June 1G, 1005.] 



SCIENCE. 



927 



charged is all in a massive-solid condition, as, 

 for example, in case a solid plug is forced out 

 of a volcano's conduit, there would be no loss 

 of pressure so long- as the plug- or obelisk 

 remained intact — except the probably negli- 

 gible part due to deci'ease in atmospheric pres- 

 sure with increase of ascent above the summit 

 of the conduit — and consequently, no tendency 

 to upward expansion. As the extruded ma- 

 terial emerged from its conduit, however, there 

 would be a conspicuous decrease in resistance 

 to lateral expansion and the mass should en- 

 large laterally, and in consequence cool most 

 rapidly by reason of the expansion of occluded 

 steam and gases, in its peripheral portion. 

 That is, there would be a tendency for the 

 extruded mass to expand horizontally on es- 

 caping from its conduit, and the energy so 

 utilized would decrease the tendency to up- 

 ward expansion. The higher the mass rose 

 above the summit of its conduit the greater 

 would be the pressure on its basal portion 

 due to its increasing weight, and for this 

 reason, also, there would be a tendency 

 toward basal enlargement. By reason of these 

 two processes, in case no loss was sustained 

 owing to explosions, dislodgment, etc., there 

 would be a tendency for the rising mass to 

 assume a bottle shape, i. e., a circular tower 

 with an expanded base. 



The two methods of cooling just considered 

 have -a joint tendency to cause the outer por- 

 tion of the ascending mass to become rigid at 

 an earlier stage than the central portion, and 

 hence to form a hollow rigid tube enclosing a 

 more highly heated and perhaps still plastic 

 central core. 



An inspection of the admirable photographs 

 of the obelisk taken by E. O. Hovey, Angelo 

 Heilprin and others, shows that it was situ- 

 ated sub-centrally in the crater from which it 

 was extruded, and adjacent to its northeastern 

 border. The side facing the center of the 

 crater was rudely concave and its outer 

 margin in horizontal section, somewhat defi- 

 nitely convex. The photographs also indicate 

 that subordinate crags within the crater and 

 in part adjacent to the base of the obelisk, 

 were so arranged that in a horizontal section 

 at the level of the crater's rim they, together 



with the base of the obelisk, formed a rude 

 circle. These deductions, together with the 

 fact that explosions occurred from time to 

 time at the base of the obelisk and, as seems 

 to have been the case, mainly on its south- 

 western or inner side, suggest that the tower- 

 like mass was a portion of the encircling wall 

 of an essentially hollow plug of rigid lava 

 constituting the new dome, which was forced 

 upward and out of the summit of the volcano. 

 That is, the obelisk was a portion of the wall 

 of a tubular plug, the greater portion of which 

 became broken and was dislodged as it rose. 

 The ' cork,' in other words, was not solid 

 throughout in cross-section, but composed of a 

 rigid enclosing wall, with less rigid and per- 

 haps still plastic material in its central por- 

 tion. 



If the above deductions from the study of 

 photographs and the observations of others are 

 well founded, they certainly indicate that the 

 cooling and consolidation of the magma in 

 the conduit of the volcano progressed from its 

 periphery toward the center. At the same 

 time, whenever portions of the summit of the 

 column were removed by explosions, or a por- 

 tion of the obelisk fell — there having been no 

 overflow of molten lava — decrease of pressure 

 on the part remaining must have occurred, 

 thus favoring consolidation throughout the 

 summit-portion of the truncated column on 

 account of the expansion of occluded steam 

 and gases as outlined above. 



It is suggestive to note also, as having a 

 bearing on the general process of volcanic 

 eruption, that relief of pressure brought about 

 in the ways just considered, would favor a 

 renewed ascent of the material remaining in 

 a conduit, and likewise a renewal of explosions 

 at its summit. 



In the light of the hypothesis that periph- 

 eral cooling was the controlling condition in 

 the case of the recent massive-solid eruption 

 of Mont Pele, several seemingly discordant 

 observations may be grouped in orderly se- 

 quence. For example, the conical mound of 

 rough, glowing lava present in the crater and 

 resembling a ' cone of eruption ' during the 

 days immediately succeeding the great erup- 

 tion of May 20, 1902, may be considered as 



