354 ON VOLCANOES AND EARTHQUAKES. 



very remarkable feature. It is easy to understand 

 the cause of these forms ; the conoidal shape, like 

 the crater, being the consequence of successive erup- 

 tions from the summit. Hence, also, volcanic moun- 

 tains change their shapes ; while the edges of the 

 craters or the summits of the cones are often demo- 

 lished, and again, sometimes replaced by new ones : 

 and thus too, eruptions at the sides produce sub- 

 sidiary mountains with craters of their own, which 

 modify the form of the principal mass. These se- 

 veral events have occurred in -ZEtna, in Vesuvius, 

 and in Vulcano. In America, they present a great 

 diversity of forms. But the Moon will convey the 

 most tangible idea of many of these varieties ; since 

 every thing is here displayed as in a model, and often 

 in a very explicit manner. 



The accurate form of a perfect crater is that of an 

 inverted conoid ; and, on Cotopaxi and Teneriffe, 

 they are surrounded by walls of lava. The bottom 

 of the crater is generally irregular ; being sometimes 

 a plain, of greater or less extent, containing small 

 irregular cones, among which, during periods of 

 comparative repose, various volatile matters are dis- 

 charged. Thus the present cone of Vesuvius appears 

 a new production, raised within a more antient cra- 

 ter, of which Somma is a remaining portion. The 

 size of the Crater rarely bears any proportion to that 

 of the mountain : in Vulcano it is very large, in Te- 

 neriffe small ; as it is in many of the volcanic moun- 

 tains of the Andes, although that of Cotopaxi is of 

 great size. The depth varies after every eruption ; 

 and, in Vesuvius, these changes are frequent, though 

 the crater of Teneriffe has for a long time appeared 

 very constant. In some eruptions, a volcanic cone 

 has suddenly disappeared, leaving a lake in its place ; 



