50-2 
J)RS. T. ANDERSON AND J. S. FLETT ON THE ERUPTIONS OF THE 
three or four days, and were on the wliole similar to those witnessed on Pelee during 
the present year, except that fissui'es were not formed, and no flows of hot mud are 
spoken of in the accounts. This year the preliminaiy stages of the eruption of the 
Soufriere lasted only for a little more than 24 hours. 
When one eruption follows another after a brief lapse of time the preliminary 
symptoms may be indistinguishable from the closing phases of the previous outbreak. 
Tins was tlie case on the Soufriere on May 9th, and on Pelee on July 9th there was 
notliing, except a slight increase in the amount of steam discharged, to give warning 
of tlie approacliing outbui'st of the avalanche of incandescent dust. Equally sudden 
was the eruption of May 18th at the Soufriere, of which no symptoms were visible 
at nightfall—aljout 7 o’clock—yet at 8.30 an immense cloud of steam suddenly 
towered into the air, with vivid lightnings and loud detonations. 
The only danger during these preliminary stages is the suddenness of the mud 
lavas that may flow down the valleys, burying the houses on their banks, as 
liappened at the Usine Guerin near St. Pierre this year. Even the vegetation 
of the hill, though covered with fine dust, suffers very slightly, and a few .showers 
of rain will remove all the ashes and restore the beautv of the foliage. In 
Martinique, and also in St. Vincent, it is possible that heavy falls of rain took place 
on the higher parts of that mountain during the earlier outl)ursts of steam, but these 
were local and did no leal damaoe. 
o 
3. 77/e Climax and Descent of the Black Cloud .—The preliminary stages gradually 
or suddenly pa.ss into the culminating phases of the eruption. The great black cloud 
wells out of the crater and rushes down the .slojjes, obliteiating all trace of vegetation, 
annihilating every living thing in its path, and leaving l)ehlnd it a desert of a.shes. 
Its appearance is usizally terribly sudden, a few minutes previously the volcano may 
have been onitting oidy a little steam and fine du.st, perliaps more than usual, but 
not enougli to awaken any general alarm. 
On the morning of May 8tli the “Iloddam” had just dropped her anchor; the 
sailors were leaning against the ludwarks, watching the great column of grey smoke 
winch rose from the volcano, wliile the fine dust was gentlv fallino- on citv and 
harbour. “ Then came a sudden roar that .shook the earth and the .sea. The 
mountain uplifted, blew out, was rent in twain from top to bottom. Emm the vast 
cha.sm there belched zqz high into tlie sky a column of belching flame, and a great 
black pillar of cloud. That was all—just the one big roar of the shattering- 
explosion, one flare, and then the cloud, shooting out from the rent, rushing 
down the mountain side on to the doomed city.” * 
It will be seen that in Martiniipie little warning was given, and the residents in 
St. Pierre were uneasy, Init not madly excited. Things looked no worse than they 
had done often during the previous five davs. It Avas the feast of the Ascension, 
and many were hurrying to church. So sudden, so unexpected, vais the great 
* Captain FuEKMAX, “ The Doom of St. Pievre,” 
‘Pearson’s Magazine,’September, 1902, p 316, 
