C. (Conrad) Keller.

Madagascar, Mauritius and the other East-African islands online

. (page 1 of 17)
Online LibraryC. (Conrad) KellerMadagascar, Mauritius and the other East-African islands → online text (page 1 of 17)
Font size
QR-code for this ebook

.iyii\.U Hi J. i U O


,<J^(»«^,-S5-^| ,


Gift of







yv^ 7







Professor Dr. C. KELLER

^Vith 3 Coloured Maps and 64 Illustrations


P A T E R N O S T I'. K S (j r A K i:.




Printed in Holland.

^ l(p. 9

4 816 7


The task of describinu^ the East- African Island-W'orltl
must be regarded as a grateful one in itself, for every-
thing unites in lending an exalted charm to the subject.
In these islands Tropical Nature displays her magic in
all its fulness, and their history is replete with remarkable^
incidents. If I were master of the power of expression
of a Bernardin de St. Pierre, whose descriptions of the
scenes of nature are unsurpassed, I would make the
attempt to give an adequate picture of the mighty
Tropical Nature — but I feel that I must confine myself
to a pale and realistic sketch of that lovely island-world,
and I thus stand in need of indulgence.

It is now twelve years since I visited the Seychelles,
the Mascarenes and Mauritius, regions which had lieen
till then but little visited. Hvents have since that time
brought the East African Archipelago, especially Mada-
gascar, into the foreground of European interest. As
during my journey I had devoted myself almost exclusivelv
to working out special questions of Natural Science, I
have been under the necessity of discussing matters foreign
to my pursuits.

Fortunately, earlier workers in the same field are not
wanting. The Mascarenes, for example, have been
described again and again. Alfred Grandidier. of I'aris,
has devoted his whole life to the exploration of the
colossal island of Madagascar, and his magnificent work
forms a rich mine of information

I am personally greatK' indebteil to this enu'neiu Ireiich
geographer for having furnished me with ettective
recommendations to the hrench authoritit.-s in Mada-
gascar and thus eminently facilitating my studies. I am
also indebted to him for sending me his })ortrait. wliich


is inserted in tliis volume in honiao;e to the great services
he has rendered to the investigation of Madagascar.

While the manuscript was in the press, there appeared
the valuable contributions by the German traveller
Dr. Voltzkow on the hitherto imperfectly known islands
of Juan de Nova and Aldabra.

It has been considered desirable that the islands lying
in the west of the Indian Ocean towards Australia should
also be incorporated in this work. I have not visited
these islands, but we possess sufficient information about
them from recent German, French, and English expeditions,
and of this I have made use.

The illustrations are partly from photographs taken by
myself. A number of the pictures of Madagascar, many of
which have indeed been published in another form, were
borrowed by me from the ''Revue generale des Sciences",
which not long ago published as a special number an
excellent paper entitled '' Ce qu'il faut connaitre de
Madagascar". The pictures of the Seychelles are nev/,
and for them I am indebted to the great kindness of
Dr. A. Brauer, who took very successful photographs on
the spot and forwarded plates to me to make use of.

Other illustrations, admirably executed by Dr. Voltzkow,
have also been turned to account.

C. Keller.


There is no book in existence which presents so complete and trustworthy
an account of ^ladagascar, the Mascarenes and the smaller islands of the Ocean
to the East and South East of Africa as this worlc of Professor Keller. To the
scientific thoroughness and accuracy of the German professor he ad is the vividness
and realism due to personal travel and investigation. His theories of the ethno-
logical history of Madagascar appear to me to be unanswerable, although they
differ from those of most writers on the subject, and even in some respect from
those of Grandidier, with whom, however, he is generally in accord.

The geologist, the botanist, the zoologist, the meteorologist, the ethnologist, and
the historian will all find new and interesting matter in the book, treated in an
eminently rational and scientific tone, and embellished with skilfully selected

H. A. Xesbitt.












































C O N T E i\ T S


Situation and Extent of Madagascar . . i

History of Discovery and Exploration . . 3

Surface and Geological Structure ... 12

Cllmate 23

Flora and Fauna 29

Population 59

Politics and Religion 90

History of European Colonization . . . loi
Organization and Govern.ment of the

French Colony of Madagascar .... 118

Produce 122

Commerce and Modes of Communication . 136

Towns and Villages 147

s.maller isl.a.nds of the type of madagascar 153

The Mascarenes i66

Reunion 178

Mauritius 189

Rodriguez 197

The Seychelles 200

The Aldabra Isles 210

Austral- African Islands 217

New Amsterdam and St. Paui 218

The Prince Edward Isles 224

The Crozet Isles 226

The Kerguelen Isi^es 228

Heard Island 237

Principal Authorities Consulted .... 239

Index 240



Chart Opposite i

Madagascar and the East-African Islands after Edrisi, 1153 8

Pedro Retnel's Map of Madagascar, 15 17 Opposite 8

Alfred Grandidier g

Geological Map Opposite 18

Geological Section from Majunga to Suberbieville (after
Leon Suberbie). i. Recent Alluvial Formations. 2. Early

Alluvial Deposits. 3. Chalk. 4. Jurassic Rocks 20

Madagascar Screw-Palms {fandanus) 30

Travellers' Tree {Ravenala madagascariensis) after Sibree ... 32

Raphia-Palms {Rophia madagascariensis) 34

Baobab {Adansonia Grandidieri) from the Re\ue des Sciences . . 38

Babakota {Indris brevicaiidatus) 42

Ruffed Lemur {Lejjiur varitis) 43

Ringtailed Lemur {Lcimcr cattd) 45

SiFAKA iPropit/iccus) 45

.\L\dagascar Tenrec {Caitetes) 48

Gigantic Ostrich of Madagascar {.Ppyoniis iiigens) (Revue des

Sciences) -2


Excrement of the Earthworm Kyiwtus Danoinii (greatly

reduced) eg

Ethnological Map of Mada*gascar Opposite 60

HovA Women (after a photograph by Dr. A. Voltzkow) Opposite 64

A Kanusikv-Sakalava of Morondova (from a photograph by

Dr. .A. Vohzkow) 65

A Sakalava Woman 5^

Sakalava Women (from a photograph l)y I )r. A. Voltzkow) Opposite 68

Bara (from the Revue des Sciences) 69

A Tanala (;irl (from the Revue generale des Sciences) Opposite 72

A Betsimisaraka with Bamboo-Guitar (from the Revue des

Sciences) .... „ .




A Sakalava House 78

Burial Place of the Betslmisaraka (from the Revue generale des

Sciences) 87

A HovA Slave Woman in Imerina 91

View of the Former Roval Palace in Antananarivo (from the

Revue des Sciences) 93

General Duchesne 116

Madagascar Agricultural Labourers 124

A HovA Woman at the Loom (after J. Sibree) . . . Opposite 132

Mode of Travelling in a Filansana 137

River Boat in East Madagascar (from the Revue des Sciences) 139

Madagascar Boat with Outrigger 140

Bazaar in Tamatave Opposite 142

Landing Place at Point Hastie in Tamatave 144

Street in Tamatave Opposite 148

The village of Mvhasoa in Eastern AL\dagascar 150

Andavakoloko on Nossi-Be (from a photograph by Dr. A.

Voltzkow) Opposite 156

Hellville on Nossi Be 157

Nossi-CuMBA — Village of Ampagorina (from a photograph by

Dr. A. Voltzkow) Opposite 158

Map of the Comoro Isles 160

Vegetation of Reunion Opposite 166

A Creole Country House in Reunion 173

Ravine near St. Paul (Western side of Reunion) . . Opposite 178

^L\p OF Reunion 179

CiLAOs and Piton des Neiges (Reunion) 180

St. Paul, the Oldest Settlement in Reunion . . . Opposite 186

Map of Mauritius 190

Map of the Island of Rodriguez 197

Map of the Seychelles 201

View on the shore of the Island of Mahe with Masses of
Granite; a Creole settlement in the foreground (from a

photograph by Dr. A. Brauer) 202


photograph by Dr. A. Brauer) Opposite 204

Vegetation on the Island of Mahk (from a photograph by

Dr. A. Brauer) 205

View of the Island of Mahe from the North; in the Back-
ground THE Peninsula of Port Victoria (from a photograph

by Dr. A. Brauer) Opposite 208



Map ok the Aldahra Isles 210

Vegetation on Cocoa-Nut Island, Aldabra Isles (from a

photograph by Dr. A. Voltzkow) 214

Map of New Amsterdam 218

Map of St. Paul 220

Map of the Prince Edward Isles . . . , 224

Map of the Crozet Isles 226

Map of the Kerguelen Isles 229

Kerguelen Island: Precipitous Cliffs with Ijasaltic Columns

(from the Tour du Monde) Opposite 230

Kerguelen Island: View from Royal Sound 231

Pringlea Antiscorbutica (from the Report of the Challenger

Expedition) 234

Map of Madagascar (after Hansen) Opposite 238


x^LTHOU(;n the island world which is rang'ed before
die mainland of Africa in the Western part of the hidian
Ocean is far and away less imposing- in extent than the
Archipelago in the North East of this oceanic region,
yet it is probably not in any way inferior in the indixidu-
ality of its characteristics.

Man and Nature here combine to form a world of
itself, which, while markedly deviating from the con-
tinental character, exhibits a singular mixture of the peculi-
arities of African and Asiatic life, and this fundamental
characteristic recalls, so to speak, the geological histor\'
of the East African islands. Certain points in the distri-
bution of animals led long ago to the supposition of
an ancient connexion both with Southern Africa and
Southern Asia, and in recent times our increased geological
knowledge has made this supposition seem more and
more probable.

Of course the history of the formation ol the seveial
islands is neither uniform in character nor synchronous.
Madagascar and the Seychelles ap[)ear to be the oldest
members of the group ; their insulation began in the
mesozoic period and is connected with the formation of
the extensive sea-basin which is now filled In* the


Indian Ocean. These two island districts are old
mountain tops which have remained above water, their
connexion with the African continent ceasing from the
eocene period. The Comoro Isles and the Mascarenes,
as well as the small islands lying far out in the ocean,
are of volcanic nature and were probably raised above
the surface of the water at a later date. The long
duration of this isolation has not been without a far-
reaching influence on the organic world. The indigenous
genera are strikingly numerous, but the animals which
immigrated in the miocene period and are so characteristic
at present of Tropical Africa, are wanting.

People have tried to ascribe a really ancient character
to the human inhabitants, and to explain them as
primitive forms standing between the races of Africa and
Southern Asia. This was of course overshooting the
mark, and the hypothesis did not correspond with the facts
of the cases. Modern investigations have led to far
more sober conclusions. The settlement of the East
African Islands by man is of comparatively late date, having
taken place, as regards a part of it, in later historic
time, and it exhibits a motley chart of population, as
the people belong to very different race elements and
to no less than three different quarters of the Earth.
No sufficient proof has yet been given of the existence
of an autochthonous race. Lying beneath the magical
sky of the Tropics these islands display a wealth of
natural beauty which has long been renowned and which
indeed cannot be surpassed in grandeur.

Some of the islands have long attracted the attention
of Europe as suitable districts for colonisation, and just


at present European capital and European enterprise
are attempting in an increased measure to inaugurate
remunerative colonization in these regions. If, as there is
every probability, Africa becomes more effectively opened
to European influence, these islands will assuredly come
more and more to the front.


tVim .^cmrunM^tjvt. i Co iw Zund,^

Prinii-d 'bj'Vf. t A K-.Toliiiston. Edinburgh i London.




The mighty island of Madagascar, whicli the French
Creoles of the neicrhbouringf Mascarene Isles call La Grande
Terre, lies nearly parallel to the coast of East Africa
from N.N.E. to S.S.W. The Mozambique Channel,
which separates it from the Continent, is some 250
miles across. It is the largest island in the western
portion of the Indian Ocean and, if we omit Green-
land, is only surpassed in extent by New Guinea and
Borneo, being 230,000 sq. miles in extent. Its shape
is elliptical, the major axis, running north and south,
is 1000 miles in leng-th, while the greatest breadth,
at the latitude of Foule Point, about the middle of
the island, separates the east from the west coast
by 375 miles. It extends over nearly 14 degrees of
latitude, from Cape Amber, the most northerly point, in
11° 59' 52", according to Grandidier, to the most southerly.
Cape Ste. Marie, in 25° 38' 55" S. lat. The positions
of these places on the old map by Pedro Reinel, dated
1 51 7, are tolerably near to those given by modern
observations. The most westerly point, in the Bay of
Fandivotra, lies in 43° 11' E. long., and the East Cape
in 50° 27' E. long. In comparison with this insular
colossus the islands which lie near, as well as those
far out in the ocean, sink into insignificance. A visitor
receives the most imposing impression b)' looking
at the coasts in bright weather from the eastern side
of the island, because there the mountains are toler-
ably near to the coast. Mountain ridges running north


and south rise in an amphitheatre from the plain on the
coast like motionless waves \ they get higher and higher,
and in the distance, in sharp outline against the blue of
the sky, one may recognize the primeval forest of the
Upper Mountain region.

The western side of the island is far more level and
is thus less impressive.



In ancient times our acquaintance with the surface of
the Earth was very limited, and until the beginning of
the Christian era practically nothing was known of East
Africa beyond Cape Guardafui, In the Second Century
two Greek navigators, Theophilus and Diogenes, sailed
round this Eastern Promontory and reached a harbour
which they called Rhapta. Soon afterwards another
sailor, Dioscorides, sailed yet further south and reached
Cape Prasum (on the Mozambique coast?). According
to the account of these mariners, as Ptolemy informs us,
the island of Menuthias lies not far off, and can be
reached in two days either from Rhapta or from Cape
Prasum, while it is about the same distance to the
Pyrolean Archipelago. This latter name points to islands
of a volcanic nature and may be held to signify the
Comoro Isles, and in that case Menuthias cannot mean
anything but what is now called Madagascar, a conclusion
rendered still more probable by other items of inform-
ation. When we are told that Menuthias is covered
with forest and possesses rivers, Zanzibar and Pemba are
excluded ; while the statements that the only wild beasts
found are crocodiles, that large tortoises live there and
that the natives make use of dug-outs as boats, agree
thoroughly with a description of Madagascar.

At the beginning of the Middle Ages the Arabs repeat-
edly obtained more accurate information concerning the
island ; the Arabic geographer Edrisi, who collected
reports from traders and navigators, mentions in his map


an island called Chezbezat, which is obviously identical
with Madagascar. Other Arabic authors use other appella-
tions, as Serendah, or El Komr (Moon-Island).

The first to make use of the name Madagascar was
the Venetian, Marco Polo. He never visited the island,
but depended entirely on the information of the Arab
mariners. It will not be without interest to quote his
description word for word. He says : '' Madagascar is
a larcre and beautiful island a thousand miles distant from
Socotra. It is 4000 miles in extent. Its inhabitants are
Moslem and all live by trade. Every day they slaughter
a large number of camels for food. In this island, as in
Zanzibar, which lies to the south of it, there are more
elephants than anywhere else in the world, and there are
found there also leopards, lions, giraftes and wild asses.

''The forests abound in red sandal-wood and on the
coast ambergris is collected. This is a product of the
whale, which is numerous in those seas.

"Many ships sail thither and obtain rich profits from
silks and other articles ; there is an important trade in ivory.

" Madagascar and Zanzibar are the most southerly islands
visited by the Indian mariners, but the current towards the
south is so strong that the return voyage is difficult.

" In the countries which lie yet further off is found the
' Griffin ', or Roc, which is not, as related, half bird and
half lion, but is rather a gigantic eagle, which covers with
its wings a space of thirty paces and carries off elephants
in its talons ; these it drops down from a height and then
feeds on the crushed flesh. The Khan of Tartary
possesses a feather of this Roc, nineteen spans in length,
and two wild-boar tusks each weighing fourteen pounds."

We may assume that in these statements of Marco
Polo there is a mixture of truth and falsehood. The imagin-
ative Orientals had communicated many gross exaggera-
tions to the Venetian. Many of these statements may, how-
ever, refer not to Madagascar, but to Magadoxo on the


Benadir coast, the inhabitants of which, like all the Somali,
breed camels for food. Lions, giraffes and wild asses are
frequent on the East Coast, but are not to be found in
Madagascar, nor is ivory to be obtained there.

On the other hand, the account of a giant griffin receives
some real support from the discovery of the remains of
gigantic ostriches {/Epyornis inaximus^ yE. ingcns)^\\\(\Q)i\
once inhabited Madagascar, as these huge cursores were
apparently still living in those days, or at any rate they
still lingered in the memory of the people. If we consider
how fond the Arabs are of exaggeration, we shall under-
stand these strange additions to the simple facts. Of course
the /Epyornis could not fly; the so-called feathers of
which the Khan possessed a specimen, were, according
to the opinion of Grandidier, which appears very probable
to me. nothing but bamboo stems which the Arabs con-
sidered as the shafts of large bird feathers. The 14-
pound tusks also cannot be accepted as coming froni
the island; they were either hippopotamus teeth, or, as
appears to me more likely, the huge tusks of the East
African wart hog {Pliacoclicerits c?tliiopiciis)\ these are
often offered for sale on the sea-coast as curiosities.

It was in the beginning of the Sixteenth Century that
more accurate information was first obtained in Europe.
The epoch-making expeditions of the Portuguese under
Vasco da Gama and Bartholomew Dias had reached the
south point of Africa and had discovered the sea route
to India by the end of the Eifteenth Century, and thus,
through the neighbouring Arab traders on the Mozam-
bique coast, had heard also of Madagascar.

In the year 1505 King Manuel of Portugal fitted out
an expedition of 22 ships and 15,000 men. This sailed
in March 1505, under the command of Dom Erancisco
de Almeida, the first Viceroy of India, and was com-
missioned to establish fortresses for the protection of
the Portuguese trade in Sofala and Ouiloa. In the follow-


ing year Almeida sent ships back to Portugal with spices
as a token of his successful voyage ; these ships were com-
manded by Fernando Soares, and on their home voyage
from hulia they reached the east coast of Madagascar,
Feb. 1st, 1506, and thus Soares is the real discoverer
of the island. On the loth of August of the same year
1506, the Portuguese Joao Gomez d'Abreu discovered
the west coast of Madagascar and sailed into the Bay of
Formosa, probably between Point Barrow and Point Croker.
As it was the day of St. Lawrence the island received
the name of San Lorengo, and this appellation is gener-
ally given to it in the maps of the beginning of the
Sixteenth Century. The navigator Tristan da Cunha
received information about the island from one of his
captains, Cuntinho by name, who had to put into a
harbour in Madagascar for shelter; he then visited several
points of the west coast and reached the north point,
which at first he designated Cape Natal.

Gomez d'Abreu after rounding the most northerly
Cape, sailed along the east coast, landed near Matatane
and even left some Portuguese behind on the coast,
where in the following year, 1 509, Diego Lopez de Sequira
effected another landing, afterwards sailing along the coast
as far as the Bay of San Sebastian. Thus different points
of the island became known in rapid succession and were
visited by the Portuguese on their voyages to hidia.
Th(^re was, however, no permanent occupation of the
island, perhaps only for the reason that the Portuguese
already had sufficient colonial possessions in South Ame-
rica, Africa, and Eastern Asia. Between 1595 and 1598
the Dutch landed in Madagascar, but were apparently
frightened away by the heavy losses sustained by their

Towards the middle of the Seventeenth Century there
began, as will be described later, attempts at coloniza-
tion on the part of the French, who designated the


island ''He Dauphine," an appellation which did not,
however, gain permanent acceptance.

The name ot Madagascar seems to be of foreign origin,
coming probably from East Africa. The Malagasy them-
selves did not know this name originally, and when they
wished to designate the island as a whole they used
different circumlocutions, as, Izao rehetra izao or Ny taiiy
rehetra^ i.e. '' all this country ", or even Ny anivon ny
riaka (literally "-in the country that lies in the midst of
the sea"). It is peculiarly interesting to us that early
travellers give us wonderfully glowing accounts of the
island and of its marvellous fertility, and sometimes
consider the inhabitants as the happiest people on \\\(t

The erowth of our knowledoe of the district can be
accurately followed by means of the numerous maps in
existence, most of them, however, of doubtful value. Alfred
Grandidier has taken the trouble to make a collection
of these maps and to reproduce them in an excellent
manner in numerous pages of his '' Histoire de la
Geographic de Madagascar." The map by Edrisi of 1 153
is in great degree a product of the imagination, and even
Martin Behaim, who depended upon the confused infor-
mation of Marco Polo, gave quite an arbitrary shape to
the island on his globe. The first useful map, the first
which gives the general outline of the island with any
correspondence to the reality, is of the year 1517. It
was executed in Seville by the celebrated Portuguese
geographer, Pedro Reinel. A great number of later maps
are obviously only copies of this, and even the large
scale productions of Placourt (1656) and of Henjowski,
so renowned for his extravagances, are obviously based
upon the map of Pedro Reinel.

We may recognize a real advance in the cartogra[)hic
sketch of d'Apres Manevillette (1775). It is confinc^d
to the circuit of the coast and gives no kind ol topo-

Madagascar and the East-African Islands, after Edrisi, A.I). 1153.

Map of Madaj^rascav l)y Tolro ReiiR-l. 1517.

To fail' p. S.


crraphic details as to the still wholly unknown interior. We
are indebted for important details to Owen, who in 1824

1 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Online LibraryC. (Conrad) KellerMadagascar, Mauritius and the other East-African islands → online text (page 1 of 17)