SOUFEIEEE, AND OX A VISIT TO MOXTAGXE PELEE, IX 1902. 4G9 
great importance, as affecting the suj^ply of water to the sugar works and rum 
distilleries. 
The material which filled up these valleys is called “ lava,” hut there can l)e no 
doubt it was sand and ashes. Shephard expressly states that the Wallibu was 
dammed up with sand, and that some years afterwards “ the gradual accumulation of 
water burst through the sandy barrier and carried away many negro houses in its 
jjrogress ; 32 slaves belonging to Wallibu Estate were washed into the sea by the 
torrent.” 
In short, it is clear that after the eruption of 1812 the upper parts of the Wallibu 
and Eabaka Valleys were in very much the same condition as at present, filled with 
deep deposits of sand, which obstructed the lateral valleys and caused lakes of water 
and of mud to collect there. Obviously, on the night of April 30th, in the height of 
the eruption, an avalanche of sand swept down the mountain and lodged in the 
ravines on the south side of it. With it there must have been a black cloud and a 
hot blast, only the eruption was less violeait than that of this year, and the black 
cloud had lost its burden of hot ash, and lifted off the surface of the ground before 
it reached the Carib Country, so that it mounted into tlie air over the heads of the 
inhabitants, and tliey narrowly escaped a sudden and painful death. 
The avalanche of dust must have welled over the southern lip of the main crater, 
and we learn, from Shephakd and other sources, that after the eruption tlie inteiior 
cone had vanished. , Major Crichton is said to have visited the crater three days 
after the eruption,and to have found in it a lake of water. It was then in very 
much the same condition as before the eruption of this year—that is to say, its depth 
Iiad increased to about IGOO feet, and it had a highly concave, basin-shajjed floor. 
It is, of course, possible to argue that this discharge of sand took the form of 
a mud lava, and not of a dry hot avalanche, hut this explanation will.not meet the 
exigencies of tlie case when vn consider the phenomena on the north and west sides 
of the mountain. About midnight on April 30th, when the epoch of maximum 
activity supervened, a great outburst was noticed on the north-west side. I'his 
marked apparently the formation of the new crater. A current of red-hot matter 
discharged from it, passed down the valley of Larikai to the north of Moine Ilonde. 
It is also stated [see p. 4G4) that the lava,I “ immediately after boiling over the orifice 
and flowing a short way, was opposed by the activity of a higher point of land, over 
which it was impelled by the immense tide of liquefied fire that drove it on, foiniing 
the figure of Y in grand illumination. Sometimes, when the ebullition slackened, or 
was insufficient to urge it over the obstructing hill, it recoiled back, like a refluent 
billow, from the rock, and then a.gain rushed forward, impelled by finsh su])plies, and, 
scaling every obstacle, carrying rocks and woods together in its course down the 
slope of the mountain, until it precipitated itself down some vast ravine concealed 
* P. Foster Huggins, ‘ An Account of the Eruptions of the St, Vincent Soufriere,’ p. 94. 
+ Blue Book, p. 94. 
