THE BUILDING OF AN ISLAND. 65 



releases this water, which comes away as steam as soon as the pressure on 

 the overlying mass is sufficiently reduced by the approach of the highly 

 heated matter towards the surface. The same may be said of the carbonic 

 acid gas which is often found to come awav from the craters and occasionally 

 the sides of volcanoes. 



There is good reason to suppose that the more important eruptions, even 

 the mightiest, are caused in a similar way to the smaller and more frequent 

 explosions in the crater of Stromboli. When a volcano remains (juiet for 

 some time, the molten rock in the upper part of the vent cools down and 

 becomes solid ; but when the mass below melts again the same pressure as 

 before is set up, and as the melting process proceeds upwards the weight of 

 the overlying rock at last becomes too small to prevent the explosion, which 

 then takes place, shattering the plug at the top of the vent to fragments and 

 throwing these high into the air, sending up at the same time a portion of the 

 molten rock, blown into fine particles. The rush of steam and gas from the 

 vent is often powerful enough to carry the finer particles several miles high, 

 so high that in the West Indies they reach the region of the return trade-wind, 

 where they then drift eastwards with the current, and may travel, as in the 

 case of the several recorded eruptions of the St. Vincent volcano, several 

 hundreds of miles in that direction, the St. Vincent ash falling not only on the 

 island of Barbados, but on the decks of ships far out on the Atlantic. If they 

 are not thrown to so great an altitude as that named above, they may yet be 

 shot so high up in the trade-wind current that thev are carried along by it for 

 hundreds of miles, as was the case during the Mont Pelee eruptions in 1902, 

 when ash from the mountain was brought down by showers of rain in St. 

 Croix and St. Thomas, Martinique being distant more than 300 miles to the 

 southeast, and even in Jamaica, more than a thousand miles from the scene of 

 the explosions. 



The heavier ash and the stones of various sizes are, of course, not carried 

 so far, but fall around the vent. It is in this way that the cone and the crater 

 are formed. The theor)^ of their formation is indicated in Fig. 30, where we 

 must let the few lines showing how the cone is built up, stand for thousands 

 of thin layers of stones and ashes. The successive showers of stones and ash 

 deposit most of their material not far from the vent, some of it slides down 

 again inwards as well as outwards and in this way the cup-like depression as 

 well as the cone is formed. 



It sometimes happens in volcanic outbursts that the molten rock wells up 

 into the crater, breaks part of it down, and then moves slowly down the 

 mountain side. The flow constitutes a stream of lava, a true volcanic rock. 

 Such a stream is genei"ally full of small holes, caused by steam and gas ; but is 

 denser towards the middle than on the surface. Such flows of lava sometimes 

 occur within the mountain instead of down its sides, the molten rock spreading 

 itself in sheets between the beds of ash ; sometimes the pressure within causes 

 upright cracks around the vent, and into these the molten matter is forced 



