126 



NA TURE 



[December ii, 1902 



LETTERS TO THE EDITOR. 



[The Editor does not hold himself responsible for opinions ex- 

 pressed by his correspondents. Neither can he undertake 

 to return, or to correspond with the writers of, rejected 

 manuscripts intended for this or any other part of Nature. 

 No notice is taken of anonymous communications. ] 



Suggested Nature of the Phenomena of the Eruption 



of Mont Pelee on July 9. Observed by the Royal 



Society Commission. 



Although Dr. Anderson and Dr. Flett were able, at the 

 largely attended meeting of the Royal Society on November 20, 

 to add little to what they had published in their preliminary 

 report three months ago, beyond exhibiting the very full and 

 excellent series of photographs of the affected regions of the 

 Soufriere and Mont Pelee eruptions, they succeeded in 

 exciting renewed interest in the problem of the nature of that 

 eruption of Mont Pelee on the evening of July 9, which 

 they had the exceptionally good fortune to witness under most 

 favourable conditions. The photographs and perfect description 

 of this particular outburst give it an unsurpassed value as a con- 

 tribution to the scientific history of volcanoes, and the Royal 

 Society has therefore the greatest reason to congratulate itself 

 upon the success — a success almost beyond the most sanguine 

 expectation — of its commission to Drs. Anderson and Flett to 

 visit the scene of these eruptions. 



We can now hardly hope [that any fuller knowledge of the 

 nature of an eruption of the kind witnessed by these geologists 

 will be forthcoming through future observations. What is now 

 to be done in order to clear up what remains obscure is experi- 

 mental work in the laboratory. To me it seems that only one 

 point requires investigation before we shall have a definite con- 

 ception and understanding of the phenomenon at the base of 

 such outbursts as those in the West Indies, as well as that of the 

 Bandaisan eruption, or rather explosion, in Japan, closely 

 similar to them in its essential features. 



From the text of the published report, modified a little 

 in the accounts given at the meeting, we know that, after spas- 

 modic bursts of steam, dust, and stones, and discharges of torrents 

 of water and mud, the climax of the eruption came as the 

 welling-up in the crater and overflow, like that of a liquid, of 

 red-hot dust, which descended the mountain side, at first 

 relatively slowly, but with ever-increasing velocity, like an 

 avalanche of snow. This avalanche of incandescent sand was 

 accompanied by a dense cloud, black as night, which soon con- 

 cealed it from view and swelled out in convolutions with terrible 

 energy until it reached perhaps one mile high and two broad. 

 After this, it ceased to enlarge and gradually lost its dense black- 

 ness through ash settling down and leaving nothing visible but 

 white steam. 



There was, therefore, (1) a flow of incandescent sand down to 

 the sea, mainly by gravitation, but with a velocity apparently 

 surpassing that of a torrent of water ; and (2) the expanding 

 motion of the superincumbent, black cloud, together with its 

 rapid motion along the course of the stream of sand, from which 

 it never lifted. Just after the overflow of sand from the crater, j 

 there must have been an enormous outrush of steam and, per- 

 haps, other gases, and this will have had some effect in driving 

 the cloud through the air ; but the progressive formation and 

 the appearance of the cloud forbid the belief that this effect 

 could have been considerable. That the cloud enlarged upwards 

 rather than laterally was due to its consisting of heated steam, 

 for although the dust which it carried with it will have impeded 

 the velocity of its expansion, it will not have lessened its 

 extent. 



There can be only one conclusion drawn as to the cause of the 

 free motion and rapid rush of the torrent of sand and of the 

 swelling, convoluting cloud, and that is the continuous evolution 

 of water vapour from every particle of the moving hot sand. Pos- I 

 sibly some other gas may also have escaped, but if so only in 1 

 relatively small quantity, as otherwise the water vapour would I 

 not so easily have condensed and become visible. Violent 

 friction between the issuing steam and the solid particles may 

 sufficiently account for the extensive electric dischrrges. The 

 continuous escape of this water from the particles of the hot 

 sand, at such a high temperature, even though in small quantity, 

 would surround every particle with a compressed atmosphere of 

 steam sufficient to keep it apart from all others, and thus pro- 

 duce a quasi-liquid mass which, on account of the density of 



NO. 1728, VOL. 67] 



the sand, would gravitate strongly and at the same time would, 

 by virtue of the interstitial compressed steam preventing all 

 rubbing together of solid particles, give the mass its marvellous 

 mobility. That this would be so is easily borne out by facts 

 familiar to the chemist and physicist. One of these was, 

 indeed, brought up by Sir William Ramsay in the discussion 

 which followed upon the reading of the papers, namely, the 

 behaviour of precipitated silica when heated, which, however, he 

 attributed to a movement of particles in gases similar to that of 

 Brownian movements of particles in liquids. When any fine 

 dust or powder, which is non-coherent whether cold or hot, 

 gives off sufficiently fast a gas or vapour when heated, it will, 

 when smartly heated, swell up and become mobile, sometimes 

 almost as mobile as liquid ether, keep a horizontal surface when 

 its containing-vessel is tilted, and admit of being poured like a 

 liquid into another vessel. Because of its frequent presence in 

 the work of inorganic chemical analysis, precipitated silica is, 

 perhaps, the most widely known example of this behaviour. 

 In ordinary circumstances, the silica acts in this wayalmost wholly 

 in consequence of its continuing to liberate up to even a blowpipe 

 heat the water always present in some form of combination 

 with it. Probably, too, it and all such light powders owe for 

 a moment part of the movement of their particles from each 

 other merely to the rapid expansion of the air in the interstices of 

 the powder when the containing vessel is quickly heated, but 

 the escape of hygroscopic or other moisture is obviously the 

 principal cause. When the silica is kept steadily heated, it 

 loses most of its mobility. Other hygroscopic or vapour-con- 

 densing powders behave similarly ; very finely divided charcoal 

 powder is generally a good example ; magnesia alba is another, 

 which gives out both carbonic acid and water. 



Light bodies are naturally best fitted for the observance of 

 this phenomenon, bufmanganese binoxide when evolving oxygen 

 shows it, and even platinum black will throw up dust and en- 

 large. Indeed, it is a common phenomenon for a slightly coherent 

 powder suddenly heated in a platinum crucible to float in motion 

 as a moulded mass in an atmosphere of gas generated from 

 itself by the hot walls of the crucible. Not inapposite instances 

 of the power of escaping vapour to hold up bodies is that 

 familiar phenomenon of liquid water or alcohol assuming the 

 spheroidal state, that is, rolling about on a hot plate without 

 touching it, being couched on a bed of its own continuously 

 evolved vapour. Where experiment is now wanted is to find 

 out what andesitic minerals will, under great pressure, combine 

 chemically or physically, but intimately, with water at a red out 

 heat and then retain it sufficiently when the pressure is released 

 for an appreciable though short time to elapse before the 

 regeneration of the steam is ended. 



A modification of the explanation here given suggests itself 

 which would do away with the necessity for the existence of such 

 combinations of water with rock materials. It is that as the 

 incandescent sand flowed over the soil, it generated the steam 

 from the damp earth or hydrated rocks beneath it in such 

 quantity as to buoy up the sand from the soil and separate its 

 particles. In accordance with this view would be the observ- 

 ation that the hot sand visibly (that is, without obscuring cloud) 

 poured over the lip of the crater and then as it fl iwcd down 

 obscured itself in cloud. On the other hand, the escape of gas 

 or vapour caused by cooling is not an unknown phenomenon, 

 while against this view is the difficulty to explain when holding 

 it the production of the sand within the crater. Drs. Anderson 

 and Flett speak of the dust as lava blown to pieces by the 

 expansion of the gases it contains. I would suggest that the^pro- 

 duction of the sand just in that way is inconceivable ; for if the 

 lava had been molten, it would have been scattered in drops and 

 vesicles in all directions, and only if solid would it have become 

 dust, while in either case it would not have remained as a 

 mass of sand, but have been scattered to the winds. The pro- 

 duction of sand or dust, if it really was produced in the crater, 

 will have been a disintegration of rock masses by the pressure 

 diffused through them of the condensed water, with which they 

 were impregnated and perhaps combined, a disintegration 

 leading up to the falling to dust of the masses while they were 

 still under sufficient pressure to prevent scattering. 



The strong escape of steam from the sand would, of course, 

 carry up much of the dust with it and thus constitute the black 

 cloud, while its cauliflower-like expansions were apparently 

 only an exaggerated form of what is to be seen over a seething 

 cauldron or a stream of boiling water. Edward Divers, 

 November 22. 



