallic state at certain depths, and the penetration 

 of water, to conceive volcanic phenomena to be consequences 

 of this decomposition and the heat thus developed. But ad- 

 mitting such to be the case, the hydrogen liberated would 

 naturally issue from the craters in a strongly-heated state, 

 and, coming in contact with atmospheric oxygen, burn there, 

 giving rise to lofty flames, which would form a part of all 

 volcanic phenomena. 



While several observers will not admit the existence of 

 such flames, still others have sometimes recognised them. 

 However, if the oxygen of the alkalies and alkaline earths 

 which constitute such a considerable part of lavas were de- 

 rived from the decomposition of water, the evolution of hy- 

 drogen should be a very common phenomenon, and the quan- 



