SOUFPJEEE, AND ON A VISIT TO MONTAGNE PELEE, IN 1902. 
399 
materials were scattered irregularly, some being found on the beach below the bouses, 
others on the opposite side towards the hill. These facts all point to the capricious 
action of lightnino’ from the cloud. 
Similarly, at Turema and at Wallilju, the factory chimneys were knocked down. 
As at hotli places many of the trees are still standing, and as at Turema the houses 
are comparatively little damaged, there can be no doubt that the chimneys were not 
overturned by the blast, and it is generally believed to have been the eftects of 
lightning. It is j^rohable also that some of the huts which were set on tire were 
really ignited in this way. 
The first reports of the catastrophe stated tliat the deaths were practically all the 
result of lightning. This is certainly not tlie case, but it is sufficient to show how 
rapid were the fatal effects as a rule. On the other hand, there can he no doubt that 
the lightnings were the cause of many fatalities. We were told by one man, who 
was looking out of the window of the rum cellar at Orange Hill, that he saw a 
woman starting to run across the yard to the building from one of the huts. Tliat 
instant there was a bright flash, and she fell dead. The coi’pses of some of the 
animals which perished in the fields gave evidence of having been struck by lightning, 
and everywhere on the devastated estates it Is easy to find trees which show the 
effects of the same agency. 
At Owia, according to Mr. Effingham Dun, small and large pelihles were falling 
about 1 o’clock, and from 1.30 to 1.45 there was a rain of hot liquid mud. At this 
time there was a nauseating odour of sulphuretted hydrogen, but it only lasted about 
lialf an hour. It now became very dark, and from about 2 o’clock to 3 o’clock the 
heat was almost suffocating, " and I had to throw water about the house to make 
breathing possible.” At Owia the damage done was comparatively slight. Tlie crops 
were injured but not buried, and none of the inhabitants were killed. 
That the black cloud surmounted the rampart of tlie Somma wall, Avhich rises to the 
north of the crater and poured down the northern slopes of the mountain, is most 
cleai'ly proved by the evidence of the occupants of the boat already mentioned, which 
at about 2 o’clock was off Windsor Forest on Its way from Campobello to Chateaube- 
lair. They saw a dense, impenetrable mass streaming down the Larikai and other 
valleys to tlie south of them, and turned to retrace their course, but Immediately 
afterwards a similar cloud was seen descending the valley at Grand Baleine, so being 
caught between two fires, they had nothing left but to stand directly out to sea. 
The Grand Baleine Valley is the largest on the north-west quadrant of the liill, 
and the main volume of the cloud seems to have coursed along It like a fluid. 
Between this valley and the estates of Fancy and Owia a series of ridges Intervenes, 
and these apparently sheltered the north-eastern corner of the Island and deflected 
the main force of the blast. In this quai'ter tlie destruction was less complete than 
in any other section of the hill. 
At GeorgetoAvn no lives were lost, and it seems certain that the deadly black cloud 
