500 
DRS. T. AXDEEROX AXD J. S. FLETT OX THE EEUPTIOXS OF THE 
stripped of leaves and branches ; but beyond the limits covered by the great black 
cloud no effects are produced, other than those consecpient on the rain of ashes 
which precedes or follows the avalanche. Except for the dust avalanche, there is 
nothing unusual about these eruptions. They have, indeed, man}" points in common 
with those of volcanoes of the ordinary explosive type, to a sub-group of w"hich they 
evidently belong. 
It is possible for us at the present time only to compare the history of the different 
outbursts of Pelee and the Soufriere in order to ascertain what features they have in 
common which are distinct from those of other eruptions, and to form a general idea 
as to what are the stages of their action, and what variations are possible in the type 
they represent. Even for this purpose the data are not yet so full and exact as w’e 
hope they ultimately will be, when the various scientific men wdio are engaged in the 
study of these volcanoes at tliis moment have completed their labours, and given the 
world the results of their investigations. But sufficient is knowm already to enable 
us to formulate certain preliminary conclusions and working hypotheses. 
We wall avail ourselves, in the first place, of the results of our own observations in 
St. Vincent and Martinique, and of our experiences during the ei'uption of Pelee on 
July 9th, 1902. But the Beports of the Commissioners appointed by the French 
Academy of Sciences, and of the various parties of American scientific men, already 
quoted hy us, have furnished many additional facts of importance, and not a few 
theoretical suggestions, wdiicli we shall be able in some cases to adopt as in accordance 
wdth the o|)inions w^e have been led to form l)y our owm investigations. It is only by 
the collection of a large Ijody of facts from authentic sources, well and carefully sifted, 
that this ne^v branch of the science of volcanology w"ill ultimately be established on a 
secure basis, and its results and conclusions entitled to the confidence of scientific 
men. 
The Stages of the Eriqhions. 
We have in a Pelean eruption certain distinct stages to recognise and describe. 
They may be classified as follow"s :— 
1. The premonitory sym})toms wdiich herald the volcanic outbreak, but are attended 
by no actual emissions from the volcmio. 
2. The preliminary stages, in wdncli activity has been resumed and discharges are 
produced from the crater. Tlieir violence increases more or less rapidly, and 
they lead up to— 
3. The climax ot the eruption, which is manifested l)y the appearance of the 
avalanche of incandescent sand and the passage of the great black cloud. 
4. The concluding stages, during which the volcano sinks iiito more or less complete 
repose. 
These stages must be admitted to be more or less artificial and arbitrary. In some 
of the eruptions they c.ui all be recognised as distinct, in others there is a tendency 
