234 VULCANISM 



(1902) outbursts in Martinique and St. Vincent, and the symptoms 

 of activity at the same time in other places, seem to point clearly to 

 sympathy. On the other hand, the independence of some neighbor- 

 ing vents, as those of Mauna Loa and Kilauea in Hawaii, is extraor- 

 dinary. These two volcanoes are only about twenty miles apart, 

 the one on the top and the other on the side of the same mountain 

 mass. The crater of Loa is about 10,000 feet higher than that of 

 Kilauea, and yet, while the latter has been in constant activity as 

 far back as its history is known, the former is periodic. The case 

 is the more remarkable because of the greatness of the ejections. 

 The outflow of Mauna Loa in 188=5 formed a stream T. to 10 miles in 



<J \J 



width and 45 miles in length, with a probable average thickness of 

 100 feet, and some of its other outflows were nearly as massive. 

 Besides this massiveness, there were extraordinary movements of 

 the lava within the crater, if the testimony of witnesses may be 

 trusted. But throughout these great movements in the higher 

 crater, the lava-column of Kilauea, 10,000 feet lower, continued its 

 quiet action without sensible relation to its boisterous neighbor. 

 No difference in specific gravity that could account for a difference 

 in height of 10,000 feet has been observed or can be presumed. It 

 seems a necessary inference, therefore, that the lava-columns in 

 the two volcanoes have no connection with each other, or with a 

 common reservoir. The tops of some lava-columns stand about 

 20,000 feet above the sea, while others emerge on the sea-bottom 

 far below sea-level. This range of elevation tells its own story as 

 to the independence of vents. 



Eruptions seem to be somewhat more common when atmos- 

 pheric pressure is high than when low, doubtless because the in- 

 creased atmospheric weight on a large area of the crust, aids in forc- 

 ing out lava and volcanic gases. This can be effective only when 

 other forces have almost accomplished the result. Eruptions seem 

 also to be more common when tidal strains favor them, for like rea- 

 sons. In the same class are probably to be put the effects of heavy 

 rains. Such factors are regarded as mere incidents, of no moment 

 as real causes of vulcanism, but of some value in determining the 

 moment of eruption. 



Periodicity. Most volcanoes are intermittent in their action, 

 long periods of dormancy intervening between periods of activity. 

 Some volcanoes supposed to be extinct have renewed their activity 

 with terrific violence. Their periodicity awaits an explanation, 



