928 



SCIENCE. 



[N. S. Vol. XXI. Xo. 546. 



the congealed summit of a column of fresh 

 lava which rose in the volcano's conduit. Be- 

 neath this hot and steaming initial plug, as 

 may reasonably be assumed, the heat increased 

 and the solid lava passed gradually into still 

 plastic magma below. Cooling continued not 

 only at the summit of the column as it rose on 

 account of radiation, the escape of steam and 

 gases from fissures, etc., and as has been sug- 

 gested, on account of the expansion of steam 

 and gases occluded in the magma, but also, 

 and at a maximum rate, adjacent to the walls 

 of the conduit principally through the influ- 

 ence of conduction and the ingress of surface 

 water. The rise of the column, owing to pres- 

 sure at its base, caused it to protrude from its 

 encircling crater, but portions of its rigid wall 

 fell or were shattered by steam explosions in 

 its central part and only the remnants left 

 standing reached a conspicuous height. The 

 changes in the position of the obelisk as ob- 

 served from time to time seem consistent 

 with the explanation just offered, since the fall 

 of one prominent spine or obelisk would leave 

 some other portion of the rim conspicuous, 

 and as it in turn was forced upward, seem to 

 take the place of its fallen iwedecessor. 



The fact that the outer surface of the 

 obelisk, i. e., its northeastern side, was striated 

 vertically, owing to friction against the wall 

 of the conduit from which it was protruded, 

 is evidence that there was but little if any 

 opportunity for steam to escape from below in 

 that portion of the periphei-y of the crater. If 

 the rising plug fitted its conduit equally well 

 all about its contact with its enclosing conduit, 

 it is diiRcult to understand where the numer- 

 ous steam explosions which are known to have 

 occurred adjacent to the base of the obelisk, 

 were located — in case the plug was solid 

 throughout its summit-portion. On the other 

 hand, if it is assumed that the inner part of 

 the rising column was less rigid than its outer 

 portion, and as is consistent with this idea, 

 perhaps even plastic at the center, it is legiti- 

 mate to infer that the central portion ' of the 

 summit of the rising column was removed by 

 steam explosion, at the same time that the 

 more prominent crags on the crest of its rigid 



peripheral portion were caused to rise high 

 in the air. 



Another series of facts which demands con- 

 sideration in the above connection is that the 

 rate of ascent of the dome and of the obelisk 

 was not the same. Explosions occurring in 

 the summit of the dome removed portions of 

 its mass and a similar result was brought 

 about in the case of the obelisk by the dis- 

 lodgment of fragments. Varying conditions 

 were thus introduced, but so far as can be 

 judged the rate of ascent in either instance 

 did not respond to these changes. Why the 

 energy exerted in causing the massive-solid 

 extrusion was not all consumed in elevating 

 the core of the plug, where the resistance from 

 weight was the least, and where the tempera- 

 ture was highest and consequently the tend- 

 ency to plasticity the greatest, is far from 

 clear. This is one of the considerations which 

 advocates of the ' fresh lava hypothesis ' to 

 account for the production of the Pele obelisk 

 are called upon to. meet. But in this connec- 

 tion, as in reference to the rapid rate of cool- 

 ing demanded by that hypothesis, an adequate 

 explanation does not seem to be at hand. 



Sequence of Events. — The sequence of 

 events during a massive-solid volcanic erup- 

 tion, and the leading variations in what may 

 be termed the normal process due to secondary 

 conditions, may provisionally be grouped as 

 follows : 



If the magma forced upward from deep 

 within the earth, through a volcano's conduit, 

 becomes highly viscous or nearly solid before 

 reaching the surface, it may be forced out of 

 the conduit and expand in the crater to which 

 the conduit leads. The amount of expansion 

 will depend principally on the degree of 

 plasticity of the mass, the range being from 

 such a degree that the material will flow un- 

 der the influence of gravity, to rigidity under 

 the pressure present. The distinction between 

 an effusive and a massive discharge is that the 

 material extruded in the former instance is 

 sufficiently mobile to flow and possibly to form 

 a well-defined stream, and in the latter in- 

 stance is so viscous or even rigid that lateral 

 motion does not result. Between the two there 



