June 17, 1904.] 



SCIENCE. 



927 



SPECIAL ARTICLES. 

 THE MECHANISM OF THE MONT PELEE SPINE. 



The growth of the Mont Pelee spine is es- 

 sentially an eruption of solid rock. Geologists 

 are familiar with the eruption of solid ma- 

 terial in the form of lapilli and dust, but the 

 escape of lava in a continuous solid mass is a 

 novel phenomenon. Naturally much interest 

 is felt in its explanation, and the subject has 

 been discussed by several geologists. So far 

 as the literature has come to my attention, it 

 has failed to include a factor which appears 

 to me to be of prime importance, and I take 

 the liberty, therefore, of contributing to the 

 discussion, even though my knowledge of the 

 Pelee eruptions is altogether at second-hand. 



The phenomenon to be explained is the 

 gradual issue of a column of rock several 

 hundred feet in diameter and having the full 

 cross-section of the throat of the volcano. 

 This rock is so hot as to be incandescent, ex- 

 cept the immediate exterior, which may be 

 supposed to be cooled by contact with the air; 

 but it clearly is not molten, for the mass as a 

 whole is so rigid as to support its own weight 

 with a height of more than 1,000 feet. Whiile 

 it grows by rising, its height is also reduced 

 by the breaking away of fragments above. 

 Allowing for this reduction, the total length 

 of the extruded column has been estimated at 

 3,000 feet. 



It seems to me quite clear that this process 

 of extrusion is, properly speaking, volcanic 

 eruption; that molten magma rising from 

 some deep source undergoes a change of phys- 

 ical condition in the conduit, and is thereby 

 enabled to issue in solid form. The process 

 is so rapid as to preclude the hypothesis that 

 solidification results from loss of heat by con- 

 duction. Even if surface water finds its way 

 to the walls of the conduit, and is able there to 

 cool the exterior part of the rising column, 

 there can be no appreciable effect on the in- 

 terior part of the column within the short 

 time indicated by the history of the eruption. 

 The suggestion that surface water is absorbed 

 (as steam) by the lava and the lava is thereby 

 cooled, encounters a double objection : (1) 

 That rising lavas, undergoing relief from pres- 

 sure, are in condition for discharging, instead 



of absorbing, gases; and (2) that the diffusion 

 through the rising magma of water absorbed 

 in the periphery of the conduit would require 

 as much time as the cooling of the magma by 

 conduction. 



I ascribe the solidification, instead, to the 

 escape of gases originally contained in the 

 magma, that is, of gases contained before the 

 magma rose in the conduit. Steam is as- 

 sumed to be the principal gas ; but the nature 

 of the gas is not important to the mechanical 

 theory. As the gas passes from the condition 

 of absorption into the free condition, forming 

 bubbles in the magma, it is greatly expanded, 

 and this expansion consumes energy. The 

 case is analogous to that of a body of air 

 rising through the atmosphere and becoming 

 cooler by reason of expansion. In that ease 

 the energy expended in the expansion is fur- 

 nished by the heat of the air itself, and the 

 result is a lowering of temperature. In the 

 expansion of gas within the magma the energy 

 is furnished by the heat of the magma, with 

 the result that the magma is converted from 

 a liquid to a solid condition. There may be 

 other results from the withdrawal of heat 

 from the magma. If its temperature was 

 originally above the temperature of liquefac- 

 tion then it may be cooled as well as solidified. 

 Or the process may, perhaps, go somewhat 

 beyond the accomplishment of solidification, 

 and give to the solid a temperature slightly 

 below the melting point. But all that is 

 necessary to the hypothesis is that the with- 

 drawal of heat from the lava suffices to change 

 it from the liquid to the solid condition. 



If this view is correct, then the remarkable 

 feature of the process involved in the produc- 

 tion of the Pelee spine is the arrest of the ex- 

 clusion and expansion of the gas at the precise 

 stage necessary for the solidification of the 

 magma. Usually it either falls short or 

 passes beyond; in the one case producing a 

 liquid magma charged with bubbles; in the 

 other bursting the solid vesicles and blowing 

 their fragments into the air. The rarity with 

 which the process is arrested at the completion 

 of solidification is probably to be ascribed to 

 the fact that it interacts on itself. The 

 amount of gas which can be held in solution 



