364 ON VOLCANOES AND EARTHQUAKES. 



the geologist, are the rocks and minerals which they 

 eject ; as these throw light on many obscure and dis- 

 puted points in the history of the unstratified rocks. 

 I shall not enumerate the long list of volcanic minerals, 

 as they are to be found in all the treatises on minera- 

 logy. The only question of interest here, is the mode 

 of their production ; and I have sufficiently shown, that 

 the greater number are the produce of fusion, though 

 some are formed by watery infiltrations into the cavi- 

 ties of the scoria ; a case identical with that of the 

 generation of the arnygdaloidal nodules in the trap 

 rocks. If it is not always possible to distinguish the 

 two cases, the following general rule may still be 

 commonly applied to them. When the minerals form 

 part of the solid rock, whether it be scoriforrn or not, 

 and particularly when they impress and interfere with 

 each other, they must be considered as the produce of 

 fusion ; where, on the contrary, they occupy the 

 caverns of the scoria, whether filling them or not, they 

 will generally be found to arise from posterior watery 

 infiltration. 



The unfused rocks ejected by volcanoes require no 

 particular notice, as they may consist of any of the 

 numerous substances that chance to lie in the way of 

 the volcanic explosion- It is easy to understand how 

 they may be thrown out little altered, as has sometimes 

 happened to limestone containing shells ; and how by 

 falling back into the crater, and being then exploded 

 again, they may bear, in a greater or less degree, the 

 marks of heat, or even of superficial fusion. The dust 

 and fragments may, in the same manner, be the pro- 

 duce of the natural rock, or of the antient solid lavas 

 which form the crater and the internal cavities of the 

 mountain. Hence the puzzolana arid the tufas of 



