A Volcanic Phenomenon. 331 



although larger than the average, was so perfectly free from 

 surrounding debris that a post-eruption visitor might well have 

 been pardoned for considering it the least important of the 

 entire series. It was, however, the seat of all the heavier 

 explosions, each of which hurled one or two bombs of stiff, 

 incandescent lava to a considerable distance, but without any 

 accompaniment of ash or other detritus. On the morning of 

 March 30 bombs as large as a meter in diameter were ejected 

 (fig. 2), and these explosions produced the flashing arcs which 



Fig. 2. 



were in all respects identical with those at Vesuvius. Until 

 they were pointed out to him my guide did not observe them, 

 and it is quite possible that I myself might not have seen them 

 had I not known what to look for, as the absence of a darker 

 setting and the lesser distance of the viewpoint rendered them 

 more difficult of observation than at Vesuvius. Some fifteen 

 were clearly seen by both of us, all the stronger explosions 

 producing them. 



What is the nature of this phenomenon ? The only hypo- 

 thesis which seems in accord with the observed characteristics 

 sets forth a proposition which, at first thought, may seem 

 almost startling, viz., that we have to do with visible sound 

 waves. -According to some far from accurate measurements 

 made on the spot the velocity of propagation of the ares 

 seemed to correspond with that of sound. We may certainly 

 assume the outburst to be spherical or at all events globular, 

 as though a huge soap bubble were rapidly blown from the 



