516 
DRS. T. ANDERSON AND J. S. FLETT ON THE ERUPTIONS OF THE 
great black clouds in Martinique, and on the night of July 9th the wet ash which 
fell ujDoii our decks smelt of sulphurous acid, though not strongly. Sulphuretted 
hydrogen is, of course, combustible, but would occasion no bright flame, and its 
presence cannot be blamed for the devastating forces which levelled the town. 
We may also be certain that other gases were there, some of them highly 
inflammable, but it is impossible to believe that they were present in relatively 
large amount. In this line of inquiry the classic researches of Fouque on the 
gases of the andesitic magma of Santorin are the best guides. Hydrogen and 
various compounds of hydrogen and carbon were ascertained by him to be present 
in that case, and there is every probability that they were here also, but there are 
no data to enable us to judge what were their composition, proportions, and 
abundance. They may probably have occasioned tiivial explosions, but could 
only have ignited when mixed with air—that is to say, on the outer fringes of 
the cloud. They were not its motive force, and could not have supplied the energy 
which launched it on its errand of destruction. 
Hydrochloric acid has been mentioned by some as having been observed in 
connection with certain of the eruptions. No doubt it was there, but as it has 
left no visible effects, so far as we knowq we cannot suppose it to have been 
23lentiful. Nitrogen is another probable component, and so also is carbon dioxide, 
which accompanies practically all volcanic outbursts ; but at the present moment 
little can be said about these, for, though we may admit that they were present, 
the evidence resi^ecting them is altogether of a negative kind. There is nothing 
to show that they played any important part in the mechanism of the eruptions. 
It has been suggested by Professor Verrill"^ that the dissociation of steam 
within the volcano produced an explosive mixture of h^xli'ogen and oxygen which 
combined with violence on emerging from the mouth of the crater, and that this 
explosion explains the disastrous force which razed the city of St. Pierre. That such 
dissociation takes place to a certain extent is made probable by the researches of 
FoUQU^,t but, owing to the enormous pressures, this could have affected only a small 
part of the watei- in tlie magma, which, moreover, was at a temjjerature not vei'y 
much higher than that at which steam begins to be dissociated at atmospheric 
pressure. 
The crystalline minerals which were floating in the magma before eruption 
show that it was at a temperature of less than 1200° C., and this renders it 
probable that the amount of dissociation produced would Ije quite inconsiderable. 
The magma of Santorin, to judge by the nature of the jDroducts, is considerably hotter 
than that of Pelee at the moment of effusion. It is also certain that when the lava 
broke forth, cooling would be continuous and not instantaneous, and the gases would 
* ‘American Journal of Science,’ 4th Series, vol. xiv., p. 72 (July, 1902); also by Professor IsR.4EL 
C. Russell, ‘Century Magazine,’ vol. Ixiv., p. 790, September, 1902. 
t ‘ Santorin et ses Eruptions,’ p. 232. 
