382 
DES. T. ANDERSON AND J. S. FLETT ON THE ERUPTIONS OF THE 
mid-day, and earthquakes Avere ex 2 :»erienced about tlie same time by those at the base 
of the volcano, ljut were local. A violent exjjlosion alarmed the whole neighbour¬ 
hood about 2.40 P.M., and no doubt remained regarding the reality of the eriqition. 
Others followed at intervals of two or three hours, accomioanied by loud noises, the 
most severe being at 7 o’clock, and at midnight. Flames, or the reflection of fire on 
the steam cloud, v’ere noticed first at 4.30 p.:m., and thereafter were seen repeatedly 
bv various observers. 
We shall now consider the evidence of a j)arty of fish sellers who left Chateaubelair 
and crossed to Georgetown on the forenoon of Tuesday. They followed the track 
already mentioned as leading uj^ the crests of the spurs to the lip of the crater. The 
party consisted of several black or coloured women and one boy. Fish are obtained 
only on the leeward side of tlie island, as tlie heavy seas on the windward side render 
small boats unsafe, and these women were accustomed regularly to cross the i.sland, 
and were in consequence very well acquainted with the a23pearance of the j^ath and of 
the crater under ordinary conditions. We cross-questioned them at the refugee camji 
at Barrualli, in piesence of Mr. T. M. ]\IcdJONALD, Avliose assistance was invaluable, 
as it is only after long acquaintance witli tliese ijeoj^le that one can enter into tlieir 
ways of thought and understand their methods of exjjressioii. 
They left Chateaubelair about 6 o’clock in the morning, and in ordinary course 
must have reached the edge of tlie crater Ijetween 10 and 11 a.m. Tliev had no 
reason to susiDect that an eriqjtion was imminent till they were just below the summit, 
when they heard rumbling noises, and felt the hill shaking. The smell of sulphur 
(sulj^huretted hydrogen) from the lake was much stronger than usual, and on looking 
down into the crater they saw that the water was much discoloured, in 2 )laces red, in 
other ^jlaces milky, but elsewhere bluish-green, as usual. “ One part milk, one part 
blood, rest all same as before.” The red colour was probably due to the .stirring up 
of de^^osits of reddish sand, which would be jjroduced by the weatheiing of the red 
ash beds in the crater -wall; the inilky water owed its colour to the de})Osits of fine 
preci|)itated sul 2 )hur which must have gathered on the lake bottom since the last 
eruption. They all state that the level of the lake was almost the same as l)efore, 
tliat no overflow had taken jd^ce, and that within the crater the bush was withered 
but not burnt. 
All agreed that the water Avas boiling and steaming strongly, especially in the 
centre of the lake. Some say they saAv stones being tliroAAm up, but several 
Avitnesses maintained that they saAv a stone in the centre of the lake, and it Avas 
ai'ound this that the sdeam AA-as rising. The last statement is A-ery important, but its 
meaning aauII be better discussed Avhen the history of the eruiotion has been traced a 
little further. 
Another AA-oman travelling by the same path, but in the opjAOsite direction, passed 
the crater about mid-day. She also saAv the AA’ater discoloured. It smelt strongly of 
sulphur, and Avas Ijoiling out in the centre of the lake, not at the sides. As she Avas 
