Full text of "Mont Pelée and the tragedy of Martinique : a study of the great catastrophes of 1902, with observations and experiences in the field"

See other formats

:CN

= 00

100

•CD

MONT PELEE

AND

THE TRAGEDY OF MARTINIQUE

SECOND EDITION

Photo. Hcilprin

PELEE IN ERUPTION— AUGUST 24, 1902
The entire crater working

MONT PELEE

AND

THE TRAGEDY OF MARTINIQUE |» J

A STUDY OF THE GREAT CATASTROPHES OF 1902, WITH
OBSERVATIONS AND EXPERIENCES IN THE FIELD

ANGELO HEILPRIN

PRESIDENT OF THE GEOGRAPHICAL SOCIETY OF PHILADELPHIA

Vice-President of the American Alpine Club ; Fellow of the Royal Geographical Society of

London ; late Professor of Geology at the Academy of Natural Sciences

of Philadelphia, etc.

«Attet> ?- BY THE AUTHOR

&*1 €>;

ILLUSTRATED WITH PHOTOGRAPHS LARGELY TAKEN

PHILADELPHIA AND LONDON

J. B. LIPPINCOTT COMPANY

1903

Published January, 1903

ELECTROTYPED AND PRINTED BY J. B. LIPPINCOTT COMPANY, PHILADELPHIA, U. 8. A.

TO
PROFESSOR EDUARD SUESS

PRESIDENT OF THE IMPERIAL
ACADEMY OF SCIENCES, VIENNA

PREFACE

In presenting to his readers the following pages dealing
with one of the most noteworthy, even if lamentable, inci-
dents in the world's history, the author feels that he must
do so with the apology that the work is only partly done.
The magnitude of the phenomena that are associated with
the Pelee eruptions, and the obscurity in which many of
the facts pertaining thereto still remain, will necessitate
further research before the episode can be made fully
known in all its relations, and probably some of the con-
clusions here set forth will have to be modified in the light
of future investigations. But the history as it stands may
be considered measurably complete, and it has the advan-
tage, at least, of being based largely upon personal observa-
tion.

The author's two visits to Martinique were made after
an interval of three months, in the latter part of May and
again in August, and during these times he enjoyed unu-
sual opportunities for the prosecution of his work. The
pleasing courtesies of the people of Martinique helped
largely to whatever of success was obtained, and contributed
a degree of comfort in labor the absence of which would
have been sorely trying. During the later visit it was the
author's privilege to be a close witness of the second great
death-dealing eruption of Mont Pelee, and he had thereby

vii

viii PEEFACE

the marked advantage of being able to make his investiga-
tions in a newly-culled field.

The author feels himself under obligation to many, on
and off the island, who in one way or another proffered
assistance, and to these collectively he extends his thanks ;
but the history of personal travel would not be complete
without a special acknowledgment being made to his friends
of Vive, Assier and Trinite, who neglected no effort to
insure comfort to himself and to his associates. These are
MM. Fernand and Joseph Clerc, Lagarrigue de Meillac,
Teliam de Chancel, and, not least, Mile. Marie de Jaham,
the affable hostess of the Clerc establishment. A special
expression of thanks is also due to M. Louis des Grottes,
of the Habitation Leyritz, United States Consul Louis H.
Ayme, whose many kindnesses brought a ready introduction
of the author to the island, and M. Ivanes.

The illustrations that accompany the work are largely
from photographs taken by the author himself, and many
of them represent, in a way that has probably not been
possible before, the consecutive stages in the paroxysmal
eruption of a very active volcano. Other photographs were
obtained through the kind permission of Messrs. Under-
wood & Underwood, of New York, whose representative
in Martinique was for a while associated with the author in
his studies of Mont Pelee.

Angelo Heilprin.
Geographical Society op Philadelphia,

December, 1902.

CONTENTS

CHAPTER PAGE

I. Impressions op Martinique 1

II. Saint Pierre and its Buins 16

III. The Cataclysm of May 8 35

IV. Days op Fear and Trembling 60

V. The Last Day op Saint Pierre 73

VI. Vicar-General Parel's Chronicle 85

VII. After the Conflagration 109

VIII. Vesuvius and Pompeii — A Parallel 121

IX. Across the Island to Assier 140

X. To the Storm-Cloud of Pelee' s Crater 151

XI. The Geography of Mont Pelee 166

XII. Pere Mary, Cure of Morne Eouge 189

XIII. Clouds of Passage 197

XIV. A Second Visit to Martinique 208

XV. Battling with Pelee 216

XVI. A Night of Illumination, and the Destruction of

Morne Eouge and Ajoupa-Bouillon 227

XVII. The Soufriere of St. Vincent and the Afterglows .. 244

XVIII. The Volcanic Eelations of the Caribbean Basin 257

XIX. The Phenomena of the Eruption 271

Appendix 319

Notes 329

Index 333

ILLUSTRATIONS

PLATES

PAGE

Pelee in Eruption — August 24, 1902 (Frontispiece)

1. The Pitons du Carbet with Pelee's Ash-Cloud in the

Distance 4

2. Street-Sweepers — Fort-de-France 8

3. Saint Pierre and Mont Pelee 16

4. Along the Roxelane^Saint Pierre 20

5. The Eed-Tiled Roofs op Saint Pierre 30

6. Rue Victor-Hugo, May 14, 1902 38

7. Cathedral of Saint Pierre 46

8. Cathedral in Ruins — Saint Pierre , 58

9. Fac-Simile of First Page (Last Issue) of Les Colonies. 68

10. Saint Pierre in Ruins 80

11. Mud-Flow of May 5 88

12. The Silent City 110

13. Bodies in Basement of House 124

14. A Martinique Pastoral — Assier 140

15. Evening Glow on Pelee's Pennant 150

16. Great Ash-Cloud Turning Day into Night 160

1-7. On the Volcano's Devastated Slope 168

18. The Majesty of Pelee's Inferno 180

19. The Morne de La Croix 188

20. A Deluge of Boulders 204

21. Morne Rouge and Pelee 218

22. Before a Shrine — Morne Rouge 230

23. The Darkening Cloud of June 6 244

24. The Enveloping Ash-Cloud of June 6 256

xii ILLUSTRATIONS

PAGI

25. Pelee in a Paroxysm, June 5 ?_^^266

26. Towering and Mushroom-Shaped Clouds 274

27. A Cocoanut-Gatherer — Assier 282

28. Pele*e " Smoking" prom the New Fragmental Cone 288

29. The Heavens Aglow, May 26 294

30. The Issuing Blasts from the Crater 304

31. Progressive Development op an Eruption 312

32. Progressive Development op an Eruption 312

33. Approaching Climax op Eruption 318

34. Ash-Cloud Three Miles Overhead ; Ash-Cloud Eising

from the Crater 318

35. A Sudden Blow from the Crater 326

36. The Devastated Eegion of Martinique (Map) 337

TEXT

Fort-de-France and the Pitons du Carbet 3

In the Savane of Fort-de-France 7

Martinique Woman 12

The Forest Solitude 15

The Theatre — Saint Pierre 18

Rue Victor Hugo — Saint Pierre 23

The Cathedral of Saint Pierre — August 23, 1902 •. . 28

Pelee in the May Eruption 40

Iron Bridge across the Roxelane — Saint Pierre 49

Cable Chart of the " Pouyer-Quertier" 55

Recovered Cable Strand Enveloping Branch of Tree 57

Opposite the Riviere Seche 68

Basse-Pointe— May 30, 1902 79

Thrown Statue of " Our Lady of the Watch" 98

Saint Pierre Burning 108

Rue Lucie — Saint Pierre 131

Burial- Vault — Saint Pierre 133

ILLUSTRATIONS xiii

PAGE

Bodies on the Terrace Road 137

Refugees on the Road — Grande- An se 143

Street Scene — Lamentin 148

Ash-Cloud of Pelee — From near Assier 154

A Burst from the Crater 163

Basin of the Lac des Palmistes and the Shattered Morne

de La Croix 171

Pelee Smoking from Fragmental Cone 179

On the Road to Morne Rouge 191

The Bomb-Scarred Slope of Pelee 221

The " Roraima" Burning 226

Saint Pierre and Mont Pelee in 1766 270

The Island of Martinique 328

MONT PELEE

THE TRAGEDY OF MARTINIQUE

IMPRESSIONS OF MARTINIQUE

My first view of the unhappy island whose misfortunes
have so deeply roused the sympathies of the world was in
the early morning of the 25th of May, two and a half
weeks after one of the greatest tragedies recorded in history
had been enacted on its shores. The Fontabelle was then
steering her course close in shore, but it was not until we
had passed the nimbus of the great ash-cloud that Pelee
was throwing out to sea that we began to distinguish the
features of recognizable land. The island in front of us
was not a tropical paradise, but a withered piece of the
earth that seemed to be just emerging from chaos. Every-
thing was gray and brown, sunk behind a cloud which only
the mind could penetrate ; there was nothing that appealed
restfully to the eye.

The landscape was barren as though it had been graven
with desert tools, scarred and made ragged by floods of
water and boiling mud, and hardly a vestige remained of
the verdant forest that but a short time before had been

2 IMPEESSIONS OF MAETINIQUE

the glory of the land. Great folds of cloud and ash hung
over the crown of the volcano, and from its lower flanks
issued a veritable tempest of curling vapor and mud.
Lying close to its southern foot, and bathed in the flame
of a tropical sunshine, was all that remained of the once
attractive city of Saint Pierre — miles of wreckage that
reached up from the silent desert of stone and sand, show-
ing no color but the burning grays that had been flung to
them or that had formed part of mother earth.

We entered the harbor of Fort-de-France shortly after
eight o'clock, and took our place beside the white flank of
the Suchet, whose work in the catastrophe of the 8th had
made it famous among its craft. Two other men-of-war,
their sheets drooping from the foreyards like linen in a
Neapolitan passaggio, were also sweltering under the genially
tropical sun, with schools of gars and dog-fishes swirling
about their hulls. The city did not at this time impress
me as being particularly concerned in the havoc that lay
so close to its doors. It being the Sabbath day, the busi-
ness streets wore the usual dress of pleasurable inactivity,
and only in the savane, or open square, and in and about
the hotels, was there anything to remind one of a serious
life. Uniformed members of the army and navy, city and
state officials, black and white, newspaper editors, scientists,
and others were gathered around in groups, discussing the
two important topics of the day — the elections that had
recently been held and the possibility of Pelee's activity
invading the city. The volcano itself is not visible from
the lower part of Fort-de-France, but its great white cloud,

IMPEESSIONS OF MAETINIQUE 3

whose towering at all times attracted the attention of some
eyes, helped to keep it in evidence, and supplied a never-
failing ferment for conversation and argument.

I took lodgings in a promising location at the inner end
of the green which surrounds the statue of the Empress

Photo. Heilprin

FORT-DE-FRANCE AND THE PITONS DE CARBET

Josephine, and where my room opened up on tiled roofs
and circling corridors, and the distant flowing curls of the
volcano. The hotel was disorganized and the service gone,
but for this Pelee was properly held responsible, for its
recent eruptions, especially that of May 20, had created
a degree of consternation among those who did not permit
themselves to believe that security was assured by distance

4 IMPEESSIONS OF MAKTIMQUE

which could be realized only by those who had lived
through the recent occurrences of the unfortunate island.
It required but a warning to set the population in panic,
and many thought it a wise precaution to place themselves
where warnings were not a necessary prelude to a peaceful
living.

Fort-de-France, which is now, after the destruction of
Saint Pierre, the most important centre of population in
the island of Martinique, occupies part of the northern
face of one of the best harbors of the Lesser Antilles, and
is backed by the heights of Carbet on the north. It lies
close to the water's edge, with only two to five feet level
between it and the surface of the sea, and thus invites to
itself a form of catastrophe which has more than once
visited other parts of the island. On the night of the
great eruption of August 30, the sea rose close to the
outer border of the savane, directly abreast of the main
hotel. The lower parts are built on made ground, and it
is, therefore, with just fear that the people look to a pos-
sible ras de maree.

The city has little to show for itself as a municipium
of nearly eighteen thousand inhabitants, the seat of govern-
ment, and the depot of naval and military stores. Until
the conflagration of 1890, which destroyed its major por-
tion, it was built chiefly of wood, but since that time stone
and rubble form the principal materials of construction.
The only conventionally interesting sites or locations are
the city green or savane, hardly cared for but ornamented
with a number of stately and regenerated royal palms ; the

: I

i r-
i m

■ >

- CO

IMPEESSIONS OF MAKTINIQUE 5

allees of rubber, tamarin and giant sabliers (Hura crepi-
tans) ; the cathedral, and the shedded market, where may
be observed at close range the faces of all nationalities
known to Martinique, and a Babel of voices heard at
nearly all hours of the day.

Beyond this there is little to attract, although many
interesting phases and pictures of life can be picked up by
those seeking new impressions, especially along the banks of
the picturesque, even if not wholly pure, Riviere Madame.
Apart from the H6tel-de-Ville, there is no commanding
edifice of any kind, whether official or private, and the
shops that aspire to a degree of worldliness are few in
number. With scarcely an exception, the streets of the
city are narrow and have the restricted sidewalks that
belong to most tropical cities of this class. Each has its
own surface-water, serving as a store to those needing it
and as an expurgator of accumulated and accumulating
filth.

The houses are chiefly of rubble and plaster or stucco,
with pitched roofs, and the greater number are of two or
three stories. There are few among them that can lay
claim to architectural effect, and they lack wholly the
attractive features that belong to Spanish and Mexican
houses. On the surrounding heights, where many of the
wealthier people reside and enjoy fresh air, there are
residences of finer pretence, and some of these are charm-
ingly inviting, in their garden approaches. The focus of
social life of the city is the savane, with its bordering allees,
and the great expanse of unadorned grass. Here, late in

6 IMPEESSIONS OF MAETINIQUE

the afternoon of almost every day, may be seen what there
is of the fashion and wealth of the city, the little gather-
ings of French men and women, their promenades, saluta-
tions and dress, recalling in miniature the life of Europe.
Necessarily, these gatherings are only of nutshell dimen-
sions, and however they may partake of the atmosphere of
true France, they give one only the feeling of being exiled,
for the life that surrounds is foreign in every way.

Four-fifths, or more, of those whom one sees are yellow
or black in color — mixed Creoles, mulattoes, negroes, and
coolies — the true Martiniquians, if one chooses to call them
such. Except about the hotels and as representatives of
the government, army and navy, white men are in evi-
dence merely as points of reference. Nearly all the
municipal offices, from the lowest to the highest, are filled
by representatives of the colored or black race, and the
same holds measurably true of the offices held under the
rule of the government. One of the two regular jour-
nals of the city, La Colonie, is edited and published by a
man of color ; the librarian of the Bibliotheque Schoelcher
is likewise colored. The condition existing at Saint Pierre
at the time of its destruction was different. It was the
city, par excellence, and it housed the wealth and aristocracy
of the island.

Hearn, in one of his brilliant color pictures of the
people, characterizes them as being a population of the
" Arabian Nights/' many colored, but with yellow as the
dominating tint. He invests them with a glory that is not
at all times theirs, but on the whole they are kindly in

IMPRESSIONS OF MARTINIQUE

spirit, the women, more particularly, graceful and dignified
in bearing, and both sexes sufficiently alive to the recogni-
tion of their worth. The men do not differ radically from
other negro and mulatto types that are distributed through-

'{■fir

/Bft mBc

Bk
■J 1

IN THE SAVANE OF FORT-DE-FRANCE

out the south, except that they are softer in character and
more gentle in their ways, an inheritance, doubtless, of
French associations. It is different with the women, who
appear immediately as a race apart. Of unusual height,

8 IMPEESSIONS OF MAKTIKEQUE

supple and straight as their royal palms, these proud
products of Martiniquian soil at once arrest attention ; and
while one could readily challenge the contention that they
are the "fairest of the fair," it may be admitted that some
of their types are imperiously attractive, and that a voice
more beautiful than theirs or one better qualified to charm,
cannot be found as a quality belonging to any other
race.

In striking contrast to the degree of unattractiveness of
its capital city, is the island of Martinique itself. Situated
in a quarter of the globe where nature knows no limit to
her work, and where the tares and stubble of erratic
growth have not yet developed sufficiently to deface, it comes
to the eye, save where desolating death has latterly laid its
hand, a picture of charming loveliness — peaceful but ex-
uberant. Its gently swelling outline does not remind one
of the crags and cliffs of Capri, of Ischia and of other
Mediterranean islands ; nor do its heights recall the nearer
•mountains of Cuba, Jamaica or Porto Rico. The landscape
is that of the Lesser Antilles, diversified in its own way,
and breathing its own atmosphere. Dominica, near to it,
has perhaps most of its fine nature, and St. Kitts sur-
passes in quiet repose ; but the unfortunate French island,
now writhing in the coils of the dragon that wrought its
earlier fabric, has a charm of its own, which its neighbors
have failed to cultivate, or which, with them, has, perhaps,
already ceased to exist.

It may not be difficult to find islands that are more
beautiful, more winsome, than Martinique, but it is less

Underwood & Underwood, Stereos. Photo., New York, Copyright, 190a

STREET-SWEEPERS OF FORT-DE-FRANCE
The Savane

IMPEESSIONS OF MAETINIQUE 9

easy to find one that is quite its equal. It has the softest
of summer zephyrs blowing across its fields and hillsides ;
swift and tumbling waters break through forest and plain ;
and mountain heights rise to where they can gather the
island's mists to their crowns. There are pretty thatched
cottages, nestling in the shade of the cocoanut, mango and
bread-fruit, and decked out with bright hibiscus and Bou-
gainvillea ; and fields of tobacco and patches of coffee and
cacao, added to bright cane, tell of a degree of prosperity
that most of the other islands do not have.

Seen from the sea, the island rises up into a series of
bold or even rugged prominences, with hanging slopes of
beautiful woodland, and fields of sugar-cane running into
their midst. The lesser heights swell up like huge camel-
humps from the confused landscape, giving a charming
background to the village sites that lie about them. During
the middle hours of day obscuring clouds generally hang
over the mountains, but in the early morning the summit
cap of Pelee, the loftiest eminence of the land, can generally
be seen dominating the landscape.

Until within the last few years the forest-primeval
clothed the mountain slopes from base to summit, but to-day
little remains of the true grands bois. A. woodland of ex-
quisite luxuriance, and showing the distinctive features of
a tropical vegetation, may still be seen and felt along the
deep waterways of the interior ; but the hand of man has
been steadily wiping out the glories of wild nature, to put
in their place the more humble picture of cultivation.
Fields of brilliant cane lie in the south, in the east and in

10 IMPEESSIONS OF MAETINIQUE

the north, and from their product the island returns most
of what wealth it has to its inhabitants ; and still humbler
plantations of cassava, bread-fruit and banana surround the
domestic cottage.

Martinique is the second in size of the group of beauti-
ful islands known as the Caribbees, and lies four hundred
and ten miles due north of the main mouth of the Orinoco
Kiver. Its softly rugged heights, and somewhat loftier
elevations, the mornes, rise from an almost immediate depth
of water of four to six thousand feet, and have for their
nearest neighbors Dominica on the north, and St. Lucia on
the south, each separated off by a billowy sea of twenty to
twenty-five miles. Nearly the whole of the island, except
where in local patches the coral-animal has built up its
reefs, is of volcanic origin — the soil, the hills, the stream-
boulders all bearing testimony to the action of volcanic
forces which were in operation thousands of years ago.
We possess no positive information of any eruption having
disturbed its surface prior to 1792, when, in the month of
January, a feeble activity, comparable to that of August,
1851, gave indication of the life that still rested within.
The present active point of the island is Mont Pelee, a
mountain of only Vesuvian proportions", whose broad foot
defines nearly the whole north shore.

Rising to four thousand two hundred feet, or somewhat
higher, its summit dominates the whole island, save where
the line of sight is cut by the bold and hardly less signifi-
cant peaks or Pitons of Carbet — ancient volcanic knobs
three thousand nine hundred and sixty feet in elevation —

IMPEESSIONS OF MAETINIQUE 11

that lie north of Fort-de-France. History records no ac-
tivity on the part of these mountains, nor from the still
conical Vauclin in the south.

Though so important among its neighbors, Martinique
is hardly more than a garden-spot, for it covers less than
four hundred square miles, and the greater part of it could
be packed into the area that is covered by the first city of
the United States. Lying well within the tropics, it has
all that a resourceful nature provides, and man has done
much — not too much, some will say — to improve what
nature has left undone. He has cut beautiful roadways
through meadowland and forest, around cultivated fields
and gardens and on the seashore cliffs high above the
surging waters. He has removed most of the forest, and
put in its place the cultivated field. Wherever we turn the
eye, it falls upon a peaceful living, and there is little to re-
mind one that man may be in want, and that the necessaries
of life are not justly distributed. But withal, the island is
not wholly a paradise, for it has had its earthquakes, its
cyclones, and its inundations ; and now must be added to its
unfortunate assets the most destructive volcanic outburst
that has ravaged any one region. The earthquake of 1839,
which wrecked one-half of the capital city, Fort-de-France,
and cost the lives of no less than four hundred people, is
still a part of modern history ; but the terrible cyclone of
August 18, 1891, which blotted forty hamlets from the map
of the island, lies much nearer to our own day. It is, in-
deed, remarkable, seeing how numerous in the past have
been earthquake disturbances of one kind or another, that

IMPEESSIONS OF MAKTINIQUE

the late volcanic cataclysm should have been so nearly free
of seismic movement of any kind.

In this island world of three hundred and eighty square
miles there lived before the eventful 8th of May one
hundred and ninety thousand people, or five hundred to

MARTINIQUE WOMAN

every square mile — about the same number to the square
mile as is found in England and Wales, and two and a
half times that in France. The number now living has
been lessened by about a sixth. Though not quite so de-
spairingly wrecked as some of its sister islands, Martinique

1MPEESSI0NS OF MAETINIQUE 13

shares in their decadent misfortunes. Capital is lacking for
new enterprises, and energy wherewith to obtain capital.
The production of sugar and rum, with its small mar-
gin for profit in some parts, and the absolute loss en-
tailed in the cultivation of the cane elsewhere, remains the
chief industry of the island, and were it not for the ex-
treme fertility of the soil, and the fact that a small and in-
dependent living can still be made from patches of earth
that have not yet been bonded to sugar-estates, the land
would soon go impoverished in the way of the other beautiful
islands of the Lesser Antilles. As it is, despite its many
misfortunes and vicissitudes, Martinique remains a com-
parative garden-spot, and the eye falls with delight upon
the pieces of cultivation — of banana, bread-fruit, cocoanut,
cassava, and Carib cabbage — that lie about on the hill-
sides, in the hollows, and along the roadside, and give a
living to thousands who have no work beyond their garden
palings, and hardly more within them.

M. Bourgarel, in the Econofniste Europeen, notes that
of the area of the island now under cultivation — forty-
seven thousand hectares out of a total of ninety-eight thou-
sand— approximately twenty thousand are given over to
the cultivation of the cane, which is little more than it was
in 1867 (eighteen thousand five hundred and sixty-five).
From that year until 1886, when the sugar crisis materially
checked the prosperity of the island, the development of
the cane-growing industry was steady for nearly every year,
the hectareage finally reaching twenty-eight thousand four
hundred and fifty. At this time, therefore, compared with

14 IMPKESSIONS OF MAETINIQUE

what it was at its maximum, sixteen years ago, the industry
has fallen short by almost exactly thirty per cent. ; and
now, with the devastation that has taken place in the
northern section of the island, where are situated many of
the most thriving plantations and some of the largest
usines of the colony, and the added uncertainties of work
that necessarily follow such a storm, the product will be
reduced very much further. It may be that this condition
will in the end work to the advantage of the island, for
it is certain that it is capable of rising to other industries
that, in the present condition of the sugar problem, must
yield more largely in profit, and open the way to a material
progress which confinement to a single enterprise cannot
permit.

Martinique, though well supplied with excellent interior
roads, which place its different locations in easy union with
one another, is entirely lacking in the means of rapid com-
munication. Excepting the small private roads, that op-
erate individually in the different plantations, there is not a
line of running railroad, whether steam or electric, on the
entire island. Inland transportation and carriage are had
by means of an antiquated coach-service and by individual
porterage, both men and women being willing servants to
this form of labor. The heavy, lumbering ox-cart, with its
double-yoked team, is still a part of the scenery of the
Martinique roadway, and may remain such, so far as
present indications point, for some time still in the future.
The modernizing of the island, while it has brought with it
a certain number of " improvements' ' — the electric light,

IMPRESSIONS OF MARTINIQUE

telephone and telegraph — leaves many things still un-
touched, and fortunately among these, the desecration of
the landscape. This will continue charming, and with it
the soft atmosphere that gives it color.

Photo. Heilprir

THE FOREST SOLITUDE

SAINT PIERRE AND ITS RUINS

Lafcadio Hearn, in his work on the West Indies,
gives the following description of the city he knew so well :

" The quaintest, queerest, and the prettiest withal,
among West Indian cities ; all stone-built and stone-flagged,
with very narrow streets, wooden or zinc awnings, and
peaked roofs of red tile, pierced by gable dormers. Most
of the buildings are painted in a clear yellow tone, which
contrasts delightfully with the burning blue ribbon of
tropical sky above ; and no street is absolutely level ;
nearly all of them climb hills, descend into hollows, curve,
twist, describe sudden angles. There is everywhere a loud
murmur of running water, pouring through the deep gut-
ters contrived between the paved thoroughfare and the
absurd little sidewalks, varying in width from one to three
feet. The architecture is that of the seventeenth century,
and reminds one of the antiquated quarter of New Orleans.
All the tints, the forms, the vistas, would seem to have been
especially elected or designed for aquarelle studies. The
windows are frameless openings without glass ; some have
iron bars ; all have heavy wooden shutters with movable
slats, through which light and air can enter."

Saint Pierre, which at the time of its destruction was
the most important commercial town of the island of
Martinique, was also the earliest French settlement on the

SAINT PIERRE AND ITS RUINS 17

island, having been founded by Esnambuc as far back as
1635. It lay on an open roadstead, without harbor advan-
tages of any kind, and directly appressed to the southern
foot of Mont Pelee. Its position relative to the destroy-
ing volcano was very similar to that which Herculaneum
and Pompeii bore to the ancient Vesuvius. The early
establishment of the settlement, its beautiful position, and
the fact that it was the natural outlet to one of the rich-
est cane and cacao districts of the island, doubtless led
to its supremacy over every other location, and made the
absence of a harbor a matter of secondary importance. It
was the home of the bankers, merchants and shippers.
Many of the largest planters had seasonable homes here,
and had built beautiful villas along the height of Morne
d'Orange, the Reduits, and Trois Ponts. Out of a total
population for the city proper,1 as reported in the census of
1894, of nineteen thousand seven hundred and twenty-two,
probably not less than from five thousand to six thousand
were whites. Indeed, some who profess to have known the
city well, assert that the white population could not have
numbered less than eight thousand, — or more than is
contained in the capitals of most of the Lesser Antilles
collectively. Saint Pierre is described as having been a
city of gay and open life, and with a moral tone perhaps
considerably lower than that of most tropical cities. How-
ever this may be, it is certain that the city was the
attracting focus of the island, and to it gravitated all
classes of the island community, especially those who had
been favored by fortune's wheel. It is sometimes referred

SAINT PIEEEE AND ITS EUINS

to as the most beautiful city of the West Indies, but apart
from its charming location and the manner of its construc-
tion, in rising tiers lined to the surrounding heights, there
would seem to be little to justify this extreme idealization.

THE THEATRE— SAINT PIERRE

Although boasting of a number of stately, even imposing,
edifices, such as the cathedral, town-hall, military hospital,
club and theatre, and several attractive promenades and
squares, the city, as Hearn describes it, was in the main
old-fashioned, with narrow streets, stone and stucco houses

SAINT PIERRE AND ITS RUINS 19

of two and three stories, and steeply pitching roofs of red-
tiling. It was closely pressed together so as to keep out the
tropical heat, and had the benefit of abundant shade-trees
both in the jmblic ways and the numerous house gardens.
The streets were lit by electricity, as they are to-day in
Fort-de-Frame. On the heights outside of the main city,
especially along the valley of the romantic Roxelane, the
better-to-do had erected charming villas, and embellished
their sites with gardens of luxuriant vegetation. The
wrecks of some of these still remain, sufficiently to show
their attractive features. Saint Pierre was the educational
centre of the island, and its Lycee was diplomated with the
rank of similar institutions in France. One of the most
notable institutions of the city was the botanical garden,
near the foot of Mont Parnasse, which at one time had the
enviable reputation of being the most beautiful of all the
lesser botanical gardens of the tropics. Many of the
plants of tropical cultivation in the famous Jardin des
Plantes of Paris had been obtained from this garden. Of
late years, however, the Saint Pierre garden had been but
indifferently cared for, the arboretums had run to wild
jungle, exquisitely beautiful in the wealth and exuberance
of tropical vegetation, while the science of cultivation was
being but little attended to. The lovely waterfall remained
as the chief attraction to the people.

Along its ocean frontage, Saint Pierre had a length of
about two miles, extending from the Anse north of Carbet
to beyond the Roxelane River. Its parts were respectively
designated the Mouillage (towards the south), named from

20 SAINT PIEEEE AND ITS RUINS

the place of debarkation arid landing ; the Centre ; the Fort,
north of the Roxelane ; and the Trois Fonts, situated along
the latter river and east of the Centre. The Mouillage
was dominated by the abrupt height, constructed of ancient
lava or basalt, known as the Morne d'Orange, along whose
sea-face the road from Carbet descends.

The picturesque rock-bedded Roxelane, whose source is
in the southwestern slopes of the Pelee buttress, traversed
the city in its northern quarter, and was crossed by a num-
ber of bridges, two of which, both of them apparently firm,
still span the lower course. Above it, on hanging walls, as
it were, were located some of the most attractive villas of
the wealthier classes. Beyond the Riviere-des-Peres on
the north followed the suburb of Fonds-Core. The foci
of the active and social life of the city were the Mouillage
or landing, with its hundreds of casks of sugar and rum ;
the savane or city green ; the Place Bertin ; and the Rues
Victor Hugo and Bouille. A single line of cars helped
the city to rapid transit.

When I visited Saint Pierre towards the close of May
and in early June the weather was very hot. The sun
beat down with intense energy, and we wondered how the
city could have maintained its favor with the Martiniquians.
Situated on the leeward side of the mountains, the site
lacks wholly in the advantage that is offered by the trade-
winds to the locations on the east coast. There were also
few public gardens and breathing places, which must have
contributed much to the discomfort of the summer inhabi-
tants. It was this that made Morne Rouge, only four miles

ALONG THE ROXELANE-SAINT PIERRE

SAINT PIERRE AND ITS RUINS 21

distant, the resort of Saint Pierre. Charmingly located at
an elevation of fourteen hundred feet above the city, on a
ridge uniting Mont Pelee with the contreforts of the Pitons
de Carbet, and looking down over both the Atlantic and
Caribbean waters, it received the softening winds from the
east, and gave to its inhabitants in a tropical clime the
blessings of a temperate region. Morne Rouge is said to
have housed at times not less than from two thousand to
three thousand people coming from Saint Pierre.

On the evening of August 30, when Mont Pelee again
swept out its fiery tongue, and laid to waste one of the
most charming spots in the whole island of Martinique,
Morne Rouge met the fate that overtook Saint Pierre. The
city was wiped out, and the greater part of its population
annihilated. Besides the church, whose noble spire still
rises mockingly over the blighted landscape, only a few
houses remain ; gardens and woodland were swept out of
existence. In the place of all this is a desert — perhaps
more soft than that of Saint Pierre, but reading the same
history.

The traveller who to-day visits the site of Saint Pierre
sees hardly more than a mass of tumbled ruins. Where
before were the Rue Victor Hugo, with its rows of two-
and three-storied, pitched-roofed shops and residences, and
the Rue Bouille, are heaps of concrete and boulders, piled
three and five feet, and more. The Place Bertin is known
by what remains of its fountain, and by the prostrate trees
that have stretched themselves in parallel lines to the south.
Tier after tier of rubbled bulwark rises up to the surround-

22 SAINT PIEEEE AND ITS KU1NS

ing heights, but above, as well as below, there are only
ruined walls, with heaps of decay lying between them.
Not a roof remains to indicate that any habitation ever had
a cover ; not a chimney to recall the cheer and welcome of
the fireside. The eye follows long lines of half-standing
walls, more like the arches of ancient aqueducts than parts
of buildings, the greater number to-day running parallel
with the ocean front. There is little that rises above two
stories, and hardly anything to half that level. Flats of
ash rise up here and there to what may have been roof
corners, elsewhere the covering is so light that the old
paving-blocks come to the surface. At intervals bits of
polished mosaic paving appear through the ash, showing
where attractive house gardens had been located ; stone
garden-posts and flower-stands lie about, and with them
fragments of decorative railing. The old club bathing
establishment is still there with water in its basement, but
its broad flights of steps, with the great flower- vases stand-
ing on either side, lead only to heaps of broken stone and
mortar. We see the great palm that stood in the court of
the Saint Pierre Club, but only as a charred stump rising
from its garden of desolate debris. These and other land-
marks help to frame a picture of the city which seems
destined never again to rise from its ashes.

When I visited Saint Pierre on the 25th of May, five
days after the second great eruption, the color of life had
been entirely driven from it. Everything was gray or of
the color of baked and mudded earth, little different from
the stern landscape which adjoins on the north and north-

SAINT PIERRE AND ITS RUINS

east. There were no pinks, or yellows, or blues that give
the life to habitations in the tropics. Save for the small
ants that were already beginning to crawl about and recon-
struct for themselves new homes, the ruins gave out no

r JM

**'** ■d#^*^

RUE VICTOR HUGO— SAINT PIERRE

evidence of the living, whether of man, of beast, or bird.
An impressive silence, disturbed only by the human scaven-
gers who were prowling about for observation and study,
prevailed everywhere ; and not even the angry volcano to
the northeast, with its hurling clouds of mud and ash,

24 SAINT PIEEEE AND ITS RUINS

interfered with the general quiet of the scene. Compared
with Pompeii, Saint Pierre appeared ten times more ancient.
The green and fertile slopes of Campania, with their nest-
ling cottages and cultivated fields, are here wanting ; they
existed once, and not many days before, but they had
passed for the time. These make modern even an ancient
field. In Pompeii the eye has had restored to it the special
activities of man ; he reads the life of the household, hears
the clamor of the market-place, follows the debate in the
Forum, and gambles on the wheels of the chariots as they
whirl around the circus field. In Saint Pierre, for those
who have not known it before, there is nothing of this.
Though its walls are modern, though everything that per-
tains to their construction and everything that has been
found within is modern, the city itself looks as though it
had been deserted at a time when man was still prepared
to be a wanderer, long before the beautiful sculptures of
Pompeii had been carved, long before the paintings had
been put on walls to charm and adorn.

For two miles or more the ruins continue ; you know
the streets by their standing walls, you recognize some of
the houses by what the walls still carry. Here is the cor-
ner of the cathedral, there the municipal building, and
farther to one side the wall of the military hospital. Only
a few days before it still bore the clock, with the hands
marking eight2 minutes of eight, which told the precise
time at which the catastrophe took place.

We followed clumps of charred tree-trunks along what
was the ocean promenade, and from them passed to the

SAINT PIERRE AND ITS RUINS 25

square or Place Bertin, where, in the shade of its lofty
trees and around its attractive fountain, the populace met
for recreation and business. What is there to-day? Great
tree-trunks stretched in line, their branches buried in dust
and turned almost to coal, their roots pointing to the moun-
tain that brought such devastation.

We found twisted bars of iron, great masses of roof
sheathing wrapped like cloth about the posts upon which
they had been flung, and iron girders looped and festooned
as if they had been made of rope. We climbed over and
under ruins, over roofs and into cellars, and everywhere
was the same lifeless quiet. Great heaps of rubbish lay
on all sides of us, and on every side they bore evidence of
the terrible force that laid them low. We seemed to be
wandering through a city that had been blown from the
mouth of a cannon, and not one that had been destroyed
by any force of nature.

Yet stranger things were found here. We stumbled
upon little cups of china that were still perfect in all their
form, upon corked vessels in which water remained pure
and unchanged, and upon little packets of starch in which
the starch granules remained as when they were first put
in. It seemed remarkable that the great storm that had so
ruthlessly stamped out the life of man should have pro-
tected and left unharmed these little things that belonged
to his household. Here, in the chemist's shop, were some
of his things, untouched. Even from the spigot of the
street fountain cold water was still running, as it ran of
old. Here lay bundles of clay pipes, with the clay un-

26 SAINT PIEEKE AND ITS EUINS

burned, in nearly the same places where they had been
offered for sale across the counter. High up in the town
I found the sounding-board of a piano, with many of its
strings still tightly wound about their pegs.

All this seemed more like a dream than a reality. As
bits of beautiful mosaic paving came out of the ashes, we
asked ourselves, Are these never to be trod again ? Are
there to be no more flowers and plants in the gardens about
which bits of fence-railing remain ? Are the glad faces no
more to be seen of those who sat on the porches and veran-
das, where only broken columns now stand ?

We wandered sadly along. One of our party told us
that a group of bodies lay near. Yes, in the bath-room of
a private house lay six, burned in flesh until they were
hardly recognizable as bodies. A woman was stretched
on her back at the bottom of the bath-tub, with her left
arm thrown out as if to grasp something in her bitter
anguish. Near by was an infant, hardly too large to be
carried in the arms, and beside it the body of another
woman, crouched as if in agony and despair. To this
room probably all had retired, expecting a moment of
relief from the tornado of death that swept over them.
We came across another group, eight in number. They
told the same history as the first.

The thousands of bodies that lie here have been partly
burned, and nearly all are buried — buried by the con-
tinuing fall of ashes from the volcano. It is a strange
fate that the mountain whose eruption cost the lives of so
many should also give to them their natural burial. It

SAINT PIERRE AND ITS RUINS 27

continues in its work of activity as if nothing had hap-
pened, mocking the beautiful world that surrounds it.
Miles high into the air it is still puffing its steam and
ashes, and from its interior still issues that deep thunder that
more than once before gave warning which was not heeded.

What to many must appear most singular in connec-
tion with the terrible catastrophe of May 8, is that the
stroke of death followed a course that left little behind to
tell its own history. The student of geology wanders
among the ruins of a former prosperous city, and seeks in
vain for those signs of volcanic and seismic activity which
are and have always been associated with the destruction-
dealing powers of volcanoes. He searches in vain for the
rifts that may have tumbled the miles of buildings — in
vain for the lava-flows with which history has associated
Etna and Vesuvius. A force of men could almost dig out
this modern Pompeii in a day or two, so feeble in most
parts is the ash that has impounded the streets, so gently
soft the material that the great volcano has vomited out.
Yet on every side is the most hopeless wreck that can be
conceived of — a picture of absolute ruin and desolation that
has perhaps never before been witnessed. " Whence and
how ?" we ask ourselves, and the question still remains in
a measure unanswered, and may forever remain with only a
partial solution.

The aspect of the ruined city as I found it at the time
of my first visit differed considerably from that imme-
diately following the 8th of May, and had manifestly been
largely shaped by the eruption of May 20. After its

28 SAINT PIEEEE AND ITS KUINS

first destruction, although the extinguishment of life was
complete, rows of houses were left standing almost intact,
notably in the central quarters of the city. Photographs
taken several days after the catastrophe plainly show this
feature, as well as other features of equal significauce, and

Photo. Heilprin

THE CATHEDRAL OF SAINT PIERRE— AUGUST 23, 1902

permit us to make an interesting comparison and study
of the results determined by the two eruptions. Many
roofs were still in position, the massive building of the
mayoralty carried its overhanging cornice, and the Hopi-
tal Militaire its walled (now historic) clock. Many signs
remained on the buildings, and there were other evidences
of an only recently passed activity. At the later day, all

SAINT PIERRE AND ITS RUINS 29

this had changed. The second blast, that in intensity was
nearly, if not fully, the equal of the first, laid to ground
what still remained high, and gave to the city that dis-
tinctive oriented aspect which it now presents. The greater
number of the massive walls run parallel with the sea, or
in line to the volcano ; and there are few that have been
preserved in their full height that take a direction at right
angles to this. It would thus seem that the destroying
force of the eruption of May 20 expended its main
energy along a north and south line, shattering everything
that was more directly opposed to its course. This was not
BO markedly the case on May 8, when much of the force
was directed radially. It is easy, however, to exaggerate
the importance of the testimony carried by this alignment
of walls ; a bird's-eye view of the ruins, like that obtained
from Morne d'Orange, shows a far greater number of the
transverse walls standing and more regularity in the streets
than appear to the eye of the stranger wandering among
the debris. The city, in fact, is clearly outlined in its
north-to-south and east-to-west streets.

The force of the destroying power was stupendous, and
wrought a ruin the like of which is paralleled only in the
path of a violent tornado. The most massive machinery
was bent, torn and shattered ; house-fronts, three and four
feet thick, crumbled and were blown out as if constructed
only of cards. The great cathedral bell lay buried beneath
the framework of iron which had supported it, tossed from
the church to whose chimes it had so long added its sweet
music.

80 SAINT PIEEEE AND ITS KUINS

Our examination of the ruins showed plainly, what
indeed had already been noted before, that the destruction
was almost entirely superficial. The destroying agent swept
the surface, but left almost untouched that which was be-
neath or buried within it. There were no displacements
due to earthquake tremors, as, in fact, there were no earth-
quakes that could properly be called such. It was this
remarkable superficial current which left intact the contents
of safes and burial vaults, of material that had been placed
in subways, and permitted water that had been contained
in large stoppered vessels to remain unchanged. For days
after the eruption, cool water continued to flow from the
faucets of the basement wall of the H6tel-de-Ville and
from other fountain-heads and hydrants of the city ; and I
am assured by Signor Parravicino, Italian Consul at Barba-
dos, who early searched the ruins for a lost daughter,
that this condition already existed on the 10th of May.
Still eight days later, water was found issuing from a house-
pipe, cool and potable.

Except on the broad principle of a fortuitous happening,
it is difficult to account for the anomalous conduct of this de-
stroying blast, — its deadly stroke at one place and its avoid-
ance of action at another. Tree-trunks, though burnt and
bereft of all their appendages, were left standing in what must
necessarily have been the centre of the storm ; bunches of
clay pipes, exhibiting no traces of either burning or scorch-
ing, were left at many points where manifestly they had
been put on sale ; and packets of starch and cereals were
passed so as to leave their contents undisturbed. Some

SAINT PIERRE AND ITS RUINS 31

cases have been reported where objects had been fused in
their coverings, when the coverings themselves had re-
mained untouched. A correspondent who visited Saint
Pierre about ten days after its destruction speaks of finding
a bird, dead, but unchanged in its plumage, lying at the
bottom of a wooden cage, which still hung seaward from
the balcony of a shattered house ; and there seems to be
enough evidence to sustain the statement that alongside the
body of a charred man was found a box of matches, the
contents of which had escaped ignition. The wonder, in-
deed, is that with such peculiarities or vagaries in action,
the destruction of human life should have been so absolute.
Manifestly a number of causes, rather than a single one,
contributed to the general destruction.

It would be difficult to indicate any quarter of Saint
Pierre which suffered less than any other, unless, possibly,
it be a part of the city of the Fort. Here, although buried
beneath a roofing of ash, there is still a semblance of con-
tinuity maintained, and from a distance the aspect is that
of deserted walls built against a hillside. Although nearer
to the volcano than any other part of the city, it may still
be reasonably assumed that the tornadic draught had not
in this section developed to the extent that it did in the
south. On the other hand, the quantity of ash and mud
covering the ruins north of the Roxelane is far in excess of
what it is in the other quarters, and in some places rises
well up to and over what would be the roofs of the houses.
This is also true of the near section of the city lying con-
tiguous to the Roxelane on its south side. We were sur-

32 SAINT PIERRE AND ITS RUINS

prised to find, on the 25th, that the iron truss bridge across
this stream was standing, and I found it still firmly intact
at a later visit, on August 24, when, according to report,
it should have been long in ruins.

The destruction of Saint Pierre is such that the greater
number of the building-sites are unrecognizable even to
those who were most familiar with the city — or could be
located only after a careful and comparative study. Of all
the buildings destroyed the cathedral almost alone presents
an architectural front, the stone coursing being retained on
the front elevation, with the statuary niches, and parts of
their contained statues. The walls of the building were
the most massively constructed of all in Saint Pierre, and
permit us to understand the degree to which they have
been preserved. On the other hand, the wreck of the
building generally only emphasizes the strength of the
blast which swept it to its doom. A number of the more
prominent structures have been identified by their step-
approaches, which in most cases have remained intact.
This, with the cellar-ways, is all that remains of the
Theatre in the northern part of the city. It is almost idle
to speculate upon the number of ways in which the masonry
of Saint Pierre was shattered and thrown to the ground.
That the greater part of the destruction was the result of
a direct impact from the visiting shocks — annihilation in
the path of a tornadic current — cannot be questioned ; and
it is merely a point to what degree this annihilation had
been hastened or furthered by the action directly upon
mortar of intense heat, and of possible electric strokes.

-•>

SAINT PIERRE AND ITS EUINS 33

This is a consideration, however, that seems to have no
approach at this time.

The ash that in its entirety covers Saint Pierre is incon-
siderable, and the quantity in no way justifies the extrava-
gant accounts that have been published regarding it. Ex-
cept where helped by mud-flows, or where it has accumu-
lated in wind-drifts, or in wall-fans, it rarely exceeds three
or four feet; and over the greater part of the city its
measure, even after later falls, is hardly more than a foot
or two ; in many places it is much less. It is true that
rains have considerably lessened the quantity since the first
fall, but perhaps not to an extent more than has been com-
pensated for by subsequent discharges. Les Colonies* a
Saint Pierre journal, reported that already on May 2,
fifteen inches (forty centimetres) of ash covered the savane
of the city, but this is probably an accidental overstate-
ment of the quantity.

From such evidence as it was possible to obtain, I
should assume that the greater number of the bodies found
at Saint Pierre were destitute of clothing, which had
either been burnt off or swept off in the passage of the
tornadic blast. In a number of places, located nearer to
the margin of the field of destruction, as on the heights of
Trois Ponts, or those beyond towards Morne Rouge, and
again southward towards Carbet, many clothed bodies
were recovered ; and on some of these the clothing had
hardly, if at all, been disturbed. Even in the same wagon-
side the clothed and the unclothed were found associated.
The searching power and penetration of the death-dealing

34 SAINT PIEEEE AND ITS KUINS

agent are thus brought impressively home to us, and the
conditions give a clue as to what must have been its nature.
There is a fair agreement in the report that asserts that
in a large number of cases the bodies were found with the
head turned to the ground, and many had the hand placed
over the mouth and the nostrils. The latter condition is
certainly expressive of a desire to avoid a gaseous or heated
inhalation. The thrown condition of the body can reason-
ably be explained on the supposition that the people gen-
erally turned their backs, whether in flight or otherwise, to
the dragon of death that was pursuing them, and were
then prostrated forward by the sweep of the tornado.
Bodies were found in unusual numbers at the intersections
of streets, and particularly so in the Place de Mouillage,
where the people had gathered to seek spiritual shelter in
the shadow of the cathedral cross. Nearly all of the
bodies had at the time of our visit been removed through
burning, calcining, or otherwise, or been buried beneath
new deposits of volcanic ash.

Ill

THE CATACLYSM OF MAY 8

The cataclysm of May 8, 1902, by which a mountain,
hitherto obscure, was suddenly brought into fame, stands
unparalleled in the history of volcanoes for its appalling
nature and the conditions which surround its existence.
Nor, indeed, is there anything that is properly compara-
ble with it. Papandayang, in Java, in its great eruption
of 1772, is assumed to have wrecked forty villages or more ;
and Asamayama, in Japan, eleven years later, was perhaps
equally destructive.3 But the data associated with the his-
tories of these mountains are to an extent of questionable
authority, and leave much room for inquiry ; and in
neither case, while the evisceration of the earth was stu-
pendous, was there a material destruction of the type that
is reflected in the wrecking of Saint Pierre. The violent
eruption, in 1888, of Bandai-San, in Japan, whereby a
quarter of the summit of the volcano was swept avalanche-
like over a populous district, was thought to have been
responsible for the loss of several thousand lives ; but the
official surveys show the number of killed to have been
less than five hundred.

It is certain that the victims of the eruption, in August,
1883, of the minor volcano of Krakatao numbered up-
ward of thirty-six thousand. In this extraordinary cata-
clysm, whose far-reaching phenomena were noted and

36 THE CATACLYSM OF MAY 8

studied at more distantly removed points of the earth's
surface than the phenomena of any other eruption, the
explosive force was most prodigious, and the result of a
kind which even the scientific mind was slow to recognize.
An island annihilated, the report of the explosion trans-
mitted thousands of miles over the earth's surface, and
clouds of ash kept suspended for a year or more in the
upper zone of the atmosphere — these were some of the
features which impressed upon the geologist and physicist
for the first time the full immensity of the power that was
resident in the volcanic recesses of the globe. It has been
estimated that eighteen million cubic metres of earth ma-
terial were disengaged from the earth in the course of this
eruption. Much the greater part of the destruction of
human life was consequent to the washing of the adjacent
island-shores by rapidly-following " tidal" waves, whose
translation to distant parts of the globe was phenomenal.
The rise of this flood-water was in some parts over a
hundred feet.4

The volcanic event that probably to most minds will
first suggest a comparison with the catastrophe of Mont
Pelee is the destruction of Pompeii and Herculaneum by
Vesuvius, and certainly no other appeals so forcibly
through its tragic aspects and the relations which attach to
a civilized life. The physiographic construction of the
land and the position of the destroyed cities, moreover,
permit of a certain geographic parallel being established
between the two episodes. Pompeii was located one mile
farther from Vesuvius than Saint Pierre was from Mont

THE CATACLYSM OF MAY 8 37

Pelee, and both volcanoes, so far as can now be told, were
of almost exactly the same height. The luminous, but
necessarily brief, description of the events surrounding the
eruption of Vesuvius that is given to us by the younger
Pl.ny, which is the only reliable information that we
possess of this historic event, leaves the student of geology,
even with the testimony that is obtainable from the ruined
walls and their contents, still in doubt as to some of the
main features of the catastrophe. These, indeed, are so
obscure and are brought out with so many aspects in the
light of the events of Saint Pierre, that it has been thought
well to give them special consideration in a chapter devoted
to a comparison of the phenomena in the two cases.

The destruction of Saint Pierre came to the city not
unheralded. For days before, the volcano had been vio-
lently active, and the form of activity that it assumed was
of a kind that should have immediately suggested disaster.
Other volcanoes, like Vesuvius and Etna, have similar
paroxysms, and are not particularly feared ; but their his-
tories are long known, and their modern periods of inac-
tivity are brief compared with even the last phase of
inactivity on the part of Pelee. It was in early May and
late April a closed mountain, which suddenly broke from
its anchorage. Vast columns of steam and ash had been and
were being blown out, boiling mud was flowing from its
sides, and terrific rumblings came from its interior. Lurid
lights hung over the crown at night-time, and lightning
flashed in dazzling sheets through its cloud-world. What
further warnings could any volcano give? A blind but

38 THE CATACLYSM OF MAY 8

impressive belief that nature would not harm, joined to
appeals against common sense made by a few who thought
they knew best, held the population to its doom. Not
even the discovery of a newly-formed crater-cone, made
ten days before the eruption, seems to have in any way
counselled a fairer judgment of coming events.

Statements conflict as to the earliest time when Pelee
gave signs of a renewal of activity, but there is no question
that evidences of unrest, whether in light emissions of vapor
or in rumbling detonations, had been apparent to a few
several months in advance of the catastrophe. The earliest
authentic record that I have been able to find of an actual
observation is contained in the note-book of M. Louis des
Grottes, an accurate student of nature, who made the
ascent of the volcano to the Lac des Palmistes on March
23, and noted his observations with care and precision.
Looking down from the summit of the mountain, the
Morne de La Croix, a fairly clear view was obtained of the
basin of the Etang Sec, and it was plainly seen that it was
sending out vapors at several points. A strong and incom-
moding odor of sulphur was remarked by the observers
even at their elevated position.

Following the habit of those ascending the volcano to
inscribe impressions on the walls of the little chapel of
" Our Lady of the Lake" {Notre Dame de V Etang), which
stood beside the mountain tarn, the record was placed:
" Aujourd' hui, 28 Mars, le crater e de V Etang See est en
eruption" (this day, March 23, the crater of the Etang Sec
(dry tarn) is in eruption). M. des Grottes' note-book

THE CATACLYSM OF MAY 8 39

account of his excursion, a translation of which appears, by
permission, in another chapter, is particularly instructive,
since it gives the only clear statement, so far as I know, of
the surface conditions of the mountain at a very near period
preceding the eruption.

The wholly accordant observations made by MM.
Lalung and Roger Arnoux, members of the Astronomical
Society of France, residents in Saint Pierre, and communi-
cated by them to Camille Flammarion,5 make it practically
certain that the first true opening of the volcano was on
April 25. The crater, whose position in the basin of the
Etang Sec is clearly established by M. Arnoux, then sud-
denly broke into eruption, throwing out showers of rock-
material to heights of one thousand to thirteen hundred
feet above the mountain.

During the latter days of April, when, as appears from
the letter of Mrs. Prentiss, wife of the American Consul,
the fumes of sulphur were so strong that horses were falling
in the streets, and the day of the catastrophe there were the
usual alternations of manifestations which attend volcanic
outbreaks, with a rapid convergence to a climax. The
cataclysm had presented all its antecedent phases, and the
final stroke, when it came, although accomplishing its work
with unheard-of swiftness, was not that of a bojt from a
clear sky. At two minutes after eight o'clock, of the time
of Fort-de-France, the morning of the fatal May 8 saw a
destructive cloud issue from the fermenting volcano, sweep
with almost dazzling velocity to its lower slopes, and fall
upon Saint Pierre. The fiery messenger of death had done

40 THE CATACLYSM OF MAY 8

its work, and a sheet of rising flame told that the work was
complete.

Judge Pub. Co., New York Copyright

PELEE IN THE MAY ERUPTION

There are few among the living who were eye-witnesses
from first to last of the full phenomena that construct this
extraordinary cataclysm, or who were permitted to follow
the sequence of events with an intelligence that was not dis-

THE CATACLYSM OF MAY 8 41

turbed by incidents likely to affect the reason. The fright-
ful and wholly unprecedented nature of the happenings
have helped to obscure the facts, and to inject into them an
interpretation which is not permitted by a more rigid analy-
sis of the testimony that is presented. On the main points
of the tragedy, the testimony given by the officers of the
French cable-ship Pouyer-Quertier, which was at the time
of the disaster eight miles abreast of Saint Pierre, grap-
pling for one of the lost cables, appears to be the most
trustworthy ; and it is confirmed in its principal details by
the testimony of other observers, notably the late Cure of
Morne Rouge, Pere Mary, Monsieur Fernand Clerc and
MM. Arnoux and Celestin, members of the Astronomical
Society of France, whose points of observation were widely
separated from one another, and removed from threatening
danger at the time. The nature of this testimony is so
accordant, that it may be readily accepted as the foundation
upon which a scientific conclusion must be based.

At almost precisely two minutes after eight, of the time
of Fort-de-France, a working message was sent off from
the Pouyer-Quertier to the Martinique capital, but it brought
out no reply. This was the same minute of time in which
the final word was received at Fort-de-France from Saint
Pierre. It is manifest, therefore, that the difference of
local time between the two cities was ten minutes, the
Hopital Militaire regulating the time for Saint Pierre.
From the moment that the great black cloud issued from
the volcano it was followed by the officers of the Pouyer-
Quertier, who noted that its forepart became luminously

42 THE CATACLYSM OF MAY 8

brilliant as it approached the sea. In an instant after
everything was ablaze, and flames shot out from seemingly
all points of the city as if from a single brazier. Light
detonations, following one another in rapid succession and
coming from the direction of Saint Pierre, were a part of
the phenomena of the ignition, and it is safe to assume that
they marked passages in the exploding cloud.6 Only one
flash of lightning was noted, and that was thought by some
to traverse the cloud in a vertical direction from below up-
ward. No flame of any kind was observed previous to the
ignition of the city, nor was any fire-sheet seen to traverse
the air in advance of the descending cloud. The further
incidents of the cataclysm were unobservable, inasmuch as
the land was immediately veiled in an impenetrable cloud
of ash and smoke, and the Pouyer- Quertier, itself threatened
by showers of ashes and fiery cinders, was obliged to seek
safety in flight.

A mournful spectator of the tragedy that was being
enacted below was M, Roger Arnoux, a member of the
Astronomical Society of France, who from his commanding
position on the Mont Parnasse, removed awhile from danger,
calmly surveyed the most important field of the volcano's
activity. He, too, had noted the death-carrying black
cloud sweep like a serpent's tongue after its prey, and he
also observed its rolling motion. No trace of flame was
visible at any part of its course.

M. Roux's account of his observations, transmitted to
Camille Flammarion, is published in the Bulletin de la
Societe Astronomique de France (August, 1902), and pre-

THE CATACLYSM OF MAY 8 43

sents in very graphic form the terrible denouement which
he was forced to observe, and in which was involved the
loss of a father, mother, brother and sister. The account
is clearly that of one trained in observation, and it alone
presents in specific detail the course of the phenomena from
an early hour to its close.

" Having leff Saint Pierre," writes M. Roux, " at about
five o'clock in the evening (May 7), I was witness to the
following spectacle. Enormous rocks, being clearly dis-
tinguishable, were being projected from the crater to a
considerable elevation, so high, indeed, as to occupy about
a quarter of a minute in their flight, and describing an arc
that passed considerably beyond the Morne Lacroix, the
culminating point of the massif. About eight o'clock of
the same evening we recognized for the first time, playing
about the crater, fixed fires that burned with a brilliant
white flame. Shortly afterwards, several detonations, simi-
lar to those that had been heard at Saint Pierre, were noted
coming from the south, which confirmed me in my opinion
that there already existed a number of submarine craters
from which gases were being projected, to explode when
coming in contact with the air.

" Having retired for the night (May 7-8) at about nine
o'clock, I awoke shortly afterwards in the midst of a suffo-
cating heat and completely bathed in perspiration ; knowing
my nerves to be agitated, I concluded that it was only
uneasiness that troubled me, and again retired. I awoke
about eleven-thirty-five, having felt a trembling of the
earth, but no other person in my house being about, I

44 THE CATACLYSM OF MAY 8

thought that my nerves had possibly deceived me, and
again went to sleep, waking at half-past seven. My first
observation was of the crater, which I found sufficiently
calm, the vapors being chased swiftly under pressure of an
east wind. At about eight o'clock, when still watching the
crater, I noted a small cloud pass out, followed two seconds
afterwards by a considerable cloud, whose flight to the
Pointe du Carbet occupied less than three seconds, being at
the same time already in our zenith — thus showing that it
developed almost as rapidly in height as in length. The
vapors were in all regards identical with those which were
being ejected nearly all the time from the crater. They
were of a violet-gray color, and seemingly very dense, for,
although endowed with an almost inconceivably powerful
ascensive force, they retained to the zenith their rounded
summits. Innumerable electric scintillations played through
the chaos of vapors, at the same time that the ears were
deafened by a frightful fracas.

" I had at this time the impression that Saint Pierre
had been destroyed, and I wept over the loss of those whom
I had left the night before. As the monster seemed to near
us, my people, panic-stricken, ran to a neighboring hill that
dominated the house, begging me to do the same. At this
moment a terrible aspirating wind arose, tearing the leaves
from the trees and breaking the small branches, at the
same time offering strong resistance to us in our flight.
Hardly had we arrived at the summit of the hillock when
the sun was suddenly veiled, and in its place came an
almost complete blackness. Then only did we receive a

THE CATACLYSM OF MAY 8 45

fall of stones, the largest of which were about two centi-
metres of average diameter. At this time we observed over
Saint Pierre, and in the quarter which I could determine
to be the Mouillage, a column of fire, estimated to be four
hundred metres in height, which seemed to be animated
with a movement of rotation as well as with one of transla-
tion. This phenomenon lasted for two or three minutes,
and was followed by a shower of stones and of mud-rain,
which pressed the lower herbage to the soil and even some
of the smaller shrubs. This torrential rain lasted for about
a half an hour. . . . Relatively to a rain of fire, of which
much has been spoken, I observed nothing of such nature,
although we followed the phenomena in their entirety.,,

The intensity of this early eruption of Mont Pelee will
always be judged by the extent of the destruction that it
wrought — the wrecking to tumbled ruins of an entire city
of twenty-five thousand inhabitants, or more ; the annihila-
tion of some adjoining suburbs ; and the destruction of
eighteen vessels that were in the roadstead at that time.
One of these was the English cable-ship Grappler, and an-
other, the passenger and freight-steamer Roraima, which
had passed to its anchorage less than two hours before.
The loss of life can only be stated approximately, and the
figure given may fall two or three thousand wide of the
truth. The official census of January, 1894, gives for the
city of Saint Pierre a population of nineteen thousand seven
hundred and twenty-two ; and for the commune twenty-five
thousand three hundred and eighty-two. The later parish
registers place the population somewhat over twenty-seven

46 THE CATACLYSM OF MAY 8

thousand. With one or two exceptions all those who had re-
mained in Saint Pierre perished, but it is known, and placed
beyond question by the published statements contained in the
Saint Pierre journals, that hundreds had left the city prior
to the catastrophe, seeking safer quarters elsewhere.7 This
depletion of the city's population seems to have been more
than made good by numbers of refugees who had fled to
Saint Pierre for protection, and by an influx of people from
Fort-de-France and elsewhere, who had come to attend
special cathedral service on the day of the Ascension.
Assuming, then, the full population of Saint Pierre, one is
perhaps justified in accepting the belief of Vicar-General
Parel, expressed in a letter to the Bishop of the diocese,
that the full number of the dead could not well have been
less than thirty thousand. In this estimate, which some
profess to believe on seemingly not very good grounds to be
much too small, would be included the killed in the suburbs
and outskirts of Saint Pierre, and those on board the dif-
ferent craft that lay about in the roadstead. The annihilation
of so large a number of lives in a very few minutes — in not
more than three to five minutes for much the larger body
— renders impressively appalling the nature of this cata-
clysm, and suggests problems in geological dynamics that
have yet to be solved.

The area of actual destruction that was involved in the
immediate catastrophe was not very large, most of it being
contained in the sector that would be bounded by the lines
drawn from the crater of the volcano to the arise immedi-
ately north of Carbet and Sainte-Philomene, the whole

THE CATACLYSM OF MAY 8 47

being comprised in an area of about eight square miles.
Within this zone the destruction of life and habitation
was practically absolute. Immediately outside of it the
measure of life-destruction remained much the same, but
the mechanical force of the tornadic blast had been largely
spent, and it permitted habitations of nearly all kinds to
stand without disorganization. As a rule, the line of de-
marcation between the outer zone of the singed vegetation,
where there was little or no destruction beyond the tempo-
rary effacement of the vegetation, and the non-affected
region is sharply defined, and one that can be easily fol-
lowed, sweeping over highland and lowland alike, even from
a distance.

Where the course of the tornadic blast was thrown
across narrow but high-walled valleys, a " haven of refuge'
was sometimes found in the lee of the nearer or hangin
wall, the plane of destruction passing overhead and reach
ing the opposite side without descending. This conditio 1
is seen in one or more of the arises (bays) north of Cai oet ;
and in the later eruption of August 30 the same condition
was repeated, the destructive blast passing over Fonds
St. Denis and singeing the highland forest beyond. The
longest line of destruction on May 8 was from thr, crater
to the north point of Carbet, almost exactly seven and one-
half miles ; the storm-blast there passed into the sea, and
naturally we can only conjecture as to what would have
happened had the land projected farther to the westward.
The condition of the ruins in the southern part of Saint
Pierre gives no indication that the force of the blast had

48 THE CATACLYSM OF MAY 8

nearly spent itself at that point, or that it had even ma-
terially weakened.

A comparison of the energy that was expended in the
Pelee cataclysm with that of other eruptions of note is
hardly permitted by reason of the diversity of the condi-
tions which this comparison touches. The statement has,
indeed, been made that, apart from its destructive and
death-dealing quality, the eruption of May 8 was not of
great power or magnitude. This is judged by the fact
that the discharge of ashes was not, or did not seem to be,
notably large, that there was no lava-flow — indicating an
absence of elevatory power in the lifting or expanding force
— and that there were no earthquake disturbances of any
moment. The comparison is, however, an entirely gross
one, since it is made between conditions that are in no way
accordant with one another. The explosive force that so
thoroughly wrecked a compact city two miles in length, or
nurtured a tornadic current, with a sweeping velocity of
one to two miles a minute, to accomplish this work, must
have been prodigious ; and while we do not as yet fully
understand the nature of this destroying cyclone of wither-
ing heat and gas, and the precise manner in which it was
accomplished, it is easy to believe that had the explosive
force been directed in its work to the inner walls of a
closed volcano instead of to its outer surface, the catas-
trophic details of the eruption would have been very dif-
ferent from what they have in fact proved to be. It is
also true that the greatest cataclysmic eruptions have been
unattended with lava-flows, or they had them only of

THE CATACLYSM OF MAY 8

minor degree. Krakatao, Bandai-San, and Coseguina are
instances of this kind, and dispose of the notion that the
power of a volcano is measured by the elevatory force that
it possesses to raise lava.

It has been impossible so far to estimate, even for the
purposes of an argumentative comparison, the quantity of

N*\w Ir,

& .< i

Photo. Hellprin

IRON BRIDGE ACROSS THE ROXELANE— SAINT PIERRE

ash that was thrown out by Pelee in its great eruption.
The island of Martinique occupying a position in the direct
course of the trade (and anti-trade) winds, with no large
land-mass lying even remotely (except very distantly) on
either side, it may be inferred that most of the ash has been
lost on the surface of the open sea, carried out directly to a

50 THE CATACLYSM OF MAY 8

distance of perhaps several hundred miles. The bark
Beechwood, travelling from Salaverry to New York, has
noted in her log-book (under date of May 8) passing
through a cloud of volcanic ashes in latitude 13° 22' ;
longitude 49° 50' W. ; about six hundred and sixty miles
eastward of Martinique. This would seem at this time to
be the farthest distance from the island at which these
volcanic products were noted in any quantity ; but the
determination is not entirely free from doubt, since these
same ashes may in part be a residuary product from the
earlier eruption of the Soufriere in St. Vincent. The
greater portion of the Soufriere ashes of May 6 and 7,
measured by the quantity that fell over Barbados, appears
to have travelled with the anti-trade winds — or, at least,
against the trade-wind — and this was also the case with the
dense ash-cloud of Pelee which we observed on May 25.

It is unfortunate that little or no notice was taken in the
early days of the Martinique eruption of the " after-glows,"
which certainly must have existed, in order to obtain some
measure at least of the quantity of the finer ash that was
thrown into the higher regions of the atmosphere. The
projectile force of the May eruption is represented to have
been very great, carrying the ash-cloud several miles into
the air ; and, if so, the high distribution of the finer ash
must have been considerable. I am informed that at the
island of Saint Croix, two hundred and fifty miles distant
in a direct line, brilliant glows appeared almost immediately.
On my second return voyage from Martinique I observed
brilliant glows on September 9 in about latitude 26° 30'

THE CATACLYSM OF MAY 8 51

north ; on September 10, in latitude 30°, and on September
11, in latitude 33° 45', longitude 71° west. The last
position is about fourteen hundred miles north-northwest of
Mont Pelee. The evening following was cloudy and no
observation could be taken. There is no question that these
glows, which came up to their full intensity and magnificent
brilliancy about thirty to forty minutes after sunset, were
the culmination of the Antillean eruptions, and probably
of those that had taken place only a few days before, but
whether of Pelee (August 30) alone, or of Pelee and the
Soufriere (September 3-4) combined, cannot positively be
told. The latter condition seems more likely, as the great
ash-cloud of the Soufriere on this occasion took a northerly
direction, and swept completely over Martinique. On the
afternoon immediately preceding the evening eruption of
August 30, I estimated, roughly perhaps, the elevation of
the Pelee ash-cloud to have been between six and seven
miles, which is still considerably less than that of Krakatao
in 1883 ; but it is seemingly fully equal to the height of
any other volcanic cloud that has been carefully observed.
It was then flowing almost directly northward, or somewhat
east of northward, and towards the region where the after-
glows were subsequently observed. This certainly helps to
link the after-glows with this eruption. And yet it would
be impossible to affirm in the absence of earlier observations
that the glows may not have been in part an accompaniment
of the first eruption, left over, and slow in coming. The
Krakatao after-glows were very tardy in their appearance
in some places.8

52 THE CATACLYSM OF MAY 8

Statements vary, and will continue to vary, regarding
some of the phenomena that were developed coincidently with
the shooting out from the volcano of its destructive blast.
Pelee was almost immediately veiled in an impenetrable
mantle of ash, and the entire region was in obscuration,
which probably sufficiently explains the discrepancies that
appear in the statements of different observers. One of the
most interesting of the observations made is that relating to
the formation of a counter wind — one coming from the
direction opposite to that of the destroying tornado— a
phenomenon which had already been noted by the observers
of the Tarawera eruption, in New Zealand, in 1886. M.
Celestin, in his account of the Martinique disaster published
in the Bulletin of the Societe Astronomique de France
(August), describes this suddenly-appearing wind from the
south as a vent impetueux, une veritable bourrasque, before
which the trees were bowed to the ground ; and M. Roux,
evidently referring to the same wind, says that it tore the
leaves from the branches of the trees, and even broke the
smaller branches. On the morning of June 6, at the time
of the eruption of the great asji-cloud from Pelee, which
was travelling with intense velocity southward, I noted the
regular clouds of the atmosphere, in a much lower stratum,
flying swiftly towards the volcano. They seemed to be
pulled towards its active point. What the precise significance
of these counter currents may be, I do not profess to know,
but they may be bound up with a condition of atmospheric
rarefaction or vacuum formed in the immediate compass of
the volcano.

THE CATACLYSM OF MAY 8 53

Regarding the destroying blast itself, of which a fuller
consideration is given elsewhere, it can only be said in this
place that it was tornadic in the violence of its sweep, of an
intensely high degree of temperature, explosive in action,
and necessarily gaseous in construction. To what degree it
may have been charged with the earthy products of erup-
tion brought to a condition of incandescence cannot now
be determined, and probably never will be determined
with certainty, but it seems positive, from the statements of
Captain Freeman, of the Roddam, and Chief-Officer Scott,
of the less fortunate Roraima, that a rain of burning ashes
was an immediate accompaniment of the explosion, and
was perhaps directly responsible for the burning of most of
the shipping in the roadstead. We are told that the Rod-
dam " was covered from stem to stern with tons of powdered
lava, which retained its heat for hours after it had fallen.
In many cases it was practically incandescent, and to move
about the deck in this burning mass was not only difficult,
but absolutely perilous." In Captain Freeman's recital it
is said that a wall of fire swept over the town and bay,
striking the Roddam broadside, and with such force as to
nearly capsize her. A probably more correct interpretation
of this phenomenon would be that the swiftly-descending
volcanic cloud was surcharged with incandescent particles
(and burning flames of gas?) and thus gave the appearance
of a solid wall of fire.

Professor Hill, in his report to the National Geographic
Society,9 has already well stated that the cataclysm brought
no important change in the topography or contour of the

54 THE CATACLYSM OF MAY 8

island, even in the quarter in which the volcano is im-
planted. The bays, valleys, gorges and ridges remain prac-
tically the same, and the new features, minor to the land-
scape at large, are those which have been added through
the eruptive processes of the volcano — the cratera.1 cone, the
deposits of ash, cinders, etc. There has been no appre-
ciable rise in the general island-surface, and no subsidence
either. The volcano, except for the loss of a portion of its
culminating Morne, stands as it did with its full height.
Some of the ravines and gorges of Pelee have unquestion-
ably been deepened and widened, but no important new
forms of this structure have been noted. It may be that
along some parts of the western coast of the island there
have been " drops" in the ocean basin — one such has been
noted at the mouth of the Precheur River — but if subsi-
dences of this class at all, they are wholly of localized
extent and without special significance. The great abyss
that had been reported formed westward of Martinique on
the line of the Puerto-Plata cable has been shown by the
soundings of the French cable-steamer Pouyer-Quertier to
be non-existent. The severed ends of the cable were found
in depths closely corresponding with those that had been
previously established by the cable-steamer Seine, in 1896,
for the approximate positions from which the strands were
recovered. Admiral Gourdon, Commandant of the Naval
Force of the Atlantic, has favored me with a tracing of the
operations of the Pouyer-Quertier, made by Captain Thi-
rion, which is here reproduced, and also with a brief letter
addressed to him by the latter officer, in which the oppor-

THE CATACLYSM OF MAY 8

CABLE CHART OF THE " POUYER-QUERTIER'

56 THE CATACLYSM OF MAY 8

tunity is taken to deny the reported finding of a marked
oscillation in the ocean bottom. It states : " Contrary to
the false reports that have been circulated touching the enor-
mous differences of depth that were thought to have been
found along the line of the cable, the plan which I trans-
mit to you indicates that the depths have but little varied."
(Translation.)

The same letter, however, contains the interesting state-
ment that evidences of a sub-oceanic disturbance or boiling
{eboulements) are not wanting, especially indicated in a
zone of one thousand to one thousand and fourteen fathoms,
and it is thought that to it may be due the constant and
successive breakages of the cable.*

The remarkable condition in which one of the cable-
ends, coiled and knotted about a trunk or thick branch of a
tree, was brought up from a depth of six hundred fathoms,
and the fact that a buoy anchored in three hundred fathoms
was lost, seemingly sucked under, almost immediately after
it had been placed, go far to sustain this supposition. A
disturbance or ebullition along the sea-bottom is, indeed,
something that one could readily expect as a concomitant
of the Pelee eruption, and it ought not to surprise us if a
condition of active eruptions were at any time discovered to

* " Neanmoins, dans le quadrilatere qui figure sur le caique, et
qui represente la zone sillonnee par notre grapin pendant plusieurs
jours, nous sommes fort portes a croire que des eboulements se sont
produits, eboulements qui ont d'abord brisee notre cable, puis Font
enseveli, et nous mettent aujourd'hui dans 1' impossibility de le crocher,
dans une zone encore mal definie."

TIIK CATACLYSM OF MAY 8 57

exist in the greater shore depth about the island. Oceanic
disturbances of greater or less magnitude have been noted
to have taken place about seven miles westward of the island
of St. Lucia ; and sea-captains claim to have remarked a
material change in the course of the currents sweeping

RECOVERED CABLE STRAND ENVELOPING BRANCH OF TREE

along the west and north coasts of Martinique. Unfortu-
nately, the observations which record these assumed dis-
turbances still lack full confirmation. On the other hand,
the sinking of a portion or of several portions of the sea-
bottom adjacent to the northern parts of the island of St.
Vincent, incident to the eruption of the Soufriere, seems to

58 THE CATACLYSM OF MAY 8

be a well-established fact ; but even here the full extent of
the subsidence or subsidences remains unknown.

What particular relation the eruption of Pelee bears to a
condition of general catastrophism in the Caribbean region
is discussed in a later chapter ; here it can only be said
that it followed as a culmination to events which had been
marked by such important passages as the destruction, by
earthquake, of Chilpancingo, in southern Mexico, in Janu-
ary of the same year ; the destruction, also by earthquake,
of Quezaltenango, in Guatemala, on April 18 ; the minor
volcanic eruptions in Nicaragua and Costa Rica ; and the
immediately preceding eruption (May 6 and 7) of the Sou-
friere of St. Vincent. Some seismologists and vulcanolo-
gists have attempted to draw a parallel or correlation be-
tween the events of the western Mediterranean basin and
the somewhat similar ones — to which Vesuvius, the sub-
volcanic ebullitions along the coast of Spain, and the nu-
merous earthquakes in the Balkan Peninsula have given
expression — occurring in the east or true Mediterranean,
but it is plain to see that the broad range and indiscrimi-
nate distribution of manifestations of like kind that can be
brought into such a " time" correlation — as, lor example, the
strong earthquakes in Finland of April 10-11 ; the strong
earthquake at Lake Baikal, April 12 ; the eruptions of Re-
doubt and Illiamna, Alaska, in April-May ; and the earth-
quake of Shemaka, Caucasus, April 17 — destroy any value
that such a comparison might have, unless, indeed, it is
made for the purpose of demonstrating that earthquake and
volcanic phenomena the world over are on the ascendant,

THE CATACLYSM OF MAY 8 59

and that we have reached a particular moment in the
earth's history when the outer crust is being specially agi-
tated. There are no known facts in geology that can be
adduced in opposition to a demonstration of this kind, any
more than there are facts that might be said to directly sup-
port it. For the region about Pelee itself, however, it is
evident that a condition has developed which is new to its
modern history, and one that opens a serious consideration
of facts in the geology and geography of a large section of
the earth's surface which have hitherto almost escaped at-
tention.

IV
DAYS OF FEAR AND TREMBLING

At precisely two minutes after eight of the fatal May 8,
as marked by the time of the capital of Martinique, the
single word " allez" was sent over the wire from Saint Pierre
to Fort-de-France. It came as a request to finish a message
travelling in the opposite direction. This was the last com-
munication that was received by the outside world from the
ill-fated city previous to its destruction.

When that final word left Saint Pierre, it would appear
that there was no particularly disquieting circumstance to
presage impending disaster. The good-natured operator was
at his post, attending in the usual way to the business of his
office. Yet, for days before, enough had taken place to
make the less strong fear and tremble, and to cause many
anxious hours to those who could not be comforted by scien-
tific explanation or newspaper analysis. In any country
but Martinique the symptoms of uneasiness to which Mont
Pelee gave expression would have impressively counselled
flight ; but in this island of tropical dreams and sunshine
the warnings went for practically naught. A feeling of
strange security had impressed itself upon the people, for,
as appears from an announcement contained in Les Colonies,
the more important daily journal of Saint Pierre, a large
excursion had been planned as late as the 1st of May for
the summit of the mountain, to take place on the fourth of

DAYS OF FEAR AND TREMBLING 61

that month. What pathos is carried in the words : " If the
weather be fine, the excursionists will pass a day that will
long be kept in pleasant remembrance I" Only once before
in the lifetime of the oldest inhabitant of the island had the
volcano exhibited an uneasy temper, but it was recalled that
the eruption of 1851 had been without destructive character,
and with hardly enough life to it to cause discomfort even
to those approaching within close range of its fires. It was
thought reasonable, except by one or two, to whom volcanic
manifestations were more than passing shows, that this
eruption would be merely the echo of the one of the past,
and that no disastrous consequences need be feared. So
late as May 7, Les Colonies, which, for political reasons,
appears to have been particularly interested in holding the
inhabitants to their city, continued to scoff at those who
meditated flight. Earlier numbers of the same journal
describe the condition of panic which prevailed throughout
the many darkening days and nights of the city : men,
women and children moving and wailing, only to return in
most cases to their homes, to be lured again to a feeling of
fancied security. A wiser few had left for good, seeking
refuge in the quiet atmosphere of Morne Rouge, whose com-
manding heights, packed closely to the foot of Mont Pelee
itself, surveyed the beautiful roadstead and the intercepting
declivity that descends to the water's edge.

Only in the light of the later occurrences can one picture
the dreary forecast of what was then impending, the unusual
appearance of the city as it had already existed for many
days, and the higher resolve which prompted the inhabitants

62 DAYS OF FEAE AND TKEMBLING

to abide by the counsel of a few who undertook the work of
reassurance for the many. A city choked with sulphur, its
streets blocked with falling and fallen ash, and with a
burning and thundering volcano standing at its threshold
— this is the picture of Saint Pierre during the latter days
of April and early May, the city whose gayety had been
compared to that of Paris, and its life to that of Pome. In
the latter days of April, as is made known through a letter
written by the wife of the American Consul, Mrs. Prentiss,
the conditions then existing must have been all but unbear-
able, for as she writes : " The smell of sulphur is so strong
that horses on the street stop and snort, and some of them
drop in their harness and die from suffocation. Many of
the people are obliged to wear wet handkerchiefs to protect
them from the strong fumes of sulphur." The odor of
sulphurous gases had already been perceived three months
before, but seemingly not until April 23, when there was a
slight fall of cinders, did the volcano give external evidence
of an active existence. On that day a number of distinct
shocks were felt, causing the houses to rock and dishes to
fall from their shelves.

The student of a later day can plainly see that from
this time on to the fatal eighth, the succession of events
was rapidly hurrying to a climax. The activity of the
volcano was no longer localized, nor was it confined to one
form of demonstration. On April 25 smoke was noted
issuing from the summit vent; the crater had opened, and
a storm of rock and ashes was hurled into the air from the
ancient pot known as the Soufriere of the Etang Sec. A

DAYS OF FEAK AND TREMBLING 63

second eruption on the day following caused considerable
disquietude, and by this time the covering of ash was a part
of the landscape. This earliest incident of any importance
that is connected with Mont PeleVs reawakening presents
itself in an interesting form through the observations made
after a brief interval of Messrs. Boulin, Waddy, Decord,
Bouteuil, Ange and Berte, which are recorded in the issue
of Les Colonies under date of May 7. Ascending the
mountain (April 27) by way of the Petite Savane and
Morne Paillasse on a little-travelled and much overgrown
path, these investigators found to their surprise that the
normally dry bed of the Etang Sec or Soufriere, which
had remained all but peaceful during the eruption of 1851,
and whose most advanced claim to activity lay in the emis-
sion of sulphurous vapors, was now in a condition of fer-
ment. A sheet of water, estimated to measure roughly two
hundred metres (six hundred and fifty feet) in diameter,
occupied the centre of a hollow or basin (cuvette), which in
itself had a basal diameter of three hundred metres.
Along one side of this picturesque lakelet, which more
than once before had come into being to mock its own
name of Etang Sec, rose a diminutive mountlet, hardly
more than thirty feet high, whose summit threw out long
trains of steaming vapor. The travellers observed a bril-
liantly shimmering surface appear at times beneath the
crowning vapor, while an almost continuous fall of water
was cascaded into the surrounding and lower-lying lake.
This small volcanic cone, whose crateral diameter was
assumed to be approximately fifty feet (fifteen metres), had

64 DAYS OF FEAE AND TKEMBLLNG

not been noted before, nor is there any reference to it con-
tained in the much earlier descriptions of the volcano. It
thus becomes particularly interesting as helping to localize
the rift whence issued the destroying force of the fatal May
day, and one of the points of main weakness in the volcano.
When it was first observed the noise of boiling matter came
loudly from within.

The days of assumed security continued to come and go,
bringing anxiety to many minds, and a still sterner resolve
to others to resist to the end. Light falls of ashes which
to some must have called up visions of distant Pompeii and
its destroyer Vesuvius, began to fasten a wintry look upon
the streets, while distant rumblings followed ominously
close upon one another. Through the obscured daylight
the eye could still follow the course of the unchanged land-
scape, but the ear noted the fall of rushing waters. The
idyllic Roxelane, so dear to the youth who knew no other
water but its own, had risen to a wild torrent, and on the
other side of the plain of the northern city roared the
Riviere des Peres. There appeared to be nothing to give
to these streams their temper, for no marked eruption is
noted at this time, but the waters came impelled with a
wild fury, and spread wreckage along their course.

The quiet of May 1 was followed by a day that largely
changed the aspect of the country. In the columns of
Les Colonies may still be seen the announcement of the
excursion planned for the summit of Mont Pelee, but a
white coat of ashes had covered the streets already in the
early hours of morning. It was like real winter this time.

DAYS OF FEAR AND TREMBLING 65

The beautiful Jardin des Plantes, which had furnished
such rare treasures from the tropics to its parent in Paris,
lay buried with its palms, its ravenalas, rubber-trees, and
mangos, its giant cactuses and red hibiscus, beneath a cap
of gray and white — the same as the noble avenue of tropical
shade-trees on the Place Bertin. The heights above the
city were white-gray, and Grande Savane had several
inches of a3h lying over it. The country roads were
blocked and obliterated, and horses would neither work nor
travel. Birds fell in their noiseless flight, smothered by
the ash that surrounded them, or asphyxiated by poisonous
vapors or gases that were being poured into the atmosphere.

The following days, the 3d and 4th of May, could
hardly have been those of assurance to the inhabitants, for
the volcano continued to tremble and to roar, and to throw
its heated ashes over at least a part of the city.

" The rain of ashes never ceases," remarks Les Colonies
(May 3). " At about half-past nine the sun shone forth
timidly. The passing of carriages is no longer heard in
the streets. The wheels are muffled. The ancient trucks
creak languidly on their worn tires. Puffs of wind sweep
the ashes from the roofs and awnings, and blow them into
rooms of which the windows have imprudently been left
open. Shops which had their doors half-closed are now
barred up entirely. The following business houses are
closed to customers : the maisons Saint-Yves, Deplanche,
Doliret, Reynoird, Boissiere, Cdlestin, Constance Esope,
Boulange\ Guichard, Dupuis et Cie., Vinac, Andrieux,
Villemaint, Lejeune, Delsuc, Lalanne, M&louze, Lathifor-

66 DAYS OF FEAE AND TEEMBLING

diere, Crocquet, Bazar du Mobilier, Bazar Sans Rival, etc.,,
The same issue of this journal announces the postponement
of the excursion to Mount Pelee in the following words :
" L' excursion qui avait ete organisee pour demain matin
riaura pas lieu, le crater e etant absolument inaccessible.
Les personnes qui devaient y prendre part seront avisees
ulterieurement dujour ou cette excursion pourra etre reprise"

One can hardly picture at this time a scene of more
hopelessly impending ruin ; for what the volcano had thus
far spared, or seemed disposed to spare, the torrential waters
of the descending streams threatened to take to themselves.
The sea is described as having been " covered in patches
with dead birds. Many lie asphyxiated on the roads. The
cattle suffer greatly — asphyxiated by the dust of ashes. The
children of the planters wander aimlessly about the court-
yards with their little donkeys, like little human wrecks.
A group goes along hesitatingly down the Rue Victor
Hugo. They are no longer black, but white, and look as
if hoar-frost had fallen over them. . . . Desolation, aridity,
and eternal silence prevail in the countryside. Little birds
lie asphyxiated under the bushes, and in the meadows the
animals are restless, — bleating, neighing and bellowing
despairingly."

The Riviere Blanche, which flows off the southwest-
ward slope of Mont Pelee and discharges two miles north
of Saint Pierre, was one of the eaux bouillants or turbulent
waters, sweeping relentlessly to the sea. In part of its
valley was enacted, on May 5, the first chapter in the
tragedy of Mont Pelee. Near the mouth of this stream, in

DAYS OF FEAR AND TEEMBL1NG <i7

a tongue of flat-land that unites its bed with that of the
Riviere Seche, was located one of the largest and most
profitable sugar establishments of the island. The Usine
Guerin had stood as a type for what it represented through
long years of toil and conquest, and its tall chimney looked
proudly over the fields of cane that circled about it, the
grands bois of the mountain slope, and the blue waters of
the near-by sea. Few of the great chains of wheels were
longer running, for the Riviere Blanche had given warning,
and the warning was for once heeded. Had the language
of the river been entirely understood, thirty or more human
lives would have been saved from the destruction that so
swiftly overtook the establishment. Hardly had the mid-
day hour passed on that eventful 5th, when the gates of the
volcano were drawn, and a flood of boiling mud was sent
hurling down the mountain side to be flung from it into the
sea. In three minutes it had covered its three miles to the
ocean, and within that time had left nothing visible of the
Usine Guerin but its chimney — a post projecting from a
desert of black boiling and seething mud. In this way
Pelee began its work of death.

It was needless to ask whence came the mud ; it could
plainly be traced to the position of the Soufriere or 6tang
Sec.10 A care-worn observer was at this time following the
occurrence from the estate of Perrinelle. For days he had
been observing the volcano, turning a watchful eye to every
new phase of action that was presented. He felt within
himself how insecure was the ground that was trod in the
shadow of a burning volcano, and pointed out to his stu-

68 DAYS OF FEAE AND TKEMBLING

dents at the Lycee the menacing force that was always
present. Professor Landes, alone, of the Commission that
was subsequently appointed by the Governor of Martinique
to inquire into the condition of danger, seems to have fully
realized the geological relations then existing, and it was a
fatal moment when, contrary to his better judgment, he

OPPOSITE THE RIVIERE SECHE

united in the counsel which advised a peaceful abidance
with the events that might follow. From his position at Per-
rinelle, Professor Landes observed the torrential character
of the Riviere Blanche, which was hurling along blocks of
rock, estimated in some instances to weigh as much as fifty
tons ( ! ) ; and at the same time he noted a white seething

-i JO&l KXV" ANNMj

L£

".UMtRO lO CENTIMES.

MERCREDI 7 MAI 190?

E S

PAKAI

OLONIES

LICVI.V DBLt m\utim jn:

''' v ' '•

•

in a

.1 Mai

FERN4ND GMSRG

AUTOUR

i:\T\s:n:(iNii:

V 1

■«nar< «ur Ij lift* ii

rnaf cunlrf laqMik
iwaMcpar

0. 01 v \\

•

LES VOLCAXS

1 au 7.""'.' ''*! ; 'I!< '' '"Ju""' '''"

tttrr, too —

EJ ty***' <*«»«aa*BotaalM»,Oa art

■HI— !■■!

UNE INTERVIEW DE M.

LANOES

I ffaaaHrMOMdn «ns man* »rmtr»
pa.r <xmr U ».»» ok v-
•<(«•' i« ■»■ r » "" J" -
Ho„. at gaaaj Ikanl JJJJJ^J ' "•

■■■raaa, ,1 '

■

ga*4aai i" • ah aaat 1 taaartaM 4 I

\.\ \l<<

■ i t, e >...„ , „ ,,,.„' ...',r. ;.

•MfaaUai((ccaai<)f U liv«rt b^Mlni,

I — ~

FAC-SIMILE OF THE FIRST PAGE OF THE LAST ISSUE OF LES COLONIES (MAY 7)

DAYS OF FEAE AND TEEMBLING 69

mass discharge with express-train velocity from the position
of the Etang Sec, and sweep down the mountain in the
plain between the Rivieres Blanche and Seche. This was
the avalanche of boiling mud and water that fell upon the
Usine Guerin and annihilated it with its unfortunate in-
mates. There can hardly be a question that the explana-
tion of the occurrence as given by Professor Landes is the
correct one : the Etang Sec, filled with the product that was
discharged into it by the newly-formed vent, broke through
one of its sustaining walls, and emptied itself of its boiling
contents. This condition makes intelligible the enormous
quantity of mud that was precipitated at one time, the
thickness of which in some parts of its flow was probably
not less than one hundred to one hundred and fifty feet.
The coast-line between Sainte Philomene and Fonds-Core
was materially extended by its discharge, and is to-day un-
recognizable in its contours to those who knew the region
best.

To the greater fear that was brought to the inhabitants
by the volcano was now added that of a " tidal" wave.
For a short time, indeed, it looked as if the city were to be
swept by the sea, for the waters, following a long recession,
rose high upon the beach, and penetrated even to the Place
Bertin. When the great mud-flow of the Riviere Blanche,
shortly before half-after- twelve, local time, plunged into the
sea, the latter withdrew three hundred feet or more, per-
haps driven to this distance by the impounding force. A
yacht, the Precheur, was overturned at her anchorage five
hundred feet from the shore. The transgression of the

70 DAYS OF FEAK AND TKEMBLING

ocean was fortunately a quiet one, and it left the prome-
nade, the landing-place, and the central Place without in-
flicting serious damage. " A flood of humanity," remarks
Les Colonies, " poured up from the low point of the Mouil-
lage. It was a flight for safety, without knowing where to
turn. Shop-girls were fleeing with bundles, one with a
corset, another with a pair of boots that did not match ;
and all in burlesque attire which would have evoked
laughter had the panic not broken out at so tragic a
moment. The entire city is afoot. The shops and private
houses are closing. Every one is preparing to seek refuge
on the heights." At this time the roaring of the volcano
continued almost without intermission, relieved at intervals
by concussional shocks that told that something was doing.
Saint Pierre had been left in night darkness. For
many days the disturbed condition of the atmosphere had
interfered with its electric illumination, and it was largely
by the aid of brilliant flashes of lightning, which came
with almost blinding effect, that the terror-stricken inhabi-
tants were enabled to grope their way through the thicken-
ing streets — to inquire, to search, and to find not. Many
had by this time fled to the hills, and others had left the
city and island for stabler shores, where there were but
faint echoes of the terrible detonations that broke from the
mountain. On the day following the destruction of the
Usine Guerin, Pelee was shrouded in heavy cloud, and its
ashes and cinders fell over a wide country, extending from
Macouba, on the north coast, to Saint Pierre and beyond.
The vegetation of forest-land, savanna and plantation was

DAYS OF FEAR AND TREMBLING 71

burned, and the cane and cocoa-nut were bowed to mother-
earth under the load of ash and mud that had fallen. The
country had already before this come to wear a strangely
withered aspect, for much of that which was growing had
been stripped of its leaves and branches and otherwise de-
nuded. Some of the surface waters had disappeared,
whether sucked up by the volcano or not cannot be told,
and pieces of land been deserted by cattle and other animals
whose manner betrayed an anxiety of mind akin to that
which agitated man. During these many days the atmos-
phere had remained singularly impassive, the barometer at
Saint Pierre indicating at the noon hour a pressure of seven
hundred and sixty-one or seven hundred and sixty-two
millimetres, the fluctuation at this hour during many days
confining itself to hardly more than one millimetre.

An intelligent analysis of the situation prevents one
from understanding how with the conditions prevailing at
this time at Saint Pierre, with a roaring and erupting vol-
cano rising from its very foot, a placid attitude could have
been maintained that still counselled remaining, and scoffed
at the notion of a departure. Where on the island, the
inhabitants are asked editorially, could a more secure place
be found in the event of visitation by an earthquake ? The
earthquake, for which the poor people had trembled from
day to day, came not. In its place came that which was
wholly unexpected, and which, in fact, could not have been
foreseen. A commission appointed to investigate the con-
dition of the volcano reported that there was nothing in its
activity that warranted departure from the city. The posi-

72 DAYS OF FEAK AND TEEMBLING

tion of the craters and of the valleys opening on the sea
was such, they said, that the safety of Saint Pierre was
absolutely assured (" la position relative des crateres et des
vallees debouchant vers la mer permet d'affirmer que la se-
curite de Saint Pierre reste entidre").

This report was virtually, and perhaps willingly, en-
dorsed by the unfortunate Governor, who, lured to its creed,
embarked on that tour of personal examination to which he
and his wife both fell victims. A far keener foresight was
that of the captain of the Italian ship Orsolina, who on
that 7th of May, contrary to the protests of those whom he
was serving, and the threats of the customs officers, decided
peremptorily to sail out with his half cargo, and turn his
stern to Pelee. He knew what Vesuvius was, he said, but
he felt that Pelee was much that Vesuvius was not.

Att>*

THE LAST DAY OF SAINT PIERRE

Wednesday, May 7, opened one of the saddest and
most terrorizing of the many days that led up to the final
catastrophe.

Since four o'clock in the morning Pelee had been hoarse
with its roaring, and vivid lightning flashed through its
shattered clouds. Thunder rolled over its head, and lurid
lights played across its smoking column. Some say that at
this time it showed two fiery crater-mouths, which shone out
like fire-filled blast furnaces. The volcano seemed prepared
for a supreme effort. When daylight broke in through
the clouds and cast its softening rays over the roadstead,
another picture of horror rose to the eyes. The shimmer-
ing waters of the open sea were loaded with wreckage of all
kinds — islands of debris from field and forest and floating
fields of pumice and jetsam. As far as the eye could reach, it
saw but a field of desolation. This was the early awaken-
ing of the day before the end, and one can hardly picture a
more disheartening opening of a new day. For days the
strenuous editor of the provincial paper, Les Colonies, had
been admonishing his readers to pay little heed to the vol-
cano, to regard its work more in the light of a nature-study
than of something to be feared. One reads with a feeling
of gentle pity an article on volcanoes that is published in
the last issue of this journal. It is printed on the first

74 THE LAST DAY OF SAINT PIEEEE

page, and in the first column, and tells of the general phe-
nomena of vulcanisra. With a blind faith in the righteous-
ness of things, the same issue (May 7) publishes an inter-
view with Professor Landes, of the Lycee, in which that
unfortunate scientist is made to appear as saying that there
was not more to be feared at Saint Pierre from Mont Pelee
than there is at Naples from Vesuvius. One can hardly
credit this belief to a man of the scientific standing of
Professor Landes, and it is easily possible that the conclu-
sion that is inferentially drawn from the interview was con-
structed by the editor of the journal, and on perhaps justi-
fiable premises.

The following is the full text of the interview as it ap-
pears in the journal :

AN INTERVIEW WITH M. LANDES.

M. Landes, the distinguished professor of the Lycee, has been
pleased to grant us an interview yesterday, apropos of the volcanic
eruption of the Montagne Pelee and the phenomena which preceded
the catastrophe of the Usine Guerin.

The following is the result of our conversation.

On the morning of the 5th (May), M. Landes observed torrents
of smoke escaping from the summit portion of the mountain, from
the locality known as the Terre Fendue. He observed that the Eiviere
Blanche was periodically swelling, and that it was running with five
times the volume of water that the high floods normally furnish. It
was hurling along blocks of rock some of which must have weighed
fifty tons.

M. Landes was stationed at the habitation of Perrinelle and
searched at twelve-fifty for the Etang Sec ; he noted a whitish mass
descend the slope of the mountain with the swiftness of an express
train, and enter below the valley of the river, where it marked its

THE LAST DAY OF SAINT PIERRE 75

course with a thick cloud of white smoke. It was this mass of mud,
and not lava, which submerged the Usine.

Later on, at the foot of the Morne Lenard, it appeared to M.
Landes that there was a new branch and that it possibly threw out lava.

M. Landes holds that the phenomenon of Monday is unique in
the history of volcanoes. It is true, he tells us, that the mud lavas
develop with very great rapidity, but this catastrophe was determined
rather by an avalanche than by a flow of mud lava. The valley has
received the contents of the Etang Sec, whose dyke having broken,
permitted of the fall of the muddy waters from an altitude of seven
hundred metres. If, as a surprising fact, there is no trembling of the
surface under the influence of this enormous fall, it is simply because
the sea has acted as a buffer.

It follows from the observations of M. Landes that yesterday
morning (May 6) the central mouth of the volcano, situated over the
higher (summit) fissures vomited out more actively (though intermit-
tently) than ever pulvurulent yellow and black matter. It would be
advisable to leave the neighboring valleys and to locate rather on the
elevations in order to escape submergence by the mud lava, as was
the fate of Herculaneum and Pompeii. Vesuvius, adds M. Landes,
has made but few victims. Pompeii was vacated in time, and there
have been but few bodies found in the engulfed cities.

Conclusion : The Montagne Pelee presents no more danger to
the inhabitants of Saint Pierre than does Vesuvius to those of Naples.

An editorial note, which is less confident in its tone
than other notes that had previously been published, sup-
plements the interview with the following : " Nevertheless,
this morning, the mountain being uncovered, the Morne
Lacroix shows in its lower part, on the side of the Etang
Plein, a gash one hundred metres in length and forty
metres in height, making possible the fall of this promi-
nence, and with it the production of an earth tremor."

76 THE LAST DAY OF SAINT PIEEEE

The other events that are chronicled in this last issue
of the Saint Pierre paper throw a vivid light upon the con-
ditions prevailing in the surroundings, and still further
darken the mystery of the quiet resolve to abide by the
events that were rapidly hurrying to a climax. There
were floods and torrents of boulders, villages inundated
and annihilated, and the ocean rising and falling in un-
known swells. The brighter days of springtime were made
black with the falling ash, thunder and lightning held sway
over the mountain heights, and the air was no longer fit for
man to breathe. Yet even in this late day, with the city
in panic, and with the visions of destruction made real
through the happenings of many days, the editor of Les
Colonies asks its readers : Why this fright, and why pre-
paring for flight ? He asks this question at the end of a
brief editorial paragraph which succinctly portrays the
condition of panic then existing, and which is as follows :

THE PANIC AT SAINT PIERRE.

The exodus from Saint Pierre is steadily increasing. From
morning to evening and through the whole night one sees only hurry-
ing people, carrying packages, trunks, and children, and directing
their course to Fonds-Saint-Denis, Morne-d' Orange, Carbet, and else-
where. The steamers of the ComjDagnie Grirard are no longer
empty. To give an idea of this mad flight, we give the following
figures. The number of passengers which on the line of Fort-de-
France was ordinarily eighty a day, has risen since three days to
three hundred.

We confess that we cannot understand this panic. Where could
one be better than at Saint Pierre ? Do those who invade Fort-de-
France believe that they will be better off there than here should the

THE LAST DAY OF SAINT PIERRE 77

earth begin to quake? This is a foolish error against which the
populace should be warned.

We hope that the opinion expressed by M. Landes in the inter-
view which we published will reassure the most timid.

It is difficult to analyze or to understand the motive
that prompted the publication of this appeal. Was it really
given out as the expression of a personal conviction in the
security of the place? Or was it, perhaps, a pennant
thrown to the wind to assist in the election of a candidate
to the French Chamber of Deputies, whose battle was being
actively fought by the editor? The editor lies dead, and
there is no one to answer for him. The same number of
the journal contains the names of the members composing
the commission that had been appointed by the Governor
to report upon the Mont Pelee eruption. They are those
of Lieutenant-Colonel Gerbault, chief of artillery and presi-
dent of the commission ; M. Mirville, head chemist of the
colonial troops ; M. Leonce, assistant engineer of colonial
roads and bridges ; and MM. Doze and Landes, professors
of natural science at the Lyc£e of Saint Pierre. It is an-
nounced that the labors of the commission would be made
known to the public. There was, alas ! enough to report,
but no one to report it.

Of the condition of affairs about Saint Pierre at this
time Les Colonies prints the following paragraphs :

THE PRECHEUR RIVER.

The Precheur River overflowed its banks yesterday and the day
before, and has carried with it enormous masses of rock. A very
curious phenomenon was noted to take place at its mouth. Sound-

78 THE LAST DAY OF SAINT PIEEEE

ings made at this point yesterday indicate that a large excavation
(cavity) has been formed. The water which had hitherto at that
point a depth of one metre has now eight metres. The cause of this
excavation has not been ascertained.

THE RIVIERE DES PERES.

A similar condition, the result of a terrible overflow, is found at
the mouth of the Eiviere des Peres. Yesterday evening, at about
seven o'clock, the flood increased and was flowing with dark water,
which was thought to be a simple rise brought on by the rains.
Presently there came a torrent which swept with it great quantities
of bamboo, and later, trees and giant blocks of rocks, which are still
to be seen in the bed of the stream. The bridge of the estate of Per-
rinelle has disappeared, buried, as it were, beneath the boulders of
rock. If the walls of the property had not been fortunately strong
enough to resist the pressure, the stables would have been carried
away by the torrent. This first overflow lasted until about ten
o'clock, when it began to diminish, only to commence again at two
o'clock in the morning.

It is to be reported that at its discharge the water of the river is
engulfed in the enormous cavity which has been cut at this point,
and that it carries down with it all the vegetable and mineral debris
which it has swept up in its course. A little beyond, the current reap-
pears at the surface of the sea, still laden with this debris.

THE OVERFLOW OF THE ROXELANE.

The Eoxelane overflowed in its turn at about seven o'clock yester-
day evening. This sudden rise was due to the heavy fall of rain on
the surrounding heights. The river holds in suspension all the ash
that it has caught up, and is consequently of a dark color. Great
quantities of dead fish have been observed at its mouth.

AT BASSE-POINTE.

The river of Basse-Pointe has overflowed since yesterday and
flows with black water. It is reported — but we have no means of

THE LAST DAY OF SAINT PIERRE 79

confirming the report, as the telegraph wires are everywhere broken
— that several houses have been carried away by the waters.

AT LORRAIN.

The Oapot, whose waters have been slightly discolored, is now
flowing so muddy that the mouth of the river is full of dead fish.

Photo. HeOprta

BASSE-POINTE— MAY 30, 1902

About one hundred and fifty kilos of dead and torpid fish have hern
taken from the irrigating canal of Vive.

MUDDY RAINS.

Yesterday, throughout most of the day, there fell in the north
a fine blackish rain, which was so charged with ash as to make the
carrying of an umbrella a matter of discomfort.

80 THE LAST DAY OF SAINT PIEERE

A RESCUE.

A fisherman named Thomas assisted M. Renus in the rescue
which we reported yesterday, and which was of a particularly peril-
ous nature. The boat which contained MM. Dupuis-Nouille the
younger (fils), Louis Claude, Elysee Fleurisson and three other pas-
sengers, and was manned by M. Stephane Larade, was upset and
broken by the muddy torrent and the numerous tree-trunks that
were swept along with it.

THE DEAD.

Contrary to reports that had been circulated, the body of Mile.
Pauline Fleurisson has not yet been recovered. We have to report
among the dead two children of M. St.-Just Prosper, one still at the
breast and the other sick, who were in a boat near to that of M.
Eenus.

Following these news-notes is a brief list giving the names
of subscribers to a general relief fund, and the amount
of subscriptions that had up till then been made. A last
balance shows eight hundred and fifty-eight francs, fifty
centimes, to which 107.75 francs are now added, making
a total of 966.25 francs. On another page of this same
number of Les Colonies is a belated account of an ascent
of Mont Pelee made on Sunday, April 27, by MM. Boulin,
Waddy, Decord, Bouteuil, Ange, and Eugene Berte, which
shows in sufficiently plain language the critical condition
which had been reached by the volcano. Although this
account appears at so late a date, and is edited, it cannot be
assumed that it had intentionally been suppressed by the
editor, who had before this published many alarming re-
ports of occurrences that were taking place. It may be
that he attached little importance to the narrative, and

THE LAST DAY OF SAINT PIERRE 81

perhaps it had not before been submitted in official form.
The journal makes the announcement that the Day of
the Ascension being on the morrow, the stenographic
courses, as well as the adult course which was planned for
the following Friday, would be postponed until Thursday,
May 15. The editor then adds for his own paper :

" Our offices being closed to-morrow, the next issue will
appear on Friday."

Saint Pierre knew no further Friday, and even of the
Thursday it had but a few short hours. It knew not on
this day the fate that awaited it on the morrow, and it
clung to the hope that a good end would still come. The
city went to sleep hoping but fearing, fearing and not know-
ing ; and it was the last sleep, except that of eternal death,
which the city had.

The 8th of May brought little welcome to Saint
Pierre. PeleVs thunders had ceased for a while, but the
hope that this gave was only to the wakeful few, for already
at four o'clock, two hours before the shadows of night had
lifted, an ominous cloud could be seen flowing out to sea,
followed in its train by streaks of fiery cinders. At half-
past six the Roraima, her decks turned to hoary gray by
the ash that had fallen over them, came into port, taking
her place with the eighteen other good craft that at this
time lay in the roadstead. She anchored to her last
berth.

The sun had risen in its course perhaps twenty degrees
above the horizon when the roaring of the dark-shadowed
mountain began anew. Hundreds of agonized people had

82 THE LAST DAY OF SAINT PIEBRE

gathered to their devotions in the cathedral and the cathe-
dral square, this being the Day of the Ascension, but prob-
ably there were not many among them who did not feel
that the tide of the world had turned, for even through the
atmosphere of the sainted bells the fiery missiles were being
hurled to warn of destruction. The fate of the city and of
its inhabitants had already been sealed.

The big hand of the clock of the Hopital Militaire had
just reached the minute mark of seven-fifty when a great
brown cloud was seen to issue from the side of the volcano,
followed almost immediately by a cloud of vapory black-
ness, which separated from it, and took a course downward
to the sea. Deafening detonations from the interior pre-
ceded this appearance, and a lofty white pennant was seen
to rise from the summit of the volcano. With wild fury
the black cloud rolled down the mountain slope, pressing
closely the contours of the valley along which had pre-
viously swept the mud-flow that overwhelmed the Usine
Guerin, and spreading fan-like to the sea. In two minutes
or less it had reached the doomed city, a flash of blinding
intensity parted its coils, and Saint Pierre was ablaze. The
clock of the Hopital Militaire was halted at seven-fifty-two
— a historic time-mark among the ruins, the recorder of
one of the greatest catastrophic events that are written in
the history of the world.

Thus had Pelee done its work. The mountain that
only a few days before had been clothed with all but pri-
meval forest nearly to its summit crown, was largely a
desert waste, scarred with burned timber, gray with ash and

THE LAST DAY OF SAINT PIERRE 83

water, and bleeding with black mud. Its waters, charged
to many times their natural force by the volcano's steam-
cloud, had graven deep channels into its flanks, and were
pouring their rock debris into villages and across habitations
which the volcano itself had spared. Precheur lay beside
in avalanche of boulders, vainly searching for a part of
the beautiful meadow upon which formerly grazed its goats
and cattle. The church stands with its half torn away, but
that which remains is more than is left to most of the
houses. In Basse Pointe boulders of eight feet and more
lie about the rubbled walls, and over the bridge of the
Riviere Basse Pointe flows the turbulent mountain torrent
that before this had meekly followed a rivulet's bed.

In and about Saint Pierre the work of death and de-
struction was accomplished in a few minutes. Thirty
thousand bodies lay among the ruins to tell the story of
that terrible day, turned to brown and black crusts — some
showing signs of a momentary struggle, the greater number
without evidence of any kind to indicate that they had
stirred after the fiery blast had once struck them. In the
houseways and in the streets, it was the same reading of
the almost instantaneous death. The burning buildings,
we ire told by Captain Freeman, of the Roddam, stood out
from the surrounding darkness like black shadows. All
this time the mountain was roaring and shaking, and in the
intervals between these terrifying sounds could be heard the
cries of despair and agony from the thousands who were
perishing. A few living forms, lit up by the lurid light of
the conflagration, were distinguishable running distractedly

84 THE LAST DAY OF SAINT PIEEEE

about the beach, only to meet death awaiting them at every
turn. Day had suddenly turned into night, but this night
brought with it no calm.

The final details in the passing of Saint Pierre were the
torrential rain that followed closely upon the destruction
and the general conflagration which continued for several
days. At the end of this time the city was laid to smoul-
dering ruins, coated with ash-paste, and looking as if built
of adobe plaster. What had before been the vivid coloring
of houses of the tropics was now an ashen gray — the color
of earth, cold, bleak and burned. Centuries seemingly had
passed between yesterday and to-day.

VICAR.GENERAL PAREL'S CHRONICLE

It was my pleasure, when at Viv£, to meet M. Parel,
Vicar-General of Martinique, who at the time of the de-
struction of Saint Pierre was officiating in the place of the
then absent Bishop of the diocese. The day-by-day record
of events that were then transpiring, and which M. Parel
communicated to the Bishop, paints with deep emotion the
incidents of the appalling cataclysm, and furnishes some
of the most remarkable chapters written in the history of
any event. M. Parel has kindly placed at my service a
copy of his note-book entries, and given permission for their
translation and publication. They appear here in full.

" Fort-de-France, May, 1902.
" MONSEIGNEUR :

"Such a catastrophe as this is utterly unheard of; it
has no parallel in history. Yet despite the general con-
sternation that prevails, I shall send you a daily summary
of events. You are familiar with the configuration of the
mass of Mont Pelee. The mountain commands the entire
northern part of the island, enclosing numerous valleys at
its base, and is the source of many streams, here somewhat
inaccurately called rivers, which course in all directions
from Saint Pierre to Grande Anse. You are aware that
Morne Lacroix (thirteen hundred and fifty metres in alti-
tude) is its highest peak, plainly visible in clear weather

86 VICAK-GENEKAL PAEEL'S CHEONICLE

from Saint Pierre, and that at its base lies the old crater
known as Etang Sec — Dry Pond — in contradistinction to
another lake situated on the opposite slope, the waters of
which are always high.

" Friday, April 25.

" On Friday morning, April 25, although the weather
was very clear, the crest of the mountain was capped with
dazzling white vapor. As at six-thirty in the morning I
boarded the ship leaving Saint Pierre, where I had spent the
previous day, and set out for home, I had an opportunity
to admire the spectacle that presented itself. Despatches
announcing a volcanic eruption had preceded my arrival
at Fort-de-France. The occurrence excited everybody's
wonder. Excursionists immediately set out for the crater,
which for so many centuries had slumbered peacefully, and
had but once, in 1851, given signs of existence by a harm-
less rain of ashes which fell over night on Saint Pierre.
The Fathers of the College were not among the last to reach
the mountain. From the summit of Morne Lacroix, they
discovered that the Etang Sec, which inclines its basin-
shaped bowl towards Saint Pierre, was filling up with boiling
water and emitting a sulphurous smell.

" Friday, May 2.

"Eight days later the nature of the eruption had
changed. Instead of vapor the mountain was now vomit-
ing ashes. At six o'clock in the morning, I received
the following despatch from the Curate of Le Precheur:
' Serious volcanic eruption ; since morning we have been
under ashes ; we ask for prayers.'

VICAK-GENERAL PARELS CHRONICLE 87

"At half-past eleven the following night, the city of
Saint Pierre awoke to the noise of frightful detonations, and
to one of the most extraordinary spectacles of nature, — a
volcano in full eruption discharging an enormous column of
black smoke, traversed by flashes of lightning, and accom-
panied by ominous rumblings. A few moments later a rain
of ashes poured down upon the city, and also, though in
less degree, upon Fort-de-France and the remainder of the
island.

"May 3.

"At dawn on Saturday morning, the whole settlement
found ashes lying thick about it, penetrating even into the
houses. As another despatch, more alarming than that of
the previous day, had reached me from Le Precheur, I left
at eight o'clock for Saint Pierre. I found the city covered
with ashes as if with gray snow. Thick wreaths of black
smoke hurled themselves upward. At intervals of six hours
the cannonading of the mountain redoubled in intensity. In
a downpour of ashes, which spread a strong odor of sulphur,
I visited Sainte Philomene, Le Precheur and Morne Rouge,
the places nearest the volcano. The villages were filled
with country people fleeing from the hills to the coast. The
churches remained crowded ; the curates baptized, listened
to confession, and attempted to sustain the courage of the
terrified people. I endeavored to reassure the inhabitants.
In the afternoon there was a frightful panic in the midst of
the ceremonies at the Cathedral. With outstretched arms
the people besought the priests for absolution. The colleges,
the Lycee, the schools, were disbanded.

88 VICAE-GENEEAL PAEEL'S CHEONICLE

" May 4.
" On this day the wind changed and the rain of ashes
moved towards the north and poured down upon Ajoupa-
Bouillon, on Basse-Pointe, on Macouba and on Grande
Riviere. Saint Pierre breathed more freely for a moment.

" May 5.

" Since morning the Riviere Blanche, so called from the
milky iridescence of its waters, which had been for some
days rising in an alarming manner, suddenly became a
threatening, muddy torrent, whose turbulence attracted all.
At the same time, a column of vapor rolled down from the
valley in the flank of the crater. ' A new crater is forming,'
was the cry. No, it was an avalanche of black, smoking
mud vomited forth by the crater ; swelled by successive
discharges it became a rolling mountain, as yet unseen while
it tore its path through the deep gorge, but the moment . it
approached the delta in which was situated the Usine
Guerin, its approach was betrayed by a great roar and by
a column of vapor. Those who witnessed the spectacle
shouted impetuously, ' run for your life !' It was too late.
In one brief instant, the avalanche had engulfed the factory
and the villas of the proprietors and employes alike. Over
a radius of several hundred metres, and even over the
neighboring hills, spread incandescent mud, several metres
in thickness. M. Guerin fils, his wife, M. Duquesne, the
head overseer, and twenty-five employes or servants were
overwhelmed. The chimney of the factory, slightly bent,
bears solitary witness to the disaster. This was about noon.

"At the same instant, along the whole roadstead of

VICAK-GENERAL PARELS CHRONICLE 89

Saint Pierre, the sea receded as though affrighted. It left
the ship Girard, which plies between Fort-de-France and
Saint Pierre, high and dry. Then suddenly the ocean,
rising mountain high, rushed back, breaking over the Place
Bertin, and even over some of the principal streets, and
spreading alarm far and wide throughout the city. The
inhabitants fled for refuge to the heights. Twenty minutes
later calm reigned once more.

" When the news reached Fort-de-France, the Suchet
was instantly put into service by the Governor, who was
anxious to visit the scene of disaster. I attempted to
secure passage, but was courteously refused, as it was feared
that my presence might only increase the panic.

"Tuesday, May 6.

" I could not leave, therefore, before the departure of

the regular boat at eight o'clock on the following day.

Accompanied by the Abbe Le Breton, I went to the Riviere

Blanche. This stream, now a raging torrent, crashed along,

carrying with it broken rocks, trunks of trees, and smoking

mud. With its trail of smoke, it resembled a locomotive

plunging headlong into the sea. I observed the slopes of

the volcano covered with mud and rock and ploughed into

vertical gashes by the waters which poured from its mouth.

The two peaks encircling it formed a valley which collected

the waters, whence they dashed forth in zigzags, to form the

seething torrent before us.

« May 7.

" Since four o'clock in the morning, when I was awak-
ened in my room at the S&ninai re-College by loud detona-

90 VICAE-GENEEAL PAEEL'S CHEOMCLE

tions, I have been watching the most extraordinary pyro-
technic display : — at one moment a fiery crescent gliding
over the surface of the crater, at the next long, perpendicu-
lar gashes of flame piercing the column of smoke, and then
a fringe of fire, encircling the dense clouds rolling above
the furnace of the crater. Two glowing craters from which
fire issued, as if from blast furnaces, were visible during
half an hour, the one on the right a little above the other.

" I distinguished clearly four kinds of noises ; first, the
claps of thunder, which followed the lightning at intervals
of twenty seconds ; then the mighty muffled detonations of
the volcano, like the roaring of many cannon fired simulta-
neously ; third, the continuous rumbling of the crater,
which the inhabitants designated the ' roaring of the lion f
and then last, as though furnishing the bass for this
gloomy music, the deep noise of the swelling waters, of all
the torrents which take their source upon the mountain,
generated by an overflow such as had never yet been seen.
This immense rising of thirty streams at once, without one
drop of water having fallen on the seacoast, gives some
idea of the cataracts which must pour down upon the sum-
mit from the storm-clouds gathered around the crater.
When day lighted up the roadstead of Saint Pierre, a cry
of amazement arose. As far as the eye could reach, it was
covered with floating islets, spoils of the mountain, the
forests and the fields, with trunks of gigantic trees, pumice-
stone, wreckage of every sort, discharged by the overflow-
ing torrents.

" I was obliged to go to Sainte Philomene and to Le

VICAR-GENERAL PAREL'S CHRONICLE 91

Precheur to give to the curates of those two places, along
with my encouragement, the aid which I had promised
them for their parishioners. But there were no longer any
bridges or roads. Accompanied by Father Fuzier and
Father Ackermann, I made my way in a boat through the
dangerous wreckage, which rendered our passage slow and
difficult. The point of the Riviere Blanche, of Lamarre,
and of the Precheur disappeared in the sea through suc-
cessive erosions, and under the combined shock of the
waves and the furious torrents. All those diluvial waters,
black and laden with mud, in tumbling into the sea, instead
of covering it as in stormy days with a muddy coat, barely
tinged it with a light yellow streak, and then seemed to
engulf themselves with their banks as if they were molten
lead. Every incident of that sad vigil was extraordinary.
I found the two men worn out with fatigue, pale from
want of sleep — always in their church, busy in preparing
their people as though for a great sacrifice, but full of ardor
and of courage, and, under the very jaws of the volcano,
faithful to their trust. Half of their parishioners had fled
to Saint Pierre, where the barracks and the schools had
been put at their service by the Governor.

"As for myself, believing it my duty to return home for
Ascension Day, I resisted all persuasions to remain and
took the boat from Saint Pierre at half past two, promising
to return the following evening, or at the latest on Friday
morning. The boat was filled with people fleeing from
Saint Pierre. I stepped out of the row-boat which carried
me over from Le Precheur just in time to embark. Was

92 VICAR-GENEBAL PAREL'S CHRONICLE

my good angel guarding me ? Or would it not have been
better to die than to survive ?

" Thursday, May 8, Ascension Bay.

" This date should be written in blood. Towards four
o'clock in the morning a violent thunderstorm burst over
Fort-de-France. Towards eight o'clock the horizon on the
north and in the direction of the volcano was black as ink.
The clouds raced across the sky towards the northwest. The
sky grew darker and darker. Suddenly I heard a noise as
of hail falling upon the roof and on the leaves of the trees.
A great murmur arose in the city.

" At the church, where eight o'clock mass had begun, a
frightful panic took place. The priest alone remained. At
the same moment through the night which shut us in thun-
der pealed, pealed continuously, appallingly. The sea re-
ceded three times for a distance of several hundred metres.
The boat which was leaving for Saint Pierre returned
affrighted. I went out on my balcony to see what was hap-
pening, and I noticed it was being covered by a hail of
stones and ashes still hot. People stood petrified at their
doors, or rushed distractedly through the streets. All this
lasted for about a quarter of an hour, a quarter of an hour
of terror.

" But what was taking place at Saint Pierre ? No one
dared to think. . . . Communication by telephone had been
cut off abruptly in the middle of a word. Some asserted
that they saw above the mountains which separated us from
Saint Pierre a column of fire rising to the sky, and then

VICAR-GENERAL PAREL'S CHRONICLE 93

spreading in all directions. The most terrible anxiety filled
our hearts. At eleven o'clock the ship Le Marin set out
to reconnoitre, and was witness to the most appalling spec-
tacle imaginable. Saint Pierre was a vast brazier of fire.
The news which burst upon the city at about one o'clock
sounded like the funeral knell of Martinique and evoked an
indescribable cry of horror. I shall not attempt to depict
such scenes ; it requires the pen of a Dante, or the elo-
quence of a Jeremiah. I am told that a ship is about to
leave to collect the wounded. I am fortunate enough to
obtain passage in it with one of my vicars. The police and
the gendarmes cannot restrain the crowd which struggles to
embark. The expedition is composed of the Prosecutor of
the republic, of an officer, and of a platoon of marines. It
is impossible to believe in the reality of so terrible a disas-
ter. We cling to every theory that permits us to hope. At
least, we think, a large part of the population will have had
time to flee. When, at about three o'clock in the afternoon,
we round the last promontory which separates us from what
was once the magnificent panorama of Saint Pierre, we
suddenly perceive at the opposite extremity of the road-
stead the Riviere Blanche, with its crest of vapor, rushing
madly, as on the previous day, into the sea. Then a little
farther out blazes a great American packet, which arrived
on the scene just in time to be overwhelmed in the catastro-
phe. Nearer the shore two other ships are in flames. The
coast is strewn with wreckage, with the keels of the over-
turned boats, all that remains of the twenty or thirty ships
which lay at anchor here the day before. All along the

94 YICAE-GENEEAL PAEEL'S CHEONICLE

quays, for a distance of two hundred metres, piles of lum-
ber are burning. Here and there around the city, upon the
heights and as far as Fonds-Core, fires can be seen through
the smoke.

" But Saint Pierre, in the morning throbbing with life,
thronged with people, is no more. Its ruins stretch before
us, wrapped in their shroud of smoke and ashes, gloomy
and silent, a city of the dead. Our eyes seek out the in-
habitants fleeing distracted, or returning to look for the
dead. Nothing to be seen. No living soul appears in
this desert of desolation, encompassed by appalling silence.
When at last the cloud lifts, the mountain appears in the
background, its slopes, formerly so green, now clad in a
thick mantle of snow, resembling an Alpine landscape in
winter. Through the cloud of ashes and of smoke dif-
fused in the atmosphere, the sun breaks wan and dim, as
it is never seen in our skies, and throws over the whole
picture a sinister light, suggestive of a world beyond the
grave.

"With what profound emotion I raise my hand above
these thirty thousand souls so suddenly mowed down, buried
in this terrible tomb to sleep the sleep of eternity.

" Beloved and unfortunate victims ! Priests, old men
and women, sisters of charity, children, young girls, fallen
so tragically, we weep for you, we the unhappy survivors
of this desolation ; while you, purified by the particular
virtue and the exceptional merits of this horrible sacrifice,
have risen on this day of the triumph of your God to
triumph with Him and to receive from His own hand

VICAR-GENERAL PAREL'S CHRONICLE 95

the crown of glory. It is in this hope that we seek the
strength to survive you.

" In this desolation the troop of soldiers sent to the res-
cue could do nothing. We returned, utterly dispirited, to
Carbet. New sensations and indescribable scenes awaited us
there. Here, in a single house, are heaped up fifteen bodies.
In another spot are dying men, horribly burned. Women
and young girls, their flesh tumefied and falling into shreds,
die as they reach the ship. Fathers mourn their children,
wives their husbands. Many of these are returning from
the country, ignorant, as yet, of the horrible truth. We
wished to hide it from them, but they divined it. The cries
which ring out break the heart. Many lost their reason.
For four hours embarkation on a dismantled sea goes on
continually. The Suchet and the Pouyer-Quertier come to
our aid. We reached Fort-de-Franee at ten o'clock in the
evening.

" It is time to explain to you how the terrible catastrophe
occurred. This, however, is not quite so easy as you may
imagine; firstly, because none of those whom the scourge
-truck escaped to tell the tale, and secondly, because those
whom the scourge spared were doubtless too much over-
come by the scene which they had witnessed to agree en-
tirely in their descriptions. Here, however, is all that I
was able to ascertain as fact :

"Since early morning of this day, the 8th of May, tin
rumblings of the volcano grew more disquieting, the dis-
charges of ashes blacker and denser. The anxiety of the
people about the mountain and in the city, then in gala

96 VICAB-GENEKAL PAKEL'S CHKONICLE

state, increased from moment to moment. Suddenly at ten
minutes of eight, as the hospital clock — providentially pre-
served among the ruins, as if to mark for all time the in-
stant at which the justice of God was meted out — bears
witness, a tremendous detonation resounded throughout the
entire colony and an immense mass was seen bursting forth
from the crater and hurling itself upward with extraordi-
nary velocity. The black coils of the appalling column,
rent by electric discharges, unrolled, expanded and dissi-
pated, and, impelled by an invisible force, moved on to
discharge, at a distance, the incandescent matter contained
within them. But suddenly, from the midst of these dense
masses, a spout of fire detaches itself, beats down upon
Saint Pierre like a hurricane, and envelops the entire city,
its roadstead and suburbs, from the promontory of Carbet
to Morne Folie, near Le Precheur, as if with the meshes of
a horrible net. On the surface about the city it describes a
regular curve of from two to three kilometres. It is impos-
sible to give any idea of the atmospheric disturbances cre-
ated by this hurricane of fire. What did it contain ? Matter
in fusion ? Gas ? Boiling vapors ? All of these at once ?
God knows ! ' Everything went down before it/ said to
me one who witnessed the sight from a favorable po-
sition, 'and at the same instant, everything took fire.'
Deep night fell over the land, but it was immediately illu-
mined by the dread fires of this veritable hell. From the
grass of the meadows, from the crops of the countryside, to
the great trees, to the houses and buildings of the city and
its suburbs, to the very ships anchored in the roadstead,

VICAR-GENERAL PAREL'S CHRONICLE 97

over earth and over sea, raged one vast conflagration, con-
suming thirty thousand human lives. In this awful tu-
mult, how terrible must have been the moment of agony
of a whole people ! What pen can ever paint the lamenta-
tions which ascended at that moment from the heart of a
dying city to the bosom of a merciful God !

"While the whirlwind of fire shot out by the crater
moved towards the south and the west, increasing its
destructive force and spreading its ravages, another phe-
nomenon, worthy of notice, stopped it in its course. Two
powerful atmospheric currents, laden with rain, held in
reserve up to that moment by some unseen but providential
hand, suddenly moved from the southeast and from the
north, and precipitated themselves on both sides of the flam-
ing spot. Circumscribing it with a clearly defined line, they
cooled it to such a degree that we could see people about
the line of demarcation struck on one side by burning
missiles, while on the other side, and at a distance of only
a few feet, nothing fell but the rain of muddy ashes and
heated stones, which descended in all directions.

" Whatever natural explanation of these phenomena we
seek, we are always confronted by a combination of truly
mysterious circumstances. It is evident, however, that a
power capable of regulating the forces and laws of nature
presided over the cataclysm, and that after having for a
moment liberated the unrestrained force of evil, at the next
it commanded the homicidal cloud to cease its destruction.
1 So far shalt thou go, and no farther/ it said. ' Here shalt
thou break the tide of thy anger/

98 VICAK-GENEKAL PAEEL'S CHEONICLE

• " Friday, May 9.

" I have just sent two priests, Father Woetgli and Abbe

Auber, with the expedition to Saint Pierre to pronounce

absolution and sprinkle holy water over the bodies which

are already being buried or cremated. While I was thus

THROWN STATUE OF "OUR LADY OF THE WATCH"

engaged, the French mail-coach arrived, containing the
Abbe Duval, the Vicar-General of Guadeloupe, and Abbe
Amieux, Curate of the Cathedral of Basse-Terre, whom Mon-
seigneur Canappe, as soon as the disaster became known,
kindly sent to us, laden with the precious burden of his
condolence and sympathy for Martinique. I shall attempt,
Monseigneur, to draw up the balance-sheet of the disaster.

VICAE-GENEEAL PAEEL'S CHEONICLE 99

"According to the statistics of the parishes of Saint
Pierre, entered in the or do of the diocese, the city had a
{population of about twenty-seven thousand souls. Adding
to this number about two thousand refugees from the sur-
rounding communes who sought safety here, at least five
hundred sailors from the ships anchored in the roadstead,
and finally the thousand victims in the parishes of Carbet
and Le Precheur, we obtain a total of more than thirty
thousand dead. Taking into account the fact that a large
number of inhabitants, especially women, had for two or
three days prior to the disaster been leaving Saint Pierre,
I feel that my estimate of the dead of thirty thousand is
as nearly accurate as possible.

" It was not the will of God, Monseigneur, that the
Bishop of the diocese should be the principal victim. And
who, among us, does not thank God for your providential
departure ?

" What need is there to name among the victims of
this horrible sacrifice the chief of the colony, M. Mouttet,
his worthy companion, Colonel Gerbault, or the twenty-
four priests whose names you already know ; — eleven of
the secular clergy, thirteen reverend fathers of the order
of Saint Esprit ? What need is there to mark out for your
pity all that group of young vicars, of young but distin-
guished professors, Le Breton, Bertot, Anguetil ; the rev-
erend fathers, Le Galbo, Demaerel, Fuzier, Ackermann,
and that sainted company of seventy-one religious women,
twenty-eight sisters of Saint Paul de Chartres, thirty-one
sisters of Saint Joseph de Cluny, ten sisters of the Deliver-

100 VICAR-GENERAL PAEEL'S CHRONICLE

ance ? And how many more ! Of the many professors of the
Lycee but five remained ; of the colonial boarding-school,
the directress alone survived. All those who escaped death
happened, of course, to be away from Saint Pierre. Digni-
taries, magistrates, merchants, honorable and Christian
families all had fallen before the destructive scythe.

" I said before that the colleges, schools and pensionnats
were disbanded. There remained, however, the two orphan
asylums, the workshop and the asylum of Saint Anne.
Teachers and scholars alike were engulfed.

" This is the moral balance-sheet, one which can never
be sufficiently deplored.

" Saturday, May 10.

"As a result of the loss of the chief magistrate of the
colony, and of so many other civil and military officials,
the Government is in a state of disorganization. The head
of the Board of Health declares that there is not the least
danger in waiting until Monday, the 12th, to begin the
cremation of the bodies which lie buried beneath the rub-
bish. And in addition — who can believe it ? — preparations
for the elections on the morrow, at least in the district of
Fort-de-France, are going on ; that of Saint Pierre no
longer exists

" M. Lhuerre, the General Secretary, is by decree tem-
porarily filling the office of Governor. The gentlemen
from Guadeloupe and I, thanks to the kindness of the
provisional Governor, secure passage on the Suchet, which
sails for Saint Pierre to examine the vaults of the bank.

VICAR-GENERAL PAREL'S CHRONICLE 101

The commander and the officers of the Suchet welcome us
very politely.

" Off the coast of Saint Pierre, the hull of the Ameri-
can ship is still burning, emitting a strong smell of putre-
fying flesh. Armed with disinfectants, we disembark at
the Place Bertin, a short time since so full of life, and
walk over the wreckage. It is a huge mass of rubbish
heaped up in indescribable confusion. Here and there
tumefying bodies, horribly contorted, show signs of terrible
agony in their twisted and contracted limbs. Beneath a
tamarind-tree, whose branches could not protect him, lies
the body of an unfortunate man stretched on his back, his
head thrown down, his hand raised to heaven in supplica-
tion, his entrails bared to view, his limbs torn and
shrivelled. That gesture of supplication alone consoles
us for the heartrending picture. God was merciful to him.
May he rest in peace ! At my suggestion, a photograph
was taken of the body.

" We find difficulty in reaching the Cathedral, as it is
impossible to recognize the streets. The interiors of the
houses, some of whose walls still stand, are blazing and
smoking braziers. The heaps of stone, iron, lime, ashes
and rubbish of all sorts burn our feet. It is danger-
ous to touch the pieces of charred wall, which crumble at
the slightest pressure. One of the square towers of the
Cathedral, with its four bells, is still erect, cracked through-
out and quite unapproachable. The tower on the left fell
to the ground with its great bell. The statue of the Virgin
which decorated the facade appeared to lie intact among

102 VICAE-GENEEAL PAEEL'S CHEONICLE

the rubbish in front of the Cathedral. The walls of the
church, with the exception of some portions of the apse,
are no longer in existence. We forced our way into it
through the Rue du College, and found several bodies half
hidden under the ruins. Here, as everywhere, a large part
of the victims are buried under the heaps of rubbish.

" We could not reach the altar, whose tomb seems intact,
although hidden beneath the confused mass of stones and
ashes. I regret that the director of the mission could not
grant me the two men I asked for to assist me in making
excavations. But who would think that men, worse than
jackals, coming from no one knows where, would prey
upon the unhappy city and complete by pillage the work
of destruction begun by fire.

"What shall I tell you of the parsonage? All this
block of buildings is practically levelled, and beneath its
ruins are buried our dear brethren, on whom we cannot
even bestow the honors of sepulture. I entered the Epis-
copal building by the wall opening on the savane. I
could have left it by walking out over the houses of the
Rue Coraille. Some portions of the walls at either end of
your Episcopal Palace remain, Monseigneur ; the middle
portion is razed to the ground. A piece of the wall of the
chapel is still intact, also. The safe, with all that it con-
tained at the time of your departure, is charred. There
your three servants perished. I could not find them. The
trees of the plantation are torn, bent towards the south and
partly burned. On my return to the Place Bertin, I at-
tempted to distinguish the church of the Fort, but in vain.

VICAK-GEKEKAL PAEEL'S CHEONICLE 103

The Seminaire College is completely wiped out. I am
told that in the Centre it is impossible to distinguish the
spot where the church stood. Balance-sheet of our losses :
— Your Episcopal residence, your Cathedral, all the
churches of the city of Sainte Philomene and of Trois
Ponts, your magnificent Seminaire-College, the workshop,
the Orphan Asylum, the parsonages, all the coffers of the
factory and the Episcopate, the coffers of the ecclesiastical
pension lists, etc.

"After having secured the treasures of the bank, the
Bucket was commissioned to aid in the evacuation of Le
Precheur. In a heavy rain of volcanic ash, two hundred
agitated people embarked. Two boats filled with women
and children capsized at the foot of the ship. The sailors
of the Suchet, with heroic bravery, rescued all.

"May 11 and following days.

"It is impossible, Monseigneur, to describe all our
anxiety and perplexity, and our hardships. I shall resume
my account of the days following the catastrophe by
relating the principal facts.

" While fire was devastating Saint Pierre, Le Precheur
was deluged by water. At eight o'clock in the morning
the Precheur River overflowed the parsonage, the town
and the church, which are now buried under one or two
metres of sand. The Abbe Desprez saved the Holy
Sacraments, but could not perform the Ascension Day
ceremony. All of the parishioners who remained were
gathered together on the 12th, and he and the mayor were

104 YICAK-GENEKAL PAEEL'S CHKONICLE

the last to leave the place, which was no longer habi-
table.

" Monday, May 19.

" I send two priests daily, with a party of men engaged
in cremating the bodies, to say a blessing over these poor
remains, but for the past three days the mission has re-
turned without being able to disembark. The violence of
the volcano appears to be increasing in intensity, and the
mountain is vomiting out masses of ashes which cover the
colony. To-day the mission was able to land, but the
rally was sounded immediately. A severe eruption took
place.

" Here in Fort-de-France, twenty-five kilometres as the
bird flies from the crater, we are living in the midst of
ashes, and, I may add, in a continual state of excitement.
Basse-Pointe, after several reprisals, was finally inundated
by the waters of its river. Several houses were carried
away and there was one victim.

" All the bridges from Basse-Pointe to Grande-Riviere
are down. These places are completely deserted. There
is no one remaining at Grande-Riviere or Macouba. The
curates of both these towns are here. The curates of
Basse-Pointe and Ajoupa-Bouillon spend the night at
Grande-Anse, and return home every morning to hold
mass. There they remain the entire day to aid the few
who have not yet deserted their homes.

"As to Father Mary, he very courageously remains
practically alone in Morne Rouge, beneath the jaws of the
monster and under the guidance of Notre Dame de la

VICAE-GENEEAL PAEEL'S CHEONICLE 105

Deliverande. I wrote to congratulate him, but there was
no longer postal connection. Should he succumb, he will
only learn in heaven that we admire him. Fort-de-France,
as well as the whole southern portion of the colony, is full
of refugees. An attempt is being made to distribute them
equally in the different townships, but we are still providing
shelter for seven thousand.

" Tuesday, May 20.

" Another date for Martinique ! As on the preceding
days, I appointed two priests to go to Saint Pierre. Will
they finally be fortunate enough to recover the sacred
vessels of the different churches ? Alas ! behold what has
happened. At quarter past five, while I was dressing, I
suddenly heard two loud detonations of the volcano, deeper
and more prolonged, I believe, than any which have yet
been noticed. I called to Abbe Recourse, who, since he
gave up his home to a family of refugees, has had a room
below mine. ' The volcano is angry/ I said ; * something
is about to happen/ At the same instant, in the distance,
above the peaks of Carbet, in the direction of Mont Pelee,
I saw, darting from a dark spot in the sky, rolling flashes
of fire accompanied by the muffled and continuous rumble
of thunder.

" Then above the black spot I saw the first coils of the
terrible column rising upward. Again I called M. Re-
course, ' Come and see, come quick V Then together, and
not without fear, we watched the sight. The meteor rose,
mounting higher and higher into the sky, unfolding its
spirals, reaching incredible heights, and then advanced

10(5 VICAK-GENEKAL PAEEL'S CHEONICLE

towards us, spreading out on all sides, shrouding the loftier
points, unrolling and unrolling until it stood directly above
our heads. We felt that the last moment of Martinique
had arrived. What would come next? Were we to die
beneath the flames as had Saint Pierre, or to perish beneath
ashes as had Pompeii ? We were prepared. We con-
tinued to watch the vast cloud and its dense whirls, which
the rising sun bathed with red. I was on my knees before
the window, awaiting the will of God. Suddenly, just as
in a theatre the curtain is drawn across the stage, a vapor
cloud spread below the aerial cloud, and shut it out from
us entirely. But the city, the city which was scarce awake,
where was it ? First a deafening tumult and a distracted
rush for safety. No one remains ! I am mistaken. The
church was regarded by many as a place of refuge. The
crowd surged to the very altars, and in what costumes ! It
is only with the greatest difficulty that the two vicars,
selected to go with the mission to Saint Pierre, can continue
mass. The third vicar ordered the five or six thousand
people assembled to pray with arms crossed on their breasts.
The sight is indescribably touching. These are in very
truth scenes that accompany the destruction of the world.
A quarter of an hour at least has passed, passed in agony.
Then follows the hail of lava and ashes. As the first
stones fall, I look about for flames ; but I am soon reassured.
We were frightened, that was all ; besides that we could
make a fine collection of volcanic stones, some of them the
size of an egg. Nearer the volcano even much larger ones
were found.

VICAE-GENEEAL PAEEL'S CHEONICLE 107

" But though we were safe, what was the fate of the
neighboring parishes ? The Suchet set out instantly to re-
connoitre.

" Its report was as follows : — ' The phenomenon which
resulted in the destruction of Saint Pierre had been re-
enacted and in the identical localities. Whatever walls
still remained standing in the doomed city were again
swept by a whirlwind of fire. Not a stone remained on top
of another. Some houses within the circle marked out by
the first scourge were struck and raked by the flames.
There were no new victims. A tidal wave ravaged the
Grande Anse of Carbet and carried away some houses.
The people who had remained at Fonds-Saint-Denis, Car-
bet, and Morne Vert fled towards the south. The curates
have just arrived. I learn, too, that the brave Father
Mary has at last left Morne Rouge. He was the last to
depart, leading the band of gallant followers who remained
faithful to him. A severer overflow would have destroyed
Basse-Pointe, which was already abandoned. The exodus
is of the entire north of the island towards the south.'

" Wednesday, May 21.

" The consequences of this new disaster are incalculable.
Since yesterday, all the families who were beginning to
regain their confidence are plunged into the deepest des-
pondency. They are embarking by thousands for St.
Lucia, for Guadeloupe, Trinidad, France and for America !

" It is no longer the exodus of the north to the south,
but of all Martinique to foreign lands. Such, Monseigneur,

108 YICAE-GENEEAL PAEEL'S CHEONICLE

is the life which we lead. Whatever the reasons for which
Providence has willed that I should witness these events,
I can hut follow the example of Father Mary and of his
fellow priests of the northern parishes. I shall he the last
to leave Martinique.

"G. Parel."

SAINT PIERRE BURNING

AFTER THE CONFLAGRATION

Acting-Governor Lhuerre's official report of the

catastrophe of the 8th of May is as follows (translated

extract) :

"Fort-de-France, May 11, 1902.

" The night of the 7th-8th passed without incident ; the
official cablegrams which arrived from Saint Pierre between
six and eight o'clock in the morning reported the situation
to be unchanged. It was at this time that the frightful cat-
aclysm which overwhelmed the city and people of Saint
Pierre took place.

" At eight o'clock in the morning, just as the Girard-
line steamer was about to leave the city for Saint Pierre, an
immense mass of white clouds, rolling in gigantic spirals,
was perceived from Fort-de- France in the direction of Mont
Pelee; at the same instant the cable and telephone lines
connecting Saint Pierre with the capital were broken, the
barometer registered an abrupt fall, and a ' tidal' wave was
felt along the coast.

" In a few minutes clouds obscured the entire sky ; a
rain of stones, some of them weighing twenty grammes, beat
down upon Fort- de-France, followed by a rain of ashes
which lasted until near eleven o'clock. The steamer
Girard, which had left the city for Saint Pierre a quarter
past eight, after the ' tidal ' wave, continued on its course as

109

110 AFTEK THE CONFLAGRATION

far as the heights of Case-Pilote, which is exactly half way.
There, stopped by the stones and ashes which fell in con-
siderable quantity, it turned back to return to Fort-de-
France.

" It set out anew towards ten o'clock (after the great ex-
citement which had been aroused at Fort-de-France had
calmed), but after passing the point of Carbet, a terrifying
sight burst upon the passengers. At the base of the vol-
cano, which was shrouded in a cloud of smoke and ashes,
the entire coast for a distance of about five kilometres, from
the ' Minoterie' Blaisemont, situated a little north of Car-
bet, to Pointe-Lamarre, beyond the town of Sainte-Philo-
mene, was in flames ; the trees as well as the isolated houses
of the country were devoured by fire ; a dozen vessels in
the roadstead of Saint Pierre, of which two were American
steamers, burned at anchor. The coast seemed deserted ; on
the ocean nothing floated but wreckage. The heat stream-
ing from this immense conflagration prevented the boat
from proceeding, and it returned to Fort-de-France at one
o'clock in the afternoon, bringing with it the sinister tidings.

" As I had had no communication with Saint Pierre
since eight o'clock in the morning, and in view of the
serious events which had happened so far, I ordered the
commander of the Suchet to go there and place himself at
the service of the Governor, M. Mouttet. The Suchet arrived
at about half-past twelve. Towards three o'clock its com-
mander, Le Bris, was able to land at the Place Bertin,
which was covered with bodies. The heat which was disen-
gaged by the smoking ruins prevented his prosecuting his

THE SILENT CITY
From the Morne d'Orange

AFTEK THE CONFLAGKATION 111

investigations farther. He picked up along the coast the
wounded who had miraculously escaped, and proceeded to
Carbet, where he also took on board the injured, who were
removed to Fort-de-France. At three o'clock I sent a boat
to Saint Pierre on which was the Procureur of the re-
public, M. Lubin, who was commissioned to report the situ-
ation to me.

" M. Lubin landed at Saint Pierre and assured himself,
as had also done the commander of the Suchet, that the
entire population of the city had been wiped out. He
finally reached Carbet, where the population of the neigh-
boring villages, and many of the wounded who had escaped
the disaster, had gathered. A line of transport steamers
was organized between Carbet and Fort-de-France to carry
the unfortunate victims speedily to the capital.

" As to the exact circumstances which accompanied the
catastrophe, it would be difficult to give them with pre-
cision. It seems from the evidence gathered as well from
the few survivors as from the people who watched the cata-
clysm from a distance, that towards eight o'clock in the
morning, following, doubtless, a fissure in the flanks of the
volcano, a spout of fire burst over Saint Pierre, causing the
instant death of the entire population and setting fire simul-
taneously to all the houses of the city and all the ships in
the roadstead. The opinion of all who have up to this time
visited these scenes is that not one of the inhabitants who
at the hour of the catastrophe were in Saint Pierre has
escaped death.

" The number of victims in Saint Pierre alone is esti-

112 AFTEK THE CONFLAGKATION

mated at twenty-six thousand. But to this figure must be
added the inhabitants of the suburbs who succumbed, so
that the entire number may without exaggeration be reck-
oned at thirty thousand."

To the prompt action of the officials of Fort-de-France,
in sending vessels of inquiry and service to the scene of the
catastrophe, is due the saving of a few lives from the ocean
wreckage and from points immediately adjacent to the gen-
eral destruction. The number thus brought out seems not
positively to be known, but in most part it was composed of
the crews, officers and others who happened to be at the time
on board the different vessels anchored in the roadstead,
and who immediately were thrown into or sought refuge in
the water. Of the handful of immediate survivors from
Saint Pierre itself, who dragged themselves or were carried
out to points of safety on the landside, it would seem that
nearly all ultimately succumbed, and history generally re-
cites but a single survivor of the conflagration — the prisoner
Ciparis. The official report of the Procureur of Martinique
dealing with the efforts made to render assistance to the
afflicted, and addressed to the Procureur General under date
of May 10, gives a vivid picture of the early conditions of
the burning city, and of the obstacles that interposed to the
work that was contemplated :

" I left by the steamer Rubis at half-past two in the
afternoon, with a company of thirty men of the troop com-
manded by Lieutenant Tessier. . . . Among others Abbe

AFTEK THE CONFLAGKATION 113

Parel, accompanied by one of his vicars, took passage on
board the ship.

" After we had passed Case-Pilote we observed that the
sea was strewn with wreckage, and the Rubis was obliged to
slacken its speed in order to avoid breaking the helm. We
also noted some groups of people.

" We approach Car bet ; to our great astonishment there
are comparatively few people on the shore. Saint Pierre is
enveloped in a cloud of smoke and flames, especially in the
northern portion, known as the Fort.

" Saint Pierre and its suburbs seem to us a heap of ashes
and ruins. The roadstead contains nothing but an immense
quantity of drifting wood. Two iron-clad steamers, com-
pletely dismantled, tilted towards the land, with their boats
partly lifted from their pegs, have become the prey of the
flames. Not a trace of the hull of any sailing-vessel ; not
a boat ; we see only three or four coasting vessels of Basse-
Pointe, pirogues, their keels out of water, capsized ; on the
coast and in the surrounding country not a living soul !

" A dozen people took refuge on the rocks between Saint
Pierre and Carbet ; the launches of the Suchet went to their
relief. We knew at once that these people belonged to the
crews of the lost boats.

" I asked the captain to approach as near as possible to
Saint Pierre, and then, having a boat lowered, the lieu-
tenant, the ensign (ship-ensign Hebert, of the Suchet), and
I steered for the city itself. We landed a little beyond the
Place Mouillage ; the desolation there is complete and we
had to force our way to the Rue Bouille.

114 AFTEK THE CONFLAGKATION

" In this neighborhood we found bodies scattered every-
where, some of them distended by gases, and not carbonized ;
as to those who regained their homes, they seemed to us to
be completely charred. It is impossible to penetrate into
the interior and to reach the main street of the city, the
Rue Victor Hugo. In fact, to do so would be to walk over
a glowing brazier.

"We reembarked and landed at the Place Bertin.
There, too, are bodies swollen by gas, but not carbonized.
The hands are not shrivelled ; death seems to have been
swift and free from suffering. At this place are a dozen
bodies, one of them, that of a woman, with a beam lying
across her limbs.

" The quays exist no longer ; the trunks of the trees are
no more. The lighthouse of the Place Bertin, about twenty
metres in height, is razed to within three metres of the
ground. The interior staircase of iron which affords egress
appears to have been broken. The stones which remain are
uncharred, and the iron of the staircase has not been affected
by the fire. The grating of the fountain is twisted ; a dis-
torted spout still gives out water.

"We attempted to make our way through the Hue
Lucie, but the heat was so suffocating that we were obliged
to abandon the effort. Regaining our ship, we set out to
pick up the refugees of Carbet.

"From our examination of the ruined city I conclude
that the phenomenon which destroyed it was produced with
such suddenness and intensity that there was no chance of
escape ; the ships in the roadstead, which were under high

AFTEE THE CONFLAGEATION 115

pressure, notably the two cargo-boats and the Girard-line
steamer Diamant, which had just arrived at Fort-de-France,
could not evade it, and foundered or burned. The absence
of any massing of bodies in the Rue Bouille and the Place
Bertin, a street and square surrounded by extremely
populous houses, and the appearance of the bodies in
an uncarbonized condition, obviously prove that no panic
preceded the destruction ; if it had been otherwise, the
entire people would have hurried to the streets. Every-
one died on the spot where he was overtaken by the cata-
clysm.

"The appearance of our boats off Carbet attracted to
the shore about four hundred people, among them a score
of wounded. I found on inquiry that not a single one of
the people came from Saint Pierre ; all were from Carbet.
The town was not set on fire, but seemed to have been de-
vastated by water. The people all along the shore implored
to be taken along.

" M. Mauconduit gave me the following account of the
phenomena as he had seen it. He was at home in the
neighborhood of Carbet, overlooking Saint Pierre, when
towards eight o'clock in the morning his attention was
attracted by an immense sheaf of flames starting out of the
volcano. He saw no cause for uneasiness, but suddenly
noticed a spout of smoke advance, pour down upon Saint
Pierre, and completely cover the city. A very violent south
wind sprang up which dissipated the smoke, and at the same
time flames burst forth on all sides. Everything took fire
at the same moment : the roadstead, the city, and the sur-

116 AFTER THE CONFLAGRATION

rounding country. All this could have lasted but a few
moments ; there was no time to flee.

" In this manner I explain the phenomena observed by
me in the city itself, — the bodies of people appearing to
have perished without suffering, the small number of corpses
in the street (there are none except those of passers-by),
and the destruction of the lighthouse of the Place Bertin,
which had not during the time been touched by flame.
The city must have been asphyxiated, the fiery spout having
exhausted all the air that could be breathed. Perhaps there
was also a mingling of explosive gas, for at Fort- de-France
we heard loud detonations. I returned to the capital with
the conviction, I may say with the assurance, that not a
single resident of Saint Pierre could have saved himself.

" I am of the opinion that all the inhabitants of the
region lying between Sainte Philomene, Fonds-Core, Trois
Ponts, Morne Abel, Morne d'Orange, and the Quartier
Monsieur inclusive, have disappeared."

The incidents connected with the escape of the negro
prisoner from Saint Pierre form one of the most striking
episodes in the destruction of that city, and furnish a per-
sonal experience which is of interest in the light that it
throws upon the problem of the catastrophe. So unique a
record is perhaps not to be found in all the pages of history,
and even from the lighter vein of romance it would be diffi-
cult to extract anything that has more extraordinary rela-
tions.

From Thursday until Sunday Auguste Ciparis was

AFTEE THE CONFLAGRATION 117

lingering in the dungeon of the city jail, knowing nothing,
beyond his own wounds, of the world's tempest that had
rocked over him. He was burned to flesh and bone, but
he knew not that others had been burned like him, and
more. His cell was windowless, and all that could be seen
of the outer world came by way of the grated aperture in
the upper part of the door. No sound penetrated to his
cell, not even the tread of the keeper's footsteps came to
relieve the silence of this desolate abode.

When I was at Morne Rouge on June 2, I knew that
Ciparis was still confined there in a temporary lazaret which
had been established by the faithful priest of that district,
Pere Mary, but circumstances did not permit me to see him
at the time, and, unfortunately, the opportunity for an
interview never again presented itself. It was a good
service to history, however, to have the statement of this
negro taken by so accurate a recorder as Mr. George Ken-
nan, who had preceded me by a number of days, and who
has placed the facts which he gave to me personally in a
published form {Outlook, July 26, 1902). At the time
of Mr. Kennan's interview Ciparis was still showing the
effects of the frightful burns which his back and legs had
received, but was sufficiently composed to give a clear and
dispassionate account of his sufferings and of the physical
conditions that presented themselves to him. As he stated
his own experience, he was waiting for the usual break-
fast on the 8th, when it suddenly grew dark, and imme-
diately afterwards hot air, laden with ash, entered his room
through the door-grating. It came gently but fiercely.

118 AFTEK THE CONFLAGKATION

His flesh was instantly burned, and he jumped about in
agony, vainly calling for help. There was no help to come.
The heat that scorched him was intense, but lasted for an
instant only, and during that time he almost ceased to
breathe. There was no accompanying smoke, no noise of
any kind, and no odor to suggest a burning gas. The hot
air and ash were the working demons that tore his flesh.
Ciparis was clad at that time in hat, shirt and trousers, but
his clothing did not take fire ; yet beneath his shirt, the
back was terribly burned, and his body gave out the odor
of burning flesh. It is difficult to conceive of a lasting
agony greater than that which was suffered by this man.
For three days and more he had been without food of any
kind, and his only sustaining nourishment was the water
of his cell. This appears to have been unaffected by the
entering hot wave. During his long imprisonment he fre-
quently shouted for help, but the cries of " Save me !" were
answered only by the groans of anguish that followed. It
continued this way until the following Sunday, when it
chanced that searching parties neared his place of im-
prisonment. He heard voices, and renewed his cries for
help. The voices were those of two negroes, who, when
they satisfied themselves that the sound that came to them
was from a human being, immediately began the task of
rescue. The refuge was broken open, and in a short time
the half-dead prisoner was brought to free air.

The history of the "prisoner of Saint Pierre," while
most interesting in its details, is to an extent shorn of its
romance by the later discovery of at least one other sur-

AFTEE THE CONFLAGKATION 119

vivor, Leon Compere-Lean dre, also a negro, whose experi-
ence, as given to a representative of the Temps, is published
in the Bulletin of the Societe Astronomique de France (Au-
gust, 1902, p. 352). Leandre, who was a shoemaker by
trade, is described as being about twenty-eight years of
age, strongly built, and with a robust and vigorous aspect.
"On the 8th of May," he says, "about eight o'clock of
the morning, I was seated on the door-step of my house,
which was in the southeastern part of the city, and on the
Trace road (the road from Saint Pierre to Fort-de-France
which abuts, almost in the centre of the city, upon the
street Petit- Versailles). All of a sudden I felt a terrible
wind blowing, the earth began to tremble, and the sky sud-
denly became dark. I turned to go into the house, made
with great difficulty the three or four steps that separated
me from my room, and felt my arms and legs burning, also
my body. I dropped upon a table. At this moment four
others sought refuge in my room, crying and writhing with
pain, although their garments showed no sign of having
been touched by flame. At the end of ten minutes, one of
these, the young Delavaud girl, aged about ten years, fell
dead ; the others left. I then got up and went into another
room, where I found the father Delavaud, still clothed and
lying on the bed, dead. He was purple and inflated, but
the clothing was intact. I went out, and found in the court
two corpses interlocked : they were the bodies of the two
young men who had before been with me in the room.
Reentering the house, I came upon two other bodies, of two
men who had been in the garden when I returned to my

120 AFTER THE CONFLAGRATION

house at the beginning of the catastrophe. Crazed and
almost overcome, I threw myself upon a bed, inert and
awaiting death. My senses returned to me in perhaps an
hour, when I beheld the roof burning. With sufficient
strength left, my legs bleeding and covered with burns,
I ran to Fonds-Saint-Denis, six kilometres from Saint
Pierre. With the exception of the persons of whom I
have spoken, I heard no human cries ; I experienced no
degree of suffocation, and it was only air that was lacking
to me. But it was burning. There were neither ashes nor
mud. The entire city was aflame."

These single escapes from Saint Pierre only put into
more prominent relief the extraordinary nature of the
death-dealing blow, whose harvest was relentlessly com-
plete, and permitted practically no one to escape its path.
The condition k, indeed, almost inconceivable, for the
marvel is not that there should have been two isolated cases
of seemingly miraculous preservation, but that there were
not many more of the same kind. A scorch-blast that
clears all human life before it, and leaves in places un-
touched objects that are normally thought to be most
destructible, has many things for its characteristics which
science has still to learn.

VESUVIUS AND POMPEII: A PARALLEL

Whatever position may be assigned to the eruption of
Mont Pelee and the destruction of Saint Pierre in the cate-
gory of volcanic catastrophism, it is certain that in the
popular mind the phenomena of May 8 will most gener-
ally be associated with those of seemingly similar nature
which caused the overshadowing of Pompeii and Hercu-
laneum. Nor, indeed, can the scientific mind well turn
from this comparison, however differently the facts in the
two cases may present themselves. The annihilation of
two cities of almost exactly the same population, of nearly
equivalent position in relation to their destroyer, and the
suddenness of the paroxysm which forced the destruction
of life and property, instinctively suggest this parallel, if
nothing more. The two events, for the moment at least,
stand apart in the history of the world.

The historic records of the past are, unfortunately, of
such a nature as to compel the acceptance of a generous
supply of uncertainty in their consideration, and the
weighing of evidence the truth of which in some instances
cannot be established. The exact information that we pos-
sess relating to the destruction of the Roman cities is so
meagre that one is almost tempted to say that it does not
exist at all. In the writings of the younger Pliny alone

have we a contemporary statement describing the Vesuvian

121

122 VESUVIUS AND POMPEII: A PAEALLEL

eruption of the year 79, but in his famous letters to Tacitus
this observer makes no mention, by name at least, of either
Pompeii or Herculaneum. This extraordinary lapse has,
indeed, given reason for the belief that Pompeii and Her-
culaneum were not destroyed at this time, nor even neces-
sarily at the same time, as had been forcibly argued by
Lippi, in his work, "Fu il Fuoco o L'Acqua che soterro Pom-
pei ed Frcolano?" (Naples, 1816); for it has been thought
inconceivable that an observer so careful in recording facts
as was Pliny should have failed to note the principal inci-
dent in the events that he was describing. His position at
Misenum was such as to command under ordinary conditions
the sites of both the Roman cities ; and he could not have
failed to obtain information of so important a fact as a
destruction from runners, or from the very persons who
brought to him the details of his uncle's death. On the
other hand, Pliny's object in writing his letters to Tacitus
being mainly to give an account of his uncle's experiences,
it may not have been thought necessary at that late day,11
the writing following the event by many years, to refer to
a general calamity whose nature must have been known
to everybody. Much of the narrative bears evidence of
having been compiled from memory, and a memory that
was perhaps in a measure faulty. The statement, for ex-
ample, that the eruption of Vesuvius took place on the
" ninth of the calends of September" (corresponding to the
24th of August of our calendar) has been corrected by
some commentators to read the calends of December, and
for the reason, as claimed, that certain fruits found pre-

VESUVIUS AND POMPEII: A PAEALLEL 123

served among the ruins of Herculaneum were of a kind
that in the region of Campania do not appear before the
month of October. This discrepancy was already noted by
Professor Gaetano D'Ancora as early as 1803, in his work,
" Storio-Fisico degli Scavi di Ercolano e di Pompei ;" but
this one fact could with equal force be used as an argu-
ment to sustain the view that Herculaneum was not de-
stroyed at the time of the great Vesuvian eruption.

If Pliny makes uo direct mention of the fall of Pom-
peii and Herculaneum, yet very nearly the first sentence in
his Epistola XVI might perhaps be construed as noting
that catastrophe : Quamvis enim pulcherrimarum clade ter-
rarum, ut populi, ut urbes, memorabili casu quasi semper
victurus oceiderit. Such an interpretation, if sought for,
can, indeed, easily be found in the works of several of
Pliny's translators. Earl Orrery and Melmoth both trans-
late the passage above quoted in such a way as to make it
presumable that Pliny makes a direct reference to the main
fact of the catastrophe, although not stating it in word.
Earl Orrery's translation appears : " For although his fall
was attended by the destruction of most beautiful terri-
tories, seeming, as it were, destined to be remembered
equally with those nations and cities who perish by some
memorable event." *

Were the information that Pliny conveys all that we

* A better rendering would perhaps be : " For although he per-
ished in the destruction of these fairest of lands, yet he was destined
to survive forever just as cities and peoples visited by some great
catastrophe."

124 VESUVIUS AND POMPEII: A PAEALLEL

possessed regarding the Vesuvian eruption of 79, one would
be, indeed, well justified, even with the favor of the pas-
sage above quoted, in doubting that the overwhelming of
Pompeii and Herculaneum were part of the same event.
But the historic narration of Dion Cassius, even with its
fanciful details, can hardly be considered otherwise than
as supplying the deficiency which is left to us by Pliny, for
it could not well have been constructed without a knowledge
of facts then existing, even though many generations had
passed since Pliny's writing. This view is made the more
probable, seeing that no great eruption of Vesuvius is noted
in the interval between the year 79 and the appearance of
Dion Cassius's history ; nor do historians make reference to
any important destruction of towns or villages during this
interval.

Dion Cassius, and after him nearly all historians, refer
the destruction of life and property in Pompeii to an over-
whelming fall of ashes and cinders, and the accompaniment
of asphyxiating gases. The city itself is largely in ruins,
and this condition gives emphasis to Pliny's statement that
violent earthquakes were a part of the phenomena to which
he was witness. But we are not informed what amount of
damage was wrought by these earth-tremors; nor do we
positively know how far the ruin that had been inflicted
by the earlier earthquake of 63 had at that time been re-
paired, although it would seem from the studies of Over-
beck and others that most of the better houses and public
edifices had been fully restored.

The condition of the ruins at Pompeii does not entirely

VESUVIUS AND POMPEII: A PAEALLEL 125

forbid a comparison with those of Saint Pierre. There is
much that is suggestively common to both places — much
that is different. The form of destruction that is repre-
sented in Saint Pierre is much more violent, cataclysmic,
than in Pompeii, but the character of the ruins in the latter
city does not entirely remove the suspicion that some other
agent besides simple ashes and possible earthquakes may
have been at work in their formation. Had the data that we
now possess been available to the earlier writers on Pompeii,
it is not unreasonable to suppose that a somewhat different
interpretation might have been given to the destruction
than the one that has come down to us from the days of
Dion Cassius ; for the particular kind of destructivity which
characterized the eruption of Mont Pelee had not been
recognized before, and consequently could not have been
used as an alternative in analvsis to the evidence which was
carried by the covering of ashes and lapilli. The crum-
bled condition of the city, for example, where not identified
with the seismic movement of 63, has invariably been ac-
cepted as proof of new earthquake disturbances, but the
dislocation of the whole of Saint Pierre, without earth-
quake movements of any kind, makes necessarily doubtful
this interpretation of a portion of Pompeii's history.

The statement that the lives lost in Pompeii, which may
have been from eight hundred to fifteen hundred, was the
result of a sudden or steady overwhelming by hot ash cer-
tainly appears plausible on its face, and it gains strength
through the further statement that the elder Pliny, in his
effort to render assistance to the inhabitants of the threat-

126 VESUVIUS AND POMPEII: A PAEALLEL

ened region near the foot of Vesuvius, was unable to effect
a landing with the vessel or vessels of his fleet ; hence the
journey over to Stabise, on the opposite side of the bay,
where lived Pliny's friend, Pomponianus, and where the
Roman fleet-commander succumbed. But that which con-
flicts with this assumption is the fact that most of the in-
habitants of Pompeii had in truth left the city prior to its
destruction or overwhelming — a condition that is indis-
putably proved by the small number of bodies or skele-
tons that have been found in the ruins, and which stands in
opposition to the fanciful statement of Dion Cassius that
the inhabitants were destroyed while witnessing a perform-
ance in the theatre (theatres). Nor is this fact wholly in
harmony with the statement of the younger Pliny that a mes-
sage of danger had been received from Rectina,12 asking for
delivery, because all avenues of escape, except that of the
water, had been cut off from her location.

A further disturbing fact in this analysis is the circum-
stance that many of the bodies unearthed in the course of
modern excavation were found in attitudes of action or
motion, of full composure, and of seeming indifference to
impending danger. The reconstruction, however fanciful
it may be, of the baker standing over his oven, with the
baked bread alongside of him ; of the slave disputing with
his master, the latter holding the bunch of keys in his
hands, of the sleeping dog, etc., is a familiar theme and
chapter in the history of Pompeii ; but it is also the em-
phatic counterpart of the history of Saint Pierre, and the
two can be justly thought to read the same episode in

VESUVIUS AND POMPEII: A PAEALLEL 127

nature. It is not reasonable to assume that these poor in-
habitants of a city, which was already deserted by the
greater portion of its population, could have died a death
resulting from progressive incineration. Why have re-
mained, it can properly be asked, in the face of such accu-
mulating danger ? Why not have followed the balance of
the population elsewhere?- One can more readily believe
that those destroyed in Pompeii were slaves, hirelings, and
others, who had returned from some point of safety to re-
move or recover needed household-goods and articles of
luxury, and that during this visit, not thought at the time
to be particularly dangerous, they were suddenly anni-
hilated. Evidence favoring this assumption can easily be
found in the circumstance that many of the bodies, when
recovered, were found lying in the ash high above the
ground-surface. This was the condition in one of the most
interesting of the Fiorelli finds — the four bodies recovered
in the Via del Balcone Pensili.

The covering of ashes and lapilli that overlies Pompeii
has a general thickness of fifteen to twenty feet, the greater
part of which is more commonly assumed — although dis-
puted by many — to be the resultant of the single eruption
of Vesuvius in 79. The lowest stratum of eight feet
seems to be composed almost entirely of loose lapilli or
pumice, having a generally uniform appearance and com-
position. Those who hold to the view that Pompeii's de-
struction was one of simple incineration (or of incineration
lielped by seismic disturbances) point to this covering of ash
and cinders, which is indeed very heavy, but there is hardly

128 VESUVIUS AND POMPEII: A PAEALLEL

a way at this time of ascertaining what proportion of this
impounding material is representative of a single volcanic
eruption, or of a series of eruptions which may have fol-
lowed one another in fairly rapid succession. The absence
or presence of stratification is not an unfailing test in this
matter, as the loose fragmental material of volcanic dis-
charges is rapidly readjusted, especially where its position
is almost directly under the cloud of torrential rains. A
"stratifying" ash in such position might readily be eaten
out where the lapilli or pumice would remain intact. Such
is certainly the condition which one finds to-day at Saint
Pierre, after a considerable number of ash-falls and heavy
rains.

Geologists who have studied the Pompeian field have
lent themselves facilely to the theory of an overwhelming
sheet of ash and cinders largely on the ocular evidence that
is presented to them by the covering matrix and by the
general form of Vesuvius itself. It has come to be a recog-
nized belief with these investigators that the modern form
of Vesuvius, as distinguished from Monte Somma, dates
from the eruption of 79, which blew off the head of the
ancient volcano, or true Vesuvius, and gave us the double
mountain which so picturesquely dominates the landscape
of the Neapolitan Gulf. Had a cataclysm of this nature,
comparable in magnitude with the cataclysms of Papanda-
yang in 1772, of Krakatao in 1883, and of Bandai-San in
1888, actually taken place, it would assuredly have fur-
nished material sufficient to bury most of the cities of Cam-
pania situated on the side of the overthrow ; but it could,

VESUVIUS AND POMPEII: A PAEALLEL 129

at the same time, not have escaped the attention of the
younger Pliny, nor have been eliminated from his graphic
observations on the volcano's activity. The spectacle would
have been too momentous, too terrifying in its various
aspects, not to have produced a profound, an indelible im-
pression upon the mind of the young investigator. And it
would have been impossible in the face of such a destruc-
tion to quietly pen the lines : Nee defuerunt quifictis menti-
tisque terroribus vera pericula auger ent (" there were those
who magnified the real dangers by imaginary and false
terrors." — Orrery).13

In truth, about the only reason that geologists have for
assuming this decapitation of Vesuvius in the year 79,
the theory of which has been so carefully elaborated by Sir
William Hamilton, Breislak (" Voyages dans la Campanie,"
1801), Johnston-Lavis and others, is the statement of
Strabo that in his time the summit of the mountain " was
for the most part level, and wholly barren, covered with
ashes, etc.," supported by the further fact that the ancient
writers generally, speaking of Vesuvius, make mention of
a single summit only. This ignores, and perhaps justly,
whatever weight may have been carried by the earlier
accounts describing Spartacus's refuge in the crater of the
volcano. It must be admitted, however, that this form of
evidence is very slender, and wholly insufficient to establish
so important a premise. Mountain forms are, of all the
objects in nature, the most difficult to describe, and probably
with most persons the more imposing point of view will
constitute the basis or nucleus of a full description. Vesu-

130 VESUVIUS AND POMPEII: A PAEALLEL

vius as it exists to-day, or as it was a few years ago when
the summit was once more in the condition of a flat plain
without crateral hollow, is so prominent an object beside
Monte Somma that it could readily be taken to be the pict-
ure ; and it may be questioned if the greater number of
persons visiting Naples to-day are aware of the presence of
two mountains. Indeed, Delia Torre himself, in his "His-
toire et Phenomenes du Vesuve" (1760), after exhausting all
the evidence in favor of the position afterwards assumed by
Hamilton, justly asks : " After all, who can state that the
ancients in describing Vesuvius with a single summit did
not refer their descriptions to positions whence the moun-
tain really appeared single, as it does to-day from many
points ; and that there were not other positions, just as to-
day, whence the mountain appeared with two summits."*

This argument is certainly unanswerable, and goes to
the pith of the inquiry. While it does not permit us to
summarily dismiss the text-book illustrations of the recon-
structed Vesuvius of the time of Titus, it is sufficient to
make their value exceedingly dubious and to make more
than questionable the accepted teaching of geology.

The discovery among the Pompeian frescoes of a num-
ber of pictorial representations of a mountain-form almost
certainly that of Vesuvius, has thrown some additional

* " Qui pourrait dire, d'ailleurs, si les Anciens, en decrivant le
Vesuve avec un seul sommet, ne Vont pas observe des endroits aVou il nous
parait encore tel aujourd 'hui, qui sont en assez grand nombre ; et s'il riy
avait pas alors aVautres lieux, comme il y en a encore a present, d1 oil il
parut en avoir deux."

VESUVIUS AND POMPEII: A PARALLEL 131

light upon this subject, and should, it seems to me, be con-
clusive in the proof that the pictures offer. But the con-
clusions, which, with a general discussion of the subject,
are set forth at length by Enrico Cocchia in his paper, "La
Forma nelle Pitture e Descrizioni Antiche" published in the
"Atti delta Reale Accademia di Archeologia, Lettere, e Belle

RUE LUCIE— SAINT PIERRE

Artf' of Naples (XXI., 1900-1901), differ, and supply
argumentative conditions which the facts themselves do not
warrant. The three copies which are furnished by Cocchia
show plainly, however otherwise deficient, a rounded or an
acutely conical mountain, which, if applied in evidence,
immediately disposes of Strabo's contention that the summit

132 VESUVIUS AND POMPEII: A PAEALLEL

was a flat plain — unless, indeed, by this term the Greek
geographer meant to convey a plain of any size, such as
Vesuvius had a few years ago when the crateral hollow was
first filled in. There is not in any of them the remotest
suggestion of the truncated colossus which appears in most
of the geological restorations, and is made to accommodate
the Strabonic plain. The picture in the Latrario Pom-
peiano, where Vesuvius is placed under the protection of
Bacchus, does, indeed, confirm Strabo's description of the
vine-clad slopes of the volcano, but in this picture the
mountain is acutely conical, and conforms almost absolutely
in contour with the picture in the Delia Torre collection
depicting Monte Somma immediately previous to the great
eruption of 1631. Surely it is not to be conceived that a
cataclysm, such as is argued for the year 79, could have
left the major part of the mountain with its old outline
corresponding with that which appears nearly sixteen hun-
dred years later. One can, I think, safely accept the con-
clusions of Beloch and Nissen that the evidence that we
now possess is entirely insufficient to support the theory of
the Vesuvian decapitation ; or farther, that evidence of this
character does not exist.*

* Geologists have gone still farther, and point to the immense
amount of debris that lies beyond (southward of) Pompeii, and which
has presumably converted the Eoman city from a port to an inland
town, as evidence of this terrible destruction ; but the close investiga-
tions of Eosini and Euggiero show unmistakably that the extension
of the land outward into the bay, and the consequent lengthening of
the course of the Sarno (the ancient Sarnus), are the result of accre-

VESUVIUS AND POMPEII: A PAEALLEL

133

Dismissing these various negative points, which are
more forcible, perhaps, in rendering doubtful the accepted
version of the fall and destruction of that city than as
establishing a positive correlation with the facts of Saint
Pierre, it remains to be seen what there is to be found in

BURIAL-VAULT— SAINT PIERRE

the writings of Pliny to suggest a correspondence between
the phenomena of Vesuvius, which he so carefully observed,

tional growth extending through centuries, and independent of vol-
canic catastrophism. This view is accepted by Overbeck in his
11 Pompeii, in seinen Gebduden, Alterthumern, und Kunstwerken" 1884,
in collaboration with Mau.

134 VESUVIUS AND POMPEII: A PAEALLEL

and those of Mont Pelee. It will be recalled that the
main feature in the eruption of Pelee was the great black
cloud, luminous in part with incandescent particles or with
burning flame, which shot out from the crater, rolled down
the mountain-side with dazzling velocity, and fell upon the
doomed city, destroying it and the life that it contained
almost instantly. In his Epistola XX Pliny writes : " Ab
altero latere nubes atra et horrenda ignei spiritus tortis
vibratisque discursibus rupta in long as Jlammarum Jiguras
dehiscebat : fulguribus illm et similes et majores erant."
Whatever latitude may be given to a true interpretation of
this sentence, it is remarkable that the two translations of
Pliny's works which are generally accepted for their
strength, those of Melmoth and Earl Orrery, give to the
passage a rendering which, were it accepted literally, would
establish so close a correspondence between the phenomena
of 79 and those of 1902 as to make it difficult to resist the
conclusion that they were fundamentally alike. Melmoth's
translation of the passage appears as follows : ".On the other
side, a black and terrible cloud, bursting with an igneous
serpentine vapor, darted out a long train of fire, resembling,
but much larger than, the flashes of lightning." And Or-
rery's : " On the land-side a dark and horrible cloud,
charged with combustible matter, suddenly broke and shot
forth a long trail of fire, in the nature of lightning, but in
larger flashes." Either description in its application to the
destroying cloud of Pelee might be taken to replace the
descriptions of the officers of the Pouyer-Quertier, Captain
Freeman, M. Arnoux, or Pere Mary, while it in no way

VESUVIUS AND POMPEII: A PAEALLEL 135

interprets the phenomena of the ordinary ash-cloud that
belonged either to Pelee or to Vesuvius. This black cloud,
with its " trail of fire/' differing from lightning, might
reasonably be taken to be the correspondent of the " wall
of fire" to which the unfortunate sea-captains in the road-
stead of Saint Pierre refer in their narratives.

Pliny clearly wishes to distinguish between this terri-
fying cloud and the ordinary ash-cloud of the volcano, for
he adds : "Nee multo post ilia nubes descender e in terras,
operire maria" (" not long afterwards, the cloud descended,
and enshrouded the sea") ; and still later : " respicio
densa caligo terges imminebat, quce nos torrentis modo in-
fusa terra3, sequebatur" which appears in Earl Orrery's
translation as : "I looked back. A thick dark vapor just
behind us rolled along the ground like a torrent, and fol-
lowed us." To make it certain that this was not the ordi-
nary ash-cloud, the further fact is stated that : " the ashes
now began falling, although in no considerable quantity"
{"jam cinis adhuc tamen rarus"). It would be difficult
to construct a description more thoroughly according with
that given by the commandant of the Pony er-Quer tier for
the descending, rolling cloud of June 6, which is repre-
sented to have been the exact counterpart of the one of May
8; or one more thoroughly inapplicable to the ordinary
phenomena of eruptions.

It is to be noted as a singular fact that in this first re-
ported eruption of Vesuvius there was no emission of lava
— a condition very different from what appears in the later
cataclysms of that volcano. On the other hand, it is not

136 VESUVIUS AND POMPEII: A PARALLEL

unlikely that there may have been extensive mud-flows, and
that one of these overwhelmed Herculaneum, just as the
great mud-flow of May 5 from Pelee overwhelmed the Usine
Guerin. Many of the earlier observers, and notably Lippi,
of the Herculanean Academy of Naples, have argued for
this form of destruction of the city, and assuredly their
view is more plausible than that which assumes the ordinary
fall of cinders and ashes united with the volcanic waters or
rains into puzzuolana. The discrepancy between the condi-
tion that exists here and that at Pompeii is too great to per-
mit of the acceptance of this explanation.14

Students of Pompeian history have always been puzzled
to account for the condition of undress to which the casts
of so many of the bodies entombed in the ruins give evi-
dence, and the speculative, if not wholly ingenious, view
has been set forth (accepted by Fiorelli, Overbeck and
others) that the affrighted inhabitants of the doomed city
sought to facilitate their movements in flight by casting off
their clothing. Some of the casts show the bodies to have
been absolutely naked at the time when the mould or impres-
sion in the encircling matrix was taken. Can it for a mo-
ment be assumed that under a rain of hot or heated cinders
or of a fall of cinders of any kind, a fleeing community
would divest itself of a last garment, expose the naked flesh
to fiery missiles or to the direct blows of rapidly- falling frag-
ments of rock ? Those who have experienced the force of
impact of these erupted fragments will appreciate the ex-
treme improbability of any such condition. A far more
plausible explanation of the situation is to be found in the

VESUVIUS AND POMPEII: A PAEALLEL

137

history of Saint Pierre. Many of the bodies recovered from
its ruins \^ere also in a state of complete nudity, having
been divested of their clothing by the force of the tornadic
blast, in a manner precisely similar to that which so re-
peatedly happens in the course of the cyclones or tornadoes
of the Western United States.

BODIES ON THE TERRACE ROAD

There are a number of other resembling facts that asso-
ciate the bodies found at Pompeii with those at Saint Pierre,
and one has but to glance at the illustrations of positions to
be struck by their remarkable identity. We find the same
attitudes in posture, the identical ones in death. Groups
of bodies have been found in Pompeii (as in the House of

138 VESUVIUS AND POMPEII: A PARALLEL

Diomed) as well as in Saint Pierre, but in the former city
they have been mostly unearthed from cellars or basements,
and it has been assumed that the refuge from the descend-
ing shower of lapilli and ashes was sought in these subter-
ranean shelters, where death from starvation, asphyxiation
and enclosure finally overcame the unfortunates. One may
well pause before committing one's self to this theory, for it
hardly appears probable that people, seeing the enveloping
nature of the falling material from the volcano, would de-
liberately seal themselves up in such a way as to forcibly
close off retreat. It is far more natural to assume that only
a momentary shelter, suggested at the time of the unrolling
of the great black cloud which is described by Pliny, was
sought for by the paralyzed multitude — the history that is
developed in the ruins of Saint Pierre, where groups of
bodies have been found in basements and elsewhere, hud-
dled together, seeking protection from the approaching
dragon of death.

Overbeck, Mau and others refer to the deformed ves-
sels of pottery and glass-ware that were found in some
of the desolated habitations, and, adhering to the theory
that there was no extensive conflagration in Pompeii, con-
clude that this deformation was the result of a chemical
and physical process continuing through long time, ages, a
change that has also affected the coloring of many of the
larger Pompeian frescoes. It should be noted that pre-
cisely the same deformation is a characteristic of the wares
removed from Saint Pierre, and was brought about in the
period of a few hours or less, in the midst of a great con-

VESUVIUS AND POMPEII: A PAEALLEL 139

flagration, which destroyed color here and left it untouched
elsewhere.

The foregoing analysis of the facts appertaining to the
Pompeian catastrophe is practically all that is permitted to
us in our present state of knowledge. It allows clearly of
the assumption, even if it does not supply an adequate
demonstration of the fact, that the city and its inhabitants
were destroyed somewhat in the manner of Saint Pierre, by
an explosive or tornadic blast, and not through simple in-
cineration, as has been generally assumed by historians and
geologists. The evidence supporting this conclusion is
found in the Plinian narrative, in the ruin characteristics
of the two cities, and in the wholly accordant condition in
which so many of the bodies were found. There are no
facts known to us at this time that can properly be said to
invalidate this general conclusion ; and it can truly be said,
that were a historian or geologist to wander through the
ruins of Saint Pierre, and note his facts in the absence of a
knowledge of what really took place, he would almost cer-
tainly come to the conclusion that the ruin, death and
desolation which there prevail were brought about by
causes identical with those which wrecked the Roman city.

IX
ACROSS THE ISLAND TO ASSIER

It required no more than a glance to convince me
that to properly understand the geography of the Pelee
eruption an ascent of the mountain would be necessary. Its
past relief was too imperfectly known to permit of a tabu-
lation of its active points on old charts, and even the major
parts of the surface were distinguishable only with diffi-
culty, not wholly relieved of doubt. Some little effort to
come in closer contact with the volcano had been made
before my arrival in Fort-de-France, but the continuance
of eruptive blasts, and the descending smoke and showers
of ashes that swept the side of Saint Pierre, rendered the
attempts to approach abortive, and committed the investiga-
tor to a distant study of his subject. I felt that the eastern
or opposing face of the mountain, where the favor of the
steadily blowing trade-wind would be obtained, held out a
better prospect for success, and accordingly laid my plans to
cross the island.

It was my good fortune at this time to meet M. Fernand
Clerc, one of the wealthiest cane-growers in Martinique,
whose independent thought and action had saved him from
the catastrophe of the 8th, and whose principal sugar estate
at Vive seemed to me most favorably situated for the studies
that I had planned. Having himself, as a first pioneer
after the eruption, made a partial examination of the vol-

140

Expl.Heilprin

A MARTINIQUE PASTORAL— ASSIER

ACKOSS THE ISLAND TO ASSIEE 141

cano, he naturally felt a more than kindly interest in my
work, and with rare hospitality placed at my disposition all
the comforts that his estate offered, and even generously
undertook to send me out. On the following morning, May
29, a mud-bespattered wagonette, not unlike an abbreviated
Vienna fiacre, was in waiting for us — Mr. Leadbeater
accompanying me — and with it came a number of letters
commending us to various servants along the road, and in-
structing for a through journey. Our route lay across some
of the most interesting sections of the island, singularly
beautiful in their vistas and verdant in the glories of a not
too oppressively tropical vegetation. The volcanic ash had
fallen here and there in small quantity, only sufficiently to
give a gray touch to patches of soil, but not to the extent
of seriously injuring the fields of brilliant cane. After
leaving the heights above the Bay of Fort-de- France, the
road descends to Lamentin, whence it again ascends to gain
the high point of Gros Morne, surveying the Bays of
Robert, Galion and Trinite, and from Trinite closely skirts
the shore-line to the end. The detour circumscribes nearly
a third of the entire island, and gives charming vistas of
cultivated field and forest, of mountain peak and plain, and
of a ragged seashore, with outlying peninsulas, reefs and
islands. It covers about thirty-five miles. The road leaves
Fort-de- France by the native village, with its shaded cot-
tages and beautifully flowered patches of garden, and
almost immediately enters the open, where scattered groups
of habitations take the place of the village sites. These are
everywhere planted with the usual selection of tropical trees

142 ACKOSS THE ISLAND TO ASSIEE

and shrubs, many of them serving in the economy of the
household, others grown for ornament only. The natives
are passionately fond of showy flowers, and there exists
hardly a house-site that is not decorated with the magnifi-
cent scarlet hibiscus, or has not its bunches of Bougainvillea
and jessamine.

After an hour and a half we arrived at Lamentin, a
commune of nearly eleven thousand inhabitants, and now,
after Fort-de-France, the most populous location in the
island. It is for Martinique a place of considerable com-
mercial importance, being the outlet of perhaps the richest
cane region of the colony, and having the advantages of
direct steamboat communication with Fort-de-France. Like
many of its prototypes in France, it is a city essentially of
one street, with the main road entering at one end and
leaving at the other. This being market-day, with market
held in the open as well as in the hall set aside for its pur-
poses, the traversing thoroughfare was thronged, and it was
with difficulty that we could force a way through. Our
mission had already been made known to the people, and
naturally they hung close to us, trying to draw from our
superior experience (!) such comforting assurances as might
help to allay even a modicum of their anxiety. Of course
the volcan was on everybody's lips ; its possibilities and
limitations were accepted and debated at all corners, and
there were few among the inquiring multitude who did not
seem to feel that we could in some way stand between the
catastrophe and the mountain. With the city's people were
a number of refugees who had come in from the region of

ACEOSS THE ISLAND TO ASSIER

143

devastation and were no longer willing to trust themselves
to PeleVs capers. Affrighted, they still carried on their
heads what little belonged to them — frequently nothing
more than could be packed into a small panier, at other
times with a hen or rooster added, and the usual covering
bonnet. They had walked twenty and thirty miles, and

REFUGEES ON THE ROAD— GRANDE-ANSE

were still searching for an abiding-place where their sleep
would not be interrupted by PeleVs rumbling.

During our short halt here, we were pleasantly enter-
tained in the house of one of the leading French families,
whose members plied us thickly with questions bearing
upon Mont Pelee. Like everyone else in the region,

144 ACEOSS THE ISLAND TO ASSIER

they were deeply interested in the different problems that
the volcano had brought out, and this interest was a near
one, for even at this distance the ash had covered, even
if lightly, much of their estates, and the flying cinders
warned of a greater destruction. I explained in some detail
the nature of our mission, and offered the assurance that an
opened volcano was ordinarily not as dangerous as one that
was closed ; but I am certain that the good people, despite
their polite assurance that my explanation was quite accept-
able, continued to believe that an active volcano was about
as bad as it well could be. The geological conception was
to them not nearly so impressive as the picture of ashes and
cinders.

Beyond Lamentin the road continues pretty well mount-
ing, until it gains the summit-crest, whence a most striking
view is obtained of the Bay of Trinite and of the peninsula
of Caravelle stretching far out to sea, with islands and islets
dotting the water south of it. The truncated summit of
Vauclin was easily distinguishable in the south, and in the
extreme west the eye fell upon the bold knobs of the Pitons
de Carbet ; there was nothing visible of Pelee or of its great
ash-cloud. Many of the stream-beds that we crossed were
blocked with boulders of basalt or diorite, evidently ob-
tained from ancient lava-flows or from dikes which unite
the different volcanic masses to one another. Outcrops of
this rock are numerous in the hillsides, and a number of
them furnish the repair material for road-paving. All
through Martinique the work of the department of Ponts-
eU Chaussees is kept well in evidence, not alone in the cutting

ACKOSS THE ISLAND TO ASSIEE 145

and surface repair of roads, but in the regulation of the
running waters, and the construction over them of massive
bridges of iron and masonry. The beautiful forest and
mountain road between Morne Rouge and Ajoupa-Bouillon,
which was partially destroyed in the eruption of August 30,
exhibits the engineering qualities of that department to
great advantage and pleasantly reflects the service that the
government pays to the country.

We arrived at Trinite, where we changed animals
for the second time, shortly before two o'clock. Trinite,
which counts about eight thousand inhabitants, is the most
available port situated on the eastern side of the island.
It has an excellent and well-protected harbor (whose ad-
vantages, however, are to an extent lost in its windward
situation), and is served by both sail-craft and steamers
of large draught. The town has its pretty shaded place,
from which leads off the great long street with its rows
of low and closely-built houses, some plastered, stuccoed
and tinted yellow, pink and blue — others, and the greater
number, of wood. As at Lamentin, everybody was in the
streets. Our septuagenarian hostess, a person of color
but of exquisite manner, who helped us to the comforts
of her little inn, explained the bustle of the streets and
the events that had been transpiring. The difficulty of
understanding the French-Creole patois made this explana-
tion very welcome, although it was evident that the excite-
ment was all about the mauvaise montagne. We were
again nearing the gray ogre, and the closer we came up

with it the more absolutely did its doings engross the minds

146 ACKOSS THE ISLAND TO ASSIER

of the people. It was Mont Pelee in everything. Every
thunder detonation, every flash of lightning, was unques-
tionably a part of Pelee — the torrents and high seas were a
part of the same destructive monster.

Beyond Trinite the road follows closely the line of the
coast, leaving it here and there only for short distances to
climb over the buttressed prominences that project into the
sea. It is a charming piece of roadway, commanding ex-
quisite retrospects over land and water, and with now and
then a distant view of the lofty mountain summits. For
long distances it is bordered by continuous lines of rubber-
trees and elsewhere by fields of luxuriant cane. Much of
the vegetation of the coast lies hard-pressed against the
rocks, blown to them by the almost continuous east winds
which come in from the sea, and appears in recumbent
masses, at first suggesting a covering of moss and creeping
plants. At many points along the sea the old deposits of
Mont Pelee, masses of agglomerate and tuff, and of rusted
fields of decomposing lava, show up in bold cliffs and road-
banks, much of it appearing in the form of the orbicular
disintegration that is so distinctive of certain eruptive
rocks.

We passed through the towns of Sainte Marie and
Marigot at a wild gallop, compelling a roadway among the
crowds that had assembled. From this point onward, to
Grande- Anse and to Assier, the roadway was thick with
refugees, who came of all ages and sizes, each carrying
something that belonged to the household. They had left
their little huts and habitations, their smiling gardens, and

ACEOSS THE ISLAND TO ASSIEE 147

cultivated patches of cassava, "cabbage," and banana to
seek shelter with friends, or where a new nature would
again help them to a living. No one, or but few, had
lost anything so far, but it was feared that a day of reckon-
ing would come to the eastern side of the mountain as it
had already come to the western. The gardens and house-
tops had been grayed with ash, they had noted the forest
break under the load of mud that had been flung upon it,
and they had seen the sugar-cane flattened out over acres
as if it had been swept by a tornado. The poor people
had also seen their gentle streams racing in tumultuous
torrents, sweeping out their banks, and hurling great
boulders against hamlets and villages. It was time for
them to leave, they thought. From near and far, they
came. Habitations were deserted and the work in the
fields was stopped. The magnificent growths of cane were
left to do their own work with nature — to grow, ripen and
decay. There was none to cut the stalk for the large mills
nearby, and the gladdening smoke no longer issued from
the tall chimneys of the usines to mark the hours of labor.
Thousands of the peasantry had left their homes and many
of the settlements showed hardly more than closed doors.
The few who could better afford the luxury were riding in
mule-carts and ox-carts, some few on horseback, but the
greater number, young and old, were trudging along on
foot in the manner of the ancient patriarchs.

We arrived at the estate of Assier, to which the hospi-
tality of M. Clerc had commended us, just as the sun was
casting its last rays upon the tall cloud that Pelee was

148

ACEOSS THE ISLAND TO ASSIEE

throwing into space. The day had worn itself away so
that little was to be seen of the distant landscape beyond
gray color and mass. Ragged banana-leaves and drooping
cocoanut-crowns were silhouetted against the western sky,
but the eye no longer distinguished between fields of tares

Photo. Heilprin

STREET SCENE— LAMENTIN

and cane, and even the motley groups of refugees, who wen
fleeing from the shadow of the " bad mountain,'' w^ere onl;
with difficulty discernible on the open roadside. The vol-
cano was rolling out from its crest-line a volume of clouc
and ash that fairly bewildered the senses. Far up, tw<
miles and more, the column of white curling vapors wj

ACKOSS THE ISLAND TO ASS1EE 149

still mounting upward — lifting, rolling and unrolling, until
it lost itself in the general obscurity that surrounded it
It seemed to be by itself, severed from any connection
with mother earth. We were away six miles as the crow
flies, and yet had to toss our heads far back to see the arch-
ing summit-vapors thin out and melt into the cold blue of
impending night. No sound issued from the bosom of the
mountain, and only back of us could we hear the ocean's
distant roar, and above, the gentle rustle of the mango-
leaves as they dropped their still lingering crusts of ash.
Could it be that this wonderful, almost silent nature was
the same that but a few days before had wrought one of
the greatest catastrophes which the world's history records ?
As I watched the pensive face of our genial hostess,
Mile. Marie, who looked long and steadfastly at the volcano,
following the clouds of smoke until they had vanished in
the blue of night, I noticed an air of sadness come over it.
She too had lost what was dearest to her in that catastrophe.
Yet, with a faithful allegiance to her trust, and without
counting the moments of danger that were at all times hers,
she remained to do her duty to the Clerc household, and to
add to it what little of cheerful comfort could still be had.
The house stands, shaded by its great rubber-trees, aged by
a hundred and fifty years and more. On one side it looks
over to the not very distant ocean, and on another across a
deep valley which falls to the Kiviere Capot, and rises be-
yond to the slopes of Mont Pelee. Some ragged banana-
leaves rise up in front of its bounding wall and there are
tall and stately cocoanut-palms, others bowed down by

150 ACEOSS THE ISLAND TO ASSIEE

ash, and rows of the graceful filao or casuarina. Bleating
sheep and goats gambol about, unmindful of the storm of
life, and sleek cattle crop the herbage that still remains to
them. A more ideal location for a sojourn could hardly be
imagined, and we thought ourselves more than fortunate to
have been invited to it. The Usine Vive, which was to
furnish the mounts for our further journeys, was hardly two
miles distant. Its position of near proximity to Mont Pelee,
and the fact that it lies in a low level at the mouth of the
Capot River, had caused it to be temporarily abandoned for
night quarters, and a more congenial location was found at
Assier. It was only four days before our coming, on the
night of May 26, that Mont Pelee had gone through a
paroxysm of action that caused more than one mind on this
side of the mountain to waver, and to ask itself, When is
the time? Its extraordinary electric illumination on that
night, and the red effulgence that glowed through the
ascending cloud, appeared to indicate a new storm, and it
was not without reason that people, with so much that was
already back of them, became apprehensive of their safety.
It was then that the decision was arrived at to remove
from Vive. That same evening a frightened multitude was
hurrying over the road with hardly enough of night-light
to give their flight a course. Men, women and children,
black and white, on foot, on horseback and in wagon — all
took a common course, to leave a long shadow between
them and the volcano ; and Pelee continued to hurl out its
lightning and thunder.

Photo. Heilprin

THE EVENING GLOW ON PELEE'S PENNANT
From Assier

TO THE STORM-CLOUD OF PELfiE'S CRATER

The morning of May 31 was chosen for the ascent of
Mont Pelee. Our friends at Assier had prepared the little
that was needed for this journey, and looked after our
mounts, which we obtained at Vive, and the services of
three Martinique boys, who were deputed to accompany us.
Disturbed by mosquitoes and the anxiety that surrounded
our contemplated journey, we had put in a somewhat sleep-
less night, and it was with little happiness that we pro-
ceeded to carry out our plan of attack. The heavy rains
of the day before had blocked our effort to visit the Falaise,
and report still had it that flood-waters had cut the road
of the Capot. How much more the volcano was account-
able for we could not know.

The morning broke radiantly clear, and we felt measur-
ably encouraged to our work. Mont Pelee, with its miles
of towering steam, was sharply outlined against the western
sky, and seemed to look peacefully and kindly to the sun-
lit landscape that surrounded it. It was puffing only white,
and there were no wicked yellow and black clouds to tell of
its wrath.

We were on our way shortly before six o'clock. The
route lay, by circles and zigzags, westward, crossing the
Riviere Capot, and then through fields of cane and open
meadow-land, passing the village of Morne Balai with its

151

152 TO THE STOKM-CLOUD OF PELEE'S CEATER

clumps of cocoa-palms and bananas, its growths of cassava
and cane, and the blood-red hibiscus flowers scattered over
the cement and thatching that now hardly knew an inhabi-
tant. The closed door told of the flight from the home-
stead.

A little after nine o'clock we emerged upon the open
slope of the volcano, at an elevation of perhaps twenty-one
hundred feet above sea level. Ahead of us a long ridge-
line, broadly undulating at first and then contracting into a
fairly narrow arete, travelled almost directly westward to
the summit of the mountain. Its gray and desolate surface,
which only recently had been beautifully clothed with grass
and forest, rose from us at an angle of fifteen to twenty
degrees, gradually becoming steeper as it neared the top,
where scoriae, boulders and angular fragments of ejected
rock took the place of the ash of the lower slopes. At no
point did the gradient exceed thirty-five degrees. Travel-
ling over this ridge was not difficult, and we rapidly rose to
heights which commanded charming views of the receding
landscape, the blue ocean dashing its white surf against the
vertical cliffs of the coast, the muddy Capot and its out-
flowing sweep of chocolate, and, in the dim distance, the
Presqu'Ile de la Caravelle. On either side of us was a
fairly deep ravine, cut by the tumultuous waters which
sweep down the mountain's slope, the sides hanging with
broken and desolated gray forest, too dead to be sought now
by the few birds that had remained in the region. We
looked over into the adjoining chasm of the Riviere Falaise,
hoping to locate the new "crater" that had broken out

TO THE STORM-CLOUD OF PELEE'S CRATER 153

beneath the Trianon. The walls stood up like a burnt scar,
but there was peace inside, and not even a puff of vapor in
which to read the history of the mud-torrent that the day
before had run wild through the lower country.

We left our animals in charge of one of the Martinique
boys at an elevation of about two thousand two hundred
feet, and slowly pushed on to the summit. The ascent was
an easy one, even if fatiguing at times to the heart and
lungs, and presented nothing more difficult than the long
slopes of some of our own Appalachian peaks. The course
was direct, without zigzags of any kind ; and had it not been
for the particular conditions which existed at the summit,
the " climb" would have been without color-incident of any
kind. As it was, we knew only inferentially what was
taking place at the top, and were even in doubt as to
whether the summit could be reached at all. Up to this
time sky and weather had been most favorable, but the bat-
tered volcano had begun to gather to its crown the island's
mists, and its own clouds hung ominously over the summit.
In a short half-hour the parting-line between the land
and sky had been blotted out, and the balance of our ascent
was made in cloudland. A discomforting rain fell upon us,
and when we finally reached the summit of the mountain,
shortly before eleven o'clock, the weather was decidedly
nasty. My aneroid indicated an elevation of three thousand
nine hundred and seventy-five feet. We were standing on
what had been assumed to be the rim of the old crater, on
the rim of the basin that contained the Lac des Palmistes.
Between shifts in the clouds we obtained spectral glimpses

154 TO THE STORM-CLOUD OF PELEE'S CRATER

of the opposing mornes or pitons, their ragged lines rising
perhaps two hundred feet higher, and of the flat basin that
stretched off to their bases. But of the lake there was
nothing. So much of the basin as we could see was abso-
lutely dry, its floor brought up to a nearly uniform level
through the fragmental discharges from the volcano. At

Photo. Heilprin

ASH-CLOUD OF PELEE— FROM NEAR ASSIER

the point where we reached it there was a clearly marked
border rising two to three feet above the floor.

It was evident at a. glance that the old " crater," con-
trary to general belief and scientific report, had not been
blown out. It remained where picnic parties, seeking its
beautiful waters, annually found it to be, where the blue

TO THE STOKM-CLOUD OF PELEE'S CEATEK 155

lobelia adorned its banks, and where dwarf palms, suc-
ceeding to luxuriant forest, told the land of the tropical
sun. To-day not a trace of vegetable growth remained,
not even a lichen found attachment on the rough-surfaced
rocks that broke out from the scoriated floor. This, at
least, was what my observation told me. We sought in
vain the position of the vent whence issued the miles of
steam and ash that formed the spectacle of the morning, of
the evening before, and of every day since the eruption of
May 2. It should have been near to us, but where was
it? We could clearly hear the rumbling in its interior,
the grondement of continuing work, but the eye failed to
penetrate the sea of clouds that enveloped us, and made
our field of search necessarily limited. Ordinarily we
could see but a score of yards ahead, and frequently not
that far, and in the tempest that swept the mountain we
dared not attempt the actual exploration of the summit.

A crash of thunder, that seemed to rend the very heart
of the mountain, broke the storm upon us, and silenced
all other sounds. In an instant more a second crash, and
the lightning cut frenzied zigzags across the blackened
cloud-world of quivering Pelee. Then a third and a
fourth, and the pitons rolled the echoes to one another like
artillery fire. There was no need to look at one another —
we knew that we were in a storm-world of our own. What-
ever was taking place, was being acted immediately about
us. It was a strange sensation this, sitting not knowing
exactly where and having as an unseen neighbor one of the
mightiest destroying engines of the globe. The rain de-

156 TO THE STOKM-CLOUD OF PELEE'S CRATEE

scended in merciless torrents, and the lightning cut blind-
ing flashes about us. We sat bowed over our instruments,
to give them partial covering, but our clothing, so far as
protection to ourselves was concerned, might almost as well
have been in the sea. We hoped for a change, but there
was none. Our boys were unhappy and trembling in fear
of the volcano, and silent tears appealed for a descent.
They knew as well as we did that there could be but a
short interval between us and the fiery caldron, and they
knew, perhaps better than we did, that some of the detona-
tions which we had preferably referred to thunder were in
reality the warning notes of the volcano. Leadbeater and I
were not yet ready for the descent. That for which we had
climbed the mountain had eluded us, and yet could hardly
be more than a stone's throw away. We knew not pre-
cisely the condition, and dared not search ; but we thought
that a favoring gust might lift the clouds, and permit us to
see ahead. It did not come. My barometer had indicated
no gathering storm, no more than did the barometer of
Saint Pierre during the eruptions preceding the event of
May 8, and indicated no change now. The compass on the
crater rim showed, however, a variation of from thirty de-
grees to forty degrees eastward, the north needle being
turned sharply in the direction of Vive.

Three-quarters of an hour of PeleVs storm was suffi-
cient. It was perhaps the most trying of any like period
that I had, up to this time, experienced, and thinking it
useless to remain longer on the summit, I decided upon a
retreat.

TO THE STOEM-CLO UD OF PELEES CRATER 157

We were both storm-beaten and mind-beaten. A day's
effort had yielded little beyond permitting us to say that
we had reached the summit of the mountain. The descent
was as rapid as the conditions of the atmosphere and moun-
tain would permit, but it was not easy work. The deluge
had graven uncomfortable hollows and fissures in the vol-
cano's sides, and running streams of mud and water had
taken the place of the hard slope of the early morning.
There was no longer a secure foothold anywhere, and it
was with difficulty that we kept from sliding into the gorge
that lay on both sides of us. By the time we reached our
mules, which had been taken to a lower level by the fright-
ened attendant, the storm had partially lifted, and to our
surprise, looking beneath the clouds, we found the Falaise,
which had been running quietly on our up-journey, seeth-
ing with steam, and threading its course to the Capot and
to the sea in a long train of curling and puffing vapors.
We followed with our eyes the circuit of the steaming river
for miles across the still fairly green country, watching
the vapor columns as they wildly tossed and bowed, but
hearing no sounds beyond those of our immediate neigh-
borhood. The scene was an extraordinary one, and one
that could only be compared in its effect to a chain of loco-
motives steaming in line. At this time we thought that
Pelee had broken out on the side turned to us, and was
disengaging its mud directly into the trough of the Falaise.

Our experience on the narrowed summit of Pelee
during this first ascent was so novel and so personal in its
sensations that it seems only natural to place here the im-

158 TO THE STOKM-CLOUD OF PELEE'S CKATEE

pressions of my associate, Mr. Leadbeater, as he has re-
corded them elsewhere. No apology is, therefore, necessary
for introducing this portion of his graphic narrative :

When we reached the edge of the old crater, at an elevation of
about four thousand feet (the basin that had contained the Lac des
Palmistes), it rained in torrents. We waited about fifteen minutes,
hoping it would clear up and enable us to see something. Suddenly
there crashed out of the very air above our heads a cannonading so
terrific that the mountain seemed to quake and tremble before it.
It took us some minutes to realize that it was a peal of thunder.
Then it commenced to thunder and lightning incessantly, and the
thunder followed so quickly after the lightning that they seemed to
come simultaneously. The awful lightning flashes came in sheets and
bolts of fire and were blinding rather than illuminating. Indeed, the
thunder was so loud that we could feel the ground heave, as it were,
under us, and the air about us vibrate. It rained so hard we could
not see ten feet away, and so awed were we by the thunder and light-
ning, and so oppressed by the hot, sultry atmosphere, that we did not
know but that we were being overwhelmed by another eruption. I
placed my camera on the ground and lay upon it to keep it dry.
But it rained through my clothes, and it must have penetrated even
through my body, for the camera was soaked. Those frightful
minutes when I lay on the ground shielding my camera, with the
rain descending in perfect floods of water — I never knew it could rain
as it did then — with the appalling thunder-charged flashes playing
incessantly about me and the very air quivering with the rapidity of
the detonations, and but a few feet away the seething, sweltering
crater of the most destructive volcano the world has ever seen, will
always stand out in my memory as a weird and horrible dream. At
last we could bear it no longer, and started to come down the moun-
tain, following our tracks as best we could. While descending the
mountain we found that the heavy rains had washed gorges in the
mud-covering of the mountain two to three feet deep, and in the

TO THE STOKM-CLOUD OF PELEE'S CEATER 159

blinding rain we frequently stopped on the edge of one of these
gullies, which, suddenly giving way, caused us to slip and slide most
of the way down. When we got to the end of the " hogback," where
we left the mules, they and their keeper had gone. We found them
later on farther down the mountain standing in the bright sunshine.

Our day's work, while giving to us many novel and
imperishable sensations, had terminated unsuccessfully.
We had been repulsed by the volcano, mudded and
drenched in a way that severely cautioned us in any further
effort not to inquire too closely into nature's hidden secrets.
The great caldron of blowing steam and ash had not been
reached, or even seen, although we could hardly have been
more than a hundred yards from its border. The question
still remained, where and how was it ? The evening wore
off quietly as that of the preceding day, and Pelee once
more presented itself in its form of grand and unconquered
magnificence. I studied carefully its vast steam-cloud, with
its ominous puffs of yellow and brown, and attempted to
locate the precise position of its emergence ; but what we
saw this evening, we had seen the evening before, and also
on the evening before that. The lesson still remained to
be learned, and I determined upon another ascent for the
following day.

Kennan, Jaccaci and Varian, three other investigators
of the phenomena of Mont Pelee, had by this time come in
from Morne Rouge, and, inspired by the extraordinary
workings of the volcano which they had witnessed there
and at Vive, had also determined upon an ascent. We
joined forces. As on the day before, the mounts were

160 TO THE STOKM-CLOUD OF PELEE'S CRATER

obtained at Vive, which also furnished the somewhat larger
number of attendants and carriers who were to do duty for
us. We left the latter place shortly after seven-thirty. Our
route, except in some narrowing curves, was virtually the
same that we had travelled the day before. Once past
Morne Balai, we followed the direct course to the eastern
arete, up which we somewhat laboriously picked our way.
The ascent, owing to the still soft and completely rifted
condition of the surface brought about by the heavy rains,
was considerably more fatiguing than on the previous day,
but reaching the summit was merely " a pulling away at it,"
with plenty of stops to take breath and ease the heart's
action. The heat of the open sunlight was, however, very
trying, and it was intense on the exposed slope of cinder
and ash. There was not even the whisper of a breeze.
Mr. Jaccaci succumbed to an early attack of acute dizziness
or vertigo, and was obliged to abandon the ascent. When
we came up to the old-crater rim, the Lac des Palmistes,
shortly before eleven o'clock, the weather and mountain
conditions were desperately like those which ushered in the
storm of the preceding day. The aged mountain had again
buried its head in cloud and vapor, and growling thunder
reverberations held out little hope that we should be able
to accomplish more than we had already done. Of the dis-
tant lowland only parting patches could now be seen, and
before long even these were blotted out by mist and rain.
On the top it was all cloudland, and with squally rains
coming and going in quick turns.

We caught fleeting glimpses of the opposing mornes

TO THE STOEM-CLOUD OF PELEE'S CEATEE 161

that rimmed in the basin at its farther side, but as yet saw
nothing that gave more than a feeble indication as to where
might be the line of the working crater. My aneroid read-
ing, without correction for temperature, gave for our posi-
tion— the same that we had occupied the day before — four
thousand and twenty-five feet, which satisfied me that the
old level of the mountain had been maintained, and that
there had been, contrary to what had been reported, no
subsidence as the result of the catastrophic explosion of the
8th. It is true that the piton which bore the cross on the
Morne de La Croix had tumbled as the result of a fracture,
but this loss to the mountain of perhaps fifty to one hun-
dred feet in no way disturbed the general aspect or mass
of the volcano. The shallow trough of the former Lac is
now floored with angular blocks and fragments of ancient
volcanic debris, forming part of the former stock of the
volcano, and with recently ejected scoriae, lapilli and mud-
ash. These built up the outer face, for three hundred feet
or more, of the top portion of the main cone. I took the
temperature at several points on the lake-floor and over the
rim of the basin and found it to be, at two or three inches
below the surface, 124° to 130° F. ; at one point, at a greater
depth, the mercury rose to 162°. It was evident that this
high temperature, about 60° above that of the air, was
merely that of the ejected material which had not yet had
the time to cool. Puffs of steam and sulphur vapor were
issuing from a number of surface vents, and from beneath
great boulder masses whose ragged and heated surfaces were

scarred with yellow sulphur blotches, and gave evidence of

162 TO THE STOEM-CLOUD OF PELEE'S CEATEE

having only recently been hurled to their places from the
volcano's mouth.

We waited patiently for a lifting of the clouds, and it
came at last. Below the mountain's clouds we could clearly
mark out the ascending column of steam, with its flocculent
whorls rolling in upon themselves and upward. The posi-
tion of the crater had been located, but alas ! it was for
hardly more than an instant. The scene had shifted and
disappeared. We were once more in cloudland, waiting
and hoping, with our Martinique boys impatient of their
assumed trials.

An angry cold wind was now swirling around both sides
of the mountain, and with it came a seemingly hopeless
rain. All of a sudden a gust cleared the summit, and a
white sunlight illumined the near horizon. It seemed
hardly more than three hundred feet from us. Across the
steaming lake-bed, little mindful of its puffs of vapor and
sulphur, we dashed to the line above which welled out the
steam-cloud of the volcano, and almost in an instant stood
upon the rim of the giant rift in whose interior the world
was being made in miniature. We had reached our point.
We were four feet, perhaps less, from a point whence a
plummet could be dropped into the seething furnace, witness-
ing a scene of terrorizing grandeur which can be conceived
only by the very few who have observed similar scenes
elsewhere. Momentary flashes of light permitted us to
see far into the tempest- tossed caldron, but at no time was
the floor visible, for over it rolled the vapors that rose out
to mountain heights. With almost lightning speed they

TO THE STOKM-CLOUD OF PELEE'S CRATEE 163

were shot out into space, to be lost almost as soon as they
had appeared. Facing us, at a distance of seemingly not
more than two hundred and fifty feet, danced the walls of
what appeared to be the opposing face of the crater, and
somewhat nearer the ragged white rocks, burnt-out cinder
masses, whose brilliant incandescence flashed out like

Photo. Heilprin

A BURST FROM THE CRATER

beacon-lights some days after the fatal 8th, and even at
our later day illumined the night-crown of the volcano
with a glow of fire. We could not tell at the time if they
were part of a cinder-cone, or merely an accumulated heap
that had been piled upon itself. The spectacle was a stu-
pendous one, — like a wild tempest raging everywhere. We

164 TO THE STOKM-CLOUD OF PELEE'S CEATER

stood silent, overawed in its presence. The ground trembled
at times, but never with any degree of force. We felt no
inconvenience from either gas or steam. A low rumbling
detonation, broken at intervals by louder bursts, crept about
the hidden floor of the interior, from which also issued the
sounds of clinking, falling and sliding cinders, the hissing
of the emerging steam — sounds which one would fain de-
scribe were it possible to do so. I tried so far as it was
possible to localize the issuing sounds, but the "blanket-
ing" by the enormous masses of swirling steam prevented
this ; everything seemed to come from everywhere, with
no marked accentuation in any particular quarter. Occa-
sional gusts of wind cleared the foreground, and displayed
the giant smoke-column in grand magnificence.

Our Martinique boys appeared to be as much impressed
by the scene as we ourselves were, and for a time lost all
fear of the awakening dragon. We found that we were
standing on the edge of a vertical, perhaps even overhang-
ing, cliff, and not feeling disposed to remain longer than
was necessary to make note-book observations and take pho-
tographic views, left rather precipitately for lower regions.

I felt that finally I had stood over nature's great labora-
tory, and been permitted to study some of its workings.
Many years before on Vesuvius I had gazed into the crater
funnel, and watched the molten magma of the earth rise
and fall, but the scene was one that could not compai
with this, grand and inspiring though it was. I attempted
to locate the axis of the vent as nearly as the direction of
the largely-obscured walls and the position of the basin of

TO THE STOKM-CLOUD OF PELEE'S CKATEE 165

the Lac des Palmistes permitted, which was north to south,
slightly southwest. The magnetic needle, which showed
such a marked deflection on the border of the lake-basin,
was normal or nearly so. The form of the crater was at
this time that of a caldron-rift, pitching steeply downward,
and opening in a direction a little off from the line to Saint
Pierre. The length could be only roughly approximated,
and at no time could we positively ascertain the extreme
boundaries. There can be no question that it traversed the
position of the narrow rift known as the Fente, or the Terre
Fcndue, which had been a feature of the mountain since the
eruption of 1851, perhaps considerably preceding that event
in its existence.

The fact that, standing on the rim of so active a crater,
we were not inconvenienced by any marked excess of tem-
perature seems rather remarkable, and might be thought
to find its explanation in the very rapidly ascending masses
of steam — the condition of continuous atmospheric dis-
placements which it brought about. But even these were
little appreciable where we stood, which was more like a
region of almost absolute calm, despite the storm that raged
in its centre, than one of flickering disturbance.

XI
THE GEOGRAPHY OF MONT PEL^E

Mont Pelee, or more properly, Montagne Pelee, which,
prior to the eruption of May 8, was barely known beyond
its own little territory, occupies with its contreforts nearly
the whole northern section of the island of Martinique. Its
humble height, hardly equalling that of famous Ben Nevis
of the Scottish Highlands, does not permit it to loom up
lofty, but it holds it crown veiled in mist during the greater
part of the day.

The volcano derives its appellation of " bald mountain"
from a bare spot which it carried about its crown at the
time when it was first described by Dutertre in (about)
1640; and inferentially this characteristic is reviewed by
Father Labat, who in his " Nouveau Voyage aux Isles de
VAmerique" (the Hague, 1724) refers to himself as the
"pere (or Mont) pele" (bald father). Just where or how
this bald spot on the volcano was located cannot at this
time be determined, for it is certain that in the past, as well
as in the recent present, the mountain was covered with a
luxuriant vegetation quite to its summit; and I am in-
formed that even the precipitous face of the Morne de La
Croix was similarly garnished. Felix Lombard, in his
paper " La Martinique et les Erreurs des Geographes"*

* Eevue Scientifique, August 9, 1884.
166

THE GEOGEAPHY OF MONT PELEE 167

dwells emphatically upon this characteristic of the moun-
tain, and asserts that at the time of his writing the volcano
was visible in its full extent, and entirely covered with
most vigorous green {vert le plus vigoureux) .

The " grands bois" or what have frequently been re-
ferred to by writers as the forest primeval, with all the
wealth and luxuriance of vegetation that a tropical nature
can supply, were the glory of the mountain. MM. Lep-
rieur, Peyraud and Rufz, who composed the scientific com-
mission that investigated the eruption of August, 1851,
speak in their report of the magnificent woodlands of Bro-
melia, Melastome and Gay Lussacia that they were obliged
to traverse, and which only thinned off to lighter woods
near the summit. Hearn describes the same forest forty
years later, and refers rapturously to the beauties and fasci-
nation of the tangled mazes which held one at almost every
step. Even so late as the closing days of April of the
present year the woodland was in nearly its full magnifi-
cence. Little or nothing remains of all this on the sides
where the outflows took place, and it is astonishing how
completely the traces of the vegetable growth have been ex-
tinguished. Mont Pelee shows up from the west and south-
west naked as though it had never known a cover. But far
beyond the line of absolute destruction, the tree-growth has
been crippled, grayed and laid to low measure by the ash
that has fallen upon it. On the eastern face of the moun-
tain the zone of destruction, previous to the great eruption
of August 30, covered only the middle and upper slopes ;
and the forest, though battered and burned, remained stand-

168 THE GEOGRAPHY OF MONT PELEE

ing in part. To-day this has also passed, and the slopes lie
bared as on the side turned to Saint Pierre. Three days
before the August eruption I noted the cindered forest of
the Falaise gorge, at an elevation of from seventeen hundred
to eighteen hundred feet, returning to life, with brilliant
greens decking the new crowns. The revivifying tree-ferns
were especially beautiful. New life was also beginning to
clothe the ridge-sands of the Rivieres des Peres and Seche
on the southwest. All of this has disappeared — most of it
extinguished absolutely. Of the growth of palms at the
summit of the mountain, and the clumps of fern and lobelia
that in the early part of the year delighted the visitor to
the Lac des Palmistes, not a vestige remains — nothing to
indicate that such, or other, vegetable growths could ever
have existed.

The summit of Mont Pelee, which commands a superb
view of the island and of its surrounding ocean, was prior
to the May eruption constructed in greater part of a small
lake-basin and of a line of bounding heights lying on its
western and northern sides. The highest of these, which
bore the cross * that was placed upon it by the late Pere
Mary, was the Morne de La Croix, whose height is roughly
assumed to have been two hundred and fifty or three hun-
dred feet. Its elevation above the sea is generally stated to
have been four thousand four hundred and twenty-eight
feet, which is the measurement of Dupuget15 in 1796 ; but
the determinations of the scientific commission of 1851 give

* Replacing the more ancient one.

Photo. Heilprin

ON THE VOLCANO'S DEVASTATED SLOPE

THE GEOGRAPHY OF MONT PELEE 169

for the full height of the volcano only four thousand one
hundred and ninety feet (twelve hundred and seventy-
seven metres) , which conforms closely to my own baro-
metric values, and, I believe, more nearly represents the
true altitude.

The lake itself was a shallow pan of water whose sur-
face lay but little below the bounding lip of the basin on
the eastern side. Leprieur, Peyraud and Rufz, who visited
it in 1851, immediately after the eruption of that year,
describe it as being about three hundred paces in cir-
cumference, and resting on z floor of mud and pumice
fragments. Their estimate is, I believe, an approximately
correct one, although the lake is sometimes described as
having been very much larger. Beyond the position that
it occupies, there is little to suggest for it the nature of a
crater-lake, which it is very generally assumed to be, and it
may be reasonably questioned whether it had this structure.
Labat refers to this summit lake in his work published in
1722 (1724), but it would seem that its crateral origin was
assumed only after the publication by Jonnes of his paper
"Explorations Geologiques et Miner alogiques du Volcan
eteint de la Montagne Pelee." * The reference to the lake
is, however, not very clear ; and the statement that the
" great crater is now converted into a lake" (translation)
may very properly refer to the crater on the southwest side

* Bulletin Societe Philomatique de Paris, 1820, p. 8. Pelee was
ascended by La Condamine, who also made a measurement of the
mountain, but I have been unable to obtain the record of observations
made on this visit.

170 THE GEOGRAPHY OF MONT PELEE

and to the Etang Sec. Jonnes could hardly have referred
to the summit lake as occupying a large crater.16

At the time of my visit, this attractive mountain tarn,
which for many years had been the central point to pic-
nicking parties of an extensive region around, and to which
an excursion had been planned in Saint Pierre for the 4th
of May, had disappeared, and no trace of the waters re-
mained. The basin itself had been largely filled in with
matter ejected from the volcano, so that the floor lay only
from two to three feet below the rim on the eastern side.
The floor was still steaming over most of its part, and it
gave out a peculiar " steamed" odor of mineral oil. I esti-
mated the distance across the basin to the foot of the Morne
de La Croix to be about three hundred feet. In just what
manner the lake- water was thrown off as the result of the
first eruption cannot be known ; but it is reasonable to
assume that the greater part of it may have been steamed
off by the heated ejecta that were thrown into it. There
is nothing to support the view that it was in any way
sucked into the crater and became a determining factor in
the explosion. The lake-basin remains intact, and has
undergone no changes beyond that of infilling.

Of the ejecta covering the lake-floor the greater part was
at the time of my first visit constituted of mud-ash and
angular blocks or fragments of andesite, trachyte (?) and
diorite, with here and there scattered boulders and
" bombs" of large size and composite character, and rep-
resenting the ancient stock of the volcano. Some of these
were coated with sulphur crusts, and others with iron-

THE GEOGEAPHY OF MONT PELEE 171

chloride. Steam was issuing vigorously from their sur-
faces, and equally so from the general floor of the lake-
basin. In crossing the basin we were obliged to thread
our way carefully between these steam-jets, which were
still numerous, especially towards the side of the crater.
My thermometer, thrust two and three inches beneath
the surface, gave a temperature of from one hundred

Photo. Heilprin

BASIN OF THE LAC DES PALMISTES AND THE SHATTERED
MORNE DE LA CROIX

and twenty-four to one hundred and thirty-two degrees,
and at a somewhat deeper point, one hundred and sixty-
three degrees. At the time of my last visit to the Lac
des Palmistes, August 30, the summit of the volcano was
so completely shrouded in clouds, steam and ash that it
was impossible to make topographic observations of any
kind ; nor, indeed, in the face of the raking fire of bombs,
did I feel disposed to penetrate beyond the crest of the

172 THE GEOGKAPHY OF MONT PELEE

mountain. The changes that may have taken place on this
little plateau-summit as the result of the more recent ac-
tivities of the volcano are, therefore, unknown to me ; but
so far as I could judge of them from a distance, they could
not have been very marked.

The plateau-summit of the volcano, which is thus partly
occupied by the basin of the Lac des Palmistes, slopes off
southward in the direction of Morne Rouge, and " spills" off
on the east and southeast in a gradual coalescence with the
outer slopes o^ the mountain. On the southwest, where it
falls in with the crest of the crater, there is a rise of a few
yards, and then follows the plunging wall of the crater.
The westerly (or crater) wall of the Morne de La Croix
drops into the basin of the Etang Sec at an angle (in its
upper part) of hardly less than seventy degrees — a pic, to
use the expression of French investigators. How much
of the Morne de La Croix has fallen, no one positively
knows ; but it is certain, as could easily be determined by
a comparison of contours, that not nearly so much of it
disappeared in the early weeks following the eruption of
May 8 as was generally supposed. I should rather believe
that its height was lessened only by from fifty to seventy
feet, instead of the one hundred and fifty as claimed. The
fall of the piton itself — i.e., of the pinnacle surmounting
the Morne — seems to have been finally accomplished on
May 24, as is published in U Opinion of Fort-de-France,
in despatches from Morne Capot, as follows : one-twenty
p.m., " Autre fragment Piton s'est ecroule ;" and eight p.m.,
" Piton disparu completement"

THE GEOGKAPHY OF MONT PELEE 173

Mont Pelee has not the conical outline of the typical
volcano, but is elongated on a northwest and southeast axis,
with the highest point lying in the northwest. It is plain
to see in this direction that it is only part of a former
larger mountain, whose buttressed masses lie still farther
to the north, and of which the Morne Siberie and the
Piton Pierreux, the latter nearly two thousand feet in ele-
vation, are still prominent relics. The sea face is on this
side abrupt and precipitous, presenting ragged bluffs and
promontories, with some detached islands and island points.
Standing off some little distance from this side of the
coast, the spectator obtains the only symmetrically con-
toured outline of the volcano, and notes the majestic extent
of its great flanks as they sweep over the whole forefoot of
the island. The gently falling slopes to the interior, being
usually free of complication and rising with low gradients
of from fifteen to twenty-five degrees, are exceedingly
pleasing to the eye, and conform to the picture of many of
the other volcanic mountains of the Lesser Antilles. To-
wards the southeast, Pelee sends out a long ridge to unite
with the mass of the Pitons de Carbet, the point of second
elevation in- the land (three thousand nine hundred and
sixty feet), and thus builds out, with the peaks of Carbet
and their long slopes, nearly the whole mountain relief of
two-thirds of the island of Martinique. The volcano itself
covers a surface area of about fifty square miles.

The singular manner in which the mountain has been
cut up into ridge-backs and deeply separating water-ways,
all radiating from almost the exact centre of the volcano,

174 THE GEOGKAPHY OF MONT PELEE

may liken it to a many-rayed elevated star. Some of
these ridge-backs are sharp enough to permit them to be
called aretes, and they fall off rapidly into the troughs that
lie on either side. Streams of various degrees of strength
occupy these troughs, and in nearly all cases have indi-
vidual courses directed to the sea. On the east side alone
are they tributary to a major water, the Riviere Capot,
which rises in a fairly deep basin several miles to the east-
ward of Saint Pierre, and defines approximately the eastern
boundary of the region that is dominated by the volcano.
All of these streams, with the exception of the Capot, have
courses of not more than four or five miles in a direct line,
but despite this condition many have proved wildly destruc-
tive during periods of heavy rains. More than one settle-
ment keeps in sad memory the picture of ruin which water
and rock have wrought. Basse-Pointe and Precheur, with
their acres of giant rocks, their canoned streets and battered
walls, read impressively this side chapter from the history
of Mont Pelee. I determined the height of the flood- water
of the Falaise, at its confluence with the Capot, to have been
at least thirty-five feet above the normal level of the two
streams. The extended flood-plain is at this point covered
with giant boulders, many of them five feet and more in
diameter, and all of them rounded as though they had
travelled for many miles. I measured some exceptional
blocks that were from eight to twelve feet in length. All
the masses were volcanic, — basalts, andesites, trachytes,
pumice, etc., — and represented the old stock of the volcano
or of its predecessor. Their perfectly rounded and planed

THE GEOGEAPHY OF MONT PELEE 175

forms, which were unmistakably not due to weathering or to
internal disintegration, suggest for them a possible oceanic
location, placed high on the slopes of the volcano as the
result of land-elevation. They seem to be more than the
simple surface deposit of earlier eruptions. The condition
of the Falaise was similar to that which we observed in the
case of the streams of Precheur and Basse- Pointe.

Not less than twenty-five streams, about one-half of
which have been dignified with the name of Riviere, radiate
from the slopes of Pelee, and the greater number of these
occupy deep ravines or well-defined Thalwegs. The most
important of those flowing to the northern side of the island
are the Grande Riviere, the Riviere Macouba, and the
Falaise.

The Precheur, on the west, is responsible for the de-
struction, on the 6th and 7th of May, of the village of
the same name, situated near its mouth. Farther south are
the Blanche, Seche, and Riviere des Peres, the last-named
limiting Saint Pierre on its farther side, and separating
it from the faubourg of Fonds-Core. The Roxelane alone
of Pelee's waters entered Saint Pierre. Built up in part
with walls, and surmounted by attractive gardens and villas,
it formed perhaps the most picturesque feature of the
city.

Of all these various waters, the Riviere Blanche, by
reason of its close association with the crater of the volcano,
has become the most noted in the later history of Mont
Pelee. It was into the channel of this stream that the
boiling mud from the £tang Sec, which wrecked the Usine

176 THE GEOGKAPHY OF MONT PELEE

Guerin on May 5, and brought the first casualties of the
eruption, was precipitated. In its upper part, and just
below the crater-basin, its course is directed through a
narrow and deeply-incised ravine, with steeply-sloping walls,
the abrupt contours of which, so far as the facts given to me
indicate, were fashioned subsequent to the events of early
May. This whole slope of the volcano, which is ragged
and torn through many eruptions, is wildly terrifying in
aspect, and much of it is sprinkled with boulders which
have latterly been shot out from the forming cone, or rolled
down on its outer face.

The great feature of Mont Pelee that has been accent-
uated as the result of its recent activities is the crater, whose
caldron lies southwest and west of the summit of the moun-
tain, and rises up directly under the lee of the Morne de La
Croix. The steep face of the Morne plunges into it at an
angle of seventy degrees or more. The feature is not an
entirely new one, as it is clear from the topographic descrip-
tion of the region given by the commission of 1851, and
from observation made just before the main cataclysm of
this year, that a soufriere or crater-basin existed on the site
of the present one already at the time of its earlier erup-
tion. This is the basin of the Etang Sec, situated as nearly
as I can state it from an approximate eye-measurement
taken at an elevation of sixteen hundred feet, and from the
observations of others, about twenty-four hundred to twenty-
six hundred feet high on the southwestern slope of the
mountain, and consequently from fifteen hundred to sixteen
hundred feet below the lowest part of the rim of the crater-

THE GEOGRAPHY OF MONT PELEE 177

wall on the east side (the plateau surface of the Lac des
Palmistes).

The altered condition of the mountain, combined with
the vagueness of past descriptions, has made it difficult to
recognize the exact topographic features as they had been
previously determined, and which appear to have been
known to the inhabitants of Saint Pierre and of the imme-
diate surroundings almost alone.* Hence many errors have
crept into the descriptions that have already been given.
It was my good fortune to have with me as companion on
my partial ascent of the volcano on August 24 one who was
thoroughly familiar with the old mountain, and could with-
out hesitation locate the main features in a comparison with
the old topography. From our point of observation at the
head of the arete which abuts against the caldron that
contains the source of the Riviere Seche, at an elevation of
sixteen hundred feet, and directly in face of the mouth of
the old crater, we had a clear view of the summit of the
mountain, of the great central cone, and of the deep gorge
of the Riviere Blanche, where it issues from below the
lower lip of the crater. The relations of all the different
parts were thus made clear to us.

The most accurate description of the crater that has been

* The lack of precise knowledge regarding the points of Mont
Pelee is well shown by the narrative of the " guide," Julien Eomain,
contained in Les Colonies in the issue of May 5, which places the
Morne de La Croix on the western side of the crater-basin, and the
fitang Plein (Lac des Palmistes) still farther west of the Morne de La
Croix I

178 THE GEOGKAPHY OF MONT PELEE

published, so far as I know, is that of Dr. E. O. Hovey,
contained in his Preliminary Report on the Martinique
and St. Vincent Eruptions, and which appears in the Bulle-
tin of the American Museum of Natural History (October
11, 1902). The author there correctly recognizes a circum-
valent rising valley partially surrounding a central or sub-
central active cone of cinders, and bounded on all sides
from the southeast to the northwest by high and precipitous
walls, composed of ancient tuff-agglomerates and lava-beds,
which culminate in the impending andesitic mass of the
Morne de La Croix. This great encircling wall, which falls
on the southwest to an elevation of about thirty-two hun-
dred feet, and in some parts is retained as an acutely narrow
ridge, reminding me forcibly of the encircling wall of the
Nevado de Toluca, in Mexico, is perhaps a little more than
a mile in length. Dr. Hovey estimated the width of the
entire basin at its summit to be about half a mile, which, I
believe, cannot be far from the truth, and agrees well with
the earlier measurements that had been made for the cuvette
of the Etang Sec. When I first reached the rim of the
crater on June 1 the opposing wall appeared to be only a
few hundred feet distant, but we could not at that time,
owing to the surcharging of the crater with steam and ash,
recognize that this wall was part of a central cone. The
steam-cloud was rising in vast swirls from the eastern
floor of the caldron, and obscured everything (excepting in
fleeting vistas) but the immediate foreground.

The crater pitches steeply downward from the northeast
to the southwest, conformably with the line of its axis. Its

THE GEOGKAPHY OF MOJSIT PELEE

179

northwest boundary is emphasized by the prominent lava-
mass of the Ti-Bolhommes, whose finger-pinnacle, rising to
perhaps four thousand feet elevation above the sea — a veri-
table " devil's thumb" — is a marked figure in the summit
landscape of the volcano. The present form of this moun-

Photo. Heilprin

PELEE SMOKING FROM FRAGMENTAL CONE

tain seems to have considerably changed since the days
before the eruption of May 8, as I cannot find on any of
the earlier photographs a form corresponding in outline
with that which now appears ; and my associate from Fort-
de-France assured me that the dent summit had been very
greatly developed. It may be that it was formerly covered

180 THE GEOGKAPHY OF MONT PELEE

in part by agglomerates, whose removal has now exposed
a greater and sharper surface.

The active central cone of the crater, a vast accumula-
tion of fragmented rock, cinders and ashes, which has built
itself up since the latter days of April, occupies basally
almost the entire basin of the ancient Etang Sec. It gives
the appearance of being a vast talus heap, and its steep
slopes, of thirty to forty degrees or more, were at the time
of our studies on August 24 being constantly carved by
trailing masses of rock and boulders. These ejected blocks,
some of which could hardly have been less than from fif-
teen to twenty feet across, were being hurled down to the
head of the gorge of the Riviere Blanche, and they swept
vast clouds of dust into the steam that was issuing in puffs
from nearly all points of the surface. The whole cone
was at times in ferment, and it was manifest that steam was
being blown or forced through the entire thickness of wall.

The great steam pennant of the volcano, however, issued
directly from the absolute summit of this cone, and clearly
located the position of the open chimney. As it emerged
at the top it covered practically the full summit of the cone,
and rose almost continuously with the outer slopes. It was
thus made impossible to determine, even approximately,
the dimensions of the actual opening, the steam manifestly
escaping through a part of the summit cap itself. But the
width of the main centrally-ascending steam-column could
not well have been less than four hundred or five hundred
feet, which is what I roughly estimated to be the diameter
of the summit of the cone. The spectacle of Mont Pelee

Photo. Heilprin

THE MAJESTY OF PELEE'S INFERNO
The crater from the crater-rim — June i, 1902

THE GEOGKAPHY OF MONT PELEE 181

smoking away from its beautifully-defined chimney-pot was
a superb one, wholly different from anything that I had
seen on my first visit. Measured by the eye, and on many
excellent photographs which I was fortunate in taking, I
should say that the summit of the cone at the time of our
observations rose to the full height of the Morne de La
Croix, which it hid from view, and, therefore, considerably
above the general summit of the mountain.17 It overtopped
the pinnacle of the Ti-Bolhommes by at least one hundred
and fifty to two hundred feet. Its full constructed height,
measured from the lowest point of its talus base, in the south-
west, can hardly be less than fifteen hundred or sixteen
hundred feet — all of it raised up since the latter days of
April. My ascent of the volcano on August 30, the day
of the great eruption, was largely for the purpose of more
accurately determining the measure of this giant cone, but
the conditions of the atmosphere on the summit were such
that the parts of the volcano could not be seen. No part
of the cone, nor of the Morne de La Croix, is in view from
Vive.

On August 24, when I attempted the ascent from the
side of Saint Pierre, two black " horns," one standing ver-
tically on the southeastern border, and the other projecting
horizontally, were clearly discernible with a glass to border
the crest of the cone. They may have been segregated
masses of cinders, fused in such a way as to present the
appearance of compact rock. It is possible that to their
incandescence was due the two fiery lights on the crown
which we observed in the early morning of the 22d, when

182 THE GEOGRAPHY OF MONT PELEE

the Fontabelle closely skirted the coast. This is merely a
surmise, as I had no means of ascertaining if, in fact, the
"horns" were incandescent; but a similar association be-
tween summit glows and incandescent knobs had been re-
marked in the earlier days of the active volcano.

The wholly accordant observations of Landes, Roux
and others, which I have elsewhere considered (Chapters
III, IV and V), leave no room for doubt that the seat of
activity and destructivity on the 8th of May, and during
previous days, was the basin of the Etang Sec ; therefore,
the caldron on whose base is now implanted the great frag-
mental cone. It is also equally certain that the real open-
ing of this basin was on April 25, when a heavy ash and
steam-cloud was seen to issue from it ; but it can hardly be
doubted that a minor eruptivity, beyond the simple emission
of sulphurous and aqueous vapors, may have existed before
this time. Nothing is known of the size or characteristics
of the cone — if the present cone existed at all at the time —
when the eruption took place ; and therefore the allocation
of the destroying blast to a definite point in the basin,
whether to its floor or to the opening in a rising cone, re-
mains to a degree speculative in its value. It appears to
me most probable that the blast issued from the basal floor
of the basin, rather than from a constructing cone — a view
which is measurably well sustained by the condition oi
violent, one might truly say paroxysmal, activity whicl
this portion of the crater still maintains. I was witness
a rare exhibition of a violent eruption from this quarter oi
June 5, and on the following day occurred a still moi

THE GEOGKAPHY OF MONT PELEE 183

violent eruption, whose characteristics were absolutely par-
allel with those of the cataclysm of May 8, as observed by
the officers of the Pouyer-Quertier. The great puffs of yel-
low and brown-black " cauliflower" cloud are, indeed, much
more closely associated with the explosions from the floor of
the crater-basin than they are with the outflows from the
summit vent ; at least, this has been my observation, ex-
tended over many days. When directly abreast of the crater
on August 24, and viewing it from an elevation of from
thirteen hundred to sixteen hundred feet, the outbursts were
remarkably forcible, and left no doubt in my mind that they
were not secondary in their action — i.e., rising, as some
have supposed, from accumulated masses or heaps of heated
cinders, but direct from openings in the floor of the main
crater.18 The lower and upper discharges were always
clearly distinguishable from each other in their fundamental
characters, and the former were much more violent and
paroxysmal, and usually much more heavily charged with
ash. When thrown out in volume and with force they built
out a landscape of terrible magnificence, the yellow and
almost black whorls rising like huge cauliflower heads, with
amazing swiftness, and spreading out far and wide over the
mountain slope. When the outburst was accompanied with
much steam, the vertical column frequently expanded out
into domed and mushroom-shaped masses. The rapid
transformation of these forms, and the chaotic cloud-world
that swept round them, were bewildering. While we did
not have the opportunity to accurately measure the heights
to which these clouds ascended, I should say, judged by the

184 THE GEOGKAPHY OF MONT PELEE

eye alone, that they must have at times reached nearly, if
not quite, two miles. During the eruption of June 5, which
we witnessed from close range, the ashes must have been
flung to still greater heights.

In the heavier outbursts that we witnessed in the latter
days of August, almost the whole crater-basin was filled by
the materials of these basal discharges, and the vigor with
which the great bursts came up, and in independent col-
umns, makes it appear as if there were not only a single
opening on the floor of the caldron, but that several such
may have existed. On August 30, and for two days follow-
ing, the entire basin of the crater was " smoking'' continu-
ously in one vast united column ; and it was the energy of
this rapidly rising mass, rasping the surrounding walls of
the volcano, which was doubtless in great measure respon-
sible for the terrific noise which the volcano gave out. As
I have elsewhere stated, this steam-column was found to
rise from the border of the crater with an initial velocity of
from one and one-half to three miles per minute ; and it has
been suggested that the quantity of steam and vapor that
rose in any one particle of time may have been the equal of
the full quantity that was being thrown out from all the
steam-jets of the world collectively, including those of
steamboats, locomotives and all of the forms of steam-
engines. After a careful study of the territory that was
affected and the conditions of the smoke-clouds as they
issued from the crater-basin — the crowding oyer of the
vicious black and ruddy puffs to the side of Morne Rouge
— I think it not unlikely, despite the apparent contradic-

THE GEOGKAPHY OF MOOT PELEE 185

tion that is presented by the overtopping summit- vent, that
the destroying blast of August 30 also issued from the floor
of the basin, and perhaps from the same opening whence
issued the destroying cloud of May 8. On the other hand,
the fact that the destroying blast now for the first time
swept over the mountain, when the cone reached the level
of the full summit, naturally suggests an association with
this structure.19

In associating the present activity of Mont Pelee with
parts of the volcano that were concerned in the eruption
last preceding this one, we have as a basis for study and
comparison only the report of the Scientific Commission
of 1851, MM. Leprieur, Peyraud and Rufz. From this
report it is made clear that none of the existing vents had
part in the earlier eruption, which in itself appears to have
been hardly more than a warning, with a localized area of
destruction immediately about the explosive points. There
were at the time of the investigation of the commission
three active craterlets, two situated at an elevation, as deter-
mined barometrically, of eight hundred and eighty-three
metres, and the third, which was seemingly the largest,
although measuring only one and a half metres across, situ-
ated some distance farther down the slope. This is thought
to have been the seat of the ancient Soufriere.* The position
bore directly east of Precheur, from which it was distant

* " Mais nous voulions visiter encore un troisieme crater e que nous
voyions fumer aussi a quelques centaines de metres plus bas dans la meme
ravine, et qu'on nous disait avoir pour siege Vancienne Soufriere.'1 Erup-
tion du Volcan de la Montagne Pelee, p. 9 .

186 THE GEOGKAPHY OF MONT PELEE

seven kilometres in a direct line. The vent nearest to Saint
Pierre was distant ten kilometres from that city. These
several openings, which were found in a condition of semi-
activity on August 9, were located in a ravine of the
Riviere Claire, a northwestern or right-hand tributary to
the Riviere Blanche, and at positions which can probably
no longer be identified. The commission did not consider
them to be active points of the main eruption, but assumed
for these a considerable number of other craterlets lying in
an adjacent valley, and at positions whose general or medial
elevation above the sea is placed at eight hundred and six-
teen metres. These were found to be all dormant.

That none of the several points of activity or past-
activity that are here referred to are in any way identifiable
with the Etang Sec (the focus of the recent outburst) — a
correspondence which has generally and not unnaturally
been assumed — is thus plainly indicated by the geographical
position outside of the actual basin of the Riviere Blanche,
and in the further narrative of MM. Leprieur and Peyraud
(p. 16), which states that these investigators visited the old
lake-basin for the purpose of making additional observations
on what was assumed to be another and still more ancient
crater of the volcano (" Sans visiter V Etang Sec qui passe
pour un autre cratere plus ancien du volcan"). This is,
indeed, a very important statement, for it shows the eruptive
point or points of the volcano to have shifted their positions
since 1851 towards the side of Saint Pierre. Naturally,
this condition fastens an added degree of insecurity upon
the mountain. The lake, instead of being dry (as its name

THE GEOGRAPHY OF MONT PELEE 187

signifies — dry tarn), was found to contain considerable
water, the quantity of which was estimated to be about five
times that contained in the summit lake (the Lac des
Palmistes : " remplie au jour oil Us le visitaient par une
masse oVeau considerable et a leur estime cinq fois plus
grande que dans le lac superieur" page 16), an overcharge
which the guides attributed to an unusual fall of rain during
the past winter season. The elevation of the lake was
determined barometrically to be nine hundred and twenty-
one metres (three thousand and twenty-five feet), corres-
ponding closely with the level of the most elevated of the
craterlets which had been located in the more distant ravine
(* Ainsi cet e tang-sec se trouve presque a la meme elevation
que les bouches superieures du volcan placees dans une ravine
plus eloignee. Rien d'ailleurs netait change dans ces lieux
au dire des guides, on ne remarqua nifente, ni eboulement").
The belief that Mont Pelee has had but a single erup-
tion recorded in its history prior to the one of May, 1902,
a supposition that is universally held in Martinique, is erro-
neous, the volcano having passed through a moderate parox-
ysm on January 22, 1762. A fairly extended account of
this eruption is published in the Journal des Mines, of Paris
(Vol. Ill (1796), pp. 58-59 of Part xviii), as an annotation
to Dupuget's paper : " Coup-dJceil rapide sur la Physique gen-
erate et la Mineralogie des Antilles" and appears from the
notes of an eye-witness, Aquart, communicated to M. Du-
puget. Earthquakes and the emission of sulphurous odors
and vapors in considerable quantity were an accompaniment
of this eruption, whose disturbing seat was in a number of

188 THE GEOGEAPHY OF MONT PELEE

craterlets situated at an elevation of about five hundred
toises (three thousand feet — consequently, closely corre-
sponding with the altitudinal position of some of the vents
of the later eruptions). Much vegetation was burned or
singed, and a number of opossums were killed. It is said
that the earth was riddled with holes, and many sulphur
aspirators were opened. At a lower level of some five hun-
dred to six hundred paces distance there was a flow of hot
black water (mud ?). The account concludes with the sig-
nificant statement : " This ancient eruption of Mont Pelee
seems to have had its entire effect on the western side
[of the volcano]. That quarter is completely overturned
[wrecked] . . . whereas on the side opposite the surface is
less torn." *

There can be little doubt that this earliest recorded erup-
tion of Pelee was from a part of the mountain not far re-
moved from the position of the 1851 eruption, if, indeed, it
was not absolutely coincident with it. The general charac-
teristics of the two eruptions appear to have been identical.

* " IS explosion ancienne de la Montague Pelee parait avoir porte
tout son effort du cote de Vouest. Cette partie est entierement bouleversee.
. . . tandis que du cote oppose le terrain est moins brise" P. 59. The
year of this eruption is more generally given 1792, but there can
be no question that the event took place at an earlier day, as the
facts were communicated to Dupuget while he was on the island,
and his voyage was made in 1784-1786. This relation is well set
forth by Professor Mercalli iu the Atti della Societa Italiana di Scienze
Naturali of Milan, XLI, p. 313, 1902.

-I
I
m

2
O
3J

om
E o
g rn

[5:

« s
IS

s 2

O 33

3 m

J? H

« o
m

♦ *v>

XII

PftRE MARY, CURfi OF MORISTE ROUGE

On June 2, following our second ascent of Mont Pelee,
Mr. Leadbeater and I made our pilgrimage to Morne Rouge,
where the faithful Cure, Pere Mary, had done so much to
relieve the anxieties of his little flock, and to serve them
with the necessaries of life. There were still four hundred
remaining in the beautiful mountain village, which before
had counted nearly four thousand inhabitants, and for these
food had to be provided. We took with us letters that were
thought to help the Father in his good work, and which
conveyed to him blessings for the work that had already
been accomplished. It was no easy task to remain for
weeks at this post on the hillsides, directly facing Mont
Pelee, and not farther removed from it than by about two
miles. The forest-land about had been broken and singed,
and in his own little town the gardens showed where ashes
and cinders had fallen. The smoke of Pel£e towered up to
mountain heights, and covered with shadow the belfry of
the beautiful church which was the heart of Pere Mary.
The volcano's thunders rolled and broke, and bright sparks
showered into the sky the fire of the raging earth ; but the
good priest remained, unmoved by the dangers that sur-
rounded him. From his upper world he had looked down
upon fair Saint Pierre on the morning of its destruction, and

189

190 PEEE MAKY, CUEE OF MOENE EOUGE

had seen the black cloud roll out from the volcano, and cast
its death-mantle over the doomed city. He knew better
than others what was the veiled language of the burning
mountain, but to him this language stirred only sympathy
to the afflicted, and a heart to do good to all. Every day
hundreds came to him to ask and to be given, and every
night his fervent prayers asked that a deliverance be
granted them, for there was little of food remaining. For
many days one could see nearly to the end, and what was
left came out in pitiful morsels.

The aged Father, who was clad in his cassock when we
arrived, grasped us warmly by the hand and bade us wel-
come in the shelter of his plainly-boarded presbytery. He
read through the letters that we had brought for him, and
again gently welcoming us, invited us to break bread over a
bottle of wine. " This is the last bottle," he said, " but,
oh ! what matters that, we are nearly to the end of our
bread and meat. God be merciful !" With tears in his
eyes he turned to us and bade us partake. It was with
much difficulty that I could force myself to join in the light
repast, for it seemed like stealing the life-food from many,
but Pere Mary insisted. Fortunately, we had brought our
own lunch, and were thus able to leave on the table a fair
compensation for what we ate. A large part of the popula-
tion of Morne Rouge had already for days been living on
fruit and biscuits, and a very little of that ; and it was evi
dent that, unless more relief came in and at an early day,
starvation would be staring the poor people in the face.
We promised the Cure that we would try to expedite the

;

PEEE MAKY, CUEE OF MOENE EOUGE 191

transmission of relief from Fort-de-France, and at that he
felt happy.

After attending to the wants of some who had come to
him for assistance and advice, he led me through his church,
and up into the belfry portico, where he explained the
country at large and the part which the volcano had taken

Photo. Heilprin

ON THE ROAD TO MORNE ROUGE

in destroying it. It was a truly beautiful landscape, with
its rolling woodland, its scattered habitations and thatched
cottages, its gardens of palm and banana ; but it was easy
to see how much had been lost through the cindering of the
vegetation. Pere Mary directed my attention to a change
of contour in the upper part of Mont Pelee, where three

192 PEEE MAKY, CUEE OF MOENE EOUGE

tooth-like prominences showed a saw-edge rising over the
slope that looked southward from the summit. These were
not in existence before the fatal eruption, nor did they be-
come visible till many days afterwards. They were white,
and looked as though they had been burned out, but at
night-time they shone out with red fire, and made bril-
liantly luminant the crown of the volcano. With little
doubt these points of rock were the protuberances that we
had seen rising out from the crater at the time that we
made the second ascent (June 1).

The site of Morne Rouge, occupying the crest of the
long ridge which unites Mont Pelee with the contreforts of
the Pitons de Carbet, and with the circling heights of the
Morne Vert back of it and the projecting knob of the
Calebasse on the opposite side, is perhaps the loveliest in
the entire island of Martinique. From it the eye surveys
nearly all the forms of Martinique landscape — not crowded
together as to harass the mind, but opened out into charm-
ing vistas of receding lowland and gently undulating moun-
tain slopes — and gathers in the more distant waters of the
Atlantic with those of the blue Caribbean Sea. Northward
lie the forested gardens and orchards of Ajou pa-Bouillon,
and southward the equally verdant slopes of Fonds-Saint-
Denis, the Reduits and Mont Parnasse. The city itself
had in a way been one of opulence, for many of the
wealthier inhabitants of Saint Pierre sought here, at an
elevation of fourteen hundred feet, the cooling breezes of
summer, and the benefits of a healthy, even if somewhat
humid, climate — a relief from the oppressive heat of the

PEEE MAEY, CUEE OF MOENE EOUGE 193

western lowland. At the time of our visit it presented a

largely desolated appearance, most of the houses being

closed and giving no signs of the living. The families

that remained were unsettled and weary, undecided whether

to remain longer or to seek shelter elsewhere. Knowing

that we had been to the summit of the volcano, old and

young came to their garden-fences to gather from us such

information as might tend to determine them in their

course. " Was Pelee still active V " Was it still to be

feared ?" " Can we remain, or must we go ?" — these were

the interrogatories that were put to us by people who were

sad of heart and trembling for that which the morrow

might bring. I gave them such consoling words as the

conditions seemed to warrant, — alas ! they were not many,

— and they seemed pleased with this paltry relief. One old

man took my hand and pressed it gently to his lips, then,

turning to his family, said in low and sad words, " Heaven

be praised, we are still living." A bright boy of about

sixteen, who met us on the road near Ajoupa-Bouillon,

addressed me in English, and begged that I inform him for

the comfort of his aged and infirm mother. I asked him

for his residence, which was pointed out near by, and also

where he had acquired his mastery of the foreign tongue.

He replied, in the Lycee of Saint Pierre.

The road by which we had come from Vive to Morne

Rouge is one of the best in the island. It follows for some

distance the main Capot, and enters well into the heart of

the mountain country. Along it the traveller is treated to

an enchanting display of tropical vegetation of palms, tree-
is

194 PERE MAEY, CURE OF MORNE ROUGE

ferns and bamboos, of heliconias, melastomes and rubber-
trees, — of giant foresters, cased in cables and creepers,
holding out their naked branches as if* asking for food and
light ; of star-massed epiphytes and orchids, and great
bursts of scarlet and blue blossoms. We follow along deep
barrancas musical with their tumbling waters, and shrouded
beneath an almost impenetrable maze of foliage. The
silence of the woodland was most impressive. A few
lizards here and there slid along the tree-trunks, and occa-
sional blackbirds hurried across the open, but there was no
song or voice of any kind. The world of life was hushed
in the silence of the dreary solitude. A stray land-crab
edged its way across the open road to clear our path, but of
the once dreaded fer-de-lance we saw nothing.

How little did we think at the time of this first visit to
Morne Rouge that in a few short weeks the town would
cease to exist, and that with it would pass the good Father,
whose ennobling life, consecrated to charity and humanity,
had set an example to be followed by the world. At nine
o'clock, or a few minutes after nine, of the evening of Au-
gust 30 a tornadic blast, similar to that which had destroyed
Saint Pierre, swept over the crest of Pelee, and in hardly
more than a minute, perhaps even in less time, Morne
Rouge was swept from existence and burning. I was at the
time at the Habitation Leyritz, on the northeastern foot of
the mountain, about five miles distant, watching the extra-
ordinary electric display immediately overhead. A lighl
shower of lapilli and ashes was falling, but not sufficiently
to obscure the night. The volcano continued roaring as il

PEEE MARY, CURE OF MORNE ROUGE 195

had done during the whole of the day, and most of the day-
previous, but there were no distinctive detonations audible.
Of a sudden, a great red glow shot high into the sky and
told that something had happened. Morne Rouge had been
shattered, and much of what remained was burning.

When the blast first swept furiously through the town
Pere Mary was in the presbytery, and it was only when this
home was aflame and no longer habitable did he seek the
shelter of the adjoining church. It was while going the
few paces from the one building to the other that he was
stricken — burned like the other poor creatures who were
either dead or dying. With a strong effort Pere Mary suc-
ceeded in dragging himself into his dear sanctum, where
he was found the following morning, suffering in agony, yet
sufficiently composed to ask after the welfare of his little
flock. Nearly or quite twelve hundred of these had already
perished. A number of small houses and the church were
what remained intact of Morne Rouge. The wounded Cu-
rate was removed to Fonds- Saint-Denis and thence to Fort-
de-France, where, at eleven o'clock on Monday morning,
September 1, he expired. On the following day Vicar-
General Parel delivered the funeral address, and the capital
of Martinique paid homage to the man whose name was in
the future to be a part of its history.

The unwavering heroism which bore Pere Mary to his
life-saving task carried with it no reward beyond that which
is reaped from a consciousness of having done the highest
work of man. But for his presence, and his kindly and
soothing words, serious panics would have repeatedly broken

196

PEEE MAEY, CUEE OF MOENE EOUGE

out in the northern part of the island, for it was not Morne
Rouge alone that his voice touched, but other settlements
were guided by it in their thoughts and action. " He very
courageously remains practically alone in Morne Rouge/ !
writes Vicar-General Parel to the Bishop of the diocese,
" beneath the jaws of the monster and under the guidance
of Notre Dame de la Delivrande. I wrote to congratulate
him, but there was no longer a postal connection. If he
succumbed, he will only learn in heaven that we admire
him.,,

xni

CLOUDS OF PASSAGE

On the morning of June 6, when events had shaped
themselves to a fairly peaceful turn, Fort-de-France was
again thrown into panic by the cries of " le volcan ! le
volcan /" Men and women, with tiny children clinging to
them and cursing the day that brought them misery, were
running wildly about, homeward or outward, according to
the degree of fear that had taken possession of their minds.
Others of more mature thought were anxiously watching
from street corners, while not a few were invoking the
aid of heaven through prayer and lamentation. That
which had given cause for this excitement was a new
and vigorous outbreak of Mont Pelee. Rushing from the
hotel to the street, I observed the sky darkened by a vast
cloud that was drawing over it. It advanced with bewilder-
ing velocity, spreading out like a giant fan as it propelled
its way southward, and in a few seconds the whole of Fort-
de-France, and all of the island that lay beyond, were in
shadow. The twilight of an eclipse had settled over us.
The spectacle of this advancing ash-cloud, like a huge oc-
topus overspreading everything, was supremely beautiful,
almost overwhelming, and for a moment we were lost to the
portentous secret that it carried. The gaze of everyone
was directed upward — watching, hoping, fearing. The end
appeared to have arrived for some. The cloud came in

197

198 CLOUDS OF PASSAGE

silence, and it followed its course in the same way. Except
where illumined by the sun into a dazzling white border,
its color was a cold and forbidding gray-black.

What had happened ? was the question that was thought
and asked by everyone. So large an ash-cloud had not
been known since the fatal 8th of May, and many seemed
to think that not even then was there anything comparable.
I roughly estimated the height of its course to be not less
than five miles above us, and it may have been more. The
normal air-clouds were then swiftly flying in the opposite
direction, heading for the volcano, and giving the appear-
ance of being attracted by it. Their plane was far below
that of the clouds reaching over them. An ashen pallor
hung over the capital for five hours after which the city
again emerged into a dim sunlight.

Having acquired some local reputation as a vulcanolo-
gist through my ascents of Mont Pelee, I was besieged for
an "opinion" on this new manifestation of the volcano's
activity, and the probabilities or possibilities that were
follow as a consequence. Many poor souls had led thei
selves to believe that the day of judgment had finalb
arrived, or was at least in sight. Others, more moderat
in their measure of the impending catastrophe, only ij
quired if there still remained time to pack and leave,
made an effort, and perhaps with some success, to alL
.their fears by unconcernedly pointing my camera to t]
sky, but the collecting crowd became uncomfortably lar^
and I moved on. At this time the cry came along that tl
sea was rising, and this gave cause for additional alari

CLOUDS OF PASSAGE 199

A similar occurrence was remembered as part of the phe-
nomena of May 8, and it was thought that possibly that
event had been repeated. The ocean-level had, in truth,
risen three feet, and the city could spare but little more.
Fortunately there were no bad results following this rise,
and the waters fell just as rapidly as they had come. The
day was almost exactly one month after the destruction of
Saint Pierre, and many wise people had prophesied that a
second destruction, similar to the first one, would take place
at this time. Here was the confirmation.

Later in the day, as the gray tint of the city wore off,
the people regained part of their confidence and once more
settled down to the quiet of their normal existence. Only
on the ocean-front were crowds still assembled, looking at
the ashen cloud as it floated off to St. Lucia, and at the
peculiarly colored steel-blue sea that lay beneath it. To-
wards evening, by invitation of one of the officers of the
French Cable Company, I visited the cable-ship Pouyer-
Quertier, which had again been out grappling for the lost
cable, The commander, Captain Thirion, kindly supplied
me with some of the details of the explosion, which were
subsequently embodied in an official report, and with
a number of the larger fragments of pumice with which
the vessel had been bombarded. The decks were still
coated with ash, although in no great quantity. At ten
o'clock, when the activity of Pelee became specially notice-
able, the Pouyer-Quertier was out six miles from land, a
little to the north of Saint Pierre. At that time a lofty
ash-cloud was seen to issue from the summit crater, and

200 CLOUDS OF PASSAGE

almost simultaneously a thick black cloud rolled down
the valley of the Riviere Blanche and forced its way to
the sea.

The aspect and manner of the volcano were precisely
what had been noted on May 8. The black cloud kept
close to the surface of the earth, rolling vertically, and
descending with a velocity which, if not as great as that of
the cloud of May 8, was swift enough. It made a course
of five miles over the sea, and accomplished its full flight
of seven miles in twenty minutes. Its dissolution on the
surface of the sea appears to have been accomplished with-
out explosion or ignition of any kind. The Pouyer-Quer-
tier had by this time sought safety in flight, for it had been
approached to within a mile by the threatening cloud, and
the fall of cinders and ash was heavy.

The eruption was noted by a slight barometric fall, but
it may be that the fluctuation of the recording needle was
the result of a jar or seismic movement, rather than of a
true atmospheric displacement. The fluctuation was four
millimetres. The following is the official account (transla-
tion) of the eruption as it was posted in the cable office of
Fort-de-France on the following day :

Fort-de-France, 7th. — During the eruption which occurred yes-
terday morning the French Cable Company's repairing steamer
Pouyer-Quertier was five miles off the mouth of the river Blanche
grappling for the cable. At the moment a thick cloud appeared
unfolding itself vertically and horizontally at the same time. This
cloudy mass was as large as the one observed during the eruption on
the 8th ultimo. Lightning was to be seen constantly flashing from

CLOUDS OF PASSAGE 201

the sea towards the sky. The lower part of the cloud travelled very
rapidly towards the southwest. In less than twenty minutes it had
travelled seven miles ; then the volumes became larger and the sky
darkened all over. A strong smell of sulphur spread while a thick
shower of stones and mud fell on the Pouyer-Quertier. During this
phenomenon the barometer fell suddenly from 765 to 761.5 and then
rose to 766. It was with great difficulty that the Pouyer-Quertier suc-
ceeded in escaping from the danger of asphyxia and fire.

The record of the Pouyer-Quertier made it clear that
the explosion of the morning was a serious affair, the full
measure of which could not be had from a study of the
magnificent ash-cloud alone. On the day preceding the
occurrence we were on a tour of survey along the northern
coast of the island, and had then an opportunity to study a
new phase of the volcano's activity. We had chartered a
small steamer, the Rubis, to take us to Grande-Riviere and
Macouba, and on our return landed in front of the great
mud-flow which had overwhelmed the Usine Guerin and
now lay like a huge glacial tongue between the Rivieres
Blanche and Seche. At the point where we landed the
flow had been fissured out by the sea, and here, as at several
other points, gave out strong jets of steam, with a feebly
sulphurous odor. I coaxed some of these with a stick, but
found them harmless. The whole mud-flat was still densely
steamed within, but the surface had hardened and largely
crumbled into dust, and gave out but little gaseous vapor.
We followed in the line of the crater, keeping close to the
course of the Riviere Blanche for about three-quarters of
a mile, when the conditions of the weather became such

202 CLOUDS OF PASSAGE

that it was thought prudent to return. Our observations,
although we obtained a clearer view of the upper gorge
of the Riviere Blanche than we had had up to that time,
revealed little of consequence. We found the Riviere
Blanche muddy and turbulent, but not hot, and steam was
issuing from it only where heated parts of the bed or
banks were newly exposed by undercutting, and permitted
of the rapid formation or evolution of vapor. At such
points great puffs were being forcibly blown off, and their
periodicity in action easily suggested the presence of true
fumaroles.

We had leisurely made the descent of the mud-flat and
barely began to pull our dinkey from the shore, when the
cry went up, " look at the volcano !" We turned our eyes
in the direction of Pelee, and the sight that met them was
truly terrifying. The crater, whose peaceful method had
lured us to a farther point than was perhaps entirely safe,
had suddenly broken into eruption, and was hurling wild
sheets of yellow cloud to the northward. They came roll-
ing and puffing with great fury, and in an instant almost
the whole northern face of the mountain was in turmoil.
Over every slope the smoke was shifting and rising, here
and there in sweeping threads, as. if tossed from a prairie
fire, elsewhere twirling itself into lofty columns and pyra-
mids or mushroom caps — rolling black and yellow with
the angry ashes that were being carried out by them.
Five minutes before we had been walking about in the sun-
shine of this same mountain slope, not fearing that any-
thing could disturb our mission; and now the mountain

CLOUDS OF PASSAGE 203

was lashed in fury, and buried in the dark shadows which
its new life had called forth. The scene was an extra-
ordinary one, and one that was made doubly impressive by
the rapidity with which it was brought about.

Turning our eyes to the channel of the Riviere Blanche,
we found the stream a seething mass of chocolate and black
mud. It came down dancing and boiling, a steaming cata-
ract that had been shot to the sea. In hardly more or less
than three minutes, it had leaped its two miles from the
crater to the sea, which it entered with what seemed to be a
vertical wall. Its surface rose and fell in angry billows, and
great vertical jets were tumbled out of it like fairy geysers.
We could hardly convince ourselves that this raging and
burning mass, roaring in its fury, and turning the sea into
angry eddies, could be the same water which we had so
leisurely followed for three-quarters of a mile on a bank that
was now no longer visible.

I think it beyond question that the increased volume
and change in the character of the water of the Riviere
Blanche were directly the result of a discharge of mud or
water from the volcano. There was no rain of consequence
that could have been thought to be responsible for this
condition, nor was the volcano itself capped with any
particularly heavy resting cloud, a strong southerly wind
dressing the mountain to the north. We found later in
the day that the Riviere Seche was also running with hot
mud.

The eruption which came thus suddenly was evidently
the prelude to that of the day following, when certain feat-

204 CLOUDS OF PASSAGE

ures, like that of the great rolling black cloud, were pre-
sented which were not developed in the earlier affair. It
was a magnificent demonstration of the power that remained
with Mont Pelee, and its effect was not calculated to give
comfort to those who had persuaded themselves, contrary
perhaps to their own feelings, to believe that there was
nothing more to fear from the volcano. But Martinique
had been favored this time. The black cloud, which had
merely left a shadow to follow its swift flight across the
island, dropped its ashes on St. Lucia, and veiled its land-
scape with the darkness of night. At one o'clock in the
afternoon the royal mail steamer entered the port of Cas-
tries with its search-lights directing the course.

Before making this survey of the Riviere Blanche flow
we had landed at Precheur, the town which had first felt
the overpowering gloom of the impending catastrophe of
May 8, and whose heart had been eaten out by torrents of
rock and boulder. For several days before the destruction
of Saint Pierre many of its inhabitants had fled thither,
thinking to secure a safer refuge in the larger city. Its own
atmosphere had been blacked with the falling ashes, and
made poisonous with the vapor of sulphur. The Precheur
River had risen over its banks, swept off scores of habita-
tions, and deluged the city with acres of boulders. The
settlement practically ceased to exist. Most of the destruc-
tion had, indeed, been accomplished before the event which
wrecked the city that was sending out aid.

The following correspondence published as a letter from
M. Duno-Emile Josse, and appearing in the issue of Les

Photo. Heilprin

A DELUGE OF BOULDERS
Basse-Pointe, May 29, 190a

CLOUDS OF PASSAGE 205

Colonies for April 26, gives a vivid picture of the first
coming of the storm :

Grand-Fonds (Precheur), April 25, 1902.
The Editor of Les Colonies, —

Permit me to ask a column of your esteemed journal in which
to recite to the public a curious phenomenon which has surprised the
inhabitants of our quarter of Mont Pelee.

Since more than three months we have felt the odor of sulphur
which has caused considerable disquietude with the inhabitants, and
finally led to the recognition that it came from the Soufriere. It
increased steadily in force and quantity, and threw us into great fear,
as the older inhabitants well recalled the Soufriere in the eruption of
1852 [1851]. In fact, at about eight o'clock of Wednesday morning,
April 23, the earth trembled with a sufficiently severe shock. Yester-
day, the 25th, at about the same time, it was noted that the atmos-
phere was darkening, and almost immediately it turned as if into an
eclipse of the sun, accompanied by a deep growling (grondement). . . .
All of a sudden, a lbud detonation, like the firing of a cannon, was
heard; the sky appeared to be in places on fire, and there was a
continuous fall of fine and white ashes which the volcano was vomit-
ing out, and which spread more particularly over the Grande-Savane,
Grand-Fonds, Bois-Lezard, and Gros-Morne. These ashes were so
abundant that at two metres distance people were unable to dis-
tinguish one another.

The affrighted inhabitants, snatching with them their children
and valuables, ran bewildered, as if stricken with blindness, to the
right and to the left, returning to their houses, crying, praying, and
at the same time asking assistance from near neighbors, who, them-
selves paralyzed by fear, were unable to respond to the appeals of
their co-citizens. This condition remained for over an hour before a
calm again settled. The rain of ashes lasted for about two hours.

Unfortunately [in an effort to reach the Soufriere] as we came
within a certain distance of it, we were obliged to return. We were

206 CLOUDS OF PASSAGE

in face of a steam-cloud that could be likened to that of several high-
pressure furnaces [hauts fourneaux] united, and which was in part
white, in other parts black.

Duno-Emile Josse.

Our visit to Precheur on June 5 was to a deserted
settlement. A handful of people remained, to guard and
watch over the few houses that persisted along the ocean
front, away from the distributing course of the Riviere.
Mud and boulders lay three and four feet deep over the
floors of habitations that still carried their walls and roofs,
or clustered about tree-trunks and bushes that formed part
of the square and of the savane. The church rests with
only its half on solid ground. We went to the old bed of
the Precheur, and found its course occupied by a huge
tongue of land packed on its surface with a wilderness of
giant rock-boulders. The sight was an . astonishing one.
Far up in the line of the former meadows were acres upon
acres of strewn rock, packed like cobble-stones in a street-
paving. One could say that nearly all the boulders were
large — many of them five feet and considerably more in
greatest diameter, others measuring eight and ten feet. I
observed several that were twelve feet in length. With
few exceptions, all had rounded contours, a character im-
pressed upon them by ancient erosions ; and the greater
number were encased in a heavy coating of dried mud, two
or three inches in thickness. The material was chiefly
andesite and basalt or diorite.

We know that this rock avalanche was brought down
by the stream from the flank of the volcano, but from what

CLOUDS OF PASSAGE 207

precise form of deposit ? Before this I had been at Basse-
Pointe and noted nearly the same conditions — a town, the
greater part of which had been annihilated by its inun-
dating torrent of boulders; and later, on August 27, I
studied the effects of the flood- waters of the Falaise, im-
mediately above the confluence of that stream with the
Capot. In a flow of hardly three miles, possibly consider-
ably less, the stream had brought down a veritable wilder-
ness of giant rounded and planed boulders. I determined
the position of high-flood to have been thirty-five feet or
more above the normal level of the stream. A large
boulder was at this time still lodged in the crown of a tree
standing at the mouth of the Capot. A broad flood-plain
had been excavated since my earlier visit, and great em-
bankments were built up to passing levels of the water.
A truly wonderful side-piece to the phenomena of Pelee !

XIV
A SECOND VISIT TO MABTINTQUE

I visited Martinique for the second time in the latter
days of August, and found a greatly improved local condi-
tion then existing. The large number of refugees who had
been quartered and fed in Fort-de-France were back in
their homes in Morne Rouge, Ajoupa-Bouillon, Grande-
Riviere and Macouba, and but a lingering few remained
behind. The repatriation of the deserted country had been
ordered by the government, and the government had been
advised by a scientific commission. It is true that a violent
eruption of Pelee had taken place as late as July 9, but
this had now been passed by more than a month, and even
at the time of its coming it brought little anxiety to the
inhabitants. Those who before had been most frightened
were eloquent in their description of the wonderful electric
displays, while the phenomena of eruptions generally were
being discussed with absolute sang-froid and had come to
be regarded by many as curiosities of nature, to be studied
and admired, rather than of necessity to be feared.

The streets had much life returned to them, the service
in the hotels had been improved, and nearly all the visiting
foreigners had left for distant shores. A new Governor had
been installed in office, and the Government House was
busy with officials, couriers and attendants of all kinds. A
new journal, borrowing the name of La Colonie from its

208

A SECOND VISIT TO MABTINIQUE 209

unfortunate predecessor in Saint Pierre, Les Colonies, and
appearing three or four times a week, had been established.

The battle of politics was again being hotly waged, and
with the same intensity and personal feeling as at the time
of the destruction of Saint Pierre. It seemed as if the
lesson of the volcano had already been entirely forgotten.

I took my quarters, as before, at the Hotel Ivanes, over-
looking the Savane. Great changes had been reported from
Saint Pierre. The streets were represented to be largely
filled up with ash and lapilli, and the walls hardly recog-
nizable, having rotted and fallen. Going out the following
day, I did not find the change nearly so marked as it had
been reported, though a few of the old landmarks were with
difficulty recognizable. Lapilli and ash had fallen in con-
siderable quantity, so that many of the older rubble-heaps
were masked, and the streets levelled out into passable road-
ways. The Rue Victor Hugo was open to wagons over
part of its course and was being travelled by a cart at the
time of my visit. The ash-covering had deepened in places
to four and five feet, and the surface was almost everywhere
made up of loosely aggregated lapilli.

Rains, excavations and lootings had exposed many ob-
jects that had before been buried, and skeletons, or parts
of skeletons, were freely lying about. We found many
skulls scattered over the Place Bertin and in the avenues
leading out from that section of the city. The northern
part of the Centre or Mouillage was heavily cloaked in ash.
At a few places some few signs of a systematic excava-
tion after treasure were noticeable, but in general the aspect

210 A SECOND VISIT TO MAETINIQUE

of the destroyed city was the same as when I had left it,
except that here and there patches of green were beginning
to appear, and clumps of new banana-trees and cane, im-
planted upon the old gardens or the decay of roof-tops, had
risen to three and four feet, and other bits of vegetation
were noticed elsewhere. It was evident that, left to itself,
the desert would soon be covered by tropical vegetation. I
do not think it can be disputed, despite some seemingly con-
tradictory analyses that have been made or reported, that
the ash contains much fertilizing principle, and was wholly
to the advantage of vegetable growth. In the interior of the
island we found the vegetation surpassingly rich and ver-
dant, and this was particularly noticeable in the districts of
ash-fall, so much so as to attract the general attention of
the cane-growers. On the lower flanks of Pelee itself the
cindered and burnt forest was breaking out into brilliant
green.

On Sunday, August 24, Fort-de-France had its first
earthquake. It came at nine-twenty in the morning and
caused considerable consternation. The city had thus far
been free from this form of disturbance, and not even on
the fatal 8th of May did it experience a shock. This first
manifestation of seismic activity very properly brought a
new fear to the inhabitants and for a moment placed politics
in the background. Clocks had stopped, tables had tilted,
and doors had opened ; crockery here and there fell from
shelving, and ceilings swayed as if suspended in free air.
The evidence pointed to a markedly horizontal concussion,
but no one seemed able to distinctly locate the quarter of

A SECOND VISIT TO MAETINIQUE 211

first impact. The Meteorological Observatory of Fort-de-
France registered the oscillations as northwest to southeast,
with a total duration of twenty seconds. I was at the time
in Carbet, just south of Saint Pierre, and experienced
nothing ; and so far as I know, no one in Carbet felt more
or knew until long afterwards that an earthquake had taken
place. Whether this earthquake was in any way associated
with the renewed activity of Mont Pelee, which developed
shortly after, can hardly be told, although this condition
of dependence naturally suggests itself. The weather at
the time of the occurrence was superb, and the barometer
indicated no disturbance.

I had with me on this day a small party assisting in the
study of the southwestern slope of Pelee. My purpose was
to determine accurately the positions of the parts of the
mountain as they had been known before the May cata-
clysm, and to follow the development of the newly-formed
fragmental cone, which had grown to prodigious dimensions.
For the accomplishment of this purpose I had planned an
ascent quite to the summit of the volcano," but we were
baffled by the deep gorge of the Riviere Seche, which had
opened considerably, and whose nearly vertical sides seemed
to present an impassable barrier. Our course was over the
gentle ash-slope that forms the water-parting between the
Rivieres Seche and Des Peres, and continues in line with
the southeastern wall of the crater. Much new vegetation
had grown out from it.

We reached at our farthest point an elevation of ap-
proximately sixteen hundred feet, or two hundred feet

212 A SECOND VISIT TO MAETINIQUE

above the ridge-line of Morne Rouge. The crater, whose
lower lip was still six hundred or seven hundred feet above
us, opened out in our direction, and gave us a splendid
opportunity to observe the contours and development of the
active cone that was implanted upon its basal floor. The
summit of this cone, which was. smoking in the fashion of a
factory or locomotive chimney, rose apparently fully one
hundred and fifty to two hundred feet above the finger-
pinnacle of the Ti-Bolhommes, and completely obscured,
from our side, the Morne de La Croix. I should say that
it rose at the time to the full height of the mountain.
Great boulders were racing down its slope, and trailing
clouds of ash-dust after them, but there were no continuous
outflows, whether of mud or lava.

The volcano was not particularly active when we came
to our final point, although puffs of steam were issuing
freely from the walls of the cone. No eruption was notice-
able coming from the floor of the crater. Before an houi
had passed, however, angry ash-clouds began to blow u]
from below, following one another in rapid succession, an<
all passing off northwestward with the wind. For twc
hours or more they continued to rise, steadily increasing in
intensity, and unfolding in beautiful, cauliflower masses.
They shot out from the crater at moderately low angles, but
with great force, and before long the summit of the moun-
tain, northward of where we stood, was shrouded in
chaotic mass of floating and shifting vapor. The quantity
of ash carried was very large, and the clouds were forbid-
dingly dark in color — red, brown and almost black. At

A SECOND VISIT TO MAKTINIQUE 213

this time the summit vent was also blowing up with vigor
and it gave us a splendid opportunity to observe the courses
of the two forms of eruption, whose relations have already-
been discussed in an earlier chapter.

The illustrations which follow, reproductions from a
series of photographs which were taken at intervals of a
minute and less, vividly portray the developing eruption —
perhaps more rapidly and consecutively than has ever be-
fore been possible for an eruption of magnitude. The
scene of awe-inspiring grandeur which they depict is in-
describable.

Failing to reach the summit of the volcano from the
side of Saint Pierre, I again moved over to the northeast,
the side whence the earlier ascents were made, and once
more imposed myself upon the open-hearted hospitality of
our friends of Assier and Vive.

The wheels of the usines were again working, and the
great high chimneys were proudly curling their smoke over
the verdant greens of meadow-land and cane. The day
after my arrival, I went over with the representing Maire
of Ajoupa-Bouillon, M. Kloss, to inspect the large cacao
estates of that district, and to ascertain, if possible, the
nature of the so-called Trianon crater in the Falaise, a
short distance beyond, at the foot of Mont Calebasse.
Much has been written about this crater and its explo-
sions. I had described it also,* and characterized the
eruptions coming from it as being of the true crateral type.

* McClure's Magazine, August, 1902.

214 A SECOND YIS1T TO MAETINIQUE

When my observations were made I was not near enough to
clearly ascertain its features, and relied for my determination
largely upon the observations of others, those who had been
quite up to it, or, at least, looked in. A closer examination
of the gorge at a time when it was entirely free from vapor,
and with the assistance of one who well knew its topo-
graphic features before the great eruption, leads me very
strongly to doubt the crateral origin of the outbursts — a
doubt which has already been expressed by Lacroix and
others. Most of the topography of the gorge as it now
exists I am positively assured existed long before May 8,
and some or all of the forbidding caldron-like holes at the
head of the cirque are recognized as part of the old features.
While it is not impossible that minor craterlets may have
existed in the upper gorge of the Falaise, just as they did
in the side ravine of the Riviere Claire in August, 1851, I
prefer to believe that the explosions that I and others
noticed as coming from the gorge, and which I have
likened to puffs from locomotive chimneys, were due to a
secondary forcing of steam from accumulated masses of hot
ashes and cinders to which surface water in some way or
another had gained access. At any rate, it is a very sus-
picious circumstance that during the active period of the
volcano in the latter part of August and early September
the Falaise should have remained perfectly quiet, whereas
the Precheur and the Grande Riviere, over whose bed lapilli
and much hot ash had been deposited, were running wildly
with steam.

Immediately after my return from the Trianon I

A SECOND VISIT TO MAETINIQUE 215

planned for the ascent of Pelee, which was accomplished
on the 30th (August), on the evening of which day the
volcano broke into a paroxysm of wrath, and destroyed
much of the beautiful country which I had first travelled
over. Little did we imagine, as we wandered through the
dense thicket of forty thousand cacao plants, as we brushed
aside the hanging vines and creepers of the gorge of the
Little Capot, that before four days were to pass a gray
desolation would again sweep over the landscape — that the
harvest of death would so soon be gathered from the living
world. It so came to pass ; and even on that very night
the red fire-glow of the volcano was seen playing on the
crown.

XV
BATTLING WITH PELEE

Throughout the whole of Friday, August 29, Pelee
kept up a continuous growl. The sound came to us like
the rumbling of wagons crossing a bridge, and at times like
distant thunder. M. Louis des Grottes, our host at the
Habitation Leyritz, where we had been installed the evening
before, felt uneasy, and thought that many days might
elapse before an ascent of the mountain could properly be
attempted. On the evening of Monday preceding there had
been a wonderful exhibition of volcanic pyrotechnics, and
everybody spoke of the great " flames" that were seen to
shoot out from the crater, of the volcanic corona and of
scintillant stars ; and since then the volcano had been con-
tinually in unrest. Refugees were seeking the roads at all
points, and the north of the island was once more in the
condition that marked the early days of May. The Habita-
tion Leyritz lies on the northeastern foot of Mont Pelee, at
an elevation of about five hundred feet above the sea, and I
had selected it, on the invitation of M. des Grottes, as a base
of operations alternative to Ajoupa-Bouillon or Morne Balai,
it being more closely approached to the volcano than either
Assier or Vive. Its position is delightfully in the path of
the ocean breezes, and its stately cocoa-palms are only
four miles distant, a vol oVoiseau, from the active crater of
the volcano. When we arrived there shortly before sun-

216

BATTLING WITH PELEE 217

set the hour of rest had already been proclaimed to the
workers on the estate, and inquisitive groups of coolies and
dark Creoles lingered and loitered about, some chanting the
evening hymn. The little Martinique blackbirds whistled
out their beautiful and mellow notes until late in the even-
ing, and after that, except for the roar of the volcano, the
" silence of night" was left to the minstrelsy of the tree-
toads.

We were up some time before the rising sun, and saw
the day break fair, with a gentle breeze sweeping over the
tops of the nodding cane. A few bad clouds were chasing
after the eastern horizon, and others hung over the black
peaks of Carbet, but they went the right way for us, and
they augured well. The difficulties that attend starting in
the tropics delayed our departure until after six o'clock, an
hour that seemed early enough for the kind of day that
promised. An hour before that time it was still dark. At
Morne Balai, which we reached in half an hour, my little
party, consisting of Julian Cochrane and myself, with three
foot attendants, was joined by seven volunteers, who felt
that the spirit of the volcano had been controlled by us,
and believed that they could courageously follow in our
footsteps. One of these I had well known from my earlier
ascents, and he stood as prophet and informant to the
others, basing his superiority upon a very fragmentary
knowledge of English. Our purpose to study the great
cone that had so rapidly built itself up in the heart of the
crater was perhaps unknown to the joining party, but they
held their courage well throughout the greater part of the

218 BATTLING WITH PELEE

day. Alas ! poor souls, they little expected that the tongue
of the fiery dragon would visit their homes ere night had
fairly fallen, and bring sorrow and death to the heart of a
peaceful and quiet-loving community. When I last rode
through the garden-lanes of Morne Balai everything was
deserted — the gardens were empty and the doors of the
thatched cottages closed. New ashes had fallen since the
day of Saint Pierre, and the inhabitants lacked the courage
to remain. Life had now come back to the village, and
how beautiful this morning were the copses of banana, of
palm and breadfruit, the hedge-rows, and the great blazing
blossoms of the hibiscus ! A more charming village scene
could hardly be found.

Our course up Pelee was from this point the same that I
had taken on my previous ascents, over the easy arete that
forms the central eastern ray of the volcano, and lies a little
northward of the ravine of the Falaise. The conditions of
the ascent on this day were surprisingly favorable, and w
were able to make use of our animals up to a height of
nearly two thousand three hundred feet. A light growth of
grass had begun to cover the arid slope of ash and cinder,
and the blackened forest of the ravine slopes was also
touched on the crown with green. The beautiful tree-ferns,
more particularly, gave evidence of this new life, and they
promised to restore in a short time to Mont Pelee that
verdure for which the mountain had been dear to the Mar-
tiniquians. It was evident that the burned forest was not
absolutely dead, and its greens were already being picked
by troops of blackbirds, fly-catchers, and the hirondelle-

•.

BATTLING WITH PELEE 219

mouche. Myriads of green and green-and-black caterpillars
were cropping the new vegetation. They had found a com-
fortable home in this newly regenerated upper world, and
were making the best of their time. It was evident that
the volcano had blown to them a good wind. Such sudden
visitations of insects to recovering volcanic regions have
been noted before, and have brought many problems to the
entomologist which still await solution.

We left our animals shortly after eight o'clock, and at that
time the volcano was raging. The steam-cloud roared out
of a seething furnace and swept the summit from our view.
Back of us dark-blue shadows were checkering the receding
landscape, but the ocean was the blue and green of the
coral reef, and lovely Morne Rouge was bathed in warm
sunshine. Nearer to us Ajoupa-Bouillon, slumbering in
sunlight and shadow, lay almost at our feet. We picked
our way leisurely up the cinder slope, but it was evident
that ejected bombs had recently scarred its surface, for
there were furrows and troughs and great boulders where
none had been before. We also noted a number of the
puzzling crater-like shallow pits or hollows which some
have thought to associate with falling rocks, others with
earthquake phenomena. In a few minutes more we were
in the storm-cloud, with only bits of landscape to fol-
low us as companions. The great knob of Morne Jacob
appeared and disappeared, and at intervals we could glance
into the deep gorges on either side of us, but of the summit
of the mountain there was nothing. Our Martinique asso-
ciates were uneasy, for from the invisible gray ahead came

220 BATTLING WITH PELEE

the terrific voice of the volcano. There were no accentu-
ated detonations, but a continuous roar that was simply
appalling. I thought on ray previous ascent to have heard
something, but this time it was the old sound multiplied
a hundred-fold. No words can describe it. Were it pos-
sible to unite all the furnaces of the globe into a single
one, and to simultaneously let loose their blasts of steam, it
does not seem to me that such a sound could be produced.
It was not loud in the sense of a peal of thunder, but of
fiery and tempestuous storm, that could best be compared
with the blowing of the ocean's wind through the shrouds
of a full-rigged ship, only ten times that. The mountain
fairly quivered under its work, and it was perhaps not
wholly discreditable that some of us should have felt any-
thing but comfortable.

Where was all this ? we asked ourselves. In front of us,
but invisible. My aneroid gave for our elevation three
thousand four hundred feet — therefore we were only six
hundred feet below the summit-level which marked the
position of the Lac des Palmistes. There appeared to be
no barometric disturbance, nor was the compass-needle
affected. A whistling bomb flew past us at this time, but it
left but a comet's train in our ears, for it could not be seen.
We took it first for a flying bird, but its course was soon
followed by another, and then came the dull thud of its
explosion in air. Deep down the ravine we could hear
the scattered parts tumbling, sliding and crackling. We
could no longer deceive ourselves as to the character
of the struggle into which we had entered. The ominous

BATTLING WITH PELEE 221

clicks in the air told us what we might at any moment
expect.

We moved up slowly, hardly more than a few paces at
a time, but with hope given to us in the occasional rifting
of the clouds. Time and time again the summit crest ap-
peared beneath the rolling vapors, and it really seemed as

l'h,, t.>. licllprin

THE BOMB-SCARRED SLOPE OF PEL^E

if the cone, of which we were in search, would suddenly
come to view. When we had reached about three thousand
eight hundred feet the fusillade of bombs became over-
poweringly strong, and we were obliged to retreat. We
were in battle. The clouds had become lighter, and we
could at times see the bombs and boulders coursing through

222 BATTLING WITH PELEE

the air in parabolic curves and straight lines, driven and
shot out as if from a giant catapult. They whistled past
us on both sides, and our position became decidedly uncom-
fortable ; many of the fragments took almost direct paths,
and must have been shot into their courses as a result of
explosions taking place above the summit of the volcano.
They flew by us at close range. Descending perhaps one
hundred feet lower on the slope, we took shelter under a
somewhat rolling knob and waited for a possible cessation
of the fusillade. A glance at my men showed that they
were thoroughly frightened, and most of them were making
quick tracks to a lower level. A lull favored a further
effort. Not wishing to incur any responsibility in a call
for company in what appeared to be a rather hazardous
enterprise, I made a second attempt by myself, keeping
my body as close to the ground as was possible. The
clouds soon separated me from my associates, and all of
visible nature that was left to me was a patch of slope and
the shifting vapors. Mr. Cochrane's figure was the last to
disappear. The roar of the volcano was terrific — awful
beyond description. It felt as if the very earth were being
sawed in two. In about a quarter of an hour I reached a
point just below the summit — the crest of the old lake
basin — which was being heavily raked by the fire of the
volcano. I could see no more than before. Everything
was as if in a surging sea, and neither the cone nor what
was left of the Morne de La Croix was visible. I crouched
down to the ground, but to no purpose. It was useless to
remain longer in the open fire, and I descended to join my

BATTLING WITH PELEE 223

associates. Mr. Cochrane was near at hand, working his
camera and seemingly indifferent to the encircling storm,
but the negroes had gone far below, carrying our provisions
with them. I was surprised, indeed, that they should have
retained their courage for so long a time, for Pelee had been
unusually active for a number of days, and if men ever
feared anything, it was this grim monster of Martinique.
But most of them had remembered my earlier ascents, and
they childishly seemed to feel that there was shelter in my
wake.

Shortly before noon a sudden lifting of the clouds re-
vealed the volcano in all its majestic fury. For the first
time since we reached its slopes were we permitted to see its
steam-column — that furious, swirling mass ahead of us, tow-
ering miles above the summit, and sweeping up in curls and
festoons of white, yellow and almost black. It boiled with
ash. The majestic cauliflower clouds rose on all sides, join-
ing with the central column, and it was evident that the
entire crater was working, bottom as well as summit, and
with a vigor that it would be useless to attempt to describe.
Higher and higher they mount, until the whole is lost in
the great leaden umbrella which seemed to overspread the
whole earth. I estimated the diameter of the column as it
left the crest of the mountain to be not less than fifteen
hundred feet, and its rate of ascent from one and a half to
two miles a minute, and considerably greater at the initial
moment of every new eruption. Great exploding puffs
were following one another in rapid succession, and they
told the story of what was going on inside the volcano.

224 BATTLING WITH PELEE

Cochrane and I were not the only ones to be inspired
by this extraordinary and bewildering spectacle. Our
Martinique men seemed equally overcome by a grandeur
of nature, terrifying as it was beautiful, which they had not
before seen, and of their own accord initiated a new effort
to reach the summit. We climbed back to our former posi-
tion, but the bombardment was too strong for us, and we
thought best to desist. The prospects for study were any-
thing but promising, and it was thought unnecessary at
this time to take further risks. Of our party of twelve
there were now only four left on the upper slopes of the
volcano, but we still hoped for one more chance. For a
half hour or so we took refuge in a hollow sufficiently deep
to about clear our heads, and waited. But even the pleas-
ures of a mountain lunch did not quite make this place
restful, for the bursting bombs flew thick to one side, and
we were too eager to watch the flying fragments to permit
ourselves a free moment. Every scattering mass brought
us to our feet, only to see and hear the fragments plunging
into the abyss that lay to one side. Cochrane and I moved
a piece higher up, and then abandoned the effort. " Where
did this last block burst?" I asked of my associate, and
before my question was answered we were spattered with
mud from head to foot by a great boulder, hardly smallei
than a flour-barrel, which fell within ten feet of us,
less.

When we reached the lower slopes we were covered
with ash and mud. For an hour or more we were nearly
beneath the centre of the great ash-cloud, whose murky

BATTLING WITH PELEE 225

masses hung at a dizzy height above us. Its mantle-sheet
carried darkness to Macouba and Grande Riviere, and far
over Dominica and Gaudeloupe the black mass still swept
out to sea. I believe that the ash-cloud must have been
fully six miles above our heads. It rolled out a few peals
of thunder, but we observed no flashes of lightning. The
ash fell lightly, and coming mixed with water soon con-
solidated into a paste. It had the temperature of the
surrounding air — was not warm. There were no large
particles. The coarser material fell miles from us, at
positions situated more nearly under the periphery of the
cloud.

It is singular that even at the point where I was nearest
to the issuing steam, a distance of probably less than four
hundred feet, no marked atmospheric disturbance was per-
ceptible, nothing to even remotely suggest a cyclonic or
suctional whirl. One could readily have expected some-
thing of this kind to occur. Nor do I believe that there
was any noticeable elevation in the temperature of the air.
Unfortunately, the single thermometer that I had with me
had broken earlier in the day, and, therefore, my note on
this point rests solely on a personal impression. Certainly
there was no emphasized change in temperature. I could
detect no gaseous emanations, except, perhaps, a very feeble
taint of sulphur.

When we again got on the level ground back of the

Habitation Leyritz we were startled by a most violent erup-

ion from Pelee, a great shaft of steam and ash being sud-

lenly shot out to a most marvellous height, perhaps not less

226

BATTLING WITH PELEE

than five or six miles. It went up as a distinct column of
its own, swiftly distancing the other cloud-masses by which
it was enveloped. It was a prelude to the incidents of the
evening that followed.

THE RORAIMA BURNING

XVI

A NIGHT OF ILLUMINATION AND THE DESTRUC-
TION OF MORNE ROUGE AND
AJOUPA-BOUILLON

We arrived at our shelter a little before five o'clock,
somewhat to the relief of the household, who had become
apprehensive regarding our safety. Early in the evening
the big blaze from an incendiary fire announced the destruc-
tion of the case de bagasse of the Habitation Pecoul, but it
gave us little concern, as our cane-fields were sufficiently
removed to insure them from contact with the flames. Still,
M. des Grottes thought it advisable to examine the premises,
and he rode down with his brother more, perhaps, as a
pastime than as a necessity, returning for a late evening
meal. While still seated at the table, a flash of lightning
and a dull thud told us in an instant that something was
happening. We were out at once. This was a few min-
utes, perhaps a quarter of an hour, after nine o'clock.
The volcano was still distantly growling. The heavens
were aglow with fire, electric flashes of blinding intensity
traversing the recesses of black and purple clouds, and
casting a lurid pallor over the darkness that shrouded the
world. Scintillating stars burst forth like crackling fire-
works, and serpent lines wound themselves in and out like
travelling wave-crests. The spectacle was an extraordinary
and terrifying one, and I confess that it left an impression

227

228 A NIGHT OF ILLUMINATION

of uncomfortable doubt in our minds as to what would be
the issue. One could not but feel that a tremendous de-
struction was impending.

The number of forms in which the illumination ap-
peared was bewildering, and I can only recall a few the
picture of which presented itself to my eyes with precision :
short, straight, rod-like lines, wave-lines, spirals, long-armed
stars, and circles with star-arms hanging off from the bor-
der like so many tails. In addition to these were the scin-
tillant stars to which reference has already been made, and
the blinding flashes of normal or zig-zag lightning. There
were no peals of thunder, but a continuous roar swept
through the heavens, mounting with crescendos and fall-
ing off with alternating, far-reaching diminuendos. Some
pretend to have heard a feeble crackling, like that which
is so often heard in association with an auroral display, but
I am not sure that I could record this condition, which
may easily have existed, among my own experiences. Tl
flashes were bewilderingly numerous, and the singular fori
interwoven with one another in such a way as to make local
ization difficult. The scintillant stars alone appeared
have a place of their own, nearer the border of the great
cloud, and perhaps in the highest parts of it. Directly
over the summit of Pelee there was little to be seen. Who
is there to tell us what these peculiar flashes are? Are
they electric, or are they the flashes of burning gases ? It
would, probably, be easy to determine their nature by
means of the spectroscope, but this form of examination has
not yet been made. It is certain that most of them are

A NIGHT OF ILLUMINATION 229

not connective discharges, for they run through, or are
contained in, individual clouds of small dimensions. The
phenomena appear to be identical with those which were
noted to accompany the great eruption of Tarawera, in New
Zealand, in 1886.20

As our eyes feasted upon this scene of majestic gran-
deur, we almost lost sight of the fact that ashes were falling
about us. A great pattering of pumice and lapilli had
ushered in the storm, and for a while it sounded as if we
were in a tropical hail-storm. Only the fragments first
thrown were large, a few an inch or more in size, and those
following were like peas and lentils, and then like sand.
But even the smaller particles came down with much force,
and the flesh stung as it was touched by them. They were
all angular bits of andesite or trachyte, white and gray in
color. We were out in our bared heads, but it was soon
found necessary to protect them. The fall lasted somewhat
over an hour, or to nearly half-past ten. All motion in
the atmosphere ceased at this time, and for once the location
Leyritz lost its usual refreshing coolness. The falling ash
felt warm, but M. des Grottes's thermometer failed to indi-
cate anything special.

It was not given to us to close the night quietly. The
flashing sky above and the falling ash had yet a comple-
ment. For over an hour the southwest was glowing fiery
red, and patches of lurid light moved themselves into the
black of the volcanic cloud. No flame was visible, but it
was only too evident that fire was devastating somewhere.
Morne Rouge lay to the same point of the compass, and we

230 A NIGHT OF ILLUMINATION

intuitively asked ourselves if it could be that town aflame.
Ajoupa-Bouillon lay a little to one side, almost adjoining us,
and if it were on fire we could easily have seen the flames.
When we retired for the night, M. des Grottes had decided
to desert the habitation. Pelee was too close to us, and too
active to be sought for as the simple ornament which
it had been designated by the Scientific Commission of
1851. Most of the working inhabitants of the planta-
tion had betaken themselves to the coast immediately after
the first storm of the evening, terror-stricken with the un-
ceasing roar of the volcano and the flashing lightning, and
my own men had joined them in their mad flight. All
thoughts of a new exploration of the summit of the volcano
on the morrow had vanished. It was not without appre-
hension that the great door of the manse was closed that
night. I did not quite share M. des Grottes's fears that
there might be no one in the morning to open it, but the
hours for rest were spent mainly in thinking.

The night-air was almost without breeze, so different
from what we had had up till now. I tossed around until
about one o'clock, sleeping in snatches, but hardly
resting. At this time there was another sharp pattering
of cinders, and I moved up to the window, only to see
darkness. On another side the sky was flashing bright
tongues of light, but I saw nothing of it, and knew not
that it was taking place. Before retiring again I had
clear my bed of ashes, for the covers and pillows were bein^
rapidly filled, and a new fall was only just beginning
The poor tree-toads, despite everything, were still chirping,

>.*.^ — .'•.•"'- —

Underwood & Underwood, Stereos. Photo., New Y
BEFORE A SHRINE— MORNE ROUGE

A NIGHT OF ILLUMINATION 231

and manifestly to them life was not a burden, nor even a
piece of anxiety. I do not know to what extent it is true
that before the eruption of May 8 the animals of the field
and forest gave signs of uneasiness, and summarily left
their homes in search of new quarters. Nothing of this
kind appears to have been noted on this side, which is in
itself not conclusive evidence denying the condition re-
in rted, and I know that on Sunday morning the blackbirds
were, as usual, gambolling about the cocoanut crowns, and
sending out their joyful notes to greet the rising sun.

Before the morning had yet broken, news reached us
that the fiery tongue of Pelee had carried death and deso-
lation to Morne Balai. The flash of nine o'clock, when the
heavens were glowing and scintillating with fire, was the
lumen that showed the path to the pretty village which we
had left hardly five hours before, and from which weeping
messengers had now come to ask for aid. I immediately
rode out with M. Edouard des Grottes to ascertain the extent
of the casualties and what in fact had taken place. We
had hardly a mile to go, even with the windings of the
path, and were soon conducted to the scene of the disaster.
In one of the low thatched cottages two bodies were stretched
out stiff in death, and near by others were lying groaning
in agony from the terrible burns they had received. Still
others, which we did not see, were in the neighboring cases.
We gave such comfort as reassuring words could offer, but,
alas ! of what value are they ? M. des Grottes arranged
for the care and removal of the wounded, and we then left.
One of those who had been with me on the mountain in

232 A NIGHT OF ILLUMINATION

the afternoon was a victim to the volcano's wrath, and his
body lay not far from the hut where we had halted for a
few minutes for a friendly chat, and which was now flat
with the ground. It had tumbled with the volcano's blast ;
others like it had fallen under the weight of ash that had
been showered upon them.

My parting from Morne Balai was a sad one. It was
hard to realize that this pretty little village, which appeared
to me so joyful a few hours before, should now be clouded
in the shadow of death — death driven to it by the same
force whose enigma I was attempting to penetrate. As we
looked down upon it from the slopes of Pelee, it lay so
peacefully embowered beneath its clumps of verdure, ap-
parently so far from danger's door. Nature had turned her
hand and heart, but this was only a part of the history of
the night before. Ajoupa-Bouillon, Morne Rouge, Morne
Capot, the heights of Bourdon, were wrecked, or had been
entirely wiped out, and with them two thousand more of
Martinique's inhabitants were sent to their graves. On all
these sites we had gazed in the quieter afternoon ; we had
noted the fleeting cloud-shadows passing over them, and
seen the smiling fields and forests that bound them into one
vast sea of green. Desolation had swept all this into gray
and black. The very slope that we had travelled over was
culled in the fiery blast, and wreck and ruin were every-
where. Our own escape was, indeed, a very narrow one,
for the blast swept the land to both sides of us, and even
descended to the rear of the Habitation. Good fortune
much more than management gave to us our place of refuge.

A NIGHT OF ILLUMINATION 233

It was only when we reached Vive that the full extent
of the catastrophe was made known to us. The great sugar
estate had once more set her wheels moving, and from the
lofty chimney curls of smoke were again peacefully flowing
over the verdant fields of cane. The Riviere Capot, whose
debordements had been so much feared and had caused so
much damage, was no longer a dangerous stream, and confi-
dence came to all who felt that the worst of Pelee was over.
Its work was thought to belong to the south and to the west,
and few feared, even in the face of the magnificent pyro-
technic display of later days, that anything serious could
happen on this side. Refugees had been returning by hun-
dreds to their abandoned habitations, and the silence of
desolation once more woke to the voice of the living.

In front of the great Usine, when we arrived there
this time, crowds of refugees coming from Basse-Pointe,
Macouba and Grande-Riviere, and from minor hamlets in
the interior, had assembled, and travelling parties were all
over the roadways. Afoot and on wagon, everybody was
going, with no one having a good word for the country.
Improvised ambulances were being sent in to Ajoupa-
Bouillon, and since the earlier hours of morning the
wounded were being brought out in scores and sent down
to Grande-Anse to be placed under government treatment.
The good people of the Usine were doing everything that
under the circumstances could be done to alleviate the suf-
ferings of those who were still living, but, unfortunately,
for many their work came too late, for they died on the
roadway. And perhaps it was best that it was so, for

234 A NIGHT OF ILLUMINATION

death removed from the body an agony that cannot be con-
ceived, while the chance for recovery was all but nil. Less
than four months had elapsed since the catastrophe of May
8 overwhelmed Saint Pierre, and the tragedy was being
enacted over again.

M. Joseph Clerc kindly invited me to join him in a
survey of the situation at Ajoupa-Bouillon, and we rode
out almost immediately after my return to Vive. The
village of Ajoupa-Bouillon lies on the eastern foot of Mont
Pelee, in a direct line not more than one mile from the
more recognized slope of the volcano, and at elevations
ranging from about eight hundred to thirteen hundred
feet above the sea, some extreme parts rising possibly
higher. It is connected on the inner side with Morne
Rouge by one of the finest roads in the island, which before
the catastrophe of this day was bordered by a woodland of
singular beauty. Its houses were mainly of wood, but
there were others of a more substantial construction, and
nearly all had gardens of their own. A graceful church
steeple, still standing, rises up from nearly the highest part
of the village. Four days before this second visit I had
come out with the acting Mayor of the village, M. Kloss, to
look over his large cacao estates, and to join in an excursion
to the Trianon, the site of a former hospital camp, situated
directly above what had been assumed to be a new crater in
the gorge of the Falaise. At that time Ajoupa-Bouillon
looked more attractive than I had ever seen it before. The
vegetation was at its best, and seemed to have profited by
the ash that had been thrown over it in the early days

A NIGHT OF ILLUMINATION 235

May. Not here alone, but all over the north this extreme
of "pushing" fertility was noticeable, and everybody
remarked upon the luxuriance of growth which distin-
guished field as well as forest. To this end, at least, of
adding fertility to the soil, the volcano may have con-
tributed, and done something to redeem its bad name.
To-day, alas ! much of this had gone. In place of field
and forest there were desolate plains, gray-scarred, ash-
covered, and bleak almost as the African desert. We
looked over to the mountain-heights and down into the
valleys and gorges, and everywhere the eye fell upon ruin
and desolation. Only back of us and in the farther dis-
tance was there enough of verdure left to remind us of the
past.

The force of the destruction was extraordinary. Be-
fore we reached the main scene of the catastrophe the wreck
was already fully indicated in a number of houses which
were laid flat with the ground, and in overturned trees with
buttressed roots lying to the side of the coming blast.
Boards were found completely penetrated by others that
had been shot through them. It was evident at a glance
that it was the history of Saint Pierre over again. The
/one of destruction began a short distance above and be-
yond the church, and extended almost without interruption,
so far as we could see from the heights, to Morne Rouge.
Looking over to the site of that town, we saw before us
nothing but a withered plain, with arid slopes on one side
of it, and slightly green mornes on the other. Cattle and
horses were lying on their backs, with their legs rigidly

236 A NIGHT OF ILLUMINATION

extended into mid-air. A few more fortunate beasts, wit]
raw flesh protruding from their tightened hides, were
moving aimlessly about, as if dazed by the conditions that
now surrounded them. Clear up to the low saddle between
the Morne Jacob and the Calebasse the eye followed the
bleak landscape, and it was plain to see that the tornadic
blast had this time lined its course over this arete, instead of
confining itself to the zone of the Riviere Blanche on the
opposite side of the mountain. The first houses that we
examined had simply collapsed. They occupied their own
ground and were merely a mass of sticks and roof material,
covering all that the houses contained — inmates probably as
well as their belongings. We put our ears to the ground
and to the planking, but could hear no sound. Off on a
side-lane we passed a little cottage apparently untouched
on the exterior, and hearing deep moaning we entered. A
poor woman, of perhaps thirty years, was rolling in agony
in one corner of a dark room, her flesh terribly burned and
hanging in places from the bone. She called incessantly
for water to relieve her excoriated throat, but it could not
be furnished. M. Clerc sent immediately for the gendarme,
to have her removed where friendly care could be adminis-
tered to her, but she died shortly after our leaving.

We entered another case near by. A dim taper illu-
mined a nearly black interior sufficiently to permit us to
see a writhing figure being tended by the hand of one who
was left probably dearest to it. The cries of pain were
heart-rending. Flies were swarming everywhere about the
place and the odor was almost unbearable, as the precaution

A NIGHT OF ILLUMINATION 237

had been taken to keep the door closed. A body, relieved
from anguish, lay stiff in another corner. We passed from
this to another house and saw the same picture repeated.
In reply to inquiries put to him by M. Clerc, one of the
inmates, perhaps less terribly burned than some others,
stated that he had been struck by the hot blast at the mo-
ment of opening the door of his case, which he had done
assuming that the storm had passed. Instantly the fiery
air enveloped him, and he felt the sensation of choking.
There seemed to be no air to breathe. His flesh was as
if baked and steamed, with raw red masses appearing
where there was no longer skin. The clothing had re-
mained untouched. I inquired if he had noted gas of any
kind. He replied in the negative, except to the extent
that a feeble sulphurous odor, already appreciable in the
earlier part of the evening, could be detected. We obtained
almost exactly the same history from an adjoining cottage.
In some cases, perhaps even a large number, where the
cottages had the doors and windows firmly closed, and
were able to withstand the force of the tornadic ferment,
there was little or no injury done. In the greater num-
ber of cases, however, it is certain that the fiery breath en-
tered even where every opening avenue had been secured.
This was also the case, as I ascertained later, at Morne
Rouge.

There was here, as at Saint Pierre, the same reference
to the feu, or fire, but it was evident that only a heated or
a luminous blast was conveyed by this designation, and
nothing burning with a flame. It seems certain that in

238 A NIGHT OF ILLUMINATION

some instances the darkness of the interiors was actually
illumined at the time of the entry of the hot blast, and
some claim to have seen electric discharges traverse the
room. I think this condition exceedingly likely, and have
always believed that localized lightning must have played
an important part in the destruction of life at Saint Pierre.
There was no evidence at Ajoupa-Bouillon of anything
having burned with a flame within the village itself nor
in the surroundings. One-half or more of the settlement
had been scorched or swept out of existence, but there had
been no fire of any kind. The sticks and planking of the
cottages showed no change to the eyes, except that they
had become gray, mainly, perhaps, as the result of the
splattering with ash. Even the dry palm-thatching had
remained intact, with no evidence of true burning of any
kind. The trees and bushes that still stood in and out
of the village had their leaves, clinging to the twigs and
branches, shrivelled up and turned to gray and umber.
Nothing had been carbonized, although the sap had been ex-
terminated and the smaller twigs broke fragile. I searched
in vain for any indication of active terrestrial gases, and
could detect no trace of any gaseous odor, not even that of
sulphur.

The destruction of Ajoupa-Bouillon took place almost
immediately after nine o'clock of the previous evening.
It was also the time of the destruction of Morne Rouge
and the invasion of Morne Capot, and there can be no
question that all the havoc that had been wrought on this
fatal August 30 was the result of one explosive blast,

A NIGHT OF ILLUMINATION 239

whatever may have been its exact nature, or of a series
of such blasts following rapidly upon one another. It is
singular that we, who were passing the evening at the abso-
lute foot of the volcano, much closer to it than some points
at had been destroyed, and remarking upon the magnifi-
nce of the electric display, absolutely above us, should
arely, if at all, have noted the detonations which pre-
ceded, accompanied or followed the explosions. At St.
Kitts, two hundred and seventy miles northwestward, the
booming of the volcano sounded at this time like the
cannonading of a naval combat in which the largest guns
were being used; and the same observation was made at
Port-of-Spain and elsewhere in Trinidad, at a somewhat
farther distance in a direct line southward. In Fort-de-
France hardly more than the continuous terrific roar of the
volcano could be heard, and it was this, together with the
illumined ash-cloud, which threw the inhabitants into con-
sternation and initiated the new panic. I confess my ina-
bility to satisfactorily explain this singular disposition of
the sound-waves, as every explanation that has suggested
itself to me seems to meet with some objection. It is not
the distance at which the detonations were noted which
imposes the difficulty to the problem, but the fact that so
transcendent a sound, originated with explosive violence,
should hardly have been noted in or near the epicentral
region. Is it an extreme condition of sound-shadow ? Or
has the cavernous and " blanketed" condition of the vol-
cano something to do with this ? Or are we forced to admit
a series of paroxysmal deep-seated explosions occurring

240 A NIGHT OF ILLUMINATION

in the horizontal conduit of the volcano, and immediately
antecedent to the vertical discharge ? The latter condition,
apart from any relation to the present inquiry, is, of course,
well possible, and even very likely. The acoustical rela-
tions of the May 8 eruption were similar to those of the
later day, and it is interesting to note that Alexander von
Humboldt, referring to the eruption of the Soufriere of St.
Vincent in 1812, remarks upon the same peculiarity of
sound-carriage — the eruption being more distinctly audible
at a distance from the island than near to it.21

The conditions of time did not permit me to visit Morne
Rouge, and my only glimpse of the destroyed city was
obtained in sailing out from the island. The sole struct-
ure visible was the stately church and its sharp steeple,
always so prominent as seen from the site of the northern
Saint Pierre. A part of the roof had been lifted, but this
could not be seen — nor the other remaining houses which
told of the former existence of a city whose population
ranged from three thousand to four thousand or more.
Like its sister city, Saint Pierre, to whose wealthier inhabi-
tants it ministered the cool of mountain breezes and the
solace of verdant fields and forests, Morne Rouge was wiped
out — razed to the ground and in part burned aflame. The
glare of its fire was plainly visible to us at the Habitation
Leyritz. The country on all sides of the town was deso-
lated, and nothing remained of the beautiful greens which
gave the charm to the location. The whole Calebasse slope
was swept clear, and far off, on the heights of Fonds-Saint-
Denis and over nearly to the Pitons de Carbet, could we see

A NIGHT OF ILLUMINATION 241

the entering-wedges of the scarred vegetation. Pelee had
wonderfully increased its zone of force.

There would appear to be at this time no way of closely
approximating the casualties at Morne Rouge, although it
is all but certain that at least twelve hundred perished. On
the morning of the fatal day, as I was informed by one of
the Brothers associated with the Vicar- General of Martin-
ique, M. Parel, two thousand one hundred rations had been
distributed by the government officials, the bulk of the pop-
ulation being still held on the list of the sinistres of May 8.
It is thought that several hundreds must have sought more
secure quarters (where ?) during the day, when the activity
of the volcano became unbearable, and of this number
probably the greater part was saved. The Vicar- General
himself believed that from twelve hundred to fifteen hun-
dred perished, excepting perhaps fifty or sixty, all who re-
mained up to the hour of nine o'clock. Many of the
corpses were swept far from the site of the catastrophe,
others remained buried under the debris that lies over
them, and still others were burned to a crisp mass. Save
the church and two or three other buildings, all the houses
of the town were of wood.

A particularly sad moment in the annihilation of Morne
Rouge was the taking away of Pere Mary, the good curate of
the church, whose faithful work in ministering to the wants
of those who stayed during the storm of May 8 will long
be remembered in the history of Martinique. He had only
recently returned from Fort-de-France, and now perished
with nearly all those who had returned with him, thinking

242 A NIGHT OF ILLUMINATION

that danger had passed. When the presbytery was on
he sought the shelter of the church, but was struck by the
hot blast before that building could be reached. He suc-
ceeded, however, in dragging himself into the interior, and,
with terrible suffering, stretched himself upon a bench.
Here he was found at four o'clock of the following morn-
ing, still fully conscious and expressing anxiety for his
flock. He was removed to Fonds-Saint-Denis, and thence
to Fort-de-France, where he expired at eleven o'clock of the
morning of September 1 — a man honored by all.

At the Hopital of Fort-de-France I had the advantage
of an interview with a lovely French girl of perhaps seven-
teen years, Mile. Desiree Martin-d'Harcourt, who had been
brought down as one of the wounded from Morne Rouge,
and who gave me a very intelligent statement of her impres-
sions of what had taken place on the evening of the 30th.
Her mother, more burned than herself, and also her
brother, were being cared for in the same room. The fam-
ily had retired for the night, not being able to stand the
strain which the roaring of the volcano imposed upon them
any longer, and firmly secured the house, closing everything.
Shortly after nine o'clock a dull detonation was heard, and
the outer shutter (sous-le-vent) was released from its bar
fastenings and swung open. Instantly the hot blast entered
and commenced its terrible rasping work. Mile. Desiree
was confident that it was luminous or electric in character.
Refuge was sought under the beds, and mattresses were
hauled down to cover the protruding feet. At this time,
thinking that the storm had passed, Mme. Martin-d'Har-

A NIGHT OF ILLUMINATION 243

court opened the door, only to admit a second and stronger
blast, to which she nearly succumbed. All experienced
extreme difficulty in breathing, but the sensation of choking
was only momentary. Sulphurous odors were strongly
perceptible. The Martin-d'Harcourt home was one of the
better properties of Morne Rouge, and doubtless owed its
escape from destruction to superior construction, as it stood
sufficiently exposed to the storm. Mme. Martin-d'Harcourt
succumbed to her wounds the day following my visit.

XVII

THE SOUFRlERE OF ST. VINCENT AND THE
AFTERGLOWS

The disposition of my time in the Lesser Antilles did not
permit me to conduct investigations in St. Vincent, and
my only view of the Soufriere was from the deck of a small
coastwise steamer coming from and going to Martinique on
May 26 and 27. The atmosphere was, fortunately, clear,
and we obtained as we approached and passed the mouth of
the Wallibou River an almost unobstructed picture of the
great volcano, whose cloud was drifting eastward, and of a
large part of the plain that marked one of the most noted
areas in the region of volcanic destruction. Puffs of steam
were rising from many parts of the Wallibou Valley, and
great blasts, coming at almost regular intervals, denned
positions of greatest accumulation of the ejecta, to which the
running waters spasmodically found their way. These sec-
ondary explosions were similar to those which appear in
the basin of the Riviere Blanche, in Martinique, but they
were much more numerous here than at any time, as I had
observed them, in the latter locality, except immediately
after the eruption of August 30. The more irregularly
cloaked and incised plain of the Wallibou, turning the
waters to widely-diverging courses, sufficiently accounts for
this.

In the great valley that lay ahead of us, and that in it

244

-4
*

^£*/ "

iB^^fa

i«f .jrifegJB

-vip

m :

*'. *

Jm± ■•-*-•,

Underwood & Underwood, Stereos. Photo., New York, Copyright, 190a
THE DARKENING CLOUD OF JUNE 6, 1902

THE SOUFEIERE OF ST. VINCENT 245

upper part is turned towards the Soufriere, were two great
arms of black mud, which hung like receding glaciers in
their expanded beds. They were overflows from disrupting
waters, and had their black color from being wet. In gen-
eral appearance they were like black lava-flows, and it was
to understand how, from a distance, they should origi-
nally have been taken to be such. At the time when I first
passed Mont Pelee a similar sheet of black mud, slowly
creeping down to the ocean front, occupied much of the
surface of the plain between the Rivieres Blanche and
Seche.

The phenomena of the great eruption of the Soufriere
on May 7, one day in advance of that of Mont Pelee, have
been carefully studied by a number of investigators and
shown to be of fundamentally the same nature as those of
the Martinique volcano. As in Martinique, there was no
lava-flow; but in its place, or representing it, there were
extensive mud-discharges, some of which appear to have
had their origin in the lake which before the cataclysm
occupied most of the deep depression of what is known as
the "old" crater. The observations of Dr. Hovey and
others point to this crater, which lies southwest of the
"new" crater, or the crater of 1812, as the seat of the
activity of this eruption. Its dimensions are stated to be,
approximately, nine-tenths of a mile in east-and-west diam-
eter and eight-tenths of a mile from north to south, with a
depth to the crater-floor of from sixteen hundred to two
thousand four hundred feet. The areal dimensions of the
ealdron are, therefore, vastly greater than those of the

246 THE SOUFKIEKE OF ST. VINCENT

crater of Pelee, perhaps four times as great. The surface
of the new and shallow boiling lake which in the latter part
of May and from June to August occupied the deepest part
of the crater-floor was estimated to be only twelve hundred
feet above sea-level, whereas the sheet of water that pre-
ceded it, and which had been famous for its beauty, rose
seven hundred feet higher (to nineteen hundred and thirty
feet.) The floor of the crater was thus about twelve hun-
dred feet lower than the lowest point of the crater of Mont
Pelee, the basin of the Etang Sec.

The Soufriere is somewhat less high than Mont Pelee,
rising to four thousand and forty-eight feet, according to the
generally received measurements. Like Pelee, its foot con-
tours a great part of the northern shore of the island, and
from near its summit radiate off a large number of streams,
nearly all of which take individual courses to the sea. The
summit bears two craters, the " old" crater, which has
already been referred to as the seat of the volcano's latest
activities, and the " new" crater, or the crater which was
active in the eruption of 1812, and which lies to the north-
east of the much larger ancient vent. The two are sepa-
rated by a saddle which descends to three thousand five
hundred and fifty feet or lower. A gentle plateau, similar
to that which contained the Lac des Palmistes on the sum-
mit of Pelee, extends eastward from the larger crater, and
passes to the south of the minor crater of 1812.

Unlike the eruption of Mont Pelee, that of the Sou-
friere does not appear to have been heralded by antecedei
outbreaks ; or, if there were such, they went by unnotice

THE SOUFKIERE OF ST. VINCENT 247

The earliest awakening symptoms, except rumblings, were
observed only two days in advance of the cataclysm, on
May 5 — the day on which the Usine Guerin was over-
whelmed in Martinique. On that day the water in the
crater-lake was observed by fishermen who had crossed the
Bummit of the mountain to be discolored and agitated, but
it was not until the day following, Tuesday, the 6th, that
the working powers of the volcano were put forth in earnest.
Great clouds of vapor were thrown out towards evening, and
the crown of the volcano is described as having been illu-
mined by a glow of "fire." The first explosion with loud
detonation was noted shortly before three o'clock of the
afternoon. On the following day, when the main destruc-
tive work of the volcano was accomplished, the explosions,
accompanied by vast discharges of ash, bombs and boulders,
and associated with electric displays and detonations of the
most intense energy, followed one another in rapid succes-
sion. They began early in the morning, and it is thought
that the first appearance of solid matter ejected by the vol-
cano was noted shortly after six o'clock. The steam col-
umn was roughly estimated to have reached an altitude of
thirty thousand feet. The most violent paroxysm would
seem to have occurred shortly after ten o'clock, but others
of nearly equal intensity succeeded during several hours,
or until nearly two o'clock, when a considerable part of the
island was hidden behind a vast, reddish-purple curtain,
which swiftly advanced over the land and descended upon
the sea. A furious fusillade of stones and boulders, a large
part of them intensely heated when they fell, was kept

248 THE SOUFKIEKE OF ST. VINCENT

up during most of this time, and was, doubtless, respon-
sible for a considerable loss of life.

The official estimate of the loss of life resulting from
the Soufriere eruption places the deaths at thirteen hundred
and fifty. Most of those who " weathered the storm" had
taken refuge in basements and cellars, and had firmly se-
cured themselves behind fastened blinds and doors.

Many of the general phenomena noted in the Martinique
eruption were also observed here, and the effects of tornadic
hot blasts sweeping off the mountain were marked in the
same way as on Pelee and in Saint Pierre, masonry being rent
asunder, trees overturned and stripped of their covering
and appendages, and flesh scorched and tumefied. There
was no single concentrated blast, however, such as that
which annihilated Saint Pierre.22 The discharges, seemingly
less powerful than that from Pelee on May 8, and more
properly comparable with those of August 30 which an-
nihilated Morne Rouge and other seats of habitation, were,
so far as we can judge of the effects produced, consecutive
in action, following one another at not long intervals, and
radial in the lines of their destruction.

The following account of the eruption, written two days
after the major event, is furnished to the Barbados Bulletin
(May 12) by the Rev. J. H. Darrell, of Kingstown, an
eye-witness of some of the phenomena which he describes :
" At seven a.m. on Wednesday, the 7th instant, there was
another sudden and violent escape of pent-up steam, which
continued ascending till ten a.m., when other material began
to be ejected. It would seem that this was the time when

THE SOUFKIEKE OF ST. VINCENT

the enormous mass of water in the lake of the old crater*
emitted in gaseous condition. . . . The mountain heaved
and labored to rid itself of the burning mass of lava heav-
ing and tossing below. By twelve-thirty p.m. it was evi-
dent that it had begun to disengage itself of its burden
by the appearances as of fire flashing now and then around
the edge of the crater. There was, however, no visible
ascension of flame. These flame-like appearances were, I
think, occasioned by the molten lava rising to the neck of
the volcano. Being quite luminous, the light emitted was
reflected from the banks of steam above, giving them the
appearance of flame.

" From the time the volcano became fully active, tre-
mendous detonations followed one another so rapidly that
they seemed to merge into a continuous roar, which lasted
all through Wednesday night, yesterday (Thursday, the
8th) and up to six-thirty a.m. this morning, the 9th instant.
These detonations and thunderings were heard as far as Bar-
bados, one hundred miles distant, as well as in Grenada, Trin-
idad and the south end of St. Lucia. At twelve-ten p.m.
on Wednesday, I left in company with several gentlemen
in a small row-boat to go to Chateaubelair, where we hoped
to get a better view of the eruption. As we passed Layou,
the first town on the leeward coast, the smell of sulphur-
etted hydrogen was very perceptible. Before we got half
way on our journey, a vast column of steam, smoke and
ashes ascended to a prodigious elevation. The majestic
body of curling vapor was sublime beyond imagination.
We were about eight miles from the crater as the crow

250 THE SOUFKIEKE OF ST. VINCENT

flies, and the top of the enormous column, eight miles off,
reached higher than one-fourth of the segment of th
circle. I judged that the awful pillar was fully eight mil
in height. We were rapidly proceeding to our point o
observation, when an immense cloud, dark, dense and
apparently thick with volcanic material, descended over
our pathway, impeding our progress and warning us to
proceed no farther. This mighty bank of sulphurous
vapor and smoke assumed at one time the shape of a gigan-
tic promontory, then of a collection of twirling, revolving
cloud-whorls, turning with rapid velocity, now assuming
the shape of gigantic cauliflowers, then efflorescing into
beautiful flower-shapes, some dark, some effulgent, others
pearly white, and all brilliantly illuminated by electric
flashes. Darkness, however, soon fell upon us. The sul-
phurous air was laden with fine dust that fell thickly upon
and around us, discoloring the sea ; a black rain began to
fall, followed by another rain of favilla, lapilli and scoriae.
The electric flashes were marvellously rapid in their mo-
tions and numerous beyond all computation. These, with
the thundering noise of the mountain, mingled with the
dismal roar of the lava, the shocks of earthquake, the falling
of stones, the enormous quantity of material ejected from
the belching craters, producing a darkness as dense as a
starless midnight, the plutonic energy of the mountain
growing greater and greater every moment, combined to
make up a scene of horrors. It was after five o'clock when
we returned to Kingstown, cowed and impressed by the
weirdness of the scene we had witnessed, and covered with

;

THE SOUFBIEBE OF ST. VINCENT 251

the still thickly falling gray dust. . . . The awful scene
was again renewed yesterday (Thursday, the 8th) and again
to-day. At about eight a.m. the volcano shot out an immense
volume of material which was carried in a cloud over
Georgetown and its neighborhood, causing not 6nly great
alarm, but compelling the people by families to seek shelter
in other districts.' '

The frightful intensity of the Soufriere eruption is made
plain from this description, which agrees well with the des-
cription of other observers, and is perhaps the most exact
of those that have been made public. A just comparison
with the great eruption of April 30, 1812, an eye-witness
account of which, reproduced from the London Evening
Mail of June 30, 1812, appears in the Appendix, is hardly
possible at this late day, but it is evident that the volcano
had lost but little, if at all, in vigor during the ninety years
of its repose.

On the evening of September 3, immediately following
my return to Fort-de-France from Vive, we were treated to
a new form of volcanic excitement. Far out to sea, south-
ward, vivid flashes of lightning were illuminating a corner
of the heavens. They followed swiftly upon one another,
and zig-zagged across broad stretches of a practically cloud-
less sky. We wondered if it could be the Soufriere of St.
Vincent again in eruption. That volcano, unlike Pelee,
had practically died down, and for weeks past had barely
given signs of a life within it. Visiting parties had
wandered over its craters and descended quite to the edge
of the water that again filled the crateral hollows. There

252 THE SOUFRIERE OF ST. YINCENT

was little to tell that an impressive activity had but
cently ceased. We were distant in direct line nearly
ninety miles from the Soufriere, and it hardly seemed pos-
sible that the brilliant flashes could have come from that
distance ; but if not from there, whence ? As the evening
advanced, the flashes became more and more brilliant, and
their localization to a very limited area of the sky left no
further doubt that the monster of St. Vincent was again in
eruption. My windows in the Hotel Ivanes opened out
upon this spot, and gave a splendid position whence to view
the display. Directly in line, but in the opposite direc-
tion, was the darkened mass of Mont Pelee. The spectacle
was both terrifying and impressive. From the Diamant
and the southern part of the island the great glow of the
eruption was plainly seen, but from Fort-de-France, up till
one o'clock of the following morning, only the lightning
flashes were visible, and these were brilliant. To about each
twelve or fifteen Pelee responded with one blinding flash,
so intense as to seem to open the heavens. A green sky
appeared in the flash, and for a fraction of time, consider-
ably longer than any I had ever experienced with ordinary
lightning, the eye was paralyzed and saw nothing.

This extraordinary spectacle, a contest in heaven, as it
were, between two titans, continued almost uninterruptedly
for three hours and more, Pelee gradually increasing the
number of her responding flashes. Shortly after one o'cloc
a great red light in the south announced the culmination,
and the Soufriere had released itself of the energies tha
had been stored within. The eruption was seemingly more

THE SOUFKIEKE OF ST. VINCENT 253

intense than that of May 7, but, moved by anticipatory
warnings, the government found the opportunity to notify
the inhabitants of the surrounding districts of their danger
and to call them into locations of assured security. Up to
the time of my leaving Martinique there were no casualties
reported. It is a noteworthy, even if not necessarily sig-
nificant, fact that this was the first of all the strong erup-
tions of the Soufriere which followed a big eruption of Mont
Pelee, the others (such as those of May 6-7 and 19) usually
preceding by a day or so the nearly concurrent disturbance
of Martinique.

Shortly after five o'clock of the morning, the edge of a
black cloud could be plainly seen advancing upon Marti-
nique from the south. It was the ash-cloud of the Soufriere,
which slowly but surely crawled in upon us. By seven
o'clock it had passed over Fort-de-France, and clung so
over Mont Pelee that the frightened inhabitants of the city
thought that their own mountain had been in eruption. At
eight o'clock the sun was covered, and it remained so until
nearly three in the afternoon. During all this time a gray
gloom hung over the city, the heavens being leaden or pur-
plish in color, but there was nothing approaching true
darkness. The general sensation was similar to that expe-
rienced during a total eclipse. White objects on the sea
loomed up with remarkable brilliancy and stood out sharp
against the background of blackened sky. The people
were naturally terrified, and once more the thoroughfares
were crowded with observers anxious to know their fate.
The canopy overhead was almost exactly like that coming

254 THE SOUFKIEKE OF ST. VINCENT

from Pelee which I had observed in the morning of June 6,
only that it was less dense and naturally moved much more
slowly. As on June 6, there was a marked lowering of the
temperature, and throughout the day a most genial atmos-
phere was maintained. Some of the inhabitants, endowed
with a specially acute olfactory sense, claimed to have
smelled sulphur, but I could detect nothing of this nature.
There was no fall of ash over the city, and but little over
any part of the island.

While on Pelee in the afternoon preceding the eruption,
I satisfied myself that volcanic ash was not necessarily a
triturated product derived by abrasion from blown-out
larger pieces (cinders or lapilli), and that it leaves the
bowels of the volcano in the form of fine powder in which
it floats out to distant parts in the murky cauliflower clouds
while they are surcharged and in the flotation of the main
cloud. The propelling power of the ejecting steam-column
was such that no extensive triturating process could take
place anywhere within its reach. The particles are mani-
festly formed by explosive action within the bowels of
the volcano, and are shot out spasmodically as successive
eruptions take place within the deep-seated conduits.
Much of the continuously ejected particles may even be
formed through direct abrasion by the rising steam-column
of the side-walls of the crater. This abrading force is at
times certainly prodigious, and it must produce some dis-
ruption.

The quantity of ash thrown out by Pelee, if measured
alone by what fell upon the island, would not seem to have

THE SOUFKIERE OF ST. VINCENT 255

been very heavy, much less in quantity, indeed, than ash-
falls that have been associated with minor volcanic erup-
tions elsewhere. Except in close proximity to the volcano,
.is at Precheur and at Saint Pierre, where a foot or two
may have accumulated as a result of two or three consecu-
tive discharges, the ground scraping can generally be meas-
ured in fractions of an inch. This meagreness of deposit
is in part accounted for by the height of the ash-cloud,
which carries the discharge to more distant parts. That the
quantity of ash contributed to the atmosphere by Mont
Pelee was in fact very considerable is proved by the mag-
nificence of the afterglows which followed the setting of the
sun hundreds of miles beyond the shores of Martinique.
We observed the red and orange skies, similar to those
which were associated with the eruption of Krakatao in
1883, for successive evenings on our return from the island,
on September 9, in about latitude 26° 30" north and longi-
tude 68° 30' west (as computed from the midway determi-
nation of the steamer's position) ; September 10, in latitude
30° and longitude 69° 30' ; and September 11, in latitude
33° 45' and longitude 71°. The last position is fourteen
hundred miles in a direct line from Mont Pelee. Unfortu-
nately, the evening of September 12 was cloudy, and it was
impossible to ascertain at that time the further limit of the
display. Weeks later, however, and well through the month
of October, I observed the glows, with still fairly brilliant
coloring, both in New York and Philadelphia. The glows,
as we noted them, began about twenty minutes after the
passing of the sun's disk, and acquired their greatest inten-

256 THE SOUFRIERE OF ST. VINCENT

sity of coloring, a brilliant orange and yellow, in ten or fif-
teen minutes after that time, the period between the sun-
setting and the first luminosity being one of grays and
blues. Before the intense glow itself appeared the higher
reaches of the sky, extending to about 70° from the horizon,
were suffused in pale pink and lilac, which intensified with
the growth of the glow and became nearly brilliant. I had
never noted such an extraordinary coloring of the sky be-
fore, and it appeared every evening, although diminishing
in intensity as we proceeded northward. The finest display
was obtained on the 9th, when our vessel was nearly oppo-
site Jupiter Inlet, on the Florida coast.23

It is impossible to say to what extent the Soufriere
of St. Vincent contributed to the making of these won-
derful phenomena. The ash-cloud of September 3 and
4 was certainly a heavy one, and its driftage was north-
ward, so that there can hardly be a doubt that it contrib-
uted largely in material to the zone of suspended particles
that analyzed the sun's rays. I was informed by a resident
of the island of St. Martin, near St. Thomas, that the red
afterglows had been almost continuous in the northern
waters of the Caribbean basin ever since the eruption oi
Mont Pelee on May 8.

XVIH

THE VOLCANIC RELATIONS OF THE CARIBBEAN

BASIN

Geographers owe to Karl von Seebach and to Profes-
sor Eduard Suess, especially the latter, the first clear state-
ment regarding the structural affinities of the islands
composing the Greater and Lesser Antilles, and their rela-
tion to the two continents lying on either side of them. In
a masterly way Suess has drawn a parallel between the
orographic lines of the European and American Mediter-
ranean basins, and shown how the features that are domi-
nant in the one are made representative in the other.
In both regions we recognize areas of marked and long-
existing weakness in the earth's crust, and in which break-
ages have been progressively taking place and still continue.
Continental masses have broken sectionally into these areas,
and their fragments lie in part scattered about as the islands
of archipelagic seas. Mountain chains have been sundered,
<lisrupted and drowned in the forming oceanic trough, but
their pinnacles also rise at times as islets or ridges from the
surface of the sea. The Eurafrica that was at one time a
single continent is now Europe and Africa; the mountains
of the Alps-Apennine system that swept continuously into
Africa and Asia are now segmented and sectioned, and we
know them in part as the mountains of Sicily, the isles of
Greece, the Atlas Mountains and the Sierra Honda of

17 267

258 VOLCANIC KELATIONS OF THE CAKIBBEAN BASIN

Spain. Around this vast region of weakness, of bodily
subsidences, great ridges have been towered up, and it is
these mountains which are now in part undergoing break-
age. Professor Suess has shown, and in a way that cannot
be easily contested, that where these great continental break-
ages are taking place they are associated with volcanic and
seismic disturbances, as, indeed, one would be obliged to
assume on any theory that connects volcanic outputs with
pressure exerted by an outer crust or shell upon a molten
interior lying a short distance below it, or holds that vol-
canic discharges take place along lines of weakness where
escape of material from the earth's interior is made easy.

We find in and about the Mediterranean basin the active
volcanic cones of Vesuvius, Etna, Stromboli and Santorin,
and the extinct, but hardly less than modern, Castellfullit
Mountains of Catalonia, Spain, the Euganean Hills of
northern Italy, the Alban Mountains of central Italy, the
Tokai and Sator Mountains of the northern Hungarian
plain, and the loftier summits of the Caucasus, Elbruz and
Kasbek, dominating a basin that is structurally a continu-
ance of the Mediterranean. In all these cases it is found
that the volcanoes, whether new or old, stand closely by the
mountain range whose development or destruction brought
them into existence, and usually they define the inner or
concave side of their trend. It was there that the greatest
pressure was exerted and relief from pressure found.

It is not now difficult to recognize a broad parallelism
between the western included waters of the Atlantic basin,
the Caribbean and Mexican Seas — which may properly be

VOLCANIC RELATIONS OF THE CARIBBEAN BASIN 259

termed the American Mediterranean — and the two basins
of the Eurafrican Mediterranean. Both seas lie between
continents, the American less directly so than the Euro-
pean. In both the depth of water is strictly oceanic (up-
ward of twelve thousand feet), and both have lofty moun-
tains associated with them in some part of their periphery.
Again, both have their island groups or lines, and the vol-
canoes that lie close to their shores, whether on them or off
them. It was a brilliant generalization in geology which
assumed that the islands of the Antilles were, in the main,
merely disrupted parts of a once continuous land area,
whose orographic relief was constituted by one of the main
lines of South American mountains ; that the Sierra Merida
of Venzuela, itself a direct continuation of the eastern
chain, or Cordillera Oriental of the Andes, was formerly
continued through the peninsula of Cumana into Trinidad
and the Lesser Antilles, and from there projected into
Porto Rico, Hayti, Cuba and Jamaica. Since the making

»(' these mountains the line has been sundered at different
points by breakages and subsidences, and elsewhere so
:< drowned" within itself as to leave no trace of a surface

xistence. The fate of the mountain ridge beyond the Blue
Mountains of Jamaica and the Sierra Maestra of Cuba is

lot known with full certainty, but the system may be as-

i nncd on fairly secure grounds — as indeed the identity in

Iithologic construction almost proves — to be projected in
Irowned ridges to the Central American coast, and thence
>ntinued into the lofty masses of Honduras and Guate-
uila as the southeastern expansion of the true continental

260 VOLCANIC KELATIONS OF THE CABIBBEAN BASIN

Cordillera — the chain that virgates at, or near, Zempoalte-
pec, in the State of Oaxaca, and continues northwestward
as the Sierra Pacifico or Occidental of Mexico.24

Whatever may be the exact relations of the low line of
heights of the Isthmus of Panama and of the higher eleva-
tions of Costa Rica, it is certain that they have little in
common either with the main Andes in the south, or with
the Rocky Mountains in the north, and seemingly they are
only a secondary or insular ramification which has been
forced up between bounding lines of pressure, or been left
standing as a part of a broken arm of the Cordillera. The
Antillean relations that have been sketched above assume
as one of their expressions the not improbable eastward ex-
tension of the ancient Pacific border, perhaps even to a
position not far removed from the western contour of the
Lesser Antillean islands as it exists to-day, and touching
the southern confines of what are now Cuba, Hayti, Porto
Rico, etc. Beyond this border may have stretched east-
ward or northeastward, to a long distance, a continental
area that was largely continuous with South America.
And for any facts that geology has to show to the contrary,
this eastward extension of the southern continent may well
have continued, as has been argued by some geologist
quite into the Old World, uniting at least with Africa ; fo
there is good reason to believe that the southern basin
the Atlantic Ocean came into existence only at a later day.

The islands of the Lesser Antilles as we to-day recog-
nize them are constituted of two groups, an easterly and
westerly, which in close position form a crescentic line ei

■

VOLCANIC RELATIONS OF THE CARIBBEAN BASIN 261

tending from Trinidad to the eastern extremity of Porto
Rico, or across seventeen degrees of latitude. The outer or
Atlantic islands, which occupy the convex side of the cres-
cent, are fundamentally of limestone or conglomerate con-
struction, joined to more ancient igneous and metamorphosed
rocks, and are of a continental type, while those of the
inner side are volcanic, and, counting from their principal
members, — Saba, St. Kitts, Nevis, St. Eustatius, Redonda,
Montserrat, Guadeloupe, Dominica, Martinique, St. Lucia,
St. Vincent, Grenada, — about a dozen in number. These
volcanic islands, which all bear evidences of recent volcanic
activity and belong to a period of no great geological an-
tiquity,— perhaps nowhere more ancient than the Middle
Tertiary, — unquestionably define one of the lines of greatest
weakness in the Caribbean or Antillean region, and they
stand implanted upon or adjoined to the old continental
basement, whose fragmented parts still appear in such re-
mains at St. Thomas, St. Croix, Auguilla, Antigua, the
eastern island of Guadeloupe, and part of Barbados, —
islands of sedimentary construction, and which after their
subsidence have in part been built up by organic growth
and volcanic discharges. No more extraordinary series of
volcanoes is to be found anywhere than that of this inner
line of islands, which have sometimes been designated the
Caribbees, and nowhere is a volcanic disposition to be found
that is more beautifully identified with terrestrial move-
ments, whether of subsidence or breakage. The Lesser
Bunda Islands, Japan and the Aleutian Islands alone pre-
sent parallels. Both on the east and the west, i.e., on the

262 VOLCANIC KELATIONS OF THE CAKIBBEAN BASIN

Atlantic and Caribbean sides, the islands rise rapidly from
deep water — more rapidly on the inner or western side —
and between each two placed north and south, although
the interval may not be more than twenty or twenty-five
miles, or even less, the separating water has in most cases
a depth of at least three thousand feet, and frequently
much more. The islands, again, present the extraordinary
peculiarity of having their highest summits brought to ap-
proximately equivalent heights, or at least to levels which
have no marked preeminence ; thus, Saba, which is hardly
more than a rock rising from a fairly deep sea, is 2000
feet high; Mount Misery, on St. Kitts, is 4300 feet;
the Soufriere of Montserrat, 3000 feet; the Soufriere of
Guadeloupe, 4070 feet ; Diablotin, on Dominica, 4740 feet ;
Mont Pelee, on Martinique, about 4300 feet ; the Sou-
friere of St. Lucia, 4000 feet; and the Soufriere of St.
Vincent, 4050 feet. It is not possible to say at this time
to what extent these different volcanic masses may be united
with one another in the trough of the sea, and there form a
continuous volcanic ridge with elevations of seven thousand
or eight thousand feet rising out from it. It would seem
more likely that their connecting bond is the continental
basin, on whose crest, or along whose fractured parts, the
volcanoes have been built up. This conception is seemingly
more in harmony with what we know of the linear dispo-
sition of volcanoes elsewhere, as, for example, in the penin-
sular and insular tracts of extreme Asia, the Aleutian ]
lands, etc.

In assuming in the Caribbean and Gulf basins tw(

VOLCANIC RELATIONS OF THE CARIBBEAN BASIN 263

great subsiding areas, one is not necessarily forced to the
assumption that their origin as such dates from the same
period of time, any more than we accept that the two basins
of the Mediterranean were necessarily formed contempora-
neously, or that the eastern basin is of the same age as the
Black Sea. But they have become isochronic, so far as
their present dynamics are concerned. They break, squeeze
and press, and as a resultant, lands are folded up and
volcanic discharges brought to the surface. There are no
facts in geology that are more difficult to establish than those
that are associated with the first appearance or making of
land-masses and the causes which have brought them into
existence; and much room for doubt must always be per-
mitted in the interpretations of the conditions that suggest
themselves in inquiries of this kind. In the case of the
Antillean region, however, it may be assumed as fairly well
established that the singular peninsular extension of the
United States, the State of Florida, is the resultant of a
lateral thrust, with upfolding, brought about by the sub-
sidence or deepening of the Gulf Basin ; and one may
accept with nearly equal certainty a like or correlative ex-
planation for the existence of the peninsula of Yucatan.
We may, indeed, assume with De Montessus the hypothesis
that the comparatively recent upheaval of parts of the
Lesser Antilles is in itself merely the expression of an up-
thrust between two subsiding basins — the Atlantic on one
side and the Caribbean on the other.25

Were we to seek for an absolutely homologic equivalent
of the American Mediterranean basins in the Mediterranean

264 VOLCANIC RELATIONS OF THE CARIBBEAN BASIN

region of Eurafrica, it would be impossible to find it, sin<
the continental relations of the two regions are not wholly
alike, nor are the mountain parts similarly placed. But it
is immaterial how the individual parts are placed geographi-
cally or how they are interrelated — their geologic aspect oi
Antlitz is fundamentally the same. M. Michel Levy has
latterly made a comparison between the two regions, and ha*
assumed a homologic equivalent between the Caribbean am
the Gulf basins on one side and the iEgean and Black Se*
on the other — the Black and Gulf seas being the include(
basins in the two cases, the Dardanelles, Bosporus and thi
Strait of Yucatan the connecting waters, and the volcani<
Caribbees and the Candian islands the concave outer rii
marking the breakage of the main basins. This comparison
is interesting as it recognizes an existing homology, but it
can hardly replace the broader comparison which is forced
upon us by the larger regions of which the Euxine-iEgean
is merely a part.*

The boundaries of the region of weakness that is included
within or touched by the Caribbean-Gulf basins may be
roughly drawn from the western coast of Mexico to the
Lesser Antilles, or over an east-and-west extent of thirty-
six degrees of longitude, and from the northern parts of
South America to Porto Rico and the lower parts of the
Mississippi Valley. Practically the whole of Central
America is included in this region, whose area may be
approximately put at twice that which is represented in the

* Revue Generale des Sciences, June, 1902.

VOLCANIC KELATIONS OF THE CARIBBEAN BASIN 265

Mediterranean region of Europe. Nearly the whole of this
tract, and much of the region that immediately adjoins it,
is characterized by violent seismic and volcanic disturb-
ances, and probably no region of the globe, with the ex-
ception of that of the Molucca Seas, has been witness to
greater catastrophic events and to a grander concentration
of volcanic figures than this one. One has only to recite a
few of the more salient events of modern date in the course
of these phenomena to properly punctuate the history of
this region : the eruption of Jorullo, in Mexico, in 1759 ;
the destruction, by earthquake, in 1773 of the city of Gua-
temala (Antigua) ; the formation of the volcano of Izalco, in
Salvador, in 1793 ; the earthquake of Caracas, in 1812 ; the
eruption, in April, 1812, of the Soufriere of St. Vincent;
the catastrophic eruption in 1835 of Coseguina, in Nicaragua
— one of the most violent eruptions recorded in history ; the
destruction by earthquake of Cartago, in Costa Rica, in
1841 ; and the rapidly following events of this year :
January 16, destruction by earthquake of Chilpancingo, in
Mexico ; April 18, destruction by earthquake of Quezalte-
nango (and other towns), in Guatemala ; and May, the erup-
tions of the Soufriere and Mont Pelee, in St. Vincent and
Martinique.

There is perhaps nothing that so clearly establishes the
unity of the Gulf- Caribbean region as a region of far-
reaching instability as the broad range of its seismic and
volcanic phenomena and the correspondent relations which
they teach. No succession of events could present this fact
more lucidly than the events of the early part of this year,

266 VOLCANIC EELATIONS OF THE CAKIBBEAN BASIN

1902, when disturbances of one kind or another were de-
veloped over a linear area of nearly or quite two thousand
five hundred miles, extending from Colima, in Mexico, on
the west, to Martinique on the east. The areal distribution
of these occurrences is, indeed, so vast that one is almost
prompted to deny the existence of any true relation binding
them together ; but the evidence obtained from similarly
concurrent events in former periods of time leaves no room
for doubt that the association, which naturally fastens itself
upon the mind, does in fact exist. The synchronism in the
time periods of the eruptions of the Soufriere of St. Vincent
and Mont Pelee, as developed in their recent activities, is
too patent to permit of any question being raised as to their
relation to a common disturbing cause ; and perhaps not
before has such a close relation been recorded. The cata-
clysm of May 8, in Martinique, was preceded by one day by
the main eruption of the Soufriere, which, however, con-
tinued in nearly full activity for twenty-four hours after-
wards ; the Pelee eruption of the 20th of the same month
was preceded, with a nearly equal time interval, by a
second eruption of the Soufriere ; while the second death-
dealing eruption of Pelee on August 30 was followed four
days later, and after an established period of quiescence, by
what seems to have been the most violent of all the recent
eruptions of the Soufriere, on September 3-4.

A careful inquiry and examination made at several of
the other volcanic islands lying in the chain of the Lesser
Antilles, St. Lucia, Dominica, Guadeloupe, Montserrat,
and St. Kitts, all of which have soufrieres or craterlets

VOLCANIC KELATIONS OF THE CAEIBBEAN BASIN 267

emitting sulphurous or heated vapors, establishes the in-
teresting fact that their points of activity were not even to
the slightest degree influenced by the eruptions of early
May — the crateral bodies of water, whether standing or
boiling, retaining their old temperatures, and giving out
neither more nor less of vapor. This condition is made to
appear the more surprising in the case of the Soufriere of
St. Lucia, an island that stands half-way between Mar-
tinique and St. Vincent. The island thus appears side-
tracked, so far as the existence of any connecting fissure
may be postulated. It should be noted, however, that the
position of the St. Lucia Soufriere is not longitudinally
concurrent with the positions of Pelee and the Soufriere
of St. Vincent, lying considerably to the eastward. And
it is remarkable, or at least noteworthy, that just westward
of this island, seven to ten miles beyond the coast, marked
oceanic disturbances, taking place at the time of the
great land eruptions, were observed, and were considered
to point to true eruptions having their origin on the sea-
bottom.

As in 1812, the great May, 1902, eruption of the
Soufriere was preceded by violent seismic disturbances in
the northern part of South America, particularly ac-
centuated in Colombia and Venezuela, and in closer chro-
nologic harmony by the great earthquake which on April
18 destroyed the city of Quezaltenango, in Guatemala —
seemingly the most destructive earthquake in the western
hemisphere since the one which in 1812 wrecked Caracas.
So close, indeed, is this association, and so intimately cor-

268 VOLCANIC EELATIONS OF THE CAKIBBEAN BASIN

related appear to be the volcanic and seismic phenomena of
the vast Caribbean region, that Professor Milne has ventured
the suggestion that it was this earthquake, or rather its
prophetic force, which brought about the eruption of Pelee.
However possible or impossible it may be to prove the cor-
rectness of this view, it is certainly very interesting an
suggestive.*

As regards the intensity of the volcanic and seismic
conditions of the Gulf-Caribbean region, it has frequently
been asserted by geologists and others that it is rapidly on
the decline, and that we could look to a comparatively near
period when a full or nearly full condition of stability
would be established. That there has been a marked dimi-
nution in these phenomena since a prehistoric period, when
the volcanoes were first formed, or for a long period after
their formation, does not, it seems to me, admit of doubt ;
but I fail to find the evidence that points to any recent de-
feasance of power or to that near future of quiet repose
which is assumed to follow dormancy. In various papers
discussing the relative merits of the two interoceanic canal
routes, Nicaragua and Panama, I have sought to point out
the fallacy of the notion that a half century or more in the

* A violent earthquake with sharp detonations was noted at d
rupano, on the Venezuelan coast, on the night of August 30, at about
nine o'clock. It is an interesting fact that almost coincidentally witl
the construction of the volcanic cone in the Lake of Ilopango, in Sal-
vador, there were violent seismic disturbances, with a southwest
northeast movement, in the Vuelta-Abajo district of Cuba (Januar
22-23, 1880).

VOLCANIC RELATIONS OF THE CARIBBEAN BASIN 269

history of an active or sem inactive volcano serves as a
proper guide to the elucidation of the possibilities of such
volcano or that it is necessarily in any way a measure of
the volcano's potential energy. It seemed to me far more
probable, seeing that we had in the 1835 eruption of
Coseguina one of the greatest paroxysms of the earth's
history, that the volcanic and seismic phenomena of at least
a part of the Caribbean region gave indications of an in-
crease rather than of a decrease of power, and I pointed
out the bearing of this condition on the problem of canal
construction. Since the appearance of these papers, the
world has been startled by the destruction of Chilpancingo,
on January 16 ; the destruction of Quezaltenango, on April
18 ; the eruption of the Soufriere on May 7 ; and the
death-dealing eruptions (besides other eruptions of almost
equal intensity, May 20, June 6, July 9) of Pelee on May
8 and August 30. These, together with the long-continued
eruptions of Colima, in Mexico, now extending through a
period of ten years, appear to me to be part of one and the
same general disturbance in a localized, even though vast,
area of the earth's crust. As to the future, and what par-
ticularly concerns the forces of the Lesser Antilles, it is
<liflicult to postulate; but there does not appear to me any
good reason for assuming that we are about to enter upon a
condition of peace. Rather should I believe that we may
be facing a period of long-continued, even if interrupted,
activity ; and that we may even be nearing a period whose
distinguishing characteristics may be cataclysmic. The
Caribbean basin is recognizably one of breakage, and its

270 VOLCANIC KELATIONS OF THE CAKIBBEAN BASIN

phenomena can easily be those that result from this con-
dition.*

* Since the above was written comes the intelligence of renewed
outbreaks (October 15-16) of the Soufriere and of the violent erup-
tion of the volcano of Santa Maria, or of a minor cone near by, in
Guatemala, standing close to the field of Quezaltenango. A loss of
life of five thousand is reported — a number that may possibly be
exaggerated.

SAINT PIERRE AND MONT PELEE IN 1766

XIX
THE PHENOMENA OF THE ERUPTION

The general characteristics of the great eruption of
May 8 may be briefly summed up as follows: For two
weeks and more prior to the event Pelee had been in rap-
idly increasing activity, emitting clouds of ashes and sul-
phurous vapors, and opening its crater on the southwestern
flank of the mountain (in the ancient basin of the fitang
Sec) on April 25. At this time the sulphur vapors had
accumulated in such quantity in Saint Pierre that respira-
tion was made difficult, and animals dropped dead in the
streets of the city. On May 2 the ashes had so far ob-
scured the roads as to compel a cessation of traffic, and
three days later, shortly after noon on May 5, occurred the
discharge of the avalanche of boiling mud which over-
whelmed the Usine Guerin. This stream, travelling with
express-train velocity, issued from the basin of the Etang
Sec, and followed down the course of the Riviere Blanche.
From this time up to the 8th, during which interval torrents
of volcanic water were deluging and destroying towns and
villages, — Precheur, Basse-Pointe, etc., — the unrest of the
volcano was rapidly travelling to a climax, and on the morn-
ing of the fatal day, without particularly active symptoms
presaging the storm, the blow fell with almost lightning-like
swiftness. The issuing explosive and exploded cloud left
the crater at almost exactly eight o'clock, and at two min-

271

272 THE PHENOMENA OF THE EEUPTION

utes after eight the destruction of the city had been accom-
plished. Saint Pierre fell before a hot tornadic blast, whose
sweep could not have been less than from one to two miles
a minute, — perhaps much more, — tumbled into ruins, and
was in greater part consumed by an immediately following
conflagration. A not particularly heavy fall of ashes and
lapilli came close upon the wake of the destroying blast,
and almost at the same time a fall of rain, whose duration
appears to have been less than an hour.

In this destruction, with few exceptions, all the inhabi-
tants were annihilated, and all the evidence points to the
conclusion that in by far the greater number of cases death
was either very swift or almost instantaneous. Some few
lingered on, and two appear to have entirely survived.
Death may have been due to a number of causes, directly
related to the crushing of a city under the force of a vio-
lent hot tornadic blast, but primarily it appears to have
been the result of scorching and asphyxiation (the inhala-
tion of an extremely heated vaporous [or gaseous] atmos-
phere). The measure of the work done by electric dis-
charges has not yet been clearly determined. Seemingly
not less than thirty thousand lives were lost in this catas-
trophe, representing the entire population of Saint Pierre
and the people of a number of adjoining faubourgs and set-
tlements, the zone of most destructive devastation being
measured on the ocean front by the interval which sepa-
rates the anse immediately north of Carbet and Sainte-
Philomene. In the middle line or zone of the sector of
devastation the destruction, following the area of concen-

THE PHENOMENA OF THE ERUPTION 273

trated force, is necessarily most complete. In it the houses
have been most thoroughly wrecked — the human bodies
most thoroughly annihilated. Few of the corpses showed
any vestige of clothing covering the body ; and none
directly within this zone, excepting the prisoner Ciparis
and a certain Leandre, appear to have been so little
burned as to be able to survive their wounds. Laterally to
this zone of greatest destruction the force of annihilation
was a gradually decreasing one, to the end of permitting
houses to stand and the corpses to retain their covering;
and in the further exterior, to inflict wounds of a purely
scorching nature which were not necessarily fatal or even
of consequence.

The zone of absolute destruction is a comparatively
small one, and probably does not much exceed eight or nine
square miles ; but considerably beyond it extends a region
of minor devastation, over which the vegetation has in great
measure been destroyed, temporarily at least, by singeing,
cindering, and the weight of fallen ashes. The explosion
of May 8, while being responsible for the destruction of the
life of Saint Pierre and of its associated settlements, is only
in part responsible for the ruined aspect of the city as we
now see it ; the eruption of May 20, which was perhaps as
forceful as the one that preceded it by twelve days, gave
new characteristics to the ruined city, and the condition of
orientation which it to-day presents.

Whatever seismic movements may have accompanied or
preceded the great catastrophe, it is certain that, if they
existed at all, they must have been of very minor conse-

274 THE PHENOMENA OF THE EEUPTION

quence, otherwise some record beyond a passing notice
would have been made of them in the Saint Pierre jour-
nals. And it is a fact that no earthquake shock was noted
at Fort-de-France on the morning of the 8th, nor, indeed,
at any time previous to August 24, six days before the
second death-dealing eruption of Mont Pelee. In this
negative aspect the eruptions of Pelee seem to differ from
those of the Soufriere of St. Vincent. The barometric
records kept at Saint Pierre indicate a remarkable atmos-
pheric stability during several days preceding the storm,
the mercury column registering regularly, up to and in-
clusive of the 7th of May, seven hundred and sixty-two
millimetres, only once falling to seven hundred and sixty-
one ; it may be that early on the 8th, as the sudden move-
ment of the needle in M. Clerc's aneroid possibly indicates,
there was a sudden or marked fall, but of this we have no
record ; nor is any abrupt change, except that represented
by a momentary depression of three millimetres, indicated
in the registry of the Meteorological Observatory of Fort-
de-France. It is certain that a heavy counter-gust swept
to the volcano immediately after the outburst, probably
drawn to the mountain by a condition of partial vacuum
which followed the displacements in the atmosphere due
to the successive explosions — the condition that in St.
Vincent, during the Soufriere eruption, permitted windows
to be smashed in by outflowing boulders and lapilli on the
side turned away from the volcano.

As a marked negative feature of the Pelee eruptions is
the absence of lava-flow, a characteristic which also marked

THE PHENOMENA OF THE EKUPTION 275

the earlier eruptions of 1851 and 1762. Yet the early-
history of the volcano plainly shows that the prehistoric
eruptions were largely accompanied by extravasation of
flowing magma, which in their later stages or periods was
mainly andesitic in character. That a molten magma now
rises well into the throat of the volcano is indisputably
proven by the ejected pumiceous particles that are so
freely ejected and by much of the exploded glassy ash, as
well as by the lavaeform bombs that lie about in fairly
large numbers. The fact that this contained lava was not
thrust out as a flowing sheet from the mouth of Pelee can
hardly be taken, in a comparative study, as a measure of
the force of the volcano, as manifestly the power to lift will
be largely determined by the weight or height of the column
to be lifted ; and our present geological knowledge does not
permit us to state this for Pelee or for any other volcano.

It hardly admits of a doubt that several of the later
paroxysmal eruptions, those of May 20, 26, June 6, July 9
and August 30, for example, shared the general character-
istics of the one of May 8, or were similarly constructed.
The personal observations of the officers of the Pouyer-
Qtiertier made on the eruption of the second date, of Drs.
Flett and Tempest Anderson on that of the fourth, and my
own on the fifth, point clearly to this conclusion. The
main phenomena were either in whole or in part the same.
My investigations and inquiries made on August 30 and
September 1, immediately before and after the issuance
of the tornadic blast which annihilated or invaded Morne
Rouge, Ajoupa-Bouillon, Morne Capot, Morne Balai and

276 THE PHENOMENA OF THE EEUPTION

the heights of Bourdon, and swept another two thousand of
Martinique's inhabitants from existence, confirm me in the
belief that the principal agent in this later eruption, and not
unlikely also in the earlier ones, was superheated exploded
steam, charged in part with particles of incandescent or
glowing matter. To the showering of the latter upon the
combustible substances of Morne Rouge was due the partial
destruction by conflagration of that city. Whatever acces-
sory gases, besides sulphurous (or sulphuretted-hydrogen?),
may have assisted in the work of asphyxiation or otherwise
killing, has not been ascertained, nor is it known that there
were any such. The simple condition of superheating and
steaming can probably sufficiently explain all the cases of
asphyxiation and scorching, or of death where it was not
brought about through contact with burning or incandescent
particles, electric strokes, crumbling walls, and the violence
of a fully sweeping tornado. The inhaling of an atmos-
phere of the intense heat of many hundreds of degrees, in
places with a temperature possibly much exceeding one
thousand degrees, means practically almost instantaneous
death, and that pronounced heating of the air-passages and
excoriation of the lining membrane of the throat and bron-
chial tubes which were associated with the pitiful cries for
water and the sensation of no air to breathe.

The geologist will never be wholly certain as regards the
precipitating cause of the catastrophe — or more broadly,
the series of catastrophic events of which the eruptions of
Mont Pelee formed only a part. In the chapter on " The
Volcanic Relations of the Caribbean Basin,' ' I have

THE PHENOMENA OF THE EKUPTION 277

tempted, following Suess and others, to point out the genetic
connection of the different island groups of that region, and
their relation to a past orographic unit and continental
disruption. The numerous disturbing incidents, whether
volcanic or seismic, that have latterly crowded themselves
into the history of this zone or region — as, indeed, they had
already done two or three times before in a period of a hun-
dred years — together with the unquestionably interrelated
manifestations that developed as a part of the synchronic
movement, lead one to believe that all of these disturbances
have a common origin, whose initiative is to be found in a
readjustment of the floor of the Caribbean Basin. This
broad zone of weakness, developed along the northern
confines of the South American continent, and between the
fragmented parts of the ancient Andes (Lesser Antilles) on
the east and the Pacific coast of southern Mexico on the
west, including within its area the greater part of Central
America and the tracts of Mexico lying south of the plateau
(whose permanency as a " region of concussion" has latterly
been well shown by Deckert in his paper : " Die Erdbeben-
herde und Schuttergebiete von Nord-Amerika" *), is seem-
ingly still in a condition of continuous oscillation, and doubt-
less of much fracturing and reacting subsidence. Along its
edges of greatest weakness, and where relief from strain can
most easily be had, do we necessarily seek for the greatest
development of volcanic activities. It is also there that, on
any theory that associates volcanic phenomena with the

* Zeitschrift Gesellschaft fur Erdkunde, Berlin, 1902, pp. 367-389.

278 THE PHENOMENA OF THE EKUPTION

accession of oceanic waters to seats of potential magmatic
force within the earth's interior, we should expect to meet
with violent or paroxysmal outbursts.

The Force of the Explosion. — It is hardly possible, ex-
cept in a very indeterminate way, to establish a comparison
between the explosive force which marked the eruption of
May 8 and that of other great eruptions whose histories are
fairly well known to us. If the measure of this force is to be
read merely from the mechanical work of volcanic decapi-
tation and evisceration, in the amount of solid material that
was thrown out, in the height of projection of some of this
material, and the concussions to which these projections
gave rise, then the eruption of Pelee stands probably far
down in the scale of volcanic catastrophism, being surpassed
by Papandayang (1772), Asamayama (1783), Skaptar Jokull
(1783), Tomboro (1815), Coseguina (1835), Krakatao (1883),
Tarawera (1886), Bandai-San (1888), and perhaps even by
many of the eruptions of Vesuvius, Etna, and Mauna-Loa.
The erupted material of Pelee was not particularly large,
and probably even considerably less than that thrown out
by the Soufriere on the day preceding. The volcano had
been well opened nearly two weeks in advance of the cata-
clysm, on April 25, and the crater had been throwing out
great quantities of ash and lapilli almost unremittingly
since that date. At the moment of the catastrophe, it
would seem that no very great part of the mountain was
raised or hurled into the air. A comparison of ancient and
modern landmarks shows unmistakably that whatever
change was imposed upon the summit or general contours

THE PHENOMENA OF THE ERUPTION 279

of the mountain, this change did not affect the broad aspect
either of the slopes or of the former crest-line, nearly all
the old topographic features having been retained, although
emphasized in part. It is not unlikely that in this erup-
tion some considerable portion of the periphery or floor of
the crater was actually blown out, the fragments coming
from the destruction of which may have constituted the
gerbe de rochers which has been described by M. Thierry
(Comptes Rendus, July 7, 1902, p. 71) and others as having
been projected several hundred feet above the crest of the
volcano.

However easily one may force a comparison between the
expended force of different eruptions, based upon the value of
their mechanical effects, a study of correlative results shows
that this form of comparison is not wholly free from error,
and may lead to serious misconceptions. Thus, comparing
the eruption of Mont Pelee with that of the Bandai-San,
in 1888, we know that the amount of solid matter thrown
out by the former was, indeed, very small. The discharge
of the latter, on the other hand, has been assumed by Pro-
fessors Sekiya and Kikuchi in their official report to have
been one billion five hundred and eighty-seven million cubic
yards, distributed over twenty-seven square miles of surface
(Journal College of Science, Imperial University of Japan,
III, 1890, pp. 91 et seq.)* Yet despite this vast dislocation

* This quantity (1.20 cubic kilometres) is just one-fifteenth of that
which has been assumed to represent the outthrow of Krakatao in
1883 (4.3 cubic miles), and hardly more than one-hundredth (!) of

280 THE PHENOMENA OF THE EEUPTION

and the great tornadic tempest to which it gave rise — a tor-
nado moving with a velocity assumed to have been not less
than ninety miles an hour — the damage wrought, estimated by
the Pelee standard, was (although very great in itself) fairly
insignificant. Only one hundred and sixty-six houses were
destroyed, completely or partially, and less than five hun-
dred (four hundred and sixty-one) lives lost. Nor, indeed,
were the " frightful" detonations that accompanied the
explosion heard at any great distance, — to windward, not
more than thirty miles.

The force of a blast such as that which, in the case of
Mont Pelee, annihilated a compactly built city along a
direct line of nearly or quite two miles can hardly be esti-
mated. Its measure can well be taken from the excess or
non-development of the ordinarily associated volcanic phe-
nomena, as these seemingly gave way to a form of erup-
tivity whose force-centre lay in a different path. It is
reasonable to assume that had Pelee been a sealed mountain
up to the time of its first great eruption, the mechanical
effects of disruption might have presented themselves on a
scale vastly more imposing than that on which they were
actually found. Professor Judd, reviewing the character-
istics of the Krakatao eruption, — which he assumes to have
been developed on a " much smaller scale than several other
outbursts which have occurred in historic times," — asserts

what (28.6 cubic miles) Yerbeek believes must have been blown out
by Tomboro in 1815. — Koyal Society Keport on Krakatao Eruption,
p. 439.

THE PHENOMENA OF THE ERUPTION 281

that "in the terrible character of the sudden explosions
which gave rise to such vast sea- and air-waves on the
morning of the 27th of August, the eruption of Krakatao
appears to have no parallel among the records of volcanic
activity.' ' We may say in the same way of Pelee, that in
the intensity and swiftness of its death-dealing blast, the
vast disturbance caused by it in the magnetic field, and the
extraordinary brilliancy and remarkable character of its
electric phenomena, the eruptions of May 8 and of later date
stand unique in the records of volcanic manifestations.

Distribution of the Products of Eruption. — It has
already been stated that the eruption of May 8, as well as
the eruptions of later date, were entirely free of open lava-
flows, and that the solid products of eruption consisted ex-
clusively of mud-materials, lava-bombs, boulders, lapilli,
pumice and ash. None of the more massive ejecta were
thrown to any great distance from the volcano's mouth.
Their location, except where subsequently disturbed, is
almost exclusively on the upper slopes of the mountain, at
distances usually within close range of the summit; and
those of larger size, measuring several feet in diameter or
very much more, where occupying a more distant position,
have in most cases undergone secondary transportation by
rolling down the nearly unobstructed slopes. When nearly
opposite the lower lip of the crater on August 24, just in ad-
vance of a fairly powerful eruption, I was witness to giant
boulders or rock-masses sweeping down the exterior slope of
the great fragmental cone. Some of these, I believe, could
not have been less than twenty or thirty feet across — per-

282 THE PHENOMENA OF THE EEUPTION

haps even considerably more. Where rolling over an open
or unobstructed course the distance covered was fairly-
great, perhaps reaching to two or three miles. The rising
plane that forms the parting between the Rivieres Blanche
and Seche, and over which swept the mud-flow of May 5,
was, when I passed it on September 6, a week after the
eruption of August 30, checkered with great boulder-masses,
some of them of very large size, and bearing testimony to
an extraordinary propulsive force resident in the volcano.
For, whether rolled to their present positions or directly
thrown to them, they must have risen through the volcanic
chimney. Doubtless, some of these ejected rock-boulders
were merely fragments of the united or fused cindered
masses that in part construct the summit of the eruptive
cone ; but others were as unquestionably true ejected
masses that had been hurled over the crest, just as they
were at the time of the eruption of May 8.

When I first reached the rim of the crater on June 1,
at a time when the caldron was swirling with steam and
vapor, it seemed to me and to my associates that a part of
the central fragmental mass (cone of activity) was con-
structed of a vertical wall, and so this feature is represented
in an article published by me in McClure's Magazine
(August, 1902), and illustrated by that very accurate art-
student of nature, George Varian. The feature was a
puzzling one, and, unfortunately, it could only be seen in
snatches through rifts in the enveloping clouds. When
opposite the cone on the southwest side on August 24, as I
have elsewhere noted, two giant " horns" — one vertical and

Expl. Heilprin

Underwood & Underwood, Stereos. Photo., New Yoric, lopyritfut, 190a
A COCOANUT-GATHERER— ASSIER

THE PHENOMENA OF THE ERUPTION 283

the other horizontal (but projecting) — protruded over the
summit of the cone, appearing perfectly black ; but even
rith a powerful glass their characteristics could not be de-
termined. Professor Lacroix manifestly saw something of
the same kind at a later day, for in a report published in
the Comptes Rendus (October 27, 1902) he says that the
cone does not appear to be constructed entirely of ejected
material, but to be formed in part of very pointed and ver-
tical-sided needles, which recalled the front of the andesitic
flows of Santorin. Can we here be dealing with a vertical
upthrow and partial overflow of flowing lava? Lacroix,
indeed, hazards the assumption that we may have before us
the construction of a cumulo-volcano.*

Of the ejected material of the volcano that was thrown
to a greater distance than five or ten miles there do not

* More recently (Comptes Rendus, November 10) Lacroix has
declared that the cone is solid, without central orifice, and that the
normal lofty pennant does not issue from its summit, but from the
sides and from the interval which separates the cone from the bound-
ing outer wall of the crater. This is an interesting observation, and
shows, if it is accurate, that the cone has undergone material change
since August 24. At that time, as my photographs plainly prove, the
pennant was rising centrally from a truncated cone, whose outer walls
were mainly of a fragmental character. Only at intervals before the
eruption which we witnessed later in the day was it supported by the
side-columns of steam. Yet I suspected at the time, from the way
in which the smoke-column ascended, that the chimney must have
been blocked, which prevented a free and open flow. It is not un-
likely that a blocking of this kind frequently takes place, and is
accessory to some of the paroxysmal outbursts which distinguish
volcanoes of this class.

284 THE PHENOMENA OF THE EEUPTION

appear to have been many fragments that were larger than
an egg ; nearly everything was, indeed, very much smaller
— particles measuring an inch or less. The finer ash was,
of course, drifted off to great distances. Practically the
whole of Martinique received some sort of a deposit. Per-
haps the farthest distance at which the drifting ash of Mont
Pelee has been noted in the lower regions is two hundred to
three hundred miles, although there can be no question that
the areal distribution is much more extensive than would
seem to be indicated by these limits. The inquiry in this
field is necessarily complicated by the discharges from the
Soufriere of St. Vincent, whose driftage preceded that (of
the main eruption of Pelee) by one day, and by the number
of discharges which preceded the main incident. Whether
applying either to the Pelee or the Soufriere it is interesting
to note that : on May 8, six hundred and sixty miles east
by south of Pelee, in latitude 13° 22' N., and longitude 49°
50' W., a falling dust was noted by the barque Beechwood,
bound from Salaverry to New York.

On May 8, two-thirty a.m., the barque Jupiter, from
Cape Town, reported receiving dust at a distance of nine
hundred and thirty miles east-southeast of St. Vincent
(Meteorological Office Pilot Chart, November). This seems
to be the farthest distance of driftage on the sea which has
been observed, and if the materials are referable to the great
eruptions, then manifestly they are part of the eruptions of
the Soufriere and not of Pelee. The time period would
then indicate a velocity of travel of nearly sixty miles an
hour, nearly treble that which (as will be seen farther on)

THE PHENOMENA OF THE ERUPTION 285

may be assumed in the passage of the upper dust-strata
which carried with them the phenomena of the afterglows.
Other observations on falling dust are contained in the
logs of the steamship Coya, bound from Montevideo to New
York (fall noted in the evening of May 7, ten-thirty o'clock
—11° 23' N., longitude 57° 52' W., two hundred and fifty
miles east-southeast of St. Vincent) ; the barque Eleanor
M. Williams, from Conetable Island to New York (fall,
May 8, three to eight p.m., in latitude 14° N., longitude
57° W., two hundred and fifty miles east of Martinique);
the steamer Porto Rico, on June 7, lying at anchor near
Ponce ; and ship Monrovia, from Rio de Janeiro (at four
p.m. of the 8th, two hundred and forty miles southeast of
Barbados). It is interesting to note that nearly all the
long-distance observations were made on the windward
side of the islands, which would seem to show that the
greater part of the dust was projected through the zone
of the trade-winds, and carried eastwardly in the path of
the alternating (or "anti-trade") winds. The royal mail
steamer La Plata (Nature, June 26, p. 203) notes falling
• lust on May 9, six p.m., one hundred miles west of St.
Lucia.

The Steam- (Ash-) Cloud. — This appears white, gray,
yellow, reddish, brown and almost black, depending upon
the quantity of ashes with which it is encumbered, the pure
white indicating a nearly pure steam-cloud. When the
volcano is in moderate activity the panache or pennant rises
in gentle outflowing sweeps, little different from the curling
smoke of high chimneys. Even in this condition the vis-

286 THE PHENOMENA OF THE EEUPTION

ible part may rise to a mile or two above the crest of
volcano. In a more violent or paroxysmal stage the ex-
tended vapor boils up or out with great force, disengages
itself in rapidly enveloping puffs and rolls, and constructs
the well-known cauliflower form of clouds. These either
rise straight up, looking as though they had been shot out
of a cannon's mouth, or spiral about in corkscrew fashion,
and give the appearance of being sucked into a central
vortex. It is then that the volcano appears in all its full
majesty — supremely powerful and terrifying. I did not
see anything that could properly be said to look like the
" pine-cloud" of Vesuvius.

The ascensive force of the steam-column is very great,
and from a number of eye-measurements that were made at
different points I should say that it frequently mounts up
to three or four and five miles. On our descent from the
mountain in the afternoon of August 30, about four and a
half hours before the explosion of that date, there was a
burst which seemed to me to carry the steam-column, nar-
rowed somewhat like a Lombardy poplar, to a height of not
less than six or seven miles. Prodigious though this may
appear, it is still very much less than the steam-cloud which
issued from Krakatao at the time of its great eruption in
August, 1883. That was assumed to rise to nearly nine-
teen miles.* On the same August 30, when the crater was

* One of the artillery officers stationed at Fort-de -France deter-
mined by instrumental measurement the elevation of the steam-col-
umn to have been five thousand metres, or almost exactly three miles.

THE PHENOMENA OF THE ERUPTION 287

boiling from all its parts, and the roar from the ascending
straight column was appalling, I timed the velocity of the
issuing stream with my watch, and found it to be from
one and a half to two miles per minute, and at intervals
even greater. Only when coming near to this column does
one appreciate the violence of its temper, the force that
has projected it into the air and keeps it there ploughing
through the other clouds that have preceded.

It becomes an interesting question to ascertain to what
extent the high flight that has been obtained is dependent
upon the propelling power that shot out the vapor, or is
merely a measure of this vapor's low gravity and expansive
power. We may, perhaps, readily admit that the far upper
zone of this pennant is " floating' ' of its own accord, and
only through consecutive concussions from below feels the
true projecting force of the volcano ; but, indeed, this ad-
mission does not very materially affect the problem, as we
have to consider in this connection not only the outer
column of steam but also that which is contained in the
throat of the volcano, and may even rise from very con-
siderable depths. The fact that so frequently the lofty
pennant is shot in a straight line entirely through the zone
of the trade-winds, as many of my photographs show, and
perhaps even through the zone of the anti-trades, naturally
proves that, at certain times at least, the propelling power is
responsible for the full or nearly full height that the cloud
attains.

In the chapter on " The Geography of Mont Pelee" I
have stated that it appeared to me that not only were the

288 THE PHENOMENA OF THE EEUPTION

eruptions taking place from the summit of the new frag-
mental cone that has been erected over the floor of the basin
of the Etang Sec, but also from still-existing parts of this
ancient floor, and I even ventured the assertion that the de-
structive blast of August 30 may have had its origin here,
rather than in the chimney-pot. I was led to this conclusion
by the violence of the steam eruptions coming from the great
depths of the crater, and their gradual crowding over to the
side turned to Morne Rouge — the location whence seems to
have issued the explosive tornado of May 8. This view
seems also to be shared by Professor Lacroix, who observes
(Comptes Eendus, October 27, 1902, p. 673) : " It would ap-
pear that it is from the interval between the walls of the
crater and the base of the cone, as well as from the flanks
of the cone itself, that the columns of gas and vapors, at
times of calm, ascend vertically to prodigious heights ("77
semble que c'est de Tintervalle situe entre les parois du crater e
et la base de ce cone, ainsi que des Jlancs de celui-ci qui sortent
actuellement les colonnes de gaz et de vapeurs qui, les jours de
calme, montent verticalement a une hauteur prodigieuse").
The plate (decimaprima, 28a) illustrating the eruption of Ve-
suvius in 1767, and contained in the " Gabinetto Vesuviano"
of Delia Torre (1797), perhaps represents the same form of
double synchronic activity.

Quantity of Ash-Sediment Discharged. — I have else-
where incidentally stated that there was probably more steam
being thrown out by Pelee at any particle of time during
the 30th of August than was escaping from all the engine
jets in the world taken collectively — from stationary engines,

Photo. HeUprin

SMOKING" FROM THE NEW FRAGMENTAL CONE
August 24, 1902

THE PHENOMENA OF THE ERUPTION 289

locomotives, steamboats, etc. Professor Israel C. Russell,
in a paper on the " Volcanic Eruptions on Martinique and
St. Vincent" {National Geographic Magazine, December,
1902), gives expression to this quantity by assuming the
areal contents of a steam-cloud rising to three or four miles
to be about 4,000,000,000 of cubic feet. He further assumes
such a cloud to be charged at its minimum with one per
cent., or 40,000,000 cubic feet, of solid matter, and that it
is regularly replaced every five minutes by another cloud
(the rate of ascent here considered being about three-
quarters of a mile per minute, which is very much less
than I found it to be on August 30). Hence, the discharge
of solid matter from the crater will be every five minutes
40,000,000 cubic feet. In all of these data I believe that
Professor Russell has understated, rather than overstated,
the conditions as they exist, and perhaps very much so, but
they serve as an interesting basis for further analysis and
comparison.* The discharge of 40,000,000 cubic feet of
solid sediment every five minutes means 480,000,000 cubic
feet per hour, and 11,520,000,000 cubic feet per day of
twenty-four hours, which is one and a half times the quan-
tity (°f sediment) that is discharged by the Mississippi River
in the course of a whole year ! In other words, if these
figures are in any way accurate, the sedimental discharge
from the crater of Mont Pelee, taken at a minimum valua-

* Professor Russell himself seems to incline to the opinion that
ten per cent, would more nearly represent the proportion of solid
matter contained in the cloud.

290 THE PHENOMENA OF THE ERUPTION

tion, is in any period of time during a condition of mod-
erate eruption more than five hundred times that of the
Mississippi River, and consequently considerably greater
than that of all the rivers of the world combined. This
daily discharge from Pelee of 11,520,000,000 cubic feet of
sediment would raise the level of a region having the area
of Martinique by almost exactly one foot. Mont Pelee has
now been in a condition of forceful activity for upwards of
two hundred days ; can we assume that during this time it
may have thrown out a mass of material whose cubical
contents are hardly less than a quarter of the area of
Martinique as it now appears above the water? One is,
indeed, almost appalled by the magnitude of this work,
and yet the work may even be very much greater than is
here stated. We ask ourselves the questions, What be-
comes of the void that is being formed in the interior?
What form of new catastrophe does it invite ? There can
be no answer to a question of this kind — except in the
future happening that may be associated with this special
condition. But geologists must take count of the force as
being one of greatest potential energy, whose relation to
the modelling and the shaping of the destinies of the globe
is of far greater significance than has generally been con-
ceived.

Flaming Gases. — I do not think that we are quite justi-
fied in denying the presence of flames in the visible phe-
nomena of Pelee. Burning gases and issuing flames having
been observed in some volcanoes, there is no particular
reason, so far as I can see, why they should not also be

THE PHENOMENA OF THE EEUPTION 291

here. The fact that most of the supra-crateral illumina-
tion, so often described as flame, is merely a reflex from
the glowing, incandescent matter below, is in no way an
answer to positive statements, coming from seemingly
careful, even if non-scientific, observers, which assert that
flames were unmistakably distinguishable in more than one
eruption. Such statements should naturally be received
with caution, but not necessarily immediately denied. If
it may be true that a part of the extraordinary electric
illumination which we witnessed on the night of August 30,
at the time of the destructive eruption, and that others
witnessed on the 25th of the same month, on August 9,
and on May 26 and 28, besides other times, was of a
gaseous nature, as some investigators pretend, then it
becomes easy to believe that burning flames may have been
seen shooting out from or burning around the crown of the
volcano. M. Roux, a member of the Astronomical Society
of France, in his report to Camille Flammarion, claims to
have seen fixed flames (" des feux fixes oVune flamme tres
blanche'1). I saw nothing that was even remotely suggestive
of flame.

Electric Illumination in the Volcanic Cloud. — The cloud
illumination which I have already described as accompany-
ing the eruption of August 30 (Chapter XVI), and the
kind which had been observed several times before as
part of the eruptive activity of Pelee,* certainly constitutes

* See August F. Jaccaci, " Pelee the Destroyer," McClure's Mag-
azine, September, 1902 ; Eobert T. Hill, National Geographic Magazine,
July, 1902 ; Kennan, " The Tragedy of Pelee," 1902.

292 THE PHENOMENA OF THE EEUPTION

one of the most interesting, and perhaps least understood,
phenomena associated with volcanic discharges. Indeed, we
are still hardly in a position to assert that the phenomena
are wholly electric, or whether they may not be in consider-
able part gaseous ; or, again, whether they may not represent
a form of electric manifestation whose peculiarities have
been induced by development in a complex gas-cloud in
place of the ordinary atmospheric one. The figures that I
have represented were sketched immediately after the cul-
mination of the storm, when the ocular impression was still
very distinct. None of the irregular figures — circles, circles
with undulating streamers, serpent-lines, straight lines, etc.
— had the full dazzling quality of the zig-zag lightning that
at times flashed through their field, but appeared extremely
brilliant, yellowish in color, varying at times to purple.
Possibly, this was an indication of the great height of the
clouds and the tenuity of the atmosphere in which they were
developed, a condition which is well known to influence the
character in color of ordinary lightning flashes. The hori-
zontal flashes, and also the serpent-lines, appeared to take
horizontal courses through the clouds, or across spaces
uniting individual fields of cloud, and manifestly their
luminant lines marked a successive development in a
progressive field. I did not myself observe any of these
" bars" terminating or exploding in an end-flash or star, as
some others have stated, but the condition might well have
existed, seeing how many rocket-like bursts appeared in
some parts of the cloud. The display lasted nearly an hour,
almost exactly the duration of the discharge of lapilli and

THE PHENOMENA OF THE ERUPTION 293

ash on the Habitation where we were staying, and during
this time there was a continuous, but not loud, roaring —
perhaps, it would be better to say, rolling — of thunder,
which in regular crescendos and diminuendos seemed to
traverse the entire field of the volcanic cloud.

Whatever may be the precise nature of these extra-
ordinary displays — and only after careful spectroscopic
analysis will we be able to arrive at a positive conclusion as
to their character — it is certain that something similar has
been observed in the eruptions of some (perhaps many)
other volcanoes. The balls of electric fire that have been
described from the ascending steam-column of Vesuvius are
almost certainly a part of the same phenomenon, although
sometimes they are referred to actually falling incandescent
boulders, the same as in the Soufriere eruption of 1812.
Professors Sekiya and Kikuchi, in their report upon the
Bandai-San eruption (1888), speak of innumerable sparks
of fire being seen through the densely falling ashes, with
characters quite different from lightning ; but these inves-
tigators refer to them as being produced " by stones and
rocks striking against each other in the air or falling on a
rocky bed. . . . We could discover nothing to lead us to
believe that there had been combustion or any other heat
manifestations." * It is singular that at the time of our own
observations all of the phenomena were overhead, nothing
appearing in the location immediately about or directly over
the crateral opening.

* Journal College of Science, Tokyo, III, 1890, p. 129.

294 THE PHENOMENA OF THE EKUPTION

Nothing of the nature here referred to seems to have
been remarked as an accompaniment of the Krakatao erup-
tion, but the report of Pond and Percy Smith on the great
eruption of Tarawera, in New Zealand, in June, 1886, leaves
no doubt that the phenomena witnessed there were identi-
cal with those of Pelee. "The electrical phenomena ac-
companying the outburst/' we are told, " must have been on
the grandest scale. The vast cloud appears to have been
highly charged with lightning, which was flashing and
darting across and through it : sometimes shooting upward
in long, curved streamers, at others following horizontal or
downward directions, the flashes frequently ending in balls
of fire, which as often burst into thousands of rocket-like
stars." * It should be noted that some of the electric dis-
play of Tarawera was "accompanied by a rustling or crack-
ling noise . . . probably the same [noise] as is heard some-
times at great auroral displays." It would seem that Drs.
Flett and Tempest Anderson observed a minor exhibition
of this form of electric discharge in the low-rolling black
cloud of July 9.

Atmospheric Stability. — It is a noteworthy fact, and one
that is wholly opposed to the view that violent volcanic erup-
tions are necessarily indicated by precedent atmospheric dis-
turbances, that none of the great eruptions of Pelee fol-
lowed any marked barometric fluctuation. For several days
preceding the May 8 eruption, including May 7, as the
Saint Pierre records show, the atmosphere was singularly

* Transactions New Zealand Institute, 1886 (1887), p. 352.

Painted by George Varian

S. S. McClure Co., Copyright. 1903
THE HEAVENS AGLOW— MAY 26

THE PHENOMENA OF THE EKUPTION 295

impassive, the barometer registering at noon seven hundred
Lnd sixty-two millimetres, and only on one day dropping
seven hundred and sixty-one millimetres. Much the
jne condition of stability was recorded by the barometer
>f the Meteorological Observatory of Fort-de-France during
several days preceding the explosion of August 30. Im-
tediately preceding the event of May 8, and also of June
>, as the observations of M. Fernand Clerc and the registry
of the Pouyer-Quertier indicate, there was a rapid fluctua-
tion with sudden fall (and equally rapid recovery), — the
depression of June 6 amounting to four millimetres, which,
I understand, is quite significant in the island of Martinique,
— but this movement may have been induced as the
result of a terrestrial (seismic) concussion rather than of
a true currental displacement in the atmosphere.* The
eruptions of Tarawera and Bandai-San likewise took place
at times of atmospheric calm, or when the barometer indi-
cated no abnormal depression, "either shortly before or
during the catastrophe" (Pond and Percy Smith). The
great cataclysm of Krakatao was preceded by a night of
raging storm, but as the volcano had really been very
active already before that date this fact loses all significance.
Breislak, describing the great eruption of Vesuvius in June,
1794, remarks upon the stability of the barometer : "Le

* The almost instantaneous barometric fluctuation noted at the
Meteorological Observatory at Fort-de-France on the evening of
August 30, and at the immediate time of the eruption, was from
three to four millimetres. See Appendix.

296 THE PHENOMENA OF THE EEUPTION

tableau des observations meteorologiques . . . prouve que
barometre rCa eprouve aucun changement sensible" ("Voy-
ages dans la Campanie" 1801, p. 216). I think that it
would not be difficult to show from the records of many
eruptions that the state of the atmosphere has little to do
with the development of phenomena of this class.

Counter Atmospheric Current. — The existence of such a
return current, or of a wind directed to the volcano, follow-
ing immediately upon the explosion of May 8 is substan-
tially vouched for in the published observations of MM.
Roux and Celestin, members of the Societe Astronomique
de France, and of others who witnessed the catastrophe at
close range. Some of these describe the wind as being
of almost hurricane force, which swept off the branches and
twigs of the trees that stood in its course, and overthrew or
swept off other objects. On June 6, when the great
ash-cloud swept over Fort-de-France, and announced the
very severe eruption that had just taken place, I particu-
larly noted the extreme velocity with which the normal
clouds of the atmosphere were sailing in a lower zone
directly to the volcano, the appearance being very much as
though they had been forcibly drawn to it. There was, I
believe, no particular movement where I was standing. At
the time of the Bandai-San eruption these counter-currents
appear to have been particularly strong, and perhaps even
did much of the wrecking. Professors Sekiya and Kikuchi
state that the "fearful blasts that wrought such havoc on the
forests and villages of the 15th of July certainly were not
counter-currents of this class, however strong these may

THE PHENOMENA OF THE ERUPTION 297

have been," but " gusts from the volcano." Yet it appears
in the testimony and report of T. Uda, of Koka'i village,
Yama-Kori, who was the nearest reliable witness to the
catastrophe, and only 3.2 miles east-southeast of Bandai-
San, that : " Soon after the eruption a great whirling wind
suddenly swept over the eastern part of the mountain with
great violence, destroying Shibutani, Shirokijo, Ojigakura,
etc." These villages are part of the seven that are indicated
in the official report as having been destroyed. This state-
ment is, therefore, directly opposed to that of Sekiya and
Kikuchi.

Pond and Percy Smith in their report on the Tarawera
eruption remark the issuance of a similar wind: "Soon
after the first outburst, and before the fall of the first stones,
a great wind arose, which rushed in the direction of the
point of eruption with great force, and was most bitterly
cold" (p. 351). The reference to the lowering of the tem-
perature is very interesting. On June 6, in Fort-de-France,
immediately on the coming of the ash-cloud and of the
counter-current there was a perceptible cooling of the at-
mosphere, perhaps by fully ten degrees, and this condition
remained for two or three hours. How much of this may
have been a direct following of the constructed counter-
wind, or due to the cutting off of the sun's rays by the
interposed cloud, I do not profess to be able to say ; but the
suddenness of the lowering of the temperature makes it
almost certain that the phenomenon was intimately bound
up with the coming of the wind. In tropical regions the
blanketing of the sun's rays more generally brings about a

298 THE PHENOMENA OF THE EKUPTION

sultry atmosphere; in this instance, it was one of refreshing
coolness, following closely upon an earlier hot air.

As regards the nature of this counter-current I do not
think it can be questioned that the explanation given by
Sekiya and Kikuchi is approximately the correct one. The
immense volumes of steam that issue from the volcano sud-
denly expand, and in doing so necessarily lower the tem-
perature of the surrounding atmosphere, and also diminish
its pressure — the steam itself undergoing partial expansion.
" To fill the partial vacuum thus produced and to equilibrate
the reduced pressure, there follows an inward rush of air
towards the crater. The strong winds commonly described
as a feature of volcanic eruptions are probably due to this
cause."

Magnetic Disturbances. — The eruption of Mont Pelee
has sometimes been said to surpass all other recorded erup-
tions in the magnitude of the magnetic disturbance which
it occasioned, the electro-magnetic waves that were shot out
causing many hours' disturbance to the magnetic needle at
distances of two thousand and five thousand miles ; and it
has been claimed that this was " the first instance that mag-
netic effects caused by eruptions of distant volcanoes have
ever been recorded at magnetic observatories." * This as-
sertion is linked with the disturbances which were observed
at the United States Coast and Geodetic Survey magnetic
observatories located at Cheltenham, Maryland, seventeen
miles southeast of Washington, and at Baldwin, Kansas,

* National Geographic Magazine, June, 1902, p. 209.

THE PHENOMENA OF THE EKUPTION 299

seventeen miles south of Lawrence, and which, noticed at
both observatories at practically the same instant of time —
corresponding to 7 h. 54 m. local time of Saint Pierre — are
naturally (though not absolutely) associated with the Mar-
tinique explosion. There was a second disturbance of these
needles on May 20, conformably with the second great
eruption of Pelee. The disturbance at the Cheltenham ob-
servatory is stated by Bauer {Science, May 30, 1902) to have
amounted at times to about 1/350 of the value of the hori-
zontal intensity (.00050 — .00060 c.g.s. units) and to from
10' to 15' in declination. Corresponding disturbances have
been noted at Toronto, Stoneyhurst, Val Joyeux (France),
Paris, Potsdam, Pola, Athens, and Honolulu {Terrestrial
Magnetism and Atmospheric Electricity, June, 1902 ; Me-
teor ologisches Zeitschrift, Vienna, XIX, pp. 316-317 ; Comp-
tes Rendus, June 16, 1902), and it is remarkable that all
of these were noted at almost precisely the same moment
of time, corresponding to 7 h. 54 m. of the time of Saint
Pierre. At Athens, as at most of the other stations, no
seismic disturbance of any kind was noted at this time ; on
the other hand, the great earthquake of Guatemala, on
April 18, was impressively registered by the seismographs
at nearly all, or all of the observatories. The most in-
teresting magnetic notation is that of Zi-ka-Wei, China, the
observations pertaining to which were made by M. de Moid-
rey, and are published in the Comptes Rendus for August
11, 1902 : " Ce jour-la, a 7 h. 58 m. (Martinique time), aprH
une longue periode de calme magnetique, noire bifilaire
nidi que un accroissement brusque de la composante horizon-

300 THE PHENOMENA OF THE EEUPTION

tale, qui reste agitee pendant huit heures environ" (p. 322).
The length of duration, and the great distance at which
the disturbance was felt, are alike noteworthy. Zi-ka-Wei
is situated almost exactly on the meridian opposed to
Saint Pierre — i.e., half round the world in distance from it.
This record is specially significant, as during the Krakatao
eruption it was considered doubtful if the instruments of
this station recorded any disturbance that could be correlated
with the events that had transpired in the Sunda Straits.*

The replies to the official inquiries sent out by the
Krakatao Committee of the Royal Society would seem to
indicate that no particular magnetic disturbances were noted
in Bombay, Melbourne or Toronto, and the perturbation
was so slight and of so doubtful a nature in the European
cities that it may be questioned if they were in any way
related to the eruption ; at all events, there was nothing
that was in any way comparable with the magnitude of the
disturbance registered at the American stations. On the
other hand, it seems that some slight magnetic variation
was noted at Para, Brazil, on the day of the Krakatao
eruption. Dr. Van der Stok, the Director of the Batavia
Observatory, noted at the time of the Krakatao eruption a
marked magnetic oscillation which he attributed to the
influence of the magnetic iron contained in the falling ashes.
Probably this disturbance was of the same nature as that

* Marc Dechevrens : "Jene sais si les irregularites magnetiques que
fenvoie out eu aussi une relation avec le bouleversement de Krakatao ; les
magnetogrammes ne montrent rien le 28" — Eoyal Society Eeport, 1888,
p. 473.

THE PHENOMENA OF THE ERUPTION 301

which I noted on the rim of the ancient basin of the Lac
des Palmistes, at the time of my first ascent to the summit
of Pelee (May 31), and when the compass-needle was de-
flected forty degrees or more to the eastward. The basin
was still extensively steaming, and was largely filled up
with ejected material from the volcano, recently cast out.*

Afterglows. — Among the interesting optical phenomena
associated with the Pelee eruptions were the remarkable
afterglows which for a fairly extended period were noted in
many and widely separated parts of the earth's surface, and
presented themselves with an intensity that almost rivalled
those which for a period of a year and more followed the
Krakatao eruption. I have elsewhere noted these glows as
having been observed by me on September 9, in approxi-
mate latitude 26° 30' N., longitude 68° 30' W. ; on Septem-
10 in latitude 30°, longitude 69° 30' ; and on September 11,
in latitude 33° 45', longitude 71° ; and again, at a much
later period, up to nearly the middle of November in New
York and Philadelphia. At the latter time, the glows were
also observed in Boston, Baltimore and other American

* An interesting possible relation existing between violent ex-
plosions and magnetic disturbance has recently been discussed by
Professor Nipher {Science, July 11, 1902, p. 64). The full record of
the magnetic disturbances connected with the West Indian eruptions
is being elaborated by the United States Coast and Geodetic Survey.
It is a noteworthy fact that nearly all of the earliest records of mag-
netic disturbance relate to the eruption of Mont Pelee on the 8th
May, and not to the earlier one of the Soufriere. This opens up an
interesting field of inquiry in connection with the physics of the two
volcanoes.

302 THE PHENOMENA OF THE EEUPTION

cities. Unfortunately, few observations were made in this
section of the United States, and in many parts the bright
loom, which appeared usually twenty to twenty-five min-
utes after the disappearance of the sun, was credited to the
smoke which permeated the atmosphere as the result of
undue burning of soft coal. In some places the glow was
also visible in early morning.

Brilliant afterglows or modified sunsets whose connec-
tion with the Martinique eruption can hardly be questioned
were noted, among other localities, at

A hundred miles westward of St. Lucia, on May 9
(green sunset, observed by the royal mail steamer La
Plata).

At Barbados, on May 11 and 14, with brilliant orange
skies, beginning at 5.30 p.m.

At Honolulu, twelve days after the eruption, with a
brilliancy of color about equal to that of the glows which
appeared in the first two weeks after the Krakatao eruption.
On July 31, as reported by Mr. S. E. Bishop {Nature, Sep-
tember 4, 1902, p. 442), the solar corona or " Bishop's
ring" was still conspicuous.

At Kingston, Jamaica, on May 25-31 and before ; with
colors reported to have been " extraordinarily rich and
beautiful."

At St. Kitts, in red color, on May 27 — being the earliest
distinctive glow noticed on the island.

Off the Venezuelan coast, between Cartipano and La
Guayra, noted by H. M. S. Gazelle, on May 10.

At Los Angeles, California, on June 22 and 23.

THE PHENOMENA OF THE ERUPTION 303

At Funchal, Madeira, on June 6, 10 and 11 — possibly-
even at an earlier period — described by F. W. T. Krohn to
have been similar to the Krakatao glows ; also on or about
July 6-7, 12-16, 26-27 and August 1-3.

At Slough, England (as observed by Professor A. S.
Herschel), on June 17, 21, 26 and later.

At Lewisham, South Kensington, and other localities in
England during late June and in July.

At Bombay, about June 25 (?).

At Morges, Switzerland, as observed by Professor F. A.
Forel (Journal Suisse, of July 10) on July 5, a brilliant
disk of a whitish-yellow light appearing thirty degrees
above the sunset point a quarter of an hour after the setting
of the sun.

In northern Italy, in early July, with streaked radia-
tions.

At Berlin, in late June or early July, with remarkable
coloring.

The characteristics of the Pelee (and Soufriere) after-
glows were similar to those of the glows of Krakatao,
although the intensity of the coloring and illumination was
probably at most points of observation less pronounced than
in the case of the glows of 1883 and 1884. As I observed
the coloring towards the middle of September, at localities
north-northwest of Martinique, a few days after the new
great eruptions of Pelee and the Soufriere, it was very bril-
liant, the orange and the red being particularly fine. The
upper border of the bright illumination faded off into a
superb and intense lilac, which, I believe, had not generally

304 THE PHENOMENA OF THE ERUPTION'

been observed as a feature of the Krakatao glows. Bishop
noticed this lilac color in 1884 in Honolulu, even in day-
time, and it is certainly due to the commingling of the
pink or roseate light with the normal blue of the sky.
Five great " shadow-beams," with broadening ends directed
to the zenith, and of almost exactly the color of the purple-
blue in the outlying field of the sky, were a distinctive
feature of the area of the glows on September 9 and 10
radiating fan-like from the position of the sun, and rising
to perhaps forty-five degrees.

The brilliancy of the glows as they were observed in
parts of western Switzerland was such as to suggest a con-
flagration, appearing "as if the whole of the west of Switz-
erland was on fire and the flames reflected in the sky."
It is singular that Professor Herschel makes the same obser-
vation for the appearance at Slough, England, on the night
of June 22, which was " an almost terrifying resemblance
to reflection in the sky of an immense distant conflagration',
{Nature, July 24, 1902, p. 294). I have been personally
informed that the same aspect of the glows was noted in
Honolulu, where many thought that the islands were
aflame. The height of the glow-producing matter has beei
estimated by Herschel to have been at different times froi
five or eight to thirteen or twenty miles, whereas the at
mosphere charged with the volcanic dust of Krakatao w*
thought to have floated twenty-five or thirty, and even forty

* Correspondence in London Daily Chronicle, dated Geneva, July
14.

Photo. Heilprin

THE ISSUING BLASTS FROM THE CRATER— AUGUST 24, 1902
Lower white clouds from base of crater

THE PHENOMENA OF THE ERUPTION 305

and seventy miles high, the uppermost particles of matter
being at that time much finer than those emitted by Pelee.

It is interesting to note in connection with the low
position of this glow-cloud that its velocity of passage, com-
pared with that of the Krakatao eruption, was also a low
one. Bishop tells us that it arrived in Honolulu ten days
after the Pelee outbreak, whereas the Krakatao glows, trav-
ersing twice the distance, arrived at the same spot in only
two days' longer time. This would give in the one instance
a velocity of about two and a half times that of the other, or
of sixty to seventy miles an hour in the case of the Kraka-
tao cloud, and of twenty-three to twenty-five miles for the
cloud from Pelee. There is seemingly no reason to doubt
that, the movement was in both cases from the east to the
west, conformably to the determinations that have been
made that the high cirrus atmospheric currents take this
course in the zone of (approximately) twenty degrees on
either side of the equator. Krohn has also assumed from
the records of Funchal, Madeira, that the rate of travel of
the Pelee cloud was on an average thirty miles an hour
{Nature, September 25, 1902, p. 540). The direction of
travel, measured by the time period, would here also appear
to have been from east to west.

The Shock and Noise of the Eruption. — Humboldt, in
dealing with the volcanic phenomena of the West Indies,
makes the interesting observation that the eruption of the
Soufriere, in 1812, was not as audible near to the mountain
as it was farther out to sea. It is certain that very few
of the inhabitants of Fort-de-France heard the explosion

306 THE PHENOMENA OF THE ERUPTION

of Pelee on May 8, or were made conscious of it through
an earth-shock or pulsation. Diligent inquiry among all
classes of people leaves me in doubt as to whether anybody
really heard it. Yet it is certain that this eruption was un-
mistakably heard at St. Kitts and St. Thomas, from two
hundred and seventy to three hundred miles distant, and in
all or nearly all of the islands of the Lesser Antilles. The
explosion of May 20 went similarly unnoticed in Fort-de-
France, whereas the detonations reported for that event in
St. Thomas, St. Kitts, Guadeloupe and Dominica were of
marked intensity. On the night of August 30 I was
located with my associate at the Habitation Leyritz, on the
northeastern foot of the volcano, not more than four miles
in a direct line from the crater, and with nothing interposed
between it and ourselves except the open, almost directly
descending slope of the mountain. When the death-dealing
explosion took place we were either seated in the open dining-
hall or were outside remarking upon the magnificence of the
electric display. Beyond hearing one or two " thuds," that
seemed to rise above the general voice of the volcano, I
doubt if any of our party of four could have localized the
explosion or series of explosions through any particular
sound or detonation. There was surely no detonation that
was particularly striking at this time. On the other hand,
the detonations heard at corresponding times at Port-of-
Spain, Trinidad, at Cartipano, Venezuela, and in the island
of St. Kitts — localities removed from two hundred and
seventy-five to three hundred and twenty-five miles away in
opposite directions — have been likened to the firing of

THE PHENOMENA OF THE EEUPTION 307

heavy siege-guns. The officers of the Fontabelle, among
others, assured me of this condition in Port-of-Spain. It
seems that the detonations were noted on the Venezuelan
coast far beyond Carupano, where rather severe earth-
quake shocks were also recorded. The report of United
States Consul Plumacher, of Maracaibo, published in the
Monthly Weather Review, gives the important record that
on the morning of the first great eruption of Pelee (May
8) terrific detonations were heard in the region of his post,
which was about eight hundred miles from Martinique.
These sounds were recognized to be not of " heavy artil-
lery," which they had been thought to be by a servant, for
"I knew that ... if all of the cannons of Venezuela
were fired together, they could not produce such sounds.
It was not like cannonading with heavy siege-guns ; it was
neither thunder, nor the strange, unpleasant subterranean
sounds of convulsions of the earth ; it was as if immense
explosions were fired high up in the clouds." This seeming
reversal of the detonations from the clouds was also re-
marked at Port-of-Spain as a feature of the detonations ac-
companying the August 30 eruption. With the intensity
of sound that appeared at Maracaibo, it is fair to presume
that the detonations were markedly audible two or three
hundred miles farther, or perhaps at a full distance from
the seat of disturbance of a thousand miles. Humboldt
states that " the frightful subterranean noise, like the thun-
dering of cannon, produced by the violent eruption of
the latter volcano [the Soufriere of St. Vincent] on the
30th of April, 1812, was heard on the distant grass plains

308 THE PHENOMENA OF THE ERUPTION

(llanos) of Calabozo, and on the shores of the Rio Apure,
one hundred and ninety-two geographical miles farther
to the west than its junction with the Orinoco" {Cosmos,
Bonn's Edition, V, p. 422) — a point fully eight hundred
miles in a direct line from the island of St. Vincent.

The peculiarity of the explosions being heard with
terrific intensity at points of distance and hardly, if at all,
near by, was also exhibited to an extent by the Krakatao
eruption, the report from which was carried to the island
of Rodriguez, three thousand miles away — the farthest dis-
tance from a point of origin at which sound has ever been
heard, or at least recorded (Royal Society Report, p. 79).
General Strachey, commenting upon this peculiarity, be-
lieves that " probably this peculiar phenomenon was caused
by the large amount of solid matter" which at the time of
the eruptions " was ejected into the atmosphere by the
volcano, and which formed in the lower strata of the air a
screen of sufficient density to prevent the sound-waves from
penetrating to those places over which it was more im-
mediately suspended" (p. 79). This explanation, so con-
trary to the results that have been obtained by Tyndall
and others in their experiments upon the transparency and
opacity of the atmosphere in relation to the passage of
sound-waves — the unexpected determination that the dis-
semination of solid particles in the air, the presence of fog,
rain or snow, etc., have little or no effect upon the trans-
mission of sound — it seems to me can hardly be the correct
one ; nor, indeed, can it find application to the conditions
which existed at the time of the eruption of August 30,

THE PHENOMENA OF THE EEUPTION 309

when I was located at the Habitation Leyritz. We were
then practically under, and not behind, the volcanic cloud,
through which came quite distinctly the muffled, but con-
tinuous, roar of the volcano. If the obscuration of sound
by solid particles was really produced, the phenomenon
must have taken place within the body or vent of the vol-
cano itself. I should rather believe that the acoustic inter-
ruption was in some way associated with an atmospheric
disintegration — the presence within it of layers of different
thermal power and differing vaporous constitution, pro-
ducing, to use Tyndall's words, acoustic clouds that are
"flocculent to sound" ("Lectures on Sound," 1875, p. 321).*
This would, however, still leave unexplained the transmis-
sion of the sound to great distances, unless, indeed, we may
be permitted to assume that the propagation of the sound-
waves has been carried to distant points through the ma-
terials of the solid crust. Can it be thought that the
sounds coming as if thrown down by the clouds, noted by
Mr. Plumacher at Maracaibo and by others in the island
of Trinidad, were reflections from lofty "acoustic clouds,"
to which the sound-waves were transmitted through the
central orifice of the volcano ? This suggestion is thrown
out with much diffidence, and only because no ordinarily

* The remarkable experiments made by the distinguished British
physicist in connection with the Trinity House have established the
existence of conditions of absolute opacity to sound in an atmosphere
that is optically transparent, and shown the fallacy of the still com-
monly accepted notion that a direct relation exists between a clear
atmosphere and the transmission of sound.

310 THE PHENOMENA OF THE EKUPTIOK

recognized theory seems to satisfactorily account for the
facts as they are present.

The remarkable atmospheric and seismic waves which
followed the Krakatao eruption seem also to have been a
part of the Pelee or Soufriere phenomena as well, but the
data that appertain to them are only meagrely in hand, and
leave little to be said regarding the full intensity of the
phenomena. Professor Henry Kelm Clayton, of the Blue
Hill Observatory, Hyde Park, Massachusetts, has noted
" some marked barographic undulations at Blue Hill on the
morning of May 7, which," it was thought, were " perhaps
connected with this [Martinique] volcanic eruption" {Nature,
May 22, 1902, p. 102). Of more startling significance is
the record of the observatory of Zi-ka-Wei, China, situated
almost exactly half around the world from Martinique,
which notes between 12.25 and 12.35, Martinique time, two
marked tremors or shocks, registered by the mercurial ther-
mometer acting as an accidental seismograph. These, as
well as the magnetic perturbations observed earlier in the
day, and which so closely correspond in time with the Pelee
eruption, are referred by M. de Moidrey to the Martinique
disturbance, and it is assumed from the hour at which the
phenomena were observed that the time of propagation of
the earth-wave was four hours and twenty-seven minutes,
giving a velocity of approximately twenty-five miles per
minute (Comptes Rendus, August 11, 1902, p. 322). This,
barring the Krakatao occurrence, is the only instance that
is known to me of an earth-tremor or pulsation having been
propagated clean through the centre of the earth to the

THE PHENOMENA OF THE EEUPTION 311

antipodal surface. The seismographs of Great Britain give
no registry for the Pelee eruption, having remained at rest,
according to Professor Milne, from the 8th until the 11th
of May.

The Nature of the Destroying Blast. — In an article
published in the August (1902) number of McClure's Mag-
azine I expressed the opinion that the destroying element
of the blast was seemingly one of the heavier (carbon ?)
gases, and that with it the force of the superheated steam
was acting only in a minor degree. At that time there
appeared to me much to support this view, although I did
not hesitate to say that the evidence upon which it was
based was far from conclusive. Particularly needful for
this demonstration was the proof of the actual existence of
such gas acting with the shattering blast ; but up to this
time none has been found. The only gases of consequence
whose presence has so far been detected among the products,
whether gaseous or mineral, of the Pelee eruptions are the
sulphurous and sulphuretted-hydrogen, the former alone
being present in any quantity.* Sulphur vapors or fumes
were oppressively diffused through the atmosphere of Saint

* Dr. Hovey has called attention to the significant absence of
chlorine gas in the analyses that have been made by Hillebrand of
the St. Yincent ash, and assumes that this may be an indication that
fresh water, and not the water from the ocean, was the prime insti-
gator of the volcanic movements. The same absence of chlorine or
of chlorine salts distinguishes the Martinique ash. I, however, found
some of the ejected boulders or bombs, both in the valley of the Rivi.
eres Blanche-Seche and the basin of the Lac des Palmistes, carrying
crusts or patches of greenish-yellow iron-chlorid.

312 THE PHENOMENA OF THE EEUPTION

Pierre for the better part of two weeks before the main
catastrophe (see Chapter III) — horses and other animals
dying from it, and respiration being made difficult for man
— and at the time of his latest visit to the crater-border, in
the month of October, Lacroix found them issuing in such
quantities, in fumarolic blasts from the crevices of the cen-
tral cone, as to make a close approach dangerous. I, my-
self, several times detected the sulphur vapors five or six
miles out at sea, but, singularly enough, failed to note their
presence, except to a very minute degree, when standing at
the rim of the crater. On the other hand, sulphur in the
falling ash of the eruption of August 30 was clearly in
evidence at the Habitation Leyritz, and we are informed by
Dr. Berte {La Geographie, September 15), of the Pouyer-
Quertier, that the air was densely charged with it when the
destroying cloud swept out from Pelee on the fatal May 8.
From these conditions one has a right to conclude that this
gas may have played an important part as an assistant in
the destruction of life at Saint Pierre, even though an
equally complete annihilation might have been brought
about without it.

The opportunity that was presented to me at the time
of the second death-dealing eruption of Mont Pelee of
almost immediately visiting the field of destruction, of in-
terrogating a number of the severely wounded, and of ex-
amining the bodies and clothing of some of the unfortunate
dead, has forced upon me a somewhat different conclusion
as to the nature or composition of the tornadic blast from
that which I formerly held, for it is now made clear that

THE PHENOMENA OF THE ERUPTION 313

the acting force — to whatever extent it may have been aided
by other forces or agents whose testimony does not appear
was superheated steam, or superheated steam charged with
ot ashes and lapilli. The evidence proving this, both at
Morne Balai and Ajoupa-Bouillon, and, as Lacroix found it
later at Morne Rouge, is, it seems to me, conclusive. This
steam was shot out as a violent blast, and its mechanical force,
withering heat and possible mixture with other gas, shattered,
asphyxiated and scorched ; and where charged with incan-
descent particles of solid matter, as in that part of its course
which overwhelmed Morne Rouge, also burned. Neither at
Morne Balai nor at Ajoupa-Bouillon did I find the faintest
indication of anything having burned with a flame, or
having been carbonized, not even the dry palm-thatching of
the cases. The trees that were left standing were dry and
largely stripped, and in the less destroyed zone the leaves
hung to the branches, shrivelled up as though having been
rapidly passed through a dry-heat furnace or a scorch-blast.
The sap from the twigs was completely gone, and the
branches and branchlets broke square across. There was
nothing to indicate the passage of combustible gases, and I
failed to find — although my examination was not made with
true minuteness — evidence of the presence or action of any
of the terrestrial gases. A number of inquiries elicited the
impression that sulphur was the only gas whose presence
was detected in the passing storm, but even its action does
not seem to have been badly felt. The scorching, reddening
or boiling, and tumefaction of the bodies plainly showed the
terribly swift and sure work of the passing steam and hot-

^TlM y

• 314 THE PHENOMENA OF THE EEUPTION

air ferment. The opening of a door or window only for an
inch and for an instant was sufficient to invite the work of
death. In this second great eruption of Mont Pelee the de-
stroying force, as is shown by the number of frail houses
that were left standing in or near the path of the storm, was
less powerful than on May 8, but its zone of destruction was
far greater, beginning almost immediately in a broad sweep
over the crest of the volcano. It may be positively assumed
that adjacent to the steam zone on either side was a zone of
simple hot air or dry destruction, in which, doubtless, many
also perished, for even here the temperature must have
ranged well into the hundreds of degrees. Professor La-
croix has, from an inspection of metallic objects that have
not been fused or undergone any material alteration, at-
tempted to ascertain the degree of heat of the Saint Pierre
blast. This, as determined by the non-fusion of the copper
telephone wires and plates, iron railings, etc., would seem
to have been not over 1900°. This method of determining
the temperature is not necessarily a conclusive one, as the
very rapid passage of a heated current over even readily
combustible objects might not inflame or fuse, whereas a
slower movement would.* But even with an air-temperature
of no more than seven hundred to eight hundred degrees
one need invoke the aid of no special agent to explain the
condition of difficult or impossible respiration which has so
frequently been testified to by those who escaped or survived

* Lacroix fully recognizes this condition, and he states that car-
tridges, rubber tubes, etc., were passed or jumped over by the hot
current.

TIIE PHENOMENA OF THE ERUPTION ^ B1& .

their wounds for a while. Assuredly the sensation must
have been one of no air to breathe, and one of the results
the burning or even excoriating of the lining of the throat
and bronchi. It was like breathing a furnace-fire, especially
where the blast was charged with burning matter.

Professor Lacroix accepts the same interpretation of the
destroying force that wrecked Morne Rouge as I have for
that of Ajoupa-Bouillon, Morne Balai and Morne Capot:
"It is not doubtful that the destruction was due to the
action of a cloud of aqueous vapor highly charged with hot
ashes. There is no reason to seek for a combustible gas ;
the trees are not burned and the palms from which the
leaves have not been forcibly torn show these to be simply
dried out" (Comptes Rendus, October 27, 1902, p. 672). It
is not difficult to apply this lesson of the later eruption of
Pelee to the special conditions of the Saint Pierre catas-
trophe. With a tornadic blast of the character of, but
more powerful than that which destroyed the five or more
towns and villages on August 30, it is easy to assume the
destruction of the city, although the swiftness and com-
pleteness of this destruction will always appear surprising.
We may, perhaps, assume as a factor in this complete de-
struction the propagation of a number of serially and
rapidly following explosions — such as Bunsen, Dixon and
others have shown to exist in an ordinarily exploding gas-
cloud.* These would surely greatly multiply the force of

* See the paper by Harold B. Dixon : " On the Movements of the
Flame in the Explosion of Gases." Proceedings Boyal Society of
London, LXX, September 20, 1902, pp. 471 et seq.

316 THE PHENOMENA OF THE ERUPTION

the exploding or initial cloud. Professors Sekiya and Ki-
kuchi, discussing the Bandai-San eruption (Journal of the
College of Science, III, 1890) properly remark that "the
tremendous explosions of steam at quick intervals lasting for
about a minute produced violent disturbances of the air, con-
sequent upon the sudden radial expansion of the liberated
volumes of steam . . . The eruption of Bandai-San may
be aptly compared to the firing of a tremendous gun — such
a one, however, as can only be forged by nature." These
authors also refer the immediate cause of the eruptions to
" the sudden expansion of steam pent up within the moun-
tain." There were no discharges following the first ex-
plosion.

The recognition of the nature of the destroying tornadic
blasts which in such swift measure swept off upwards of
thirty thousand inhabitants from the surface of the earth
still leaves untouched some important considerations bearing
upon the explosions themselves. What was the exact seat
of the explosion or explosions ? Was the main explosion
in the conduit of the volcano, whence the great internal
detonation might have been conceived to pass ; or was its
locus immediately above the crown of the volcano, with its
ascensive energy blanketed by the opaque cloud of steam
and ashes overhanging? It is not easy to explain the
downwardly-directed or oblique shots, unless, indeed, we
assume some such sort of down-throwing as the result of
pressure from above or behind. Mr. George Kennan, in
his work, " The Tragedy of Pelee" (1902), has ably discussed
this aspect of the problem, and he compares the explosion

THE PHENOMENA OF THE ERUPTION 317

to other explosions which have had their directions or in-
tensities determined by the presence of an unyielding wall
or barrier on one side of the blast. The "extraordinary
violence of the lateral blast caused by the explosion of the
Toulon powder-magazine, in March, 1899," — which, as
stated by Colonel J. T. Bucknill (Engineering, London,
May 26, 1899, pp. 665-666), appears to have exerted its
main force in one direction (" something like an accidentally
formed fougasse"), and covered the ground for a full kilo-
metre in the path of its course with rock-debris and masonry,
while hurling blocks of stone weighing four hundred-weight
to a distance of two kilometres, — is justly brought in for
comparison, and there is hardly a question that it approxi-
mately supplies the explanation to the Pelee blast. Colonel
Bucknill describes the Toulon magazine as being " so solidly
built that it practically formed a sort of a cannon or mortar ;"
and this is virtually what we find or can assume to have
been the case with Pelee. The condition of explosion may
then be stated as follows : A volume of steam with intense
explosive energy rising to the crater-mouth, blowing out in
its first paroxysm a part of the crater-floor, and then ex-
ploding in free air under a heavily depressing cushion of
ascending steam and ash, and with surrounding walls of
rock on three sides and more to form an inner casing to
nature's giant mortar. The blast was forced through the
open cut, or lower lip of the crater, that was directed to
Saint Pierre. It is interesting to note that the " overflow"
eruption of August 30 only took place after the crater-floor
had been elevated, as we are informed by Lacroix, by per-

318 THE PHENOMENA OF THE EEUPTION

haps seven hundred to nine hundred feet as the result
the accumulation of volcanic ejecta.

The black appearance and " rolling" of the destroy-
ing clouds are, as most investigators have already indicated,
due to charging with large quantities of ashes and other
solid particles. It is generally conceded that the brilliant
red glow which was noted in it by some observers, especially
in its advanced position, was merely the loom of the numer-
ous incandescent particles that were contained within the
cloud — the cloud appearing as if burning with flame. But
may it not be assumed that a part of this red coloring was
due to that property in steam under pressure which at times
permits it to acquire a red color to transmitted light ? This
condition was pointed out by Principal J. D. Forbes in the
case of escaping steam from a locomotive as far back as
1839 (Transactions Eoyal Society of Edinburgh, 14, 1839,
p. 371). Such a luminous red mass might readily have been
taken for a " descending wall of fire," such as some seamen,
like Captain Freeman of the Roddam, claim to have seen.
It is interesting to note that a hot, suffocating blast, evi-
dently of the type of that issuing from Mont Pelee, was
noted by Pond and Percy Smith in their investigations of
the Tarawera eruption (Transactions New Zealand Insti-
tute, 1886 (1887), p. 351).

APPENDIX

LETTER FROM A PROFESSOR IN THE LYCEE OF SAINT PIERRE

{Translated from the Bulletin of the Societe Astronomique de France)

Saint Pierre, Saturday, May 3, 1902, 5.45 a.m.

We are in the midst of a full eruption of Mont Pelee ; all night
the volcano has been sputtering ashes over the city, which appeared
this morning covered with a grayish shroud ; from time to time
muffled detonations are heard. It is several days since the old
volcano has been manifesting its desire to return to life, or, according
to other theories, the beginning of its deathrstruggles, by the outbursts
of vapor. The ancient lake has dried up j another has formed which
boils like molten metal. But nothing has been as impressive as the
spectacle presented yesterday.

When I went to the Lycee at eight o'clock the mountain was
clear of all clouds and vapor. At the end of my lecture I noticed
the tutors and some of their pupils pointing out the mountain to each
other. I joined them. Three enormous balls of very compact,
grayish smoke have just poured out of the crater ; the eruption con-
tinues. The wind has carried these vapors towards the Dominica Chan-
nel. At noon the eruption was renewed with greater severity. At
two o'clock stones, plainly visible, were hurled out ; at the same hour
in the morning your mother and sister were awakened by a detona-
tion ; our dogs barked. I had been sleeping soundly, but my slum-
bers were interrupted by a strange dream and a smell of sulphur.
Ashes were raining down on the house itself and the furniture is full
of them. We went out on the boulevard to see what was happening.

Nine-forty-five. The ashes blind us ; Mont Pelee is completely
invisible, being hidden by an impenetrable layer of vapors. All the

319

320 APPENDIX

surroundings are full of smoke, which sticks to the trees and falls
in an impalpable powder. We continue on our course in order to see
what has happened to the Saussines.

Near the Jardin des Plantes it becomes imprudent for your
mother and sister to proceed further. I continue. They cross th(
Savane to see the Armamets ; a bull which has escaped rushes
through the streets as though mad ; the little birds hardly know 01
which branch to perch; the pigeons are cowering in their houses; ii
the yards hens and ducks remain in their coops ; the appearance oi
the country is dismal ; it is grayer than when it rains ; in fact, it is
raining, only it rains ashes.

I arrived at the house of the Saussines; all was closed,
knocked ; the door opened but was quickly closed again ; cinders
cover the floor, the furniture, and penetrate even into the drawers.
We relate our experiences ; the night has been a very anxious one,
for it seemed as though we were awaiting suffocation. I took a cup
of coffee and we then went off with Saussine. My hat was covered
with ashes to a thickness of several millimetres ; my alpaca vest was
gray, my trousers and my shoes the same color.

From the Lycee it was almost impossible to distinguish the sea, so
dense were the emanations. People asked, "Where would it end?"
for their eyes, ears and noses were filled, and this state of affairs was
becoming intolerable. The water of the Goyave had in part given
out. My opposite neighbors left this morning. They went, as they
said, up to Morne d' Orange in order to breathe more freely. It Was
a mistake, for the conditions there were the same.

How did the next night pass ? The inhabitants of Sainte-Philo-
mene and of Precheur were terrified, and women holding their little
children in their arms were to be seen passing by.

Ten o'clock. The drum calls. We were the only ones to hold
classes. But the Governor has just arrived ; he has selected a com-
mission. (The ashes had reached Lamentin in the south.) All the
stores are closed. A dispatch reports that Fort-de-France, too, is
receiving a fall of ashes. It would aj>pear from the usual phenomena

APPENDIX 321

that we can expect a flow of lava ; but the air becomes more difficult
to breathe.

Noon. At two o'clock there will perhaps be something new
igain ; but I hasten to mail this letter.

THE CATACLYSM OP MAY 8 AS WITNESSED BY M. ROGER ARNOUX,
MEMBER OF THE ASTRONOMICAL SOCIETY OP FRANCE

(Addressed to Camille Flammarion and published in the " Bulletin de la
Societe Astronomique de France" August, 1902)

"Fort-de-France, Martinique, le 3 juillet 1902.
"Cher Maitre,

"Seul survivant* de tous mes braves collegues de la Societe
Astronomique de France habitant Saint-Pierre, le devoir me present
de vous faire part de la disparition de tous ces Societaires, y compris
mon malheureux frere Charles, employe de la Compagnie Generale
Transatlantique. Pour moi, je ne dois mon salut qu'au pur hasard qui
me conduisit sur ma propriete du Parnasse, la veille au soir, 7 mai,
personne n'ayant pu s'echapper de la ville pendant l'eifroyable sinistre
du 8 mai dernier.

" Toute ma famille a ete aneantie par le coup fatal, mon pere, ma
mere, mon frere, ma soeur ; ils etaient restes a Saint-Pierre.

"J'ai l'honneur de vous transmettre le rapport ci-dessous sur ce
que fai vu.

"Veuillez agreer, cher Maitre, l'assurance de ma tres grande

veneration.

" Signe : Eoger Arnoux."

Je remonterai dans ce recit de deux ans en arriere.
Le lundi de la Pentecote 1900, etant alles en partie de plaisir au
sommet de la Montagne, nous pumes decouvrir, mon frere et moi, ainsi

* Nous avons vu que fort heureusement deux autres de nos Societaires
actuellement a Paris : M. le docteur Jtemy N^ris et M. Th. C&estin, avaient
pris la decision de s' Eloigner du volcan. — C. F.

322 APPENDIX

que les guides qui nous accompagnaient, l'emplacement de deux petite
solfatares qui s'etaient ouvertes dans le cratere actuel dit l'Etang-Se<
Nous vimes nettement deux espaces de 30 ou 40 metres de rayoi
completement denudes, les arbres couches et brules et le sol parseni(
d'une matiere jaune que nous pensions etre du soufre ; tandis que loi
de notre premiere ascension l'annee precedente, le meme site offrait 1(
spectacle de la plus riche vegetation. Toutefois nous n'avons vu
moindre petite vapeur indiquant que ces matieres pussent etre ei
combustion.

L'annee d'apres, quelques amis ayant fait une nouvelle ascensioi
m'assurerent avoir vu au meme endroit cinq ou six petites fumeroll(
d'ou s'echappait une fumee verdatre empestant le soufre. Mais
n'est qu'au mois de mars de l'annee presente que les phenomenes
manifesterent d'une fagon appreciable et qu'on commenca a en parle
a Saint-Pierre.

Des habitants des hauteurs du Precheur racontaient sentir presque
continuellement une forte odeur de soufre, et un de mes amis habitai
le quartier du Morne-d' Orange, me certifia avoir vu de nuit vers la
du mois de mars une assez vive lueur sortant de l'entonnoir du cratere

Le temps etant demeure tres nuageux pendant tout le moi
d'avril, personne ne put se rendre compte du travail qui se faisait si
la Montagne. Certains habitants du Precheur disaient avoir entendi
des detonations, d'autres avoir vu du feu, etc. . . . et ce n'est que
dans la nuit du 25 avril qu'on fut convaincu que la Montagne s'etait
rallumee.

Etant couche vers les onze heures et demie de la nuit du 25 avril,
je fas reveille par une formidable detonation que je pris tout d'abord
pour un coup de foudre, le meme fait s'etant reproduit un instant
apres, je me levai pour examiner le ciel, trouvant singulier un orage
au mois d'avril. Sitot que j'eus regarde la Montagne, je compris qu'il
s'agissait d'une eruption. De l'endroit ou je savais etre le cratere, je
vis s'echappant une immense colonne de fumee dont le sommet
s'inflechissait dans la direction Nord-Est. Bientot apres, ce furent
des detonations et des grondements continuels, tandis que de la

APPENDIX 323

colonne de vapeur partaient des etincelles electriques. Nous resumes
alors une pluie d'environ un demi-centimetre d'un sable gris a grains
presque aussi forts que le plomb de chasse appele cendrille. L'erup-
tion dura jusque vers une heure et demie du matin et parut se ralentir
pour de nouveau recommencer vers les cinq heures, nous lan§ant cette
fois un sable plus gris que la nuit et dont les grains etaient presque
impalpables.

Les jours suivants, on voyait, surmontant la Montagne, un gros
nuage d'un gris bleuatre ayant absolument l'aspect d'un gros nuage
orageux, mais ni grondements, ni orages, ce qui faisait penser que sans
doute le cratere etant largement ouvert, les pbenomenes ne pouvaient
qu'aller en diminuant.

Le matin du 2 mai, vers les neuf heures, les memes faits signales
pour la premiere eruption, se reproduisirent (detonations, grondements,
cendres, etc. . . .) et je pus m'apercevoir que le cratere s'etait elargi
considerablement ou, pour mieux dire, que d'autres bouches s'etaient
ouvertes, mais a peu de distance de la premiere et toujours dans le
raeme cirque de l'Etang-Sec, large d'environ 300 metres et situe a peu
pres a 800 metres d'altitude.

Ce n'est que le 5 mai que la decharge du cratere commenca a se
faire par la coulee de la Biviere-Blanche. De fortes vagues d'une
sorte de boue noiratre surmontee d'une epaisse vapeur descendaient
de la Montagne, et l'apres-midi de ce jour, l'Usine Guerin etait
ensevelie sous l'une d'elles.

Le lendemain 6 mai, l'eruption semblait entrer dans une periode
d'accalmie, les vapeurs degagees du cratere ayant une moindre force
ascensionnelle, de sorte que tous pensaient que l'eruption irait
declinant, vu que la decharge se faisait normal ement.

Le 7 au matin, me trouvant a la rhumerie Berte, je causai avec
le directeur du cable anglais (M. Miller) que m'apprit que toutes les
communications telegraphiques entre la Martinique et les ilea voisines
etaient coupees. L'idee d'un cataclysme me traversa l'esprit, ear le
directeur du cable lui-meme attribuait ces ruptures de cables a des
depressions sous-marines.

324 APPENDIX

Dans l'apres-midi, on entendit a Saint-Pierre, venant de la direc-
tion sud, des detonations se succedant a de courts intervalles et pro-
voquant des vibrations aeriennes faisant trembloter les bibelots situes
aux etages. Le bruit courut alors que c'etait un navire qui s'exergait
dans les eaux de Fort-de-France, chose d'autant plus croyable que le
semaphore avait effectivement signale un navire de guerre dans le Sud.

Pour moi, je trouvai etrange la violence des commotions aeriennes.

Ayant quitte Saint-Pierre le soir vers les cinq heures, j'assistai au
spectacle suivant. D'enormes roches nettement visibles etaient pro-
jetees en Fair par le cratere, a une hauteur considerable, si bien
qu'elles mettaient environ un quart de minute a retomber, decrivant
un arc les lan^ant bien au dela du morne Lacroix, point culminant du
massif.

Vers les huit heures du soir, nous vimes pour la premiere fois au
sommet du cratere des feux fixes d'une flamme tres blanche. Peu
apres, quelques detonations semblables a celles entendues a Saint-
Pierre, se produisirent, venant toujours du Sud, ce qui me confirma
dans l'idee que j'avais deja de crateres sous-marins langant des gaz
detonant au contact de l'air.

Dans la nuit du 7 au 8, m'etant couche vers les neuf heures, je
me reveillai peu apres au milieu d'une chaleur suffocante et tout
couvert de transpiration ; sachant mes nerfs agaces, je pensai a un
malaise et me recouchai.

Vers les onze heures trente-cinq je me reveillai a nouveau ayant
senti une secousse de tremblement de terre, mais comme personne
n'avait ete reveille chez moi, je crus encore avoir ete trompe par mes
nerfs et me recouchai pour ne me relever que le matin a sept heures
et demie.

Mon premier regard a l'exterieur fut pour le cratere que je
trouvai assez calme, les vapeurs se repliant tres vite sous la pressic
d'un vent d'Est. Vers les huit heures, etant encore a regarder 1<
cratere, j'en vis sortir une petite vague, suivie deux secondes apres
d'une nappe considerable qui mit moins de trois secondes a couvrir jusqu'd
la Pointe du Carbet, en meme temps qu'elle se trouvait deja a noti

APPENDIX 325

zenith, se developpant par consequent presque aussi vite en hauteur
qu'en longueur. C'etaient des vapeurs en tout point semblables a
celles lancees presque tout le temps par le cratere. D'un gris violet,
elles paraissaient tres denses, car bien que douees d'une force ascen-
sionnelle inimaginable, elles conservaient jusqu'au zenith leurs som-
mets arrondis. Au milieu de ce chaos de vapeurs petillaient
d'innombrables etincelles electriques, en meme temps que les oreilles
etaient assourdies par un fracas epouvantable.

J'eus alors l'impression bien nette que Saint-Pierre avait ete
pulverise, et je pleurai sur-le-champ tous les miens que j'y avais laisses
la veille au soir. Comme le monstre semblait se rapprocher de nous,
mes gens, pris de panique, se mirent a courir sur un petit morne
dominant ma maison, me priant d'en faire autant. A ce moment un
vent terrible d'aspiration se leva, arrachant les feuilles des arbres et
cassant les petites branches, nous opposant meme une forte resistance
a la course. A peine etions-nous arrives au sommet du mamelon que
le soleil s'obscurcit tout d'un coup, faisant place a une noirceur presque
complete. Alors seulement, nous recumes des cailloux dont le plus
gros mesurait environ 2 centimetres de diametre moyen, en meme
temps que sur la ville de Saint-Pierre et dans la direction a peu pres
ou je savais trouver le quartier du Mouillage, nous vimes une colonne
de feu semblant animee d'un mouvement de translation et d'un autre
mouvement de rotation, laquelle trombe de feu j'estime au moins a
400 metres de hauteur. Ce phenomene dura de 2 a 3 minutes. Peu
apres les pierres, une pluie de boue s'abattit sur nous, couchant au ras
du sol toutes les herbes et meme les petits arbustes, puis ce fut une
pluie torrentielle durant environ un demi-heure.

En tout le phenomene avait dure a peu pres une heure, apres quoi
le soleil perga.

La vague que je vis s'abattre sur Saint-Pierre devait etre com-
posee d'une matiere liquide a une temperature considerable, lequel
liquide a du se vaporiser au contact de l'air, non cependant d'une facon
absolument instantanee, car je remarquai durant les deux secondes
que mit la vague a couvrir la ville, comme une petite pointe a l'avant

326 APPENDIX

de ladite vague : c'est du reste la seule facon de e'expliquer le fait, car
physiquement parlant, on ne peut guere concevoir un gaz doue de
deux forces contraires, force de chute et force d'ascension.

La foudre aussi a du contribuer a l'incendie, puisque, comme je
l'ai dit plus haut, ces vapeurs etaient sillonnees d'etincelles electriques.
De plus, par suite du degagement des gaz, il a du se produire un vide
considerable sur la ville ; lequel vide aura asphyxie les individus qui
s'etaient trouves dans des conditions particulieres pour ne pas etre
atteints par la vague. Le vent d'aspiration que je ressentis au
Parnasse, situe a 3 kilometres de Saint-Pierre a vol d'oiseau, a du
donner le coup de grace en broyant absolument Saint-Pierre.

Eelativement a une pluie de feu dont on a beaucoup parle, je n'ai
rien apercu de semblable, ayant cependant observe le phenomene dans
son entier. Quant aux matieres volcaniques (cendres, boue et pierres)
tombees a Fort-de-France et dans presque toute l'ile, elles ont du
provenir d'une sorte de fusee lancee par le volcan quelques secondes
apres la destruction de Saint-Pierre, car a aucun moment je n'ai vu
l'eruption verticale ; les vapeurs qui s'etaient precipitees sur Saint-
Pierre, ayant, dans l'espace de quelques secondes, couvert entierement
la Martinique, en meme temps qu' elles se trouvaient deja au zenith.

II s'agirait de savoir quelle a pu etre la nature de ces gaz. Poi
moi, je crois simplement a de l'eau chaude a une temperature excessive,
car peu apres l'eruption j'ai pu sentir pendant un temps assez long
une forte odeur de terre bouillie qui me conduisit immediatement
l'hypothese ci-dessus.

Eoger Arnoux.

APPENDIX 321

METEOROLOGICAL NOTES FROM THE OBSERVATORY OP FORT-DE-FRANCE AT
ABOUT THE TIME OF THE AUGUST 30 ERUPTION

August 24. — Earthquake nine- twenty -five a.m. Oscillation north-
west to southeast. Duration twenty seconds.

August 25. — Eruption visible at ten-twenty a.m. Numerous illu-
minations resembling lightning.

August 26. — Four distinct eruptions between five-thirty a.m. and
seven-ten a.m. Clouds rise to four thousand metres. Glow on the
first ascending clouds.

August 28. — Bright illumination (lueurs) after ten-thirty p.m.

August 29. — Distinct rumblings (grondements) between nine and
ten p.m.

August 30. — At one p.m. great flocculent volcanic cloud, of large
dimensions, flowing from crater northwest to southeast, halted about
half distance from crater to Fort-de-France, becoming dark gray in
centre and turning to white on border. Between twelve-fifty and
one-ten p.m. the barometer descends slightly with a V-nick °f f°ur
millimetres and an opening of 20'. No electric manifestations. Sea
calm. Violent eruption in evening comparable to May eruption.
The projected cloud appears about nine p.m., and advances rapidly,
almost to Fort-de-France. Electric manifestations remarkable, and
much more intense than on July 9, although identical in character.
Lightning more intense. Simultaneous scintillant and zigzag dis-
charges, producing an uninterrupted crepitation, from time to time
effaced by a great glow, which illumined all the clouds. Strong odor
of ozone perceptible. There was a retreat of the sea at nine-twenty
five p.m., followed by a more rapid rise (about one metre), which
covered the quais and came to the border of the savane. Barometer,
which had been depressed three millimetres, rose four millimetres in a
period of ten minutes. Light fall of ashes and small stones.

328

APPENDIX

BAROMETER I

Ten a.m. Four p.m.

Millimetres. Millimetres.

August 21, 762.6 760.8

" 22, 761.4 760.1

23, 762.0 < 760.3

" 24, 762.8 761.8

" 25, 762.6 760.4

26, 762.8 761.0

27, 762.9 761.3

28, 763.0 761.0

" 29, 763.1 .764.0

30, 763.1 761.2

31, 763.0 761.0

Total rainfall for August, 215.8 millimetres.

JJfo^rcs. g%

THE ISLAND OF MARTINIQUE

NOTES

1 (p. 17). The population of the commune of Saint Pierre was
according to the same census (1894) 25,382 (Annuaire de Martinique).
A less official census placed the population in 1901 at 26,500.

2 (p. 24). The more distinct photographs show the time on the
clock of the Hopital Militaire — the hour of the destruction of Saint
Pierre — to be 7 h. 52 m., instead of 7 h. 50 m., as generally stated (8 h.
2 m. time of Fort-de-France).

3 (p. 35). The eruption of Asamayama is stated by Milne to have
been "the most frightful eruption on record," the projectile force
having been sufficient to throw out rocks from forty to eighty feet in
some of their dimensions, and even to have cast out one rock meas-
uring two hundred and sixty-four by one hundred and twenty feet
(British Association Report, 1887). It is more than likely, judged by
the character of the attending phenomena, that the accounts of some
of the earlier eruptions, such as Papandayang, Asamayama, and
especially Tomboro, are greatly exaggerated, and to a degree even
fanciful.

4 (p. 36). Flammarion, Bulletin de la Societe Astronomique de
France, July, 1902, p. 300 ; Royal Society Report.

5 (p. 39). Bulletin de la Societe Astronomique de France, July and
August, 1902.

6 (p. 42). See the interesting paper by Harold B. Dixon " On t ho
Movements of the Flame in the Explosion of Gases" (Proceed ingi
Royal Society, September 20, 1902).

7 (p. 46). A few of the inhabitants were taken from the city in
a badly scorched condition, but with the exception of the prisoner
Ciparis (or Cilbarice) and Compere-Leandre, concerning whom further
details appear in Chapter VII, none of these seem to have survived
their wounds.

329

330 NOTES

8 (p. 51). A further discussion of the Pelee afterglows appears in
Chapter XIX, « The Phenomena of the Eruption."

9 (p. 53). The National Geographic Magazine, July, 1902 ; Cen-
tury Magazine, September, 1902.

10 (p. 67). Several of the older active points on the southwestern
slope of Pelee passed under the name of " Soufriere" or " the
Soufriere." That of the Etang Sec is not to be confounded with the
one referred to by Leprieur, Peyraud and Eufz as one of the seats
of the eruption of August, 1851. See Chapter XI, "The Geography
of Mont Pelee."

11 (p. 122). It seems to be a common belief that Pliny's letters
to Tacitus narrating the Vesuvian eruption of 79 were written when
the famous epistolarian was hardly more than a lad, less than
eighteen years of age. Pliny was witness of the eruption when he
was of this age, but the narration belongs to a much later period.
Pliny's opening reference to Tacitus as one whose writings would
confer immortality upon his own uncle, the elder Pliny, clearly estab-
lishes this point, as Tacitus, at the time of the eruption, was him-
self only about twenty-seven years of age, and had produced no work
that would entitle him to special consideration as a fame bestower.
It is more than likely that these letters were written nearly or quite
twenty years after the event; and this circumstance probably ex-
plains why no direct reference is made in them to the destruction of
Pompeii and Herculaneum.

12 (p. 126). In Orrery's translation this passage appears : " em-
barked with a design not only to relieve the people of Retince" etc.,
the personal name Rectina being evidently confounded with Retina
(also Rectina), the location on the Bay of Naples which closely
corresponds in position with the modern Resina (near Hercula-
neum).

13 (p. 129). See Johnston-Lavis, " The Geology of Monte Somma
and Vesuvius," in Quarterly Journal Geological Society of London,
XL, pp. 35-112, 1884. The author says : " If the Plinian eruption
had formed the greater part of the present Vesuvian cone, it must,

NOTES 331

besides the materials that cover Pompeii on the mountain slopes in
that direction, have ejected sufficient also to form a cone twice the
size of that of Vesuvius, to fill up the great crater, and upon the base
of this another at least half the size of that now visible" (p. 38). If
this is possible, there would seem to be no particular reason to
assume that any cone — corresponding to the modern Vesuvius — ex-
isted in 79, since a new one with full dimensions might have been
built up at the base, or through the flank, of what already existed —
that corresponding to the Somma wall. We should then have the
single summit which some think necessary to harmonize with the
older descriptions of the volcano.

14 (p. 136). See the discussion of this subject in Lyell's " Princi-
ples of Geology."

15 (p. 168). Journal des Mines, Paris, 1796, 3 (part xviii), p. 58.

16 (p. 170). The map accompanying Labat's " Nouveau Voyage
aux Isles de V Amerique" locates a lake {Lac des Palmistes ?) on Mont
Pelee.

17 (p. 181). A further discussion of this relation is contained in
Chapter XIX, " The Phenomena of the Eruption."

18 (p. 183). Professor Lacroix, in the Comptes Rendus for No-
vember 10, 1902, p. 772, states that the new fragmental cone now
overtops the eastern rim of the crater (the southwestern border of
the basin of the Lac des Palmistes) by fifty metres, and approaches
the bounding wall to within about one hundred metres. The depth of
the surrounding atrio is thought to be in places not more than one
hundred and fifty metres. The upward growth of this cone has not
been latterly so rapid as Professor Lacroix appears to believe, as my
photograph taken on June 1 shows the summit to be but little below
the crest of the volcano already at that time. Lacroix also makes the
interesting statement that the cone appeared to be completely solid,
and to have no central chimney (" entierement constitue par des roches
solides . . . Ce cone ria certainement pas de cheminee centrale ; quand il
y a peu de vent, toute les fumeroles qui sortent de cesflancs s'elevent verti-
calement et donnent Villusion d'un panache terminal").

332

NOTES

19 (p. 185). A further discussion of this subject appears in Chap-
ter XIX, "The Phenomena of the Eruption."

20 (p. 229). Pond and Percy Smith, in Transactions of the New
Zealand Institute, 1886 (1887), XIX, p. 350.

21 (p. 240). Humboldt, Cosmos, Bonn's Edition, Y, (1872), p. 422.
For a further discussion of this subject see Chapter XIX.

22 (p. 248). " But a hurricane blast of steam charged with burn-
ing dust did not sweep down from La Soufriere as it did from Mont
Pelee" (Eussell, National Geographio Magazine, July, 1902, p. 275).
This statement is perhaps not fully in accord with the observations
of some other investigators. The dominant feature of the Soufriere
eruption, as described by Tempest Anderson and J. S. Flett, repre-
senting the special Commission of the Eoyal Society, seems to have
been most strikingly similar to that of the eruption of Pelee. " Those
who were in the open air saw a dense black cloud rolling with terrific
velocity down the mountain. . . . The cloud was seen to roll down
upon the sea, and was described to us as flashing with lightning, espe-
cially when it touched the water. All state that it was intensely hot,
smelt strongly of sulphur, and was suffocating. They felt as if some-
thing was compressing their throats, and as if there was no air to
breathe. There was no fire in the ordinary sense of the word, only
the air was itself hot and was charged with hot dust." (Proceedings
Eoyal Society, London, August 22, 1902, p. 428.)

23 (p. 256). A further discussion of this subject appears in Chap-
ter XIX, " The Phenomena of the Eruption."

24 (p. 260). For a discussion of the relations existing between the
Sierra Merida and other continental mountain chains see Suess : " Das
Antlitz der Erde" 1, pp. 700 et seq.

25 (p. 263). " Les Manifestations volcaniques et sismiques dans Ic
groupe des Antilles" (Revue Generate des Sciences, July 30, 1902).

INDEX

Afterglows, 50, 255, 301-305
Ajoupa-Bouillon, 192, 213, 219;

partial destruction of, 232
Antilles, Lesser. See Lesser An-
tilles
Ash, volcanic, distribution of, 50,

288
Assier, journey to, 140-150
Atmospheric disturbance, 294

Basse-Pointe, overflow at, 78, 104,

174
Blast, destroying, nature of, 311
Bourdon, Heights of, 232

Cable, breaking of, 54-56

Calebasse, the, 213, 236

Capot, the, 157, 174

Carbet, 46, 47, 107

Carbet, Pitons of, 10

Caribbean Eegion, disturbances
in, 58; volcanic relations of,
257-270 ; a region of break-
ages and subsidences, 258-
262; earth movements in,
263-265

Ciparis, Auguste, prisoner of Saint
Pierre, 116

Destroying blast, nature of, 311
Dion Cassius, account of Vesuvian
eruption, 124

Earthquake in Fort-de-France,
210

Electric illumination, 231-234,
251-252, 291-294

Eruptions of May 8, 35-59 j dust-
distribution, 49 ; afterglows,
50 ; intensity of sound, 239 ;
force of, 278; electric illu-
mination, 291 ; magnetic dis-
turbance caused by, 298;
earth pulsation, 310 ; atmos-
pheric disturbance, 310

Etang Sec, 38; boiling up, 63;
breaking of barrier, 69 ; seat
of the eruption of May 8,
271

Falaise, the, 157 ; supposed crater

of, 213
Fonds-Core, 20, 116
Fonds-Saint-Denis, 107
Force of eruption, 278
Fort-de-France, 2-6; earthquake

in, August 24, 210

Grande-Anse, 146
Grande-Kiviere, 208, 225

Herculaneum, 36 ; destruction of,
compared with that of Saint
Pierre, 121-139

333

334

INDEX

Hovey, E. O, on Pelee's crater,
178 ; on Soufriere's crater,
245

Kennan, George, interview with
prisoner of Saint Pierre, 117 ;
on the nature of explosive
blast, 316

Krakatao, eruption of, 35-36

Lac des Palmistes, 38, 153, 160,
169-172

Lamentin, 142

Landes, Professor, interview with,
74

Les Colonies, articles in, 61 et seq

Lesser Antilles, geographic rela-
tions of, 257-270 ; volcanoes
of, 261-263 ; broad extent of
seismic and volcanic disturb-
ance, 265-270

Leyritz, Habitation, 216, 225.

Lhuerre, Acting Governor, his
official report, 109-112

Lorrain, 79

Macouba, 208, 225

Magnetic disturbances, 298

Martinique, characters of, 8-15 ;
area, 11 ; population, 12 ; cul-
tivation of cane in, 13 ; ab-
sence of railroads, 14

Mary, Pere, 104, 107, 189-196;
death of, 195, 241

Morne Balai, 217 ; partial destruc-
tion of, 231

Morne Capot, 232

Morne de La Croix, 38, 161, 168;
destruction of, 172.

Morne d'Orange, 29, 116

Morne Eouge, 20-21, 189-196;
annihilation of, 195, 232 ; loss
of life in, 241

Noise of eruption, 305

Parel, Vicar-General, his chroni-
cle, 85-108

Pelee, eruption of, May 8, 35-59 ;
of May 20, 105 ; eruption of,
compared with that of Vesu-
vius in year 79, 121, 139 ; as-
cents of, May 31 and June
1, 151-165; crater of, 162-
165 ; geography of, 166 et seq.;
height of, 168 ; crater of,
176-180; eruption of, June
6, 197 et seq.; unsuccessful
effort to ascend, 210 ; ascent
of, August 30, 216 ; electric
display of, 229 ; eruption of,
August 30, 231-243 ; sympa-
thetic relation with Soufriere,
251-252

Phenomena of eruption dis-
cussed, 271-318

Pliny, account of Vesuvian erup-
tion, 121 et seq.

INDEX

335

Pompeii, 36 ; destruction of, com-
pared with that of Saint
Pierre, 121-139

Pouyer-Quertier, report of officers
of, on eruption of June 6,
200

Precheur, 91, 103, 174; ruins of,
204-207; conditions in, in
early May, 205

Precheur Eiver, 77

Procureur of Martinique, report
of, 111-116

Eiviere Blanche, 66, 69, 88-89,

175, 201, 203
Eiviere-des-Peres, 20, 78,175,211
Eiviere Seche, 175, 211
Eoxelane, the, 20, 78, 175

Sainte-Philomene, 47, 103, 116
Saint Pierre, 15-34 ; ruins of, 21-
34; objects found in, 25-27 ;
ash-covering of, 33; destruc-
tion of, 37-49, 82; days of
fear and trembling in, 60-72 ;
last day of, 73-84; annihila-
tion, 82; number of killed,
83 ; burning of, 109-120 ; pri-
soner of, 116 ; destruction of,
compared with that of Pom-
peii, 121-139

Shock of eruption, 305

Soufriere, the, 244 et seq. ; craters
of, 245 ; height of, 246 ; erup-
tion of, May 7, 247-351;
eruption of September 3-4,
251-254; of October 16,
270

Tarawera, electric afterglows of,

229
Ti-Bolhommes, the, 212
Trianon, assumed crater of the,

213, 214
Trinite, 145

Usine Guerin, 69, 88

Vesuvius, eruption of year 79
compared with that of Pelee,
121-139 ; assumed decapita-
tion of, 128 ; ancient form of,
131

Vive, 150, 233

Volcanic relations of the Carib-
bean Basin, 257-270

Volcanoes of the Lesser Antilles,
261-263

Wind, volcanic, 52, 296

MAP OF THE NORTHERN PART OF MARTINIQUE
Showing approximate areas of destruction

The oblique shading: Zone of destruction determined by the May eruptions, with the region

of absolute annihilation shaded dark
Dotted shading: Extension of field due to the eruption of August 30
Scale of distances in kilometres

id*

PLEASE DO NOT REMOVE
CARDS OR SLIPS FROM THIS POCKET

UNIVERSITY OF TORONTO LIBRARY

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Mont Pelée and the tragedy of Martinique : a study of the great catastrophes of 1902, with observations and experiences in the field"

See other formats