Decembkr 18. 1903.] 



SCIENCE. 



793 



as it rose lost some of its heat, principally on 

 account of the cooling effect of the water which 

 gained access to it, and as it approached the 

 sununit of the mountain, changed to a solid 

 and rigid condition. Steam explosions oc- 

 curred in the congealed portion, of such in- 

 tensity that it was shattered, much of it being 

 reduced to dust, and the fragments produced 

 blown heavenward. During the greatest of the 

 explosions the projected fragments were hurled 

 to a height of four or five miles. 



In the manner briefly outlined above, the 

 rigid summit portion of the rising columns 

 within the craters of the volcanoes was re- 

 moved and fresh material was forced upward 

 from a depth to take its place, and the process 

 repeated. There was thus a transfer of ma- 

 terial from deep within the earth to its sur- 

 face just as truly as if an overflow of molten 

 rock had occurred. In fact the material 

 blown out by Pele or by La Soufriere would 

 in each instance, if melted and run together, 

 form an extensive lava flow, or one, I venture 

 to say, exceeding in volume all of the lava 

 discharged by Vesuvius during the eruptions 

 of 1872. 



The eruptions thus far considered may be 

 termed fragmental-solid discharges, except 

 that on both Martinique and St. Vincent a 

 minor portion of the material extruded was 

 in a plastic condition when blown into the air. 

 This still plastic material probably came to 

 the surface during the later stages of several 

 of the eruptions which furnished the solid 

 angular fragments. 



La Soufriere has exhibited only the explo- 

 sive phase of solid eruptions, but Monte Pele 

 has undergone another and much more novel 

 variation of the same process. Succeeding 

 the earlier and more intense explosions in the 

 summit portion of the conduit of the volcano 

 on Martinique, the solid lava in its throat was 

 forced upward in a massive condition until, 

 in spite of many losses, it stood at one time 

 about 1.000 feet higher than the rim of the 

 crater from within which it was protruded. 

 Its rate of ascent for a period of eighteen 

 days, as observed by ^lajor ITodder, was fort.v- 

 one feet per day. The massive tapering column, 

 or 'obelisk,' was approximately 600 feet in di- 



ameter near its base, and composed of solid or 

 massive rock, and not of angular fragments 

 or adhering cinder-like scoria, as in ordinary 

 cones of eruption. Light was emitted from 

 fissures in its sides, and steam escaping from 

 it showed also that its interior was still hot. 

 The fall of material from the summit and sides 

 of the obelisk and at times its nearly complete 

 destruction, may be accounted for by its struc- 

 tural weakness and by the occurrence of steam 

 explosions owing to rain-water gaining access 

 through fissures to its hot interior. A still 

 more important agencj' leading to its diminu- 

 tion appears to have been furnished by steam 

 explosions about its base. Then, too, as will 

 be understood, a huge irregular plug of solid 

 rock was being pushed out by pressure applied 

 beneath, through an irregular opening in still 

 more rigid rock, and many jars and tremors 

 must have occurred which would tend to dis- 

 lodge masses of material from the sides and 

 summit of the ascending plug. Allowing for 

 the losses due to these various causes, the ele- 

 vation which the obelisk would have attained, 

 had it suffered no loss in height, can reason- 

 ably be estimated at not less than three thou- 

 sand feet. 



The obelisk is evidence that the upward 

 movement of the material in the conduit be- 

 neath it did not cease when explosions in its 

 rigid summit portion failed to remove ma- 

 terial as fast as it rose from a depth, but 

 continued and was the cause of the forcing 

 out of the summit of the plug in the manner 

 described. 



As the rock comxjosing the obelisk was still 

 hot even after rising high in the air, the de- 

 crease in the energy of the explosions which 

 accompanied its growth can not be ascribed 

 to the lack of the necessary heat to vaporize 

 water. The only other alternative seems to 

 be that the rate at which water gained access 

 to the summit portion of the volcano's con- 

 duit diminished, or there was a marked de- 

 crease in the vapor content the rising magma 

 brought from a depth. 



Thus far, as may reasonabl.v be claimed, 

 the conclusions reached are a direct and legit- 

 imate deduction from the facts observed. We 



