42 The American Geologist. January. 190:^. 
hold us in this shower of aslil. Next morning- a new' delta 
14 X 40 feet had grown at the month of the Wallihon. This 
eruption was largely due to the swollen stream reaching a hot 
ash bed. 
On June 24th I witnessed the nuid flow from Pelee and all 
the evidence I obtained from revisiting the field to the crater's 
mouth indicates that this was due to the water sent to the ash 
beds in the Seche and Blanche direct by an eruption from the 
3 miles distant crater. No other explanation can I make hold, 
and there is much to substantiate the view. 
Origin and Action of Ash Deposits. 
Hot ejected rock of varying size from the impalpable dust 
to boulders 21 x 50 feet have been distributed over the great- 
er portion of the slopes of Mt. Pelee and the Soufriere, but 
while they seldom average above a depth of four or five feet on 
the upper surface, concentrations of from 20 to 100 feet have 
taken place within the narrow river beds. 
I am inclined to believe that one important method of this 
canyon concentration has been during great outbursts when 
large volumes of water have been erupted together with the 
heated rocks and dust, and that these masses have rushed down 
the mountain sides in great avalanches, some reaching the sea 
direct, forming deltas and even islands, others being deposited 
as a flood plain within the canyon walls, where the heavier 
boulders, gravitating to the lower portion, the upper part of 
the mud flow serves as a blanket to retain the heat. On May 
5th the Guerin factory on Martinique was buried some 60 feet 
under such a flow. The ash retained, at the time of our de- 
parture from Martinique, July 6th, sufficient heat to cause 
heavy columns of vapor whenever the canyon became flooded. 
It is by the access of water, whether by collection in heavy 
rain, by stream damming process, or by direct eruption from 
the crater or otherwise that these hot ash beds give vent to the 
geyser-like eruptions, sending out great clouds of steam and 
mud, which give the impression of a primary volcanic eruption. 
This has led to the belief that true terrestrial craters exist on 
the lower slopes. This secondary phenomenon has been the 
most instructive of the recent West Indian eruptions. It lay 
spread out, literally smoking hot, exposed to the observation of 
