EARTHQUAKES AND VOLCANOES. 



that it breaks the wall, and rushes down the side of the 

 mountain. 



72. It might be inferred from the occurrence of such tremen- 

 dous phenomena, that the form and condition of volcanic craters 

 would be subject to constant and considerable change. Their 

 variation is, nevertheless, much less than might naturally be 

 expected. Thus, for example, the ramparts of the crater of 

 Vesuvius were accurately measured by Saussure in 1773, and 

 again by Humboldt and Lord Minto in 1822-3. In that interval 

 of 50 years no considerable variation took place. 



73. The case of Cotopaxi producing torrents of water mixed 

 with scoriae and enormous blocks of ice precipitated down the 

 slopes of the mountain by the sudden fusion of the snow which 

 crowns its summit, is not singular. The same phenomena attend 

 all the volcanoes whose cones rise above the line of perpetual snow, 

 of which the chain of the Andes presents many examples. Inde- 

 pendently of these occasional inundations of the external surface, 

 these volcanoes exert a constant action even during their apparent 

 repose, by the slower fusion of the snow in immediate contact with 

 them, and the infiltration of the water through the crevices and 

 fissures of the rocks of which they are formed. Subterranean 

 reservoirs of water are thus formed at and below their bases, 

 with which the streams and rivulets of the surface commu- 

 nicate. It has been ascertained that in these dark reservoirs fish 

 multiply more largely than in the open waters, and that when 

 the caverns containing such waters are suddenly opened, as they 

 sometimes are, by the earthquakes which always precede violent 

 eruptions, water, fish, and tufaceous mud are thrown out in one 

 confused mass. Humboldt says, that when on the night of the 

 19th June, 1698, the summit of the Carguairazo, at the height of 

 nearly 20000 feet above the level of the sea, suddenly fell in, 

 leaving two stupendous peaks of rock as the sole vestiges of the 

 rampart of the crater, masses of tufa, in a liquid state, and of 

 clayey mud, containing immense numbers of dead fish, were spread 

 over a tract of about fifty square miles in extent, rendering the 

 whole space barren. 



In some cases, the quantities of this dead fish which have been 

 ejected have produced putrid fevers, fatal to a large part of the 

 population. 



74. The fiery appearance which is so often observed over the 

 volcanic craters in great eruptions is not name, as it is com TH only 

 supposed to be. It is due principally, if not altogether, to the 

 reflection of the lurid light which issues from the crater, by which 

 the clouds, vapour, ashes, and other ejected matter, forming the 

 column over it, are illuminated, but in part, also, from scoriae and 



174 



