Ha)
i

eee

"

2m,

cansae “Eo,

CORAL AND ATOLLS

Ss

ORGE CLUNIES ROS

7
vi

I

ATE G

EL

22]

eo saa
1b 7 oz. caRDEP
Invert. Z00|

CORAL AND ATOLLS

A HISTORY AND DESCRIPTION OF THE
KEELING-COCOS ISLANDS, WITH AN
ACCOUNT OF THEIR FAUNA AND FLORA,
AND A DISCUSSION OF THE METHOD OF
DEVELOPMENT AND TRANSFORMATION

OF CORAL STRUCTURES IN GENER

AUG 4998

Pht

aN

Bg BY
¢ ;
F. WOOD-JONES, D.Sc.,
F,Z,S., EVCs!
KE-ISSUE
LONDON:

LOVELL REEVE & CO., LTD.

Publishers to the Home, Colontal, and indian Governments
6 HENRIETTA STREET, COVENT GARDEN
1912

oY Iw

rite
Mal. Lav.

TO THE LATE
GEORGE CLUNIES-ROSS

PREFACE

For fifteen months I dwelt in Cocos-Keeling atoll. The
charm of so remote a spot, the romance of its story and of its
proprietor, and the fact that it was Darwin’s atoll, were the
reasons for my residence—its purpose was the medical care
of the handful of men engaged in working the cable which
breaks its journey from Australia to Africa upon the shores of
the most northern islet of the group. —_-

Upon this islet-—Pulu Tikus—I lived, tad shared its
inconsiderable strip of coral land—some thousand yards in
length—with fifteen Europeans, some twenty Chinese servants,
three Malay boatmen, and two Hindoo clerks. The outlets
for medical enthusiasm were therefore strictly limited, and
much time was to hand for the study of the place and its
‘fauna, and for speculating upon the problem of how so tiny a
speck of land came to be made in the midst of that great
waste of ocean.

I arrived early in June 1905, and lived upon Pulu Tikus
until September 30, 1906. In January 1907 I returned, and,
for a time, lived with the Governor at his home on Pulu
Selma. During all these months I examined every portion of
the atoll, became familiar with its living creatures, and made
every endeavour to appreciate the life-processes of those most
‘misunderstood of architects—the reef-building corals. . Most
of my days:-were passed in wading about the barrier flats,
sailing or paddling a boat in the lagoon, or making excursions
‘to the other islands of the atoll ring. The beauty of the clear
‘water of the lagoon and the breaking surf of the barrier made
these journeyings a perpetual delight, for the whole field-work

Vii

Vili PREFACE

of a coral-island naturalist is done amidst a splendour of
tropical sunshine and luxuriant vegetation.

In all my doings I was aided and encouraged by the
members of the Clunies-Ross family, and the kindly assistance
of George Clunies-Ross—a man born with a true love of
nature, trained by his life’s work to observe, and familiar with
every creature that lives within the confines of his domain—
was always ready for my guidance. To him, and to the
members of his family, I am deeply indebted, not only
for an intimate knowledge of many things not easily known
to a stranger, but for a splendid hospitality and a true
friendship.

When, in 1907, 1 returned to the atoll and lived under
his roof, I had the advantage of discussing with him those
many details of coral-island biology which, though they may
be guessed at, assumed, or scoffed at by a museum naturalist,
are things familiar to a man whose long life of keen observation
has been passed in their midst. )

The greater part of this account of the atoll was written
during my residence in the islands, and this fact I mention for
two reasons: first, as an apology for scant references to former
papers on the same subject; and secondly, as a guarantee that
the ideas put forward are those actually arising out of the
study of the atoll, and not of its literature.

A word must be said with regard to the native names
employed in these pages. Throughout I have used those
names to designate the islands, and the animals and plants
that live upon them, which are in everyday use in the atoll;
and for this I make no apology. I do not regret the fact that
the names here used to distinguish the islands do not
correspond with those on published charts. Did a stranger
arrive in the atoll and ask his way of any one—native or
European—to “Sczvola Islet” or “Workhouse Island,” he
would certainly not arrive at his destination ; and the giving of
such made-up names to pieces of land already possessing well-
known native names has nothing to recommend it. The names
that [ have employed are those in local and long-established

PREFACE 1X

use, and are those that every man, woman, and child about the
atoll constantly employs.

Again, with the animals and plants, a native name is
altogether useless unless it is the local word employed to
designate the species. It is vain to urge that a particular
word is not the one employed by Malays in the Straits to
distinguish a certain fish, and therefore is not good Malay ;
for the stranger who is unacquainted with the word chuchut
will glean no information about sharks in the atoll, though
in Singapore he used an altogether different word. I have
given native names for the Flora and Fauna of the group
solely as a guide for any subsequent collector, and to him no
names are helpful save those in actual local use. This I
have done because in my own case I derived considerable
help in the naming of species from the lists collected by
Dr. C. W. Andrews in Christmas Island.

Some details of the zoology of the islands I have already
published, for the most part in the Proceedings of the Zoolo grea
Society of London, and to this society I am indebted for permis-
sion to republish much of the contents of chapters viii, 1x.,
and x. and of Appendix I. For the re-use of some figures I
am also obliged to the Zoological Society, and for others to the
editor of the Badminton Magazine.

Notes on the Fauna of the islands were published in the
pages of the London Hospital Gazette, and with the kind
permission of the editor of that journal some of these notes
are incorporated here.

I owe a debt also to Mr. F. Hesse—general manager of the
Eastern Extension Telegraph Channa te the officers of the
company’s ships, and to the shore staff of the cable station, for
the use of material, and in many cases for the kindest personal

help.

PREFACE TO RE-ISSUE

Durine the short period that has elapsed since the appearance
of the first issue of this book some contributions have been
‘made to our knowledge of submarine conditions which have
a very direct bearing upon the debated question of the
bathymetrical limit of the reef-building corals. Firstly, the
“Michael Sars” North Atlantic deep-sea expedition has
determined that light-exerts a great influence on the Helland
Hansen photometer at depths of 100 metres, and some in-
fluence even at 1000 metres; in the clear waters of tropical
oceans this distance would almost certainly be greatly ex-
ceeded, so that it is safe to assume that some light asserts its
influence very far below the limit at which reef-building
corals live.

Secondly, a discussion which took place in the columns of
Nature towards the end of 1910 produced the very welcome
contribution of Mr. A. R. Hunt, who so long ago as 1882 had
conducted researches regarding the Limiting Line of Sedimen-
tation in wave-stirred areas, and determined the laws which
govern its presence.

Both these contributions afford valuable additions to the
subject-matter of chapter xxi. and furnish pe: real support
for the conclusions arrived at.

CONTENTS

PREFACE
PreFaAceE To RE-IssuE

INTRODUCTION

PART I. THE HISTORY OF THE ATOLL >

I. Tue Discovery aNp Earty History or THE IsLaNps
II. Tse Ortiein oF THe Crunies-Ross Famity
III. 1825-1854.
IV. Tue History or THE ATOLL, 1854-1871
‘V. HutsTtory or THE ATOLL FROM 1871

Native Customs

PART II. THE CORALS AND THE CORAL
| PROBLEM |
VII. Tue Earty Staces or THE Lire-History oF THE
Corats _
VIII. Tur Growrn or THE CoraLt CoLony
IX. Tue Lire-Processes or tHE Corat CoLony

X. Tue Processes or Deatu IN THE Cora CoLony

PART III. THE ATOLL AND ITS
PROBLEMS
XI. Tue Arott as a WHoLE

~ XII. Tse Supmarine Bank oN wHICcH THE AToLL Rests ~

xi

57

69
101
121

135
144

Xu
CHAP.

XIII.
XIV.

XV.

XVI.
XVII.
XVIII.

XIX.

OS

XXI,

XXII.

XXIII.
XXIV.

XXV.

XXVI.
XXVII.

CONTENTS

Tue STRUCTURE OF THE BARRIER
Tue Istanp Bracues
(A) Tue Seawarp Bracues

(B) Tue Beacues or THE LaGoon SuHore
Tue STRUCTURE OF THE ISLANDS
Tue LaGoon
Tue CLimaTe AND PrREvVALENT WINDS

Tue TuHeories or ATotL AND REEF ForMATION PUT
FoRWARD UNTIL THE PuUBLICAIION oF DaRwiN’s
THEORY OF SUBSIDENCE. E

Tue Tueories or ATott aND REEF FoRMATION PUT
FoRWARD SINCE THE PuBLICATION OF DaRwin’s

>?

Turory—Tue Tuerory or “ SoLuTion’

Tue Tueortes or AToLL AND REEF ForMATION PUT

ForRWARD SINCE 1880

Tue Meruop or DEVELOPMENT oF THE REEF Bank

AND THE REEF

DEVELOPMENT OF THE ATOLL

- Tue Arrer-Historny or THE ATOLL

Tue DrvyELOPMENT OF OTHER CoRAL STRUCTURES, AND

THE INDICATIONS OF ALTERATION oF LaNnp-LEVEL

PART IV. THE FLORA AND FAUNA OF

THE GROUP

Tue Metruops or Naturat CoLonisaTION
SHIP-BORNE AND WIND-BORNE COLONISTS

Tue Lirrorat anp Marine Fauna

212

232

278

286
302
314

CONTENTS

APPENDIX I. LIST OF THE FAUNA

MamMa.ia
AVES
ReptTivia .
PiscEes

Mo .tusca
LEPIDOPTERA
HYMENOPTERA
DieTera
CoLEOPTERA
RuYNCHOTA
NEUROPTERA
OrTHOPTERA
ARACHNIDA AND Myriavopa
CRUSTACEA

VERMES

APPENDIX II. LIST OF THE FLORA

DicoTyLEeDoNs

MonocoTyLEDoNsS

BIBLIOGRAPHY

INDEX

384

ILLUSTRATIONS

FACING
PAGE

Portrait of the late George Clunies-Ross Frontispiece
Map showing the Position of Cocos-Keeling Atoll 1

PLATES
PLATE

I. Bungalow of the Telegraph Station on Pulu Tikus 44,
II. A Colony growing as a Plate-like Growth (Montipora) 74

IL].

A Branching Growth approximating a Plate-like Form
(Madrepora) 76

. Chart of the Southern Group of Cocos-Keeling Islands 138

V. The Barrier Flats and Seaward Beaches of Pulu Tikus
at Low Tide 162
VI. Barrier Pools left by the Fall of the Tide, Pulu Tikus 166
VII. The Lagoon Shore of Pulu Tikus 172
VIII. A Typical View of the Vegetation of the Atoll 178
IX. Sand-piling of the Lagoon Side of the Island, Pu ’u Tikus 180
X. The Lagoon from the Western End of Pulu Tikus 186
XI. A Sunset in the Lagoon 200
XII. A Small Colony of Astreopora 250
XIII. A Fractured Surface of a Piece of Barrier-flat Breccia 254
XIV. The Island Rise, Sea Beaches and Barrier Flats, Sea-
ward Side of Pulu Tikus 256
XV. The Raised Steps of Breccia Platform of Pulu Tikus 280
XVI. Krakatoa as it appears to-day 290 »

XV

xV1 ILLUSTRATIONS

FACING
PLATE PAGE
XVII. Nest of Merula erythropleura 298
XVIII. The Smaller Frigate Bird (Fregata ariel) 300
XIX. The Nest and Eggs of the Philippine Land-rail 312
XX. The Ikan P’dang 322
XXI. Fauna of the Islands 346
XXII. Natives clearing away the Growth of Klenchi
(Guielandina bundoc) 372
XXIII. The Spathe of the Coconut Palm 380
XXIV. The Flower Spike of the Coconut Palm 380
XXV. Young Nuts developing on the Flowering Stems when
the Intertile Flowers have been shed 380
XXVI. Young Coconuts in stage klapa chinker 380
XXVII. Green Coconuts in stage klapa muda 380
IN THE TEXT
FIG. PAGE
1. A Wedding Group in Cocos-Keeling 52
2. Fungia: Living specimen 58
3. Fungia ; Same Specimen Dead and Dried 58
4. Montipora Colony, Living 60
5. Montipora Colony, Dead and Dried 61
6. A Section through a Portion of an Alcyonarian Coral to show
the Relations of the Zooid to its Skeleton 62
7. Larval Forms of Corals: Condition shortly after Extrusion 65
8. Larval Forms of Corals . 865
9. Two Young Growths of Seriatopora that have started their
Colonies upon a Living Oyster 66

10. A Section through a Commencing Colony of Cenopsammia
nigrescens 67
11. A Growth of Porites 73

ILLUSTRATIONS

. Example of a Coral Colony growing as an Encrusting Layer
3. Example of a Colony growing as a Tree-like Growth

. Example of an Amorphous Type of Growth

. Vegetative Reproduction in Corals

. Vegetative Reproduction in Corals

. Vegetative Reproduction in Corals

. Vegetative Reproduction in Corals

. Vegetative Reproduction in Corals

Vegetative Reproduction in Corals

. Vegetative Reproduction in Corals

Colony of Cenopsammia Willey: growing from the Lower

Surface of a Boulder

. Types of Growth of Montipora

. Types of Growth of Madrepora

. Types of Growth of Pocillopora

. A Colony of Pocillopora

. Rough-water Type of Highly Branched Madrepora

. Upper Surface (A) and Lower Surface (B) of a Partial Plate

Growth

. Diagram of Type of Growth of Madrepora when living in

water free of sediment (M. pulchra)

Diagram of Growth of Madrepora pulchra when living in a
habitat exposed to the action of sediment

. Colourless Stylopora Colony from Deep Water (8-12 fathoms)
. Colony of Poczllopora

. The Influence of Environment on Coral Growth

. Millepora Colony of the Type named complanata

. Millepora Colony of the Type named alcicornis

. Young Porites Mass grown equally round a Central

Nucleus

. Older Porites Colony in which the Lower Zooids are killed

by Pressure
b

“I
m5

Jj 4 = =
I qa

oo

107

XVill ILLUSTRATIONS

FIG. PAGE
38, Adult Colony of Porites in which the Upper Zooids are
killed by Sediment 108
39. A Small Colony of Agaricia 109
40. Type of Repair of Madrepora when the “ Dominant Apical
Zooid”’ is not destroyed 1s
41. Type of Repair of Madrepora when the “ Dominant Apical
Zooid”’ is entirely destroyed 113
42. Montipora Colony repairing a Branching Growth by an
Encrusting Layer 117
3. Colony partly killed by Deposition of Sediment 119
44, Colony showing Death by Sediment overtaking it from
below 126) 9
45. Specimen showing the Tunnelling of a Living Coral Colony
by Boring Parasites 127
46. The “ Stings” caused by Contact with a Colony of Millepora
alcicornis | 132
47. The Lagoonlet of Pulu Luar at High Tide 140
48. Chart of Keeling Atollon 142
49, Section of the Island and Barrier according to Darwin 152
50. Transverse Section of the Submarine Ridge and the Cocos-
Keeling Atoll 152
51. Section of the Atoll and Submarine Slopes according to Dr.
H. B. Guppy 152
52. Theoretical Section to show : i. The Breccia Platform (black) ;
ii. The Island Pile of Coral Fragments and Sand 153
53. Theoretical Section of Funafuti Atoll 153
54, Living Specimen of an Aleyonarian Coral 161
55. The Ikan Kakatua Ijou (Scarus) 165
56. Holothurians feeding at the Bottom of a Barrier Pool 166
57. Western End of Pulau Tikus 174
58. Diagram to Show the Prevalent Winds of the Atoll 203

59. Illustration used by Darwin to make clear the Stages of
‘Subsidence ” in Atoll Formation 219

ILLUSTRATIONS LX

FIG. PAGE
60. Photograph of a Coral Colony that has Living Zooids over
its Entire Surface 240
61. Three Figures showing the Formation of a Crescent-shaped
Island from some Original Obstruction to the Waves 258
62. Chart of the Cosos Atoll 260

63. Photograph of a Boulder of Alga-covered dead Coral Rock — 265

64. Diagram to illustrate the Development of Keeling Atoil,

Stage I. . 272
65. Diagram to illustrate the Development of Keeling Atoll,
Stage IT. Ble
66. Diagram to illustrate the Development of Keeling Atoll,
Stage ITI. 274
67. Diagram to illustrate the Development of Keeling Atoll,
Stage IV. 275
68. Diagram to illustrate the Development of Keeling Atoll,
Stage V. 276
69. The Robber Crab Bl
70. Small Crab that holds in its Claws two small Bouquets of
Living Algz 319
71. A Mixed Catch of Cocos-Keeling Fish 320
72. Landing a small Black and White Shark on the Seaward
Beach of Pulu Tikus 321
73. A Baracouta taken on a Trolling Line in the Lagoon 323
74. The Ikan Pareh (Dicerobatis Eregoodoo) 324
75. The “Green Fish,” Ikan [jou (Pseudoscarus) 326
76. The Ikan Buntal (Tetrodon Patoca) 328
77. Egg of Gygis candida, laid on the Branch of a Grongang
Tree 340
78. Gygis candida sitting on its Egg 341

79. Young Gygis candida, soon about to take its first flight 342

INTRODUCTION

I HAVE endeavoured to give in these pages an account of a
very restricted and very isolated piece of land.

I have attempted to put on record as fully as possible its

history, its physical conditions, and the state of its Flora and
Fauna.

There are many reasons why this work should be done, and
why, to-day, a census should be taken of the forms of life which
have found a footing upon this tiny speck of land surrounded
by a vast waste of ocean.

Before 1825 the isolation of the place was complete ; for
the next eighty years it was scarcely broken; but to-day the
intercourse with the greater world is on the increase, and with
this freer intercourse comes change—change in many unlooked-
for ways.

_ Steamship routes open up new roads for animal cimigrants ;
and nothing is more strange than the manner in which the
balance of Nature may be upset in these isolated spots by the
advent of what is apparently an inconsiderable addition to the
Flora or Fauna. If any argument were needed for the utility
of compiling uninteresting lists of the forms of life to be found
upon the atoll, the story of Christmas Island would suffice.
Were it not for the careful record of the Flora and Fauna of
that island, made in 1897 by Dr. C. W. Andrews, a remarkable
page in Nature’s story would be lost for ever.

In the case of Cocos-Keeling it is fortunate that no less
than three naturalists have left records of the state of the atoll
over a fairly long period, and the accounts of Darwin, Forbes,

and Guppy form invaluable landmarks.
XXl

XXIl INTRODUCTION

The human story of the settlement is also ripe for
chronicling, for much of the story of the Clunies-Ross family
is forgotten, and my fortunate access to the journal of the
Pioneer justifies the recording of this romantic story in its
complete form.

With regard to the atoll itself, and the problems with
which its origin and its physical conditions are surrounded,
this human story is of importance, for three generations of
opservant men have watched its changes and marked the
tendencies of its physical history.

I have endeavoured throughout this book to keep separate
the observations I recorded in the atoll and the inferences
which I drew from them; and this for the reason that, should
the inferences prove ultimately to be incorrect, the observations
may still stand as an index of the state of the atoll which may
be of use to any subsequent investigator.

The record of the cyclones that have wrought havoc in
the atoll has been given in the chapter dealing with the
meteorological observations, but I regret that the story of
their destruction, as told there, is not complete. On November
27, 1909, a cyclone, unprecedented in violence, swept over the
group. The damage wrought in a few hours of storm is
most vividly described in the Governor’s own words:

“The barometer fell as low as 27°92—lower than I have
ever known it fall before. We knew the day before that we
were sure to have a very heavy storm, and we worked all day
to secure and prepare for it; but unfortunately the whole of the
preparations came to naught. At about six oclock in the
evening the cyclone was on top of us, and by eleven o'clock that
night the centre passed over our isiands. After that we had
a lull for about half an hour, and then came the final blast,
which carried away and finished everything. Searcely a single
thing withstood its fury. The cyclone was accompanied by
great waves (sea-water, carrying sand with it, passed through
the tower of my house, which is fifty feet high), and these
waves left hardly anything standing. The whole of the
villages, as well as the working sheds and stores, were levelled

INTRODUCTION XXIll
to the ground, or wasted and carried away by the sea. I could
scarcely credit my eyes when daylight came; the wreckage was
so thorough and complete that the islands were unrecognisable.
Of my plantations of coconuts I am sure that not more than
one tree in a hundred stood it; in consequence I have lost
800,000 trees, more or less. As for lighters and boats, more
than 40 per cent. are lost, broken, or damaged.

“ Altogether only five buildings stood upright, and these
without roofs upon them. The whole of that night we were
practically under water or in the water, wet and miserably
bedraggled. Fortunately only one man was killed right off
by a falling tree, and another died ofexposure. It was simply
miraculous how the people were saved, and the tales of their
sufferings during that night are beyond belief. They were
scattered by the seas all over the islands, and one can only say
that Providence guided them into safety.

“ We are plagued by mosquitoes, flies and other insects, all
of varieties that I do not even recognise—and they pester the
whole group, and make life a burden.”

It is a sad duty to add this letter to the description I have
given of the prosperity and order of the settlement as I
last saw it. If it is permissible to extract interest from such a
disaster, I would call attention to the importance of the arrival
of presumed new species of insects with the wind. Again, the
fact that the waves swept right over the islands, and deposited
sand upon a tower fifty feet high, will serve to bring home to
those who do not believe in the power of waves to shape atolls,
and move “ negro-heads,’ that the waves of the storm-driven
blue ocean are not to be pictured at an inland fireside.

ADDENDUM TO RE-ISSUE

In April 1910, the Governor, whose ill-health had been aggravated by
the hardships experienced during the cyclone, sailed for England. He
was not fated to return. On July 7, 1910, he died in the island from
which some three hundred years before, William Keeling had set out on
his voyages of discovery. Keeling lies buried at Carisbrook, while hard
by in Bonchurch churchyard George Clunies-Ross has found his last
resting-place.

CORRICENDA
On page 39, fur “ anixety ’
eg Olly on AESISCILIL >? Gaara % wessall”
» > 145, ., ‘*Sherrard Osborne” read “* Sherard Osborn, ’
, 199, ,, “neigbourhood” 7cad “neighbourhood.”
Index. ‘Andrew Ross,” p. 286, ead “Dr, C. W. Andrews.”

read “anxiety.”

(a ee tenner =

4,

4
ee

oe a ¢
ENE Se I SS A NY a EES RI

maby

a
ie a

4

spe

us Bs Pees. es

LS CAN Hoy Tore Tey rorelan- ~~
Ayers. SPadang’, BR Simiant e aay B, “I ak
aon P Te; osileda 5 Rae eo ee ey Point Bon
SeBeroo Aunewi AOS ONY FTanghong M22
M: R r ¥ id tapas B. CM ni ar aanib Mapa Wid
wibo Ss 5 MACE S
ant Jennys 8° q.82_Indrapoor I: ae a pw S ee Rebus,
ergo Bringen® R ; 4 % Zz fis
Seaflow Ks i * ‘Gi foes Lunrpu py
url nctrapoorF. Us, M Soi T R
Sero™ CIrtbory oS RLakat IK
Masso oar jauban Fant @foco Moco % ’ Rolsitt 4) &
ox kakap ~ 8. Xr,
Si Ti
i: aig “anopo , et
of -Laeg I. Mittoh,
Bergen I.

o Js sit bat ‘get
Mego orTrtestel. Satake ‘

Pringdine eg

£fadang ud
Bend

Oceanic Current
——__—

100° 105°

Lovell Reeve

SerBP,

Paris

= 2

© is CoterteysSi-

wiR. O14 So charvos se syery She
: REL soo 4
re wy 28 Sliver, P¥

2 “Annws
Lit

Sy ~ F Po, a
9 OUP ay ~~
pe gig Enna
hi. 4 “ar :
¥ by:
3 fay
e >

MALAY ARCHIPELAGO

SHOWING
COCOS-KEELING ATOLL

ihe StanfordsS Geog! Estat, London
. Limited. :

105°

FMT TT
Hsin

AMMenpi

RH geaiundi aint

ing me Kwakt 4 E f

Clhuvibor’
apf aul %

a> &

ZG, Woe

| Doge

Point Bin 9's ‘
Tajooorf ie
Dany ! Ss
S

S
@ €

Any

al
XR Ls
ltt

%& Palembang 2
a asf)

yt
Baka,
Uttoh

“South Sh.

S,
XB
2,

Lice
Gsulombo

=f !
Vy Looting RS

ium Afatany |

tm

Oceanic Current
ee

Lovell Reevek Co.. Limited.

MALAY ARCHIPELAGO

SHOWING

COCOS-KEELING ATOLL

Stantords Geog! Estab, London

8 droitcs

Dee? Jan? & Feb?

PART I
THE HISTORY OF THE ATOLL

CHAPTER I

THE DISCOVERY AND EARLY HISTORY OF
THE ISLANDS

From what has been said by way of introduction to this
account of the atoll, it will be seen that the Cocos-Keeling
Islands are worthy of some notice from the greater world for
a twofold reason: first, for their being what they are—a
perfect example of that most wonderful of geographical forma-
tions, a coral atoll; and the second, that they hold the records
of a history of romance and adventure which rivals any of the
imagined happenings of fiction.

The kingdom of the Cocos-Keeling Islands has a strange
and attractive history—as instructive in its way as is that of
any great nation, or any story that tells of the doings of the
ruler and the ruled.

The drama of history loses nothing by reason of the small-
ness of its stage; and just as all the problems of life—all the
social questions in their most elaborate and acute forms—may
be seen outside a philosopher's front door ; so may the never-
ending story of monarch and rival monarch, of rule and mis-
rule, and of treasons and plotting, which constitutes all history,
be traced in miniature in the early doings of this little colony.
There is always a popular attraction attaching to little king-
doms, and a halo of romance about the man who, greatly daring,
has left the more beaten tracks of life to take upon himself
the rule of other races, even though his people have been

few, and his enterprise of but little import among the nations.
1 A

2 CORAL AND ATOLLS

Many such romances lie buried in obscurity, and from the
early records of the traffickings of Englishmen and Asiaties a
score of wonderful tales might be unfolded; for it is not only
in fiction that the bo’s’n who has gone ashore in foreign parts
has been made a ruler, or the shipwrecked wait become a
mighty monarch.

It is much to be regretted that such thks should ever be
lost to the world, for each tells a story of enterprise and hardi-
hood, and offers a picture of the greatest of the birthrights of
the Englishman—the power that is born in him to rule an
alien race.

From time to time stray paragraphs in newspapers, or
more pretentious articles in magazines or chapters of books,
have told the tale of the “ Kings of Cocos”; and though the
romance of the dynasty has always appealed to those who
penned them, they have been strangely lacking in accuracy.
The true details of the early days of the Clunies-Ross
sovereignty have never been published, and I am fortunate in
having at my disposal the original papers of the founder of the
Settlement. It is my great regret that these documents are
too lengthy to be am ntened in their complete form, for they
make a splendidly told story of adventure, written by the man
who lived their pages through, and they tell in every line the
fine robust ideals of Ross Primus, who risked so much on his
venture at kingship.

The history of the atoll is a thing of to-day, for it begins only
with the advent of the Clunies-Ross dynasty, and what details we
know of the pre-Ross period are obscure and of no great interest.

Only one excuse could warrant the unearthing of all the
early references to the atoll; and that is the possibility of
tracing the past history of the islands, so that some notion
of their geographical changes, and their formation, might be
gleaned. But Cocos- Keeling is not of sufficient account in the
world to possess a well-recorded history, and even were it to
be complete, as we reckon history, it would probably not help
us much in understanding the formation of the atoll, for no
human record can hope to mark more than a day in its long

DISCOVERY AND EARLY HISTORY 38

life-story. I have, therefore, been brief in my references
to the early chapters of the island history, and have only
gathered together such fragments as seem fairly reliable and
sufficiently interesting. Most of the details that are available
have been collected by Dr. Guppy, who published an account
of the islands in 1889; and others le scattered through the
Blue Books of the Straits Settlements Government. From
both of these sources I have taken such odds and ends as help
to fill in the gap between the time of the discovery of the
group, and the arrival of Ross Primus.

There is no doubt that the islands were discovered by the
early English navigators, and there seems most reason to
assign to Captain William Keeling—after whom the northern
atollon is named—the honour of their first finding.

In 1609 William Keeling, who was a captain in the service
of the East India Company, sailed homeward from Bantam,
and on this voyage it is supposed that he sighted the islands,
and it is certain that his course must have lain close in their
neighbourhood. And yet the actual record of their sighting
does not appear to be forthcoming, for no reference was found
to it in the account of his voyages: and no more direct
evidence than the date of the first charting of the islands, and
their name, associates Keeling with their discovery.

In any case it is probable that Keeling only saw the island
that to-day bears his name, and the southern atoll—the Cocos
Islands—were the independent discovery of another English
navigator, and was made soon after Keeling’s finding of the
northern atollon. .

Captain William Keeling was a man of some note—ana
many virtues—and in Carisbrooke Church in the Isle of
Wight a tablet and an inscription bear testimony to his many
merits. Keeling is depicted standing upon the deck of a ship,
and the tablet witnesses that “Here lyeth the body of the
right worthy William Keeling Esquire, Groom of the chamber
to our Soverign Lord King James, General for the Hon. East
India Adventurers, where he was thrice by them employed, and
dying in this Isle, at the age of 42, An, 1619 Sept. 12th, hath

4 CORAL AND ATOLLS

this remembrance heer fixed, by his loving and sorrowful wife
Ann Keeling.” And then, more picturesquely, it goes on:
Fortie and two years in this vessel frail
On the rough seas of life, did Keeling saile
A merchant fortunate, a Captain bould
A courtier gracious, yet alas, not old.
Such wealth, experience, honour, and high praise
Few men in twice as many years or daies.
But what the world admired, he deemed but dross,
For Christ: without Christ, all his gains but losse ;
For him, and his dear love, with merrie cheere,
To the holy land his last course he did steere.
Faith served for sails, the sacred word for card,
Hope was his anchor, glorie his reward :
And thus with gales of grace, by happy venter,
Through straits of death, heaven’s harbour he did enter.

Since the group has been for so long known by the name
of its discoverer, it is to be regretted that nowadays this
name is often dropped, and the group is simply called the
Cocos Islands. Not only does this curtailing of the proper
name of the islands ignore the claims of a worthy English
navigator, but it leads to much actual confusion. For besides
this group, there are others that are known by the name of
Cocos Islands :—the Cocos of treasure fame, the Cocos in the
Andaman group, the Cocos off the West coast of Sumatra, and
other smaller islands so named from their bearing coconut
palms are apt to be confounded with this atoll. So far has
this confusion been carried that mistakes have been made,
even by representatives of the Government, and the atoll has
received an official visit intended for the Cocos Island in the
Andaman group.

The southern islands—the Cocos Islands proper—were
long known by the name of the Triangular Islands, and the
name Cocos was given to them apparently by the Dutch.
The northern atollon was called Keeling’s Island, or occasionally
Killing Island, until Horsburgh, the hydrographer to the East
India Company (and after whom one of the islands is named)
united the two groups of islands on his charts, and called them
the Cocos-Keeling Islands. On many maps of the present day
the atoll is named the Borneo Coral Reefs: a name that was

DISCOVERY AND EARLY HISTORY = 5

given to the islands because Ross Primus took his household gods
to the new home in a ship he named the Borneo: but Ross
Primus himself always referred to the group as “ Keeling’s Isles.”

In the study of the early history of the atoll the most
important information might be expected from a comparison
of the older charts with those made at the present day, for by
this means we might hope to see what the modern tendencies
of the atoll growth may be.

But unfortunately the charts that are available are few,
and the islands are as a rule depicted so roughly that there is
little safety in arguments based on any changes that may seem
to have occurred.

In maps earlier than 1609 Dr. Guppy was unable to find
any reference to the atoll, but in Dudley’s “ Arcano del Mare,”
published in 1647, the islands are charted in their correct
position. The islands are also shown, but not named, in
Blaeu’s appendix to the “ Theatrum Orbis Terrarum ” of Ortelius
of 1631, though in the edition of 1606 they are not indicated.
The most important early record—because of the more accurate
charting—is that in Van Keulen’s “ Zeefakkel” of 1753, where
the islands are depicted by Jan de Marre, the Dutch navigator,
who made his chart in 1729-30. Dr. Guppy reproduces this
map, which is by no means a bad representation of the group,
but I would hesitate to attach a high value to any theories
based upon its truthfulness.

Beyond a few visits from passing navigators, there is nothing
of historic interest to be noted until the year 1825, when
according to Keating (in “Holman’s Voyage”) Captain Le
Cour of the brig Mauritius temporarily occupied the atoll; and
it is said that his name, and those of his crew, were carved on the
coconut palms. This occupation could not have been of long
duration, for on December 6 of the same year Ross Primus
landed and found the islands uninhabited, and in his account
of his settling he says no word of their recent occupation.

With the year 1825 begins the modern history of the
atoll, the arrival of the present line of settlers, and the unfold-
ing of the strange story of the early days of this modern Utopia.

CHAPTER II
THE ORIGIN OF THE CLUNIES-ROSS FAMILY

THE more recent history of the atoll tells thestory of three genera-
tions of the Clunies- Ross family; and it is a record of remarkable
human enterprise carried out in strange ways and in a strange
place, by men of no ordinary stamp of mind or individuality.

For eighty years some member of the Clunies-Ross family
—men standing out among their fellow men—has with the
most wonderful endurance and pluck, and with a singular mix-
ture of persistence and tact, ruled and fathered the people, and
developed his island kingdom.

Although this little history includes but three generations,
and comprises but eighty years, yet it is instructive of history
in general, for much of the early doings of the family are now
mere legend; much is forgotten, and much more is guessed at ;
and many are the conflicting tales that are at present current
concerning the birth of the colony and its younger days.

The history of the Ross family is one of peculiar interest,
and in variously distorted forms, fragments of it have on
several occasions appeared in print. Hach representative of
the Straits Government who has come upon the annual visit to
the islands has given some outlines of the life of the present
proprietor—and no more picturesque figure could tempt the
pen; yet of the early history of the family there is but little
that has been published that can claim any real title to accuracy.

Ross Primus was a remarkable man, and every page of his
story marks him out as one to whom a strange fascination
belongs; as an example of those strong men whose lives are
naturally interwoven with romance, and one to whom out-
landish enterprise was his commonplace, and adventure his
destiny.

6

ORIGIN OF CLUNIES-ROSS FAMILY 7

The grandsire was the prototype of the grandson, and to-day
Ross Tertius stands out a man remarkable among men—a
monarch and a father among his own people and in his own
islands ; a strong man, and an able, in any company, in any
land. George Ross to-day has all the masterful attributes of
John Ross a hundred years ago, has all his ability and his
dexterity, and his inborn power to rule: but to-day George
Ross has his enterprise centred in a quiet and peaceful tropic
isle, whilst a hundred years ago his grandsire fought and lived
in those stirring scenes enacted under the flag of the Old Hast
India Company.

Like many another notable race of men the Ross family
shows in a remarkable degree the phenomenon of inherited
prepotency, and the dominant features, both mental and
physical, of the family type show a most wonderful persistence.

So far as concerns the origin of the race that has pioneered
this strange enterprise, we need go no further back in history
than the year 1715: for the events of that year were
instrumental in shaping the after-destinies of the family, and
without the 15 and its subsequent disasters, the Clunies-Ross
family might still be treading the easy pathway of the Scottish
gentry.

The year of the ill-fated rising of the Scots in defence
of the rights of “ James, son of James II. of England,’ found
Alexander Clunies—called Clunies-Ross by right of his wife—
doing great deeds for the Clan Chattan in Sutherlandshire: but
one short year after, both he and his cause were in sorry straits.

When on February 5, 1716, James sailed from Montrose,
he left a hopeless cause behind him: and when the English
came by sea, nothing was left to the broken clans save to
disperse; and to the broken leaders save to seek their homes
—or their hiding-places.

It was a hiding-place that Alexander sought, and with his
two young sons he went to the Orkneys, to tarry till time
should heal the wound where a cannon-ball had removed
his right leg, and in some measure mend the fortunes of
his cause. But long before his wound was healed, his. cause

8 CORAL AND ATOLLS

was dead: and in the horrid hue and cry that followed, he
was forced to fly to Yell in the Shetland Isles, for even the
Orkneys could not hide him now. In Yell he settled for the
remainder of his days, and in Yell his eldest son James
completed the sad work of breaking the old man’s proud
Jacobite heart by marrying Catherine Plapen—a Norwegian
girl of plebeian descent.

This was the last straw for the old Jacobite, who had lost
his home, his cause, and his estates ; and so he cursed his son,
and taking his claymore, he broke it in two; for the blade
which had go long defended the family honour was useless now
that this honour counted for so little to him who should
defend it next. That the act might be the more complete,
he burnt the records of the clan for whose pride his eldest
son had shown so little respect; and cut off for ever the life
that he had led, and the ideals that he had represented, from
the prideless doings of his wayward son.

James and Catherine remained in the Shetland Islands
and there their eldest son John was born: John afterwards
married Catherine Clunies, who was a cousin of his, and
they also continued to live in the family’s adopted home.
The eldest son of John and Catherine was named George,
and he also married one of the many relatives that were now
inhabiting this isolated home of theirs. His wife was named
Ross, and by her he had a son, John Clunies-Ross, who was
born in August 1786, and who afterwards became the first
“ King of Cocos.”

Since no authentic account of John Clunies-Ross has
ever been published, and since in those accounts that have
appeared there has been but little agreement and but little
truth, it will be best to give the story of his origin and his
aims in his own words. The document from which I take
his autobiography is a petition to the Hon. Sir T. Bladen
Capel, K C.B., Commander-in-Chief of the East Indies; and
it was written from Cocos-Keeling for the purpose of obtain-
ing a better recognition of his settlement at the hands of the
British authorities.

ORIGIN OF CLUNIES-ROSS FAMILY = 9

He mentions that he had petitioned his late Majesty to
recognise “ Keeling’s Isles ” as a part of the British Empire, and
then he goes on to tell his own history in the following words:

“To begin then at the beginning—I am a native of
Zetland, descended maternally from a Norwegian family,
and lineally on the paternal side from one of those ignorant,
and therefore misguided, men who in the last century led
forth many of the tribes of N.W. Scotland to assist with
their swords, hundreds opposed to millions, Royal hereditary
pretension against National Constitutional rights; and were in
consequence mostly ruined, cut off, and expatriated.

“T may almost say that I was born at sea (Aug. 1785) the
walls of my parents’ bedchamber being often wetted by its
spray—most of my waking hours were however spent in and
on its waves from the age of towards three at most to thirteen,
when I proceeded to serve apprenticeship to the Greenland
whale fishery, from whence after making voyages to the Baltic
and to China, etc., I proceeded to the whaling business in the
Southern Sea department, where I was in May 1813 mate
and harpooner of the ship Baroness Longueville of London—
S. Chace master—and when she was put into the port of
Coupang in the Island of Rimer to obtain refreshments and
water before starting for England, the full cargo of spermaceti
having been acquired in the space of sixteen months from
thence, there lay in that anchorage a small Brig called the
Olivia in want of a Commander and being hired and armed as
a dispatch and coasting convoy vessel by the British Java
Government.

“The resident of Coupang applied to Mr. Chace for an
officer to supply that want,and I resolved to accept the offer altho’
I had been what is called ‘ very lucky’ in my vocation, being
chiefly induced to take that resolution by consideration of the
intelligence which had then been first received by us of the
war, commenced by the American republic against Great
Britain, and the consequent risk of capture in a single sailing
and nearly unarmed vessel all the way to St. Helena, where
only convoy was to be had.

10 CORAL AND ATOLLS

“Having received charge of this vessel, I proceeded with
her via Macassa and Beema, to Batavia, where first I saw
the Brig’s owner—and Alex Hare Esq. who was at that time
British ‘commissioner’ alias Miniature Governor General over
the Island of Borneo and ‘ Resident’ alias Governor Particular
of Banjarmassem the chief local British Establishment in that
island.”

This chance meeting with Alexander Hare was a great
turning-point in the life of John Ross, and since their subse-
quent histories were so interwoven, it will be well to trace the
life of Hare up to the time of the meeting in Batavia. As
Ross Primus himself says, “as the subsequent history of this
person became much connected with mine, and occasion has
been made by himself for its previous portion to be also
noticed by me,” it is best to give some account of Hare,
following the narrative of Ross Primus as closely as may be.
Alexander Hare was the eldest of four boys, sons of a wealthy
London watchmaker, and although the watchmaker was a pious
and respected man, his eldest son showed early those signs of
the eccentric degeneracy that marked and marred his later
life. When a young man he went to Portugal, and became a
clerk in Lisbon, and there he continued for some time; his
next move was in the direction of the Hast, which seemed always
to call him, and soon he appears in Calcutta as the agent for
his firm. It was here that the magic of the Hast first took
possession of Hare, and in all his subsequent doings we find
him becoming steadily more Oriental, steadily more eccentric.

From Calcutta he went to Malacca as- an agent and as a
merchant, and in Malacca he gave vent to all his love of
Oriental splendour, and started his slave retinue and his harem.
From now onwards, whenever Hare is mentioned in the
narrative of Ross Primus, it is always in association with the
extraordinary collection of Oriental women who were his
chattels, and who constituted his harem. He himself called
them his “ fiddle faddle, which whether wise or no, I am in
the habit of considering necessary.” Whilst in Calcutta, Hare
met Mr. Rafiles (afterwards Sir Stamford Raffles) that great

ORIGIN OF CLUNIES-ROSS FAMILY 11

and somewhat neglected man, who did so much for English
prestige in the Far Hast. Mr. Raffles was at that time
secretary to the Government of Prince of Wales Island, and
his friendship had a marked influence on the eventful career
of Hare. Not long after Hare’s meeting with Stamford
Raftles, the British expedition against Java was planned ; and
Lord Minto selected him to join the foree—and no wiser
choice could have been made. As a friend of Raffles, Hare
joined the expedition and saw that wonderful achievement
of Englishmen—an achievement whose fruits were after-
wards wasted by Enelish politicians—the British conquest
of Malaya.

It was as a result of joining this expedition that Hare
came to occupy the position in which Ross Primus found him,
for when hostilities had ceased, the Sultan of Banjarmassem
applied to the British Government to have a British resident
attached to his kingdom, and Hare was appointed to the post.
The Sultan gave to his new resident a large grant of land, for
a great area of his dominions had been depopulated by pirates,
and he hoped that a strong British resident, backed by the
weight of British authority, would restore order and prosperity
to his much-harassed people. But Hare was by no means a
strong man, and at the outset he made a very bad step by
choosing for his second in command a very ill-famed person
named Vanderwadl. Vanderwadl was a Dutch adventurer of
a very degraded type who was condemned to death for his
crimes by the great Commander Daendels, and by an
extraordinary coincidence was saved on the eve of his
execution by the capture of Fort Cornelis by the British.
Vanderwadl, saved from the gallows at the hands of his own
countrymen, was chosen by Hare to act as his deputy, and in
this choice there seems to have been little to recommend the
candidate, save that his tastes and his mode of living
coincided with the ideals of Hare.

Good results were not to be expected from the adminis-
trations of two such persons as Hare and Vanderwadl, and
the result was not a happy one. They obtained convicts to

12 CORAL AND ATOLLS

do forced labour and open up the country, and started a
general programme of misrule.

This state of things was still in progress when Ross Primus,
who had been cruising among the islands for two years on
behalf of the Government, met Hare, and was asked by him
to superintend the harbour works of Banjarmassem. Ross
accepted the position, and leaving the Olivia he commenced
his new career by taking command of the Hon. East India Com-
pany’s cruiser brig Mary Anne, a vessel of 300 tons and mount-
ing twelve 12-pounders. He continued in this famous service,
which has numbered in its ranks such men as Clive and Warren
Hastings and to whose deeds we owe the British supremacy in
India, and he left it only on the British cession of Malaya to
the Dutch. Before Malaya was ceded, Alexander Hare went
to Java, and left Banjarmassem in the charge of John Hare, his
younger brother. During the time that Ross Primus had been
engaged in his duties as harbour superintendent he had busied
himself in shipbuilding, and had laid down a ship of 428 tons,
which he had named the Borneo. The ceding of the British
possessions to the Dutch found this his masterpiece still
unfinished, and so, rather than leave the uncompleted work in
the hands of the Dutch, he remained behind for eighteen
months in order that he might in the end gain the fruits of
his labour, and have the satisfaction of seeing his own ship
launched.

The Lorneo jeft the slips an accomplished success, and
Ross sailed away in her in search of new adventures in some
fresh enterprise.

Although John Ross had built the Borneo, the ownership
of her passed largely into the hands of the two Hares, John
Hare having 55-64ths, Alexander Hare 1-64th, and Ross
8-64ths, and Alexander Hare was appointed managing
owner.

It was intended that the Borneo should sail between
England, the Cape, and Malaya, and take a share in the trade
in spices and coffees, and all those Eastern products which in
the early days so easily built up the colossal fortunes of the

ORIGIN OF CLUNIES-ROSS FAMILY 13

Eastern merchants. Ross sailed to Bencoolen and took Hare
aboard, and then steered for the home from which he had so
long been absent. At the Cape, Hare was landed, in order
that he might make arrangements for the trading schemes of
the partners. The first voyage was, however, such an ill-success
that Ross despaired of making any money by the venture, and
since he had for so long been prosperous in the East, he decided
that he would take his wife—an Knglishwoman named Eliza-
beth Dymoke—and his family to some new home.

The marriage of Ross Primus with Elizabeth Dymoke,
like all the doings of his life, had in it a strong element of
romance. It was in London that he met her and he en-
countered his future wife under rather curious circumstances.

The times were those of the activity of the Press Gang, and
it chanced to young Ross that he nearly fell into the hands of
these worthies. But although he had so strong a love of the
sea, he had no mind to set out on a voyage in this fashion,
and so after a brief encounter he took to his heels and bolted
up the street, to avoid being taken by the gang.

The chase became a hot one, and as he was in some danger
of being overtaken, he turned a corner on a sudden and dashed
into the nearest house. It happened to be that of Samuel
Dymoke, and in it the young man found shelter, for Elizabeth,
taking pity on him, hid him and eared for him during the
interval in which he dared not show himself abroad.

Ross fell in love with the girl who had befriended him and
saved him from the Press Gang, and he married her. She
and her children shared his adventures with him and prepared
to start with him for his new land.

His choice of a new home ranged over many strange
places, and in his journal are several interesting reflections
upon the various advantages and disadvantages offered by such
remote spots as Melville Island, the Falklands, Kerguelen, St.
Paul’s, Christmas Island, the Poggies, and Cocos-Keeling. Of
the Poggy Islands he actually took out a lease in 1821, a
“John Christie Esq.” being the other lessee, but the Dutch
possession prevented him from ever taking advantage of it.

14 CORAL AND ATOLLS

He seems to have had a most peculiar taste to guide him
in the choice of his new land: one thing he must have, and
that was the sound of the sea always in his ears, and the cities
of America were therefore not for him: whilst Australia he
decided against, for he felt it an injustice to his children that
they should be reared in “a country that had the taint of
convicts.”

He took his wife and children on the Borneo and returned
to the Cape (May 1825), where Hare was waiting in a ship
called the Hippomenes, commanded by James Ross, his younger
brother.

Both ships cruised off, and it was agreed that they should
meet at the Cocos-Keeling Islands. Ross Primus was the first
to reach the meeting-place (Dec. 6, 1825), and waited for his
brother until the 19th, but as the Hippomenes did not
appear, he sailed away for the Cape again, and found that
Alexander Hare had not long left. He then proceeded to
London, for he felt that, as he had decided to make his home
in Cocos, he had better make his position there somewhat
more assured; and he therefore applied to John Hare—as a
person of some authority—to get Government possession of
the place. Then, thinking that all would be well, he went to
sea again and arrived in Cocos on Feb. 27, 1827; there he
found Hare fully established with his harem, and playing the
role of despot to his heart’s content.

The trading purposes of the Borneo’s voyage were not
yet completed, and so, as Ross Primus says, “I hutted my
family on the isle, and continued my voyage.” Later in the
year he returned, and then began the dual occupation of Hare
and Ross, and all the queer stories of the squabbles of the
rival rulers in their tiny kingdom.

CHAPTER IIl

Wirn dominions that comprise the whole breadth of great
continents, and with subjects taxing man’s powers of computa-
tion, it is an easy thing for monarchs to disagree; it is none
the less so in an atoll with a population of less than two
hundred souls, and an area of a few square miles, when two
rulers attempt to wield their independent sceptres. There
were no methods sanctioned by tradition in the government of
the Islands and trouble arose early, and general disagreement
continued without remission until the dual rule came to an
inglorious end.

The two rival factions had settled in the atoll for very
different reasons; Hare elected to come to Cocos because he
was in some disgrace at home, for he had not satisfactorily
settled his financial dealings with the East India Company—
and he was glad to seek some secluded corner of the globe,
where he could dwell free from all interference from his
creditors ; and free too from the prying eye of civilisation,
that would look askance at his harem and his manner of
living. With Ross Primus it was a shrewd business instinct,
coupled with his love for the sea, that guided him to the
atoll; for he saw the commercial advantages that Cocos-
Keeling afforded.

In those days the Eastern produce, spices, coffee, and
pepper, was sold from Eastern ports with great rapidity; and
as the crops arrived at the coast only twice in the year much
time was wasted for the ships engaged in the trade, for no
reserve cargoes were kept, and a ship would have to wait long
for her load, and take it aboard at the tardy convenience of

the native merchants. Now Ross saw that a man with a
15

16 CORAL AND ATOLLS

thorough knowledge of Eastern produce, and familiar with the
dialects of Malaya, could buy with great cheapness when the
crops were abundant, and could, after storing his cargo, sell it
to homeward-bound merchantmen with great advantage to
himself. Such a storing-place he intended his new home
to be, and he designed to run the Sorneo between the
Malayan islands and Cocos, and make the atoll into a depot
for Eastern produce. It was for this purpose that he wished
the islands to be taken under the protection of the Imperial
Government :—but many disappointments awaited him. He
had left his claims in the hands of John Hare, who was to
represent his cause in London, but Hare played him false,
and no help came from that direction. Ross next made
direct petition to His Majesty King William IV. This
method also failed, and so did his attempts to persuade
the Government of Mauritius to take the islands under its
charge. It was felt that Cocos-Keeling was too small and too
remote for its protection at the hands of the Government to.
be warranted by its importance, and all the reward that Ross
obtained for his repeated petitions was the visit of a British
man-o-war. In February of 1830 Captain Sandilands arrived
at the islands in H.M.S. Comet and held an inquiry and
furnished a report, but no further steps were taken. It is
said that in the meanwhile Hare was in negotiation with the
Dutch authorities, for his purpose was to persuade them to
take over the protection of the atoll. This would seem to be
a strange line of action for Hare to take, for his feeling for
the Dutch in Malaya cannot have been friendly ; and yet a
Dutch gunboat—the Slora under the command of Mynheer
Van de Jagt—arrived in October 1829, and reported on the
condition of the islands, as a result—so it is said—of his
negotiations. Neither Ross nor Hare produced any other
result from their petitions than a visit of inquiry, and it is
not surprising that this was the case, for the entire population
of the atoll at the time only numbered 175 all told. Of
these 175 colonists, 20 were white, and 10 of these were born
in the islands: the remaining 155 were natives of Sumatra,

1825-1854 LZ

Borneo, Celebes, Java, Baly, Sumbawa, Timor, New Guinea, the
Cape, India and China. The extraordinary mixture of races in
the settlers is accounted for, partly because Hare’s women were
recruited from various Eastern nations, and partly from the
fact that the crews of the Borneo and Hippomenes doubtless
contained the usual mixture of Asiatic seamen.

It would seem that the original intention of Ross Primus,
of making his islands a depot for Eastern produce, did not last
long, for in 1837—ten ycars after the original settlement—he
had turned his attention to the natural products of the atoll,
and describes his business as an “oil factory,” and his trade
and communications were mostly carried on with Mauritius.
From this first establishing of the oil factory the whole
industry of the islands has sprung, for it is the coco palm and
its many useful products that form the sole commercial
importance of the group. to-day.

In connection with the coco palm a point of interest arises,
for Dana has said “ there is no known instance that any island
never inhabited has been found supplied with coconut trees.”
It is often said that the coco palms in Cocos-Keeling were
planted by Ross Primus, and it is pointed out that he did not
come to the atoll to work copra or oil, and did not make
mention of this industry until several years of his settlement
were passed. But Darwin, when he visited the atoll early in
1836, describes it as being thickly covered with coco palms,
and says that they constituted the whole prosperity of the
place, the only export being oil from the nut, and the nuts
themselves, which are taken to Singapore and Mauritius.
Further than this, we have seen that when Captain Le Cour
arrived in 1825—-before Ross Primus had even come to the
islands—he carved his name on the palm trunks; and in
1753 Van Keulen described the islands as being “ wooded.”

It is quite a mistake to suppose that the coconut is not
easily planted by the agency of the waves, for unlike many
other seed waifs that are borne in vain to the shores of coral
islands, the coconut will germinate and flourish within reach

of the salt water. Hyerywhere around the lagoon shores the
B

18 CORAL AND ATOLLS

process may be seen any day, and there is no reason to suppose
that any human agency was involved in the first planting of
the coconuts in Cocos-Keeling.

In the staple industry of the island Hare also took his
share, but according to the island legends it was more for the
purpose of keeping his retinue employed than for any definite
commercial aim. It is said that when his men had prepared
a quantity of oil, Hare was in the habit of going by night and
removing the corks from the casks, that he might waste by
darkness what his men had prepared by day. In this way he
ensured a sufficiency of employment for the men of his party,
and so protected himself and his mimic court from interference
at the hands of idlers.

The small faction over which Hare ruled were held by him
as slaves, for though he had been forced to take out certificates
of emancipation, the business had made no difference to the
status of his retainers. He so arranged it that the liberation
paper was served on board the Borneo at Fort Malboro in
1820, and it was he who gave an explanation of the contents
of the papers to the unknowing natives. Afterwards he made
them all sign documents declaring that they and their children
were his absolute property, and it was as his personal chattels
that they came to the islands. It was over the question of
slave workers or free colonists that Ross and Hare first fell to
quarrelling. It was the policy of the pioneer of the Ross
dynasty that all his people should be free men, “free,” as he
said, “except to commit mischief”; and many are the pious
reflections in his journal concerning the slavery methods of
Hare. Naturally enough this state of affairs caused much ill-
feeling, and the quarrel was greatly aggravated by the very
natural desertion of many of Hare’s people to the camp of

toss Primus. It was the boast of the pioneer that none were

weaned away from the rival settlement, but that protection
was freely extended to those whom ill-treatment drove to seek
a refuge with him.

As a matter of fact this neutral attitude did not last long—
at any rate on the part of the natives—for Hare’s people

wf

1825-1854 ie.

were mostly women, and those who came with Ross were for
the most part men. Among the island stories there are to-day
many well-told tales of nocturnal exploits; stories of heroic
swims, of hasty flights, and adventures in which the women
were stolen—not all unwilling—from Hare's camp, and carried
off by the men of the Ross faction.

In this way Ross’ little band of settlers grew larger by
degrees, and his influence steadily increased ; whilst Hare, with
his ever-diminishing subjects, became more aid more absorbed
in the butterfly pleasures of his mimic court. It is said that
Hare took up his residence on the islet that bears the name of
Pulu Bras, from its resemblance in shape to a piled-up heap of
rice. Pulu Bras is to-day a very small island, with an area of
but a few square yards; and in the map of Jan de Marre of
1730 it is also shown as an inconsiderable piece of land. Yet
in 1827 it is said that it was much larger, and was joined to
the island next to it by a spit of sand. Great changes must
have taken place with some rapidity in this portion of the
island ring if the legend of his residence is true; and it is
not altogether unlikely, for Pulu Bras is a pure sand island,
and therefore liable to change with every fancy of the wind
and waves; and some few years ago it is said that parts of his
house were still to be seen upon it.

The original Ross settlement was made upon the Pulu
Selma, but in consequence of the unfriendly relations with the
neighbouring party on Pulu Bras, it was moved on to Pulu
Atas, the most southern—and distant—portion of the island
ring. The Pulu Atas settlement was, however, only a temporary
one, for the silting of the southern portion of the lagoon was
taking place so rapidly that in a very few years the colonists
were forced to move back to Pulu Selma, in order to have an
adequate waterway for their boats. The moving of the Ross
settlement to Pulu Selma and the end of Hare’s brief reign
took place at about the same time, and both were recent
events when Darwin came to the islands in the April of
1836. |

The final disaster that befell Hare was a wholesale

20 CORAL AND ATOLLS

desertion of his followers, headed by one Neh Basir, who was
born in Malacca, the descendant of parents presented to
Alexander Hare by the Rajah of Banjer. Neh Basir afterwards
married a slave girl named Daphne, belonging to Hare, and
he died on June 19, 1893, at the age of 88. He formed one
of the direct links with Hare’s time amongst the Cocos-born
Malays, and his descendants to-day are numerous.

Little by little Hare’s influence diminished as his subjects
gradually dwindled away, and before the end of ten years
he finally retired from the island to die in Singapore. His
attempt to realise his ideal—to be the monarch of a slavish
Eastern court amidst the luxurious setting of a tropical coral
island—had proved a failure. His band of musicians, his
slaves, his courtiers, his harem, and his splendid sovereignty
had slowly but surely slipped from his grasp, and the more
stubborn, more practical rule of Ross Primus had ruined his”
Utopia.

Of Hare’s reign there is little but tradition in the islands
to-day, but many of the people are of course descendants of
his retinue, and there is left at least one fine silver badge,
which—popularly supposed to be the insignia of the keepers
of the harem—is the old livery badge of the Hast India
Company.

When Hare left the islands Ross Primus laid claim to the
whole group, and was styled the King of the Cocos-Keeling
Islands. It is Charles Darwin, the atoll’s most noteworthy
visitor, who gives us our first picture of the settlement—and
it is not a very flattering one.

In April 1836 the Beagle arrived in the lagoon, and
Darwin wrote in his journal an account of the atoll and its
flora and fauna that to-day, after a lapse of 70 years, strikes
one for its wonderful accuracy. Of the settlement he said:
“The Malays are now nominally in a state of freedom, and
certainly are so as far as regards their personal treatment; but
in most other points they are considered as slaves. From
their discontented state, from the repeated removals from islet
to islet, and perhaps also from a little mismanagement, things

1825-1854 21

are not very prosperous.” His notes on the natives are of
peculiar interest, and he says: “The houses of the Malays are
arranged along the shores of the lagoon. The whole place has
rather a desolate aspect, for there were no gardens to show the
signs of care and cultivation. The natives belong to different
islands in the East India Archipelago, but all speak the same
language: we say the inhabitants of Borneo, Celebes, Java and
Sumatra. In colour they resemble the Tahitians, from whom
they do not widely differ in features. Some of the women
show a good deal of the Chinese character. I liked both their
general expression and the sound of voices.” The most
remarkable fact anent Darwin’s notes is that he makes no
mention of the Zulu people who came from the Cape with the
original party. It is likely that they were not numerous, and
yet their prepotent influence has been strongly marked on the
physical stamp of the islanders of to-day.

Not only did he not mention the African element, but
he also overlooked the Papuan people from New Guinea,
and they too have left very visible traces of their woolly-
headed characteristics; probably the Chinese character that
he observed among the women was also the result of real
Chinese blood—for its influence too is well marked to-day.
This is a question of real ethnological interest, for it shows
that the parental Negro and Chinese elements were in 1836
so inconspicuous that Darwin practically overlooked them ;
and yet after 70 years of intermarrying the resulting race
may be said to depart from the typical Malay stock in two
directions: first, and most conspicuous, is the Negro type; and
second, the type that possesses Chinese characters. Some Cocos-
born families possess Negro features to a remarkable degree
and the race is a very fine one; the Chinese type is of
a physique not nearly so good—it is held locally in some
contempt—and resembles that found all over the seaboards
of Malaya where Chinese and Malay intermarry. It may be
said as a general rule that in intermarrying the African
Negro and the Chinese both swamp the physical characters
of the pure Malay—and so too does the Papuan, whose frizzy

22 CORAL AND ATOLLS

hair tends to persist in a race, despite the original straightness
of the Malay locks.

Ross Primus was not in the islands at the time of Darwin’s
visit, and the place had been left in charge of one Leisk, who
was one of the original colonists, and had been mate of
the Borneo. Leisk figures again in the history of the settle-
ment, and not in a very glorious light, but in connection with
Darwin’s visit he is only remarkable as being the author
of those errors in the great naturalist’s account of the
atoll that lent support to the Theory of Subsidence of Coral
Reefs.

It was under Leisk’s guidance that Darwin examined
the atoll, and the information that was imparted to Darwin
was certainly not reliable, and in some cases I believe
that there is no doubt that the inaccuracies of Leisk’s
statements influenced Darwin in his conception of the mode
of formation of the group. In some measure, therefore, Leisk
helped on the cause of the Theory of the Subsidence of
Atolls.

With the departure of Hare, Ross Primus, with the
mile of the whole “atoll” im” his “hands, did) not sanueuncr
fnd his path an easy one. The time of his assuming
the sole authority was when the great southern whaling
industry was still in a very flourishing condition, and when all
the scum of the maritime populations of three continents was
afloat in the Southern Ocean. In those days Cocos-Keeling
was a place of call for whalers, for fresh water and fresh
provisions were always to be had there. The ordeal of a
prolonged cruise in a whaler was an experience that every
sailor shrank from, and when, after months of hardships and
brutality, the crews got ashore in the islands, they generally
rejoined the ship with some of their number lacking. Many
are the island tales of deserted whalers and their doings,
but one runaway deserves special notice, for he became the
subject of a petition to Sir T. B. Capel, and came near to
making white man’s war in the peaceful island settlement.
This man was an American named Joseph Raymond, who

1825-1854 23

deserted from The Trusty, a whaler which had come into the
atollin January 1836. He was a bad character, who had fled
from justice in “Charlestown close by Boston in Massa-
chusetts,” having already served ten years for the murder of
a negro. Raymond, whom Ross in his petition calls “the
foreign anarchistical ringleader,” joined with Leisk in opposing
“duly constituted British authority in territory which if
not British is at all events not American.” The trouble
became serious, for Raymond induced the natives to strike,
enlisted other deserters in his cause, and did much damage
to the settlement by setting fire to the working sheds and
oil factory. It was to have a sure ground for thoroughly
dealing with the rising that Ross Primus applied to the
Admiral. Unfortunately, with the sending of the petition,
the story as written by Ross Primus in his journal ends; but
though no word is said as to the fate of Raymond, we learn
that “hed was duly banished to Batavia, aud passed out of
the island story.

A natural interest attaches to the fate of Raymond, for he
was playing the desperate game of civil war and of attempting
to usurp the authority in a place where a strony and resolute
man made his own laws and dealt his own justice. Which-
ever way the game should chance to go, it was one that had
of necessity to be enacted without the help or interference
of outsiders, for no appeal could be carried to an authority
nearer than Ceylon, where the Admiral was stationed at
Trincomalee.

When Ross first discovered the plot, he says in his journal,
“fortunately my previous suspicions and the gradual manner
in which the whole affair had opened upon me had so far
prepared me that I was not taken by surprise, in which case
I had certainly bestowed upon the villain his quietus on the
instant.” This petition tells us all we know of the affair, and
as Admiral Capel does not seem to have come to the islands
and settled the business, it is to be presumed that Ross Primus
managed to execute justice without the additional weight of
outside approval.

24 CORAL AND ATOLLS

Raymond may have been sent to Batavia, he may have
been deported in another whaler—or he may have received
his “ quietus on the instant”; in any case he disappears from
the history of Cocos-Keeling and John Ross continued his
reign undisturbed.

Besides the whalers and the Beagle other vessels came
to the island and brought visitors to Ross Primus from the
outer world, for in sailing-ship days Cocos was a comparatively
commonly visited place. For many years the Australian
horse ships called at the atoll to take on water and fresh
provisions, and their visits were fairly regular; until, as a
consequence of the dangers of shipwreck on the barrier, the
place was avoided by all boats save those in actual distress.
A vast number of vessels have from time to time gone to
pieces on the barrier, and the history of the atoll is a history
of shipwreck—even the fauna of the place tells a tale of ship-
wrecked castaways.

With the cessation of the journal our knowledge of the
doings of Ross Primus practically ends, for the sailing-ship
visitors have left no record of the condition of the islands or
the story of their ruler.

In 1842 and 1844 the atoll was visited by Dutch
navigators. Mynheer J. J. Duintjer and Mynheer J. W. Retgers
have made reference to the islands, but the petition to Admiral
Capel marks the end of the accurately recorded history of the
reign of the first settler in Cocos.

The remaining years of the life of Ross Primus were
devoted to study, and a great deal of philosophic writing fell
from the pen of this remarkable man. It is unfortunate that
a great part of his writings was destroyed by a disastrous fire
which gutted his house and consumed most of his possessions.

But some fragments remain, and one, a criticism on
Darwin’s essay on Coral Reefs, was published in the year after
his death in the Natuurkundig Tijdschrift voor Nederlundsch
Indie (deel. vii., Batavia, 1855). By a curious mistake this
work is credited in the Royal Society’s Catalogue of Scientific
Papers to Sir J. C. Ross, the Arctic explorer, and, in the

1825-1854 25
article already referred to by Dr. Guppy, to his son Ross
Secundus.

Besides this publication, Ross Primus left to the world a
two-volume treatise on the work of Malthus, a ccpy of which is
now preserved in the Governor’s house on PuluSelma. Itisa
work of great erudition written from an extreme point of view,
but, although it makes a fierce attack upon every premiss
and every argument of Mr. Malthus, it cannot be said to
detract greatly from the patiently drawn conclusions of that
astonishing and suggestive cleric.

In 1854 the pioneer died at the age of 68. Concerning his
death there is told one of the many interesting and romantic
tales that form such a feature of the history which lives for the
most part in the mouths of loyal and admiring narrators.

Upon the morning after his death, the present Governor—
then a little boy of nine—was sent in a boat to fetch the
erew of the schooner to the funeral. Although no boat had
previously put off to the vessel, he found on his arrival that
the captain and his crew were awaiting him, all dressed in
black, with full knowledge of the sad event of the previous
evening.

It seems that at sundown—when the spirit of Ross Primus
fled—a sound had rushed through the air, and to each listen-
ing man upon the schooner had come the wail of the passing
of the spirit of his beloved chief. The schooner lay far out
in the lagoon, but the news came clear and unmistakable,
and the boy rowed back a troop of silent men, who had needed
no telling of the loss which they and the little kingdom had
suffered.

Ross Primus was 41 when he made his island settle-
ment, and he ruled as King of Cocos-Keeling for 27 years.
He is buried in Pulu Selma, and Ross Tertius has honoured
the memory of his grandfather by obtaining from Scotland a
block of granite, which, carved by Cocos-born masons, makes
a fitting monument for the grave of the pioneer.

The work of this strange man is one that is hard to picture
in these matter-of-fact days, and it is difficult to properly

26 CORAL AND ATOLLS

estimate the hardihood of the man who took his family and
“hutted ” them in this remarkably isolated spot. His writings
show him to have been of a philosophic turn of mind, and
quaint reflections come into all the accounts of his extra-
ordinary endeavours; yet the philosopher was a man of hard
business capacity and of daring activity.

He ruled the scum of sailor-men and governed a staff of
coolies mostly recruited from convict gangs; and he carried
on the whole enterprise without the shedding of blood in a place
where there was no appeal to authority save that which he
wielded.

Piety is evident all through his journal, but it is a piety —
that becomes at times rudely interrupted by threats of violent
justice to wrongdoers and speculations upon the successes of
his enterprises. He was a man with a firmly settled belief in
Divine justice; but, situated as he was in an isolated coral
reef seven hundred miles from land, he by no means underrated
his ability to be its successful instrument.

The story of Ross Primus is one that should not lightly be
forgotten, for although his whole efforts were spent in a small
and out-of-the-way corner of the world, he was of the stamp
of the real pioneer and a man in every way remarkable.

CHAPTER IV
THE HISTORY OF THE ATOLL, 1854-1871

WueEn the pioneer was dead, the rule of the islands descended
to his eldest son, John George Clunies-Ross, who had for
many years assisted his father in the work of administration,
and afforded him the leisure to indulge his literary tastes.
John George Clunies-Ross was one of the children who,
with Elizabeth their mother, had been “hutted” on the
islands in 1827; and in the islands he remained through-
out his youth and early manhood. He was born in London,
and was christened at Stepney, and he was but a small
child when his father set about his enterprise of making a
settlement upon the islands which he was afterwards to govern.
His education had been received on the atoll at the hands of
his mother and father, assisted by the Scotch mate Leisk ;
and although the foundations of his schooling would appear to
have been slight, he was a man of remarkable aptitude for
learning, and devoted the greater part of his life to study.

Of Ross Secundus there is no such vivid picture as that
left behind by his father or that presented by his son, and of
his governorship there is no clear or detailed record.

He appears to have been a philosophic man—a_ good
observer and a true lover of nature—and somewhat of an
idealist. In manner he was quiet and very reserved, and
much given to a habit of silence.

In 1859, four years after his taking over the control of the
settlement, he visited his native England, and this one brief
journey home was the only occasion upon which he ever
returned to the larger world; for the rest of his life was spent
in his own atoll, and in expeditions among the islands of the
Malay Archipelago.

2¢

28 CORAL AND ATOLLS

Before he made this one excursion into the old civilisa-
tion which he had left with his childhood, the greatest event in
the history of his governorship had taken place, for on March
31, 1857, H.M.S. Juno arrived in the atoll. It was a great
occasion in the islands; with the firing of a royal salute,
Captain Fremantle proclaimed the atoll to be a part of the
British dominions; and the dream of the Cocos islanders was
realised.

At the same time, the status of Ross Secundus as an
absolute ruler was somewhat altered, for he was declared in
the proclamation to be appointed as the Governor of the
Settlement during Her Majesty’s pleasure: the Dutch flag was
no more to be hoisted upon the trading schooner of the islands,
and Ross Secundus was to be responsible for the good conduct
of the colony. |

It would appear that all the demonstration and ceremony
incidental to this visit of the man-o’-war was in reality the
outcome of a very curious mistake, for it is said that the
Cocos Island that was intended for annexation to the British
dominions was one in the Andaman group, and not the atoll
of Cocos-Keeling at all. However, the visit of the Juno
marks a well-remembered epoch in the history of the settle-
ment, and from that date the group ranks among the British
possessions.

Captain Fremantle did not steam away directly the
business of the proclamation was disposed of, but the Juno
remained at anchor in the atoll for three months, and the
events of this prolonged stay are not likely to be soon for-
gotten. It was a time of gaiety and change for all the
islanders, and one that brought them into contact with much
that was new to them, and it marked an eventful period in
their quiet island routine of life.

A curious incident happened during the stay of the Juno,
as before she steamed from the lagoon in June 1857, a Russian
man-o-war called in at the atoll, and seeing her there, saluted
the British flag and retired, though this was presumably. not
the sole object of her visit.

HISTORY OF THE ATOLL, 1854-1871 29

The departure of the Jwno left Ross Secundus as a
Governor of a colony under the British rule, but the change
of his status was merely a nominal one, for the colony was so
remote from the sovereign power, and so absolutely his own,
that the Juno virtually left him as she found him, the sole
appeal of law and order.

Ross Secundus now set about developing the island in-
dustries and carrying on the work that had been begun by
his father, and this he did with great success. The islands
flourished and the plantations of coconut palms increased,
and everything in the colony appeared to be on the high road
to prosperity. In 1862, however, one of those periodic re-
verses that have been the lot of the settlement smote the
islands, and a terrific cyclone wrecked the flourishing industries
and laid the islands waste.

The whole disaster was the work of a few hours, and yet
in this short time homes were demolished, working sheds were
wrecked, and the coconut palms, which were the sole source
of the island’s wealth, were torn up by the roots or snapped
like matchwood by the violence of the storm. The happy
dream of prosperity was turned into an actuality of hardship
and want, and all the philosophic patience which was the great
characteristic of the life of Ross Secundus was needed to again
start the building up of what was left of the little colony.
He recalled his eldest son from Scotland, and with his assist-
ance set out upon a vigorous endeavour to put to rights the
havoe wrought by the cyclone.

Soon after this event the colony was again visited by a
representative of the sovereign power, for in 1864 H.MS.
Serpent, a surveying ship, called at the islands. A survey of
the group was conducted, and the affairs of the island were
inquired into, and the damage done by the cyclone noted.

From the time of the visit of the Serpent, the history of
the rule of Ross Secundus appears to have been without great
incident, and nothing of importance is remembered to-day as
marking out the events of this period. His eldest son—the
present Governor—was working with him in the islands, and a

30 CORAL AND ATOLLS

great part of the administration of the settlement passed into
the hands of this strenuous young man.

During these years of which we have no detailed account,
Ross Secundus was engaged in fostering the industries of the
colony, and in perfecting the code of island laws compiled by
his father.

The legal code, which has been drafted by three generations
of the Clunies-Ross family, is a matter for the contemplation of
the modern jurist: for it has stood the stress of many crises,
it has risen triumphant to meet several curious emergencies,
and has proved equal to all the strains imposed upon it. In
the framing of these laws Ross Secundus took delight, and
with his son, a man of strong and vigorous policy, to act with
him and carry on the work of the islands, he passed in a great
measure from the active administration of the colony into a
position of passive headship. |

As well as a special aptitude in the details of legal enact-
ments, Ross Secundus had a great inclination for the practice
of medicine, and he became the physician of the islands. He
had, of course, no medical training, but he was an ardent
practitioner of such simple methods as were requisite for the
combating of the sicknesses of coral island life, and the
hygienic welfare of his people was his first care. Some extra-
ordinary cures are credited to him, and he enjoyed a very
exalted reputation among the people asa clever doctor and a
marvellous healer.

For all this, and for his patient researches into the natural
history of the islands, he received the native title of Zwan-
pandai, or “ the learned one,” and to this day a cowry shell, in
whose finding he took special delight, is called siput tuan (the
master’s shell) by the natives.

There can be no doubt that his philosophic bent of mind
has caused but scant justice to be done to his character by
those who have from time to time told the island story. He
has been pictured by his biographers as a dreamer, and one to
whom the practical details of life were of so little concern
that he passed his days in idle speculation, to the detriment of

HISTORY OF THE ATOLL, 1854-1871 31

the settlement and the people. But the philosopher was a
man of action too, for no one who dreams only may rule
a settlement of two hundred Malays, many of whom were of
chain-gang blood, in an atoll seven hundred miles from land.
The closing days of the life of Ross Secundus were spent
in the enjoyment of the society of his family, and in the
investigation of the natural resources of his islands.

In 1841 he had married S’pia Dupong, a Malay lady of
Royal Solo blood, and a woman of the most remarkable force
of character.

There are many stories told of her bravery and loyalty, and
she exerted a great and beneficial influence over the people.
She became the chief moral guide in the atoll, all her deeds
echoed the nobility and generosity of her nature, and she
always upheld with proud and unswerving loyalty the traditions
of the old Malay stock. To her nine children were born, and,
though she died many years before her lord, her name is one —
that will long be remembered in the islands, for she was a
good and a brave woman.

In 1872 Ross Secundus contracted Java fever. What-
ever the true nature of that complaint may be, it is one that
runs a fatal course with great rapidity; he succumbed to it
after a brief illness, and was buried in Pulu Selma.

His grave lies near to that of his father, and his memory
as a wise and kindly ruler is dearly cherished by all the
islanders.

CHAPTER V
HISTORY OF THE ATOLL FROM 1871

Or the nine children born to Ross Secundus, seven were sons:
George, Charles, Edwin, Alfred, Alexander, Andrew, and John;
and two were daughters: Isabella and Eliza.

George Clunies-Ross, the eldest child, was born in 1842,
and at the age of 30 he was called upon to take charge of
the islands, and guide the destinies of the colony after the
death of his father, He had not been very many years in the
islands with his father before he was called upon to take sole
charge, for he only left Scotland and his studies in engineering
to come to the settlement and render assistance in repairing
the great destruction wrought in the atoll by the cyclone of
1862.

George Clunies-Ross—Ross Tertius—was educated at Eliza-
beth College in Guernsey, and afterwards he proceeded to
Glasgow, where he took up, with great success, a course of engi-
neering. This advantage of a practical Scottish education he
shared with all his six brothers, for it had been the wise decision
of his father to send all his boys to Edinburgh or to St. Andrews.
The advantage Ross Tertius received, and that has stood him
in such good stead, he has in his turn handed on, and all his
four sons have been to Scotland for their education, Dollar
and Edinburgh having known most of them.

Ross Tertius came to the atoll, a young man of the most
energetic nature and in the most active period of his life, and
very soon it was apparent in the islands that in him the spirit
of Ross Primus, with all his courage and fortitude, was back
again to rule and guide the colony.

From the study of books and machinery he passed to the

study of the ruling of natives, and no better student ever
32

HISTORY OF THE ATOLL FROM 1871 33

set out to learn a subject the right understanding of which is
given to very few.

It was not enough for him that his mind was stored with
knowledge of matters never even imagined by his people; it
was not enough that he was familiar with the construction of
engines of the existence of which they did not dream, but he
early determined that what the natives could do he would do
too, and he would do better than any of them.

It is difficult to speak of the accomplishment of this self-
imposed task without enthusiasm, for on their own ground he
met the natives, he learned all their crafts, and in the per-
formance of them he excelled them all. George Ross, who
came to the atoll an ardent engineer and a student of books,
had soon learned to cast the Jala-net and throw a fish-spear
with greater accuracy than any one of his subjects; and to
this day his pre-eminent skill remains. He learned to know
the native character in all its details, good and evil, and in
addition to his keen natural intuition, and his shrewd judg-
ments of his fellow men, he came to possess a wonderful power
to fathom the devious ways of the Asiatic mind which make
for faithful devotion, or lifelong enmity towards the man who
reads or misreads them.

The knowledge that Ross Tertius gained from the study of
the natives was soon called into requisition, for during the
closing days of the reign of Ross Secundus trouble broke out
among a section of the imported Bantamese coolies.

A series of law-breakings and disturbances culminated in
the particularly brutal murder of a woman at the hands of
the ringleader of the Bantam malcontents. It was a time for
action, and George Ross’ blocd was afire to avenge the
woman’s death, and to restore order and obedience to.
authority. He begged permission from his father to go in
search of the man, that he might execute prompt justice upon
him, but Ross Secundus was a man of more judicial instinct,
and forbade any sudden and violent form of vengeance.
George Ross promised his father that if he went off into the

bush to find the murderer, and to avenge the death of the
©

34 CORAL AND ATOLLS

unfortunate woman, no word would ever come to his ears of
what might happen at the meeting. But this did not satisfy
the father’s conscience, and milder methods were adopted.

Diplomacy did not succeed, and the discontented coolies
assumed a threatening attitude. The climax was reached
when, in an angry meeting outside the Governor's house,
George Ross faced the blood-guilty leader of the malcontents.
The people seemed out of hand, and the ringleader made ene
step towards George Ross. Things were at high tension, and
swift action could alone save the situation and the lives of
those in authority. Ross Tertius acted on the instant and
struck the man to the ground with a cutlass before the eyes
of all the people.

This prompt action produced an instantaneous effect upon
the malcontents, and with the loyal co-operation of the Cocos-
born natives, the gang was broken up, and the wretched
murderer was handed over to the authorities in Java.

This event did not long precede the death of Ross
Secundus, and the command that Ross Tertius had gained
over the natives stood him in good stead when the difficult
duty of taking sole charge devolved upon him. At his father’s
death, George Ross did not tacitly assume the Governorship,
but calling all the people together he spoke to them, discussing
the affairs of the islands and asked them to elect their ruler.
The islanders were not slow to make their choice, and George
Ross was duly installed as the elected Governor of the islands.
He set about his duties with great resolution, and soon, under
his guiding hand, the islands reached a pitch of prosperous
tranquillity that they had not known before.

George Ross had already followed the example of his
father in making himself more qualified to act in every way
as the friend of his subjects by marrying a wife from among
his own people. The lady who became his wife and the
mother of his numerous children was named Inin, and she was
a member of one of the original high-born Malay families that
came to the islands with the pioneer. She was in every way
fitted to be the wife of such a man as he, for she was as

HISTORY OF THE ATOLL FROM 1871 35

brave as he was, and as loyal: and she it was who saved his
life on more than one oceasion. Like S’pia Dupong, she was a
woman of great and good influence over the natives, and nothing
unseemly was done by them, lest she should be offended, and
nothing ill-mannered, lest she should condemn the act.

In this way the influence of Inin was an extremely
fortunate thing for the islands, for though it may be done in
silence, and is apt to find but little place in a chronicle of the
doings of men, the influence of a good woman may do works
as great as those accomplished by the acts of a brave man.

_In every way Inin helped her husband in his task of
making the colony happy and prosperous, and while he ruled
the men with his masterful will, she stood as law for the
women, and ruled them by her example; and she ruled them
well. The industries of the place rapidly improved, and
George Ross and his wife took it upon themselves to teach
their people and to guide them as if they had all been
members of their own family.

The planting of coconut palms was carried out on a
large scale, work was organised, and many new schemes laid
down. Four years saw the whole aspect of the colony
changed; for the Governor introduced modern methods into
the working of his islands, and instructed his people in the use
of European tools and appliances. He imported and set up
machinery, built brick factories for the preparation of coconut
oil, which was then the staple product of the islands, The
prepared oil was sent to Java to be used as lamp-oil, for at
that time all the lighthouses of the Malay Archipelago were
lit up by the oil brought from Cocos-Keeling. Under his
careful management, George Ross had the satisfaction of
seeing his home cease to be a mere jungle-clad ring of islands,
and become a place of ordered and fertile groves of coconut
palms; and at the end of the first four years of his charge,
everything was smiling on him and his colony. But in 1876,
on January 25, the blow fell, and all the efforts of his four
years of patient and loving toil were brought to naught by
the passing of a cyclone. The cyclone of 1876 was one of

36 CORAL AND ATOLLS

the most severe that has ever visited the atoll, and it wrought
extraordinary havoc in the settlement. Coconut palms were
twisted and snapped off, and trundled before the wind like
ninepins, so that the fury of the storm cut large glades through
tbe plantations, often in most capricious fashion. Extra-
ordinary freaks were played by the storm, for whilst it laid
whole areas bare, others, quite close at hand, remained un-
harmed ; and trees were selected as victims for its fury in the
most haphazard manner. Atap roofs were snatched from
houses and carried away, working sheds were laid low, and the
factories so carefully constructed. under the new régime were
ruined.

A portion of iron from the roof of the Governor’s house
was whirled away by the storm, and made a plaything of the
wind, and to-day the scars are to be seen upon a large tree,
deep in the trunk of which it finally came to rest. The
destruction caused by the wind did not account for the whole
of the ruin, for the seas became tremendous, and overrode the
limits of the barrier, carrying before them the wreckage of all
that they could reach.

Life was preserved with difficulty, and, when the storm
was spent, the stricken islanders had but little to comfort
themselves with save what each man stood up in, and the
wreckage of his possessions.

Thus were the fruits of George Ross’ four years of work
dissipated in the brief visitation of the storm, and the
prosperous and smiling settlement was turned to a scene of
desolation and misery. But although the tropical islands are
liable to these sudden catastrophes, there is no place in which
nature is so quick to efface the effects of the damage and to
repair the loss that the few hours of storm can cause.

Neither was George Ross the type of man to despair in the
presence of this reverse, and once again he put all his energies
into the building up of the prosperity of the settlement. So
well did he succeed in this, that when, in 1879, Dr. H. O.
Vorbes visited the atoll he found a happy and contented
colony on the high road to the recovery of its lost fortunes.

HISTORY OF THE ATOLL FROM 1871 37

In 1878 the islands underwent a political change, which,
like the annexation of 1857, made but little difference to
the internal affairs of the settlement. In November 1878 a
proclamation was issued to the effect that the Cocos-Keeling
islands were annexed for the purposes of administration to the
Government of Ceylon, “to prevent any foreign Power step-
ping in and taking possession of them for the purposes of
settlement or for a coaling station.”

This change was the outcome, so it is said, of a report that
Russian agents had been examining the atoll and its neigh-
bourhood. Letters patent were granted to George Clunies-
Ross, and until the visit of H.M.S. Zephyr, eight years after-
wards, the Government of Ceylon remained as the immediate
appeal for the settlers.

In August 1885, H.M.S. Hspoir arrived at the rece
and Mr. KE. W. Birch went ashore to make a thorough
examination of the colony on behalf of the Government of the
Straits Settlements. The outcome of this visit is a very
interesting record of the condition of the settlement at the
time, but Mr. Birch did not meet George Ross, for at that
time he was absent from the islands on the first visit he had
made to Kurope since his arrival in the colony.

This first journey home was one of the most remarkable
achievements of Ross Tertius, and was a piece of hardy enter-
prise, the full reality of which it is very hard to picture.
When to-day we make ocean journeys in powerful, punctual
steamers, it is difficult to realise the daring of George Ross,
who built his own ship, and sailed her half round the globe,
that he might take his children home to be educated.

During the life of his father the vessel had been laid
down, and her design was the outcome of a great amount
of thought and careful attention to detail, but the rate of
her building was of necessity slow. She was a schooner of
one hundred and seventy-eight tons, built in thorough Cocos
fashion, with every minute detail laboured over until there
was no part of her that was not perfect. The building of
boats in Cocos is an object-lesson in these days of hurried

38 CORAL AND ATOLLS

business and scamped detail, and no vessels more thoroughly
seaworthy than these island-built boats were ever launched.

No wood is bent or strained in their making, and every
piece in their shell is cut from the natural curve of long-
seasoned ironwood, for the Cocos islander is content to spend
weeks on the fitting of a part whose making could be done in
an hour by less thorough methods. All the skilled work of
the islanders and all the best material of the islands was put
into the construction of the schooner, and for patient years
they laboured on her.

In 1884 she was ready on the stocks, was christened the
J. G. C\-Ross, and launched as the islanders’ triumph. In
Iebruary of 1884, George Ross sailed from the lagoon.
Aboard the schooner he had put his seven eldest children—
four little boys and three little girls; and taking his brother
Andrew as his officer, and some Coces-born men to work the
ship, he started on his remarkable cruise. The islanders all
turned out to see their Zuan sail away in their own schooner
to visit the outer world, and it was a moment of pride for the
colony ; a moment only equalled when they learned afterwards
that their handiwork had been examined by the Board of
Trade in London and classed as Al at Lloyd’s for eighteen
years.

Throughout the whole length of the journey, George Ross
and his brother kept watch and watch about; and as a
testimony to their seamanship is the fact that, from the day
on which she left the lagoon to the time of her arrival in port,
she did not once have her mainsail lowered, although in her
journey she encountered all sorts of weather. A stay of three
weeks was made at the Cape, and one of a few days at St.
Helena, and in August the J. G. C.-Ross arrived in the
Thames. All hands were well, no mishap had taken place
on the voyage, and George Ross took his seven children
ashore and made arrangements for their education.

In every way it was a remarkable achievement, and the
six months’ navigating and alternate watching must have
proved a severe strain on any man, but when to this was

HISTORY OF THE ATOLL FROM 1871 39

added the care of seven young children, this deed of Ross
Tertius excites the greatest admiration.

Having seen to the requirements of his little family, and
obtained an English captain to assist his brother Andrew in
sailmg the schooner to the atoll again, George Ross pro-
ceeded on his travels, returning to his island home by way ot
America,

The adventures of the J. G. C.-Ross were not ended with
her safe arrival in the atoll, for eight years after her happy
launching she passed mysteriously from the island story.
The tale of her going is a curious one, and is worth recording
as one of those real sea romances that are often more strange
than any of those that find their way into books.

Karly in February of 1892, an Italian barque, bearing
the name Lwiyi Raffo, with a crew of mixed nationalities, put
into the atoll. She was said to be on a journey from Java
to Antwerp, end she arrived at the islands in a leaking con-
dition. Soon after she had cast her anchor, a strong swell
from the N.E. arose, and the maimed and unseaworthy ship
was driven upon the rocks and wrecked.

Her crew, which consisted of eighteen of the very roughest
sort of sailor-men, were thrown upon George Ross’ hands,
and it was not long before they began to give him very serious
trouble. They created disorder in the villages, set a bad
example to the natives, and in every way became such a
nuisance in the place that there was nothing for it but to
send them packing. The Governor placed the J. @. C.-Ross
at their disposal, put aboard, for their safe-conduct, a
Norwegian captain and eight of his best Cocos-born men. On
February 29 they sailed from the lagoon, and before Andrew
Ross, who piloted her out, had left her side, the ruffians
had fallen to quarrelling. Her destination was Batavia, and
her return was looked for with anixety, but from that day to
this no word of her has ever come to the islands. Batavia
never saw her. No mast or spar was ever picked up, nor
was she ever identified afloat in any part of the world.
What happened to her no man has ever told, but in the

40 CORAL AND ATOLLS

islands the belief is that the Cocos men and their captain
were put overboard, and the Italians disguised her and sailed
away. There are features in the case that lend a probability
to this view, for when inquiries were made, it was found that
the ship named Luigi Raffo was quietly lying in Genoa at
the time of her supposed wreck upon the reef of Cocos-
Keeling. It would appear as though the ship in which the
Italians came was a pirated and disguised one, and it would be
no surprise to many people in the atoll to find the J. G. C.-
ftoss still afloat under a new name. But her loss was a
heavy one, and a great blow to George Ross and all his people,
and nine gocd men went with her and left their wives to
mourn.

George Ross did not return to the islands in time to be
present at the visit of H.M.S. Hspoir in 1885, and so the visit
of representatives of the Straits Settlements Government was
carried out in his absence. Nevertheless Mr. Birch amassed
a great deal of valuable information concerning the atoll, and
the report that he furnished to the Government is one of the
best accounts of the conditions of the place that have been put
on record. In August 1886, H.M.S. Zephyr made a visit of
inspection with the object of granting to George Ross new
letters patent, revoking those of September 10, 1878, by which
the islands had been placed under the Government of Ceylon.

On Tuesday, August 24, 1886, the new letters patent
were read to the people by Mr. A. P. Talbot. The ceremonies
of 1875 were repeated, the Union Jack was hoisted, and a
party of British bluejackets gave the royal salute and fired a
feu-de-joie. A proclamation was given to the people and the
Governor of the Straits Settlements was made their overlord,
while a grant-in-fee of the islands was given to George
Clunies-Ross. Since that time the visits of British men-
o’-war have been fairly regular, and each succeeding Commis-
sioner has noted the progress of the settlement and the
beneficence of the Governor’s rule.

The settlement of Cocos-Keeling had now become a
peaceful colony devoted to the quiet development of the

HISTORY OF THE ATOLL FROM 1871 41

natural resources of the atoll; the strenuous times of its early
days were past, and George Ross and his wife became more than
ever the father and mother. of the people rather than their
rulers. George Ross, and his brothers Charles and Andrew,
taught the men useful trades, and the carpenters of the islands
are not easily to be matched among Malay people. Black-
smith’s forges were set up, and kilns for the burning of coral
boulders into lime were made, and soon there was but little
that the island colony could not produce for its own immediate
wants. Good dwelling-houses were erected, and rat-proof
stores were constructed of bricks and mortar so that large
stocks of rice could be kept in the islands, as a reserve, for use
in times of any shortage caused by the delay of the island
schooner. The people were encouraged in manly sports, and
recular tournaments were held, at which throwing the spear
and diving were keenly contested competitions. In these
contests, George Ross far excelled the best of his people. The
love of the Malay for his boats, his knives, and his spears was
fostered, with the result that boat-building has reached a very
high pitch of development, and the men have become the
handiest boatmen and the most skilful spearers of fish to be
met anywhere. Whilst George Ross was doing all this for the
men, his wife was training the women. She took the girls to
her own house and educated them thoroughly in domestic
duties, teaching them ideals of moral conduct and habits of
neatness and thrift.

It is difficult to over-estimate the success that has attended
these efforts, or to exaggerate the degree of contented pros-
perity which reigned in the islands. By 1888 so great a
peace had spread over the settlement that George Ross looked
for other fields for the expansion of his colony.

On several occasions parties had gone from the atoll and
sailed to Christmas Island, which lies some five hundred and
twenty-five miles to the north and east of Cocos-Keeling.
The extraordinary fertility of this uninhabited place, and its
vast flocks of absolutely unsuspecting birds, were sufficient to
tempt any man to make the passage, and it was no uncommon

42 CORAL AND ATOLLS

thing for George Ross to take a party of his men to camp upon
Christmas Island. Expeditions were made to bring home
some of the valuable timber that grew in such abundance on
every mountain-slope of this fine island, and the advantages of
a permanent settlement there were apparent to the Governor.
In 1888 he therefore took his brother Andrew and a party of
Cocos-born men, and making a clearing in the neighbourhood
of Flying Fish Cove, he built houses and work-sheds and laid
the foundations of a colony.

The little band of colonists set to work at once : coffee and
coco palms were planted, the island supply of fresh water was
conserved and wells were made for its storing. It seemed to
George Ross that his ideals were to be realised, and that here
he had a territory which did not know the confines of a few
miles of barrier reef, and could at the most provide a
livelihood for a few hundred active men. Christmas Island
was clothed with virgin forests to the top of its plateau a
thousand feet above the water-line, and it had never been
inhabited or in any way exploited ; it was a field for unlimited
enterprise, and for the expansion of the Cocos-Keeling colony.
George Ross was a happy man; he dreamed of subjects to be
numbered by thousands instead of by hundreds, and a suffi-
ciency of hardy enterprise and pioneer work to shape the man-
hood of his people for many years to come. He at once took
steps to obtain a legal tenure of the island, and in 1889 he
made application to the Straits Settlements Government for
a grant of land there. But he had not reckoned with the
business instincts of men situated nearer to the sources of
information than he himself was in his out-of-the-way corner
of the earth. Specimens of the rocks of the island, collected
by Captain Aldrich of H.M.S. “yeria, and by George Ross
himself, had been submitted to experts at home, and found to
consist of nearly pure phosphate of lime.

The greed of wealth would not allow the world to sanc-
tion an idealist like George Ross to establish a rural Utopia
upon an island, the whole extent of the surface of which could
be shovelled into the holds of ships and sold for high prices.

HISTORY OF THE ATOLL FROM 1871 438

As aresult of the negotiations, the Governor, in his secluded
coral islands, had to give way to the men of business instincts
at home, and in 1891 he became merely a part owner of a
lease of the island. It was the beginning of the end of his
daughter colony, for in 1897 the working of the island was
taken over by the British Christmas Island Phosphate
Company. The plantations of Liberian coffee, of bananas, and
pepper had to go before the advance of machinery and the
spoiler. One thousand Chinese coolies replaced the colony of
Cocos-born Malays, and Sikh policemen kept order in a land
where for a time the opinion of one man had been sufficient
guarantee of good conduct.

Before the inrush of all the turmoil of wealth-getting, the
wonderful fauna has had to give way. The unique rat (Jus
Mucleari) has been utterly destroyed, and the Christmas Island
fruit pigeon (Carpophaga Whartoni) is fast following in its
footsteps.

Christmas Island has become a remarkably successful
commercial undertaking, but as a preserve for an interesting
and isolated fauna, and as an ideal place for a peaceful Malay
settlement, its day has passed. George Ross and his Cocos
men have withdrawn from the island; they have abandoned
for ever the scheme of their plantations and their larger
colony, and centred all their energies on perfecting their home
in Cocos-Keeling.

The Governor has busied himself with designing and
erecting a fine house for himself, which is a monument to
his patience and skill. He has imported building material
from England and, with his own masons and carpenters, he
has moulded upon his own designs a house the perfect con-
struction of which is his pride. He has built turtle ponds and
boat harbours, made good paths, laid down trolly lines, and
generally improved the island of his residence so that it is a
model of neatness and good order.

He has portioned out the work amongst his brothers and
sons, so that each man has his duty, and every detail receives
attention. It would be difficult to find anywhere a better

44 CORAL AND ATOLLS

ordered or more contented community than that of Cocos-
Keeling, and everything about the settlement shows clearly
the imprint of his master hand. Every scheme is his, and
every detail of its execution is the object of his thought, and
the outlet for his energies; and to this day George Ross
remains a man as quick to think, and as ready to act upon the
thought, as in the day upon which he felled the murderer and
won the hearts of his people.

In 1889 Inin his wife died, to the sorrow of the whole
population, and it was not until 6 years after that he married
Ayesha, his present wife. ;

In 1901 the quiet seclusion of his island home was some-
what broken by the advent of the cable, and its necessary staff
of workers, to Pulu Tikus; but the presence of visitors from
the outside world has not in any way upset the peaceful
routine of coral island life. Cable routes run to Pulu Tikus
from Perth in Australia, from Mauritius and Africa, and from
Batavia in Java; and so the news of all the world is known in
this quiet spot. Yet so remote is Cocos and its settlement
that even the busy messages of politics and commerce merely
pass across its coral shores, and leave it still with all its wonted
aloofness from the world and its strivings.

George Ross is still the one man in Cocos, and Cocos and
its men still possess the same detachment from the world that
has been their characteristic for the last 80 years, and it is to
be hoped that for long this state of things may remain.
Coined money is still unknown, for the parchment notes of
George Ross are the sole medium of exchange; crime hardly
exists, and without police or military, perfect order prevails.
To chronicle a tithe of the doings of George Ross is impossible,
for the life of the man has been a succession of stirring scenes,
in every one of which he has been the central figure, and in
every one of which he has played a man’s part. A strong
man and a good one, he has the rare privilege of seeing his
work perfected, and of recognising in every corner of his home
the order and prosperity he has brought about by his life’s”
work, and the peace that he may long be spared to enjoy.

‘SOMIMG ATAd NO NOWIVLIS Hd VEIOTTAL CAL LO MOTVONOST

T ALV Id

CHAPTER VI
NATIVE CUSTOMS

THERE are many customs practised by the natives of Cocos-
Keeling which are of peculiar interest, for, in common with
all Malays, they show in their ceremonies a curious mixture
of practices that are entirely foreign to their professed creed
of Islam.

At any time the Malay is but a lukewarm adherent to
the orthodox religion, and it must be owned that the Cocos
Islander does not come any nearer to the Islamic ideal of
strict observances than do most of his kin.

The doctrines of the Koran are taught to them and they
are professed Mohammedans, but, like all their race, they show
clearly in their customs strong evidence of a pre-Islamic
nature-worship.

They have a densely populated spirit world, the inhabi-
tants of which are apt to intrude themselves into ceremonies
of purely Islamic origin, and at times almost as much
attention is paid to the requirements of the spirits as to the
details prescribed by the orthodox religion. The hour of
twilight is the one in which the denizens of the spirit world
are apt to be abroad; it is a dangerous time for delicate folk
and children, but for the woman who is pregnant it has the
most terrors, and she must not be out of doors unless she
carries a knife in her hand. It is out of deference to the
spirits of animals that neither she nor her husband may kiil
any animal, for if the blood of any creature is spilled at this
time, vengeance will be visited upon the unfortunate offspring.
Concerning the reality of the visitation, there are many strange
stories current in the atoll, and the doctrine of the potency of

maternal impressions is one that always finds a ready credence
45

46 CORAL AND ATOLLS

in Malay communities. Spirits play an active part in all the
ceremonies and observances connected with the birth of a
child, and special precautions are taken to keep evil spirits far
away from new-born babies.

Beneath the pillow upon which the baby lies is hidden a
parang—the ever-ready knife of the Malay—and in the bed
itself is placed a little broom, as a warning to noxious spirits
that they must keep far from the child. The custom of
placing a knife in the child’s bed is very widespread through-
out the world, and it even exists in some parts of England
and Wales.

If, by some visitation of evil influence, an island baby .
chances to be born malformed, the handiwork of the devil is
at once recognised, and steps are taken to undo the mischief
wrought by Satan. The observances practised at the birth of
a malformed child are curious, and concerning their utility—
or the reverse—all argument is vain, for no one may convince
people who have seen and known these things, the workings of
which are the expected outcomes of their recognised cause.
A woman who gives birth to a child that is in any way
deformed does not immediately begin to deplore her hard
fate, or to assume that the harm that is done is beyond repair,
for she recognises that the devil has done the mischief, and
she knows that if she sets properly about the thing she may
make him undo his handiwork. No notice whatever is taken
of the unfortunate child, but it is covered with a cloth and
placed aside; the mother and the midwife discuss any general
topics of conversation and try to pretend that nothing has
happened. All concerned are determined that they will take
no notice of the child while it bears upon its body the imprint
of Satan’s evil visitation. And, as a rule, the trick succeeds,
for when at length the babe is uncovered, the deformity is
gone, and the child is as strong and as whole as though it
had never carried on its frame the manifestations of Satan’s
influence.

I know a little boy whose head is shaped just as are the
heads of other little boys, and yet, when he came into the

NATIVE CUSTOMS 47

world, it was to present to his horrified mother a visage
surmounted by two large protuberances like hen’s eggs—and
indeed eggs had been much in her mind of late. But he was
covered up and ignored, and the mother’s ruse succeeded: for
her next look at her babe revealed only the rounded scatp and
puckered red face common to all new-born babies. Other
instances too are quoted in the islands, and cases of webbed
fingers and clubbed feet seem easily amenable to this
treatment.

Little children die—even on coral islands this cannot be
prevented—and evil spirits are mostly at the bottom of it, and
the knowledge of the influence of the spirits gives origin to a
very curious custom, A woman who has had the misfortune
to lose her baby suspects that the devil has a design upon her
and her offspring—and a continuance of her trials only con-
firms her suspicions. If child after child is born only to die
of infantile complaints, she knows that it has been decreed by
the evil one that she shall not rear her baby, and so when next
a child is born to her she pretends that it is not hers and places
it in the road where some one is sure to see it before long.

The expected happens, for some other woman passing by
sees the bundle, and, knowing the meaning of its presence,
picks it up and carries the baby to her house. She mixes the
child among her own children, and cares for it, and pretends
that it is hers; and in this way the devil loses sight of the
child and. lets it grow up, forgetting that it is the offspring of
a mother upon whom his curse was set.

The same idea underlies the custom which is prevalent in
the islands of changing altogether the name by which a man
has always been known, after he has recovered from a serious
illness. A man who is smitten with a great sickness, and has
been near to death, feels that it would be unlucky to rise
from his bed still bearing the name by which the spirits knew
him when such ill-fortune befell him, and so he takes a new
name when he recovers; and goes among his fellow men
known by a different title from that under which he passed
when he was smitten.

48 CORAL AND ATOLLS

One of the oldest retainers of the Clunies-Ross family—a
fine man of mixed Zulu blood—whom I always knew as Ali,
began life as “ Tredin,” but the “ Tredin ” was dropped, and never
used again, after he had narrowly escaped death from fever.

The recovery from sickness is not, however, the only
occasion upon which a Cocos native indulges in a change of
name, for when the first child is born to a man and woman
they straightway drop their names and take as their own the
name they bestow upon their infant. The man prefixes “ Pah”
to his child’s name and the woman prefixes “ Mah” ; and by
the compounded names they are known henceforth. For
instance, a man named Jenal married a woman named Itam,
but when their child was born and had been named Angas,
the father was ever after known as “ Pah Angas” and the
mother as “ Mah Angas.”

It might be imagined that this custom would create con-
fusion, and lead to a great deal of uncertainty as to an
individual’s real name, but as a matter of fact it works out
quite well in practice, and leads to no inconvenience. More-
over it possesses the advantage of conferring a desirable dignity
on parenthood.

A child that has survived the chances of spirit inter-
ference during its babyhood has no important event to look
forward to until the rite of circumcision is carried out. This
ceremony takes place as a rule at about the age of ten, and the
time of its performance depends upon the boy's proficiency
in reciting from the Koran. Since the Koran is learned in
Arabic, a language foreign to the Malay, the performance is
not a test of great scholarship or great piety, and the recital
consists merely of committing a certain number of verses to
memory. ‘The boy is gaudily dressed for the occasion, and
he is decked out as only brides and bridegrooms are: like
them he has his face painted, and his eyes adorned with a
monstrous pair of black eyebrows. The feasting and cele-
brations are kept up,for three days, and at the end of the
rejoicings the boy is considered a man, who henceforth must
wear trousers. The transition to trousers 1s more abrupt

NATIVE CUSTOMS | 49
for the little Cocos boy than for his western brother, for it
consists in a sudden change from nakedness to the wearing
of the fully developed garments of manhood. He loses too
his happy privilege of bathing in a state of nature, for no
Cocos man or woman ever thinks of entering the water un-
clothed, and no more modest race exists than the dignified
Malay.

The boy now turns out to work with the men, and takes
his part in all the business of the island as a fully grown
member of society.

He is now free to marry, but he generally waits until he
is some years older, for it has been the wise policy of the
Governor to attempt to put a stop to very early marriages
among the people. The youngest bride that I knew of was a
girl named Denning, who married at twelve, and the youngest
bridegroom one ''arie, who was fifteen years old: but notwith-
standing the disfavour in which early marriage is held, Cocos
has had the honour of possessing a grandmother at the early
age of twenty-eight.

The wedding is the grandest of all the island ceremonies.
and the whole festivity is a long-protracted affair, and one in
which much polite etiquette is essential. When two people
fall in love, it is not the custom for the intending bridegroom
to propose to the lady, but all the negotiations are carried out
by his father with due formalities. The betrothal is arranged
by the young man’s father calling upon the parents of the
maiden and offering a present—the Mas kawen—which is
generally of gold, and usually nowadays takes the shape of
English sovereigns. If the match is pleasing to the girl’s
parents the present is accepted, and the pair are considered as
engaged, and they proceed to set about the preparations for
their marriage.

Should one of the parents of the intended bride be dead,
a strange custom is practised: upon the day before that
fixed for the wedding the young couple go to the grave of the
dead parent, and announce the fact that they are betrothed, so

that the spirit of the departed one may be a party to their
D

50 CORAL AND ATOLLS

contract. Upon the day of the wedding the best friend of
the bridegroom gathers all his male friends together and goes
to the bridegroom’s house, and after decking him out in finery
and painting his face a hideous yellow, leads him out to make
an attack upon the bride’s house.

The bridegroom is truly a pitiful sight, for he is crowned with
quaint headgear and adorned with flowers, his face is ghastly
yellow with kuwnyet (turmeric), and is painted with double
eyebrows and a false moustache. But he is the centre of
attraction, and he walks proudly out, sheltered beneath an
umbrella carried by his best man.

There is much firing of guns and shouting, and a warlike
spirit prevails among the young companions of the bridegroom.
Some go ahead and practise mimic sword-play, and all are
gaily dressed and in high spirits, eager to take their places in
the procession as it threads its way towards the bride’s house.
When the house is reached, the best of the fun begins, and
the gun-fire becomes brisker and the noise louder, until at
length the bridegroom mounts the house steps to claim his
bride. But this is not made easy for him; his future mother-
in-law rushes out to repulse his entry, and offers a brief but
stubborn resistance to him and his young men ;—for it is her
part to defend her daughter. .

When she fancies that she has resisted long enough, she
eracefully yields, and then she welcomes her son-in-law, and
leads him to the bride. The bride awaits him in the house,
and,in token of her submission to her future lord, takes water
in which rose petals and sweet-scented flowers are steeped,
and, with due humility, washes his feet. The rejoicing now
reaches its climax, and all who can gain entry into the house
crowd in and join in singing and feasting and general merry-
making of a very orderly and dignified character. During the
singing, the bride and bridegroom sit on the floor with a
pillow placed between them, and they eat, with great solemnity,
from a dish containing rujak, a very pleasant mixture made
of unripe fruit of various kinds compounded with sugar, salt,
vinegar, and chillies, the presence of which at the wedding cere-

NATIVE CUSTOMS > 51

mony is essentially symbolic. The eating of rujas: signifies that
these two young people, who sit sc solemnly on the floor, are
preparing to enter into the bond in which they may taste
together all the sweets and bitters of life; but I have not
heard of any man who, coming upon too much chilli, gained
timely experience from the rujak. When the singing is over
and much incense has been burned, the bride and bridegroom
rise from the floor, and set out side by side from the house
on a journey to the parents of the bridegroom. An umbrella
is held over each of them, and the procession forms for the
return visit. Upon the house steps, the bridegroom’s mother
awaits the couple, and when they have mounted to the
threshold, she takes a long scarf, called a slendang, and ties
them together round the middle, in the sight of all the people.
The happy couple stand face to face, with their noses almost
touching, whilst the bridegroom’s mother produces rice,
boiled and stained with Awnyet, and solemnly feeds them with
a spoon. It is really a curious sight, for the queer expres-
sion that the false eyebrows lend to the pale yellow faces
gives the young couple rather a sadly comic aspect, and the
business of feeding with a spoon two people tied nose to nose
is apt to appear ludicrous to a crude Westerner. The
festivities are now continued in the house of the bridegroom’s
parents, but, in the feasting that follows, the poor young
people take no part, for they sit solemnly aside, with very
demure faces, until the whole business flags and comes to an
end at about four o'clock in the morning.

After this very trying ordeal, they adjourn to their own
house, if they have one, or, if not, to the house of either of
their parents; and they embark upon their career of married
life.

It is easy to see in the procedure of the marriage cere-
mony the remains of the more primitive marriage customs
of the original tribal bands of the Malay stock. The mimic
warfare of the capture of the bride, and the feigned resistance
of her mother, are mere pale ghosts of the practices of the
days in which a man sallied out from his home, and, with the

52 CORAL AND ATOLLS

help of his friends, captured a maiden from the company
of a neighbouring tribe. The bridegroom’s friend was his
assistant in the raid, and to-day the descendant of the brother
raider is seen, in our own weddings, as the best man, though
custom has long since spared him the ordeal of making sword-

Gale

A WrEpDING GRoUP IN COCcOS-KEELING.

The bride and bridegroom are wearing the flower-decked headgear.

play, or firmg blank charges in the direction of the bride’s
mother.

Just as our best man is scen on a glorified scale in these
Cocos wedding ceremonies, so is the custom of sending out
wedding cake to absent friends displayed in its archaic form
at the feasts of the Malays. The food that is to be used at
the wedding feast—or indeed at any other ceremony of general

NATIVE CUSTOMS 53

rejoicing—is first blessed by the Jmam (priest), and then
‘distributed amongst those present, the blessing of the food
being an essential and a religious function. The food that is
blessed is, however, not all consumed at the feast, for por-
tions of it are sent to all the friends and relations who were
unable to be present at the rejoicings, in order that they may
share in the blessings and goodwill to be derived from these
occasions. The blessed food is called drekat, and custom
dictates that the utensils in which the brekat was distri-
buted shall be returned unwashed to the givers of the feast.
The sending of drekat is an important piece of etiquette,
and an extremely pretty custom, and it would seem likely
that our wedding cake, although unblessed, is in reality a
descendant of the brekat.

The other occasions of feasting, in addition to the rites
of circumeision and of marriage, are the Skada bumi, the
Banchahan, and the Slamatan.

Skada bumi is almost an equivalent for the harvest thanks-
giving, and though the Cocos Islander has but little in the
way of crops to be thankful for, he still maintains his S’kada
bumi with much enthusiasm. The whole island joins in this
festivity, for it is one of the greatest events of the Cocos
Islanders’ year, and elaborate preparations are made for it.
When the feast of the Ykada bumt is over, the food that has
been blessed is not all sent out as brekat to the living, but
some is kept reserved for spirits. Little wicker baskets are
prepared, and hung up in the branches of the trees, that the
spirits may come and take the food that has been blessed, and
so be partakers in the feast. The custom is evidently a very
ancient one designed to make offerings to the spirits of the
various crops and of the elements, and the human feasting,
though at the present time the dominant feature of the
ceremony, is probably a secondary thing.

The Slamatan is a feast given as a thanksgiving in honour of
an individual, a prayer for his safe-keeping, or a thanksgiving
for his safe return; the trips made by the Governor to England
are the greatest of the occasions for the SlJamatan to-day.

54 CORAL AND ATOLLS

The Banchahan is a feast for children, and is a very pretty
custom. The children all gather together in the house and
courtyard of the giver of the feast; some come alone, some are
carried by their mothers, and all are happy. The food is
distributed to them, each child bringing a little bowl to take
its share of the curry stuffs and rice, and great merriment
and good-humour prevails on these occasions. When each
child has shared in the feasting and merrymaking, the
grown-up people take bowls of water, and all at once set
about throwing the water over the crowd of children. A
regular fusillade of water is kept up until, in a very short
while, the elders have splashed and chased every laughing and |
happy child out of the compound. Throughout the feast
there is much burning of aromatic substances, and little
smouldering fires are placed beneath the beds of all the
children in the house, and the ashes of these little fires (abu
bunchahan) are carefully preserved. Should there be an
absent child in the family of the giver of the feast, his
portion of the food is put into a little bowl, and placed apart
upon his bed, so that he, though absent, may share in the
festivities of his family.

The ceremonies connected with death and burial are always
of peculiar ethnological interest, and some of the customs of
the Cocos Islanders are worthy of record. In common with
many Hastern races, it is a strict duty among the Malays to
watch by the body, from the moment that life has gone, until
it is finally lowered into the grave. This duty is studiously
carried out in Cocos, and in all the observances connected
with burial an extreme reverence is shown for the body of
the departed spirit. Immediately after death, the corpse is
washed and little plugs of the fluffy natural island cotton
are placed in the ears, and between the fingers. I do
not know why the cotton is placed in between the fingers,
nor do I know the reason for the curious practice of putting
a needle into each pad, when the body is that of a woman
who has died in childbirth. The origin of the custom
seems lost in obscurity, but it is only one of the many

NATIVE CUSTOMS 55
details of special tender regard paid to the bodies of such
women. .

After the washing and the insertion of the cotton-wads, the
body is wrapped seven times round with linen, and all the
jewellery that was worn during life is left upon the corpse and
goes to the grave with it. The actual funeral always takes
place by water, and the island upon which the burials are
made (Pulu Gangsa) is an uninhabited one, used for no other
purpose. The corpse is placed in a wicker coffin, and the
funeral cortége of boats follows it to the burial island, where
the ceremony is conducted by the native priest, who reads
verses from the Koran as the body is lowered by ropes into
the grave. The wicker coffin is not buried with the body, but
the corpse, swathed around with its linen wrappings, lies
stretched upon its back upon the bare coral sand; a board is
placed over the features to protect them from the falling earth,
and the grave is filled in. In the case of a woman, it is the
custom to bury the body with the head resting upon a little
pillow made up of her own hair, for a Cocos woman never
throws away the hair that she combs out at her toilet. There
is a deep-rooted belief in all this: it is exceedingly unlucky
to leave hair lying about, since an enemy may obtain
possession of it and do Jampei, or witchcraft, with it. A
woman therefore always keeps her hair, and upon it she rests
at last in the grave. A wooden tombstone marks each grave,
and it is shaped differently in the case of men and women.
The tombstone which marks a man’s grave is pointed as an
arch at the top, whilst that upon a woman’s grave is square ;
and in neither case is it elaborate. There seems to be no
especial direction in which the grave is made, and the head
is not constantly pointed in the direction of Mecca; but graves
lie at all angles on the island, being made more with regard
to the convenience of the site than to any geographical
orientation. Upon each grave is placed a Kema (clam) shell
filled with fresh water for the use of the spirit, and flowers
are strewn about the grave by relatives of the deceased.
The graves are carefully kept, and visits are paid to them at

56 CORAL AND ATOLLS

stated intervals, as a part of a regular routine of duty.
The period of mourning is determined by a very curious
custom. The wicker coffin is carried, and lowered into the
grave, by means of a piece of stout rope; and after the
burial, the rope is carefully unravelled, and portions of it are
tied about the wrists of all the relatives of the dead person.
This rope bracelet is never removed, but, in course of time, it
rots through and drops off, and then it is known that the time
of sadness is at an end, and the spirit of the departed desires
30 more mourning. Upon the whole it must be said that a
Cocos funeral is a terribly sad and affecting ceremony; the
community is so small that each loss is keenly felt; the
relationship so intimate that all are mourners; and the sad
procession of boats to the island of burial makes the ceremony
a painfully impressive one.

Jip and bas ot
THE CORALS AND THE CORAL
PROBLEM

CHAPTER VII

THE EARLY STAGES OF THE LIFE-HISTORY
OF THE CORALS

THe myriad coral colonies that grow in such luxuriance in
the warm, clear waters of Cocos-Keeling are of course the
most important of all the forms of life found in the atoll,
for of coral the whole place is formed, and no particle of the
dry land is made of anything save the bleached fragments
of dead colonies. Every island is made of boulders of coral
and coral sand; every rock is coral living or dead; and
every portion of the barrier reef is composed of the remains
of colonies torn from their beds below the sea. There are no
stones in coral islands save those brought by man, or carried
floating on the roots of trees, and the most commonplace
of stones have been regarded as being of enormous value
in some coral islands, when chance has cast them on the
beaches.

Every square inch of land in the atoll is coral, and coral
in its living splendour clothes the seaward sides of the islands,
and carpets the lagoon, wherever the conditions for its life are
favourable.

Besides being of paramount importance, the coral beds are
objects of very great beauty; but I think that the pictures
that have been painted of the splendour of living coral are

often over-coloured. The corals of the Cocos-Keeling atoll
bil

58 CORAL AND ATOLLS

are not rainbow-hued, but are, for the most part, ofa yellowish
or greenish colour, with here and there a purple colony, and
here and there a pink one. The “fairy bowers,” so often
Fie. 2 described, exist only in cer-
tain spots where the water is
sufficiently smooth to permit
the more highly branched
forms to flourish. ‘lhe most
abundant type of growth is
in the shape of a large
rounded, solid rock. It is
the clearness of the water
that gives the whole charm
to the beauty of the coral
beds, for the water is of an
Fungia: Uivine SPECIMEN. indescribable blue, and so
clear that the bottom is
plainly to be seen in depths of over 30 feet. Through the
blue water the living colonies are seen as little trees and rocks
scattered about in a con-
fusion of colour and form.
The whole scene is made
alive, by the flickering of
the moving water and the
darting forms of brilliant
fish, In a manner that jis
altogether charming; and
the beauty of the whole
water picture, rather than
the beauty of the corals
themselves, is the chief at-
traction of the coral beds.
Fungia: SAME SPECIMEN DEAD AND
Some of the corals are DRiep.
hard, and their colonies form
massive boulders or branching stony growths (Zoantharia), and
some are soft and fleshy (Alcyonacea), and these spread out
upon the rocks very much after the fashion of the fungi

Ji, Bt.

KARLY LIFE-HISTORY OF CORALS — 59

growing on trees. As only the stony colonies are of any
importance as reef-builders, they alone will be discussed. It
must not be imagined that the hard corals are dead things, or
that a great stony boulder of Porites is merely a mausoleum
upon whose surface life is flourishing. The coral that we are
accustomed to see in its dried and bleached form is only the
skeleton of a colony of zooids, each of which has its separate
parts, although it is only a member of a great compound
body.

All corals do not live in colonies, and many species of
solitary corals are to be found about a reef. The comparison
of a living solitary coral with the skeleton that remains after
it is dead will give the best idea of the true nature of the
specimens called “corals” in museums. The individual coral
zooid is most like a common sea-anemone, and if an anemone
be imagined to develop a calcareous skeleton, this skeleton
would be very much like what remains when a solitary coral
is dried and bleached.

A vast aggregation of these skeletons, all very minute and
all joined together in the form of a boulder, or a branching
erowth, constitutes what we know as “coral.” There are
many problems connected with the production of this “ coral”
and with the life-histories of the corals themselves, but of a
great many of the most important details knowledge is
strangely lacking.

How far all the myriad forms to be found upon a reef are
to be considered as separate species, or only as mere varieties,
is a point still open to debate; and a vast amount of work
remains to be done before the question may be definitely
settled.

The methods of feeding, of growing, and of reproducing—
in fact all the life-functions of the corals—are but very little
understood, for corals are usually studied when dead and
dried in museums. It is for this reason that any work done
actually upon a living reef is likely to help in the solution of
vexed questions, for no real knowledge of corals is likely to
come from the study of their dead remains, when far removed

60 CORAL AND ATOLLS

from the site on which they grew and from the actions of all
those forces which influenced their growth.

Unfortunately it is common to measure the truth of
observations made in the field by the knowledge that has
been already gleaned from the uncertain data of museum

Fie. 4.

Montipora CoLoNy, Livine.

specimens, and until this state of things is changed corals will
continue to be unpopular subjects for investigation with people
who have the opportunities of studying their forms when
alive.

The maze of literature that has been written upon the
description of corals—often in necessary ignorance of any of
the conditions of their life and growth—is devoid of any
general interest, and is not very helpful.

Observations must be made upon the living zooid if we
are to add to the real knowledge of these creatures, for it is
essential that the life-history of the zooids must be carefully

EARLY LIFE-HISTORY OF CORALS 61

followed before we can form any conception of the real signi-
ficance of their structure.

The corals, like many other animals that live sedentary
lives as adults, are actively moving creatures when young.

Fic. 5.

Montipora COLONY, DEAD AND DRIED.

Beyond the meagre chances of distribution of living fragments
by the waves, and the rare transport of colonies on floating logs,
the adult life of a coral growth is of necessity absolutely fixed
in the place where it is settled down. For when a coral larva
becomes attached to a suitable base and lays the foundation
of a colony it has no further hope of translation other than by
chance accidents as the sport of the waves. ‘The power of
active movement is lost for ever with the settling down of
the larva. For this adult inactivity the motile larval coral
compensates: the younger stages are free-swimming creatures

62 CORAL AND ATOLLS

capable of propelling themselves through the water, and of being
carried far and wide by currents, and in this way they are able
to establish new colonies at a distance from the parent growth.

It is well known to the inhabitants of the atoll that the
corals do reproduce themselves by freely floating embryos, and

Fic. 6

Tentacles of the
partly exparded

Zooid. Epithelial layer

clothing the outer
aspect of the colony.

Stomodaeum

Genital
land
situated
in one of the
mesenteciesS

Mesentery-

Tissue in which

caleiFication

is proceeding
(=)

Tissue in which
- calcification 1S

proceedi ng.

A SECTION THROUGH A PORTION OF AN ALCYONARIAN CORAL
TO SHOW THE RELATIONS OF THE ZOOID TO ITS SKELETON.

it is a common thing to hear it said that a particular wreck
or boulder became quickly encrusted with coral growth
because it happened to settle down when the corals were
spawning.

It would hardly be expected that in an environment so
uniform throughout the year’s cycle the corals would have
a definite breeding season, and yet reproductive activity
evidently undergoes periods of waxing and waning. At

EARLY LIFE-HISTORY OF CORALS 63

times it has been noticed that many colonies are ripe and are
actively giving rise to embryos, whilst at other times no
actively reproducing colonies can be collecte d.

The early stages of the young coral will be described but
briefly, and only such details given as are necessary to make
clear the manner of its life,and the manner of its attaining its
adult condition.

The ova and spermaria are developed from the epithe-
lum that covers the mesenteries within the body of the
parent zooid: some differences in their distribution occur in
different species, for in some cases only ova are found and in
some only spermaria. When both elements occur together
they may be developed separately on different mesenteries
within the same zooid, or they may both be present on the
same mesentery; any mesentery may be the site of develop-
ment of ova and spermaria, but the ova always tend to be
developed at the attached radial edge, and the spermaria at
the free central margin. When mature, the reproductive
elements are passed from the central cavity of the zooid along
the stomodeum, and through the mouth out into the water,
the passage being effected by jerking movements of the parent
zooid.

The stage of development, in which the reproductive
elements are liberated, varies considerably, for while some
observers have seen the male cells and female cells extruded
individually in the process of spawning, still it appears to be
a more general rule that viviparous reproduction is adhered
to, and a young embryo is extruded into the water. The
segmentation of the ovum, and the early stages of the
development of the larva, take place, as a rule, within the
internal cavity of the zooid, and the larve are shot out
through the mouth in successive -batches. In Maniciana
areolata, Duerden observed that a dozen or more larve were
ejected at a time, whilst in Porites clavaria, Favia fragum, and
Siderastrea radians, they came out singly or in twos and threes.
As the process of spawning advances, and as more and more
batches are ejected, the stage of development of the extruded

64. CORAL AND ATOLLS

larvee becomes more advanced, and a colony which at a spawning
time started by ejecting ova and sperm cells will give rise, as
the season advances, to successively more perfectly developed
larval forms. The process of development of the larvee within
the zooids takes place fairly uniformly all over a colony, so
that any zooid that may be examined will show the larve, or
reproductive elements, all in about the same stage of maturity.
The actual period over which the extrusion of larvee is spread
may be several days, or even a week or two, and then, after the
occasional appearance of a few stray individuals, the zooids cease
to give forth larve, and relapse again into their normal state
as members of the colony.

Zooids within the span of their lifetime are able to give
rise to more than one crop of germinal elements, for a seasonal
outpouring of the larvee does not terminate a zooid’s repro-—
ductive life. The capacity of a mature colony for reproducing
its species is therefore potentially very great. |

When first launched on an independent career, the young
coral larvee are little bodies from 1 to 3 millimetres in length ;
that is to say, they are slightly longer than the diameter
of an average pin’s head. They vary in shape according to
their species; they may be oval, elongated, pear-shaped, or
spheroidal: moreover, the larva has some power of altering its
form, and, as growth proceeds, its original shape may become
considerably changed.

The larve are active motile creatures, and their activity
begins to display itself, as a rule, immediately after their first
extrusion from the parent zooid: the source of their motile
power is the one so commonly found in lowly swimming
animals—the lashing of cilia.

The larval body is uniformly ciliated; that is to say, it is
entirely covered with filamentous protoplasmic processes which
move in a definite manner, and urge the animal through the
water: the action of the cilia is wavelike, and, as a ripple of
fine movements, spreads down the sides of the larva, much as
the legs of a millipede move in their flowing sequence.

Lt i is by the lashing of these processes that the coral larva

EARLY LIFE-HISTORY OF CORALS 65

moves forwards, and it is by means of the cilia that the
movements of rotation, that some larvee possess, are carried

Fig. 7.

LARVAL ForMs OF CORALS: CONDITION SHORTLY AFTER EXTRUSION.
(After Duerden.)

out. The larva moves along with the site of the future mouth
pointing backwards, and in those larvee whose form is pear-shaped

LARVAL Forms or CORALS.

No. 1. Larva of Favia fragum, Esper., about ten days old, but still
¢ free-swimming.

No. 2. Larva of Favia fragum, Esper., about a week after settling.
(After Duerden.)

it is usually the broad end that leads the way. On extrusion
and after a few moments of quiescence, the larvee watched in

aquaria have been noticed to begin to swim about actively,
E

66 CORAL AND ATOLLS

but, after a day or two of this life, they settle down, and
become fixed to some object by the end that is pointing
forwards as they swim—that is, the end farthest from the site
of the future mouth.

When first liberated they are opaque and solid bodies,

Fic. 9.

Two Youne Growrus or Seriatopora THAT HAVE STARTED
THEIR COLONIES UPON A LIVING OYSTER.

and it is not until some days of their free existence have
been passed that the dimple of the mouth begins to pene-
trate the solid body and hollow it out, and then the larva
swells somewhat and becomes transparent. In colour the
newly liberated larva is usually yellowish. As is so frequently
the case in the adult, this is due to the presence of Zooanthellee,
which are vegetable cells living in the tissues of the coral.

EARLY LIFE-HISTORY OF CORALS 67

These symbiotic alge are crowded most densely about the
region of the mouth, which is in consequence more deeply
coloured, but, strangely enough, after the larve have lived
their independent existence for some time, and have become
possessed of a definite stomodeum and internal cavity, these

Fic. 10.

= fe 6]
a LAL m7} fe ‘
’ e, ee 4

A SECTION THROUGH A COMMENCING COLONY OF
: Cenopsammia nigrescens,

Showing the production of daughter zooids by budding from
the sides of the parent.

Zooanthelle are passed out through the mouth and into the
water.

Although it has been said that the larva becomes fixed
after a few days passed as an active, free-swimming creature,
still, among those that have been watched in aquaria a great
majority display a singular reluctance to settling down: among
Wilson’s larvee of Maniciana many did not settle down until
they had spent three weeks swimming free, and Duerden
came to the conclusion that if fixation did not take place
within the first few days, it did not occur at all, a though the
larvee continued to swim slowly about the vessell, and even

68 CORAL AND ATOLLS

rested temporarily on its sides. These wandering larve,
though they live for several weeks, undergo no development.
They never attain tentacles, they acquire no skeleton, and yet,
if fixation does at length occur, it provides the necessary
stimulus, and growth and development proceed vigorously even
under unfavourable circumstances. A larva may in this
period of non-development be carried far by currents, and |
it is evidently an important factor in the economy of corals
that this stage of developmental inactivity should last long in
some members of the brood, for thus the chances of the spread
of a species become greatly increased. On the other hand,
a larva may never wander from its parent colony; but may
settle on some exposed portion, and start a new colony of its
ow upon its parent growth; or it may settle down in the
company of a fellow, or several of its fellows, and start a
compound colony which is in every respect similar to a growth
that is the outcome of the activity of a single larva.

With the fixation of the larva a new phase commences in
the life of a coral: the forwardly pointing end of the swimming
larva now becomes its point of attachment, and it expands
laterally to form the basal plate: the dimple of the mouth
opens up to form the stomodeum, and the original solid larva
becomes hollowed out by a central cavity. The animal
becomes more complex, but into the precise steps of this
differentiation of parts it is not necessary to enter; tentacles
are budded out, mesenteries are formed in definite order, and
the larva assumes the form of the adult zooid.

Calcium carbonate is laid down as the supporting skeleton,
which in the end becomes “coral” in the popular sense of the
word, and soon the pioneer commences the process of asexual
reproduction by budding or by fission, and so lays the
foundation of the complex body of the coral colony.

CHAPTER VIII
THE GROWTH OF THE CORAL COLONY

THE coral embryo that has settled down, and started its
processes of division and budding, becomes the father of the
colony, and the rate of growth of this colony is a matter of
some importance. Since coral colonies may be regarded as
the bricks of which the whole edifice of an atoll is composed,
it is of interest to know how fast these bricks are made
beneath the sea that will one day tear them from their beds
and heave them up to form dry land.

As a number of observations were made during the fifteen
months of residence on the atoll, and some care was taken in
recording these observations, and in eliminating errors, it will
be as well to give in brief the actual figures arrived at. All
the specimens measured were normal colonies growing in their
natural habitat, and care was taken that the processes of
measurement did not act harmfully upon the colony; in no
case was a coral moved to its site of observation from a
previous habitat. Branching forms were recorded by carefully
measuring the branches at times when wind and tide permitted
of accuracy, and by fixing a little band of copper round the
branches at a distance of 10 centimetres from the distal
extremity. Numerous branches, averaging 20 centimetres in
length, were taken on many different colonies, and the
distance of the band of copper from the tip of the branch was
measured from time to time.

It is only fair to say that the results were extremely
variable: branches would grow fast, and then pause ; and some
that for weeks showed no activity would suddenly enter on a
period of unusually rapid growth.

The reason for these fits and starts of growth is not easy
69

70 CORAL AND ATOLLS

to see, for they take place quite irregularly in different
colonies; but no colony measured in the atoll was exempt
from them. In 100 days, from December 1, 1905, to March
10, 1906, a Madrepora branch grew 3°5 centimetres; and
then from March 10 to July 24, a period of 136 days, no
addition was made to its growth, and the branch though
perfectly normal and healthy showed absolutely no signs of
activity.

In one month, from December 31, 1905, to January 31,
1906, two Porites masses showed no increase in their measure-
ments, and yet by July 31, one had added 9 centimetres to an
original circumference of 180, and the other 13 to an original
circumference of 219.

The period of the greatest activity of the Madrepora was
the period of the most. complete quiescence of the Porites and —
vice versa; and yet all the growths were measured under
identical corditions, all lived in close proximity, and were
subjected to the same influences of tide and season. The
question of the cause of this waxing and waning of coral
growth is interesting, and one that doubtless has great in-
fluences on the building of dry land, for the prolonged
inactivity of any of the more important builders would
necessarily mean an alteration in the rate of land-formation.
It is obvious that any observations that only extend over brief
periods of time cannot be taken as strictly accurate records of
the rate of growth of corals, for the measurements may chance
to be made during a passive phase, or during an active phase
of the colony.

In order to eliminate in some measure the influence of
this variability of growth, the observations were carried on for
as long as possible. The chances of the sea and of storms cut
many experiments short, but no case was used for obtaining
results in which the measured period of growth had not
exceeded 100 days. With the massive forms of growth this
method of measurement could not of course be adopted, and
direct measurements of the maximum circumference had to be
taken at intervals. When a growth of Porites has reached

GROWTH OF THE CORAL COLONY 71

the stage in which its upper surface is flattened and dead, the
varying circumference of the growth represents practically the
entire increase, and where possible such colonies were taken.

Put briefly, and in the form in which they are most readily
comparable, the results were as follows: In 100 days the
average growth of branching forms was 2°74 centimetres ; that
is to say, the distance of the copper band from the tip of the
branch was, on an average, 12°74 centimetres after it had
been in place for 100 days.

In English equivalents the growth 1s 1:09 inches in 100 days,
or roughly 3°7 inches in a year.

In 100 days the massive forms of growth had increased
on an average to the extent of adding 34 of their original
circumference to their growth.

That is to say, a massive coral 37 inches in circumference will
have increased its circumference by 1 inch in 100 days.

This increase in the size of the colony is entirely effected
by the multiplication of the number of the individual zooids
composing the colony, and the multiplication is carried out by
the division and budding of the already existing zooids. The
various methods by which this division is effected differ from
each other considerably ; and it is necessary to touch briefly
upon the chief methods by which a zooid reproduces itself.

It is as well to state at the outset, and thus avoid much
further explanation, that the different methods of division,
though highly characteristic of certain well-marked types of
growth, are not definitely and unalterably fixed ; and I think
it is justifiable to dogmatise, and state that any form of coral
may upon occasion exhibit any form of division. Since the
resulting form of vegetative growth is purely the outcome of
the type of division adopted by the zooid, then it follows that
coral colonies may grow in many vegetative forms; and,
again, these vegetative forms are not definitely fixed, for as
any form of coral may exhibit any form of division, it follows
that any form of coral may also, under different circumstances,
exhibit any form of vegetative growth.

Speaking quite broadly, a colony may grow according to

72 CORAL AND ATOLLS

five different types of vegetative growth. It may grow as
(1) a spherical mass, (2) an encrusting layer, (3) a free plate,
(4) a branching tree-like growth, or (5) a mere amorphous
lump; and though a definite inherent growth tendency is
strongly implanted in the embryo, still the demands of the
environment may call forth any type of vegetative growth.

The growth-forms are purely the results of repeated
divisions of the zooids, and so it will be seen that the relative
value, from a reproductive point of view, of the zooids in a
colony is of the greatest importance. This consideration
brings in its train the division of all the colonial corals into two
groups of normal growth-forms; for all the zooids may take
an equal share in the asexual reproduction, or, again, some
may be of greater importance than others, and the asexual —
reproductive functions may be lodged ina very few individuals ~
only. These two great divisions must be considered separately,
for the rules that may be applied to their respective methods
of growth are widely different.

Taking first the class in which every unit is of equal value,
and going back to the earliest origin of the colony, it is easily seen
that a zooid “ A” settled on a nucleus will divide into zooids
“BRB? and “Cy and “.B? wall turther divideinto <)> samdeaeere
and “C” into “F” and “G,” and so on; each newly divided
individual taking its equal share in future divisions. The
natural outcome of this state of things is that, if the site of
election of growth be a prominence, or, aS is not uncommon, a
small isolated fragment, then the equal divisions will tend to
form a spherical mass. The rapidly growing colony will tend
to surround the nucleus on all sides, and in this manner are
formed those rounded masses of Porites and <Astrwopora com-
monly found lying free in sandy pools, which, when broken
across, are seen to be formed around a central nucleus, which
generally consists of a fragment of dead and altered coral.

It is of course but natural that the true spherical form
cannot long survive in very large colonies, for the zooids
growing below are of necessity killed by pressure. The mass
will therefore become a hemisphere, and continue its growth

GROWTH OF THE CORAL COLONY 73

as a rounded boulder. This boulder-form is a very common
type of vegetative growth among the reef-building corals ; for
if the growth starts as a perfect sphere, it will ultimately
assume this form, and if the colony starts its growth on a

Fig. li.

A GRowTH OF Porites

In which the colony has grown as an approximation to a sphere.

basis that is not an isolated fragment, it will start this mode
of growth from its first beginning.

There are many different ways in which the asexual
reproduction of the zooids is carried out, for the budding and
division may proceed in various fashions. The principal types
of budding vary from each other in the actual site of origin
of the daughter zooid from the parent,and in the degree of
the final separation of the two zooids. These types, I think,
do not merit individual description here, for what was said

74 CORAL AND ATOLLS

previously is true for them all, and each, though highly
characteristic as an inherent growth-form for a certain coral,
may be assumed by many other corals.

One method of division that is important among corals
with zooids of equal reproductive value is the type charac-

Fie. 12.

EXAMPLE OF A CoRAL COLONY GROWING AS AN
ENCRUSTING LAYER.

(Pavonia.)

teristic of the Meandrine. In these corals the complete
separation of the divided zooids is never carried out by the
calcareous partition-walls, and so a more or less linear series of
zooid mouths is formed. Tentacles fringe the margins of this
series, and the resulting skeleton exhibits those well-known
fissure-like markings that have given museum specimens the
name of “ brain-corals.”

Now even such a highly characteristic type of division as

(FYIOdLINOW) HLIMOUD AMIT-ALV Id V SV DNIMOUD ANOTIOOD V AO WTd WV XU

‘TL Wily td

GROWTH OF THE CORAL COLONY 75

that of the MWeandrinw may be assumed by other corals; and
as sports, or even as localised patches in otherwise normal
colonies: other corals may divide actively and yet form no
calcareous: partition-walls,

The meaning of these assumptions of an entirely foreign

ig, 13.

EXAMPLE OF A COLONY GROWING AS A TREE-LIKE GROWTH.

(Madrepora.)

method of division is somewhat difficult to understand, but
when an Astrea exhibits Meandrine fission, it is usually a
sign that the colony is not flourishing and is in an unsuitable
site ; 1t would almost appear that the building of unnecessary
partition-walls was too great an effort for the unhealthy
zoolds.

Again, the thickness of the intervening partition between
two adjacent zooids is subject to endless variation, and in

76 CORAL AND ATOLLS

consequence the surface-pattern and the density of the coral
may vary widely within the limits of a species. The same
variation is to be seen in the level of the site of origin of the
lateral buds: this will have to be referred to later on when
the cause of the variations is considered.

Another feature that is subject to an excess of variation,

figs, lal.

EXAMPLE OF AN AMORPHOUS TYPE OF GROWTH.
(Porites.)

and must therefore be considered, is the amount of raising
from the general surface of each individual corallite. Every
corallite of the colony may be flush with the general surface,
or it may be raised from it in varying degrees, and the degree
of raising gives very characteristic appearances to the colonies.
The degree of elevation of the corallite is no safe criterion for
determining specific rank, for it is a variable factor depending
altogether on the reactions of the coral to its environment.
The portions of the coral body which lie between the actual
corallites are also important in this connection, for many

(FdOdeIdF NM) WAOA AMITALV Id V ONILVWIXOUddV HEMOUD DNIHONVUGT V AO WTdWVXo

TL HLV 1d

GROWTH OF THE CORAL COLONY 77
characters to which undue importance has been attached are
displayed here. The interspaces may be smooth or rough,
they may be sculptured in various fashions, and they may

Fig. 15.

SR,
Wiig

V7

ag? aa
NN An
Ad 4 |
yan
1 ag :

ra

ul

SE ESESSL CAEL SAG

==

|
|
|
i
i
|

ee AED EPL LA
——-$—— (| {+--+ 1

a
~_]

ian

eee

|

LN
ay

ay
is

rot

;

awe

VEGETATIVE REPRODUCTION IN CORALS.

Diagram to illustrate reproduction by equal budding: the
transverse lines show the surface-level at the time when
division took place.

be elevated or depressed; but very great caution must be
observed and the possibilities of variation must be given due
consideration by any one who would assign specific rank to
any of their forms.

78 CORAL AND ATOLLS

Although a very large number of corals have the normal
habit of reproducing equally from all parts of their surface,
still evidences may be seen, at many points of most colonies,

Fic. 16.

VEGETATIVE REPRODUCTION IN CORALS.

a. Type in which the daughter zooids become completely separated.
b. Type in which separation of the daughter zooids is less complete.

that certain groups of zooids are more active than their
fellows.

The special activity of these zooids may be due to purely
local causes ; it may be called forth merely by irritation of a

iG. 1A

a
7 NS

mn 7 ']sfu),

S
2)

RS
<S

VEGETATIVE REPRODUCTION IN CORALS.

a. I'urther stage of incomplete separation of zooids
b. Meandrine type of division.

localised portion of the surface of the colony ; or it may be
a real alteration of vegetative habit. It is an outstanding
fact in all the colonies observed for long intervals that the

GROWTH OF THE CORAL COLONY 79

growth tends to become irregular—at different times different
portions of the colony have sudden phases of active growth,
and these may be due to no observable cause. Again, some
zooids of the colony may have special advantages due to the
environment, and will grow and: divide excessively, in such a
manner that they dominate the growth of the whole colony,
and consequently modify its form. The conditions of the

VEGETATIVE REPRODUCTION
IN CORALS.

VEGETATIVE REPRODUCTION
IN CORALS.

Diagram showing an Astrea
assuming (as a sport) a Mean-
Meandrine type of division: no drine form of division. From
separation of daughter zooids. an actual specimen.

environment may call forth special reproductive activity in
any portion of the growth, and so in a colony, all of whose
units are of equal importance, some part will be found to be
growing onwards, whilst all the remainder of the zooids are
practically at rest. In this way the typical hemisphere of
such colonies may become modified as a creeping plate, an
encrusting layer, a pseudo-branching form, or a mass of mere
irregular nodules.

~ Thus, in those corals whose zooids are naturally of equal
reproductive importance, forms may arise, as the result of
physiological need, that simulate exactly other forms, the re-
productive activities of the zooids of which are normally
unequal.

80 CORAL AND ATOLLS

Turning now to the corals that constitute the second class,
and have some of their units specialised as active agents of
growth, it is at once seen that the possibilities of variation
of normal vegetative habit are greatly increased. All the
elaborate branching forms, plates, and leaf-like growths belong
to this class; and all are evolved by special peculiarities of the
growing point. The zooids that constitute the growing point
may take various forms: they may be arranged as a cluster, as
a creeping edge, or as many varieties of terminal shoots of
branches.

In the first instance, it is necessary to draw very sharp
distinctions between two subdivisions of this group. In
Group 1 come all those forms like Montipora, whose distal
zooids are the newest formed members of the colony; and in
Group 2 are included the Madrepora, whose distal zooid is the
most ancient individual in the whole growth.

In dealing with Group 1 many forms have to be considered,
for when the youngest are the active zooids their growth-cluster
may be very variously disposed, and on its disposition the
resulting vegetative form entirely depends.

When the growing zooids are arranged in linear series, a flat
growth will result, which grows from one of its edges, or from
them all; and in this way an encrusting layer or a free plate
may be formed. Corals that grow with a linear growing
point may settle down on a basis, and spread over it in all
directions, taking an exact impression of every irregularity of
its contour ; and then, reaching the limits of the basis, they
may grow from it as flat plates, spreading out from its margins.
Such corals as grow in this fashion show many curious changes
at their free margins, for although the superior surface of the
encrusting layer can alone produce zooids, still at the free
edge zooids will appear on both superior and inferior surfaces.

In connection with these partial plates a very curious fact
is always demonstrated, for the whole surface structure of the
coral body, and of the corallites, differs above and below; and
this important fact will need further reference.

A coral may start from its first beginning by growing as free

GROWTH OF THE CORAL COLONY | 81

exfoliating plates, and then it may bear zooids on both surfaces
of the plates, or on the upper surface alone. In every case
where zooids are borne both above and below, there is the
same marked difference of structure between the two surfaces.

When the linear growing point grows uniformly upwards,
the resulting growth consists of a series of vertical plates ;
and when the growth takes this form, the structure of both

Fie. 20.

VEGETATIVE REPRODUCTION IN CORALS.

Diagrammatic section to illustrate the mode of growth of Montipora :
the uppermost zooids are the most newly formed.

sides of the plate is identical, and zooids grow from both
surfaces.

A plate-like growth is formed by a uniform and con-
tinuous growing edge; but the linear series of young growing
zooids may not maintain their continuity: the growing edge
may reach a certain size and then divide, and the resulting
growth consequently takes the form of a plate, cleft at its
edges, or of a branching form, all of whose branches are given
off in one plane. This is a highly characteristic form of
growth of one type of Montipora which is abundant in the
islands, and it grades very naturally into the plate-like growths

F

82 CORAL AND ATOLLS

on the one hand, and into the complicated branching forms on
the other.

In the many highly branched forms of vegetative growth,
the growing zooids form a cluster, and this cluster divides as it
grows upwards, with the development of many growing points
at intervals on the parent stem. The cluster may be of
various shapes, and its form determines that of the stem which
results from it. The stem may be rounded or flattened, may
be thin or thick, and it may branch at frequent intervals, or
it may scarcely branch at all, so that straight rods of uniform
thickness may be characteristic of the vegetative growth of
the coral.

Another modification of this method of growth is that in
which practically the whole colony represents the growing
point, and then the entire mass grows upwards as a solid
column ; or the growing point may be confused and irregularly
distributed, and then an irregular, lumpy amorphous mass
results.

Whatever the vegetative growth may be, it is in these
cases the product of a mass of growing zooids, and these zooids
are being perpetually renewed, so that the growing point always
contains the youngest individuals in the colony.

In Group 2, however, this state of things is entirely altered,
for there one zooid, which is situated at the extremity of the
stem, and which I shall call throughout the “ dominant apical
zooid,” constitutes the growing point; and this zooid is the
parent of the entire colony.

The zooid that settles down to establish a Madrepora
colony has the peculiar innate property of perpetual growth
and perpetual youth ; and this original zooid grows up and up,
budding new zooids from its sides, until destruction overtakes
it. As a matter of fact, a Madrepora colony usually starts as
a flat growth which spreads from its edges, but this method of
growth lasts for only a very short time; and all the charac-
teristics of the “ dominant zooid”—-which is here the central
zooid—are even then well marked.

Besides possessing these peculiar physiological distinctions,

GROWTH OF THE CORAL COLONY

83

the “dominant zooid” is marked off from its fellows by a

distinctive anatomical feature, for it is sym-
metrical. Of the many thousands of daughter
zooids budded off from the “ dominant zooid,”
all are not alike; the great bulk are asym-
metrical, and are but little raised from the
general surface of the coral, but here and
there a prominent and symmetrical zooid is
eiven off. Like the “dominant zooid,” these
lateral zooids possess the power of perpetual
growth, and they are the agents in forming the
lateral branches. Under certain conditions
only very few of these prominent lateral zooids
are produced, and then the resulting vegetative
growth consists of long straight stems with
but few side-branches.

The typical form of vegetative growth of
the Madrepora is therefore a branching system,
but many variants of this form are developed
normally and abnormally. Pseudo plate-like
growths are common, and they are formed by
the anastomosis of numerous branches in one
plane. Again, the importance of the “ domi-
nant apical zooid ” is variable, and some types
branch more after the type fashion of
the Montipora, for a whole apical series
of zooids may be symmetrical.

Having now reviewed, in some

Hig. 21.

measure, the various methods of the VEGErAIIVE REPRO-
formation of the vegetative growth of | PUCTION IN CorAts.

colonies, it is necessary to see how far
these forms of growth are to be reckoned
as specific qualities. Here a great diffi-
culty arises, for an enormous amount
of time and careful experiment must

Diagrammatic section to
illustrate the mode of
growth of Madrepora:
the “dominant apical
zooid”’ is the oldest
member.

be devoted to the proving or disproving of each individual

84 CORAL AND ATOLLS

case; and though, after long familiarity with living corals,
one may feel certain that two entirely different vegetative .
forms belong to the same species, yet the conclusive proof
may be lacking.

Collecting very extensive series of variations will, in some
cases no doubt, link up extremely diverse forms, and study-
ing the repair of damaged specimens will serve to make others
clear; but experimental breeding, and the rearing of identical
species in diverse surroundings, must be the final test in most
cases. Now this is a work of extreme difficulty, for an adult
colony, when removed from its own environment and placed
in a different one, almost invariably dies, and the artificial
rearing of corals is very troublesome; my experience of corals
in aquaria is that it is very difficult even to keep them alive.

The only method that is open to every resident in places —
where corals flourish, and the one that I followed in the ©
Cocos-Keeling atoll, is the careful noting of every modification
of colonies the life-surroundings of which differ by reason of
some influence that can be easily recognised. By this I mean
that the corals of deep water and. shallow water should be
compared ; those living in the surf and those living in calm
spots should be noted; and the corals living exposed to
sediment should be Conencted with those living where no
sediment is being deposited.

These are extremes of habitat, and at first the corals of
two entirely different environments will seem to be quite
distinct, but every compromise of conditions will be found in
different spots in an atoll, and with every grade of altered
surroundings it will soon be seen that there is a modification
of coral growth; and the more completely this method of
observation is carried out, the more will the types of extremely
different habitats be found to be linked up by intermediate
forms. It will be best to study the influence of the different
conditions of life-surroundings on the form of vegetative
growth by taking the different possible modifications of
environment in order; but first some general growth
tendencies of all corals must be made clear.

GROWTH OF THE CORAL COLONY © 85

As a rule, coral zooids and coral colonies tend to grow
upwards, and the general form of vegetative growth depends
on this fact.

To this rule there are two noteworthy exceptions in this
atoll, and these corals (Cenopsammia Willeyi and C. nigrescens)

Fig. 22.

COLONY oF Cenopsammia Willeyi
/ y

Growing from the lower surface of a boulder.

generally grow with their zooid mouths turned downwards.
Now I think these exceptions to be not without interest, for
both these corals live in dark places—they prefer, in fact, the
under sides of boulders, and they possess no symbiotic algze in
their tissues. They are coloured respectively red and black ;
and I would put it forward as a speculative idea, that it is the
innate tendency of the symbiotic algee to grow upwards, rather
than any innate property of the corals themselves, that causes
the general growth tendency of the corals.

It is true that the zooids on the lower surfaces of plate-
forms grow downwards: but as a rule I believe those horizontal

86 CORAL AND ATOLLS

plates which bear zooids below are but variant forms of growth
of other types: for such corals as have no growth-form except
that of a flat horizontal plate do not as a rule bear zooids
below.

Some further remarks must be made about plate-like
growths; but here is a convenient place to refer to a rather
strange fact to which Darwin called attention, that vertical
plates grow so that they offer their flat surfaces to the
currents. The same thing is true of growths that are branch-
ing forms, for then the plane of greatest branching is at right
angles to the line of current. At first sight this seems a
strange thing, for plates growing in rough water become so
much more exposed to damage; and yet it is doubtless for the
exposure of a larger surface of zooids to the food-bringing
currents that the plate spreads in this direction. This phe-
nomenon is especially notable in the Jfillepora, the broad
Jaminze or fan-shaped growths of which are always opposed
to the line of the waves, even when the colony happens to
dwell in very rough water.

Two general rules may therefore be laid down which apply
to all the forms of growth that have been described: the first,
that all corals having symbiotic algze tend to grow upwards;
the second, that all tend to offer their greatest surfaces to the
line of currents.

We may therefore assume that every coral embryo that
settles upon a site of election, and starts the foundation of a
colony, has three inherent tendencies: it has its inherited type
form of vegetative growth, and its inclinations to grow up-
wards and to oppose its growth to currents. Now these
tendencies are affected by the nature of the water in which
the embryo settles down; and for nearly every coral there is
a modification of vegetative growth dependent on the en-
vironment: thus most corals have a deep-water form, a smooth-
water form, a rough-water form, and numerous variations »
depending upon the amount of sedimentation that is taking
place in the water of their habitat. A coral that grows in
rough water is obviously exposed to injury; and those people

GROWTH OF THE CORAL COLONY 87

who have stated that the home of election of the corals is the
surf-beaten edge of the barrier have made an error of ob-
servation and of fact. The comparatively lifeless zone of a
coral-reef is that part of the barrier which is exposed to the
maximum force of the surf, with the rising and falling of the
tide.

Such corals as do grow in the almost perpetual crash of
the surf are of a very easily recognised type; and when con-
trasted with the forms of the same species that have lived in
calmer waters, they appear to have very few points in common.
The type of vegetative growth best suited to resist the force
of the waves is of course the rounded or flattened massive
boulder; and the Porites masses are the type to which all such
growths tend to conform.

This rounded form of growth is brought about in several
different ways: in Rane. and other corals, where the equal
reproductive value of all zooids tends to the production of a
sphere, it is of course the normal habit of growth; and so such
corals have their natural home in the rougher water, and
undergo their atypical modifications when developed in other
sites,

All branching forms, when growing in rough water, are
subjected to a change of habit of vegetative growth; and this
change is a very interesting one. It will be pointed out when
the process of repair is dealt with, that injury affects various
forms of corals differently ; and all these rough-water forms
are in reality the result of perpetual injury.

It would be an obvious disadvantage to a coral to adopt a
highly branching form when living on the surf-beaten portion
of the barrier; for the colony would soon be wrecked and
broken up by the waves. There is no Montipora in the atoll
that lives by election in rough water, but occasionally colonies
become exposed to strong currents; and then the perpetual
injury to the upper portions causes the growing clusters to be
broken up and confused; and the resulting growth is an
irregular mass of short, stunted branchlets.

Alontipora in rough water may also take on a creeping

88 CORAL AND ATOLLS

habit of growth, and form encrusting layers on the surface of
dead massive colonies; but they are not corals which are at all
common in any but the calmest water. Their usual home is
the lagoon, and the greatest normal departure from their
favourite habitat is in the current-swept shallow inlets to the
lagoon; and here an endless series of modifications may be

Fig. 23.

Types OF GRowTH OF Montipora.

a, In barrier pools. 6. In deep water. c. In rough water.

collected, ranging from branching forms to mere amorphous
masses and encrusting layers.

The rough-water forms of the Madrepora are highly cha-
racteristic, and all depend on the processes which always occur
in this group when the “dominant apical zooid” is injured.
Instead of a growth that consists of few dominant zooids
situated at the extremities of long branches, a rounded mass
is formed; and it is composed of little groups of symmetrical
zooids surrounded by others that are asymmetrical—each
little group representing a branch in an extremely abbreviated
form.

From these extreme rounded masses, with branches that
are mere bosses on the general surface, every transition form

GROWTH OF THE CORAL COLONY 89

may be collected, up to the highly developed lagoon types,
with branches many feet in length.

Exactly the same results are produced in the Pocillopora,
and here the rough-water type is a flattened growth, with

Fic. 24.

a b Cc

TYPES OF GrowTH OF Madrepora.

a. In barrier pools. 6. In deep water. c. In rough water.

irregular divisions into separate lobes; each lobe representing
a separate branching system.

The Millepora also show well-marked rough-water types ;
and that named MM. verrucosa is the form which lives in the
crash of the surf.

Any coral that chances to establish its colony in rough
water exhibits, therefore, one well-marked characteristic: it
always tends to form a rounded or flattened mass; and this
for obvious mechanical reasons. It is a direct outcome of the
conditions of the environment, and any zooid must conform

90 CORAL AND ATOLLS

to it or perish. Repeated injury to the growing cells is the
determining cause of this method of growth; and though the
fully developed colony is so totally different from the calm-
water form, still the process is easily seen in the making by
means of artificially inflicted injury.

Besides the alteration of the general appearance of the
vegetative growth, the rough-water environment causes other

IEG, BD,

Tyres or GrowrH or Pocillopora.

a. In barrier pools. 6. In deep water. c. In rough water.

changes in the coral; for it is diagnostic of a rough-water
coral that its structure 1s compact and dense, and its corallites
tend to be flush with the general surface of the growth. The
question of the raising of the corallites will have to be dis-
cussed again when the action of sedimentation is gone into;
it must be stated here that though the levelling of the coral-
lites in rough-water types is doubtless partly mechanical, it is
also due to the fact that in rough water sediment does not
tend to be deposited.

In marked contrast to the rough-water types are those
forms of a species which happen to have become fixed in an
environment where the water is more or less calm; and here,

GROWTH OF THE CORAL COLONY 91

as every grade of environment is to be found, every grade of
modification of the colony is represented.

Corals grow in great luxuriance on the wave-stirred outer
slopes of the atoll, but this is a site by no means to be con-

Fic. 26.

A Couony oF Pocillopora

Whose growth is branched in complex fashion. This form is
linked to the less highly branched forms by the reduction of
its smaller branchlets.

founded with the surf-line of the barrier. They grow also in
the numerous pools of the barrier flats, in the inlets to the
lagoon, and in the lagoon itself; and each environment has
its own peculiar conditions.

In the barrier pools every modification of life-surroundings
is to be met with, for these pools are of varied depth, they
are filled with sand, or are of bare rock; and they have a
varying exposure to the wave action of the barrier breakers.

92 CORAL AND ATOLLS

Corals grow luxuriantly in most of them, and they afford the
best field for experimental work. Since their conditions
change with every cycle of the tides, they form the meeting-
ground of nearly all the commoner species, for they afford in
turn most phases of natural environment. Here are found
the highly branched Madreporw and Pocilloporw, and many
Astree. The colonies are distinguished as a rule by having
their corallites raised, for in most pools sand is being deposited.
They are also highly branched and of a bushy form, for they
are exposed to injury by moving fragments, and so branch
formation is stimulated, whilst the great development of indi-
vidual branches is limited.

These rock-pool forms show naturally the graduating
series of types which connects the rough-water forms and the
smooth-water forms; and they show also the intermediate
stages of the development of coral structures which are
intended for the resistance of the action of sediment.

In the smooth-water forms the predominant feature of the
colony is the fragile nature of the growth. Contrasted with the
rough-water forms, these colonies are extremely lightly calcified,
and their branching systems are distinguished by their deli-
cacy. Their branches are long and slender, their structure is
far more porous, and their whole appearance is quite different
from that of the colonies of the same species that chance to
reside in wave-beaten areas.

The smooth-water forms lead the way to those growths of
corals which inhabit the deeper pools of the lagoon, The deep-
water forms are the most fragile of all: their growths are
more attenuated, and their branches are given off at far less
frequent intervals. There is practically no damage inflicted
on the growing points—whether they be growing clusters, or
“dominant apical zooids”—and so lateral branch formation
is never stimulated. Several of the forms of Montipora and
Madrepora which occur in from 8 to 12 fathoms in the lagoon
are mere cylindrical stems, of great length, and with practically
no lateral branches whatever.

Besides the attenuated form and the absence of lateral

GROWTH OF THE CORAL COLONY 93

branching, one other feature distinguishes the comparatively
deep-water forms from those that inhabit the surface waters—
and this is the general absence of pigment throughout every
portion of the colony. Deep-water forms are therefore as
a rule pale or entirely colourless.

It will be seen from these instances that the form of the

: RovuGH-wWATER Type or Hicguity BrancHED Madrepora.

From an actual specimen of Madrepora pulchra taken
from the barrier.

colony varies as the outcome of the influences of the environ-
ment; and every embryo which settles in any habitat has to
comply with the demands of the physical conditions, and
modify its inherited growth tendency—or perish. Totally
different forms are produced in totally different environments,
but these forms must not be regarded as “ species,” for they
are mere variations of vegetative growths in response to the
necessities of the life-surroundings of the colony.

94 CORAL AND ATOLLS

The type of vegetative growth is affected mostly by the
physical conditions of the water in which it lives, but the actual
structure of the coral depends greatly upon the presence, or
absence, of sediment. Some very strange results are pro-
duced by waters in which sediment is held suspended, and
from which it is being deposited. Sediment will alter the
appearance of a coral more strikingly than any other influence.
The deposition of sediment is the greatest agent in causing
coral death: corals are very easily killed by even compara-
tively little sediment, and are profoundly altered by it, if they
are to successfully resist its influence.

The extent of silt formation at the surface of an atoll in
mid-ocean is hard to imagine, and it has certainly not been
appreciated by those experimenters who have attempted to
estimate the age of an atoll by catching the silt in a net as it
passes into the lagoon. Adown the submarine slopes of the
atoll, for a hundred miles east and west, the bottom was found
by the cable soundings to be strewn with fragments of coral
and finely triturated coral-sand, and itis only an uncertain and
inconstant fraction which passes into the lagoon. Silt is one of
the shifting influences of the atoll, and so may visit the coral
colonies for only a brief portion of their lives, and then partial
death and strange repair-growths result. In the lagoon, and
in some portions of the barrier pools, silt is a constant factor,
and here shows to the greatest advantage the modifications
which it is capable of producing in coral growths. Speaking
generally, silt alters the vegetative growth-form of colonies
only in as far as it produces flat-topped rock masses by killing
the uppermost zooids, and causes amorphous and irregular
growths by partly killing the uppermost growing zooids of the
growing points. But in the surface structure of the coral it
produces great and wonderful changes. Its effects are best
studied by comparing the upper and lower surfaces of partial
plates. In these plates, the upper flat surface is alone exposed
to the action of the deposition of sediment, and here the corallites
tend to be small, and to be raised from the general surface,
and the intervening spaces themselves tend to be sculptured

GROWTH OF THE CORAL COLONY 95

and complicated in various ways. Below, the corallites are
larger and are flush with the general surface, and the inter-
vening spaces are flat and plain.

This condition is entirely the result of the attempt of the
uppermost zooids to build a silt-resisting structure.

The corallites are smaller, and are raised from the general
surface in order to minimise the chance of silt dropping in

lan, BS

(A) UPPER SURFACE AND (B) LOWER SURFACE or A PARTIAL
PLATE GROWTH,

To show the differences in general structure.

and choking the zooid. The intcrvening coral body is
variously sculptured and grooved for the carrying off of
sediment which happens to lodge upon the surface of the
colony. Wonderful series of modifications are formed in this
way. A single species will show extreme variations in the size
of its corallites, and in their raising from the general surface,
when specimens collected from different habitats are compared.
The size of the corallite and its projection from the surface
are therefore not true specific features; for corals of identical
species, from sediment-carrying water and from absolutely sedi-
ment-free water, exhibit great modifications of these characters.

96 CORAL AND ATOLLS

The vegetative habit of a coral, as we have seen, is no true
index of its species. Its method of asexual reproduction, the
characters of its corallites and surface structure, and also its
coloration, are equally variable.

Coloration depends on many, and very little understood,
influences. Corals from deeper waters lose their pigment ;

Fig. 29.

ae
CH
Sey)
90. A
@:
@ “@®)
pre SS 4
C£S eS
Diagram oF Type or GrowrTn DIAGRAM OF GROWTH OF
oF Madrepora Madrepora pulchra
When living in water free of When living in a habitat exposed to
sediment. (J/. pulchra.) the action of sediment.

and corals that are struggling hard in adverse circumstances—
corals in fact that are about to die—become highly pigmented.

Corals identical in every other respect, and living side by
side, may be differently coloured. Nothing is more familiar
than the purple, brown, yellow, or greenish Porites masses which
live under exactly the same conditions, as far as can be
determined. Even one colony.may be differently coloured in
different parts. In Pocillopora there is a dimorphism of
coloration, some growths being pink and some pale brown:
the pink is a very beautiful and striking colour, and yet the
corals are identical when dead, and the zooid in both cases is

GROWTH OF THE CORAL COLONY = 97

the same. JMontiporw vary from yellow to olive-green and
brown, and yet the zooid is always of the same sulphur-yellow,
and the coral is always identical in other respects.

The general coloration of the coral body cannot therefore
be regarded as specific, and the coloration of the zooids also
shows a strange inconstancy. Although, as a rule, the zooids

1S Bile

COLOURLESS Stylopora COLONY FROM DEEP WATER
; (8-12 FatTHoms).

When growing in more shallow water its branches are brownish yellow.

of different colonies of identical species—though the colonies
may be very different in appearance—are similar, yet in one
colony the individual zooids may be very variously coloured.
There is one very striking case, which is not uncommon in
the atoll, of an Astrwa, where the zooids on the upper surface
are a fine fluorescent green, at the sides of the growth brown,
and below white. And here it would appear—as also probably

in the deeper water forms—that light had some influence on
G

98 CORAL AND ATOLLS

the production of the pigments. Despite this peculiar varia-
tion of the colour of the zooids in one colony, it remains a
fact that the zooid is the true index of the species. In all the
strange growth-forms and abnormal coloration of Pocillopora
the zooid remains constantly brown; and in Montipora, con-

ING Bo

Cotony oF Pocillopora

Whose general colour may be either a delicate pink or a yellowish
brown. In both cases the zooid is identical and coloured brown.

stantly pale sulphur-yellow. In Stylopora, whether the coral
is the thinly branched and colourless deep form, or the thickly
branched and brownish shallow-water form, the zooid is always
of the same yellow colour.

In considering the vegetative forms of the corals of an
atoll, it must always be remembered that the environment is
not a constant one. Although the coral colony is absolutely
debarred from changing the site of its growth, still the

GROWTH OF THE CORAL COLONY 99

physical conditions of its surroundings are always altering.
The terms rough water and smooth water, deep or shallow,

HiGaas:

THE INFLUENCE OF ENVIRONMENT ON CORAL GROWTH.

Series of young colonies of Pocillopora in their position of growth
upon a floating log.

and sedimenting or non-sedimenting, are therefore only
comparative ; for what to-day is a habitat free from sediment
may in the course of a few weeks become the site of a copious
deposition of silt. The rise and fall of the tide across the

100 CORAL AND ATOLLS

barrier must of necessity cause great changes in the life-
surroundings of the corals in its passage; and so a colony,
found in a calm pool, may for a part of its life be exposed to
violent wave action. This must always be borne in mind, for
the results obtained by careful collecting will not be pure.
Corals may be found of different forms, growing in close
proximity; but far from being evidence that they are different
species, and not mere varieties, they demonstrate the fact that
the physical conditions of their surroundings are inconstant, and
that, for a cycle, each form is, in its turn, the most suitable.

The very inconstancy of the environment is one factor in
showing the plasticity of the corals, for the partial death and
repair caused by changed conditions afford striking evidence of
the wonderful powers of varied building possessed by the zooids.

Experiment with constant physical surroundings must be
the ultimate test of variability, and in this connection a very
interesting case may be quoted. In the lagoon, a large
portion of a tree-trunk was floated, and made fast to an anchor
and chain; the wood was used to float a ship’s moorings, and
remained just two years in the water. When it was removed
in 1906, several colonies of Pocillopora had started growths
upon it, and they had taken up different positions around its
circumference. The colonies growing above were flattened
bosses; those on the sloping sides showed more tendency to
branch; and those below its convexity were delicate branched
forms.

Now the environments of these colonies were very different,
and they were absolutely constant; at all states of the tide,
waves broke upon the upper surface of the log, whilst the sides
were in gently moving unbroken water, and the bottom was in
comparative calm. The growths might be referred to many
so-called species, and they represent many types found in the
atoll, and yet no one may justly doubt that they are identical,
and that their vegetative growth is entirely the outcome of
their differing environment. I believe that this natural ex-
periment indicates the lines along which the real understanding
of the “species” of the corals is to be arrived at.

CHAPTER IX
THE LIFE-PROCESSES OF THE CORAL COLONY

ALTHOUGH corals as adults do not have the power or inde-
pendent motion, but must live and die in the spot where they
originally settled down, still they have the characteristic that
belongs pritmitively to all protoplasm—they are capable of
resenting injury, and of moving their parts in response to
stimuli. If, when in the course of a walk on the barrier, a
mass of coral be found the zooids of which are actively ex-
tended, it is easily seen that a very slight stimulus will cause
them all slowly, but very certainly, to retract. A light brush
of the surface or a gentle touch will cause .a slow response,
and the zooids withdraw themselves over the definite area
affected. Of the solitary corals Mungia furnishes a good
example of resentment of injury, for if a living specimen be
touched, the delicate tissues covering the rays of its skeleton
slowly shrink and become pale, and this condition spreads as a
slow and curious wave. The sensitive tissue of the creature
thins out over the exposed portions and retracts into the
spaces between the rays, so that, from being a delicately
glandular and prettily coloured mass of soft tissues, it becomes
an almost colourless piece of stone.

These movements of parts are the animal's only means of
avoiding injury, and though they afford some protection to the
more delicate parts of the zooid, a danger that threatens the
whole mass of the coral is a danger that the coral cannot
shield itself from. |

There is yet another factor in the question of the repair
and regeneration of corals, and it is the factor that is so
intimately bound up with the problem of the limitless liability

of the corals as a class to become modified and to vary.
101

102 CORAL AND ATOLLS

When a name is given to this factor, and it is stated that the
corals are a plastic group, it is not quite easy to say what that
name should rightly connote; but although the words when
used in their more common sense are not at all apt, | would say

Fig. 34.

Millepora COLONY OF THE TYPE NAMED complanata.

that the corals are an impressionable and responsive class of
animals. They are ready to comply with the demands of their
environment; they are, within narrow limits, resourceful and ~
capable of remarkable compromises between the contending
forces of inherent growth-form and alterations demanded by
changed surroundings.

Judged as we judge the higher animals, the corals are a
class of unstable individuals: we can say definitely that a
young elephant will grow up to be an elephant and no other

LIFE-PROCESSES OF CORAL COLONY 103

beast; but we cannot say that an embryo Jillepora will grow
to be a branching MZ. alcicornis and not a plate-like JZ. com-
planate or M. verrucosa ; we cannot foretell that a young colony
of Pocillopora will certainly be P. brevicornis and not P. nobilis

Fig. 35.

Millepora COLONY OF THE TYPE NAMED alcicornis.

for, depending on the conditions of its surroundings, it might
chance to be either.

This very plasticity shows itself not only within the limits
of a certain species, but in the life of every actinozooid, for
each member of a colony, or each solitary coral, shows in its
life, its growth, and its repair, all those endless conformations —
to the demands of its environment which tend to produce change
throughout the whole world of living things. An actinozooid,
then, as an individual, possesses a birthright that gives it a

104 CORAL AND ATOLLS

maximum power of repair of damage, or regeneration of
lost parts; and in the colonial forms, which are of special
importance in the economy of the class, this power is greatly
intensified.

In the group of solitary corals there is no very great
interest attached to the processes of repair. The individual
animal is at times damaged by injury, and the damage is
repaired by the laying down of new calcified material, causing
an irregularity in the symmetry of the animal. As a common
feature of repair in any living thing, it may be noticed that
the new material laid down tends to be excessive. Few
large Fungie are to be found in which some injury has not
caused the development of a quantity of irregular calcification
where the delicate tissues of the animal have been split over
the sharp edge of one of the rays. Excessive injury leads to
local death, and local death may affect a very large area of a
solitary coral without being necessarily fatal to the whole
animal. The individual has but little power of repairing
a large portion of its surface when once the area is definitely
dead—very often for the reason that parasitic algze and sponges
attack the dead area.

It is not till the colonial forms are reached that the power
of repair possessed by the corals is properly seen. In a colony
composed of myriads of individuals each actively living, each
capable of growth and repair, the very best conditions for
repair are at hand, and this is further aided in many cases
by the peculiar mode of growth of the colony. There are
some inherent characteristics of the lowly animals amongst
which the corals rank zoologically that are very wonderful,
and which are opposed both to the order of things which pre-
vails in the higher classes, and also to the popular conception
of the life-history of the reef-building corals.

I think it is fair to say that the average belief with
regard to the building of corals and coral-reefs is that the
zooids live and grow, flourish and die, and that their dead bodies
form mausoleums on which their progeny found their colony, —
and thus build islands. Now in contradiction to this is the

LIFE-PROCESSES OF CORAL COLONY 105

biological fact that the actinozooid is a living thing which knows
no time of youthful vigour, no waxing to a period of adult
life, no waning to senility—it knows no age—it practically
knows no natural death. There is no building on the dead
bodies of ancestors, no perpetual dying and new birth, A
colony of Madrepora will contradict this popular fallacy at
one glance, for, whatever the age of the growth, the parent
anthozooid flourishes till death or accident overtakes the
whole. Itis a wonderful thing—and one which is not, I think,
generally considered—that the age of some of these indi-
viduals in every colony must be excessive, even reckoned as
we reckon the age of higher animals. When we consider the
very slow rate of growth of some corals, and the great size of
some of the colonies to be seen every day on an atoll reef,
and when we rightly understand their mode of growth, and
recognise that the pioneer organism of the colony is still
flourishing there, we cannot help being struck by the excessive
antiquity of that organism as a living entity. That an apical
zooid of a branching Madrepora colony should be ten years
old seems wonderful, but these individuals are mere juveniles
when some cf the component zooids of massive growths are
considered.

Throughout the prolonged life of these lowly animals the
process of repair is a possibility, and a strange paradox is
presented in some forms, for the most aged member of the
colony shows the greatest activity in all the processes of
renewal and repair.

The coral colony increases in size by the budding-off of new
zooids and the deposition of new calcium carbonate in their
tissues, and just as this is the ordinary mode of growth, so it
is the ordinary mode of repair. The type of repair naturally
tends to follow the type of growth of the injured colony ; and
the various genera of corals might be taken in order and the
details of the repair of injury noted for each genus. But it is
more likely that some idea of the bionomics of the group
will be gathered from the general study of the phenomena
as they affect the life-history of the coral than from a survey

106 CORAL AND ATOLLS

of the manner in which the, by no means satisfactory, classifi-
cation of the types affects the phenomena.

I. Among the colonial forms there is a sympathy of individuals,
so that each member of a colony takes its share in resenting
the iyjury to a part, and by an increased activity tends to
compensate for its loss, or to assist in its reparr.

In connection with this sharing of each individual in the
fortunes of its fellows and of the entire colony, it may be
remembered that it was stated that the whole population of a
colony may suffer shock from an injury inflicted upon only a
small portion of it. Hven the zooids remote from the seat of ©
injury will frequently not re-expand for forty-eight hours after
the injury was inflicted, and this is so even when the injury
is very trivial. |

Now after the receipt of an injury, the effect produced
by this communal sympathy varies in different forms of coral,
for in a colony, as we have seen, all the members may be of
equal importance, or some may be of greater value than others
as producers and directors of growth.

(A) In a coral such as the massive forms of Porites, where
the growth-tendency is to form spherical masses, every living
entity in the whole vast crowd of active members bears an
equal share in building and inreproducing. It is this equality
of all the zooids in the community which produces the charac-
teristic spherical form of the young growth; and the equality
of the zooids, plus the receipt of injury, produces the typical
flat-topped circular rocks into which the old colony generally
shapes itself. When a mass of Porites has attained some size
as a sphere, the zooids which lie below are necessarily stamped
out of existence by the weight of the accumulated mass. It
is not often, of course, that the environment is so ideal that
anything like a perfect sphere is ever formed, but still, in
sheltered pools, many forms of corals may be obtained resting
free on the bottom, with every portion of their surfaces living.

LIFE-PROCESSES OF CORAL COLONY 107

The original nucleus has been covered equally upon all sides,
and the weight of the colony is not sufficient to cause the
death of those zooids which happen to live on the under side.
But as the mass increases in size and weight, death of the
lowest zooids must inevitably occur, and the rest of the surface-
area carrying on the compensating building will cause the
erowth to become dome-shaped. Theoretically, the dome

Fig. 36. Fic. 37.

YounG Porites Mass GROWN OLDER Porites COLONY IN WHICH
EQUALLY ROUND A CENTRAL THE LOWER ZOOIDS ARE KILLED
NUCELUS. BY PRESSURE.

shape would be the type of form of all the massive species
which follow this method of growth and division, but practically
the dome shape is far less common than the flat-topped rock,
and this is for the reason that injury to the uppermost zooids
is usual in the life-history of a colony.

When the dome has become of some size, its upper surface
becomes large enough and flat enough to form a resting-place
for sediment, and the uppermost zooids decline in activity, the
compensating growth carries the sides farther out, and the
colony tends, by the increase of the rest of the surface, to
become still more flattened at the top. Injury caused by
loosened fragments sweeping over the surface of the rocks,
and the further deposition of sediment, finally cause the
wholesale death of the zooids of the flattened tops. The
zooids round the margins now form, by their active growth,

108 CORAL AND ATOLLS

swelling lips about the plateau, and make cushion-like bosses
which tend to enclose a central flat depression in which sand
accumulates, and on which other and differently growing
species of coral may lodge and flourish.

This process may be described as the normal accident of
the life-history of those species in which the equality of the

Fig. 38.

Aputr CoLony or Porites IN WHICH THE UPPER Zooins ARE
KILLED BY SEDIMENT.

zooids of every portion of the colony is a life-condition. It
furnishes a good example of the rule of Nature’s utter dis-
regard for the life of the individual, for all those zooids on
the upper surface must certainly be killed by sedimentation
or by injury, when they have succeeded in making their
colony of sufficient size. |

The existence of these flat-topped masses of Porites has
been noted by practically every observer who has visited reefs
of living corals, and various explanations have been given for
their origin. I have come to regard these rocks as being of
the utmost importance in the study of the question of the
development of atolls, and the deposition of sediment upon a
surface flat enough to allow of its resting I imagine to be the
active agent in their causation.

I regard the finished colony of Porites as an atoll reef in
miniature—the central dead area representing the lagoon and
the raised margin representing the encircling reef; and the-

LIFE-PROCESSES OF CORAL COLONY 109

factor that calls into existence the flat-topped Porites colony
also fashions the atoll-shaped reef.

This question will therefore be again referred to in the
chapter dealing with the formation of reefs.

The same process which leads to the normal shaping of

Fig. 39.

A SMALL CoLony oF Agaricia

To show the death of the upper surface due to the dropping of
sediment upon the colony.

massive coral colonies may be seen on a small scale at myriad
points upon the surface of the growth. The actual amount of
living tissue in a large growing boulder is very small, for the
depth to which the living animal tissues of the zooid extend
beneath the surface is very slight, and yet the calcified portion
which extends from the living margin to the centre of the rock
must in no way be regarded as the accumulated and dead
remains of past generations. It is in reality the skeletons
of those zooids whose mouths are now showing at the surface.
It follows, therefore, that since the skeleton devoid of animal

110 CORAL AND ATOLLS

tissues is incapable of carrying out any repairs, a superficial
injury is most likely to lead to the death of a definite area of
zooids on the surface—this area representing the base of a
cone whose apex is at the centre of the colony. The zooids
over this area, if definitely destroyed in the extent of all their
living animal tissues, are not regenerated, but a sympathetic
growth takes place round the edges of the area. New
material is thrown out, new zooids are budded off, and the ©
dead area is finally invaded and covered from the active zooids
of the edge. It is due to this process of repair that many of
the boring molluscs become enclosed in corals, for when the
surface has been attacked and killed, the margins by their
sympathetic activity tend to bridge over the injured area and
enclose the mollusc, which finally comes to rest in a cavity
beneath the surface of the coral.

The rounded cysts found in the substance of most speci-
mens of the massive corals, and which contain an encysted
mollusc, are therefore not to be regarded as entirely the work
of the mollusc, for they are due in part to sympathetic activity
of zooids in the coral colony.

Besides boring molluscs, several species of worms attack
corals and hollow out tunnels this way and that through their
living substance, and here, too, the sympathetic reaction of
the zooids is shown. The tunnels destroy whole groups of
zooids and replace the solid skeleton of the coral by a series
of tubes, and the strength of the whole colony is greatly
diminished. Around these tubes the uninjured zooids divide
and grow with increased activity, new calcium carbonate is
thrown out, and an attempt is made to compensate for the
destruction of tissue caused by the worm. Some curious re-
sults are brought about in this way, and specimens in which
worm-boring has led to fantastic growth are always to be
found, for few corals escape their inroads. A tunnel running
superficially, or in the thickness of a plate-like growth, leads
to an increased activity of surface-growth of the zooids, and
the tunnel stands out boldly from the surface, covered and
strengthened by an ever-advancing layer of coral. In this

LIFE-PROCESSES OF CORAL COLONY 111

way a tunnel may actually bridge a space from one plate to
another, and its mouth be carried out clear of the genera)
surface of the coral; for where the tunnel goes the surface-
layer of coral ever keeps pace.

(B) The question of the processes of repair becomes
further complicated in those corals, such as the Jadrepore,
in which all the members of the colony are not of equal im-
portance. We have seen, in considering the mode of growth
of such corals, that the very first individual in a colony may
continue to flourish and lead the growth of the entire com-
munity as long as that community lasts. The original zooid
which, as an embryo, settled on the basis that formed the site
of growth may be the “directive” zooid of the entire colony,
and the apical zooid may represent the oldest living animal
matter in the community. But besides the “dominant apical
zooid, others arise at intervals, by budding from the sides,
which are possessed of more inherent vitality than their fellows,
and from their first birth they tend to grow out as new direc-
tive zooids and lead to lateral branch formation—and besides
these more virile lateral zooids are the hosts of individuals
which in the normal condition of the growth reach no greater
dignity than a uniform projection of their corallites.

Now when injury or destruction affects a portion of the
colony, it reacts on individuals whose functionating values in
the economy of the colony are not equal, and so we should
expect that the result of injury or destruction would vary
according to the different parts of the colony on which the
maximum of damage falls. And this is the case. The actual
results of repair of various injuries inflicted in experiment, or
by Nature, show more clearly the relative values and functional
activities of different portions of such a colony than will any
amount of theorising or speculation.

1. If the injury be so inflicted that the branch of a
Madrepora colony is broken transversely, and the injury is
limited to a mere fracture of the cross-section, then the repair
takes place by the activity of the “ dominant apical zooid.”

112 CORAL AND ATOLLS

There is, as we have seen, no portion of a Madrepora
colony which nermally dies, and the obliteration of animal
tissues in the proximal portion of any zooid is a late change.
If the fracture takes place not very far from the end of a
branch, the inherent vitality of the terminal zooid predomi-
nates, and it starts the repair by continuing to grow out in
the direction of its original axis of growth, and by budding
new zooids from its sides.

In a measured specimen which was fractured cleanly across
one of the main branches without other injury to the colony,
the apical zooid had, at the end of a hundred days, grown
out 1 centimetre and had budded from its sides forty lateral
daughter zooids; and the general surface of the fractured end
showed seventy newly formed corallites of old and new zooids.
During the same interval of time a branch of about the same
diameter on the same colony, which had received no injury, had
advanced by 1:5 centimetres and had added about a hundred
and twenty new lateral zooids; so that, judged as growth in
these corals must be judged, the rate of repair is a rapid one.
In this case the dominant zooid is apical, and its superior
vitality enables it to regenerate and to continue the growth
along the lines of original branching; but if the vitality of the
“dominant apical zooid ” is definitely destroyed, a very different
state of affairs is brought about.

2. If the “dominant apical zooid” is destroyed, and
especially if the damage is extensive and affects a large area
of a branch, the predominant functions of the apical zooid are
taken over by the more vigorous lateral zooids, so that there
is a tendency towards branch-formation below the site of
injury.

This state of things, when put into terms of the life-
functions of the colony, means that potentially almost any
lateral bud possesses the inherent vitality of the apical zooid,
but it is only in times of stress to the colony that this
potential power becomes actual. A stem of a Madrepora
colony may shoot up straight for the distance of a foot, and
show nothing more worthy of the name of a branch than the

LIFE-PROCESSES OF CORAL COLONY 1138

normal projections of dominant zooids scattered irregularly
over its surface; but if sufficient damage be done to it to
destroy the apical zooid as well as a fair portion of its entire
length, then the remaining part will at once start budding

Fic. 40.

Fic. 41.

Type oF Reparr or Madrepora Tyrer or Repeater or Madrepora WHEN

WHEN THE “ DoMINANT APICAL THE ‘Dominant APICAL ZooIp” Is
ZOoID” IS NOT DESTROYED. ENTIRELY DESTROYED,
Process at the end of 100 days. Process at the end of 100 days.

with vigour. The best examples of this mode of repair are
seen in those cases in which a colony is attacked by filamen-
tous alge. The thin green threads of this parasite enmesh
the branches, and penetrating into their substance, lead rapidly
and certainly to the death of the part attacked. Artificially
produced injuries also provide good examples of this mode of
repair. The first stage is noticed when, in contrast to the
dead terminal portion, the living part swells out; its dominant
zooids become more conspicuous, and their projections increase
H

114 CORAL AND ATOLLS

until they arrive at a stage at which they themselves give off
lateral buds. The number of these enlarged lateral zooids
may be very great, but not all of them ever attain the dignity
of actual branch formation.

By the end of a hundred days after the receipt of injury
the uninjured central end of a branch is roughened all over its
surface by the projections of lateral zooids, and those of them
that will form new branches have already begun to develop
secondary buds from their sides. This process cannot be
spoken of with absolute accuracy as an actual repair, for the
part injured surely dies, and its substance is never regenerated ;
but it is a process by which the sympathetic activity of the
remaining zooids in the community is called into action, and
the potential power of branch formation inherently lodged in
the dominant lateral zooids is turned into a real power. The |
value of this process in the life-saving of a species is obvious, ©
for the growth tendency of a Madrepora colony is to steadily
develop upwards, and the limit of this upward growth is either
at the water-level, or at a level at which protection from
rough water and moving fragments afforded by the shelter
of surrounding rocks is lost. Now a colony which grows up
beyond its upward limit of safety will sooner or later have its
terminal ramifications killed by exposure, or broken by waves
or moving fragments. The injury that destroys the power
of the apical zooid causes the lateral buds to branch out at
angles to the parent stem, and spread fresh zooid-bearing
surfaces far and wide in the area of safety. It is this process
that is the great determining cause of the general growth
tendency of Madrepora colonies, and the one that causes the
deep-water and shallow-water forms of the same species to
vary in their vegetative forms of growth.

3. If in a colony the injury is such that the apical zooid
is neither injured nor destroyed, but the damage is limited to
the surfaces of a branch, then the repair takes place as in
colonies in which every zooid is of equal value.

This repair is well seen after experimentally inflicted
injuries, and the process of carrying out the regeneration of

LIFE-PROCESSES OF CORAL COLONY 115

the destroyed area is very like that previously described as
occurring in those corals that grow like the massive forms of
Porites.

The first step is the active marginal growth and the
formation of an excessive quantity of new material, which, in
the form experimented on (Madrepora pulchra), is at first of a
light blue colour, and is semi-transparent. In this new
material the mouths of corallites soon appear, and the edges
become covered by a host of uniform zooids, which soon spread
over the entire area destroyed by the injury, provided that the
area is not too large, and that no alga settles on it in the
meanwhile. It may be stated here that in experimental
injury, many experiments fail for the reason that the destroyed
or injured area commonly becomes a focus for the invasion of
boring parasites, worms, molluscs, sponges, and algze, and the
pure results of injury and repair become complicated.

Where no such complication existed, areas of 5 by 5
millimetres, 20 by 8 mm., 20 by 10 mm., and 25 by 12 mm.,
were completely covered by new material with a multitude
of new zooids in the course of 100 days, whilst larger areas
were commonly attacked by sponges or algz before repair was
completed. Very much the same process is seen to occur
when constricting metal bands are placed round branches.
The zooids, where the bands exert their pressure, are destroyed,
and the general growth of the branch and the sympathetic
activity of the zooids at the margin of the injury soon tend
to cause the band to sink into the substance of the coral,
and to become completely embedded beneath the surface.

It was previously stated that the type of repair naturally
tends to follow the type of growth of the injured colony; and
this is generally true. If a coral has a particular mode of
growth, and if it be in such an environment that its mode of
growth is the one most suitable, then an injury is repaired or
a part regenerated on the same type of growth as that of the
colony. But the particular form in which a colony may be
growing may not be the form best suited for successfully
flourishing in that particular environment ; and then, following

116 CORAL AND ATOLLS

an injury, a remarkable state of things is brought about, which
may be stated as follows:

II. A coral may repair an ijury by a new growth of a different
type from that of the colony, and in such cases the reparr
growth is of a form better suited to the environment than is
the form of the parent colony.

It must be borne in mind that the physical conditions of
the atoll are not absolutely constant. Seasonal changes in sets
and strength of currents are perpetually causing alterations at
all points of the island ring, and a habitat may alter greatly in
its physical conditions in the course of a few months. When
a coral embryo settles down upon a basis, and starts the
founding of a colony, the type of growth which will result will.
be that best fitted to the environment as the embryo finds
it. But suppose that, after the founding and growth of a
colony, the physical conditions of the environment change ;
rough water is admitted, silt settles down, or the water
becomes shallow and calm: then the growth of the colony
may not be the ideal one for flourishing under these new
conditions.

In many places where the conditions are prone to vary,
the habitat of the corals may alter in its physical nature in
more or less regular cycles. Where, as at the eastern
extremity of Pulu Tikus, a spit runs out, building sheltered
pools and protecting the shore for a considerable distance, and
then later on, in the periodical cycle of currents and eddies,
is carried away, many corals must be subjected to a great
variation of environment. In such circumstances no doubt
many colonies die, for, as we have seen, a sudden change of
habitat cannot be resisted when a form is well adapted for one
definite kind of environment, and in any case the colonies
are liable to injury and partial death in their changed
surroundings.

When such injury falls on a coral not ideally situated,
the repairing growth exhibits a strange independence, and,

LIFE-PROCESSES OF CORAL COLONY 117

forsaking the growth-form of the colony, builds its repair in
the form best suited to the new conditions of the environment.

This is a strange zoological fact, that the inherent growth-
form, once stamped upon a well-established colony, continues
to be the type of growth, though it be but ill adapted to its

Fic. 42.

Montipora COLONY

Repairing a branching growth by an encrusting layer.

habitat; and yet, when the continuity is once broken and a
new start is made, the newly budded zooids can throw off the
stereotyped method, and build anew to the altered conditions.

It is facts such as these which give some clue to the under-
standing of the vast range of variation occurring within the
limits of a species, and make the establishing of a species a
matter of extreme doubt, until every possible variation which
different surroundings will stamp upon the type has been
examined. We have seen, in following the life-history of
corals, that the colony shows great adaptability, being able to

118 CORAL AND ATOLLS

mould its growth in response to the demands of its environ-
ment, and in this repair process it possesses a further power,
for it can entirely alter the structure of an established colony.

Numerous examples of this strange process may be seen.
A branching Montipora, growing in a gap in the island ring, is
found to have every colony dead or damaged in a greater or
less extent of its whole growth. The damage is probably due
to the fact that currents have altered the physical conditions
of the habitat since the founding of the colonies, and that a
greater rush of water has brought more sand and moving
particles in contact with them, for the apical branches of all
the colonies within a definite area will be found broken. The
repair of this damage invariably takes the form of an amor-
phous encrusting growth covering the débris of the dying
colony, the regenerated portion keeping pace with the destruc- -
tion, and thus keeping the colony living—but living as an
entirely different type of its species.

Madrepora colonies show the same phenomena, and very
strange repair-forms of Pocillopora growing 1n rough water as
encrusting growths may be found.

When, after repair, a Madrepora colony assumes an en-
crusting form, as it frequently does, the inherent tendency of
its growth is still evident, for rising at intervals from its flat
surface are numerous dominant zooids, which, were the oppor-
tunity afforded them, would form upward-growing branches.

It is by no means uncommon, in this process of repair, not
only for the vegetative growth of the colony to be altered, but
for the actual type of the corallum to be changed. When a
Montipora repairs its own ill-situated and dying colony by an
encrusting growth, the whole minute structure of the coral is
changed. Instead of the smooth surface over which the fairly
wide mouths of the corallites are dotted, is a coral with an
outward appearance notable chiefly for its extreme roughness,
due to the development of numerous papillz at the bases of
which open the minute corallites. The characters of the
original growth and of the repair growth are so entirely
different that they would certainly be regarded as two distinct

LIFE-PROCESSES OF CORAL COLONY 119

species. Other examples of this entire change are not un-
common, and some very strange abnormalities may be found,
which show that in repair, as in adversity and in alteration of

Fig. 43.

COLONY PARTLY KILLED BY DEPOSITION OF SEDIMENT,

Showing abnormal type of division,

environment, there is practically no limit to the developing of

the different main types of growth by any coral. It is common

in repair of injury, as it is in cases of adversity, that the

bounding walls of the newly budded zooids are never com-

pleted, and repair by a pseudo-meandrine form of fission is

often met with in many and widely separated species of coral.
2K

120 CORAL AND ATOLLS

It is common, too, to find that the repair-zooids have raised
corallites, when those of the parent colony are flat, and in these
cases it is probably silt that has caused the original damage.

The study of the repair in corals is therefore one not
devoid of zoological interest, for it shows clearly that a type
must never be considered as a species, in the way in which we
regard species among the higher animals, until it has been
seen in all its variations, and until all the possible modifications
that repair produces have been studied.

That a type like the encrusting Montipora should be in
reality the same species as the branching form would be
considered as highly improbable, but when it is seen that the
one type repairs its damage by the development of a new
growth of the other type, there is no alternative but to regard
them as identical species.

That the numerous types of Millepora and of Pocillopora
should be but variants of a single species would seem at first
sight to be very unlikely, for there is little enough likeness
between the extreme forms, and yet their processes of repair
show them capable of building to any of their diverse types,
regardless of the nature of the parent growth.

If the processes which have been described, and the con-
clusions which have been drawn from them, be accepted, they
can serve only to make clearer the great fact that morphology
—the animal’s type—is the outcome of necessity; and here
the demands of necessity bring about change, not in the life-
history of a species only, but in the life of the individual.

CHAPTER X
THE PROCESSES OF DEATH IN THE CORAL COLONY

THE subject of repair leads naturally to the consideration of
the death of the organism, for when the destructive processes
outweigh the resources of repair, then death must inevitably
ensue.

There is one fact in the life-history of corals which the
study of their processes of repair clearly brings out, and it is
this, that all the methods of regeneration are more for the life-
saving of the colony than of the individual. It is a rule with
Nature that the life of the individual is of little moment:
Nature has little care for individuals, though she strives
always to maintain the life of the species. In a _ vast
community of individuals, as is a coral colony, each separate
member is but a part of the whole body, and the preservation
of the colony is of more import than the saving of a few
individuals. A branching Madrepora grows naturally upwards
into the danger zone, and the terminal branches are inevitably
destroyed, with the sacrifice of a myriad of zooids; but the
result is a stimulus to lateral branching within the area of
safety, and the colony continues to flourish.

In every massive growth starting to develop on all sides
of a nucleus, those zooids that are budded below can never
hope to live, and those on the upper surface will in all
probability die. In all the processes of repair that have been
described, it is not the individual which is mended, for an
individual once badly damaged is not repaired ; but the loss is
made good by the growth of new zooids which assume the
functions of those lost. Repair in colonial forms does not
save the individual from death, but it preserves the life of the
colony. Now loss by death in a colony is not always repaired.

121

122 CORAL AND ATOLLS

We have seen that the flat tops of the massive growths of
Porites remain devoid of living zooids, and in several types of
growth, death of a portion of the colony is more or less usual.
Among the branching Porites it is norma] to find the lower
portions of the growth dead, and no attempt made at their
repalr.

When it is said that the partial death of a colony is more
or less usual in some types of growth, it is not in any way
meant that the progress of coral formation is a building of the
living zooids upon the dead bodies of past generations, for the
partial death is due, as a rule, to very definite outside causes.

In this atoll the greatest cause of coral death has been
quite an unusual one. but it has been a most instructive one,
for it brings out some very interesting facts in the life-history
of corals. In 1876 all the living coral of the south-east portion —
of the lagoon was entirely destroyed by the pouring out of
foul water from a supposed voleanic vent at the southern side
of the atoll. The picture presented by these denuded areas
was described by Dr. H. O. Forbes in 1879, and by Dr. Guppy
in 1888; and in 1906 there was still the same tract of dead
coral on which the efforts at recolonisation have been practically
unavailing.

This remarkable failure of the corals to repopulate a large
portion of the lagoon is probably due to the fact that during
the period immediately following the disaster various alge,
such as agar-agar and several other species, being of a faster
and more hardy growth, stepped in and took possession of the
area before the slow-growing corals could obtain a proper
footing. The growth of algz is in itself hostile to the life of
corals, and, apart from that, the alge beds in the lagoon are
the greatest factors in catching silt and piling up the shifting
sand-banks, the presence of which is so fatal to coral growth.
These two factors, aided perhaps by subsequent minor poison-
ings, have so completely paralysed all coral activity that to
the south of Pulu Selma there has been in many places no
trace of new growth; and the abnormal death which occurred
thirty years ago has remained unrepaired to this day.

DEATH-PROCESSES OF CORAL COLONY 123

Although such events as this are quite exceptional in the
life-history of corals, or of coral islands, still the after-history
of the disaster shows on a large scale the influences of those
factors which in the normal life of corals tend to bring about
their death. It is the silt and seaweed that have prevented
corals from flourishing on their old site, and the silt and
seaweed are to-day in the atoll the two great causes of coral
death.

The influence of matter suspended in the water is one of
the most far-reaching factors in the life-history of corals: it is
to resist its effects that many of the vegetative forms are
evolved; it is on account of the silt that many acres of the
lagoon are devoid of coral growth; and it is probably on
account of the presence of silt that wave action is so neces-
sary to coral life, and that the unstirred depths below about
20 fathoms are comparatively bare of coral.

Sut, sand, or suspended matter may cause the actual death of corals
im two ways: (a) It may fall upon them and choke their
zooids from above. (b) Lt may overtake them from below.

Of these two actions examples are always to be seen in
(a) the partial death of the tops of massive Porites colonies, and
in (6) the stems and lower branches of branching Madrepore
which are normally lifeless. In these cases the death is only
partial, for the reason that the colony is capable of resisting
as a whole the amount of suspended matter normally present
in the waters of its habitat ; but if the amount be suddenly
increased, the colony may be unable to resist it, and general
death ensues. Evidences of this mode of death are seen in
the gaps im the island ring where an alteration of current
brings more silt than is usual to the growing colonies, and
very interesting results may be produced experimentally. On
December 13, 1905, several healthy living colonies of rough-
water forms of Madrepora and Pocillopora were removed, with-
out exposure or injury, from their habitat of rough barrier
water, and without any delay were placed in marked sites in a

124 CORAL AND ATOLLS

sheltered sandy pool of the barrier flats. In the same pool,
which is almost completely cut off from the sea at low tides,
and then contains about 2-3 feet of water, and which is about
100 yards long by 20 wide, numerous corals live and flourish ;
calm-water forms of Madrepora and Pocillopora, capable of resist-
ing silt, being the most abundant. The conditions of life in these
barrier pools are peculiar. The pools are filled with sand, for
the fragments which are triturated by their journey to and fro
over the barrier are deposited in them, they contain the
minute green filaments of the boring alge, and at midday low
tides they become heated by the sun to 93° Fahr. and more.
Their coral fauna is practically always the same; MJadrepora
flourishes in its most highly branching forms, Pocillopora always
has a good foothold, and the other species are in plenty, but
hardly in a state of luxuriance. Béche-de-mer in hundreds
live in these pools, and crustacea, polychetes, mollusca, eels,
and the myriad brilliant fish make up the conspicuous fauna.
Of the many rough-water colonies which were transplanted
experimentally into this environment, not one remained alive
at the end of 50 days, and most were dead within a month.
The first sign by which a colony shows that its environment
is not suitable is by becoming highly pigmented. Both rough-
water forms are, when flourishing, very pale corals, being
usually of a light buff colour; but within a fortnight the
transplanted colonies had become of a dark yellow-brown, and
in Madrepora there was a more than usual tendency to lateral
branch formation. In 20 days most colonies had some portion
dead, and the dead parts became rapidly the site of growing
alge. In 30 days nearly all the numerous transplanted
colonies were dead or dying, and by 50 days no portion of
any colony remained alive. It was silt that had determined
the death of them all; the stunted, flattened types of
Madrepora and Pocillopora are both corals of a rough-water
habitat, they are used to clear water in which there is little or
no suspended matter, and not a single colony was able to with-
stand the slow but certain sedimentation of the barrier pool.
When the silt had once fairly determined their death, the fine

-DEATH-PROCESSES OF CORAL COLONY 125

boring algze completed the ruin. If any colony of branching
coral be removed entire from the lagoon, it will be found that
the lower portion is invariably dead ; but this death is in most
cases not a natural one resulting from the senility of the
zooids, but merely an index of the amount of silting up of
the lagoon which has taken place since the establishing of the
colony. Sand is ever being washed into the lagoon through
the numerous gaps in the island ring, and most decidedly the
tendency all over the lagoon is a gradual filling up, by the
deposition of finely triturated fragments. The floor of the
lagoon is fairly steadily rising, and those colonies of corals
growing in its bed are for ever being encroached upon by the
gradual rising of the sand level. The deposited sand most
certainly kills the zooids with which it comes in contact, and
the result is that the lower portion of every lagoon colony is
killed. Silt then, in this atoll, is the most potent factor in
causing coral death, and next in importance to the silt comes
the seaweed. ;

There is a green alga which, at some seasons of the year
more than at others, comes to the barrier pools in great
quantities: it is a growth of fine green threads, and its effect
on coral growth is really wonderful. A pool in which nume-
rous flourishing colonies live, quite on a sudden may show
the advent of this alga, and every portion of every colony,
which may receive a chance injury, at once becomes the site
of the growth of these fine threads. I believe that it is
always at the site of injury that the attack of the alga starts,
but its growth soon invades, and invariably kills, the living
healthy portions of the colony. A colony once fairly invaded
by this parasite rapidly dies, and yet it never succeeds in
obliterating coral growth, for as suddenly as it came to the
barrier pools it goes. Spring tides and hot weather seem to
promote its growth, or perhaps lower the resistance of the
coral colonies, for when the rock-pools are left long to swelter
in the sun, with but little depth of water in them, then the
alga seems to be most active. It is a great factor in causing

the death of the atoll corals, and ranks in Cocos far in advance

126 CORAL AND ATOLLS

of the boring sponges, worms, or molluscs as a destructive
agent.
The more obvious boring creatures do not cause damage

Bie. 44.

COLONY SHOWING DEATH BY SEDIMENT OVERTAKING IT FROM BELOW.

to the colony other than to weaken its structure, and lead to
its nore ready destruction by the action of the waves or
moving fragments; and beyond this they cannot be rightly
considered as effective enemies of coral growth. There seems
to be indeed an almost symbiotic relation between certain
boring animals and the corals that they have chosen as their
hosts, for coral growth extends and strengthens their tubes by
sympathetic growth, and the cavities of the molluscs in many
cases expand the living area of the surface of corals by causing
irritation and repair. It is an extraordinary thing to see the

DEATH-PROCESSES OF CORAL COLONY 127

extent to which a colony of Millepora complanata may be
riddled with the wide smooth-walled tubes, and yet not be
appreciably weakened, and the resultant observation on such

Fic. 45.

SPECIMEN SHOWING THE TUNNELLING OF A LivING CoRAL
CoLony BY Borinc PARASITES.

colonies is that the borings are very little harmful to the
colony.

Another cause of coral death to which much importance
has been attached is the exposure caused by the receding tide,
for it has been said that corals cannot survive even a
temporary exposure to the sun and air.

Since Darwin first claimed this as an axiom of coral
bionomics, a great deal has been taken for granted with regard
to the effects of exposure, and yet every fresh investigator
has attached less and less importance to it. Now, as a matter

128 CORAL AND ATOLLS

of fact, there is no species of coral in this atoll that is not
able to withstand an exposure of many hours to the midday
sun, with from 6 inches to a foot of its apical growth above
the water: there is no barrier species that does not normally
suffer this at midday spring tides. There are many isolated
rocks that are ordinarily exposed for two feet at low tide, on
which living corals flourish luxuriantly. When season and
winds combine to cause tides abnormally low, it is possible to
go from island to island along the barrier flats, and for the
greater part of the journey to walk in but a few inches of
water ; and if such a walk be taken during a low tide at hot
midday, the smell of the exposed coral is almost overpowering
and may be noticed far out in the lagoon. Coral has an odour
that is peculiarly offensive when the growth is exposed to the
air, and this strange odour is no sign of death, for a stinking
coral when replaced in water or when re-covered, by the
rising tide, flourishes again. At such a low tide the barrier
flats present a picture of bushes and boulders of living
coral all freely exposed to the sun, all dry, all smelling very
offensively, and yet the returning tide finds them all living as
actively as when it left them. A coral may be taken from
the bottom of the lagoon, may spend the best part of a hot
day, high and dry, in the bottom of a boat, and yet, when it
is replaced in the water, all its zooids will expand, and it will
resume all its vigour. Exposure to the sun and air between
tide limits plays but little part as a causative factor in the
question of coral death. Of course no coral could grow beyond
the normal high-tide level, but the remarkably level appear-
ance of the barrier flats is not the result of the action of sun
and air on the coral colonies so much as of the levelling
effects of the waves, and the moving fragments which they wash
to and fro. It is the grinding action of the surface waters at
their level of maximum activity which determines the limiting
level of upward coral growth, far more than the death of the
apices from the effects of sun and air. The waves that sweep
over the flats, and carry shorewards the fragments which they
have broken from the seaward margin of the barrier, are for

DEATH-PROCESSES OF CORAL COLONY 129

ever keeping the coral colonies within the limit of upward
growth ; but their action is not altogether detrimental to the
corals, for though, where island beaches are formed, many
fragments of living coral are cast ashore only to perish, still
many more, where no such dry land exists, are safely lodged
in a new resting-place. Broken fragments are swept across
the flats, they lodge in pools, they become stranded under the
lee of boulders, or are washed into the lagoon; and each of
these fragments, if not too badly damaged, will form the
nucleus of a new colony in a suitable habitat. If a large
colony of Madrepora be broken up. in the rock pool where it
flourishes, the great majority of its fragments will continue to
grow and branch out into new colonies; and if some of these
fragments are swept onwards by the waves, they will form
pioneers for the species when lodged in a suitable environment.

Freshening of the water from the excessive tropical rains
has been said to cause wholesale death among the lagoon
corals ; and in high islands where rivers flow into the sea, the
fresh water is well known to be a great cause of the absence
of coral growth. Before Darwin’s visit to the atoll it is said
that an abnormal rainfall killed many corals, and again in
1866 the fresh water is said by the Governor to have stood
for a height of several inches on the surface of the lagoon, so
heavy and continuous was the rain, Again, in May 1896, the
rains were abnormal, and the freshening of the water destroyed
the lagoon algz and fish; and this in such quantities that
when Mr. Arthur Keyser visited the islands in July, the dead
fish were still being cleared from the lagoon. There is no
doubt that the rain would have to be long continued, for all
those corals that live in rock pools are immune to the
influence of the fresh water accumulated during a heavy
downpour at low tide. When the tide is low and the rainfall
is heavy, the rock pools undergo a remarkable degree of
freshening, and so too does the lagoon if the weather is calm
and the rainfall is a sudden one.

In the lagoon the surface specific gravity may fall to 1021,
but I have never found it lower, and the perpetual churning

I

130 CORAL AND ATOLLS

of the waves prevents any marked evidences of freshening being
observable near shore. Outside, in the ocean itself, there is
such complete mixing of the waters at the surf-beaten barrier
edge that it is not likely that the influence of the rain could
be recognised.

On February 13, 1906, when ‘28 of an inch of rain fell
in ten minutes, the sp. gr. of the surface of the lagoon
dropped from 1027 to 1021, and the temperature was 77°°5
Fahr.

On February 24, after -5 of an inch of rain had fallen in
half an hour, the sp. gr. was 1023, and the temperature
82°-7 Fahr.

On January 4, after 7°4 inches had fallen in twelve
hours, the sp. gr. was 1021, with a temperature of 78°
Fahr. .

The surface of the fish-pond, which is a pit about 15 feet —
square, will show a reading as low as 1015 three days after a
downfall of 5 inches in twelve hours, although its waters rise
and fall with the tides,and the outside ocean shows no change
after the downpour.

It is therefore mainly wave-action that obliterates the
effects of tropical showers in freshening the salt water, and
the coincidence of great rainfall and dead calm must be very
complete, and very lasting, before anything approaching a
general destruction of corals could result.

Many animals have been ranked amongst the enemies of
corals, and Darwin classed some fish and the myriad Holo-
thurids as causes of coraldeath. In this atoll Dr. H. O. Forbes
has described the “ Scarus feeding in the surf on the living
coral,” and has asserted that the Kakatua and other lagoon
fish actually eat the living polyps. The observation has been
several times doubted, and, so far as this atoll is concerned, it
is certainly an error. There are no fish and no Holothurids in
Cocos-Keeling lagoon, or on the barrier, that eat coral when it
is living, though many different classes of animals contain great
quantities of dead coral in their alimentary canals. The
importance of the coral-haunting fish and the Holothurids as

DEATH-PROCESSES OF CORAL COLONY 131

factors in atoll formation is great, but it is not as destroyers
of living coral that they fulfil their role, for the coral that they
take in at their mouths is already dead.

From the study of the life of the colony in different
surroundings, and from the repair of injury, and death, in
unsuitable habitats, I think it will be seen that the number
of the true species of corals is by no means so great as is at
present supposed. There is no doubt that a great number of
our museum-made species are mere vegetative varieties, pro-
duced in response to the demands of the environment. I
do not think it is possible to determine from a fragment of
growth—often with no sufficient data—if it be a new species,
or even a new genus, or if it be a mere vegetative variety of
some already well-known species. There can be but little
utility in the naming and describing, with great minuteness, of
all these variations; for of this work there can be no end, and
persistent collecting, from even such a small area as the Cocos-
Keeling atoll, would yield such a variety of fragments as would
occupy a lifetime to describe. In very many cases one single
colony could be found to provide several types of growth, which,
if presented as fragments, would be deemed to merit individual
description as species. In such cases some factor in the
physical condition of the surroundings will show, when the
colony is in situ, the cause of these different modes of growth ;
but when the colony is transported to a museum, it presents a
very striking puzzle.

One side of a colony may be shaded from light, sheltered
from currents, or protected from silt; whilst the other side
may be exposed to all these influences: and then it is but
natural that the two sides should vary, and—knowing the
wonderful plasticity of the zooids—the great differences are
not astonishing.

Besides the occurrence of colonies which exhibit two or three
well-marked types of growth, there are those that can only be
called “undecided ” forms, and these present growths that are
intermediate in character between two well-marked and very
diverse types. Such “undecided” forms are very common,

132 CORAL AND ATOLLS

but it is the natural instinct of the collector to pick out well-
marked and well-grown forms as his specimens.

Fic. 46.

Tur ‘*STINGS’” CAUSED BY CONTACT WITH A COLONY OF
Millepora alcicornis,

In 1898, Professor Sydney Hickson made an exhaustive
study of the so-called species of the hydrocoralline JMillepore,
and he came to the conclusion that there were “very strong

DEATH-PROCESSES OF CORAL COLONY 133

reasons for believing that there is only one species of
Millepora” (P. Z. S., 1898, April, p. 256).

Previous to the writing of this paper he had come to the
same conclusion concerning the Alcyonarian genus Tubipora,
and regarded it as possible that the species of the Zoantharian
corals might have to be considerably reduced.

The Millepore are exceedingly abundant in the atoll, and
the well-marked forms Millepora alcicornis, M. complanata, and.
M. verrucosa are to be found in plenty. But these well-marked
forms are all linked up by intermediate stages, and I have
no doubt that all are variants of one species. In the atoll
the Millepore are called Karang gatal—or “itchy corals,” for
they are capable of inflicting a painful sting when touched.
The sting is like that of a nettle, but far more intense, and
grave results may follow it. It is curious that of the three
forms, M. alcicornis has the power of stinging more severely
than the other two, but this fact does not necessitate their
taking specific rank, and I do not doubt that there is but one
species of Millepora. I donot doubt either that all the forms of
Pocillopora which are found in the atoll are in reality one species ;
_ and I strongly suspect that there is only one species of Montipora

in Cocos, although its varieties are legion. The species of Madre-
pora in the islands are in reality very few; many diverse forms
are certainly identical species, but experimental breeding must
finally settle how few these species are.

It is the same throughout the whole series of the Cocos-
Keeling corals; there is a very limited number of species;
and I would account for the origin of the many varieties, and
the present confusion of their nomenclature, by the alteration
of environment caused by atoll formation.

When the origin of the atoll is considered, it will be
demonstrated that the whole structure is developed from an
original submerged bank, and on this submerged bank I
would imagine that the corals represented few species and few
varieties.

The life-conditions all over the bank were fairly uniform,
and there lived upon it Pocillopora, Montipora, and the other

134 CORAL AND ATOLLS

corals, exhibiting probably one form of growth only, and one
that is represented by an intermediate type to-day.

With the origin of the heaped-up débris that forms the
island ring, and with the formation of the barrier above the
level of the tides, a gradual change occurred, and in the place
of one uniform environment, an infinity of diverse habitats
was produced.

The rough water of the barrier, the smooth water of the
lagoon, the silting water of the inlets, and the clear water of
the ocean, were marked off from one another; and the em-
bryos of the originally similar corals had to grow dissimilar to
adapt their vegetative types to the new-formed habitats.

In this way the present infinity of types was brought
about, and wherever the environment is changing to-day new
types are developing to conform with its demands.

PART III
THE ATOLL AND ITS PROBLEMS

CHAPTER XI
THE ATOLL AS A WHOLE

I HAVE chosen to dwell somewhat fully upon the life-history
of the coral colony for the reason that in the study of the
development of atolls and reefs, a knowledge of the growth
tendencies of the colony is an essential. The question of the
formation of atolls is in reality a zoological one, for every
square inch of their structure is the outcome of the activities
of living creatures, and if we are to appreciate rightly the
building of atolls, it is necessary to understand the lives of
the zooids which produce them. In the portion of this book
dealing with the corals themselves I have endeavoured to
show that the study of corals cannot be properly conducted in
a museum, by the examination of dead specimens far removed
from the site of their growth, and from all the influences of their
environment. In these chapters I would endeavour to estab-
lish the fact that the development of atclls and reefs is,
in reality, a question to be studied by a careful inquiry into
the mode of life of the coral zooids themselves ;—that it is
not a problem to deal with in laboratories or museums, for the
laboratory can never be a substitute for the reef as a field for
experience.

In the explanation that I shall furnish for the making of
-reefs and atolls, no other processes will be invoked than the
normal growth tendencies of the coral zooids, and those effects
upon land-making that may be seen wherever the winds and

the waves act upon matter. J am convinced that the more
135

136 CORAL AND ATOLLS

we tend to divorce the study of the structures raised by the
zooids from the study of the zooids themselves, the less will
we tend to approach their true solution.

The subject-matter of these chapters may be taken as

representing the outcome of a somewhat prolonged series
of personal observations, made upon the living, growing reef,
that supports one of the most perfect atolls that is in
existence. .
Lest my practical knowledge of coral formations should
rest only upon the examination of a perfected atoll, I embraced
the opportunity, afforded by the kindness of Captain Maclear
Ladds, of the Singapore Pilot Service, of examining the many
reefs and coral structures that lie in the neighbourhood of
Singapore ;—and I was further enabled to examine, though
in no great detail, coral structures that fringe the shores of |
Sumatra, Java, Sumbawa, and the other Malayan Islands ;—
and finally to make the double journey of the whole extent of
the Great Barrier Reef of Australia.

I was early convinced that “ Subsidence ” will not explain
the development of all coral formations; and that “Solution”
will certainly explain still less ;—and, in the explanation that
I would advocate, the chief agent is one that acts upon the
living colony—and is “ Sedimentation.”

T am well aware that the harmful effects of Sedimentation
have received previous attention, but I would extend their
action, and I would make them by far the most important
agents in producing all the forms of coral structures as we find
them to-day.

The processes of Sedimentation which have been followed
in their action upon the colony will therefore be studied as
they are to be seen displayed upon a grander scale. It has
been demonstrated how the harmful action of Sedimentation
can produce an atoll-shaped rock a few feet in diameter,
and it is logical to inquire if an atoll like Cocos-Keeling,
whose lagoon is several miles across, might be produced in
the same fashion. The atoll as a whole must therefore be
examined.

THE ATOLL AS A WHOLE 137

The group consists of two separate atolls:—a small
atoll to the north, which is Keeling Island proper; and a
larger atoll, some 15 miles to the south—the Cocos atoll.
The information concerning the changes in the atoll to be
derived from old charts is practically nil, and it is a great
mistake to place any reliance on what are, after all, only rough
sketch-maps made by passing mariners. In some cases de-
ductions have been made as to changes that are presumed to
have taken place in the atoll, by noting the differences which
exist between these old mariners’ sketch-maps and modern
charts. Such deductions are liable to cause grave error. In
some cases such errors have been caused, and mention will be
made of these as the subject comes under discussion.

The group lies in the eastern portion of the Indian Ocean,
12° 9’ south of the Equator, and 96° 53’ east of Greenwich ; it
is separated by some 600 miles from Java; some 500 from
Christmas Island ; some 2000 from Cape Leeuwin, Australia ;
and by about the same distance from Ceylon.

The atoll lies in the track of the South-East Trade winds,
which blow for 300 days in every year; and in the Westerly
Equatorial sea drift, which flows constantly past the islands.

Various accounts of the atoll have appeared from time to
time, and its name must always remain associated with the
theories of Charles Darwin, for when he made his visit in the
year 1836 he saw, in certain features of the atoll, all the signs
that he had previously pictured as being the evidences of
gradual subsidence. It was the only atoll that Darwin ever
examined, and it was the one on the examination of which the
famous theory of “Subsidence” was founded.

Dr. H. O. Forbes visited the islands in 1879, and the out-
come of this visit was the interesting account which appeared
in “A Naturalist’s Wanderings in the Eastern Archipelago,”
published in 1885.

Dr. H. B. Guppy made a stay in the group in 1888, and
has published by far the best account of the actual structure
of the atoll in the Scottish Geographical Magazine, vol. v.,
Nos. 6, 9, and 11.

138 CORAL AND ATOLLS

Darwin’s stay in the atoll was a short one lasting only
ten days; Forbes stayed for three weeks, and Guppy, who saw
far more than either of the two previous visitors, lived for ten
weeks with the Clunies-Ross family. Neither Darwin nor
Forbes landed on the Keeling atollon, and its description by
Dr. Guppy was the first ever published—in fact Guppy was
the first scientific man who ever landed on the island ; and to
this day Keeling remains one of the spots least Tifiel by men
that the world can still boast.

THE SOUTHERN GROUP

THE COCOS ATOLL PROPER
CALLED ON OLD CHARTS

THE TRIANGULAR ISLANDS

Some twenty-four islands enter into the composition of
the atoll ring; twenty-three of these islands constitute a
horseshoe-shaped rim to the lagoon, and the twenty-fourth
lies in the gap of the horseshoe, where the lagoon com-
municates with the open sea.

All these separate islands (with the exception of Pulu
Luar) rise from a common stratum of level coral breccia
rock—the barrier flat; and this level stratum is in reality the
rim of the great reef that paves the bottom of the lagoon, and
falls away outside to the ocean depths. The rim of the reef is
some twenty-five miles in length, and a total of about seventeen
miles is covered by the dry land of the islands. The islands
vary in size from such considerable pieces of land as Pulu
Panjang—some six miles long and half a mile wide—to mere
mounds detached from other islands, or breccia masses heaped
on the barrier flats. The eight miles of the edge of the reef
that is not dry land varies in its nature in different parts of
the circumference, and these eight miles are the most
interesting miles of the whole of the atoll, for they still show
how the other seventeen miles have arrived at their present
condition—how, in fact, coral islands are made on the rock

PLATE IV.

tkus.

lu Pasir

— 1

+ ai

CORAL AND ATOLLS

PLATE Ivy.

Pulu Tikus.

Pulu Pasir.

Pulu Bras
Pulu Gangsa.

Pulu Selma.

B I
atu Satu ré Pulu Kechil.
i] te
fj ? Pulu Ampané.
1 oi a 8
yi 2 4 Pulu Blukok.

Pulu Kembang.

Pulu Pandan.

Pulu Gray.
Pulu Siput

Pulu Jambatan
ulu Labu.

=

~~ a — ee

Pulu Maria‘ Pulu Klapa Satu.

Pulu Blanche.

96° 52°

CHART OF THE SOUTHERN GROUP OF COCOS-KEELING

ISLANDS.

THE ATOLL AS A WHOLE 139

flat. The essential feature which constitutes the varying factor,
in the parts of the ring which are not the sites of islands, is the
presence or absence of the barrier; for the barrier, which is
the mainstay of the existence of the atoll, must always be
regarded as potential dry land, although still belonging to the
ocean. The barrier does not form a complete rim to the reef,
but, as a pioneer, it outruns the land, and is wanting in only
about three and a half miles of the whole circumference.
The parts in which the barrier fails are situated to the north
and west of the atoll ring, and the sailors’ rule, that the
entrance to a lagoon is to be found on the leeward side, is
therefore justified.

The whole atoll is in the form of an ellipse with its long
axis stretching in a south-easterly direction—prolonged in
fact in the line of the prevalent wind, a common and neglected
feature of atoll form. The ellipse is some ten miles long from
barrier to barrier, and some seven miles across its short axis.
The whole atoll therefore occupies very much the same area
as does Christmas Island, its nearest neighbour.

The largest islands lie to the south of the ring, and Pulu
Atas and Pulu Panjang are the largest land masses of the
group. Not only are they longer than any others, but they are
also much wider from sea beach to lagoon shore, they are the
highest land in the whole atoll, and are evidently the oldest of
any of the islands. The islands to the north are smaller, and,
as a rule, very much lower; and they have evidently altered
very little from their primitive condition of being mere heaps
of coral débris. Pulu Tikus is one of the lowest of all the
islands in the group, and its general elevation is only from two
to three feet above mean tide-level; its highest point is raised
some thirteen feet above the sea, and much of its surface
is below the level of spring high tides. It is difficult to realise
the extreme lowness of the islands of the atoll, and it is always
a cause of surprise to any one who sees these things for the
first time, although the fact may be perfectly familiar. As a
homely illustration of the actual condition, it is useful to take
the level of a railway line, to represent the sea, and then the

140 CORAL AND ATOLLS

platform of the station gives a very true idea of the height
to which the islands rise above it. There are other diffe-
rences between the windward and the leeward sides of the atoll.
At the southern and eastern extremity the maximum of wind

Fig. 47.

THE LAGOONLET OF PuLU LuAaR at HigH TIDE.

effect is seen, and sand covers the surface of the islands in
wind-blown ridges of some depth, so that, in places, actual
hills, thirty feet high, rise in Pulu Atas. The barrier flats are
far wider on the south side of the group, and larger boulders
are piled upon them; the sea beaches are steeper, and are
composed of larger fragments; and lastly, the lagoon itself is
far shallower in its southern portion.

It has been said that the atoll is horseshoe-shaped, and
this form is not confined to the whole collection of islands,

THE ATOLL AS A WHOLE 141

but is stamped on every constituent piece of land in the atoll
structure. This is a very important point, and any explanation
of atoll formation must also furnish some satisfactory reason
for this shape belonging to the individual islands of the atoll
ring. The incurving of the ends of the islands leads to the
formation of the secondary lagoons that are present in Pulu
Atas and in Pulu Panjang, and are foreshadowed in Pulu
Pandan and Pulu Ampang.

One of the constituent islands is in reality an atollon, and
encloses a lagoon which has altogether ceased to communicate
with the sea. This island is the leeward one, named Pulu
Luar, and the lagoonlet is situated to its leeward side; a few
feet of water rise and fall in it, reaching it by tidal percolation ;
but a débris beach cuts it off from any direct communication
with the leeward shore, where its entrance used to be. The
changes that have taken place in this very interesting atollon
will be further discussed when the method of formation of the
islands is under consideration.

THE NORTHERN ATOLLON
KEELING ISLAND

Keeling Island lies some fifteen miles to the north of the
Cocos atoll proper; it is an island that has been very rarely
visited, and it is one of great interest. It is somewhat over a
mile long from north to south, and is about half as broad as
it is long. In its present condition it is an almost complete
circle of land that encloses a shallow lagoon still communi-
cating with the open sea. The lagoon contains something
rather less than a fathom of water, and it is not the site of
coral growth, but is occupied mostly by sediment banks and
seaweed. The sea is admitted through a narrow gap on the
eastern side—and much has been made of the fact that this
Opening is on the windward shore; but a glance at the chart
will show that this is not in reality a lagoon entrance, and
that no gap in the barrier is to be found opposite the channel.

142 CORAL AND ATOLLS

The island is very densely covered with large timber trees,
mostly Pisonia inermis and Cordia subcordata ; while coconut

Fig. 48.

CHART OF KEELING ATOLLON.
(Modified from Guppy.)

palms and a wealth of undergrowth make the whole place a.
very beautiful one. Vast crowds of sea-birds (/regata aquila,
F. ariel, Sula sula, S. abbotti, S. piscatrix, &c. &e.) live on the

THE ATOLL AS A WHOLE 143

atollon. At the time of my visit, in June 1906, they were
breeding in thousands. The southern and eastern shores of
the island are steep, and truly marvellous seas crash on the
south-eastern extremity of the land. The western and north-
western shores of the island are composed of high sandy
beaches, and at the north-western extremity—the leeward
point—the barrier is deficient. The landing is a very hazar-
dous one, and without an island pilot, well used to the locality,
it would mean almost certain destruction, for the boat must be
taken in over the barrier of the western shore, on the crest of
a roller. At times, for very long spells together, no landing
may be effected. The history of the changes that have been
brought about in this very interesting atollon will have to be
further referred to; but 1t is well to say here that it shows
clearly an origin from at least two separate windward islands,
whose old outlet was towards the north and west. The
present gap which gives the sea entrance to the lagoon is the
remains of the interval between these two islands, and is not
the real lagoon inlet, nor comparable with the inlet into the
southern Cocos atoll’ The interest of the atollon lies in the
fact that it shows a further stage of development than does
the southern group, and it foreshadows the fate of atolls
generally,

CHAPTER XII

THE SUBMARINE BANK ON WHICH THE ATOLL
RESTS

THE submarine slopes, by which the edges of the atoll reef
descend to ocean depths, are perhaps the most important
features in the whole of the structure of the atoll.

Every theory that attempts an explanation of the formation
of atolls and reefs must be concerned with this decline from
wave-beaten barrier to deep-sea bottom. Some misconcep-
tion in the estimation of the gradient of this slope has
probably caused certain inaccuracies in the ideas as to atoll
formation. Since there is but very little reliance to be placed
on some of the older soundings of very deep water, the nature
of the ridge on which this atoll is situated has been always a
good deal obscured; and such concrete expressions of its slope
as have appeared in publications are shown by more recent
soundings to have been mostly fallacious. The statement that
the older soundings are often inaccurate is, I think, easy
enough to substantiate, for with old-fashioned paying-out
gear, errors were extremely difficult to avoid in very deep
water.

Ten miles of sounding-line have been paid out, and no
bottom found, in a place where the bottom was no more than
three miles distant (see Maury, &c.); and, in the neighbour-
hood of this atoll, the older soundings are often contradicted
by the more modern ones. The accuracy of the soundings of
Fitzroy around the atoll, in which he found no bottom at 1200
fathoms 6600 feet from the shore, has been questioned by
Admiral Sir W. L. Wharton (see Nature, June 19,1890). I
had not the apparatus, nor the opportunity, for making

soundings; and I do not regret the fact, for deep-sea soundings
144

THE SUBMARINE BANK 145

are not to be made by amateurs, and are impossible without a
well-found ship with modern gear. But I may make very
definite statements about the depths of the ocean, and the
submarine slopes of the atoll. Owing to the fact that I have
had the soundings, made by the various cable ships of the
Kastern Extension Telegraph Company, placed at my disposal
(due to the kindness of Mr. F. Hesse, the General Manager of
the Company), I am enabled to bring together a great mass of
accurate information concerning the nature of the bottom in
this part of the Indian Ocean.

These soundings consist of four different sets of observa-
tions. The first, and most valuable, are those made by the
cable ship Sherrard Osborne when first an oceanographic
survey was made in the region of the atoll, in order to test its
suitability to become a cable station. The second set consists
of a line of soundings, taken by the cable ship Anglia, to
the west and south of the atoll, along the cable route to
Rodriguez, and on to Mauritius. The third set was taken by
the cable ship Scotia, and the soundings follow the cable
route to the south and east to Perth, Australia. The fourth
set was made by the cable ship Magnet, when sounding over
the route taken by the cable north and east to the Sunda
Straits. These various collections of soundings therefore
embrace a very wide field of the ocean which has been brought
within our accurate knowledge, and they furnish very safe
ground for any assertions as to the contour of the ocean bed
and the atoll slopes.

It is of course unnecessary to say that there are many
reasons why these sounding lines cannot be reproduced, but
the general information derived from them is sufficiently clear
without the construction of a detailed chart of their actual
routes. The soundings are taken at frequent intervals, and
in the route from the atoll to Java there are 75 different
soundings over the 600—700-mile course; and from most
of these soundings I have examined specimens of the ocean
bottom brought up with the apparatus. or the use of these
specimens I have to thank the Telegraph Company and their

K

146 CORAL AND ATOLLS

captains, who have very kindly placed the material at my
disposal,

When we come to look at the published sections of this
atoll, the thing that is most obvious is the extraordinary
abruptness with which the reef land falls to great depths;
indeed it was thought that the landing of a cable in such a
place would be futile, for the idea is natural that it would
cross the edge of the reef abruptly, and hang into space for
perhaps a mile or more before it found any support on the
ocean bottom. To any one living on the atoll, this notion
very easily comes—that the side of the reef is sheer, and falls
to ocean depths with practically no break in its descent. I
know for certain that most.of those who come to the islands,
for a stay that averages a year, go away with the impression
that the barrier edge is the edge of a precipice of untold —
depth ;—and it certainly looks like it. |

When standing far out on the seaward edge of the barrier
reef an appearance is presented as though the great masses of
Porites, which show beneath the waves, are the extreme edge of
the land, and beyond them is the dark blue water, which, in
this ocean, tells of great depths. It seems as though at the
surf-line there was a real edge on which waves broke, and
which was the top of a cliff on which the atoll stood. From
any inconsiderable height that an island affords this appearance
may be seen to be unreal, and only apparent to the observer
when he stands at the barrier level. From any greater height,
the more gradual descent of the boulders can be seen beyond
the breakers; and if a boat be sailed outside the atoll ring,
any one may see the coral-clad hillside upon the crest of which
the atoll stands.

The barrier reef, which in its submerged portions consists
of living corals in large masses, is, for the most part, of light,
and often brilliant tints; yellow patches of coral and white
sand alternating with the darker colour of deeper holes, and
making a conspicuous bottom in the clear water of a few
fathoms depth. In from six to ten fathoms the rocks are plainly
seen, but beyond that depth the overlying waters become

THE SUBMARINE BANK 147

so deep in colour that the brilliant bottom is lost. In
places this transition is somewhat sudden, and soundings
taken a quarter of a mile from the shore show twenty-five
or thirty fathoms, whilst before half a mile is reached,
towards the south and south-west, seventy and one hundred
and ninety fathoms are recorded. It will be easily imagined
that the sudden ending of the brightly coloured barrier rocks
in the deep blue water outside gives the impression that the
visible shore on which you live falls suddenly into the invisible
depths of the ocean.

This is, however, not the case, for it is only into these
trivial depths that the fall is abrupt; it is not a fall from
shallows to great depths; and in many cases the varied
coloured bottom of the boulder-strewn reef. may be seen
several hundred yards to seaward, gradually losing itself in
the deepening blue waters.

Soundings taken a mile from the shore show a depth
varying from one hundred and ninety to three hundred and
thirty fathoms. On the route towards Perth, Australia, six
fathoms were found at 0°95 miles from the atoll and at 1°45
miles the bottom had fallen to two hundred fathoms. On the
routetowards Rodriguez thirty fathomswere found at 0°90 miles,
and two hundred at 1°05 miles. This then is the sudden
drop, the descent within the limits of a mile or so into some
two hundred fathoms. Two and a half miles from the atoll
ring the bottom has fallen to a depth of from four hundred and
fifty to five hundred and fifteen, or, roughly, an average of
five hundred fathoms at two and a half miles. At five miles
a depth of seven hundred and fifty fathoms is registered, but
on the cable route to Java the five-mile limit gives a sounding
of a thousand fathoms. In other directions one thousand
fathoms are not found till six miles away—the old sounding
of twelve hundred fathoms a mile and a quarter from the south-
east shore being, perhaps, safely neglected. Upon the actual
cable route to Australia, the bottom has only fallen to one
thousand six hundred and thirty fathoms at a distance of
63:71 miles from the atoll.

148 CORAL AND ATOLLS

Two thousand fathoms depth is found at an average of ten
miles from shore, though fifteen miles are passed towards Java,
and one hundred towards Australia, before this depth is re-
corded.. Two thousand five hundred fathoms occur on an
average at fifty miles, and an average of one hundred and
eighty miles are passed before three thousand fathoms, or
ocean depths, are reached.

Upon the actual cable lines the three thousand fathoms limit
is reached at very varying distances. Towards Java, the ocean
depth of 2950 fathoms is not met with before seventy miles
are passed, whilst towards Rodriguez one thousand miles of
the cable lie in depths less than 2910 fathoms, and the
actual cable never reaches a depth of three thousand fathoms.
Towards Australia, two hundred and fifty miles are passed, on
the cable route, before a depth of 2960 fathoms is reached ;
and four hundred and thirty before the bottom falls to 3130
fathoms. The greatest depth sounded towards Australia was
3500 fathoms, in Maclear deep, at a distance of four hundred
and fifty miles from the atoll; towards Rodriguez 3300 fathoms
were met at two hundred and fifty miles from shore, and this
depth shoaled again, at another two hundred and sixty miles
farther on, into 1000 fathoms. Towards Java the bottom
falls away to 3535 fathoms, but 3000 fathoms is the average
depth of the trough which separates the atoll from Java. The
Cocos atoll is, therefore, the summit of a great roll of the
ocean floor which rises gradually, from the ocean depths of 3000
fathoms, at an average distance of something over a hundred
miles.

Unfortunately, we do not know the soundings along the
line that connects Cocos ~-atoll to Christmas Island, but it
may be that the ridge, from which that ancient atoll has been
hoisted to a height of a thousand feet, is continuous with the
ridge upon which the Cocos-Keeling group takes origin.
Again, we have no very definite data of the condition of the
ocean bottom to the north of the southern atoll, but it seems
certain that a submarine ridge streches for fifteen miles to
Keeling atollon; and it is said that this ridge is nowhere

THE SUBMARINE BANK 149

deeper than a hundred fathoms ; indeed in calm weather it is
said that the bottom may be seen in sailing between the two
atolls.

To the north of Keeling the ridge probably stretches to Glen-
dinning Shoal—a reef of which I have no definite knowledge,
but I have been told by one of the captains of the Clunies-
Ross schooners that he has often seen the discoloration of
shallow water when on some of his numerous runs to Batavia.

It would be of great interest to sound over this northerly
continuation of the ridge to determine its limits and its
characters.

Soundings directly to the south of the southern oll are
also rather unsatisfactory, and the southern limits of the
submarine ridge are quite uncertain; but it seems likely that
the southern slope is more abrupt than that in other
directions.

Along the sounding-line to Java, I have examined samples
of the bottom brought up with the sounding-gear, and these
samples, on the whole, confirm the state of things shown in
the map of ocean bottom furnished in the Challenger re-
ports, where the atoll is shown to stand in the midst of an
area of Radiolarian ooze.

I have not examined the samples obtained by the sounding
expeditions in the other directions, but the naked eye appear-
ance of the specimens is described by the staff of the cable
ships as being successively “coral fragments,” “coral sand,”
“fine sand,’ “sand and mud”; and then, from a hundred
miles from shore and onwards, “brown mud” was the. un-
varying nature of the bottom. In the neighbourhood of the
atoll, coral fragments and pieces of Nulliporw, with a coating
of a whitish clay, composed of Globigerina and Orbulina
remains, constitute the commonest sample. Adown the sides
of the submarine slope, to a depth of over 2000 fathoms, the
remains of organisms with calcareous skeletons far outnumber
the siliceous organisms, and Globigerina and Orbulina are the
predominant forms. This is the “fine sand” and “sand and
mud”; its colour is greyish, and it has a distinct granular

150 CORAL AND ATOLLS

appearance. The “brown mud” is made up mostly of the
siliceous remains of Radiolaria, &c., mixed freely with a
brownish unorganised deposit.

From all samples there was a fairly large proportion of
this unorganised débris, which consisted, in great part, of
particles of pumice and irregular mineral crystalline forms,
and with the deposit, from greater depths, a large admixture
of the spicules of sponges was always found. From the limit
of the drop of the reef, therefore, the foundation of the atoll
would seem to be a great sedimentation slope, and certainly
not a steep peak, rising suddenly from ocean depths to a wave-
washed reef. The slope is certainly nothing like the picture
that Guppy—its last investigator—-made of it, for he said in
a letter to Sir John Murray, published in Nature, p. 237,
1889: “Here the submarine slope slopes gradually down to —
20 or 30 fathoms, but beyond this the descent is precipitous.”
The steep descent comes first, the gradual slope comes after.
We may take the slope from the reef drop, to 2000 fathoms
—that is to a distance of ten miles from the atoll—at an
almost uniform gradient of one in five. From ten to fifty
miles the slope is about one in eighty, and from that point
onwards, at about one in a hundred, to the depths of the
ocean. In other words, the Globigerina ooze slope is one in
five; the Radiolaria ooze slope is about one in a hundred.
When a section of this great ridge of ocean floor, with its flat
top of reef, is made to scale, it is at once seen that a very
wrong importance has been ascribed to certain features of an
atoll. The islands are trivial things, the lagoon merely a
slightly submerged reef, and the barrier an inconsiderable
ring around its edge; it is only by a proper understanding of
the relative proportions of these parts that a right interpreta-
tion of their origin may be arrived at.

CHAPTER XIII
THE STRUCTURE OF THE BARRIER

For the purposes of description alone, it is necessary to give
different names to different portions of the rock ledge that
forms the outskirts of the table-land of the summit of this
submarine ridge.

In the giving of many names to portions of a whole, much
confusion may be caused, and a loss of the clear idea of the
whole may be suffered in searching for the origins of the
parts.

By the term “ Barrier” is generally understood the ledge
of the rock that stretches from the seaward margin of the
land to the edge of breaking surf; and it is certainly fittingly
named. But that ledge of rock is only a portion of a much
larger bed, which forms the basis of the islands and even the
rocks of the lagoon shore. In description some subdivision is
necessary, but it must be understood that the several parts
are In continuity, have been formed by the same agencies,
have fulfilled the same functions, and constitute one homo-
geneous whole. The ledge of rock from the surf-breaking
edge to the rising land of the island is conspicuous, its more or
less level surface becomes exposed in varying degrees with the
movements of the tides, and its continuity is unbroken to the
first rise of the land. Here, at the seaward beach, its structure
becomes more or less confused; loose fragments of broken
coral boulders lie scattered on it, roughly strewn towards the
sea, piled by the waves towards the land, until the piling up
has become so considerable that the level which constitutes the
island dry land has been attained. But the barrier properly
speaking does not end here, for under the piled-up fragments

of its landward margin its solid level stratum can be readily
151

HIST TL Oe
SoM F] eTeog “UoTONAYsUOD sq UT posn o40M sSutpunos s Aoazqtyy

(6 “ON *4 ‘Toa “nbz *f0a 4099)

“‘kddad ‘GF ‘HUG OL
DNIGYONOV SAdOTY ANIMVNANAG GNV TIOLY AHL FO NOILOag

1g DIT

"1[0}8 949 Jo opis 1eyjz10 uodn
SOTIUL U9} FO SOURISIP B TOF poMOT[OF SuIog sedojs ouTIeUIQNS OYJ,

‘TTOLY DYNITAAY-SO00D
GHL GNV aodly QNIYVNANG AHL AO NOILOWS ASUAASNVA], V

‘0G “SIA

(61 ‘d ,“sfooyy [e109 ,,)
“NIMUVG OL YNIGLOONINV CCRC ORR AGS § aNV CGNV'IST HHL wo NOILOGS COSI

"JooI oY} FO apis PIVMpPUIM 9} FO
odoys snye} oY} 07 uoYRIEd oTquqoad sy pux o10q oY} Jo o718 24} MOYS OF,
beer r
adojs MOLY LLOdVNOY 10 NOLOGY IVOILANOAH,
auliewgqns
quUsseig

‘atog jo ujdeq

"EG ‘OLA

210g jo UOTIISOg

‘espe premees oy} uo dn ysvo ‘UMOYS SI Lepfnoq Sie] ouo pure “Tel11eq oy} UO YSeMe st
op oy, “pues pue syuewseaz [e090 Zo opid purjst oy, me *(yoR]q) Wo; VI Biooorg oY, “I : MOYS OF,

‘NOILOGG 'IVOILANOMH,

fated ! SP a 125%) x x Ce
Qe
es

6G OLA

154 CORAL AND ATOLLS

found. Under the accumulation of coral boulders, sand, and
vegetable débris, which forms the island itself, it still continues
unbroken; and on the lagoon side of the island again, in many
places, it juts out as extensive slabs of rock, breaking the sand
stretches which form the greater part of the lagoon shore.

This is all barrier, all one continuous layer of rock, all
formed in the same fashion; and the function which the sea-
ward edge now serves has in times past been carried on by
every portion of the layer. Although the term “ Barrier,” as
generally used, cannot perhaps be properly applied to this
entire rock stratum, still it is only by considering it as a whole
that a right understanding of its origin can be arrived at. To-
day the seaward part alone acts as a functional barrier proper ;
in the geological to-morrow it may be buried under the land
accumulation which it has helped to form; and in the geological —
yesterday that cemented layer of coral fragments which is
found everywhere under the islands was the functional agent
in resisting the waves, and in forming the dry land. Since
the seaward portion alone can be called the “ Barrier,’ and
since the whole layer will have to be referred to again, we
will call it the Breccta PLAtForM, for it is a mosaic inlay of
coral fragments, cemented together into a solid platform,
stretching from surf-line to lagoon.

The barrier is the pioneer of the island, it is dry land
potentially, and on the varying influences of wind and waves
depends the disposition of the land for which the pioneer has
laid the foundation. The dry land that the barrier has called
into existence is in reality broken pieces of barrier, for the
entire atoll is a débris formation, and the débris is recruited
from the seaward edge of the barrier for the most part.

Imagine for a moment all this débris to be removed: no
loose fragments of coral, no sand, no vegetation piled up. All
that would remain would be a flat and roughly level ring of
coral rock, firmly cemented, and making a broad and flat
margin to the lagoon; only the Breccia Platform would be
left in fact. This margin would be mostly awash; the
waves would break on its seaward edge and the ocean swell

THE STRUCTURE OF THE BARRIER 155

rush over into the lagoon, exactly as it does to-day where
the gaps in the dry land ring let the ocean and the lagoon
communicate.

The Breccia Platform is, therefore, to be seen in its simplest
form in the gaps in the atoll ring, in the places where no
islands exist, and here it is usually called “ Barrier” in its
whole extent. In such places as this it may be easily examined
at low tide, for, on a calm day, one can walk from one island
to the next through the water which washes the rock flat. At
the seaward edge of the flat is the line of surf, and the
contrast between the blue rollers on the one side and the surf-
washed rocks on the other is very marked; it is this contrast
which gives the impression that the barrier edge is like the side
’ of a house against which waves beat. Standing on the barrier,
as near to the breakers as is possible, an appearance is pre-
sented as though at a line a few yards from you the deep
ocean ended, and its blue rollers hit on a sharp upstanding
edge of rock that caught their feet and scattered them in spin-
drift among the rocks on which you are standing. But this
appearance is really a delusion, and on a very calm day, even
from the level of the barrier, discoloured patches of bottom
can be seen through and reflected in the rollers before they
break ; and from such trivial heights as it is possible to attain
on a coral island, these patches are seen to stretch a long way
to seaward of the breaker line. Moreover with the rise and
fall of the tide this surf-line moves, coming nearer to and
retreating further from the land; and no matter where the
line be at the moment, the appearance presented is always the
same—a sheer sharp edge on which deep ocean breaks.

The falling away of the rock ledge into the ocean is not in
reality such a sudden fall as has been imagined, and its falling
to the lagoon is very much more gradual. The barrier is then
a wide rock flat, falling to the ocean at its outer edge, shelving
to the lagoon at its inner edge, and extending almost entirely
round the summit of the ridge.

There can be no doubt left in the mind of any one who
watches the great ocean waves broken at the edge, and swirl

156 CORAL AND ATOLLS

and boil across the flats, only to fall back in broken confusion
to the hindrance of the next oncoming roller, that there is
something in the moulding of the barrier reef itself that makes
it so successful a resister of the ocean’s terrible and ever-acting
force. No wall made by man could ever hope to withstand
for long the perpetual bombardment of those great blue rollers
of the Trade-driven Indian Ocean. The great secret of the
success of the barrier reef is to be found in the peculiar
curvature of its wave-swept surface. From the uneven edge
on which large rocks are hurled and firmly seated, and on
which the first force of the waves is spent, the barrier runs
shorewards as a smooth Nullipore-covered platform which has
a slight curve upwards. The barrier is a flattened arch, and the
peculiar pitch of this arch is suited to the most efficient,
spending of the force of the waves. The waves that the
boulder-guarded edge has broken become further spent in
sweeping across the convex seaward margin, and finally ex-
hausted by the uneven flats which line the shore; whilst the
backwash from the spent waves seethes among the flats, and
pours backwards with increasing force as the outer slope is
reached. Every broken wave helps to be the undoing of its
successor.

The only places in the atoll ring in which this rock flat is
wanting are on the north-west, or leeward, side of the group;
and here, although no breaking surf is present, and no sea-
washed rock flats stretch outwards, the same coral boulders
lie at a somewhat greater depth uninfluenced by the waves.

The flats of the barrier are entirely composed of coral,
living or dead; and the coral that is living is a very small
fraction of the whole. The dead shells of molluscs, the cast
ectoskeletons of crustacea, the hard parts of echinoderms and
other marine animals, are too unimportant a contribution to
invalidate the statement that the barrier is entirely the pro-
duct of corals.

In certain parts of the island ring it is true that the
encrusting MVulliporw constitute a very important part of the
barrier. This Nullipora growth appears to be a very varying

THE STRUCTURE OF THE BARRIER 157

factor in atoll formation. Darwin described the Nullipora
bed in this atoll,* and Dr. Guppy also called attention to the
unusual thickness of its growth.f In describing the cha-
racters of the coral reefs of the Solomon Islands, Dr. Guppy {|
came to the conclusion that Nullipore formed only thin layers
that were unimportant in barrier building, but three years
afterwards, when he came to describe the Cocos group, he
mentions the slabs two feet thick which are commonly to be
found broken from the surface of the outlying barrier boulders.
It would appear that in these islands the Nulliporw play a
more than ordinary part, but as their presence is not universal
at all parts of the atoll ring, it is evidently not at all essential
to successful barrier building. On the other hand, the Litho-
thamnionide are particularly insignificant in the atoll, and
they cannot possibly be of the importance that, in the Mal-
dives and Laccadives, Mr. Stanley Gardiner § supposed them
to be. ‘They are found only in certain sites, far from the
barrier edge, and the total bulk of all their remains cannot
make a very great difference to the sum of the atoll’s building
assets. .

From ocean to lagoon the character of the coral flat varies,
and the.changes that are seen are natural and easily explained
ones ; they throw a considerable light on the formation of the
‘barrier, and of the whole structure of a coral island, or an atoll.

In the first place, the various forms of corals have a fairly
‘defined site of election; but for all that their distribution is
eapricious, and is only roughly to be argued for any given
spot of barrier from the knowledge gained at any other
spot. From reading of accounts of the corals of atolls, the
impression is easily gained that a remarkable order in their
distribution exists; the corals of the reef and those of the

* “The Structure and Formation of Coral Reefs,” p. 24.

+ Scot. Geog. Maga., vol. v. No. 9, p. 460.

¢ “Characters and Mode of Formation of the Coral Reefs of the
Solomon Islands,” Proc. Roy. Soc. Edinburgh, July 1886, p. 859.

§ “Fauna and Geography of the Maldive and Laccadive ee
1902,” vol. i. part 2, p..175. Fe

158 CORAL AND ATOLLS

lagoon are described, as well as those of all the intermediate
situations ; and the idea remains that a species is either a reef
coral or a lagoon coral. Now it is as well to say at once that
the distribution of the corals of this atoll does not follow any
such hard and fast lines. There are several species which are
always more abundant on the barrier, but they are also to be
found in the lagoon, just as are the more typical lagoon
corals to be found on the barrier. The essential feature in
the distribution of atoll corals is that it is a distribution of
types, and not of species; the barrier types are not species but
are the barrier forms of lagoon species, and vice versa. The
subject of the modification of the corals is an interesting one,
and must force itself on the collector as he gains experience ;
for as he first makes his collections from the lagoon and the
barrier he begins to believe that the number of species is
without end, and it is not till he has spent some time in this —
pursuit that he comes to realise that the species are in reality
few, but the adaptations of the corals to their environment
are legion. The student of corals who should see these col-
lections only when far from their site of growth would be
entirely without a clue to account for the endless range of
forms which any casual visitor to an atoll might collect. There
are many factors which tend to modify the outward form of a
coral’s growth, and to confuse any one who, without a know-
ledge of their habitat, would attempt their classification. This
point has been studied in connection with the corals them-
selves. When distribution is spoken of here, it is this dis-
tribution of types that is meant, a very different thing from a
distribution of separable species.

It has been said, and much weight has been given to the
assertion, that the nearer to the edge of the barrier one goes,
the more luxuriant becomes the coralgrowth. In this atoll, at
any rate, this is definitely not the case; and any arguments
for the formation of the atoll based on this assumption cannot
be valid. It is not necessary to assume that corals flourish best
in the surf-line because of the better aeration, or that they
receive a more abundant supply of food, and so grow better; for

THE STRUCTURE OF THE BARRIER 159

the statement that these assumptions would explain is a
statement that facts do not justify, and the assumption would
appear to be far from true. It is true that some corals pre-
dominate on the seaward edge for the very reason that, as the
area most commonly exposed to the surf is reached, many
species that cannot withstand the rough usage are left behind,
and the few dominate the picture. In that area on which the
surf beats—the seaward edge of the barrier—-living coral
erowth reaches its minimum of individuals as well as of species.
The most luxuriantly growing of all the corals are not even
represented far out on the barrier, for the simple reason that
they could not withstand the conditions which obtain there.
Towards the lagoon the corals increase in number and
variety, and in the lagoon itself are some of the most
luxuriant of all the growths, branching and laminated forms
whose habit of growth does not permit of a successful struggle
for existence in the perpetual tumble of rough water.

It must be borne in mind that these remarks apply only
to the rock flat—-the simple barrier in its whole extent—in
those gaps of the atoll ring in which no islands are formed.
_ Where an island is present, a disturbing factor comes in, for the
island is a piling up of débris which rises from the barrier. The
nearer the island therefore, the more the barrier becomes ex-
posed at low tide; and although it is by no means true to say
that corals cannot withstand an exposure of some duration to
the sun and air, still they do not flourish best on any part of
the barrier normally exposed at low tide. It is true, therefore, to
a certain extent, to say that the farther one goes seaward on
the barrier from an island, the more luxurious becomes the
growth of coral; but this is an after-effect of the formation of
the island, and must not be used as an argument in a theory
to account for its formation.

I have stated that there is a roughly defined distribution
of the types of corals on the barrier, and this distribution is
one that is easily accounted for. The conditions of life on the
seaward edge of the barrier are peculiar and severe ; the com-
petition for life takes place in a perpetual crash and rush of

160 CORAL AND ATOLLS

surf, and more than this, in an environment where, ever and
again, loose fragments come hurtling along at the mercy of the
waves. These conditions completely eliminate any delicately
branching forms of coral, and only those which grow as rounded
knobs, as stout short branches or plates, or as encrusting
layers can hope to flourish. To live at all in that tumult of
waves and moving masses of rock, only those things which are
very hard and rounded, or are very soft and yielding, can exist
free, all else must hide in crevices or creep in burrows. Of
the very soft things which can stand these severe conditions
are many of the Alcyonacew, which live in great abundance in
depressions of the seaward edge of the barrier, and of the hard
and rounded things are those most typical of all the barrier
corals, the massive Porites.

An “outer edge” of the barrier has been spoken of, but
the use of this phrase by no means implies that here the
barrier ends. The submarine slope shelves away from the
land ; its line of retreat being the barrier flats, and on the
strength of the waves and the amount of the tide variation
will depend the slope of these flats. At dead low tide the
surf-line has receded from the land until it has travelled a
hundred yards or so, and has fallen by three or four feet from
its high-water mark. At this low tide surf-line therefore will
be the junction of that portion of the coral bottom which is
subjected to the action of the surf, and the portion which lies
undisturbed below. Storms, and the action of high waves,
have at intervals displaced large rocks, and have then been
powerless to trundle them up the flats; and so the “ outer
edge” of the barrier is in many places marked by the ex-
posure of large and irregular rock masses at low tide. It
must always be borne in mind in accounting for the disposition
of the different parts of the barrier, that there is a perpetual
force acting over a wide area, but the intensity of this force varies
at different points of its activity; and its action is inconstant
throughout. At the seaward edge its lifting force is greatest,
for here the surge and swell of larger waves is felt ; it is here,
therefore; that the largest rock masses are moved. But again,

THE STRUCTURE OF THE BARRIER 161

its time of action is short, for only at dead low tide can it
exert its full force; and so it comes about that large masses
are displaced and frequently reseated on the “ outer edge” of
the barrier. It is over the intermediate part of the barrier
that the maximum force is felt, for the crashing surf has the

Fie. 54.

DRAWING OF A LIVING SPECIMEN OF AN ALCYONARIAN CORAL
WHICH IS ABUNDANT IN THE ROUGH WATER OF THE BARRIER.

Two Zooids are seen completely expanded and several others in the
passive stage between contraction and expansion.

power to act during the rise and during the fall, and at all
times, except at the very lowest spring tides, is this inter-
mediate portion a battle-ground of surf and rock. Here,
therefore, are the flats of level cemented rock, free of loose
fragments, free of jagged edges, and free too of living corals;
boulders are washed across and tend to level everything, and
only the Nullipore may flourish. It is here that the finer
particles are forced by the spin-drift, as by a sand-blast, into
every crack and rift of the shore-washed rocks; this is the
factory of the breccia. At the inner or shoreward edge of
the barrier the wave action is powerful during high tide, and
intermittent during the rise and fall. It is here that the
smaller boulders come to rest; this is the factory of the
débris which makes the island. It is easy then to see that on
the barrier, as it stretches from an island to the ocean, actual

living coral growth is not abundant. It is only in pools and
L

162 CORAL AND ATOLLS

under the shelter of rocks that corals can retain a foothold at
all, and the picture of the active corals flourishing in the crash
and swirl of the breakers is not a true one. We have seen
that, as the barrier in between the islands is followed into the
lagoon, the coral growth becomes more abundant; and again
in the ocean, outside the surf-beaten edge, the active living
coral growth increases, and in the six fathoms or so in which
it is possible to see them clearly, large masses surpassing
anything on the barrier, or in the lagoon, flourish on the
submarine slopes.

Although living coral forms so small a portion of the
actual barrier, still there is no part of the barrier that is not
made of coral, but it is the dead and altered remains of past
generations. The idea is common regarding the building of
coral islands, that the coral grows and grows, and as it dies it
forms the basis of the island; that in fact the coral is ever
growing up, and the calcified part in time reaches to such a
level that dry land is formed in the midst of the ocean. This
of course does not, and cannot, happen. The barrier has not
grown up by the growth of coral im sitw, and the island has
not come into existence through the direct activity of coral
growth. Even those few living corals on the barrier are for
the most part waifs and strays, living where they do because
they have been washed there, and, as is their wont, have
fixed themselves to the first safe anchorage that comes to
hand. In the coral beds which lie on the submarine slopes of
the atoll ridge large masses of Porites, and other forms, are
ever growing in great profusion, and the rock is being slowly
built up which will one day, during storm, be moved from its
bed, lifted by the swell, and hurled shorewards by the surf.
Many large masses are no more than lifted from their beds, and
left to stand as outguards of the barrier, their sides becoming the
site of further coral growth so that in time they become
cemented in their place by the growth of the hardier species
of barrier corals. Smaller masses are trundled farther toward
the shore, and a belt of coral wreckage is deposited from the
low-water surf-line of spring tides, to the high-water line of

Adit MOT LY SAMMI OTOd FO SHHOVAd GUVMVAS GNV SLYIA ULINIVA WAL

‘A MLV Td

THE STRUCTURE OF THE BARRIER 163

storms. It is this belt of wreckage, which happens to be
composed of coral, that forms the barrier of coral islands.
Now this boulder-strewn belt of shore varies in different parts
of its breadth ; to the ocean edge are the large masses and
deep pools with their sheltered and living corals, and inside
this is the broad smooth belt formed by the perpetual] sea
wear.

Every block on this belt becomes firmly embedded in the
platform, and cemented fast to its neighbour, and the whole
surface is levelled by the waves and the fragments which they
carry. Coral sand, formed in various ways, is driven into
every chink, and a fine breccia of dead and altered coral is
formed. The breccia becomes for the most part covered with
living Nullipore, but it is not, as a rule, the site of any living
coral growth. The landward edge remains loose, for there the
fragments are trundled beyond the reach of ordinary waves,
and on the breccia which has been formed in the past the
piles of rounded sea-worn coral fragments lie; and of such
fragments the coral island is formed. The structure of the
barrier reef varies in different parts of the atoll ring, and it is
different on the leeward and windward sides, as might be
imagined. On the windward side, in such places as the
neighbourhood of Pulu Atas, the barrier reefs are of great
width, and stretch lagoonwards as long sand flats, which
extend in between the islets as spits and shallows, far out
into the lagoon. The smaller particles are swept on the
farthest, and the fine sand formed from disintegrated corals
and shells makes a gradually extending silt which has filled
the ancient lagoonlets of the windward islands, and is gradually
shallowing the windward side of the lagoon itself. Here too
the barrier rock is evidently the most changed by extreme age,
and the coral forming the breccia is often so altered that, in the
finely tessellated fracture of the windward barrier, its original
structure is often difficult to recognise. Amongst the other
factors which tend to modify the structure of the barrier reef
are the entrance and exit places of the currents of the lagoon,
and remarkable changes take place in the more variable

164 CORAL AND ATOLLS

portions from time to time. During the early months of the
year, at the eastern extremity of Pulu Tikus, a spit composed
of rock fragments and sand, will form and extend far out into
the lagoon, only to wane gradually during the next few
months, with the deposition of sandy beaches at the western
end of the island.

The barrier is the home of a myriad forms of life, but
only those need be mentioned here which have an influence
on its architecture, and so on the formation of the atoll.
Fish in thousands live in its pools, and come and go with the
rise and fall of the surf-line; and one of the strangest and
most beautiful sights of the barrier is the crowd of brightly
coloured fish seen mirrored in every oncoming wave. As you
stand on the coral beach it seems that every wave must wash
ashore these wonderful fish, and yet, as the sea sucks back
and the foam clears, they are still quietly feeding in the clear
water where there is only just enough depth for them to
swim. Many of these fish are amongst the most important
agents in the making of the coral sand which forms the cement
of the island and the silt of the lagoon, and the most active
of them all are the beautiful Kakatua (Scarus, sp.), the most
conspicuous of the barrier fish. Darwin and Forbes, during
their visits to the islands, watched these fish, and asserted ©
that they fed on the living coral; but this is certainly not the
case. As these bright green, or cobalt-blue, Kakatua go
slowly along head downwards, and with a frequent flapping of
their bright tails above the water as the wash flows back, they
seem to be grazing among the corals as beasts among herbage ;
and if their stomachs are opened they are found to contain
coral ground down to a very fine sand. But the coral from
which this sand is made is dead, and has long been dead; for
it is the coral of the cemented breccia, and it is only scraped
off by the fishes’ hard beaks as they rasp the boulders for
their covering film of alge.

If the feeding-spot of the bright fish be noted, and the
boulders over which they have fed be examined, it will be
seen that on every rock is the characteristic bite where the

THE STRUCTURE OF THE BARRIER 165

alga has been scraped from its hold, but any living coral will
be found quite untouched. The part played by these fish in
the making of sand must be considerable, for every rock of
the barrier is scored by their hard beaks; and from ten to
fifty or more will be in each of the many shoals which pass

Fic. 55.

THe Ikan Kakatua Ijou (Scarus),

Showing the hard beaks with which they score the coral boulders.

along the surf-line with the turning tide. As the average
weight of dry sand in an individual's intestinal canal is nearly
two ounces, the bulk of coral rock converted into sand by
these creatures must be very great indeed. Other animal
agents in the production of sand are the Holothurians, the
sea-slugs or béche de mer, of many species that lie scattered
about all over the barrier and lagoon in countless thousands.
They are from three feet long to less than an inch or so, and
they are merely hollow animal tubes which act chiefly as sand-
filters. They fulfil on the barrier and in the lagoon very
much the same functions that the earth-worms do on land,
for they pass through their bodies large quantities of dead
material along with their food. The sand is taken in at the»

166 CORAL AND ATOLLS

hard mouth, triturated and separated from its living mhabi-
tants in the long alimentary canal, and ejected as a fine white
mud freed from animal and vegetable particles. The wash of
the waters must in a great measure mask any influence these

Fig. 56.

PHOTOGRAPH OF HOLOTHURIANS FEEDING AT THE Borrom
or A BARRIER POOL.

The photograph was taken through some two feet of water, without
disturbing the Holothurians in any way.

animals have, in common with the earth-worms, on the altera-
tion of level or position of the free-lying boulders, but their
handiwork must be of some importance by reason of their
very numbers. Countless worms bore the rocks and make
tunnels on their surface ; numberless molluscs live in the nooks
and corners of every boulder, and when they die their shells
are added to the sum of débris, but as workmen in the building
of the barrier their réle is an unimportant one. In addition
to the animal agencies in sand-making, is, of course, the very

YPNIGTING TIOLV JO SASVHd
AGTVAOI NEL CAL LO CNO—-SWITVd LONOOOD TO ONININWUAOND WH GNV VOUS TH LO LNUNWHOVOWONT
WHE SMOHS HdVADOLOHd SING SOWML WINd “AGL CHE TO TIVE MH AT LIT STOOd WHIMUVva

TA DLV Id

Pike! wes
hee se

alo} ea ek cite” Se aad

x Ve wn
<0 eee ny Tok S
; ; rs F
= fe ;
. ‘ Se. he ¥
, 4 th
> e
a 5
f A

% % H

( :

5 ‘
5
t vere oh 5
i ‘
, P
j r
t
i
. 1% 4 y

THE STRUCTURE OF THE BARRIER 167

much more important one of the grinding movements of the
water. This is one of the greatest forces with which we have
to deal in estimating the physical surroundings of the atoll,
for it is carried on without ceasing. Every wave moves some
fragment and triturates some particle, always tending to make
finer and finer the broken pieces of coral which are for ever
being removed from the coral beds beneath the surf-line.
The branches and boulders become chips and fragments, and
these in their turn are washed to and fro, becoming smaller
and smaller in the process, and a diminishing scale of coral
material is created, in which the first term is the growing
boulder and the last is the fine white mud.

The great quantities of sand formed in a coral atoll may
be realised when the silt banks are watched in their growth,
and when it is remembered that not a tithe of the sand ever
reaches the lagoon, but lies on the submarine slopes of the
ocean bed for a distance of several miles all round the atoll.

The structure of the barrier reef is then in reality a simple
one, it is a great platform built round the top of a large flat
plateau, and, as a rampart, it encircles the summit for a dis-
tance of some seventeen miles.

CHAPTER XIV
THE ISLAND BEACHES

(A) THE SEAWARD BEACHES.

As the barrier varies in different parts of the atoll ring, so does
the beach which rises from it, and so too do the trivial elevations
that may be dignified by the name of the “coast-line” of the
islands. This is but the natural outcome of the fact that
the different aspects of the circle are exposed to different
physical conditions. The physical forces that build the dry
land are not uniformly distributed around the atoll, and the
natural consequence is that the disposition of the land varies
as its creative forces vary. Were we to have an accurate
model of any atoll, we could tell from its examination which
was its windward side and which its leeward, which were the
islands exposed to the ocean’s currents, and which were those
protected from their scour. Since the building processes at
work on the seaward side of the islands are different from those
which shape the lagoon shore, it is but natural that the sea
beaches and the lagoon beaches should differ in their nature.
The ocean beaches consist essentially of a sloping pile of coral
débris piled up on the breccia stratum as a basis; they form
the merging-line of the island dry land and the barrier
flats.

On the ocean side there are places where no beaches can
be said to exist, for the island rises abruptly from the barrier
flats. In such places a succession of miniature terraces,
two or three in number, rise as steps of consolidated breccia
and bear on their top step the steeply piled up boulders that
make the shore-line of the island. In other places there has

been a denudation of the island land, and the tide-washed flats,
168

THE ISLAND BEACHES 169

in which the roots of former generations of coco palms may be
found, merge gradually into the island débris, there being no
heaping up of boulders to resist the sea’s encroachment ; and
here, with high tides and a strong wind, the sea will rush over
the island in an alarming manner.

Although the senna beach is usually made up of chips
and fragments of coral, still, in some places, both the lagoon
beach, and the sea beach may be composed of fine coral Sandi
Where the island is a pure sand formation it is generally
sheltered on its seaward side by the accumulation of large
boulders. Within the shelter formed by pieces of detached
barrier, and on a firmly cemented rock basis, is piled fine sand
and small débris which gradually slopes from the seaward edge
upwards to the lagoon shore. Such a structure is the beach
of Pulu Bras and of Pulu Pasir. No large pieces of coral
enter into the formation, and on either side of that portion of
barrier on which they have their basis, sand spits run out
towards the lagoon. The seaward beaches are seen in their
most perfect form to the windward of the atoll, and at the
southern point of Pulu Atas, and at the south-east aspect of
Keeling atollon, the beaches rise steeply piled to a height of
twenty feet or more. No better idea of the wonderful force of
the Trade-driven waves can be gathered than by studying these
piles of massive boulders tossed up by storms. In these
windward beaches there is very little sand, save that piled by
the wind beyond the reach of the ordinary seas.

Of whatever formation the beach may be, it is its chief
characteristic that, in its minor details, it is perpetually
changing with the cycle of months and years; and a daily
walk round an island, during a year’s progress, will note per-
petual alteration in the disposition of sand and rock, in the
distribution of silt bank and channel; each changing always
as the currents around the atoll wax and wane locally. To-day
a boulder-piled rise will merge to the sea by a sandy beach
and form a temporary spit, on which the cast ectoskeletons of
crustacea, the dead echinoderms, shells of all sorts, and alge,
will come ashore in profusion; and in a few weeks’ time it will

170 CORAL AND ATOLLS

be gone, and the breccia platform will be laid bare to the rise
of coral boulders.

The nature of the beach depends entirely on the force of
the sea which usually beats upon it. Where the waves are
normally high, and the shore is exposed to the full force of
the Trade-driven swell, the surf will rush over the submerged
flats at full tide, and exert its whole energy on the boulders of
the beach. It is in such places that the island rise is abrupt
and the beach is made up of great masses piled one upon the
other. All the sand, and the smaller débris, is washed away,
only to be deposited on any part of the shore where a normally
calmer state of things prevails. The character of the beach is,
therefore, a map, which, when studied aright, gives up its story
of the effects of the currents and winds, and the negative or
positive movements to which the shore platforms have been —
in the past subjected; but the temporary piling of sand or its
temporary removal must not be taken as evidences of actual
changes of level, or of constant action. On the seaward
beaches is thrown the flotsam and jetsam that reaches the
group from the outside world, and one of the principal items,
—one that has in many places caused a considerable alteration
in the character of the islands—is pumice. The greater part of
the pumice found in the group arrived after the eruption of
Krakatua: being washed up in 1883 in vast quantities. This
pumice, lightest of all the wrack that the sea has piled up, has
been carried for varying distances into the island from the
seaward beach, and shows, as an index, the limit of surf action in
the island building that has been reached in twenty-three years.
It occurs in great quantities as rounded sea-worn masses, some
being a foot and more in their long axis, but the majority
varying from the size of marbles to that of cricket balls.
Besides the Krakatua pumice, which lies to-day mostly on the
seaward beaches, and for a few paces into the island itself,
there is older pumice which may be found almost anywhere in
the breadth of the dry land. Pumice has been arriving from
somewhere ever since the first appearance of land in the atoll
ring, and has, during the period of its stay, undergone much

THE ISLAND BEACHES 171

decomposition. The Krakatua pumice is almost uniformly grey,
and is fresh and clean: but pumice exists far from the sea
that has become impregnated with foreign substances, and is
in many places entirely fragmented. ‘The various stages of
pumice degeneration may be traced from the sea beach to the
interior of the island. Some of the pumice has never been
grey, and rounded blocks of a black and cinder-like substance
are here and there found in the parts of the island where
pumice has been most freely washed ashore. This pumice
does not appear to belong to any one particular period, for it
is found to-day on the beaches, and in the islands, but its
composition wherever found appears to be the same, and, on
fracture, its internal part is always shining and fresh-looking,
if it be picked up on the beach or far in the centre of an
island. Pieces of wood, the timbers of ships, well covered with
barnacles and often with portions of copper still attached to
them, float ashore from time to time, and in the history of the
islands many interesting finds have been made.

Many hard seeds and seed-pods are washed ashore with
every high sea, and it is an interesting fact that there are
quite half a dozen commonly found seeds, which will grow if
planted, but which have never, by the agency of waves or
birds, succeeded in making a footing in the islands. Various
large beans are common on the beaches, and I have succeeded
in growing them, but in all the islands there is no plant of their
kind that has grown from a sea-tossed waif. The waves trans-
port the seeds to the high-water mark and leave them there,
and there seems to be no agent in the atoll to effect their
removal to a suitable site for growth tocommence. Rarer finds
are large masses of gum resins whose place of origin I do not
know. The ocean beaches are the resting-places of all sorts
of jetsam from the barrier and the coral beds which lie upon
the submarine slopes of the atoll. Shells in great variety,
coral fragments, the harder portions of crustacea and echino-
derms are thrown up in great profusion by every tide. It
has repeatedly been stated, and very often doubted, that the
dead masses of some corals will float. Dr. H. B. Guppy found

172 CORAL AND ATOLLS

many large blocks of coral in the atoll, which when returned to
the sea were found to float buoyantly.* These coral masses
are not at all uncommon on the seaward beaches, they are
mostly sea-worn and ancient, and, so far as I observed, all
belonged to one species of massive Astrwa with very large
zooids. The skeleton floats by reason of the many spaces
securely walled off by the septa. The most interesting thing
about these stranded lumps of coral is the fact that no species
at all like them is to be found living in the atoll. In the
ocean between the atoll and Java Heads, and in the Banka
Straits, I have seen these floating masses pass by the ship in
company with the smaller lumps of pumice which seem to be
for ever floating in the ocean in this part of the world.

Some of the masses are very large; one that I found on
the seaward beach of Pulu Tikus was nearly three feet in
diameter ; it was a difficult undertaking to roll the great mass
back into the sea, and yet, when at length it reached the
water it floated readily. It had been long in its position at
the top of the beach, and it must have been a time of severe
storm that tossed if beyond the reach of the normal waves;
for though it was left floating, no sea at that season was powerful
enough to heave it back again, and, after bobbing along the
shore of the island for many days, it was lost sight of. Por-
tions of this boulder would not float in fresh water, but this is
not invariably the case, and some floating boulders are freely
buoyant in fresh as well as salt water.

(B) THe BracuEs or THE LAGOON SHORE.

In contrast to the surf-beaten, rock-bound shore which
stretches along the ocean side of the islands is the calm sandy
beach—a place of lapping waves and gentle slopes—by which
the island dry land falls to the lagoon. The lagoon beach is
typically composed of sand of a peculiar whiteness which the
sunshine makes dazzling, and which is entirely composed of fine
particles of coral, mixed only with fragments of broken shells.

* Scot. Geog. Maga., vol. v. No. 6, p. 287.

K VII.

=
Se
ony

g

TERLY WIND.

SAND-PILING BY A WES

=
vy

3]
3

IKUS, TO SHOW

=)
Ss
5
&
mH
S
2
a

LAGOON SHC

0)

Hk

THE ISLAND BEACHES 173

This sand is made in many ways which will need detailed
examination, but the most important of all the agents in its
making are the grinding movements of the wave-driven
boulders on the shore platform, and the myriad beaked Scari
which are perpetually reducing dead coral masses to fine
sand.

The sand is the lightest of all the degradation products of
the coral, and is carried the farthest both by the waves and
by the wind; it is the fluctuating currency of the land, and
is deposited and swept away, accumulated and withdrawn, as
it makes and unmakes new land. In all the fluctuations of
this sand movement there is a compensation, and if a westerly
wind piles up the lagoon shore a foot higher with sand at one
point of the atoll ring, it has robbed some other part, and has
left a breccia platform exposed, instead of a gently sloping
sandy beach.

The great extent of the inner margins of the island ring con-
sists of this white sand beach on which lines of agar-agar, and
other lagoon alge, mark the excursions of the tide; but in
places it is broken by breccia platforms, exactly like the shore
platforms of the seaward barrier. These rock layers, often
broken and displaced by the sucking action of scours and
eddies, stretch out from the accumulated island sand and
run far into the lagoon, becoming continuous with the dark
patches which break the stretches of blue-green water, and
mark the site of the lagoon breccia rocks.

It must not be imagined that these breccia slabs, found on
the lagoon shore, have been formed in situ under the present con-
ditions. The only site of typical breccia formation is the surf-
beaten barrier, and the breccia slabs of the lagoon shore were
laid down in the distant past, when the present lagoon shore
was the wave-beaten outpost of the island plateau. Although
typical breeccia—the well-consolidated mosaic of coral frag-
ments —is not formed on the lagoon shores, yet the same
forces that produce the seaward breccia are always acting in a
lesser degree on the lagoon shore. The gentle lapping waves
of the lagoon shore are for ever tending to weld sand particle

174 CORAL AND ATOLLS

to sand particle, just as the surf of the ocean is always con-
solidating the larger fragments together.
Just as round the sand nucleus of the ocean barrier

Fic. 57.

THE WESTERN END oF PuLu TIKUS,

Showing the worn coral masses that are piled on the beach by the
waves. The presence of the large rounded boulders is due to the
strong currents rushing out of the lagoon entrance.

calcium carbonate is ever being deposited, so on the lagoon
beaches this nuclear increment is ever being added. The
whole question of the formation of breccia turns on the con-
sideration of the welding force of the waves and the materials
with which they have to work. On the ocean side the frag-
ments are wedged together, sand is driven home, and by the
cementing action of deposited calcium carbonate the whole is
consolidated into a massive mosaic. On the lagoon shore
the materials are in a fine state of division, and the force is

THE ISLAND BEACHES 175

not nearly so great, and so the product formed is beach sand-
stone, or beach sandstone in which are embedded coral frag-
ments after the manner of conglomerate. All degrees of
fineness and consolidation of the beach sandstone are to be
found. A fine loose sandstone represents the minimal product
of the process—the hard mosaic breccia, the maximal; and
all intermediate grades are to be found, their composition
depending entirely upon the wave force that was exercised in
their making. Some specimens show an interesting stratifi-
cation, and tell of different forces which have been brought into
play on the lagoon beaches at different periods of the island
building.

On the lagoon shores, where there is an exposure to strong
currents, as on some parts of Pulu Panjang and the western
end of Pulu Tikus, the sand accumulation cannot take place,
and here the wide areas of the beach are made up of broken
coral fragments. These fragments are water-worn from their
ceaseless movement upon each other, and they remain entirely
uncemented ; but this condition gives way to the normal, even
sandy stretches wherever the normal conditions of the lagoon
shore prevail.

The sandy lagoon beaches outrun the islands, and at their
extremities form sand spits stretching towards the lagoon.
This is a very characteristic feature of the atoll islands. The
islands tend to be crescents; the lagoon beaches exaggerate
this crescent shape, and so the extremities of the islands
curve to the lagoon as tapering sand spits. Between these
sand spits the ocean communicates, across the barrier flats,
with the lagoon. On the southern or windward side of the
atoll it is only natural that the light sand should have been
driven to its farthest extent, by wave action across the flats,
and by wind action across the islands; and here it is that we
find the broadest stretches of sandy beach, the highest sandy
shores, and the most extensive lagoon shallows. Here, too,
are found the island lagoonlets with wide areas of coral silt,
but this more fittingly belongs to the description of the island
than the beach.

CHAPTER XV
THE STRUCTURE OF THE ISLANDS

In describing the general structure of the atoll it was noted
that not only did the whole system of islands tend to be
circular, but that the general plan of the atoll—the “ fairy
ring ’—was foreshadowed in the structure of every constituent
island; and a glance at the chart will show this fact. The
most common form of the islands is the crescent, and this is
only to be expected, as many of them are considerable portions
of the circumference of the atoll ring: but this feature goes —
still further, and in many cases the extended horns are bent
inwards until the island itself becomes an incomplete circle,
and itself rises to the dignity of enclosing a lagoonlet. Pulu
Panjang encloses two such lagoons, and Pulu Atas three;
whilst Pulu Ampang major and Pulu Pandan are advanced
approximations to the circular form.

These lagoonlets are places of great interest, for they have
to some extent a flora and fauna of their own; they also show
the final stage of coral degradation, for they are composed of
a white, chalky—almost slimy—ooze, which is dry at low
tides, and partly dry for the greater part of the whole tide
cycle. The degradation of the coral sand that has taken place
in the lagoonlets is doubtless partly a sub-aerial decomposition,
and partly the life-work of the many creatures that live in
these oozy flats. The “fiddler crabs” (Gelasimus, sp.) with
one great pink claw, larger than all the rest of the crabs, live
in myriads in little holes in the white mud; and when
disturbed they beat a steady line of retreat, marked by the
rolling onwards of a pink wave, which fades away before you, as
each one seeks refuge in his burrow. The large Kapeting
Balong (Cardiosoma, sp.) lives round the margin of the

176

THE STRUCTURE OF THE ISLANDS 177

lagoons ; and many forms of smaller crustacea, sand-hoppers,
worms, &c., dwell on the flats and furnish food for the shore
birds which frequent these places.

The chief vegetation consists of bushes of “tea tree”
(Pemphis acidula, Forst., Kayu Burung) which grow on the
higher parts—islands during high water, and mounds when
the tide is out ; these and pioneer coco palms are studded all
over the white mud. Some coarse grasses (Lepturus repens,
Forst.) that bind the loose sand with their spreading roots, are
also agents, and apparently active ones, in helping to win these
lagoons to the island dry land. These lagoonlets differ of
course from the great lagoon of the atoll in that they have no
depth of water, and no growing coral; but they are formed in
the same way as their larger representative and their fate—
that of slow silting with coral sand—foretells the history of
the lagoons of atolls in general.

The approximation of the individual islands to a crescentic
form is due to a very obvious cause. At those tide-washed
gaps which separate island from island is the connection of the
outer ocean with the lagoon; for here, at high tide, the rollers
sweep across the barrier flats. very wave which breaks on the
barrier washes minute fragments of coral, and fine coral sand,
across the flats, and deposits its load of silt in the stiller
waters of the lagoons. It is but natural that the suspended
matter becomes first heaped up at the sides of the gap, for
here the current slackens first, and so the extremities of the
islands become the site of the deposition of sand-banks, which
are formed as in-curved continuations of the land. Dr. Guppy
has laid much stress on this simple phenonenon as an agent in
land fermation, and it is certainly an important one in the
shaping of the islands: in fact it is a continuation of the very
process which has, in the past, shaped the atoll itself. The
current that lays down the banks at the ends of adjacent
islands still continues to carry silt in its moving waters, and
this is ultimately dropped farther out in the lagoon, opposite
the gap. Sand-banks in the lagoon therefore, when far out,

tend to be opposite the interspaces of the island ring, whilst
M

178 CORAL AND ATOLLS

nearer shore they tend to be continuous with the ends of the
islands.

Although there is a certain amount of constancy in the
shape of the islands in the ring, there is a very great diversity
in their size; and the largest islands lie to the south of the
group.

Pulu Panjang and Pulu Atas are by far the largest pieces
of land; not only are they longer, but they are wider from sea
beach to lagoon shore than the other islands, and their seaward
margins are amongst the highest pieces of land in the
atoll. Whatever the size or shape of the island, it must
always be borne in mind that the dry land is a heaped-up
collection of débris: the islands are made of flotsam and
jetsam, of which of course the greatest constituent is dead and
broken coral. It is only to be supposed that time, with the
sub-aerial alteration of coral, the growth and death of all sorts
of vegetable products, and the action of the winds on the
lighter particles, should have done much to alter the original
appearance of the land. The larger islands towards the
south are those that have, to-day, the least of the original
features of their origin: the coral fragments have become
disintegrated and buried beneath the accumulated generations
of fallen trees, fallen leaves, and the husks of the coconuts.

It is the regulating policy of the islands that the husk of
the nut is not used for the purposes of commerce, and so no
fibre is made and no coir exported, but the husk rots where
it falls, for all the nuts are husked as they are picked up. In this
way the thickly vegetated islands have accumulated a consider-
able depth of soil.- One of the important factors in the formation
of this vegetable mould is the large Kapeting Balong (Cardio-
soma, sp.), which plays, on a grand scale, the role of the
earth-worm. It burrows the land through and through, and
undermines the heaps of husks with its tunnels—almost as
large as rabbit-holes—so that a considerable part of the fibre
becomes buried. The buried husks rot, and more earth is
formed for future generations of coconuts to take root in, and
for future generations of the Kapeting to till. Not only the

PLATE VIII.

A TYPICAL VIEW OF THE VEGETATION OF THE ATOLL.

THE STRUCTURE OF THE ISLANDS 179

Kapeting Balong but several other species of land-living crabs
have a habit of taking the fibre into their burrows, and go
deliberately burying it, and the influence of efforts at tilling
carried out by these large creatures, which exist on some of the
islands in great numbers, must be considerable. The land-
living crustacea are then the earth-worms of the group, in as
far as their function in the formation of vegetable mould is con-
cerned ; and the earth-worm which does exist on the atoll is so
very uncommon that its share is negligible.

The stem of a coco palm that has fallen lasts whole but
a very short time, the wood of the centre of the trunk is soft,
and soon by ,the agencies of the myriad millepedes (native
name—kaki vibu) and the natural processes of decay, the
trunk becomes hollowed out, and nothing but a shell of bark
is left. Every cyclone wind which visits the islands lays low
many hundreds of trees, and their rotting remains form
another source of soil, and another cloak for the coral débris.

Besides the husk of the nut, the coco palm is perpetually
shedding fronds, for as it grows up and up the lowest fronds
drop off leaving those rings on the trunk which mark the
upward growth—and these fronds are always being added
to the land accumulations. The remainder of the vegetation
of the islands is characterised by a tropical mushroom rapidity
of growth; the papias spring up fresh and green to a height
of twenty feet, and, like great weeds, die down and rot. There
is nothing more strange than the manner in which these
luxuriant trees appear after wet weather in a soil consisting of
nothing save coral chips and a little sand. At the end ofa
dry period, the advent of a good spell of rain will produce
green shoots from every fissure of the surface of the ground,
and the most luxuriant of all the green things is the papia, or,
as it is called in the islands, the Katis.

Even amongst the largest trees the Ampol (Pisonia grandis)
has been well called the “cabbage tree,’ for a branch the
thickness of a man’s thigh is not to be trusted with a man’s
weight : and no tree of any age is to be found that has not
some large portion rotted away Vegetation springs up rapidly

180 CORAL AND ATOLLS

and as rapidly dies; during droughts the islands will be
parched and brown, every plant save the coco palms will
look leafless and dead: but the next day if rain comes, the
whole appearance of one of the more barren islands like Pulu
Tikus will wholly change, and all the white coral surface
of the land will be green with sprouting seeds. The rapidity
of the cycles of vegetation is a thing to be reckoned with in
considering the alteration of the primitive structure of the
islands. |

In connection with the formation of vegetable mould, the
influence of the heavy rainfall must be considered, for the
rate of the washing of surface accumulations through the
loosely strewn coral chips is very wonderful. In Pulu Tikus
coco palms have lived and died for generations, their trunks
have decayed, their leaves rotted: and yet on the surface of
the greater part of the island no soil accumulation is to be
seen. As a matter of fact the extreme barrenness of the
islands is more apparent than real, for if a few inches of the
coral débris be scraped from the surface, the vegetable earth
is found below, intermixed with the lower layers of coral,
where it has been washed by the rain. On Pulu Tikus forty
tons of earth were imported from the Botanical Gardens of
Singapore; in 1902 it was landed, in 1905 any trace of it
was hard to find. It is such facts as these that make the
presence of phosphates deep in the soil of high coral islands
easy to explain: for every shower that falls dissolves any
surface substance readily soluble, and carries it in solution as it
percolates through successive layers of loosely packed coral,
thus giving it the very best opportunity to enter into chemical
combination.

Another factor in the after-history of the island is the
drift sand, which, under the influence of the winds, builds up
the surface of the island beyond the limit of the sea action.
Many of the islands are thickly covered with sand; some are
even entirely composed of sand, and an interesting charac-
teristic of those islands, into whose formation no large frag-
ments of coral have entered, is their uniformity of contour.

‘SOMIE ATOd ‘ANVISI AHL LO AGIS NOODV'I AHL AO YNITId-ANVS

“XI ULV Td

THE STRUCTURE OF THE ISLANDS 181

When an island is formed of the typical coral débris, consisting
of detached slabs of breccia, boulders and fragments, it is
always evident that its windward margin is its highest land,
and, on section, it is roughly wedge-shaped, the apex of the
wedge being towards the lagoon:—for the island has been
built by the waves from the ocean side, and its wave-beaten
shore is piled the highest. On the other hand, when an
island is made of sand it is moulded mostly by the wind, and
the effect of the wind is to drive ihe light sand farther and
farther towards the leeward side, so that the island rises from
the ocean shore to the lagoon beach—the apex of the wedge
being towards the ocean. So marked is this effect of wind
formation that, in the case of Pulu Bras, the seaward shore
rises as a gradual sandy slope, and ends abruptly towards the
lagoon as a sand cliff twenty-seven feet high, which is the
leeward limit of the island. On all the islands this wind-
driven sand makes some covering for the coral basis, and most
often it is collected into characteristic ridges and mounds,
making some little break in their otherwise level surface. As
the sand is being built up all over the islands, even in those
parts where wave action has long ceased to have any influence,
it must have been one of the most considerable of the agents
in raising the land from its sea-made level. The action of the
wind-driven sand has doubtless, ina great measure, waned since
the islands have become thickly planted with coco palms, and
now is probably secondary in importance to the influence of
the vegetation itself.

On the lagoon shore the sand is the only material from
which the island now builds, and the wind-driven particles are
forced back by the lapping waves of the lagoon, with the
result that a characteristic rise of sand marks the fall to the
inner shore, just as the boulder rise marks the seaward edge.

The outcome of these various building forces is that the
atoll islands have not only a characteristic geographical shape,
but they also possess a characteristic land contour. The sea-
ward edge is piled the highest, and is the mountain range of
the island—in Pulu Tikus, on the lee of the atoll, reaching

182 CORAL AND ATOLLS

thirteen feet at its highest point; and in Pulu Atas, to the,
windward, rising to a “hill” of thirty feet.

From this sea-bank the land falls to the centre of the
island, where in some instances there are many hollows below
the high spring-tide level. Towards the lagoon the land rises
again as an inconsiderable sand ridge, and thence falls gently
away to the lagoon. In Pulu Tikus, which is the island that
has undergone least change from the bare débris condition of
its first making, the greater part of the surface of the land is
bare coral fragments with a very sparse admixture of sand and
vegetable mould, and there has been very little building up of
the interior of the island. The result is that most of its
surface is raised but a foot or two above high-water level,
and a considerable part of it is below, and as the débris is
easily permeable to water, the burrows in which the Kapeting |
Balong lives are usually filled with salt water almost to their
mouths. Wherever a hole is dug, the water is found just
below the surface, and large puddles are formed in many
places at every high spring tide; whilst the water in a fish-
pond, some thirty feet from the lagoon shore, follows all the
movements of the tide with but little delay. There are many
facts of great interest in connection with this sponge-like
condition of the islands, which permits the sea to fill them at
high tide, and leave them when the tide is low.

It is this condition that gives rise to the curious pheno-
menon of rising and falling wells of fresh water, which can be
made on a great many of the islands, and to account for which
many suggestions have been put forward. It has been said
that the sea water at the height of the tide percolates into
the island, and, as it goes, it becomes filtered through the
coral so perfectly that in the middle of the island it becomes
quite fresh, and may there be tapped as good drinking water
in wells. That filtration by the accumulation of dead coral
could be so complete as to remove sodium chloride from its
solution is an assumption so bold that its acceptance could
only be justified by the careful exclusion of all other possible
causes of the freshening of the water: and as a matter of fact

THE STRUCTURE OF THE ISLANDS 183

the other and more likely causes have not been excluded.
In Pulu Tikus fresh water cannot be obtained in wells, and
Pulu Tikus, being the island that has retained most of its
original features, shows in this as in the case of other pheno-
mena the half-formed product that more easily tells its origin.
At high tide the water rises, the fish-pond fills up, the
Kapeting’s holes show water a few inches below the surface,
and puddles form in the more low-lying parts of the island.
At low tide a visit to the barrier will show this water running
out of the land again, welling up in holes and trickling down
the beach as the sponge empties. If, on this island, a hole
be dug as far from the sea as possible, the water that is met
a foot or so below the surface will be found to be either quite
fresh, brackish, or wholly salt; and it depends entirely on the
amount of recent rain which of these three conditions is
found. After a wet period, fresh water will be found any-
where towards the centre of the island, and it will rise and fall
in the hole as the tide goes through its cycles: but when the
dry weather sets in the fresh water becomes more and more
brackish, and in the course of some days becomes entirely
salt again. The reason for this is easily seen, for since the
land is so little raised above the ocean-level, the water which
the accumulated rains produce must lie as a subsoil water very
near to the surface, and, as it is lighter than the sea water, it
rises and falls as a surface layer with the movements of the tide.

Fresh water will remain separate as a surface layer even in
the open sea for a considerable time. After a tropical down-
pour in Singapore roadstead the surface specific gravity was
lowered from 1026° to 1021°, and on two occasions in this
atoll so continuous was the rain that large numbers of fish died
from the freshness of the water, for the rain stood for a height
of several inches on the surface of the salt waters of the lagoon.

The diffusion which takes place slowly in the sea takes
place still more slowly when the rain water, after percolating
through the porous débris of which the island is composed,
lies in the depressions of the compact breccia layer that forms
the basis of the islands; and in an island so porous as is Pulu

184 CORAL AND ATOLLS

Tikus, the complete mixing takes, at the fastest, a period of
many days. In the larger and more consolidated islands,
where the building up of the land level has advanced further,
this subsoil water exists as a large pool on the breccia basis,
-and is only displaced by the movements of the tide water. A
very prolonged drought would no doubt affect all the wells of
the atoll in time; but so large is the accumulation of subsoil
water that forty tons a day could be taken from a well in Pulu
Selma for the Australian horse ships, without affecting the
quality of the water, if the well were given a rest of an hour
in the middle of the day. No restrictions of any kind are
placed upon the using of the well water by the natives of Pulu
Selma, and domestic water for cooking, washing, and bathing,
as well as water for their gardens and industries, is taken
freely from these wells; and yet, even in the prolonged drought
of 1905-6, the water very fairly retained its freshness. The
water found in Pulu Luar is the best well water in the group,
and is that used by the Ross family for drinking purposes.
Under no conditions does it ever become brackish, doubtless
because its use is restricted, and the well is situated in the
centre of the island, and therefore far from the sea. The water
of Pulu Selma has.a degree of hardness that makes it difficult
to lather with ordinary soaps, though with the island soap,
made from coconut oil, it always lathers freely :—-with this
soap, however, sea water can be used for washing purposes.

The finding of fresh water in wells dug on coral islands is a
common thing, and in the low islands of the Pacific and in the
West Indies these wells are made by the natives; but the water
is apt to be scant and brackish, and this atoll is somewhat
unusually favoured.

The reason for the constant freshness of the wells is, of
course, the heavy annual rainfall, and the yearly seventy inches
does away with any necessity to collect water in holes in coco-
nut palms, or to regard water as a great luxury, as is the case
in some lowislands. But on such of the islands as Pulu Tikus,
where no wells can be made, all the water must be condensed
for drinking, and collected from the roofs for domestic purposes.

CHAPTER XVI
THE LAGOON

In accounts of atolls it is often found that the lagoon is
regarded and described as though it were a pit—a sort of
lake that is situated on the summit of a submarine moun-
tain; often it is supposed to be the water-filled crater of a
volcano.

So long as such ideas concerning the nature of the lagoon
are held, no description of it is at all likely to depict its real
features, for no descriptions written under this belief will ever
make it apparent that the lagoon is really the top of a broad
reef, around whose edges a ring of islands is raised. In this
description it is assumed throughout that the lagoon is a
slightly submerged reef, whose surface has undergone some
modifications owing to its partial enclosure by the island ring.
The lagoon is a sheet of water with an area of some twenty-five
square miles, and over the whole extent of its area the after-
effects of its enclosure by the islands have not been equal. The
result is that to-day the lagoon may be divided, for the purposes
of description, into two parts: a southern shallow part; and a
northern deeper part. It is of importance to note that the
shape of this southern part is exactly that already described
as belonging to the sandy lagoon shores of individual islands.
If all the southern islands, from Pulu Ampang on the east to
Pulu Panjang on the west, be imagined as one continuous reef,
then the shallow part of the iagoon is exactly what we would
expect its lagoon shore to be. It is also of importance to
notice that this crescentic shallowing of the southern portion
is gaining on the deeper northern part, exactly as the cres-
centic sand beaches of the individual islands gain on all parts

of the lagoon. The same process is evidently at work on a
185

186 CORAL AND ATOLLS

larger scale, and a line of transition from coral shoal to sand-
silted lake is sweeping, as a crescent, from south to north.

The entrance to the lagoon is situated to the north of
the atoll, and here the bottom of the lagoon merges with the
comparatively shallow areas of the coral ridge, and is seen in
its most luxuriant growth, and in its most primitive condition.
Between Pulu Luar and Pulu Tikus the barrier is entirely
wanting, and in this gap the original coral bed lies at a few
fathoms’ depth with its luxuriant mixed growth of massive
corals undisturbed by violent wave action. This coral bed
is, no doubt, the type of the original condition of the whole
area of the ridge now occupied by the southern atoll and the
northern atollon of Keeling Island.

The shallow southern part of the lagoon extends in the
form of a great horse-shoe, skirting the inner shores of the
islands, and invading at its margins the northern deeper part.
Its area may be roughly taken as slightly more than half the
total lagoon area, or about thirteen square miles. Large areas
of its most southern parts are dry at low water, and covered by
a few feet of water only at full high tide. The more northerly
and central parts of the shallows are covered by anything less
than a fathom of water at low tide. The bottom of this part
is, however, not at all uniform, and deeper hollows are studded
about the central portions of the greater part of the southern
shallows.

Along the inner shores of the large islands, Pulu Atas and
Pulu Panjang, the long sand flats are dry for a mile or so at
low tide, and at full high tide a great part of this flat has no
more than an inch or two of water on it. Towards the shore,
these flats become continuous with the oozy white mud which
collects in the lagoonlets of these two islands, and towards the
lagoon they slope gradually to the slowly deepening water,
which is clearly indicated as a bright green belt, merging after-
wards to the characteristic deep blue.

Bushes of Pemphis acidula and pioneer coconut palms
stand here and there as outposts of the land: and it is the
policy of the islands that these pioneers are not interfered

“LUISI GNVS TIVNS
HHL “YISVd ATOd SGUVMOL ONIMOOT ‘SOMITE ATOd HO GN NYAISAM AHL WOU NOODWT AHL

“X HLV 1d

THE LAGOON 187

with ; for the advance of vegetation is the surest way of winning
dry land from the silt regions. In the reclaiming of the
lagoon shallows a grass (Lepturus repens) plays a great part,
for it binds the shifting sand firmly with its long trailing roots,
and makes firm dry land out of the useless shallows.

In 1827, when the permanent settlers first arrived in the
islands, the silting of the southern portion of the lagoon was
not nearly so far advanced as it is to-day. The schooner which
the islanders made for their use was actually built in Pulu
Atas, and launched from it; and it sailed in and out from the
settlement on the island to keep up the necessary traffic of
the atoll. Before ten years had passed, however, the silting of
the southern end of the lagoon had proceeded so far that the
boat channel became obliterated, and schooner traffic with the
settlement was cut off. It was for this reason that the site of
the settlement was moved to Pulu Selma, the island which it still
occupies, situated in the northern deeper half of the lagoon.
It would appear that the southern anchorage has never been
used for any other craft than the island-built schooner, for
ever since the atoll has been charted the safe anchorage has
been given as Port Refuge in the lagoon off Pulu Tikus, or
else without the lagoon altogether in the gap between Pulu
Tikus and Pulu Luar.

Although the whole story of the southern part of the
lagoon discloses a steady progress of the conversion of a coral-
studded stretch of water into sand flats, still, at any one point,
some indications might be seen, from time to time, which would
seem to contradict the general trend of the process. Beach sand
has been seen to be the fluctuating capital of the islands and
of the beaches: and, in exactly the same way, the silt sand of
the lagoon is the ever-varying factor in the ceaseless minor
variations, which are always going on in some part of the whole
extent of the lagoon. Sand spits and shallows wax and wane ;
banks are piled up, and then soon after scoured away, but in
the manner of these changes there is no general influence upon
the atoll structure as a whole, and purely local conditions
determine the balance of the processes at work in any given

188 CORAL AND ATOLLS

spot. To any one who has for long watched these minor—yet
striking—changes, it must be evident how fallacious are those
arguments concerning the age of atolls, or the manner of atoll
formation, which are based on the measurement of the quan-
tities of sand piled up, or denuded, in any one point in the
whole circle of more than twenty miles. Where in the island
ring the sea runs strongly through the gaps and into the
lagoon, spits are formed wherever, by some interference, the
current slackens; in this way the crescent shape of the islands
is brought about, and the great sand flats of the southern
portion of the lagoon are accumulated.

Drift sand is the main factor in causing the shallowing of
the lagoon, and drift sand it may be safely said will, lacking
some general upheaval, be the agent which will ultimately win
the lagoon to the dry land; but it is the most inconstant and
Capricious agent, and no man may say that so many feet will
- be laid down this year, and so many next, and that therefore
in a certain meaningless number of years the lagoon will cease
to exist. The very presence of the sand in the lagoon
introduces a complication into the question, for sand is one of
the most potent causes of coral death; the shallowing of the
lagoon by the accumulation of coral growth must therefore be
arrested when the shallowing by sand deposit sets in. It is
under these circumstance that the very slight building in-
fluences of Lithothamnionide come into play, for they do not
appear to be so sensitive to the influences of silt as are the
corals. In certain places at the southern side of the atoll, in
the border-line region of barrier flat and lagoon flat, the
growth of these calcareous algze occupies a fairly large area,
which is free of living coral in any quantity: but apart from
arresting the drift sand in their meshes they do not have
any influence on the rate of building. It must not be
imagined that the great southern part of the lagoon is by any
means devoid of living coral colonies, for dotted about at
intervals are some of the most luxuriant of all the coral
growths. It has been said that the distribution of the living
colonies on the barrier and the barrier flats was a distribution

THE LAGOON ~ 189

of types and not of species; and exactly the same assertion
holds good with regard to the lagoon. The same corals that
live on the barrier are also found in the lagoon. At the same
time it is quite correct to say that some types are definitely
confined to certain definite environments, and this forms one
of the most interesting pages in the study of living corals.
This subject has already been discussed and so no further
details will be added here: but it must be insisted that sand
and silt are the most important agents in the causation of
death of corals, and that their presence acts in two ways:
(1) It may fall on them and choke their zooids from above,
or (2) It may overtake them from below. Some idea of the
rate of the silting of the lagoon may be gathered by measuring
the dead lower portions of the great branching colonies of
Madrepora ; for these corals do not normally die in their lower
portions, and the dead area represents the amount of silting that
has taken place, in the immediate neighbourhood of the colony,
since its first establishment.

Sand then is the great architect of the lagoon, and while it
inevitably kills the coral growths, it more than counterbalances
for this by its own building efforts.

The still waters of the lagoon are harmful to the coral
colonies in another way, for large areas of the shallow
water are covered by great beds of alge. Where algz exist
abundantly corals do not flourish. Agar-agar and many
other seaweeds cover a considerable portion of the lagoon,
making characteristic dark patches on the bottom. One such
large bed exists in front of the eastern end of Pulu Tikus, and
many are found in the southern portion of the lagoon, and
these beds flourish to the absolute exclusion of coral growth.
There is in particular one very fine green alga that occurs
even in the barrier pools, and is particularly inimical to coral
growth. At certain seasons of the year these fine green
threads are found, as drift-weed, in every pool; and as they
wash to and fro in the tide they become entangled among the
branches of the corals. Wherever the threads obtain a fair
lodgment, a green mat-like growth is spread on the surface of

190 CORAL AND ATOLLS

the coral, and the fate of those portions of the branches upon
which it settles is sealed. The coral slowly dies, and on the
dead portion the alga becomes thickly encrusted, its filaments
spread out and are washed away to find a fresh resting-place
among the neighbouring branches. In sheltered pools, where
the process can be watched, it is easily seen that the alga is
one of the most potent factors in causing coral death; and
doubtless in the lagoon, where the beds of alge occupy many
acres, the factor is one of great importance; certain it is that
in the alge beds the corals do not flourish.

The sand and the seaweed between them have in some
measure banished the living corals from certain parts of the
lagoon, and the actual fraction of its whole area covered by
living coral I would judge to be less than the half that, in
1836, Darwin estimated. ;

Of the southern part of the lagoon, the western half is the
area that is most thickly covered with coral growth; and the
bottom of the lagoon is here chequered over with coral colonies
in a very characteristic fashion. Large masses of Porites,
with actively living zooids only present round their sides,
give rise to large, and roughly circular, flat-topped rocks;
and on the flat tops live the more active branching forms of
Madrepora. In the sand, isolated and small colonies of Porites,
all approximating to a circular form, lie scattered about; the
great majority of the colonies being far less luxuriant than
those of the same species that live in the ocean beyond the
barrier edge, or in the northern half of the lagoon, The
nearest parallel that I can draw to the condition of coral
growth on the lagoon bottom is the manner in which gorse
bushes may be dotted about upon a hillside; it is a series of
scattered groups disposed about at quite irregular intervals.
The groups of coral colonies are of very varying size, but they
nearly always contain the same set of types of vegetative
growth.

I was particularly fortunate, during my residence in the
atoll, in being able to accompany the boats engaged in clearing
the coral from the southern part of the lagoon. The great

THE LAGOON © 191

boulders are shifted with crowbars by the Malays standing in
the water up to the level of their chests, and, with much skill,
large masses are rolled and hoisted into a lighter, where the
corals and their teeming symbiotic fauna can be well studied.
The work is undertaken partly for its own utility, and partly
to obtain the boulders for enclosing turtle-ponds, and also for
burning into lime. Thanks to the kindness of Mr. Andrew
Clunies-Ross I was able to take part in these expeditions, and
the experience gained by seeing and handling so much coral
from different areas of the lagoon gives some confidence for
dogmatism as to the disposition of living corals within the
lagoon.

Time after time, as the coral was hoisted into the lighter,
the same sequence of types showed. The basis was nearly
always formed of a large colony of massive Porites, mostly
dead, and only here and there at its periphery showing signs of
activity; the older portions giving a resting-place to Astraw
and branching types of Madrepora, and, at its edges, the
Montipore and the various plate growths which are a common
modification of many species. In the sand around these
masses are the branching types of Madrepora and Montipora,
which crumble to pieces and break as the boulders are raised ;
and growing at intervals, and usually in the deeper parts—so
that the Malays must dive to procure the specimens——are
the large leaf-like growths of Echinopora lamellosa. Here too
grow the rounded stems of the branching Porites, which are of a
varying shade of greyish yellow or greyish purple when alive,
but which, when the coral is dead and exposed to the air, turn
black. Growing upon the boulders of massive Porites, and
also living free in the deeper pits, are the fine colonies of the
most delicate of all the corals—the pure white Seriatopora,
which has received amongst many other specific titles that of
caliendrum.

Of all the lagoon types of coral, those of a branching
habit are the most numerous and appear to be thriving the
best, and the type of growth of Madrepora known as pulchra
covers unquestionably more ground than any other type. The

192 CORAL AND ATOLLS

western half of the southern area of the lagoon is therefore
fairly well populated by coral colonies, and a very definite set
of types is usually found gathered together as a composite
mass of growths, these masses being dotted about in irregular
fashion all over the bottom of the lagoon.

The eastern side of the southern end of the lagoon is
more barren of corals than the western side; and this state
of things is owing to the effects of the strange happenings of
the year 1876.

As a sequel to a cyclone which visited the atoll in the
January of that year, there occurred an event which to this
day has left its imprint plainly marked on the coral growth of
the lagoon, When the storm had died down for some thirty-
six hours, it was noticed that a spring of dark-coloured and
malodorous water was welling up into the lagoon from a source
somewhere between the southern end of Pulu Selma and the
northern end of Pulu Cheplok. The foul water continued to
ooze out for some ten or fourteen days, and it slowly spread all
over the eastern portion of the southern end of the lagoon.
Dr. H. O. Forbes visited the atoll two years after this event,
and has well described the lifeless state of the area over which
this water flowed. In “ A Naturalist’s Wanderings,” p. 22, he
says: “Its whole eastern half was one vast field of blackened
and lifeless coral stems, and of the vacant and lustreless shells
of giant clams and other Mollusca paralysed and killed in all
stages of expansion. Everywhere both shells and coral were
deeply corroded, the coral especially being in many places
worn down to a solid base. Since the catastrophe, there has
been, till almost the day of my visit, no sign of life in that
portion of the lagoon; I saw very few s)he and only here
and there a new eon of Madrepora and Porites. I found
only one TZridacna alive (its three years’ growth being twelve
inches in length and thirteen in breadth).”

This description forms avery good basis for estimating the
time that is occupied by the corals in renewing their growth,
and we have a further landmark in Dr. H. B. Guppy’s account
written in 1888.

THE LAGOON 193

In the Scottish Geographical Magazine, vol. v. p. 570,
he says: “The corals on the shallow flats naturally suffered
most from the intensity of the destructive fluid, and even
when I examined this locality twelve and a half years after
the event much of it appeared a lifeless waste. The dead
trunks and branches of the arborescent Madrepores that had
once flourished here literally covered the bottom in places.
Multitudes of the shells of the 7ridacne, rotten and decayed,
strewed the ground; and even the pits displayed on their
bare slopes numbers of dead detached plates of Hchinopora
lamellosa that had once clustered thickly around their sides.
Yet it was evident enough that the corals were in places
beginning to resume their sway. ‘Thus, the flats immediately
south of New Selma have long since been clothed afresh with
rapid-growing branching Montipore and other corals; and
even where life seemed completely absent, a cioser inspection
usually discovered a few young growing corals and small
living clams. However, at least a quarter of a century will
have to sueceed the event before the growth of corals can
disguise the destructive effects of 1876.” In 1905 and 1906
I spent very many days in drifting all over this eastern area
of the lagoon. With a companion in a boat, a start would be
made from the south, and as, with the wind and tide, the
boat slowly drifts over the lagoon, every foot of the bottom
may be carefully examined. Anything that demanded close
examination was dived for, or waded to, and in this way
a very intimate knowledge was acquired of the present state
of the coral growth. Although thirty years have gone by
since the pouring forth of this destructive water, distinct
evidence of its effects are still abundantly clear; and the
time of final repair must certainly be put forward considerably
beyond Dr. Guppy’s limit. There are many places to-day
where there are practically no signs of life amongst the dead
coral masses that cover the bottom, and alge have for the
most part been the successors of the coral life.

I have watched the dead coral fields for a period of fifteen

months, and failed to detect any signs of life in some portions
N

194 CORAL AND ATOLLS

of them. I have seen the crumbling masses raised into the
coral lighter, and boulder after boulder will be rolled in
without any one of them showing the minutest portion of new
growth attached to it. To the south and west of Pulu Selma
there are still large areas that are only occupied by the re-
mains of the corals killed in 1876, and the thirty years that have
elapsed seem to have accomplished but little repair when the
masses are seenraised to the surface. But before this extreme
slowness of repopulation may be taken as a fitting criterion
for any estimation of the rate of coral growth, the conditions
which have prevailed during the interval of thirty years must
be reckoned with. In the first place the alge, of which there
were many beds on the adjacent areas untouched by the foul
water, have, by reason of their more rapid growth and greater
hardihood, taken possession of a very large part of the area»
that suffered the damage. Before the more delicate and
slower growing corals could recover their activity, and re-
colonise their ground, the alge had invaded it; and as we
have seen, the advent of abundant growth of alge terminates
all activity in the coral colonies. When once the seaweeds
had fairly settled on the spot, the invariable accompaniment
of their growth took place, and sediment began to accumulate
in the beds which they had formed. This fact is clearly seen
wherever alge are settled—that the lagoon begins to shoal
with increased rapidity; alge in the lagoon play very much
the part of the grass on the shore flats, for they enmesh
the drift sand and hold it fast. The environment there-
fore becomes one of rapid silting, and no environment is more
inimical to coral growth. It is, therefore, not alone by reason of
the slow growth of the coral that this area has not become
repopulated with coral colonies; for the time that has elapsed
since the destruction is ample for very advanced growth to
have taken place.

All over the whole of the southern shallow portion of the
lagoon are deeper holes, which show plainly as darker patches
in the clear water: and a fair general picture of this part
of the lagoon is that of a wide stretch of white sand, with

THE LAGOON oe 6 185

here and there patches of coral rocks, and here and there
deeper holes. The coral rocks rise almost to the surface at
low water, and they are for the most part of a bright yellow
colour, being colonies of the most abundant type of Porites.
At intervals are beds of yellow branching Madrepore and
fields of green and white alge; the whole seen through a
fathom or so of clear green water. Upon the sandy bottom
the rills of wave action are clearly to be seen, running east
and west—the imprints on the bottom of the Trade-driven
waves of the lagoon. There are, however, very many places
in the lagoon where the coral growth is not nearly so freely
scattered as that just pictured, and in areas over which the
foul water never exerted its baneful influences are stretches of
white coral sand, where a coral mass will have no neighbour
nearer than one some fifty yards away.

From the southern shallows the lagoon deepens towards
the north, where it becomes a basin with an average depth of
from four to five fathoms. Even in this deeper part, however,
there are many spots where the growing masses of coral rise
to within a few inches of the surface at low water ;—some are
even normally exposed at all low tides. One such mass rises
from Dymoke shoal in the northern basin, opposite Pulu Selma,
and this rock—known to the natives as Batu Dua, or “The
number two rock ”—is exposed for some two feet at most low
tides. The transition from the shallow portion of the lagoon
to the deeper portion is very easily seen in the colour of the
water, and at places the line of demarkation is very sharply
cut. The clear green water of the shallow sand-covered
portion passes, often very abruptly, into a very beautiful blue,
—a clear wonderful blue, which can never be described or made
real. The entrance into the lagoon is between Pulu Luar and
Pulu Tikus, and here ships may be safely navigated by one
knowing the locality. A good anchorage stretches in front of
Pulu Tikus with from three to eleven fathoms of water. It is
possible for vessels of shallow draught to enter the lagoon be-
tween Pulu Luar and Pulu Panjang, both to the north and to
the south of Turk Reef (Batu Satu, “The number one rock ”),

196 CORAL AND ATOLLS

but the channels are intricate and liable to considerable
changes from the rapid growth of coral masses.

The growth of corals all over this part of the lagoon is
very luxuriant, and very large colonies of Porites are to be
seen in the deep blue water; here also large growths of
Millepora are seen, and both the great plates of the type
complanata and the branches of the type alcicornis are to be
recognised. Millepora colonies are as a rule absent, or
represented only by very sickly looking growths, over the
southern shallows; but in the northern basin some of the
specimens are among the finest to be found anywhere in the
atoll.

To the westward the lagoon shelves slowly to the open
sea, and here great masses of coral are to be seen growing in
the greatest profusion. The site is one in which practically no
sedimentation is taking place, for in any weather the sea is a
confused one in the gap between Pulu Luar and Pulu Panjang.
When the wind comes from the west, or north-west, very
heavy seas roll into this gap, and the whole interval is filled
by great white-crested rollers, making any entry to the lagoon
an impossibility.

The northern basin of the lagoon presents an altogether
more varied picture than the southern portion, for it shows
every phase of the transition from shoal to submerged reef. It
shows at its southern limits the destruction of coral growth
by sedimentation, and the shallowing of the lagoon by sand
accumulation ; at its northern limits, the absence of sedimen-
tation, and the shallowing of the lagoon by coral growth.
Some of the coral boulders in the northern basin have grown
with very great rapidity, and their removal from the boat
anchorage opposite Pulu Tikus is occasionally necessary.
Most of these boulders are very large colonies of Porites, which
are for the most part yellow coloured, and here grow in the form
of slightly flat-topped spherical masses. Another part of the
lagoon which is undergoing considerable change, brought about
by the activity of coral growth, is the best channel to Pulu
Selma, but here it is great beds of the typical form of

THE LAGOON 197

Madrepora pulchra which are causing the alteration. ‘l'hese
beds are bringing about a somewhat rapid shoaling of the
lagoon in this spot, for M. pulchra is a type of colony well
formed to resist some degree of sedimentation ; and not only
are the colonies growing very actively, but are holding in their
dead lower branches a great amount of drift sand. They are
therefore forming banks, with living branches actively growing
at the summit, and sand, held fast among the dead branches,
at the base. Another type of colony thriving in this environ-
ment is a plate growth of Montipora, approximating to the type
named M. circinata; it plays very much the same part as
Echinopora lamellosa does in the southern shallows, and grows
upon the sides of the banks, holding the sand back behind its
broad leaves.

Although the water is,in many places, six or eight fathoms
deep over wide stretches in the northern part of the lagoon,
the nature of the coral colonies that lie below can nearly always
be made out, and where the lagoon falls to the open sea,
between Pulu Tikus and Pulu Luar, the parti-coloured bottom
may be seen for a very great distance. The fall from the
shallow waters of the lagoon to the ocean outside is not nearly
so abrupt as has been imagined, for the entrance, lying to the
north, corresponds with the northward tailing out of the bank
on which the atoll is formed—the bank upon whose northern
extremity Keeling atollon is situated. When sailing from the
lagoon, the bottom is seen as a patchy background through
water that becomes increasingly blue; at first yellow masses
of Porites alternate with white patches of sand, and blue
patches of the deeper holes ; then the patches blend more, and
lighter yellow areas and darker blue areas alone are seen.
When still farther from the atoll only the larger patches show
at all, and the appearance is somewhat that seen when looking
from a hillside upon distant fields, differently coloured by
various growing crops. Alternations in colour are seen for
long; and the whole picture, from minute detail to complete
blurring, fades slowly. I can speak of this point with some
certainty, for I have many times been in and out of the lagoon,

198 CORAL AND ATOLLS

and have, on calm days, taken a small boat round the seaward
side of Pulu Tikus. These beds of coral outside the lagoon
are the sites of the most luxurious coral growth to be seen
anywhere in the atoll; and Porites grow in colonies whose
size is not approached by any in the lagoon, or on the barrier.

It is well known to sailors that from some lagoons the
water issues, at the falling of the tide, with such force that the
passage of a ship into the entrance may be an impossibility.
Although there is no tide rip from this lagoon which offers any
serious obstacle to navigation, still a strong current is always
running out of the lagoon, past the west end of Pulu Tikus, to
impinge on the southern and eastern shores of Pulu Luar.
The gaps in the island ring that lie to the south and east are
inlets for the Trade-driven ocean; in the run between Pulu
Tikus and Pulu Pasir the set is always lagoonwards, whatever
the state of the tide. As a consequence of this, all the flotsam
from the lagoon shores of the inhabited islands is swept away
towards the mouth of the lagoon, and it is the well-justified
island custom when anything has gone adrift to look for it first
on the shores of Pulu Luar. With the rarer westerly winds the
opposite effect is produced, and a strong current running into
the entrance of the lagoon is brought about ; the waters of the
lagoon become banked up, and much destruction of the sandy
beaches usually takes place. The wind-driven lagoon water
becomes raised above its normal level, sand is swept away, and
the old barrier stratum exposed; trees are undermined, and
bushes uprooted ; and did a visitor spend only a few days in
the atoll, after the coming of a strong westerly wind, he would
surely give the world his certain conviction that the sea was
rapidly encroaching on the land. In this way, I feel sure,
arise some of the conflicting opinions of men who spend but a few
days or hours in an atoll; for the visit may chance to be made
after a period of denudation by one wind, or after a period of
sand accumulation caused by the antagonistic wind. Were
the length of the stay to be measured by months or
years, the visitor would see the banks come back: the
ancient barrier stratum covered again with a thick bed of

THE LAGOON 199

white coral sand; and the whole beach raised again by the
agency of the wind and waves. He would even see, in all
probability, a series of little terraces made in the sandy shores
as gradually waning high tides built to their maximum levels ;
then he might possibly think that the land was steadily gain-
ing on the sea. It is always assumed in estimating any
changes in the shore-line that the surface of the ocean stands
constantly at the same level, and yet, strange though this may
appear, it is not always safe to assume this. It is apparently
one of the strange phenomena of far-out blue water that its
mean surface-level is subject to considerable variation. Of
this strange fluctuation in level, such a place as the atoll forms
a very interesting index—it is the Plimsoll-mark of the
ocean’s variations. During my stay in the islands I have seen,
for a week at a time, and without any variation in the tides,
the curious condition of the barrier never being once exposed
at low tide. The tides went through their usual excursions,
the high tide being higher than normal, and the low tide not
so low as normal; but the whole surface-level of the ocean
was set at a higher plane than was ordinary. ‘his state of
affairs is doubtless brought about by the banking up of the waters
of the Indian Ocean, and the event happened after a report had
come over the cable that a violent storm of long duration had
raged in the neigbourhood of the island of Rodriguez. It was
the time of neap tides, and so the increased high tides did not
reach the striking level that they would have done had the
event happened during spring tides ; but the phenomenon was a
very striking one. It is easy to see that if a change of ocean-
level—even of very brief duration—were to be accompanied
by a combination of high winds and spring tides, imprints
would be left on the islands which might appear to indicate a
shore movement of several feet. It would not be at all likely
that these appearances would be rightly interpreted by a
visitor to the atoll who had not seen the process in its
making.

The average tidal variation is from three to four feet, and
the tide’s turn takes, as a rule, about an hour and a half to

200 CORAL AND ATOLLS

travel from the south of the lagoon to the north—a distance
of about eight miles. The amount of tidal variation is an
important factor in determining island elevation, and this will
be again referred to in the consideration of the mode of
formation of the islands.

The lagoon being shallow, calm, and more or less enclosed,
is considerably warmer than the surrounding ocean, and some
parts—as for instance the back water enclosed by the sand
spit at the northern end of Pulu Panjang—are distinctly hot
after a day of calm and sunshine. A surface temperature
of 85° Fahr. is a very common one for the open waters of the
lagoon.

1 XI.

<q
Fe
py

LAGOON.

K

A SUNSET IN TH

CHAPTER XVII
THE CLIMATE AND PREVALENT WINDS

From its position in mid-ocean, and 12° 9’ south of the
Equator, the group derives all its peculiarities of climate. It
lies to the south of the monsoon area, that is to say, the
South-East Trade wind is not normally reflected on its course for
any constant part of the year; for though winds other than the
South-East Trades do reach the islands, there is no season of
the year in which a reversed wind constantly prevails. The
slackening of the Trades, from north to south, reaches the
group in the early months of the year, and then “doldrum ”
calms and uncertain winds are the prevailing type of weather ;
but it is quite incorrect to speak of the “change of the mon-
soon ” in the Cocos-Keeling group, although the phrase is by
no means infrequently used.

As it is south of the monsoon area, and therefore has no
monsoon seasons, so it is too far north—too near the
Equator—to be subject to obvious hot and cold seasons; there
is a period of the year normally more dry and hotter than the
remaining months, but the mid-oceanic position of the group
deprives even these seasonal variations of any really obvious
effect, the variation being, when averaged, but very slight in
amount.

All the data for these meteorological records have been
taken with the Telegraph Company’s instruments, and those
used here cover a period from Dec. 1901 to Dec. 1905,
They may be relied on both as regards the accuracy of the
instruments and of the observers.

It must be remembered, in estimating the amount of
seasonal variation to which the group is subjected, that all

the observations have been taken on Pulu Tikus, which is a
201

202 CORAL AND ATOLLS

small and comparatively sparsely vegetated island, and that
as a consequence, the rainfall, as registered for this island,
probably does not represent truly the rainfall of the group as
a whole. During those months in which the minimum rain-
fall is registered by the Pulu Tikus instrument, rain, in some
considerable quantities, falls on Pulu Atas and Pulu Panjang.
Were the records to be taken on these larger pieces of land,
which lie six miles away, I think there is no doubt that the
annual rainfall would have a higher index, and at the same
time be more uniform throughout the course of the year.

That these meteorological records help in the under-
standing of the scant, but interesting, fauna and flora is only
natural; for every living thing must reach the islands as a
chance plaything of the winds or waves; but since the true
explanation of the origin of atolls lies in the appreciation of
the physical forces with which they are surrounded, they should
help, if rightly interpreted, in understanding the method of
island building. For this reason it is very fortunate that so
accurate a set of observations is available, for the details of
the climatic conditions of the group which have hitherto been
published do not give a satisfactory working basis. .

THe PrevaLtent Winps.—The prevailing winds are the
S.E. and the E.S.E. which constitute the Trade winds. The
normal variation of the Trades includes winds blowing from
the 8.E., E.S.E., and E., and the winds from these quarters
blow on an average for 300 days in every year. It must not
be thought, however, that the odd 65 days are consecutive
days occupied by any change of the Trades, for as a matter of
fact the 300 days are very equally distributed over the twelve
months.

The wind blows from the S.E., E.S.E., or E. as follows, the

numbers being averaged from the records of four years:

IN
Jan. 23 days Feb. 18 days March 23 days April 23 days
May 23. 5, June 26__,, erally 2h ee DAWG PAC 255
Sepin 2s, Oects 289%; INOW 213} op Dec zeae,

CLIMATE AND PREVALENT WINDS 203

The early months of the year are those in which the wind
is subject to the greatest variation, and in these months

Fic. 58.

DIAGRAM TO SHOW THE PREVALENT WINDS OF THE ATOLL.

The figures opposite each point of the compass show the average
number of days in a year in which the wind comes from that
point.

northerly winds blow at intervals, and these winds bring rain-
squalls and unsettled weather generally.

204 CORAL AND ATOLLS

Although this slight variation is clearly shown in the
table, it will be at once seen that the monsoon influence is not
marked ; there is nothing that could be entitled to the name of
“the change of the monsoon.”

The Trades blow a steady cool breeze day and night, and
they become another agent, and a potent one, in keeping up
that uniformity of conditions which is the outstanding feature
of the climate of the group. They give to the islands their
peculiar individuality, their sounds of rustling palm-leaves
and breaking surf, they give them their even temperature and
steady barometer; and more than this, they have impressed
themselves on the very formation and structure of the group
itself.

The sky is seldom cloudless, partly of course for the
reason that so large an expanse of sky is always in view, but |
partly too for the reason well known to sailors, that clouds as
a rule do mark the site of coral islands. Captain Maury, in his
“Sailing Directions,’ even goes further than this, and says:
“ These cloud piles . . . are often seen to overhang the lowest
islet of the tropics, and even to stand above coral patches and
hidden reefs,‘a cloud by day’ to serve as a beacon to the
lonely mariner out there at sea.” A lower stratum of large
fleecy cumuli moving steadily in the line of the Trades, and
an upper and very high stratum of thin long-drawn cirrus,
appearing by contrast almost stationary; this is the most
ordinary sky picture, and will serve for the great majority of
days in the year. The upper stratum has, at times, a real
opposite movement to the lower, and whilst the cumulus
masses are drifting from the south and east, the overlying
cirrus comes from the north and west. This is well known to
the islanders, and is looked on, and rightly, as a sign of coming
rain. Although the northerly winds form such an incon-
spicuous constituent, when seen in the graphic representation
of the winds, still they are important winds for the islands, for
they bring the rain; they bring to-day, and doubtless have
brought in the past, new waifs and strays as tentative additions
to the fauna of the islands. After northerly winds, even of

CLIMATE AND PREVALENT WINDS 205

short duration, dragonflies are almost invariably seen on the
islands. They do not make a successful footing, and they
do not become permanent additions to the fauna, doubtless
because of the scarcity of fresh water; but they show by their
coming the lines by which pioneer species may invade distant
islands and, finding a suitable environment, become settlers.
Stray birds, as swallows and wagtails, have come after these
winds, but their stay is brief.

It is the northerly and westerly winds that bring the
cyclones which have played so important a part in the history
of the Ross settlement, and considering the extreme lowness of
the islands, and the normal force of the wind and waves, it seems
wonderful that anything should be left standing after a cyclone
has passed the group. Four recorded cyclones have visited
the islands, and on two occasions have practically left them
waste.

In 1862 a cyclone wrecked the settlement; again in 1876,
on Jan. 28, terrible damage was done, and the flourishing
condition of the little band of settlers was turned again to
hardships and privations. The third cyclone occurred on
Feb. 4, 1893; the wind was mostly from the north and north-
west, the barometer remained between 29°82 and 29°45 inches
from 6 A.M. on Feb. 4 till 6 a.m. on Feb. 6, and during that
time 30,905 coconut palms were uprooted and much other
damage was done. On March 4, 1902, a fourth cyclone
swept over the group, and it is recorded by the Pulu Tikus
meteorological records. At 9 a.m. the barometer stood at
29°40 and between 10 a.m. and noon fell to 28°95 inches.
The wind, which at first blew steadily from the south-west,
gradually worked round with increasing force until it blew
from the nor’-nor’-west with cyclonic violence. No measure-
ment of rainfall was possible as the rain-gauge was wrecked,
nor was it possible to take other observations. Over 2000
coconut palms were uprooted on Pulu Tikus alone, and up-
wards of 300,000 trees were blown down in the entire group,
and yet this was by no means comparable to the former
cyclones for its violence. In the cyclone of 1876, of which a

206 CORAL AND ATOLLS

graphic hearsay account is given by Dr. H. O. Forbes, the
aneroids are said to have fallen to 26°5 inches and then to
have stuck, and the mercurial readings became impossible as
the storm carried away the barometer. Certain it is that this
storm did great damage, and to-day scars in trees are pointed
out and carefully remembered as the exit or entrance marks
of sheets of roofing-iron; numbers of trees were uprooted and
whole areas of the islands were denuded; and what is of
greater interest, the waves rose and deposited wrack and
jetsam 150 yards above the normal high-water mark.

RatnraLu.—the annual rainfall, as taken by the Pulu Tikus
instrument, gives an average of 70°23 inches; the actual annual
amounts being :

In 1902 . : ‘ é : 5 : . 50°73 inches
SSE eG Re A OM mL GRC S|
In 1904 . , : ; ; 5 : e(tomelle |b.
In 1905 . : i : : ; ie TD TA ee

The fall of rain as registered for Pulu Tikus is probably
the minimum of any portion of the group, and the rainfall of
Pulu Atas or of Pulu Panjang could be taken at a figure
considerably higher. The annual rainfall is therefore a heavy
one, although Dr. Guppy states, after a stay of ten weeks in
August and thé two following months, that “probably it does _
exceed 40 inches in the year.”

One curious phenomenon needs especial mention in con-
nection with the subject of the rainfall, and that is the sudden
advent of a shower of rain from an absolutely cloudless sky.
When all the sky is blue and no cloud is to be seen from
horizon to horizon, save those masses lying far out to sea, a
shower will suddenly descend. On the hottest and most
calm of days the phenomenon is most prone to occur, and
the pelting of large drops of rain at such a time is a very
striking occurrence. It has impressed the native mind, for
in some Malay communities these showers are considered as
signs of ill-luck—and to be caught in a shower from a cloud-
less sky foretells a death by violence.

CLIMATE AND PREVALENT WINDS 207

As a rule the greatest rainfall occurs in the first half of
the year, April being the wettest month and October the
driest ; the average fall for each month being:

Jan. 8-03ins. Feb. 3°60 ins. March 8°37 ins. April 12°50 ins,
May 876 ,, June 7°65 ,, dil 9 3 J\hegs ACG) op
Sept. 1°42 ,, Oct el02 05 INONeuecollen. Dee, sil 5,

(The readings for November are probably exceptional, for
in Nov. 1903, 14°38 inches, the maximum rainfall for a
single day, fell. Leaving 1903 out and taking only 1902,
1904, and 1905, the average rainfall for November is 1:85
inches.)

This amount of rain is distributed over an average of 156
days every year. The actual number of days on which rain
fell being as follows:

In 1902 . : 5 : . : ; . 148 wet days
In 1903 . : : 5 ; ‘ : oO ae a
In 1904 . ; ; : : : : 6 ASE! Sse Sp
In 1905 . 3 . : ‘ ; 5 ee ee ae rs

These 156 wet days are distributed, on the average, among
the. twelve months as follows:

IN
Jan. 9:5 days Feb. 10-7 days March 14-5 days April 18°8 days
iWiany WG. alms ISO 55 Uy alse PAC OL 2 es
epimers Oct lO0Re ee me NOvas 1:3) = IDwe, © 85)

TEMPERATURE.—The mean temperature of the islands is
81° and the mean daily variation is 9:06°. It must be re-
membered in making estimates of the temperature of the
group that the maximum is a very sustained one, the greatest
heat does not last for merely an hour or two of noontide
sunshine, but, with very trivial variations, is felt during a
greater part of the day.

208 CORAL AND ATOLLS

The Pulu Tikus observations differ somewhat markedly
from those that have previously been reported from the group.
They differ, as do the barometric readings, from those that
have been published in the Blue Books of the Straits Settle-
ments, and which were based on the observations taken by
the Clunies-Ross instruments. Most of the discrepancy is
doubtless accounted for by the fact that the Pulu Tikus
readings are taken at 9 a.M.and 3 P.M., whilst the Clunies- Ross
observations are recorded at 6 A.M. only. The mean tempera-
ture is therefore actually several degrees higher than the
generally quoted 72°. Dr. Guppy’s observations on the
temperature are also misleading; he gives the mean tem-
perature as “about 78°50°”; but there is no doubt that in
the limits of his short stay he had no real basis for fixing
a mean temperature without consulting the Clunies-Ross
readings. |

The highest temperature recorded during the time that
my results cover is 94°3° on April 2, 1903, and the lowest is
61:0° on August 4, 1904, whilst the maximum daily variation
is 24:80°.

The average maximum and minimum readings for the
different months are as follows:

Max. Min. Max. Min.
Jan. ¥ or 69) EY Be GRR Feb. : . 90°4° h2eon
March . 5 BORO” (29° April. 5 SkORs 73°8°
IER? 5 5 teehay 73°0° June. 5 lo 713°
July 3 5 eae" 70:0° Aug. : 5 aPILY 69°4°
Sept. . 5 oak? Talo? Oct. ; 5 OS? Tey
Nov. . . S9c4> 2 Dee. : oo LOM C2BP

Barometric ReEADINGS.—The barometer shows steady
readings throughout the course of the year, the average
monthly variation being *220 ins. ‘The average maximum
reading of all the months is 30°128 ins., which is considerably
higher than that recorded by the Clunies-Ross instrument.

The averages of the maximum and minimum readings of
each month for four years are as follows:

CLIMATE AND PREVALENT WINDS 209

Max. MIN. Max. MIN.
Jan. . 30°091 ins. 29-991 ins. Feb. . 30:091 ins. 29-935 ins.
March . 30:062 ,, 2ORO2Kaa eon Ss BOS aS) PAO
May: +; 30:092) ,, 29:0 fae. Amines es BOOS 2B CB)
dulye 2) o0:102” ,, AU ONG. ANNE 9% BIOCIER a PASS) o
Sept. . 30°154 ,, 30°019 ,, Oct. . 307144 ,, ZS IS) 55
INOvee 2 OU 29) = ASSO 3 Dee. a alee 29-930

The highest reading recorded by the Pulu Tikus instru-
ments is 30°966 ins. in December 1904, and the lowest is
28°95 ins. during the cyclone of March 1902; but in April
1888 a reading of 28-00 ins. is recorded in the Blue Book
without any accompanying note of unusual weather.

Apart from these bare records of the instrumental readings,
some other influences of the meteorology of the place must be
told. So tiny a speck of dry land in the midst of so much water
must of necessity have some peculiarities of its climatic con-
ditions, and besides the perpetual mystery of the few incon-
siderable feet of coral débris remaining intact in so restless a
spot, there are many other individualities of the physical
surroundings of the atoll that are of interest. The Trade
winds are, of course, the outstanding feature of the atoll’s
climate, and they are the determining influence which must
always be appealed to in solving most of the problems con-
nected with the coral circle. Itis said that the region of coral
islands is usually one of “light winds and calms, even where the
Trades are blowing strongly all around them” (Capt. Maury).
It is certainly true that the lagoon and the leeward islands are
places of comparative calm, but this is more for the reason
that the ring of vegetation, 60 feet high, gives them a very real
protection than that the Trades are lulled in their neighbour-
hood. There are days, but at the most only one or two in
each year, that may truly be called calm, and for the rest the
Trades blow a steady breeze, with a force of from 10 to 25
miles an hour, making every coconut palm toss and sway, and
keeping up a steady sea within the lagoon. Exceptional times
of course occur in which the Trades die down abnormally, and

then, as for instance in the early months of 1906, there are
O

210 CORAL AND ATOLLS

prolonged periods in which the lagoon is almost unrippled, and
even the windward islands are almost deserted by the breeze.
The high annual rainfall of 70°23 inches gives rise to the
curious phenomenon of the fresh-water wells in which water fit
for drinking rises and falls with the movements of the tide. On
Pulu Tikus no fresh water may be found in wells, and though,
in the more rainy months, fresh water is to be found at a depth
of a foot or so below the surface of the land, it soon becomes
brackish, and finally entirely salt ; long rainless spells therefore
occasion great inconvenience, and the island vegetation soon
becomes parched in the hotter part of the year. It is the
presence of so much rain that prevents the accumulation of
guano on the bird-covered Keeling Island, and it is the rain
that, carrying the droppings in a state of solution through the
permeable coral, causes the deposition of the phosphates which
occur in the upper layers of the soil of Cocos-Keeling, and
which have made Christmas Island of so great a commercial
importance.

From time to time in the history of the islands long rain-
less spells have occurred, and during 70 consecutive days at
the end of 1905 only °37 of an inch fell; and this very small
quantity was distributed over nine separate showers. During
the whole of these dry periods rain-clouds are for ever
coming up and passing by the islands, and it has long been
noticed here that a rain mass will divide before reaching the
coral ring, and part pass away to the south and part to the
north. This is an interesting fact and is of such common
occurrence that an explanation must be readily to hand; and
it is doubtless this, that the overheated coral, which in the sun-
light dances and quivers with a dazzling whiteness, gives rise to
such a column of heated air that the upward-rushing mass is
sufficient to cut the oncoming rain-cloud into two parts. This
is the explanation, I have since learned, that was also furnished
by Mr. J. D. Hague to account for a similar state of things in
Jarvis Island. Quite apart from rain-storms, clouds may be
often seen to melt away as they come over the group, be-
coming thinner and thinner as they meet the hot air and

CLIMATE AND PREVALENT WINDS 211

finally vanishing as they pass over one of the larger land
masses. ‘I'he opposite phenomenon also occurs, and clouds
are formed over the lagoon by the extra evaporation from its
surface ; these are the clouds that, on calm fine days, are
noticed by the sailor to stand poised above atolls. The lagoon
is shallow, and is hotter than the ocean water; there is a
tremendous evaporation going on from all over its surface,
and the water vapour is swept upwards in the rising column
of hot air till, meeting with the colder layers above, it con-
denses and forms the private sky of the lagoon. The island
ring is in fact the one and only radiating object in the midst
of the ocean, but since it is really a sponge work it can never
rapidly cool at night, because of the great warm-water jacket
which surrounds and permeates it; the consequence is that
the deposition of dew is a very rare phenomenon indeed.
A curious state of things is therefore brought about in the
physical conditions of the islands, for the excessive radiation
from the heated surface prevents the fall of rain, and yet the
night fall of temperature of the surface of the land cannot
become so great that dew can be deposited. There is then no
compensation here, but only a vicious circle of overheated coral
and cleft rain-clouds, without the accompanying blessing of dew.

The result of dry weather when prolonged is accordingly
severe and the vegetation suffers; the undergrowth fades and
becomes parched and brown; the leaves fall from the trees, and
the coconuts drop prematurely, leaving the islands very bare
and waste. It is the opinion of the Clunies-Ross family that,
despite the increased planting of coconut palms, these dry
periods are becoming more frequent and more prolonged ;
though strange to say the wells on Pulu Selma never become
exhausted, despite there being no restriction placed on water
using.

Thunder and lightning are very rare, and when observed
the phenomenon is usually far distant; storms other than
cyclones play no part in the history of the islands, for the
absence of large areas of land to undergo rapid heating and
cooling saves them from sudden atmospheric variations.

CHAPTER XVIII

THE THEORIES OF ATOLL AND REEF FORMATION
PUT FORWARD UNTIL THE PUBLICATION OF
DARWIN’S THEORY OF SUBSIDENCE

Ir is only to be expected that so strange a geographical
formation as a coral atoll should have given rise to abun-
dant speculation on the part of thinking people; and when
the origin of these isolated rings of dry land in the midst of
great oceans came first to be considered much difference
of opinion prevailed. Even to-day it cannot be said that
there is complete unanimity. |

Atolls have always been mysterious, and I do not think
that I underrate the state of the present-day knowledge when
I say that the mystery even now is not entirely cleared away.
When, for the explanation of a fact in Nature, you may find
to hand many rival theories, whose several makers put them
forward as individually adequate, then it is certain that the
explanation of that fact is still to be made clear, either by
reconciling the contending theories or by establishing a new
one in the place of all.

Concerning the origin of atolls, the warfare of rival
explanations cannot be said to have yet ceased; and the
number of theories that have been brought forward from time
to time shows plainly that the halo of mystery, with which
the early navigator endowed the coral rings, has to this day
remained in the minds of many. ‘Travellers’ accounts of all
that they saw and heard, when far from their fellow-men,
have undergone a regular evolution, just as the travellers
themselves have done: and the evolution and progress of
modern scientific travel is easily followed in the literature left

behind by successive generations of pioneers. In the early
212

ATOLL AND REEF FORMATION — 213

days it was but natural that, after a prolonged and uncom-
fortable voyage, the sighting of a calm stretch of water girt
round with a ring of luxuriant islets should have impressed
the romantic voyager as something mysterious in which he
saw the hand of Providence. And when he came to know
that the whole strange structure was the work of living
creatures, he, following the lines of thought of his day, read
into the whole the extraordinary human vanity of the lower
creatures striving with ordered activity to create new fields
for man’s enterprise.

The coral rock of which the islands were built was the
work of living things, and these living things, although he
did not appreciate them zoologically, he knew to be at work
creating dry land in the midst of oceans. The building of
islands was their function, and they built them of a circular
shape, so that within the barrier which they, by their life and
death had raised, their offspring might be safely reared, and
sent forth to carry on the good work of island building. Such
were the views of Flinders, and the ordered circular building
by the zoophytes is characteristic of the time of the acceptance
of these theories. Nothing seemed more-true and yet more
wonderful than the little beasts conspiring together to make
that circle of dry land, knowing it to be the proper way to
enclose a piece of water, where they might in safety establish
a breeding-ground and a nursery for their kind—and a safe
anchorage for the mariner.

With the advent of the zoological traveller—the scientific
man who wandered over the earth during the Renaissance
period of zoology—this mode of thought became for the most
part obsolete among educated people. But it is charac-
teristic of the Romantic age of zoological travel, that the
impressions which it stamped upon the popular mind have
been lasting ones; and the reason for this is very easy to see.
The traveller of the Romantic age was a man who made a
voyage of adventure. He, as a traveller, ohserved with vary-
ing accuracy the phenomena that he met with; and assimilated
with varying credulity the stories that were told him; and

214 CORAL AND ATOLLS

then he gave his adventures and his Natural History to the
public as an interesting pot-pourri.

The scientific man of to-day can reach the popular mind
in no such way; he has no record of stirring adventure; he is
as a rule a specialist whose bare results are not acceptable to a
wide public. The result of all this is that when Natural His-
tory is served up for that period of childhood when Nature’s
wonders are of interest to nearly all, and when most people
gather their ideas of Nature, it is from the writings of the
Romantic age that the greater part of the information is
culled. It is in travellers’ tales that we as children
delighted, and the plates that adorned these tales bid fair to
outlast the tales themselves. All thisis but natural, for it must
be admitted that it is the fault of modern zoological literature
that it is devoid of interest, except to a very limited number —
of specialists. JI do not think that the man exists who, merely
possessing a thirst for information and a general love of
Nature, could read with any pleasure the mass of modern
literature on the corals and coral islands. It is easy to recall
the time when we were told or when we read of coral islands,
and revive a thought association of busy living creatures
building, building—always building and dying—in the midst
of the ocean. By their efforts these living creatures made the
islands where the waving palm trees grow, and where the
weary mariner may rest in the calm lagoon, and gather in
the green island groves all the splendid fruits of the tropics.

It is, I believe, a common enough idea that the coral
creatures build coral, much as bees build honeycomb, but the
comb is to them their dwelling-place and their mausoleum, and
the comb is raised up and up from the depths of the sea as
the busy creatures build below. Such ideas are of course no
longer tenable, and serve only as landmarks of the time when
the whole of Nature’s happenings were explained by reading
into the great puzzle the direct intervention of Providence on
man’s behalf, and by giving to the lowest of creatures most
of the emotions and all the instincts of human _ beings.
Although this mode of thought has long since departed from

ATOLL AND REEF FORMATION — 215

all ordered investigations of scientific subjects, still, in the
history of the theories of atoll formation, it lasted long; and
many careful researches must be made and accepted before
an explanation so easy and so comfortable is quite forgotten.

It was in a great measure the increasing knowledge of the
conditions of the ocean bottom that tended most to modify
the prevalent ideas of the nature of atolls, and as we have
learned more and more of ocean soundings so have our ideas
of coral islands been slowly corrected. Many of the theories
that were put forward in the early days are not without
interest to-day, for in many cases the modern theories are but
compromises, in which the remnants of an older explanation
are easily detected.

One of the older theories that must claim some share of
attention—for, in a modified condition, it survives still—is that
which seeks to explain the usual form of atolls by assuming’
the island ring to be a coral growth built around the crater of
a volcano, The well-known cone shape of the typical voleano
gives rise to the characteristic form of the atoll; the lagoon is
the water-filled crater, round which the islands stand in a
more or less perfect ring, being built wherever the rim of the
erater comes sufficiently near to the surface of the sea. This
theory received authority by appearing in Lyell’s “ Principles of
Geology” (vol. 11.) in 1832; but, even if no rival theory had
so soon been destined to overthrow it, the many and obvious
objections to it would have rendered its general acceptance
impossible. No terrestrial voleanoes are known with craters
so vast as are many atoll lagoons, and no mountain chain
so strangely level as are the low island groups scattered in
the Pacific Ocean. Despite the improbability of this view,
the majority of the stray visitors who have come to this atoll
have fancied that somewhere in the lagoon was the deep hole
of the crater, and have believed themselves to be living on the
rim of a vast volcano. The Rev. E. C. Spicer, who as naturalist
accompanied H.M.S. Hspoir in 1885 on behalf of the Straits
Government, was struck by the appearance of the narrow strip
of land and the enclosed lagoon, and said “the resemblance

216 CORAL AND ATOLLS

of the whole to a giant crater is very striking.” Even to-day
it is held by some that atolls may be formed on ancient
craters; Prof. A. Agassiz says “atolls may also be formed
upon the denuded rim of a voleano crater as at Totoya and
Thombia in Fiji.”* These however, must be very special
cases and do not justify any general statement.

It was with the voyage of the Beagle that the era of the
specialist zoological traveller dawned, and it was as the result of
the visit of her great naturalist to this group in 1836 that
modern scientific thought was directed towards the question of
the origin of atolls.

Although it is commonly assumed that Darwin formulated
his views of atoll formation as a result of his study of Cocos-
Keeling, still he definitely states that “the whole theory was
thought out on the West Coast of South America, before I had.
ever seen a coral reef.” + Cocos-Keeling was the first atoll
that Darwin ever saw, and the only one he ever examined ;
and in it he believed he found evidences of those events that
he had pictured as constituting the past life of an atoll. To
some extent, doubtless, Darwin was influenced by what was
told him in the islands, and the whole of this information was
certainly not accurate, but the observations made by Darwin
are to-day as true as when he recorded them.

Of Darwin's theory of atoll and reef formation it must be
said, that it was a piece of reasoning more masterful than any
of its rivals, and more complete in its detail; but it must
also be said that the theory of subsidence when invoked in the
formation of the Cocos-Keeling atoll does not meet with any
support. The theory of SuBsIDENcE must be referred to in
some detail; but it is necessary at once to say that many of
the views of Darwin’s successors were views that Darwin him-
self had originated, and the validity of which he had admitted
under certain conditions: in the discussion of the origin of
reefs and atolls this fact is often overlooked.

* See Nature, 1903, p. 547.
+ See the Introduction to Charles Darw'n’s ‘Coral Reefs,” by J. W.
Williams (and other quotations from the same source).

ATOLL AND REEF FORMATION — 217

As the theory of SUBSIDENCE ushers in the modern epoch
of coral-reef theories, it will be best to explain, before these
views are discussed, what problems the question offers, and
what details must be explained by a theory that is to be a
useful one.

In order to save wearying repetition, these problems will be
briefly taken in order, and the merits and the shortcomings of
the various theories will be measured for each individual
problem. In every case the consideration of the subject falls
naturally into three divisions: the views held before Darwin’s
day, Darwin’s work itself, and the theories put forward in
opposition to Darwin’s representing three distinct epochs in
the evolution of our ideas of coral-island formation.

THE First Pornr that requires an explanation is the
nature and origin of the submarine bank on which the atoll
rests :—i.e., on what structure did the corals first start to
build the islands? The question of the nature of the “ basis ”
assumes great importance from the fact that the depth of water
in which the reef-forming corals can grow is strictly limited.
The basis of every atoll must therefore be an elevation, rising
so near to the surface that reef-building corals may flourish
on 1t.

This fact consequently presupposes the existence of a very
large number of submarine mountains, dotted all over the
Indian and Pacific Oceans, and all with their peaks approxi-
mately of the same height.

Now no terrestrial formation such as this is known to exist ;
and it was this fact that led Darwin—rather than assume
such an apparently impossible condition of submarine con-
figuration—to bring forward his well-known theory of subsi-
dence. But it must always be borne in mind that he
admitted that, were such banks at hand, they could furnish
the basis for atolls.

To Semper he wrote (Oct. 2, 1879): “I always foresaw that
a bank at the proper depth beneath the surface would give
rise to a reef which could not be distinguished from an atoll
formed during subsidence”; and again to Alexander Agassiz

218 CORAL AND ATOLLS

(May 5, 1881): “I have expressly said that a bank at the
proper depth would give rise to an atoll, which could not be
distinguished from one formed during subsidence.”

It is often overlooked that Darwin ever admitted any
alternate hypothesis to that of subsidence, and it is well, there-
fore, to quote his own words to show that he fully recognised
a state of things which, subsequently brought forward anew,
became a rival theory. Not only did he admit that, were
banks existent in the ocean of the necessary depth below the
surface, they would become the sites of atolls in a coral-
bearing sea; but he further acknowledged that, were signs of
elevation and not depression common in atolls, his theory
would not be valid. To Semper he wrote: “ But I fully agree
with you that such cases as that of the Pelew Islands, if of at
all frequent occurrence, would make my general conclusions of —
but very little value.”

The Pelew Islands are peaks in the Caroline group, where
Semper found coral reefs raised up from 400 to 500 feet, and
only 60 miles away from them are typical atolls.

Having seen what was Darwin’s attitude towards other
views of atoll formation, it will be well to consider in some
detail his theory of subsidence; for it is almost entirely in
connection with the question of the foundation of the atoll
structure that his theory has its individuality and its
importance.

No better idea of it can be given briefly than Darwin’s own
diagram, which shows at a glance the main points of the
explanation that he furnished for all coral formations.

In the first place, a high oceanic island is pictured,
situated in a sea favourable to coral formation; on the
submarine slopes of this island corals will flourish and form
a reef around the land. (See Illustration A.)

This first stage is of course no fictitious state of things,
for many instances of high islands surrounded by barriers are
known.

Now if the whole structure be imagined to sink slowly,
the corals of the barrier will grow, and compensate for the

ATOLL AND REEF FORMATION — 219

subsidence, and Darwin assumed that the corals of the outer
edge would grow best; for he said that the corals love the
surf, where they are better aerated and more abundantly fed.
There is no compensating factor for the sinking land, and if
this process continues for some time, the second stage is

Fic. 59.

THE ILLUSTRATION USED BY DARWIN TO MAKE CLEAR THE
STAGES OF “SUBSIDENCE” IN ATOLL FORMATION.

arrived at, and the island remains, surrounded by a moat, the
lagoon channel, which is again encircled by the barrier reef, now
standing out some distance from the island. This second stage
is again a geographic reality. (See Illustration B.)

From this stage the atoll is easily derived ; the subsidence
continues, the reef outside compensates by coral growth, the
land, uncompensated, finally disappears within; thus the
typical lagoon is formed :—a salt lake, in which a peak of
land is buried. Such, in brief, is the theory of subsidence.

The basis of the atoll is therefore, on this theory, a sinking

220 CORAL AND ATOLLS

one ; and, in an advanced case, is a bed of ever-thickening coral ;
for as the land continues to sink, the upward building of the
coral adds new layers to the accumulating coral bed.

THE Second Point that any theory attempting the expla-
nation of atoll formation must make clear is the regularity of
the tendency to a circular shape. So constant is this feature
that the explanation of it must be intimately bound up with
the proper history of the development of the structure.
Darwin’s theory possessed this charm, that it carried with it,
in the picture of the disappearing peak with its outlying reef,
an idea of the development of a ring formation, which is very
easy to understand.

The atoll shape comes about in the making of the structure
by subsidence, and is afterwards maintained by the corals
of the outer edge of the reef growing faster than those on the —
inner edge. The corals of the central portion are those that
fare worst both as regards their supply of oxygen and of food,
and they also have to contend against the adverse influences
of sediment and impure water; in this way the lagoon
becomes a stereotyped feature of the atoll. The vigorous
growing edge of the reef afterwards became the site of an
accumulation of wave-tossed débris, which gradually rose into
the dignity of island dry land; and so a circular ring of
islands was formed around a central lagoon. The formation
of all the characteristic features of a typical atoll is, there-
fore, very naturally explained by Darwin’s theory of the reef
foundation.

As a side issue of this aspect of the question, a THIRD
POINT arises, for it is obvious that the shape of the individual
islands demands some explanation. In Darwin’s theory there
is no special provision for dealing with this problem.

THe Fourts Pornt is that any theory which deals with the
formation of atolls, and attempts to explain their formation,
must also take into account the other coral formations—the
fringing reef and the barrier reef. In respect of this detail
the theory of Subsidence was quite satisfactory, for the explana-
tion brought all these structures into an ordered sequence.

CHAPTER XIX

THE THEORIES OF ATOLL AND REEF FORMATION PUT
FORWARD SINCE THE PUBLICATION OF DARWIN’S
THEORY—THE THEORY OF “SOLUTION”

THE voyage of the Beagle had given Darwin the information
necessary for completing the development of his theory of
atoll formation. ‘he voyage of the Challenger furnished the
material for the origin of the rival attempt at the solution of
the problem, and the formation of a definite new theory to
replace that of Darwin. |

On April 5, 1880, Dr. John (now Sir John) Murray
published, in the Proceedings of the Royal Society of Edinburgh,
a paper on “The Structure and Origin of Coral Reefs and
Islands.”

The new theory thus brought forward differed essentially
from the theory of “Subsidence” in practically every detail,
for Murray started with the knowledge—which was not avail-
able for Darwin—that submarine banks did exist in the
ocean. In this new theory, the basis of the islands or reefs
was not a sinking one, and in the place of the sinking peak of
land, Murray pictured a volcanic elevation of the floor of the
ocean. This elevation reaches to the depth limit of coral
growth by one of two processes: if it were originally pro-
jecting above the water it becomes worn down by the action
of the waves, and if it is too far below the surface, it becomes
raised by the deposit upon it of Globigerina and Pteropod ooze.

“The soundings of the Zuscorora and Challenger have made
known numerous submarine elevations; mountains rising
from the general level of the ocean bed, at a depth of 2500
or 3000 fathoms, up to within a few hundred fathoms of the

surface. Although now capped and flanked by deposits of
22

222 CORAL AND ATOLLS

Globigerina and Pteropod ooze, these mountains were most
probably originally formed by volcanic eruption ” (p. 507).

Such, in brief, is the basis which Murray furnished for the
building of reefs and islands. Upon this basis the corals form
a reef, and at the outer edge of the reef the corals grow
best, because here they are enabled to obtain most food.
Murray gave up the idea that the corals of the edge of the
reef thrived most because of their better supply of oxygen,
but he retained the notion that the oceanic currents brought
more food to the edge of the reef, and so enabled the corals
to grow more freely. The lagoon becomes formed, and con-
tinues to grow in size, by the “ solution” of the dead coral which
forms the central part of the reef—in Murray’s own words:
“Larger quantities of lime are carried away in solution as a
bicarbonate from the lagoon than are secreted by the animals.
which can still live in it; the lagoon thus becomes widened
and deepened” (p. 511).

The whole theory of atoll formation on the volcanic
mountain base is summed up in this way in the original
paper. “That when coral plantations build up from sub-
marine banks they assume an atoll form, owing to the more
abundant supply of food to the outer margins, and the re-
moval of dead coral rock from the interior portions by currents
and by the action of the carbonic acid dissolved in sea water”
(p. 517). This theory has been further elaborated by Murray,
by Murray and Irvine (Nature, June 12, 1896), and by James
G. Ross (Wature, March 15, 1880); and it has received new
embroideries from the work of Mr. Stanley Gardiner in the
Maldives and Laccadives.

The prevalent shape of the constituent islands of the atoll
ring appears to be the subject of no special attention in this
theory, and it is not apparent how other coral structures, as
fringing reefs and barrier reefs standing far out from shore,
are made at all easy of understanding.

The views of Mr. Stanley Gardiner have been put for-
ward in his work on “ The Maldives and Laceadives,” published
in 1902, and more fancifully in an article in Nature (p. 371,

ATOLL AND REEF FORMATION — 223

1904). In these publications, at least four distinct methods
of atoll formation are postulated, and these are briefly stated
as follows :

Mertuop No. 1.—An elevation of the ocean floor becomes
covered with sediment; it reaches the depth at which reef
corals flourish, and a reef is developed upon it. “ Boring
organisms enter on its central part and cause the rock to
decay. Sand-feeders follow and triturate up the fragments,
throwing a constant stream of fine mud into suspension in the
water, to be removed by tidal and other currents. Assisted
by the solution of coral in sea water, a lagoon is formed in the
centre of the reef, and passages are cut later from it along the
lines by which its muddy water escapes.”

MerHop No. 2.—High land rising from the ocean is first
present, this high land becomes disintegrated and broken down,
and finally, by the action of currents, reduced to a level
plateau some depth below the surface of the sea. On this
plateau Mr. Gardiner pictures an atoll to be very readily
formed. “At depths below 50 fathoms itis obvious that from
the first the organisms on the periphery of the bank so
formed would grow more rapidly, and so an atoll as such
would directly arise. The whole action might proceed with
extreme rapidity.”

MerHop No. 3.—A high oceanic island is pictured with a
fringing reef around its shores, denudation and solution remove
the island land, and the reef stands out from the shore as a
barrier reef. ‘Subsequently the edge of the terrace grows
outwards and its inner part is removed as in (1), forming a
barrier reef. Eventually the original island, owing to similar
causes, disappears, leaving an atoll. This method is of quite
wide occurrence in areas where elevated coral reefs are found.”
The stages of progress in this third method are, therefore,
those pictured by Darwin, but their manner of development is
widely different.

MetuHop No. 4 grants that subsidence alone, or subsi-
dence plus the agencies brought into methods Nos. 1 and 2,
may cause atoll formation.

224 CORAL AND ATOLLS

The development of all coral structures therefore presents
no difficulties if the work of Mr. Gardiner is accepted, for the
choice of four different ways of building these strikingly
uniform formations is presented.

Not only is a choice of four methods given, but the
question of the development of an atoll from a reef is made
simple because in one case “boring organisms” conveniently
“enter on its central part,” and in another because it is
“obvious” that the corals grow faster round the periphery.
The situation is further made more satisfactory by incorporating
Darwin's theory with the ideas which Darwin deprecated, and
which are generally used in opposition to his conclusions.

It will be noticed that in Method No. 3, the theory—which
is an extension of that of Murray—has been used to explain
the presence of barrier reefs, though I do not know if all.
barrier reefs are supposed to arise in this way by supporters |
of these theories.

This particular method (No. 3) of making an atoll appears
to me to need for its acceptance an utter disregard for every-
thing that is at all probable in connection with the story of
the warfare of sea and land. The picture of the high island
towards the completion of the process, when, after having
stood resisting in a troubled sea, it so conveniently crumbles
to pieces within the calm of an encircling barrier reef, appears
to me to be contrary to all natural laws.

What form of denudation and solution could be invoked
to remove a large oceanic island when its barrier reef of
soluble coral remains intact ?

It must be a widespread and potent force, for “this
method is of quite wide occurrence in areas where elevated
coral reefs are found.”

In Method No. 1 the forces consist of “ boring organisms,”
‘sand-feeders,” and “the solution of coral in sea water.” It
cannot be “the sslution of coral in sea water,” for that would
remove the barrier and leave the island; it cannot be wave or
current action, for the barrier has now protected the land ;—all
that remains is the “ boring organisms” and “ sand-feeders.”

ATOLL AND REEF FORMATION — 225

The fact of the cutting down of land below the level of the
sea by currents has been called into question by geologists.*
Admiral Sir W. L. Wharton has given instances in which
the process has been carried out. Graham’s Island in the
Mediterranean is a notable example of this; but it must be
remembered that the destruction of a loosely piled volcanic
structure, cast up to-day and washed away to-morrow, is a
very different thing to the removal of an oceanic island girt
about by a barrier reef.

Professor Alexander Agassiz has also added the testimony
of his great experience in favour of this method of preparation
of the submarine base. In a paper read before the Royal
Society in 1903, he says: “ Denudation and submarine erosion
fully account for the formation of platforms.” Agassiz also
demands a rising, rather than a sinking, of the areas over
which coral formations are found, and in the same paper he
says: “Throughout the Pacific and Indian Oceans, and the
West Indies, the most positive evidence exists of a moderate
recent elevation of coral reefs.”

Dr. H. B. Guppy is another believer in ne theories of
Murray; he brought his views forward in a paper published
in the Proceedings of the Royal Society of Edinburgh in July
1886. The paper was the outcome of his investigations in the
Solomon Islands, and in addition to finding confirmation of
Murray’s views, he also postulated a movement of elevation.
Dr. Guppy doubted the power of coral growth to raise a reef
to a greater height than some five fathoms below the surface,
and, at p. 367, he speaks of “the inability of detached sub-
merged reefs to raise themselves within the constructive power
of the breakers without the assistance of a movement of
elevation.”

When in 1888 Dr. Guppy came to examine the atoll of
Cocos-Keeling his views, at any rate with regard to the
importance of the effects of “solution,” seem to have slightly
changed. In his account of the atoll (op. cit. p. 578) he says:
“T do not, however, consider this [solution] as a very effectual

* See Mr. Ernest Schwarz, Nature, 1904, p. 581.
P

226 | CORAL AND ATOLLS.

repressive agency in the present closing scene of the atoll.
The accumulation of coral débris is too patent a fact. When
we walk through a field of Madrepore and sink knee-deep into
the friable fees of many preceding generations of this coral,
we realise the fact that dead coral accumulates and is but
partially removed in solution. Still there can be no doubt, as
shown by. Murray, Irvine, J. G. Ross and others, that the
solvent action of sea water has a repressive effect on the rate
of reef growth.”

Dr. “Guppy, i in the course of his examination of this atoll,
made an important contribution to the study of coral structures
by formulating a very definite theory to account for the pre-
valent shape of the constituent islands of the atoll ring; this
theory will be mentioned again, for it affords a valuable addition
to our knowledge of coral formations. |

Since the theory of Solution, in some of these modified |
forms, is the one that obtains the greatest general acceptance at
the present day, it is necessary to examine its virtues, and its
shortcomings, in some detail. The theory is put forward to
explain (1) the formation of the original lagoon, and (2) how
that lagoon “ becomes widened and deepened.”

With regard to (1), I think that the theory does not in
reality offer an explanation of the method of atoll formation,
and with regard to (2), the condition that it would explain is
contrary to experience, for atoll lagoons ae as a rule, to
become smaller and shallower.

Then, even if the conditions presumed to be brought Siem!
‘by this agency were conditions that experience showed to be
real, there is no definite proof that this agency could bring
them about; there is, so far as I can see in the published
writings of the advocates of this theory, no direct observation
of the working of “solution” made actually in an atoll lagoon.
Sir John Murray, in his original paper of 1880, at p. 511, says
“larger quantities of lime are carried away in solution as a
ipieadeonaite from the lagoon than are secreted by the animals
which can still live in it.” In a paper published in 1890 * he

* Murray and Irvine, Nature, June 12.

ATOLL AND REEF FORMATION — 227

says, however : “ Turning now to the lagoonsand lagoon channels
of coral islands, 2t 2s believed * that large quantities of carbonate
of lime are in the same way being dissolved from these shallow
basins as well as from the deposits of the deep sea, but under
somewhat different circumstances. In the case of a shell
falling to the bottom of the sea, it is continually brought into
contact with new layers of water, which has the same effect as
if a continuous stream of water were passing over the shell.
In the case of the lagoons this last is what takes place. The
water which flows in and out of the lagoons twice in twenty-
four hours passes over great beds of growing corals, and from
all the observations we have is largely charged with carbonic
acid, owing probably to the larger number of animals on the
outer reef over which the water passes on its way to the lagoon.”
Throughout the writings of Sir John Murray much stress
is laid on “the water which flows in and out of the lagoons
twice in twenty-four hours,” and this for the reason that it
gives the operations of solution a better chance to act. It is
true that the level of the lagoon rises and falls with the cycles
of the tide, and that the alterations in level give rise to local
currents in the inlets of the lagoon; but this does not affect
the whole lagoon, and to imagine the process as anything
akin to that which happens as a shell falls through the depths
of the sea is a mere fallacy.
~ When a shell falls from the surface of the sea to greater
depths it passes through new layers of water, each, with the
increasing depth, being more potent for solution. In the
case of the lagoons this last is what does not take place. The
rising tide raises the lagoon waters, the falling tide lowers
them, but the waters of the lagoon are not changed twice in
twenty-four hours. The action of “ solution” in the lagoons of
atolls may therefore be accepted as being only a theoretical
action, and the actual proofs of its working do not seem to
have been established.
Although there seem to be no published observations to
prove that material is “carried away in solution” from a
** Italics by present author,

228 CORAL AND ATOLLS

lagoon, still Sir John Murray has shown quite conclusively that
matter is bemg removed in suspension. At p. 515 (op. cit.)
he says: “ Very strong currents run out of the entrances into
lagoons and lagoon channels, and when the tow-net was used,
it showed that a large quantity of coral detritus was being
carried seawards.”

This is an easily verified observation, but it needs some
amplification, for it is only true for certain inlets, or at
certain states of the tide. If other inlets had been visited, for
instance those on the windward side, or if the turn of the
tide had been awaited, then currents running into the lagoon
would have been detected, and their burden of detritus would
have been found to be even greater. In the Cocos-Keeling
atoll, Dr. Guppy estimated “that at least 5000 tons of sand
and débris derived from the breaker edge of the reef are
annually transported by the currents through the passages —
between the islands into the lagoon, nearly all of which is
deposited at or near the lagoon’s margin.” The importance of
this water-borne sand will be dealt with more fully, but the
undoubted fact of its coming to the lagoon in greater quantities
than it leaves it, negatives the idea that, even in suspension,
matter is commonly robbed from a lagoon.

Besides this failure of the common history of atoll lagoons
to show any signs of the action of “solution,” or to demand
any such explanation for their origin or development, there
is an inherent improbability in the theory as a whole.
Darwin, on the very first publication of the theory, detected
this weakness, for in a letter written to the late Mr. T. Mellard
Reade on September 22,1880 (quoted in Nature), he says: “It is
astonishing that there should be rapid dissolution of carbonate
of lime at great depths and near the surface, but not at the
intermediate depths where he places his mountain peaks.”
This is of course a very pertinent point, and one that does
not appear to have had any definite answer, although it was
again raised by Dr. R. von Lendenfeld in Nature of May 8,
1890. Dr. Lendenfeld sums up the stages as follows:
“ First a limestone cone is.built, because the lime is deposited

ATOLL AND REEF FORMATION — 229

more rapidly than it can be dissolved. Then a lagoon is
formed because the solution exceeds the accumulations, and
this on the same spot, and in still shallower and less power-
fully dissolving water.”

The late Mr. T. Mellard Reade has also pointed out that if
solution does in reality form the lagoon at the top of the
bank, it ought to dissolve the bank itself much more easily
(Nature, March 22,1888), and Mr. Hedley in the same journal
(August 4, 1904) has also shown this weakness of the theory.
As a matter of fact the work of Murray and Irvine appears to
show very clearly that at the surface of the sea calcareous
deposits take place, whilst at greater depths they tend to
become dissolved ; certainly calcareous deposits take place in
atoll lagoons. It is easily seen that brecciated rock and sand-
stone are formed in the lagoon by the actual deposition of
calcium carbonate. The deposition of calcium carbonate is a
very widespread phenomenon in the lagoon, so that, granting
that sea water dissolves coral, its action cannot be a very
destructive one where rapid deposition of calcium carbonate is
taking place.

There are some features of a fully developed atoll which
have been urged in support of the theory of solution. In
Sir John Murray's first paper, he says (p. 515): “At the
Admiralty Islands, on the lagoon side of the islets on the
barrier reef, the trees were found overhanging the water, and
in some cases the soil washed away from their roots. It is a
common observation in atolls that the islets on the reef are
situated close to the lagoon shore. These facts point out the
removal of matter which is going on in the lagoons and
lagoon channels.”

Now with regard to the first statement, it is fair to say
that it is identical with Darwin’s observations in the Cocos-
Keeling atoll, which gave him such strong confirmation of
his own theory. ‘That any real weight could be attached to it
was denied by the objectors to Darwin’s theory—and quite
rightly, for the condition is only one of those temporary
phases of atoll life which come and go with seasons and winds.

230 CORAL AND ATOLLS

I am quite convinced, as a result of watching the Cocos-
Keeling atoll for fifteen months, that these temporary phases
are misleading, for there is no evidence to show that they
represent the normal process of development. The observa-
tion can therefore carry no more weight when urged to support
the theory of solution than when it was used as evidence of
the reality of subsidence.

With regard to the position of the islets on the reef, it will
be pointed out, in studying the development of the atoll, that
this is entirely the outcome of the force of the waves which
first raised the islets, and the subsequent seaward growth of
the reef edge.

Since the examination of the fully developed atoll lends no
support to this theory, it is natural to inquire whether the
study of its earlier stages will yield any evidence in favour of ©
the theory that its lagoon is caused by solution. One of
the most important additions to the study of coral structures
was made by Admiral Sir W. J. L. Wharton, who pointed out
that the great reefs of the China Sea, and elsewhere, had an
atoll form when far below the surface of the sea. These
ringed reefs are not in reality “drowned atolls,” but are reefs
in the process of development, but it is difficult to see how
solution can possibly produce the condition found in them.

From my prolonged observations of the Cocos-Keeling
atoll, I am convinced that the theory of “solution” does not
at all account for its development, its present features, or its
future history.

Not only does no feature of the atoll yield any support
to the theory, but agreat many of them definitely contradict it.
The lagoon shore is steadily gaining on the lagoon, all round
the atoll ring. The lagoon is steadily shoaling by the deposition
of sand brought into the lagoon by currents. Calcium carbo-
nate is being deposited in the lagoon, making beach sandstone
and conglomerate, and causing the old dead fragments to
become harder, and heavier, than they were during life. The
dead colonies of coral are not dissolved and removed from the
lagoon, for it is definitely known that great numbers of such

ATOLL AND REEF FORMATION — 281

colonies have been standing for thirty years—an observation
of very great importance in this discussion. The present
condition of Keeling Island and of Pulu Luar is prophetic of
the future of the whole atoll: the evidence it affords is in
direct opposition to any hypothesis of a deepening or widening
of the lagoon by a process of “ solution.”

CHAPTER XX

THE THEORIES OF ATOLL AND REEF FORMATION
PUT FORWARD SINCE 1880

In the space of more than twenty years that has elapsed
since the first publication of the theory of “solution,” several
other views upon coral formations have been put forward ;
but although many observers have dissented from the “ solu-
tion” theory, no new concrete idea has been formulated to
account for all the phases of coral formations. In the dis- —
cussions which have taken place upon the subject it has been,
as a rule, a question of accepting the “solution” theory in
the place of the older views of “subsidence”; and yet there
are many observers who appreciate the fact that “ subsidence ”
will not explain all coral structures—and who yet do not
accept “solution” as an explanation for the typical form of
atolls.

Professors A. Agassiz and Semper were early opponents of
the idea of a subsiding basis; but, in 1903, Agassiz said (Nature,
1903, p.547) that in his work “no attempt was made to establish
any independent general theory.” It would appear that the
general tendency was to revert to the ideas of Chamisso—views
that were put forward in “ Kotzebue’s First Voyage,” long before
Darwin formulated his famous theory. Chamisso supposed
that the surf was the home, by election, of the reef-building
corals, and so those colonies upon the outer edge of the reef
grew faster than those within—and in this way a circular
reef was formed. What was lacking in Chamisso’s theory
was a base upon which to build this reef. The failure was
entirely due to the scant knowledge of the ocean bottom at the
time when the theory was put forward: when Darwin over-

came the difficulty he adopted the views of Chamisso to a
232

ATOLL AND REEF FORMATION — 233

great extent. It is not true that corals live best in the
surf: the observation has been repeatedly made that the surf-
line is practically barren of living coral growth: but it 4s
true that the corals of the outer—submarine—edge of the
reef grow most luxuriantly, and Chamisso’s ideas were in the
main correct.

The structure that Chamisso erected upon an unknown
base requires but little elaboration when Semper, Agassiz, and
Murray have demonstrated the foundation to be a reality—
and: for the making of this structure neither subsidence nor
solution is called upon.

The idea that the reef corals delight in the crash of the
surf must be given up, and so the original suggestion of a
better aeration of the outer coral colonies, causing their more
rapid growth, has also been abandoned. Combined with this idea
of a more abundant supply of oxygen to the outer corals is
the supposition that they are also in a position to obtain more
food, and so to thrive better than those which get the water
only after it has passed over the outer rim. Semper attached
great importance to the currents which flow about a reef, and
Professor Hickson, in the Celebes, noted the effects of the
eurrents upon coral growth. Mr. G. C. Bourne, from a study
of Diego Garcia (Proc. Roy. Soc., 1888, p. 458), came to the
conclusion that currents were among the most important
agents in causing the shape of coral reefs, and he says, “ The
strength and direction of currents appears to me to be the
main influence of coral growth ; that the behaviour of currents
on meeting an obstacle with sloping shores explains the super-
abundant growth of corals on the outer slopes of a reef,
whether submerged or awash; that the growth of corals on
the periphery of a bank being in great excess of the growth
in its interior portions is sufficient to explain the formations
known as atolls and barrier reefs.” Yet the workings of this
particular current influence still await an explanation.

These views therefore agree in placing Chamisso’s bank on
the base that Agassiz, Semper and Murray pictured; but the
method of the more rapid growth of the edge is modified.

234 CORAL AND ATOLLS

Agassiz and Semper both showed that frequently the base was
a rising one, but Semper admitted that at times it was not
impossible for atolls to be formed on sinking bases. He says
(“ Animal Life,” English Trans., International Scientific Series,
vol. xxi. 1906, p. 456): “I regard it as quite possible that
under certain circumstances a subsidence may be combined
with the formation of atolls, and even that it may once have
been the sole cause of their formation ; but I cannot admit that
subsidence is alone sufficient to explain all the conditions and
relations of coral reefs, or even of predominant importance.”

The theory of subsidence was accepted, and strongly ad-
vocated, by Dana, and he even went a step farther than its
originator, for he was unable to conceive that, even if banks
did exist at the proper depth, an atoll structure could be
developed upon them without the intervention of a movement
of subsidence. ‘This possibility Darwin had always allowed.
He said: “If such a bank lay a few fathoms submerged, the
simple growth of the coral without the aid of subsidence would
produce a structure scarcely to be distinguished from a true
atoll; for in all cases the corals on the outer margin of the
reef, from having space and being freely exposed to the open
sea, will grow vigorously and tend to form a continuous ring,
whilst the growth of the less massive kinds on the central
expanse will be checked by the sediment formed there, and
by that washed inward by the breakers; and as the space
becomes shallower, their growth will also be checked by the
impurities of the water, and probably by the small amount of
food brought by the enfeebled currents, in proportion to the
surface of living reefs studded with innumerable craving
mouths.” This I consider to be one of the truest pieiunes
ever drawn of the development of an atoll reef, but Darwin
did not imagine it to be the mode of formation of the majority
of atolls ; and Dana observed : “ If such patches of submerged
bank existed, the lagoon structure would still be as inexplicable
as ever” (op. cit. p. 220).

The newest epoch in the investigation of coral structures
is opened by the detailed study of the atoll of Funafuti carried

ATOLL AND REEF FORMATION — 235

on by no less than three successive expeditions. The object
of the examination was the making of a bore through the
island and its basis in order to determine if the necessary
depth of coral reef, which a theory of subsidence demands, was
found to exist in reality. In the volume in which the results
of this successful boring are published (“ Atoll of Funafuti,”
Roy. Soc., 1904) the core obtained, to a depth of 11144 feet,
is described in its entire length, and the conclusion appears
to be that the whole extent of the core had been formed at a
distance of no more than 50 fathoms from the surface, and
consists of a continuous formation of reef rock. Reef corals
were found as casts in the lower limits of the core, and if these
corals are in the position of their growth, then a gradual sub-
sidence must have taken place during the formation of the
atoll. On the other hand, to my mind, the description of the
core reads far more like a section of a talus bank, with surface
forms of corals and foraminiferz, piled in disorder as a débris
slope. Since the bore was driven on the extreme—windward
—edge of a reef some ten miles across, there is no reason why
this should not in fact be the talus slope of an originally
smaller reef, the primitive boundaries of which are to be found
“somewhere nearer the middle of the present lagoon. If the
original submarine bank had grown seaward pal half a mile
in all its long history, its talus bank would have been pierced
by the bore, and its condition would in all probability have
been just what is seen in the actual bore.* This is made easy
of understanding by means of a diagram drawn to scale. (See
Fig. 53, p. 153.)

This objection is not urged in the spirit of destructive
criticism, for it seems to me that some atolls may have been
formed on subsiding bases: but to my mind it renders it quite
impossible to say that any process of subsidence caused the
development of this atoll of Funafuti. More recently Mr.
Hedley has shown—and here the evidence appears to be con-
clusive—that the Great Barrier Reef of Australia is the site

* On this point see also Hedley, Natural Science, xii., March 1898,
p. 174.

236 CORAL AND ATOLLS

of definite subsidence. It does not necessarily follow that the
features of this coast-line are brought about by the growth
of coral reefs on a sinking basis, for, as Hedley says (“ Coral
Reefs of the Great Barrier,” p. 10): “It may be allowed—
though Darwin deprecated the idea—that the continental shelf
was ready prepared with numerous banks representing eroded
islands, just reaching to within the required distance of the
surface, when the first coral builders came.” ‘The coral reefs
are definitely established upon a sunken basis, but that is not
the same thing as saying that the sinking caused their present
disposition.

In this same paper, Mr. Hedley carried on the work of
Guppy and of Kramer in the appreciation of the force of the
winds and waves in the formation of crescent and horse-shoe
islands and reefs, and, on the whole, this wind formation is one
of the most distinct advances made in the study of coral
structures. |

I think it may therefore be said that there is no one
generally accepted method by which it is agreed that all coral
formations are brought about. Some reef spots show signs of
elevation, and some of depression, and some fail to give any
indication of change, so that land movements may not properly
be claimed to be the cause of any features that are constant.
Upon a stationary basis again, there is no generally accepted
method for building the characteristic features of the fully
devoloped atoll or barrier reef.

CHAPTER XXI

THE METHOD OF DEVELOPMENT OF THE REEF
BANK AND THE REEF

SEDIMENTATION IN THE OPEN OcEAN.—It is a statement which
will not be disputed that sediment is for ever dropping from
the surface of the ocean. When the real nature of the mate-
rial which composed the floor of the ocean was made known,
and the hard remains of a number of animal forms were
recognised, it was at once asked: do these forms live at the
depths at which their remains are found, or are they surtace-
living creatures, whose heavier parts have dropped from the
surface waters after the death of the animal? The tow-net
revealed the animals during life as surface-living forms; and
many investigations—especially the Challenger reports—have
made all the processes of deep-sea deposit quite clear.

Not only are the troughs of all the oceans lined with ooze
of slightly varying nature, but the elevations which arise from
them are also clothed with the same deposit. There is then a
perpetual shower of sediment from the surface of the ocean.
Calcareous and siliceous remains of pelagic organisms, volcanic
dust, and other less considerable constituents form a never-
ceasing rain of particles from the surface-water to the ocean
depths. They drop, and line the ocean bed: they drop, and
cover ocean slopes and elevations, and come to rest wherever
the absence of wave action will let them settle.

This certain knowledge prompts a very natural inquiry as
to the extent to which this sedimentation building may take
place. Supposing some elevation to be present on the ocean
floor, or supposing more sediment to be deposited in any one
particular spot, by reason of some ocean current; how far

below the waves would the rising mound come to a standstill ?
237

238 CORAL AND ATOLLS

In other words, where, below the surface of the sea, will
dropping particles fail to come to rest, and cease to form banks ?
We do not know, and beyond saying that somewhere in the
ocean there is a level, below which particles may rest and
sedimentation take place, and above which particles are pre-
vented from settling by the movements of the surface waters,
—we cannot definitely go. This level is one of such great
importance that I propose to look upon it as a definite plane,
occurring over every ocean, and to give to it the name of the
LimitInG LINE OF SEDIMENTATION,

We may attempt to find out where this line may be ex-
pected to exist by ascertaining how far beneath the surface of
the ocean the effects of wave action are felt; but on this
point our knowledge seems to be rather scanty. Dana, in
his “ Corals and Coral Islands,” 1875, has collected a certain
amount of information concerning this question, and he quotes
M. Siau (Comptes Rendus, t. xii. p. 744) for the assertion that
ripple marks are present at a depth of 94 fathoms to the
north-west of St. Paul’s Roads. Various other estimates
have been made, and 83 fathoms, 50 fathoms, and 15 fathoms
are some of the figures arrived at. De la Beche is respon-
sible for the statement that sand and mud are stirred up, so
that they discolour the water, at the depth of 15 fathoms ;—
it is evident therefore that this depth is above the Limiting
Inne of Sedimentation.

It would not be expected by any one with a knowledge of
the sea that the level of the Limiting Line of Sedimentation
would be the same for all waters; for the normal rollers of —
the Trade-swept ocean must affect a far greater depth than
the short seas of land-locked basins. It would not be ex-
pected even that it would be at the same level at all times
in the same sea, for in monsoon areas it would be probably
different in two monsoon seasons. Still, for all parts of all
oceans there is an average line separating the wave-stirred
upper layers from the still, quiet waters of the ocean depths.

SEDIMENTATION BanxKs.—lIt is, therefore, an easy matter
to picture the development of a sedimentation bank in the

DEVELOPMENT OF THE REEF 239

ocean. It may, perhaps, rise from ocean depths, in places
where the oceanic currents deposit more abundant sediment,
or it may build on some already prepared elevation; we can
at any rate assert that, wherever its formation starts, it will
certainly end at the Limiting Line of Sedimentation. Great
numbers of these banks are known, and their very number does
not warrant their mention. They may be built, perhaps, on
volcanic elevations (Murray), on denuded volcanic islands
(Wharton), or on denuded remains of ancient land (Stanley
Gardiner); or they may be built on inequalities of the ocean
floor not included under these heads, or be raised from ocean
bottom by sedimentation alone. I do not know that we have
much evidence to show what they may be; and I do not think
that we have any reason to believe that they are all volcanic.
For the explanation of the development of coral structures
I do not think it is to be greatly deplored that we cannot
assign a common origin to them all, for in their development
they are all essentially sedimentation products, and so have to
adhere to the laws governing its deposition.
Tae BaTHyMETRIcAL Limits OF THE REEF - BUILDING
Corats.—It is next natural to inquire at what point, below
the level of the waves, may the growth of the reef-building
corals start ; and on this matter again our evidence cannot be
said to be complete. Dana collected all the evidence at his
disposal, and placed the limit at 20 fathoms; Darwin, from
his data, set it at from 20 to 30 fathoms; and Stanley
Gardiner, from a series of dredgings in the Maldives and
Laccadives, has confirmed Darwin’s estimate. By the careful
work of Basset-Smith upon the Tizard and Macclesfield banks
(Ann. Mag. Nat. Hist., 1890, p. 353), the knowledge of the
distribution of the reef-builders has been greatly increased,
and the bathymetrical range has been found to be far more
extensive than was supposed. Im a series of dredgings, in
depths between 31 and 45 fathoms, Basset-Smith found 19
species of living corals, and 12 of these species are typically
reef species. The most remarkable result is the certain know-
ledge that a species of Favia was flourishing at a depth of

240 CORAL AND ATOLLS

45 fathoms. The actual details of the species, as given in the
paper, are as follows: “ Stylopora, 1 species; Favia, 1 species
at 45 fathoms; Pavonia, 1 species; Leptoseris, 1 species ;
Phyllastrea, 1 species; Psammeora, 1 species; Montipora, 3
species, one in 44 fathoms; Rhodarwa, 1 species; Alveopora,
2 species.”

It is therefore safe to say that, in this part of the China
Sea at any rate, the species that we meet with at the surface
may live quite well at something over 40 fathoms. Nor
would I put this depth as the absolute limit of their range,
for the results of dredging are always somewhat, uncertain, and
the experience of the difficulty of obtaining specimens from
beds that can be actually seen must make any one cautious of
assuming that negative results of dredgings mean necessarily
absence of coral growth. I have seen beautiful branches of.
Stylopora come up on a fishing line from 12 fathoms, after I
had failed to bring up specimens with a far more elaborate
contrivance.

THE REASONS FOR THE RESTRICTED BATHYMETRICAL RANGE
oF REEF-BUILDING CorALs.—Dana suggested the lowering of
the temperature of the water at greater depths, but the
temperature of depths far below the range of the reef-builders
is not sufficiently low to exclude coral growth; and Dana
admitted that temperature could not account for the whole of
the phenomenon.

It has been said that it is because the conditions for
forming calcareous skeletons are only present at comparatively
shallow limits; and it is pointed out that deep-sea corals are
lightly calcified. There is, no doubt, some truth in this, but
30 or 40 fathoms would seem a narrow limit to set as the
maximum depth at which satisfactory calcareous skeletons
could be formed by the reef-building corals, when solitary
species, as Caryophyllia clavus, which I have taken in quanti-
ties from a cable at 380 fathoms, and which has been obtained
at much greater depths, possess a highly calcified corallum.
Again, it is possible that the Lithothammonide and Nullipore
flourish, and build their skeletons, in great masses in depths far

DEVELOPMENT OF THE REEF 24]

below the bathymetrical limits of the reef-building corals
(Stanley Gardiner, op. cit. p. 175).

Mr. Stanley Gardiner has said that, in determining the
cause of the 30-fathom limit, “the real factor is the light,
which to reach the commensal algze has to penetrate the tissues
of the polyps as well as the water” (op. cit. p. 177). Now,
without wishing to controvert the assertion that corals “feed
mainly by their commensal alge,” I would point out the fact
that light is certainly not essential for their feeding. It has
been my invariable experience with the corals in this atoll
that they are actually absolutely nocturnal in their activities.
It must be noticed by every one who visits a reef that, during
the hours of daylight, the zooids of the corals are not ex-
panded; they are in a passive phase, from which a further
stage of retraction may result on stimulus—but they are
practically never expanded. It is most instructive when
examining a reef to see great soft masses of Alcyonaria with
myriads of zooids expanded and waving in the current, and
alongside it a rock colony of Porites in which not a single
member shows activity.

It is also one of the most striking things to examine the
reef corals by night, and see the extent of alteration their
appearance undergoes, when the full expansion of their zooids
occurs. Mr.J. EH. Duerden observes: * “As a general rule the
polyps are not expanded to their full degree during the day,
either on the reef or in the laboratory; but the process begins
immediately after sunset, and full expansion is maintained for
the greater part of the night. In the morning the polyps are
again found retracted.’ These observations were made in
Jamaica,and so the habit is not one that is due to locality, and
I think we may therefore regard it as the fact that the reef-
building corals are active and that they feed in the dark. I
agree with Mr. Duerden in the observation that artificial light
appears to have no effect on zooids already expanded in the
dark. Again, the corals are certainly not entirely dependent
on their commensal alge for their food-supply, for any one may

* Memoirs National Acad. Sci., vol. vili., 7th Memoir, p. 419 (1902).
Q

242 CORAL AND ATOLLS

see the reaching out of tentacles, and the engulfing of particles,
when a crushed-up shell-fish is dropped upon a colony by night.
Solitary Lungiw, which I have kept alive, will take into their
mouths portions of the flesh of oysters considerably larger than
a pea.

On the strength of this positive evidence, I therefore
discard light as being the determining cause of the bathy-
metrical limit of distribution of the reef-builders.

It now remains to inquire if there is any other factor
which can with more justice be claimed as a cause of the
absence of the reef-building corals below 30 or 40 fathoms.
Karly in my examination of the atoll, I concluded that it was
undoubtedly the presence of sedimentation. Everywhere, in all
parts of the reef, the most patent fact in the economy of the
corals was that sedimentation was the most potent cause of |
coral death, and the most important influence upon all phases
of their existence. I therefore postulated a region in the
ocean where reef-builders could not live, because sediment
was always falling upon them and remaining undisturbed upon
them; and another in which they could flourish, because
particles were not allowed to remain and choke them, owing
to the stirring action exerted by the waves. I arrived at this
conclusion in ignorance of the fact that two definite pro-
nouncements, of an almost similar nature, had already been
made. Agassiz had said: “There seems to be no simpler ex-
planation of the limited bathymetrical range than that of the
baneful action of silt held in suspense near all reefs”; and
Sir A. Geikie had said that the depth to which currents cut
down land “is probably nearly coincident with the lower limits
of the reef-builders.”

So influential a factor is sediment in the life-history of
all reef-building corals that it is one of the greatest causes
of the many modifications of vegetative growth embraced
within the limits of a “species.” Attention has already been
called to this in considering the coral colony, and it is un-
necessary to recapitulate the evidences of the various changes
that sediment brings about in the appearance of a colony. It

DEVELOPMENT OF THE REEF 243

is certainly the greatest enemy to activity, and the most
potent cause of death in corals.

It may therefore be said to be an axiom in coral bio-
nomics that an area in which sedimentation is taking place is one
in which the reef-builders will not flowrish. It is then possible
to say tbat reef-building corals will not live below the
Limiting Line of Sedimentation, and this without special regard
to the nearness of reefs, or the presence of land-cutting
currents, but as a general statement. I have said that we do
not know precisely where this line may be, and that it will
vary in different places from purely local causes, but wherever ~
it may be reef-coral growth will be excluded from the depths
below it, and the surface water above it will—other condi-
tions being favourable—offer a suitable site for their develop-
ment. We may therefore lay it down as a law that the
bathymetrical limit of the reef-building corals extends to the limiting
line of sedimentation, and that this exact level will be the
outcome of purely local conditions.

It may be objected that, if the reef-building corals are pre-
sumed to be unable to live below the limiting line of sedimen-
tation, how is it that many species of solitary corals are found
in far deeper water; for if sediment is the real cause of death,
how do some species resist its action ?

There seems to be but little doubt that it is the common
type of growth of these deep-water corals which explains their
ability to withstand the action of silt. Most of them are flat
discs, or more or less wide-mouthed cups: they are not, as a
rule, honeycombed with a series of small openings, as are the
reef-builders. It would seem at first sight that this type of
growth was perhaps the most ill adapted to resist the effects of
sedimentation, for a larger proportion of the zooid surface is
exposed to the dropping particles. The large living specimens
of the Fungiw, which lie scattered about in sheltered pools on
the barrier flats, afford the best examples of the extreme degree
of flattening of the corallite, and exposure of the zooid: the
study of the habits of these Fungiw offers the explanation of
their ability to live in a silting environment.

244 CORAL AND ATOLLS

If such a specimen be watched, as sand particles are
poured over its surface, it will at once be seen that there
is a very perfect mechanism for ejecting the silt. The
delicate tissues, which clothe the rays of the coral, shrink
back from the intruding substance; and a curious wave
spreads along the partitions, from the central end towards the
periphery, and the foreign body is borne along in the wave.
The same process is employed in the ingestion of food ; but in
that case the particle is passed on from the periphery to the
centre. If an oyster is crushed, and its contents scattered
over the surface of a large Fungia, the whole of the soft
tissues of the coral shrink together at first, as a result of the
stimulus, and then, slowly, successive contractions and ex-
pansions sweep the individual particles towards the central
portion of the disc, where they disappear from view. Later on,
the small bright fragments of oyster shell, which were taken
in along with the food, are seen to be migrating towards the
periphery, where they are thrown off. The Fuwngia therefore
protects itself by getting rid of the sediment which comes in
contact with it; and the mechanism must need to be a very
effective one, for /ungiw will live in pools in which vast
numbers of particles are moving about, and the skeleton of a
dead Fungia always collects large quantities of sand.

It is probable that all those deep-sea corals which have a
flat or wide-mouthed type of growth get rid of the foreign
substances that fall within their mouths, after the same
fashion as the Hungiw; and so they are indifferent to the
presence of suspended matter in the water, provided that the
deposition of sediment does not proceed at such a rate that
it overwhelms them entirely. With the small mouths of
the reef-builders this process may be quite impossible, and
sediment that settles on a colony cannot be got rid of in this
way ; the individual corallites become filled with sediment, the
zooid becomes choked, and death ensues.

The definite knowledge that sedimentation areas are
certainly excluded from reef-coral growth, whilst wave-stirred
areas are particularly favourable, marks an important stage in

DEVELOPMENT OF THE REEF 245

the study of the problem, for it now becomes evident that
oceanic banks may rise to the Limiting Line of Sedimentation ;
and from that line the reef-building corals may start their
building operations.

Tue Formation oF REEFS.—Theoretically, where the one
process leaves off the other begins; but, practically, it is most
probable that a host of other creatures help to bridge over the
gap between the cessation of the building by sedimentation and
the building of the corals of the reef. It is probable that
the Nullipore, Lithothamnionee, and other calcareous alge
outstrip the reef corals in their bathymetrical range, and they
probably begin to clothe the summit of the sedimentation
bank. As they grow, fragments of their growth die, or become
broken off and are added to the foundation of the reef.
Animals with hard tissues live on the reef, and their remains
become added to the débris; the whole surface becomes
raised, and at the same time consolidated. Particles, which
could find no resting-place upon the bare exposed bank,
doubtless become held fast in the uneven surface of the growth
that has now clothed itssummit ; and sedimentary action may
again add its increment, although the whole .structure is now
within the wave-stirred area. It is probably in this fashion
that the foundation of the reef is laid and the immediate
clothing of the sedimentation bank is brought about—the
lower forms making a platform upon which the corals them-
selves actually build.

The pioneers of coral growth are probably not those species
which are represented at the wave-washed surface of the fully
developed reef. Many deep-sea corals have been found to
clothe submarine banks, and the work of Basset-Smith and
J. Y. Buchanan has established the ability of these corals to
raise banks by their own growth from very considerable depths.
It is these corals (Dendrophyllia, Heliopora, Lophohelia, &c.)
which first contribute, by their building, to make a coral reef
on a sedimentation bank.

Sires OF REEF CorAL BUILDING OTHER THAN SEDIMEN-
TATION Banks.—Since the reef-building corals may start their

246 CORAL AND ATOLLS

growth upon any bank that may happen to lie at or above
the Limiting Line of Sedimentation it follows that any eleva-
tion of the ocean floor, which rises to this plane, will furnish the
corals with a suitable basis. I do not doubt that many other
formations than the slowly formed sedimentation banks become
the basis of coral structures; for land that has sunk beneath
the waves will also furnish the necessary building ground.
Land that has risen by earth movements to a station within
the wave-stirred zone, or loose volcanic structures that have
been denuded by the waves, may also become the sites of reef-
coral growth; and doubtless in many cases they have done so.
In these cases the top of the platform may rise near enough
to the surface to be directly clothed with the reef corals, or the
intermediate building of calcareous alow, and deep-water corals,
may take place; or it may not reach so high a plane, and the
processes of sedimentation may be called in to fill up the
necessary interval.

It matters not what the base may be, so long as its plat-
form comes within the wave-stirred area. It may happen,
therefore, that the corals begin their growth on platforms
already having an upward or a downward earth movement,
and these movements will doubtless continue after the develop-
ment of the reef, with varying effects upon its structure.

It is contended that, for the development of a coral reef,
some basis that reaches the wave-stirred limit is essential; and
on such a basis a reef will form, when the other conditions,
such as the temperature of the sea, are favourable. In
some places, I do not doubt, a downward earth movement has
brought a base to the necessary depth for reef development :
the Great Barrier Reef of Australia is almost certainly an
example of this. In other places, an upward earth movement
has probably brought up some elevation sufficiently near to the
surface, and coral reefs have become developed on the basis
thus brought within their domain. The loose structure of sud-
denly erupted oceanic volcanic islands may doubtless be cut
down by waves and current, to some height at, or above,
the Limiting Line of Sedimentation; but the cutting down of

DEVELOPMENT OF THE REEF 247

continental land to a depth of 150 fathoms by currents in
mid-ocean (Gardiner) requires.a greater measure of proof than
has been adduced. These may be the sites of coral reefs,
but in mid-ocean it would seem probable that the banks upon
which reefs and atolls are formed are essentially sedimentation
structures ; and those which do not actually start as purely
sedimentation banks ultimately develop the characters of such
foundations.

THe AFTER-DEVELOPMENT OF THE REEF.—We have there-
fore arrived at the stage in which a bank has become clothed
with a growth of the true reef-building corals. The whole
patch has risen towards the surface, and consists, in all
probability, of a scattered series of colonies, lying in irregular
confusion about the top of the bank. This is of course a well-
known stage, and great numbers of such reefs are present in
seas that are favourable to coral growth.

It remains now to follow the progress of this coral reef
between the Limiting Line of Sedimentation and the surface
of the sea; and to see what factors will bring about the
changes that are known to be developed in this process. It
is here maintained that in this stage again the action of
sedimentation plays the leading part in the history of the reef.
In order to study the action of sedimentation upon reefs,
some idea of its workings may be gained by following the
process as it affects colonies. This question will first be
dealt with.

There is a very common condition of certain coral
colonies which has received special attention in the works
of both Darwin and Semper, namely, the state of the
flat-topped rocks formed by the old colonies of Porites.
As actually seen upon the reef, the colonies consist of
great boulders of rock—for it is the large colonies that best
exhibit this form—the tops of which are flat, or even some-
what hollowed ; the flat surface is destitute of living zooids,
but it is ringed about by a raised, rounded edge of healthy
members of the colony. The dead flat surface of the
rock is generally occupied by a layer of sand, and, in the

248 CORAL AND ATOLLS

lagoon, alge and other forms become established upon it.
Darwin assumed that in these rocks upon the barrier he saw
the mark of the limit of coral building to tide-level, for the
dead flat top of the rock he regarded as being the result
of the exposure of the zooids above low-tide level. Semper
also took tide-level to be the cause of this formation, but he
added the accumulation of sand, and the occasional presence
of fresh water from showers of rain, to the effects of exposure
to sun and air. IJ have given an account of the development
of these flat boulders, as I have observed it in the atoll, and,
in this account, the exposure to sun and air has been excluded
as a causative factor in the great majority of cases.

A colony of Porites is composed of a number of individuals,
every one of which has like powers of division and reproduc-
tion ; it therefore follows that when an embryo settles upon a
small and isolated nucleus, the colony which it gives rise to will
tend to assume the form of a sphere. Such spherical colonies
are, as a matter of fact, often to be found lying free in rock
pools; but they can never grow to any size as spheres, and no
large colony ever has anything like a truly spherical form.
The dome-shaped rock is the type of all larger colonies: it
is developed gradually from the spherical type by the death of
all the zooids in that part of the colony that is subjected to
pressure; ze. the side of the sphere upon which the colony
was lying. A dome-shaped colony may grow to a great size,
and some of the largest of all the colonies of the type P. lutea,
which live upon the submarine slopes of the outside of the
atoll, are very perfect domes. As a rule, wherever sedimen-
tation is taking place with rapidity, the dome shape is lost,
and the flat-topped rock is produced. . Colonies in the lagoon,
deep and undisturbed by tide-level, assume this form ; colonies
in barrier pools that lie many feet below the lowest fall of
spring low tides become flat-topped, and it is evident that it
is not the low-water exposure which is responsible for the
assumption of this form. |

A dome-shaped colony grows larger and larger, its outline
becomes the segment of an ever-increasing circle, and the top

DEVELOPMENT OF THE REEF 249

of the colony, therefore, becomes increasingly flattened. When
this process has gone on for some time the flattening of the
rock enables sediment to find a resting-place upon its upper

Fig. 60.

PHOTOGRAPH OF A CoRAL CoLONY THAT HAS Livine Zoorps
OVER ITS ENTIRE SURFACE.

No zooids are as yet killed either by pressure or by sedimentation,

surface; and the activity of the uppermost zooids declines
as this aspect of the dome becomes increasingly favourable to
the settlement of silt. The flattening becomes increased by
the margins rising, till finally the central portions are entirely
killed by sediment, and a flat sand-strewn area is produced,
surrounded by the raised rolled edges of actively dividing
zooids which are untouched by sediment. Upon the barrier
there is no doubt that injury of the upper zooids, caused by
moving particles sweeping backwards and forwards in the

250 CORAL AND ATOLLS

waves, greatly hastens the progress of this change; for com-
paratively small rocks may here become flat-topped ; but even
when injury starts the process, the destructive action of
sedimentation completes it. Some importance is attached to
this method of the formation of flat-topped rocks, because
it is believed that the process helps in the understanding
of some other points in coral bionomics.

In the course of a walk over the barrier flats, it may
easily be seen that flat-topped colonies do not assume an ab-
solutely uniform plane, as they should do if exposure caused
the phenomenon, and that dome-shaped colonies may very
often be found to be raised higher than flat colonies. Cases
that are particularly instructive are met with in the sharp-
edged pools common in some parts of the barrier flats. In
these the breccia layer will be found to cease suddenly, and
an overhanging ledge forms the margin of the pool, into which
the particles that are swept across the flats are always pouring.
They may be only three feet deep at low tide, and fifty yards
or more in length. Numerous colonies of corals will be found
in them, and all the colonies of Porites that are of any size
will be flat-topped, whilst bordering the edge of the breccia
rock, three feet above them, there is a flourishing growth of
Pavonia of the type P. repens, Brug. The Pavonia lives upon
the under side of the overhanging edge, and is free of the
sediment which sweeps over the flats to be deposited over the
lower area occupied by the Porites.

Not only does the deposition of the sediment produce the
flattening of the tops of these Porites colonies, but it also
affects, in various ways, the growth-forms of other species
living in the pools; but these other modifications do not con-
cern the present inquiry.

Now a dome-shaped rock of Porites is a colony of a myriad
individuals, and it is not an illegitimate stretch of reasoning
to compare it to a reef that is composed of a myriad colonies.

The individual colony, under the action of sedimentation,
has come to consist of a flat, sand-strewn area ringed around
by a raised vigorous margin of growing coral zooids; it is

WAOLVINIW NI TIOLVY NV SINUSHYdaY
ANOTOO SIHL ‘SGIOOZ HHL NOan INAWIGAS WO ONTddONd WHEL AA GHSOVO WOVANOS UAddA Sut
LO VAUV GVAd AHL MOHS OW “TOOd daiad V HOUT NANVIL ANOIOD ITVNOD TIVAS V AO HAdVUDOLOHA

‘TIX ULV Td

DEVELOPMENT OF THE REEF 251

suggested that this effect will also be produced on the reef
itself, and there are very many facts that go to support this
suggestion. The rising reef becomes the site of the pro-
duction of much débris, dead colonies are reduced to coral
sand by the many sand-producing agencies of the reef; and
this sand becomes held in suspension in the water moving
over the surface of the reef. The reef, therefore, in a great
measure provides its own suspended matter, but added to
this is the shower of particles held in suspension in the
surface waters. This accumulation of particles will, doubtless,
find lodgment on the uneven surface ,of the coral bank; and,
in its to and fro journeyings across the bank, it will tend to
exert the maximum of its influence upon the flat central area
of the reef. In this way it is easily seen that the central
colonies of the reef, just as the central zooids of the colony,
will tend to decline in vigour, and their growth will not be
so rapid as that of the colonies situated nearer to the edge.
The growth of the reef as a whole will, therefore, tend to be
more rapid at its margins, and less rapid at the centre; thus a
somewhat concave upper surface to the bank will be developed.
The fate of the colony is again reproduced ; sediment becomes
accumulated in this central depression, and coral growth be-
comes still more reduced; and a stage is arrived at, in every
way comparable to that followed in the Porites colony, of a
flat-topped plain, ringed around with a healthy growing edge
of coral. So far, it may be objected, this development of the
reef is pure hypothesis; but it is a hypothesis that is capable
of receiving very real support from the actual condition of
those submarine coral banks which have been at all adequately
surveyed. The stage at which we have arrived is exactly that
seen in reality in such reefs as the Tizard and Macclesfield
Banks in the China Sea.

Admiral Sir W. L. Wharton has described these banks in
Nature, Feb. 23, 1888, and Dr. Bassett-Smith in the paper
already quoted. Tizard Bank is 32 miles long and 10 miles
wide : it is flat and lies at a depth varying from 30 to 47
fathoms, and its raised edge is only some 10 fathoms below

252 CORAL AND ATOLLS

the water. Hight reefs on this edge have already reached the
surface, and three bear islands upon them.

Macclesfield Bank lies farther to the north; its depth is
from 40 to 60 fathoms, and its edge.is from 10 to 19 fathoms
below the waves. Prince Consort shoal, the Great Seychelles
Bank, and a great many others present a like condition, and are
in fact atolls below the waves, representing, on a grand scale,
the condition of the flat-topped Porites colonies. Other reefs
are representatives of the other stages which we have noted in
the development of these colonies ; some are dome-shaped, and
others are flat, the activity of the central colonies not having
diminished, and the rim being as yet undeveloped.

It would seem to be possible that some dome-shaped reefs
might rise near to the surface of the sea as domes, just as we
have seen the Porites colonies do on the barrier; their history
then becomes similar to that of these colonies, and when they
reach the waves their central portions will be less vigorous
than their outer margins. .

We have now arrived at the stage at which we may picture
a coral reef risen near to the surface of the sea; its central
platform is depressed; its colonies are less vigorous in their
growth, and sediment is being deposited upon its floor. The
changes have so far been brought about by the effects which
sedimentation produced naturally upon coral growth. No
other agency is involved in this explanation than the easily
verified reaction of the living coral colonies to the harmful
effects of particles deposited upon them.

The development of the great edged reefs goes on unseen
below the surface of the sea, but the process may be realised
in watching the evolution of the flat-topped Porites colonies,
and the changes in the many smaller reefs to be found in the
lagoon. ‘There are a hundred diminutive submerged atolls to
be seen any day in the coral-covered sedimentation banks of
the eastern side of the Cocos lagoon; for every little mound
that rises near to the surface shows the condition that I have
imagined and described.

CHAPTER XXII
DEVELOPMENT OF THE ATOLL

THE FORMATION OF THE BARRIER.—The rim of the reef is now
lying a little way below the surface of the sea, and it is obvious
that, since it is raised by the agency of living coral colonies
only, there it must stay, so long as no other building factor
comes into play.

The corals themselves can never grow beyond the tide
limit, and form dry land of their own efforts; and a submerged
reef would stay submerged for ever if its rising depended
upon coral growth alone. With the advent of the reef near to
the surface of the sea, however, a very different set of building
agencies comes into play, and a very great change is introduced
in the history of the coral-clad plateau.

We have seen that the whole structure of the atoll consists
in reality of a level, and slightly raised, rim around the
margins of an extensive reef, and that on this level rim islands
of coral débris are placed at intervals. It is therefore necessary,
first of all, to follow the method of formation of this level rim,
—the barrier flats. This process of formation, again, is one
that can always be seen in the making, and the composition
of the barrier rock itself easily gives the explanation of its
origin.

When the reef-edge has risen so near to the surface of the
sea that the waves may act upon it powerfully, some colonies,
or some portions of colonies, will be uprooted from the extreme
edge, and shifted on to the tops of their fellows that lie nearer
to the centre. These accumulations of broken corals will tend
to increase, and the reef edge will become the site of many
irregular elevations, so that in time waves will break upon
some part of the edge of the bank. This is the beginning of the

253

254 CORAL AND ATOLLS

barrier. Where large fragments lie piled in uneven masses,
smaller fragments will collect, and fill in the interspaces; the
waves will wedge them home, and tend to make the whole
thing solid. Upon the lee of these elevations, grains of coral
sand and lighter particles will collect, and find their way into
the chinks left in the rock piles. A very important action
now begins to take place, for around each small coral particle,
as a nucleus, a crystalline deposition of calcium carbonate
takes place; particle is held to particle, and the fine sand that
has been driven by the force of the waves into every tiny
interspace becomes a cement that holds the coarser fragments
firmly together. The result of this action is that a con-
glomerate is made of irregular fragments, broken branches,
and coral sand; the whole mass cemented by a deposition
of calcium carbonate around the particles. The deposition of |
calcium carbonate is not confined to an encrusting layer around —
the particles; it takes place in the dead coral itself, and the
individual fragments become harder and heavier than they
were during life.

A breccia mass, as hard as concrete, is thus formed
about the fragments which some chance storm first cast upon
the rim of the reef. It is obvious that, since the process is
the outcome of wave action, it will start first of all upon the
windward side of the reef; and at the point of maximum wave
action we may expect to see the first piling up of the boulders
that will afterwards form the barrier. Ultimately, other
portions of the ring will be affected-—probably in the first
instance during storm weather—and the piled-up rim will
spread, as a horse-shoe, around the rim of the reef. The
sequence is, in all probability, somewhat as follows. A time
of storm brings waves that are abnormally powerful across
the reef spot; upon the windward side their feet will be
caught, and during the height of the storm some boulders of
massive corals will be lifted and cast upon the top of their
neighbours, where they will come to rest. Under more normal
conditions of wind and wave, accumulations will take place
about these dislodged boulders, and they, by their own growth,

SSVI AIOHM AHL AO ONIINANHO AH GNV SHIOMNVd TIVWS AO NOISOTIONI
GHEE MOHS OF ‘VIOONUE LV TA-YAINAVA AO WOUId V AO GOVAITNS CAMALOVIA V HO Hd VUONOLOHA

THX ALVId

DEVELOPMENT OF THE ATOLL — 255

or the growth of neighbouring colonies, or by the welding of
smaller fragments, will be firmly reseated on their new site,
A point on the reef is thus formed upon which the waves will
break in ordinary weather. In time the process spreads, and
instead of there being only one point upon which waves break,
a line of breakers will be formed; and as the line continues to
grow it will become a crescent, tailing off around the edge of the
reef. Smoother water will be present within the protecting
crescent, and there fragments that have been hurled inwards
from the edge will become lodged. These fragments, again,
become welded by the deposition of the calcium carbonate
around the particles driven among them, and the breccia
masses tend to become a continuous rampart. Waves will
break upon the boulder-piled edge and sweep across this
rampart, and, as they do so, will become spent :—there will
therefore be a tailing off of the breccia layer within, and we —
may expect to find sand and sediment deposited in such a
way as to build up a shore internal to the platform. With
the perpetual sweep of waves across the platform, and the
constant addition of new fragments and the deposition of
calcium carbonate, the whole becomes levelled, or approxi-
mately levelled ; its completed stage will be that of a uniform
concrete platform, ending sheer at its outer edge, and tailing
off as a débris slope within. The level it assumes will be
that of the point of the maximum intensity of wave action
at all tides. When the tide is low the waves will crash upon
the outer edge and move large masses of rock in times of
storm; but as the tide rises the surf-line sweeps on across the
flats and rushes with decreasing force to their inner edge. At
high tide, therefore, smaller fragments will be swept across the
flats and be left there with the fall of the tide, so that the
inner edge becomes raised. The intermediate part will bear
the full brunt of the surf during the rising and the falling of
the tide, and so it will become the most levelled by wave
action. The outer margin will receive the first force and the
great lifting power of the ocean waves, and so will become the
site of the biggest of the masses that the waves may lift and

256 CORAL AND ATOLLS

toss up, but are unable to move farther by reason of the waning
of their force. In this way the characters of the typical zones
of the barrier become developed.

We have seen that the first part of the barrier will be
formed at the windward side of the ring, and here its develop-
ment will be always most complete; the larger masses will
be upon its outer border, and its level zone will be far wider
than in any other part of the circle. Now, in the case of an
atoll that lies in the zone of the Trade winds, this will be a
very marked feature of the barrier ; and in the Cocos-Keeling
atoll it is particularly striking.

It will be seen that the very perfection of the windward
barrier tends still further to protect the leeward side of the
reef; and it does this with such effect that the waves are
altogether powerless to bring about the changes that produce
barrier formation. Upon the leeward side the boulders are
not raised, the breccia platform is not made, and there is no
rampart at the level of maximum wave action over which the
seas rush. There is, therefore, a leeward gap in the barrier.

Since the waves are always rolling across the ocean in
one direction, it is only their feeble echo that washes against
the lee side of an atoll in the Trade wind zone, and so the
mariner’s maxim, that the entrance to an atoll lagoon is on the
lee side of the atoll, is a very well-founded one.

It has been said, in the description of the Cocos-Keeling
atoll, that the barrier is the pioneer of the dry land, and that
wherever the barrier exists it is potentially the site of island
formation: a most important stage in atoll formation has
therefore been arrived at. With the completion of a portion
of the barrier, a turning-point is arrived at in the history of
the reef, for now, instead of remaining as a wave-swept circle,
it easily becomes a fully developed atoll. The stages of the ~
process are easily seen, in miniature, at any point in an atoll
ring ; and the whole process of island building may be studied
at any of the inlets between two adjacent islands. Just as the
first step in the formation of the barrier consisted of some chance
boulder being hurled by storms upon the top of its fellows, so

PLATE XIV.

THE ISLAND RISE AND SEA BEACHES ON THE SEAWARD SIDE OF
PULU TIKUS.

DEVELOPMENT OF THE ATOLL — 257

the island starts to form by some boulder being tossed by
the waves upon the platform of the barrier. With the advent
of any permanent obstruction to the waves a new set of con-
- ditions is brought into play.

During my stay in the Cocos-Keeling atoll, I became
impressed by the very marked influence of any hindrance to
the waves and current, and the changes brought about in this
way were fully studied. In the same atoll Dr. Guppy had
previously called attention to the facts, and his account was
the first that established the real cause of the development of
island form. In 1907 Mr. C. Hedley further followed out
these processes during his examination of the Great Barrier
Reef of Australia, and the whole sequence of events, from the
rising of some permanent hindrance to the wash of the
currents to the development of an island, or even of an atoll,
may be taken as fully established. The steps in the process
are very simple, and they may be watched in any stream of
running water; for the development of a sand-bank in a
river follows very much the same rules as those that apply to
the formation of atoll islands. The boulders which the high
seas have tossed on the barrier flat stand in the line of the
waves that sweep across the flat; the current strikes them
and rushes along their face, and dividing, sweeps on in fan-
shaped currents from their extremities. Behind the obstruc-
tion an area of calm is created, and here small particles come
to rest so that, after a time, a little sand-bank is raised. The
currents that become slackened by meeting the obstacle drop
their burden of sediment in lines that stream from its ex-
tremities, and they tend to raise a V-shaped bank, with the
apex of the V at the point of obstruction. As the original
obstruction grows larger, by the addition of new fragments
washed against it, the area of slackened current increases,
more and more sediment becomes deposited, and a bank of
coral débris and coral sand is piled upon the barrier. This is the
nucleus of the island, and such nuclei may be seen at many
points of a barrier ring.

When the nucleus has attained a somewhat greater

R

‘SpUTM JuaTvAodd 94} FO UOT}DeITp OY} SyUoSoIder MOAIE 9],
‘SHAVA\ FHL OL NOLLOALISAOQ:
TVNIDIUQ ANOS WOU ANVIST GAdVHS-INANSAYD V AO NOILVWUOT AML YNIMOHS SIAAYIY ATHY,

& G

j t

NY, ; Vv é !
|

|

SS

T9 ST

DEVELOPMENT OF THE ATOLL = 259

dignity, it will mark the limit of the march of the waves across
the barrier flats, for they will break upon its edge, and any-
thing that they may sweep over the flats will be added to its
growing pile. There is, however, a limit to the extent of the
building seaward, for the force of the waves will not permit
the resting of fragments till they have become spent by some
lengthy journey over the flats. There is also a limit to the
land building lagoonwards, for the force of the spent waves is
not sufficient to carry material any farther; and so the rim of
land must of necessity be anarrow one. ‘There is, however, one
way in which the land pile may spread—and that is by the
extension of its extremities in crescent form. The currents that
make the tailing-off of the sand-bank in the diverging lines from
a single obstruction come into play at the extremities of the
linear island, with the result that the land becomes bent
inwards, and runs towards the lagoon as incurved banks.
Meanwhile, on other spots of the barrier flats, others of these
island nuclei will be developed, and the history of them all
will be similar; save that those to the lee side may be ex-
pected to develop more slowly and to rise to lesser heights.
These nuclei will therefore grow around the ring, ever reaching
out towards each other, and tending to form a dry land circle.
But, as they approach each other, the channel that separates
them becomes filled by a more rapidly moving stream of
water; its rush becoming steadily increased as its channel
becomes narrowed. There is therefore a limit to the growth
of adjacent islands under normal conditions, for the streams
that pour through the gaps between them are too rapid to
allow of the resting of sediment. We may therefore expect
to find the adjacent extremities of two islands to be carried
lagoonwards as incurved banks, and to be separated by
a channel of rapidly flowing water. In the description of
the lagoon this condition was pictured, and the depositing
of the sediment carried by the current, as banks farther out in
the lagoon, was described.

The whole of this method of island building has been
pictured as though it proceeded regularly, and gradually, over

260 CORAL AND ATOLLS |

Fie. 62.

CHART OF THE Cocos ATOLL,

To show the effects of the prevalent winds upon island—and
atoll—formation,

DEVELOPMENT OF THE ATOLL = 261

a long series of years; but I do not imagine this to be at
all the fashion of development of the atoll. Every detail of
its structure points to the fact that the. increments to its
building have been added at quite irregular intervals: it seems
as though the land had taken a rapid stride, and then had
ceased from growing; the quiescent period being again followed
by great additions to its substance. I think it may be taken
as a law of atoll growth that the processes of formation take
place by fits and starts. The evidences that we see of sudden
building increments are without doubt the evidences of storms
of exceptional violence ; and it is no doubt the unusual events
of wind and wave that initiate every process of atoll formation,
and that give fresh stimulus to the progress of their completion.

We have therefore arrived at the stage in which bare
débris islets are raised about the edge of the reef, and in their
formation nothing, save the normal growth of coral, the
deposition of calcium carbonate, and the natural effects of the
winds and waves, is involved.

That the final formation of the atoll is entirely dependent
upon the prevalent winds and seas seems to me to be quite
clear, and for its proof the examination of the charts of atolls,
and atoll groups, is all that is needed. Professor W. J. Sollas.
has said (Brit. Ass. 1893) that clear evidence of the truth of.
the theory of Subsidence is seen in the fact that atolls tend to
be elongated in the direction in which the group to which they
belong is stretched. Now far from taking this as evidence
that a line of subsidence had taken place, I would regard it as
showing that the wind and waves that made the whole group
shaped each individual atoll of the group.

Lhe shape of the group, the shape of the atoll, and the shape ”
the individual islands 1s impressed by the agency of the normal
winds and waves ; and the stimulus of origin, and the increments
of building, are the outcome of the exaggerated violence of the normal
winds and waves.

CHAPTER XXIII
THE AFTER-HISTORY OF THE ATOLL

THE atoll, as far as we have at present considered its develop-
ment, consists of a wave-washed barrier crescent surmounted
by a series of piles of débris which are also crescent-shaped.
The islands, so far, are mere masses of broken coral branches
and loose boulders, heaped up im confusion by the waves, and a
very long period of time may be passed without any advance
from this condition being made. , kl

The “Button Islands” (between Pulu Pasir and Pulu
Bras) of the eastern side of the Cocos atoll are to-day mere
piles of débris and their development has not proceeded in
any way since they were first charted by Van der Jagt in
1829; but this is not the usual history of such structures.
As a rule, the heap of loose coral fragments becomes a con-
solidated and vegetated island ; and in the process the winds
are the most influential agents.

It is obvious that the height to which the seas may pile
the boulders is limited, and the limit depends upon the
amount of tidal variation, and the violence of the storms that
pass over the tract of ocean in which the atoll lies. The
influence of the range of tidal variation is easy to account for,
and the greater the rise and fall of the tides, the higher in all
probability will the island be; for since the tides have a
greater excursion upon either side of their level of maximum
wave force, it follows that the range of possible building
will be greater from this level when the tidal variation is a
large one. When the tides have an excursion of 3 feet it may
be roughly said that 1 foot 6 inches of broken water will be
available for land-building above the level of the barrier:

when the excursion is 5 feet another foot is added to its range
262

THE AFTER-HISTORY OF THE ATOLL 268
of activity, and so with increasing tidal excursion a greater
elevation of island-building is possible.

The influence of great storms upon the height of island-
building is easily seen, for it is simply a question of how far
up the débris slope may the largest waves of the greatest
storms hurl any additional fragments. Pumice represents
the tell-tale of maximum wave action, for a fairly large frag-
ment of light pumice is the float that the spindrift of a broken
roller will carry farthest in front of it; and pumice, in most of
the islands, has been flung over the crest of the sea beaches
and on to the island land. Storms of exceptional violence,
such as that which visited this atoll in 1876, have overridden
their rampart so completely that waves have swept into the
islands for a distance of 150 yards from their seaward margins.
It is only the possession of positive data of this kind that
makes it possible to realise how any sea could toss some of
the enormous rocks of massive corals to a height of fifteen
feet or so, after a journey of more than a hundred yards across
the reef flats.

Upon the lee side of the atoll the waves may only toss
fragments to the top of a bank some three or four feet high, and
in Pulu Tikus the most exposed portion only has been sea
built to a height of thirteen feet. Upon the windward side
the building rises higher, the wall of the sea beach is steeper,
and is composed of far larger boulders than the sea beaches of
the lee side. The rampart here may be piled to a height of
twenty feet, and the blocks that enter into its composition are
often far too large for any man to move.

When the pile of débris has risen as high as the waves
can cast their burdens, it must cease from further growth from
sea influence unless some storm of quite exceptional violence
visits the group, or unless the seaward edge of the barrier reef
grows farther seaward. Since we have seen that the distance
of the island from the barrier edge is regulated by the force
of the waves—the windward islands being farther from the
edge than the leeward islands—it is evident that the seaward
margins of the island will advance as the seaward margin of

264. CORAL AND ATOLLS

the barrier advances. The seaward advance of the barrier
edge is of necessity very slow, for a vast amount of submarine
building must take place before the wave-washed edge can
grow outwards. The barrier edge probably grows outwards as
the outcome of unusually high seas, and the evidences that are
to be seen of sudden increase of island building have probably
been preceded by similar increases in barrier building. With
the exception of these uncertain and irregular methods of
building the sea is, therefore, powerless to further add to the
growth of the island’s dry land; it is the wind that completes
the moulding of the pile of coral débris. When the building
by coral boulders and fragments ceases, the building by coral
sand begins ; in all the stages of the after-history of the atoll,
sand is by far the most important factor. The sources of this
sand are manifold. Most of the sand is made on the barrier,
where the grinding of wave-driven fragments is for ever
taking place. On the flats, the broken branches of Madre-
pora and fragments of massive colonies are for ever sweeping
backwards and forwards in the waves, and being reduced to
smaller and smaller particles. Sand is always being pro-
duced all round the atoll ring, wherever rough water moves
fragment on fragment; and vast quantities are always in
the process of production, becoming increasingly fine as the
waves carry the particles shorewards. Compared with this
method of sand production, it is probable that any others are
of quite trivial importance, but of such as do make an appre-
ciable contribution, I am convinced that in the Cocos-Keeling
atoll, at any rate, the fish of the family Scarus are by far the
most important. With their hard beaks these fish are always
nibbling the masses of dead coral upon which Nullipore and
other algze have become encrusted; they actually bite deep
grooves into the rock and the linear scores that they leave
upon the boulders are easily identified, although Europeans
usually fail to associate them with the fish which undoubtedly
produce them. ‘The most important sand-producing species
is a blue Scarus—native name Kakatua Ijou—which weighs,
as a rule, about five pounds; this blue fish feeds round the

THE AFTER-HISTORY OF THE ATOLL 265

barrier in thousands, and its attention is wholly confined to the
coral rocks of the atoll, for no other feeding-ground exists
for over five hundred miles. I have shot large numbers
of Kakatua Jjou, and have examined their intestinal con-

. Fig. 63.

PHOTOGRAPH OF A BoULDER OF ALGA-COVERED DEAD CoRAL
Rock,

To show the bites of the fish of the genus Scarus. The black line
marks the edge of the alga covering not bitten away by the fish.

tents. Every fish that I have killed has been filled with coral
sand of its own producing: the average weight of washed
and dried sand found in them is two ounces. Considering
the activity of these fish and their vast numbers, the addition
that they make to the sand-supply of the atoll must be very
considerable. Both Darwin and Forbes have described these
Scart as eating the living coral, but there is no doubt that
the observation is incorrect, for they never by any chance
touch a living colony.

The Holothurians are also a source of some sand-pro-
duction, but I do not think that their contribution is a very

266 CORAL AND ATOLLS

serious one. It is true that they contain large quantities of
sand—indeed Dr. Guppy, who attached much importance to
their influence, calculated that fifteen of them passed a ton of
sand through their bodies every year. It is quite possible
that some such figure may hold true, but I do not regard this
ton as any addition to the island’s sand capital, for the Holo-
thurian has merely passed it through its body. It was sand
when it went in and it is sand when it comes out, so I do not
think that the Holcthurian is a serious rival of the Scarus in
this respect or a great agent in sand-production. The influence
of the various boring molluscs and worms is again still less,
and their importance as sand-producers is trivial indeed, and
has, as a rule, I think, been very considerably overrated. In
the Cocos-Keeling atoll there are only two agents that produce
sand from solid coral rocks in any appreciable quantities; —
one is the waves of the barrier and the other is the |
fish of the genus Scarus, and the influence of these latter
has hitherto, I believe, been entirely neglected or very much
under-estimated.

The sand, by whatever method it is produced, is the lightest
of the particles that the waves carry—it is the finest product
of coral trituration—and it will therefore be the last burden to
be deposited by the waves. The boulders are left on the
barrier edge ; the smaller fragments are deposited on the barrier
flats or added to the island pile; but the sand is carried high
up the seaward beaches, or far out into the lagocn. The sand
that is thrown upon the island beaches becomes dried in the
sun, and scattered over the island land by the winds. The
effects of this wind-blown sand have been seen in the ex-
amination of the islands, and the characters of wind-made
islands have been noted.

The islands of the windward side of the reef will be the
first to obtain an abundant supply of sand, and their rough
coral boulders will soonest become clothed by a layer of sand.
In places, great mounds of sand will become piled up-—these
are the atoll mountains, and they may rise to a height of 30
feet under favourable conditions.

THE AFTER-HISTORY OF THE ATOLL 267

Ridges of sand will be thrown across the islands, and will
rise in the lee of any prominent boulder, and stretch away
from it on the line of prevalent wind. In time, the general
surface of the land becomes raised as a whole until, in the
case of the windward islands, there. is no trace to be seen of
the original low coral débris-pile that forms their foundations.
The leeward islands will remain long unclothed by sand, and
when the windward islands are already covered and thickly
vegetated, the leeward islets will still be mere piles of broken
coral. Some islands are built almost entirely of sand by the
agency of the wind, and then they may attain a height of
some 30 feet as pure sand-heaps; when this is the case, the
highest land is on the leeward side, as we have already seen
in the description of the islands of this atoll. I think that the
importance of wind building,in the after-history of the islands,
is generally rather neglected, but the reality of it may be
appreciated by observing that all of the older trees of the
windward islands are buried for a very considerable distance
by the sand that has been accumulated during their lifetime.
The principle which is involved in this completing of the island
characters is one that is of general application, for upon the
western bank of the Nile in Nubia I have watched every stage
of land-making under the influence of wind-blown sand. Here,
where the sand from the Western Desert is for ever blowing
over the granite land of the Nile valley, I have watched all
the details of ridge-formation, the burial of boulders, and the
general elevation of land-level, such as I had already noted in
this atoll. The arrival of sea-cast seeds is the next event in
the island history, and of all the coral island vegetation the
coconut is the pioneer. Although Dana says that “there is
no known evidence that any island never inhabited has been
found supplied with coconut trees,” still it is a fact that
coconuts have been noted on this atoll ever since its discovery
in the early part of the seventeenth century, and there is no
evidence of any inhabitants being in the islands previous to
1825.

The coconut owes its success as a pioneer to the fact that

268 CORAL AND ATOLLS *

it will start its growth wherever the waves toss it. This may
be seen at any point of the atoll, for the nuts that are dropped
into the sea will be washed up again upon some other beach,
and, if they come to rest upon a patch of sand, ee will start
their growth straight away.

Other hard sede arrive and are blown by “he wind into
places that are suitable for their growth, and an abundant
vegetation grows up. Another source of land-building is now
added, for as the vegetation increases, leaves and branches
become added to the sand accumulations, and the covering of
the coral débris proceeds with added rapidity. It might be
thought that, with the luxuriance of tropical vegetation, a
vegetable mould would very rapidly accumulate, but as a
matter of fact it is not so, and, on the islands that have no
depth of sand, the perpetual addition of dead leaves and nut
husks seems to make but very little difference. The reason
for this is the presence of the heavy annual rainfall, for as
fast as the leaves fall and rot, the rain washes their remains
into the interstices of the porous coral heap.

With the addition of the vegetation the atoll island is
completed, and the condition of a sand-built vegetated strip
of land is one that lasts for a long time in its hiss

The sand that is made by the barrier waves and by the
fish is not all flung upon the land and built up by the winds,
for some of it is “irene rushing through the inlets to the
lagoon, and being deposited from the water in various places.
We have already noted the method of incurving the island’s
ends by the tailing out of sand-banks into the lagoon, and
this process is one that continues all through the life of the
island. With the elongation of the horns, the area of
smoother water that stretches between them increases and
becomes the site of more sand deposit: in this way the
lagoon shores become widened. ‘This process again takes place
with greater rapidity at the windward side of the atoll, for
there the supply of sand is the greatest, and the lagoon beaches
are under the shelter of the islands. Wide stretches of sandy
beaches are therefore laid down upon the windward side of

THE AFTER-HISTORY OF THE ATOLL 269

the lagoon. It is easily seen how the forces that shape the
whole atoll into a horse-shoe, or an incomplete circle, are acting
on the individual islands, and that the incurving of the ends
of the islands may easily bring about the partial enclosing of
a portion of the lagoon between the horns. Horse-shoe- sine weil
islands are thus produced, and by the deposition of sand-bars
they come to enclose a private lagoon of their own. This
process is well seen in the Cocos atoll, where all stages of the
evolution are at hand.

Each island of the reef therefore tends to become atoll-
shaped, and to enclose its own little lagoon. In the Maldives
this process is seen in great completeness, and it has advanced
so far that, in the case of many reefs, a series of small atolls
stand ringed about a central lagoon. When an atoll lies in
the region of the Trade winds, it would seem as though the
islands to the lee side of the reef could never become
atollons, for although their extremities are incurved by the
water running through the inlets, the wash of the lagoon waves
would always be likely to lead to the unmaking of banks,
when the lagoon was sufficiently large to afford any sort of
sea. The Maldives are in the monsoon zone, and therefore get
a seasonal change of wind: but I notice, from the charts, that
the western atollons are most complete; and I would suspect
the west monsoon to have the greatest influence on their
building. The final stage of the process that leads to in-
dividual islands enclosing their own lagoons is the filling in of
the area that they have enclosed. Many agencies come into
play in this, and the most important is again the wind-blown
sand. The agent of second importance is the vegetation ; and
erabs of the genus Gelasmus, and Cardiosoma, also play their
part, very much after the fashion of earth-worms. A fine
white mud is produced in these little lagoons by the decom-
position of the coral sand; bushes of Pemphis acidula, coco-
nut palms, and coarse grass take root in the mud; sand
is driven about their roots, and the whole area is slowly
reclaimed.

One other fate may befall the island in the course of its

270 CORAL AND ATOLLS

after-history :—it may become joined across the reef flat to
its neighbour or neighbours. We have seen that the reason
why islands do not normally unite round the edge of the
reef is that, as the channel between them narrows, the current
of water becomes swifter, so that the deposition of sediment
in the channel is impossible. If, for any reason, the current
between two islands should slacken down, then the deposi-
tion of sediment will begin, and island may stretch across
to island; this is, of course, what actually happens. Many
things may lead to a slackening of the current in the in-
lets; a storm of unusual violence may cast boulders into its
channel, and be powerless to move them farther. Other
fragments collect against them, and the channel may become
blocked up, the wind-blown sand will then complete the work,
and join the islands across the gap.

The closure may be more gradual, and in this pradiell
blocking the beds of calcareous Ales no doubt play some part.
As a rule the inlets to the lagoon are not favourable sites for
coral growth, and some stunted colonies of Montipora alone
live in them; but here the only considerable beds of the
Lithothamnionee are found. Within the meshes of the growth
of these algee much sand becomes held, and, in many spots, it
would seem that they are steadily filling up the channels, and
will in the course of a great many years succeed in blocking
the waterway.

Besides the sand that is deposited as continuations to the
extremities of the islands, a vast amount is carried on into the
lagoon. A part of this goes, as we have seen, to the making of
broad beaches to the lagoon shores of the islands, and a
general filling in of the lagoon. ‘The sediment borne in the
middle of the stream is carried on the farthest, and, with the
slackening of the current, it is dropped as banks far out in the
lagoon. This point has received some previous notice, and
it is mentioned again as it is one of some importance ; the
proper understanding of it makes clear some of the changes in
the after-history of atolls. As a general principle it may be
laid down that the currents in the windward inlets set con-

THE AFTER-HISTORY OF THE ATOLL 271

stantly lagoonwards. These currents always carry large
quantities of coral sand, and as the stream becomes first
retarded at its margins, the first sediment to be deposited is
in lines along its sides, forming the continuation of the island
extremities which we have seen. The central part of the
stream sweeps on, and only becomes slowed down as it loses
itself in the waters of the lagoon ;—the sediment that it con-
tains will therefore be deposited opposite the gap in the island
ring.

It is easy to see that if the lagoon were only a small one,
and the islands that composed the ring were very few, then it
might chance that the windward inlet into the lagoon came to
lie opposite to the leeward entrance to the lagoon. It might
also happen that the sediment brought in by the windward
inlet would not be carried right through the lagoon and out
into the ocean beyond, but would be dropped as a bank actually
in the entrance itself. It is maintained that this is what in
reality happens in some atolls. The condition that would be
met with, when the process was completed, would be a wind-
ward entrance to the lagoon ;—but this windward entrance is guarded
by a barrier reef. This is exactly the condition that is found
to-day in North Keeling atollon. The gap is to the windward,
but the windward barrier is complete—there is no entrance to
the leeward, but the lee barrier is absent. The current that
has swept into the lagoon, through the windward gap, has
made the sand-spits incurving the land which were noted by
Dr. Guppy; and it has also raised a sand-bank, so large that
it has blocked up the real leeward entrance to the lagoon.
The sand-bank that is deposited by the current becomes raised
high enough for the waves to cast some of it above their
reach ; the wind then comes into play, and a wide sand-barrier
is made across the old lagoon entrance. The gand-bar con-
tinues to grow, and the whole of the leeward side of the atoll
becomes a wide sandy stretch, probably far broader than the
original dry land of the windward island.

- It is almost impossible to judge of the method of formation
of any atoll not actually visited and examined; but from the

272 CORAL AND ATOLLS

charts, I would expect that some of the atolls in the Laccadive
Archipelago were the outcome of this method of building.

Fic. 64,

DIAGRAM TO ILLUSTRATE THE DEVELOPMENT OF
KEELING ATOLL.

Srage I, in which two crescentic islands are raised upon the
windward side of a horse-shoe-shaped barrier flat.

The Chetlet and Kiltan atolls strongly suggest this process,
and Mr. Stanley Gardiner says “ practically all land lies on the
eastern or leeward side of the reefs, the group being com-
pletely exposed to the gales of the south-west monsoon, while
it must be largely protected by India from those of the north-
east.”

THE AFTER-HISTORY OF THE ATOLL 273

This condition he ascribed to “elevation and subsequent
erosion, the latter having completely removed all save traces of

Fic. 65.

DIAGRAM TO ILLUSTRATE THE DEVELOPMENT OF
KEELING ATOLL,

SraGe II.,in which sediment is being dropped as banks
continuing the island extremities, and also as a sand-bank
opposite the windward inlet.

the land from the western reefs.” Without an actual ex-
amination of the atoll, it is easily possible to fall into error,
but the process which has, without doubt, caused the characters
of Keeling atoll, would also seem to be quite sufficient to
explain the form of these Laccadive atolls.

In larger atolls this process does not happen, for the

sediment-carrying current drops its burden before it reaches
iS)

274 CORAL AND ATOLLS

the entrance to the lagoon. In this way the windward. part
of the lagoon will commence to shoal. The shoaling of

Fira. 66.

DIAGRAM TO ILLUSTRATE THE DEVELOPMENT OF
KEELING ATOLL.

Srace III.—The further development of the formation
of sand-banks in the leeward lagoon entrance.

the lagoon of the Cocos-Keeling atoll is proceeding steadily,
and year by year more and more sediment is laid down upon
its bed. Such shallowing is the probable fate of all lagoons.
Two agencies are at work ; coral growth tends to fill up the
lagoon slowly, and sedimentation tends to fill it up more
rapidly; but, as we have seen, these two factors cannot act in
the same area, As a matter of fact the two processes work

THE AFTER-HISTORY OF THE ATOLL 275

quite separately, and the sedimentation spreads as a wave
across the lagoon, first killing the corals and then shallowing

Fic. 67.

DIAGRAM TO ILLUSTRATE THE DEVELOPMENT OF
KEELING ATOLL.

Srace IV.—The sand-bank completed and a part of it
already made into an exposed bank by the action of the
waves.

the water by its own efforts. In time, this process
will spread across the whole area of the lagoon, and the
processes which have obliterated the secondary lagoonlets of the
horse-shoe-shaped islands will finally fill in the main lagoon.
This chain of events has of course already occurred in many

276 CORAL AND ATOLLS

atolls, and Pulu Luar in this group is an example of the final
stage of atoll-making—a stage in which a rim of slightly

Fic. 68.

DIAGRAM TO ILLUSTRATE THE DEVELOPMENT OF
KEELING ATOLL.

Srace V.—The final stage, in which a large area of

leeward land is formed, and the only inlet to the lagoon is

to the windward. The inlet is guarded by a barrier and
to the lee side of the atoll no barrier is developed.

raised land surrounds a central basin, wholly or partly dry at
all states of the tide.

We have followed the fate of the sand that becomes cast
upon the land, and of that which becomes deposited in the
lagoon, but this by no means represents the whole of the

THE AFTER-HISTORY OF THE ATOLL 277
sand which is produced around the reef. A vast quantity never
reaches the barrier flats, but falls down the submarine slope,
and adds to the sedimentation bank upon which the reef rests.
Another great contribution is removed again from the lagoon
by the currents which run out of the leeward channels, and this
too is added to the submarine bank.

In the Cocos-Keeling atoll the current that runs from the
lagoon mouth impinges upon another island—Pulu Luar—
and on its windward shores sand-building is taking place with
some rapidity. This is not, however, the common fate of
sediment removed from a lagoon, for in ordinary cases it would
be swept to sea, and dropped so as to form an elongation of
the bank to leeward. By the addition of sand and fragments
that drop from the edge of the reef, the submarine bank is
therefore always tending to grow; but there is every reason to
imagine that its growth will be very slow. The additions to
the bank will enable the reef margin to advance seawards and
so the whole atoll may become enlarged, but this must be an
extremely slow process in an atoll that has its origin in very
deep water. But I do not doubt that all atolls, and
especially those in shallower seas, grow outwards upon banks
of their own talus, and that this outward growth is an im-
portant factor in atoll construction.

The atoll has therefore reached its full development, and,
short of any actual change of land-level, it will remain at this
stage for an incalculable time, its only alteration being a very
tardy spreading of its boundaries. In these final stages of
atoll development the presence of sediment is again the
dominant factor. Sediment played an all-important part in
the early stages of the formation of the structure, and in its
closing scenes it again becomes the guiding and directing
influence.

CHAPTER XXIV

THE DEVELOPMENT OF OTHER CORAL STRUCTURES
AND THE INDICATIONS OF ALTERATION
OF LAND-LEVEL

I HAVE already said that no concrete theory of atoll formation
may be considered as at all satisfactory unless the principle
that underlies it can also be used to explain the development
of other coral formations. It therefore becomes necessary to
see what support the ideas connected with the processes of
“ Sedimentation” will gain from an examination of fringing
reefs and barrier reefs. Now I take it as a strong argument
for the truth of these views that when I came to apply them
to the formation of a barrier reef, I found that I was the third
in a succession of investigators who had quite independently
conceived that the real factor in the formation of barrier reefs
was the process of “ sedimentation.”

So long ago as 1856, Professor Le Conte had given it as
his opinion that barrier reefs were fringing reefs whose growth
was “limited on one side by the muddiness of the water, and
on the other by the depth.” *

Dr. Guppy, without knowing of this explanation, had come
to the same conclusion, and had published his observations in
1884., When in the Cocos-Keeling atoll—-and without a
knowledge of either of these papers—lI realised that the
picture I had drawn of the processes of sedimentation would
easily account for all the features of a barrier reef.

Since the reef-building corals will begin constructing
upon any platform that rises to within the area above the
Limiting Line of Sedimentation, it follows that a reef will form

* American Journal of Science, 2nd series, vol. xxiii. p. 46.
+ Proc. Linn. Soc. N.S.W. vol. ix. Part 4.

2

OTHER CORAL STRUCTURES 279

about any land whose submarine slope gives this necessary
foundation.

In this way a fringing reef will be formed, and it will be able
to extend seawards as far as its submarine base stretches within
the required limits, and after its land base is exhausted, upon a
bank of its own talus. But, near the shores of the land that
it fringes, the water will commonly carry a heavy load of sedi-
ment, which has been washed from the land by wave action ;
and so a near-shore limit will be set upon the growth of the
coral colonies. This is the most simple explanation of the
formation of barrier reefs, and since it has been agreed upon
by two investigators, who did not advance the action of
“sedimentation” as a factor in causing atoll formation, it
gains an added weight. The agency that causes barrier-reef
formation would seem to be almost certainly the same as that
which caused the kindred structure of the atoll, and Darwin re-
cognised this fact, for it would seem absurd that two separate
conditions should be involved to account for two such similar
structures. At the least, it may be said that it would appear
illogical to imagine barrier reefs to be formed by “‘ sedimen-
tation,” and the nearly allied atolls to be formed by the directly
opposed action of “ solution.”

I therefore imagine that the workings of “ sedimentation ”
upon the living coral colony, and upon the living coral reef,
will account for all the different coral structures that are met
with.

The famous theory of “subsidence” also did this, and it
did it in a most complete and masterly way; and it is natural
to inquire if in a fully developed atoll such as Cocos-Keeling
there is any sure sign that subsidence has not taken place or
caused its features. It has been said by the supporters of this
theory, that subsidence, like a conflagration, destroys its own
traces; but in reality it is not so. We have seen, in the
course of the development of the atoll, that the first formation,
after the reef has come to the surface, is the broad flat layer
of the barrier. It has also been insisted, in the description of
the barrier, that the breccia stratum is an unbroken layer from

280 CORAL AND ATOLLS

the surf edge of the barrier to the breccia rock of the lagoon

shore. ‘Too much importance cannot be attached to this fact,

which is sometimes forgotten or not appreciated. The islands

are mere piles of débris upon this level platform. It is easy to see

that the lagoon margin of this platform is the first formed effort

of the waves upon the newly risen reef, and that once formed,

it has no power to alter its level in any way. The forces that

made it are retreating, and as it was when made by the waves

at the first gale that met the reef, so it must remain. The

extreme seaward edge of the barrier represents the very latest

effort of wave-building, and so we have from seaward edge to

lagoon shore a platform that represents the sea-building

throughout the entire history of the atoll. It is true that the

lagoon edge of the platform becomes covered with beach sand,

and that the intermediate part becomes covered with the
débris pile of the island; but at any point the breccia stratum —
may be exposed below its coverings. .

Since the breccia platform, when once cotnplbted by the
waves, and finally deserted by them, can never undergo any
further building, it follows that the inner margin of the
platform will certainly give the level of the reef, at the time
of its very first meeting with the waves. If the whole atoll
has sunk, then this inner edge of the platform must occupy
a lower plane than its outer edge. The sinking, if it did
happen, would not destroy its own traces, but would have a
very definite index in this lagoon edge of the breccia platform.
As a matter of fact, in this atoll, I have failed to find on
levelling across an island any difference between the level of
the barrier flats to seaward and the breccia platform exposed
on the lagoon shores. I therefore conclude that in the whole
long history of this atoll—from the time at which its reef
first came within the domain of the breakers to its present
advanced condition—no subsidence has taken place. Therefore
I reject “ subsidence ” as an explanation of the features of the
Cocos-Keeling atoll.

Since there is no trace of subsidence to be found in this
atoll—nothing to be read in its history, or to be found in its

SSOMIL NTOd AO ALINAUYEXA NYUISVA AA AO AaIS
GUuvVMVAS WH NO WHOALVTd VIOONUNd AO SdULS GASIVA AHL MOHS OL ONIMVUG

“AX GLV Id

OTHER CORAL STRUCTURES 281

structure, to indicate it as a monument raised upon sinking
land—it becomes natural to inquire if there is any indication
of the opposite movement ; whether there is any suggestion
of an ultimate hoisting from the ocean-level, as has occurred
with its nearest neighbour, Christmas Island. As a result of
his examination of the atoll, Dr. Guppy gave it as his
opinion “ that neither of upheaval nor of subsidence is there
any evidence of an unequivocal character.” These words were
_ written in 1888, whilst in 1878 Forbes had said: “ Between
Direction Island and Workhouse Island I observed what
seemed to me signs of recent elevation” ; andin 1836 Darwin,
as has been mentioned, believed that, in the undermining of
some coco palms he observed those signs of sinking which he
had looked to find.

The evidence of sinking that he found he describes as
follows: “At Keeling atoll I observed on all sides of the
lagoon old coconut trees undermined and falling; and in one
place the foundation posts of a shed, which the inhabitants
asserted had stood seven years before just above high-water
mark, but now was daily washed by every tide.

There is no real difficulty in accounting for such diverse
opinions, given by such highly trained observers. Darwin
stayed but ten days in the atoll, and his visit was in April
1836; Forbes remained for three weeks in January and Feb-
ruary 1879; and Guppy landed in August 1888, and his
visit lasted for about ten weeks.

As the island dry land is undergoing a perpetual cycle of
making and unmaking with every set of the local currents, it
becomes evident that in order to form any just estimate of
the resulting outcome of these forces, the observer must watch
the process for a considerable time.

As a result of my prolonged stay I have no doubt that the
gain is ultimately to the land, and the opinion of the Governor
is that the islands are always winning from the sea; though
at any time in some portion of the atoll you may point out
spots in which the water is undoubtedly encroaching. The
atoll comprises a large strip of land—much larger than any

282 CORAL AND ATOLLS

casual visitor can hope to examine with any thoroughness—
and it is easy to draw the most diverse conclusions from the
inspection of restricted areas; but with time and familiarity
with more of the islands, it becomes evident that there is
a compensation in the process: but it is more than a mere
compensation, for the repair tends over the whole atoll to
outweigh the destruction.

Although much importance has been attached to the
various indications of land gain and loss of the atoll, I do not
think that most of the processes of building and wasting that
can be seen daily in progress give any real clue to the actual
earth movements to which the land may be subjected. What
we really see when we watch a sandy beach piled up, or a
strip of shore denuded and the palms uprooted by the waves,
is the fluctuating gain and loss of the islands’ floating capital.
A bank or a spit is made and unmade by a shift of wind or
current, it does not indicate a rise or fall of island-level.

We have seen besides that great alterations may be brought
about by actual temporary change of ocean-level, for, during
periods of several days at a time, the surface of wide tracts of
ocean may be raised or lowered locally by the banking-up
effects of winds and currents, and there is nothing to dis-
tinguish this temporary alteration of water-level from an actual
depression or elevation of the land. [ do not think that
this curious phenomenon of the local variation of ocean-level
is sufficiently considered; for it is a very real and a very
potent fact; it is only when your extreme elevation above the
sea happens to be a few trivial feet that its reality strikes you.
For days at a stretch during storms to the westward, the level
of the ocean in the neighbourhood of the atoll will be most
distinctly raised; and any denudation that may occur during
this time will present all the appearances of alteration of land-
level. For actual indications of land movements, other things
than the shifting sand-banks must be looked for, and evident
indices of elevation are furnished in the raised platforms of
breccia which may be found in certain portions of the barrier
ring. The breccia of the shore platforms is entirely the

OTHER CORAL STRUCTURES 283

outcome of subaqueous processes, and it is impossible to con-
ceive a breccia platform to be made in sitw beyond the reach
of the waves; indeed when the composite rock is long exposed
to the air, and rain water, it suffers a certain amount of
disintegration, and its sub-aerial life is a limited one. Since
breccia platforms may to-day be found intact, and exactly in the
condition in which they were laid down, beyond the reach of
the highest spring tides, it is evident that since the date of
their formation they have been elevated by an actual upward
movement of the land. Between Pulu Tikus and Pulu Pasir
these signs of elevation are striking. A notable fact in their
disposition is that they occur in series,one layer above the other ;
successive land movements being registered by successive steps
of breccia, which rise one above the other to a total number
of three or four, and to a height of almost as many feet.
At that part of Pulu Tikus where these steps are evident,
the island rise is thirteen feet above mean tide-level, and is
by far the highest land of the island. I can imagine no other
explanation for the presence of these raised platforms than
that afforded by the supposition that the land on which they
were laid down has been raised, as a whole, since their forma-
tion, and on the evidence of these raised platforms I presume
the Cocos-Keeling atoll to have been subjected, locally, to
successive slight waves of elevation.

The positive evidence afforded by raised seaward beaches
is not so strong, for gales of abnormal violence may alter the
configuration of the débris bank so markedly that a raised
field of broken coral may be the work of a few short hours of
storm, and the loose fragments may be easily thrown beyond
the reach of any subsequent high spring tide. The formation
of breccia platforms must of necessity be very slow, and no
sudden violence or temporary condition can cause the building
of breccia beyond the normal reach of the waves. Wherever,
therefore, raised and intact platforms of breccia are found,
we may safely argue that actual elevation of land-level has
been brought into play; but when raised débris beaches or
raised sand beaches are found, then although every appearance

284 CORAL AND ATOLLS

of elevation may be present, still it must be remembered that
the result seen may be brought about by changed conditions of
sea-level or violent storms.

When a raised breccia platform is found upon the seaward
barrier, the breccia platforms of the island and of the lagoon
shore of course share in the movement. The land undergoes
the same slight alteration of level, and the slabs of breccia on
the lagoon shore become tilted. At the eastern end of Pulu
Tikus these effects are well seen, for the old breccia platform
has become undermined by the lagoon wavelets, and has
become cracked and broken, and in many places moved from
its original site.

I have not included among the possible signs of an upward
movement those large blocks of coral rock that are found
isolated upon the barrier, and in some atolls have been given.
the name of “negro heads.” There are many enormous
“negro heads ” upon the windward barrier flats of the southern
islands of the Cocos-Keeling atoll, but I never regarded them
as being anything else than the tell-tales of the efforts of
unusual storms. The experience of living on the atoll during
a spell of cyclone weather confirmed this view, and I conceive
the storm-driven waves to be capable of tossing up the largest
“negro head ’ that I have ever seen. The power of the waves
to move these blocks has been denied by Semper (English
Trans., 1883, p. 239) and by Wilks, but the recent work of
Hedley on the Great Barrier Reef leaves no doubt that “ negro
heads” are in reality wave-cast boulders. That the matter
should become a subject of theoretical speculation seems
almost absurd, for the people who have been born in the
atoll, and have lived all their lives upon it, know of these
things, and remember the advent of outstanding “negro heads”
during the many cyclones which have passed over the islands.

An interesting point is to be noted in connection with some
of the elevated rocks of the barrier, and that is the fact that
some of them are capable of producing musical sounds. By
the action of the sea they become hollowed and tunnelled
so that the water rushing into them causes the air to be

OTHER CORAL STRUCTURES 285

driven out through holes, with the production of various
curious notes.

At the back of Pulu Tikus is a rock which at different levels
of the tide and under different conditions of wind and sea
emits sounds that range from a deep sigh to a shrill scream.

The point is of interest in connection with the island
legend that a spirit voice exists which, calling to people—
especially young children—lures them to their fate over the
barrier edge.

PART IV

THE FLORA AND FAUNA OF THE
GROUP

CHAPTER XXV
THE METHODS OF NATURAL COLONISATION

THE plants and animals which have succeeded is gaining a.
foothold upon the islands are comparatively few in number,
and for the most part the living things of the land are
inconspicuous creatures, with but little of that brightness of
colouring or luxuriance of growth which one is apt to associate
with Nature in the tropics. A great interest, however, attaches
itself to the population of a coral atoll, and every living thing,
although it may not be attractive to the eye, possesses an
added importance from the fact that it has effected a landing,
and made a home, in so distant a spot. For this reason I have
paid an especial attention to the flora and fauna, and the lists
which are added as an appendix to this chapter may be con-
sidered as fairly complete for most of the orders.

It is desirable that such a place as Cocos-Keeling should
possess a record of the state of its flora and fauna at this
time, for, with the increased intercourse with the outer world,
changes are always liable to take place; new forms may be
introduced and old ones exterminated with great rapidity.

This has already happened in the case of Christmas Island,
and were it not for the careful recording of the state of the
living things by Dr. C. W. Andrews in 1897, we should now
be in ignorance of the extraordinary changes that have taken

place so suddenly in the fauna, and we should lack the
286

NATURAL COLONISATION METHODS 287
knowledge of some interesting creatures that have gone for
ever.

No good record of the Cocos-Keeling fauna and flora is at
present available. The visit of Darwin in 1836 was too short to
enable him to make a collection which was at all complete, and
during his stay of ten days in April he noted only the rats,
the rails and snipe, one small lizard, and 13 species of insects
belonging to 8 distinct orders.

Darwin’s observations were quoted by Wallace in his
“Tsland Life” (p. 275), and until the visit of Dr. H. O. Forbes
they remained our sole guide to the conditions of life upon the
islands.

Dr. Forbes visited the islands in 1879, arriving on
January 18, and staying till February 9, and he amplified
considerably the list made by Darwin; unfortunately his
collections were lost in returning to Java, and so the additional
species which he observed have not been specifically recorded.*

On August 20, 1885, Mr. E. W. Birch, on behalf of the
Straits Government, landed and made inquiries about the
islands and their inhabitants; with him, as naturalist, went
the Rev. E. C. Spicer. The expedition visited most of the
islets, and remained for eight days in the atoll, but in the
report (Straits Blue Book, 1885) no light is thrown on the
condition of the fauna.

In succeeding Blue Books are scattered notes, made by the
Commissioners, on some of the most striking features of the
atoll fauna, but most of this information is mere interpreta-
tion of local legend, and is of no value.

Dr. H. B. Guppy visited the islands in 1888, and has
written an excellent account of their physical features in the
Scottish Geographical Mazagine (vol. v., 1889, pp. 281, 457, and
569), but he has not, so far as I am aware, published any
description of the fauna.

I collected the insects continuously throughout my stay of
15 months, and I am convinced that nothing short of a fairly
prolonged residence, even in such a tiny place as is the atoll,

* “ A Naturalist’s Wanderings in the Eastern Archipelago,” 1885.

288 CORAL AND ATOLLS

can yield a collection that is at all representative. A casual
visitor, whose stay is limited to a few weeks, cannot hope to
meet with the entire insect fauna, and nothing but protracted
collecting can give results upon which it is possible to fairly
estimate the fauna of even the smallest coral island.

I do not think that the increased number of creatures
recorded here indicates an actual increase in the fauna of the
place, but only represents a more thorough and prolonged
collecting. The Governor is of the opinion that additions
have not been made to the fauna within recent years; and
those species to be seen to-day have been for long in the
islands. Still, changes are increasingly liable to occur as time
goes on, and it is very much to be hoped that this census-
taking of the fauna will be repeated from time to time.

It is obvious that there are only two routes for emigrants
to such a place as Cocos-Keeling; a creature must either
come by water, or by flying through the air; and the entire
population may be separated into water-borne and wind-borne
waifs. The water-borne division comprises two very different
sets of colonists: the one party of those who made the voyage
by chance, and with much hardihood, at the mercy of the sea-
drifts; and the other whose members came as passengers in
ships, in company with that greatest distributor of animal types
—the merchant sailorman. The first party is the most worthy
of study. In order to reconstruct a history of the methods of
their transit, some idea must be formed of the ocean flotsam
at their disposal, and of the oceanic drifts which carry the
flotsam. No better method can be followed than the exami-
nation of the wave-cast material of the seaward beaches ; and
a daily tour of a stretch of barrier reef, for a period of several
months, will lead to the discovery of many potential pas-
senger vessels. The trunks of trees are naturally the most
abundant of all the flotsam that might serve for the colonist,
and it must be remembered that a single tree-trunk may
be the only remnant of what was once a real floating island.
The very name “floating island” creates the suspicion of
romance, and yet floating islands are realities. Many of

NATURAL COLONISATION METHODS 289

the trees that fringe tropical shores (and one of the most
notable features of tropical shores is the presence of dense
vegetation right down to the lap of ocean waves) have com-
plicated root systems, so that, above ground and below, the
trees tangle this way and that, and form inseparable masses.
The mangrove is a good example of this type of growth, for a
clump of trees will grow in so great a confusion of entangle-
ment that the separation of individuals may be impossible ;
and, moreover, the whole will be firmly braced together by
creeping plants, which run from branch to branch like cordage
to the masts of ships.

A scour of the sea-coast, or a river in flood by heavy rain,
may undermine such a clump, tear it away, and set it adrift
to sea, with all the earth still held in the meshes of the roots,
all the trees standing—a perfect “floating island.” Such
things happen, and the floating island is a not uncommon
phenomenon in the straits and: seas of the Malay Archipelago.
I know of a case, where, far out to sea, a submarine cable, on
being hauled to the surface, brought up with it banana trees
and portions of native houses, which had obviously gone to sea
as a floating island and had encountered shipwreck. 3

It is easily seen that such an island might serve for the
transport of a very large number of colonists. It is possible
even that it carried human occupants. Such islands could
not be expected to long survive the waves of the Southern
Ocean, and disaster would surely overtake them when they
had left the calmer landlocked seas of the Archipelago. But
some portion of them, crowded with the survivors of the wreck,
may float for long, and carry its living burden to new homes.

The trunks of trees are common on the seaward beaches
and each may have carried its passenger to the atoll from
some distant coast-line where its voyage was started. There
is one peculiarity about many of these trees which makes
them well adapted to the carrying of Nature’s colonists. This
peculiarity is the shape of the lower portions of their trunks.
-Most of the tropical trees, whose wood is not especially hard,

and which often grow to a great height, have very strong
Ay

290 CORAL AND ATOLLS

buttresses thrown out about the bottoms of their trunks.
These buttresses are in the form of large thin wings, which taper
to the trunk above, and below form a series of compartments,
_ like stalls in a circular stable. Within these stalls much earth
is held fast by the interlacing of smaller roots, and when such
a tree is uprooted, and set adrift to sea,it carries its earth with
it. It may carry it for very great distances, and I have seen
a “buttressed ” tree come ashore in the atoll, from whose base
a wheelbarrow-load of fine red earth might have been collected.
It must be remembered that such an incident is notable in a
place where there is no “earth,” save coral reduced to sand,
and decomposed vegetation in very small quantities. From
the roots of such a tree I have taken small stones, and these
are rarities in a land where no stones exist. It is certain that
such a tree would have many tenants when it started on its
voyage, and it is not unlikely that some would have the good
fortune to survive the passage. Its advent offers an explana-
tion for the presence of Typhlops braminus—a small snake with
an underground habit—in the atoll.

Many other sea-cast objects are stranded on the barrier.
The hard seeds of many plants are washed up on the beaches,
and these are potential colonists of the vegetable world. Gum -
resins float to the islands in large masses, and in the heat
of the sun they melt upon the barrier rocks. ‘Timbers of
ships, often with copper nails still held fast, tell the tale of
wrecks, and the ships that do not come to port. Pumice, in
all sizes and shapes, is for ever touring these seas, and in parts
of the islands it forms a considerable portion of the beaches ;
most of the pumice that comes to the atoll to-day is still the
remains of that great upheaval of 1883, when the eruption
of Krakatua altered the whole of the topography of the
Sunda Straits. This pumice has been touring the ocean for
over twenty years, and still, in the Sunda Straits, some set of
current will send whole masses to sea, and a ship will steam for
half an hour through the bobbing white balls of pumice which
are launched upon an indefinite, and an irresponsible, journey.

In the islands there is a pumice more ancient than this

AVG@-OL SUVAddV LI SY VOALVYVUM

TE
ae

TAX HLV Id

NATURAL COLONISATION METHODS 291

relic of Krakatua: it is not so clean or white, and it is found
farther inland from the present sea beaches. This is most
probably the pumice set adrift in the April of 1815, when
the unparalleled eruption of Tomboro—the great volcano of
Sumbawa—took place.

It is difficult to-day to form any conception of the vastness
of these eruptions, for tropical vegetation is so quick to ob-
literate the traces of destruction. Half a dozen times have I
sailed under the towering height of Krakatua, and failed to
see, in the jungle-clad coast-line of the Straits, the evidences
of the great tragedy that buried the town of Anjer and all its
inhabitants for ever. On July 9, 1883, Dr. H. O. Forbes
steamed past the mountain, and described its “ blazing crater.”
But little more than a month after, it had buried Anjer, slain
an incalculable number of people, swept vast coast-lines with
an all-devouring tidal wave, and altered its own appearance so
profoundly that a hundred fathoms of water marked the site
of part of a previously towering island. The crater has ceased
to blaze, and the wonderful luxuriance of tropical vegetation
has healed over all the dreadful destruction of twenty years
ago. Krakatua is now a slumbering giant that towers above a
peaceful stretch of water. Terrible as was the eruption of Kra-
katua, it was trivial compared with that of Tomboro, although
time has almost effaced that event from short-lived Eastern
memory. At Bangoewanghie, in the eastern end of Java, the
ashes covered all the land to a depth of eight inches, although
that beautiful Javanese town is two hundred and forty miles
distant from the volcano. For twelve days Tomboro was
active, and in that short time 12,000 persons were buried
beneath the burning ashes, and 200,000 more fell victims to
starvation and exposure.

The blocks of pumice set adrift by these eruptions have
been navigating the Hastern seas ever since ; they have visited
many shores in the course of their travels, and have consti-
tuted a mighty fleet of passenger vessels for the use of Nature’s
colonists.

A floating thing more strange than pumice is coral—coral

292 CORAL AND ATOLLS

in the form of large massive rocks which will float buoyantly
in the sea. This coral is a species of Astrea, the zooid of
which is a very large one, and nothing but its dead form is
ever seen in the islands, The secret of its presence is the
fact that it floats to the atoll, and is in reality an intruder
upon the island fauna. The reason of its floating when dead
is the presence of the numerous air-spaces, formed in its sub-
stance by the growing zooids depositing thin septa at the lower
limits of their bodies. Since the zooids are for ever growing
upward, they seal off cell after cell, and when they are dead,
their mausoleum is a great honeycombed mass of rock, which
leaves its old moorings—and after a long life spent in one
spot, sets out to sea. It is not likely that this floating coral
could be a very useful agent in the colonising of the islands
to which it floats, but it might easily increase the coral fauna,
for it is a regular thing in coral bionomics that a young colony
starts its growth upon an older one.

These floating objects have almost certainly been the means
of introducing a portion of the fauna of the atoll—and the
group of water-borne colonists, which are not dependent upon
man for their transit, is the one upon whose coming some light
may be thrown by an attempt to trace their source.

Two methods of investigation are open to the dweller on a
coral island. He may try to identify the objects which come
ashore, and so determine their place of origin; and he may
cast bottles into the water, and try to find out where they go
ashore and are picked up. In this way some idea of the
ocean drifts which pass the islands may be arrived at; we
may find out where they come from, and whither they go.

It might be thought that a determination of the species of
those plants, the seeds and trunks of which we find washed up,
would give the clue to their native land; but, unfortunately,
the information to be derived from this source is not very
definite. The distribution of the littoral plants of the tropics
is so very wide that even when a seed is definitely diagnosed
its exact place of origin cannot, in some cases, be limited even
to a continent. It is certain that many of them come from

NATURAL COLONISATION METHODS 293

Australia, and from the islands of the eastern end of the
Malayan chain; and most of the plants which have become
established in the atoll are represented in Timor, and in Java,
It is certain also that the pumice came from Java, and from
Sumbawa; for the advent of the Krakatua pumice is well
remembered after the eruption. The floating coral has, like
all its kind, a wide distribution, and does not help in the
determination of the direction of the sea-trend.

One piece of flotsam cast on the beaches of the southern
islands tells far more definitely of its origin. A boat came
ashore some years ago, which in its lines differed from any
that the men of Malay countries build. It was in very
perfect condition when it was first found, and the carving on
it was so well preserved that the Governor had no doubt at all
that it was a war canoe from New Zealand, a craft with which
he was familiar.

It is, therefore, safe to conclude that the currents which
bring the floating things to the atoll come from the east, and
such ocean drifts are, of course, charted. Through Torres
Straits the waters of the Pacific Ocean sweep into the Indian
Ocean, right to its western limits at the shores of Africa.
The waters that wash the northern shores of Australia and
the southern shores of the Malayan Islands are, therefore, the
carriers of the sea-drift which is likely to be cast up in Cocos-
Keeling atoll.

The track of the drift which leaves the islands is more
easily followed, and, in order to learn something of it, I set
adrift a number of bottles from the lee side of the atoll. In
order to make the experiment yield a true index of drift
action only, I loaded the bottles so that they floated with none
of their surface exposed, on which the winds could act; and
as they bobbed about in the waves, only a very small portion
of the neck came out of water. Within each bottle I puta
note, asking the finder if he would help me by writing to me
and telling me of his find. I enclosed a small coin to reward
him, and to cover his postage; I sealed the message in a test-
tube, put it in the bottle, and launched the whole aftair,

294 CORAL AND ATOLLS

Many bottles went to sea, being launched a few at a time,
during a period of two months. I started the experiment in
November 1905, and by June 1906, a reply had come to
hand. It was sent by Lieut. G. Piazza, the Resident of Brava,
Ttalian Somaliland, and it told that a bottle, launched on
November 15, 1905, had been found at a place called Ras
Day, at Brava (1° 06’ 8” N.; 44° 01’ 52” E.), on May 27,
1906. My little note came back again none the worse for its
sea travel of over three thousand miles. No other reply
came in for more than a year, and then, much to my surprise,
the Resident of Brava sent me another letter telling me that
another bottle had been found, in exactly the same spot as
the last, on July 11, 1907; he again enclosed my note, and
strangely enough it bore the same date—November 15, 1905
—as the first: I was glad to notice that my kind correspon-
dent had meanwhile ceased to be Lieutenant, and was now
Captain, Piazza. |

No other messages ever came back, and considering the
chances of disaster that are likely to overtake a floating
‘bottle—the chances of it breaking in its landing, and the
‘chances of its going ashore where it would not be noticed—
it is very wonderful that two ever did survive a passage of
almost the whole breadth of the Indian Ocean.

It is a very wonderful thing that, after so long a journey,
two of the bottles should have arrived at the very same point
on the coast of Africa; and I would dearly love to examine
the coast-line of Italian Somaliland, in the region of Ras Day,
‘to see if there was not some waif that had succeeded in making
a new home there, and that could tell clearly of a Malayan or
an Australian origin. I do not doubt that some such colonist
would certainly be found :—some typical Asiatic or Australian
intruder on the African fauna and flora. A sea immersion of
one hundred and ninety days would still leave seeds fertile
and ready to start growth in a new home, and it is possible
that some animal waifs might also survive the passage.

The drift that strikes the coast of Africa turns south, and
‘flows down the eastern shores of the continent ; and it would

NATURAL COLONISATION METHODS 295

appear that there is a southern return current, for a bottle,
launched in Mauritius, is said to have reached the atoll. But
I do not think that this is a possible route for colonists, for the
bottle almost certainly came to the atoll from the eastward,
after completing an enormous circuit of the Indian Ocean.
The length of this journey would almost certainly prohibit the
success of the route as a path for colonisation. Upon the
experiment of the drifting bottle, we may base some calcula-
tion as to the time which would be occupied by an animal
floating from the nearest land to the atoll ; and we may reckon
that, if the strength of the drifts to the east and to the west
of the atoll are at all comparable, a colonist must face a sea
journey of some forty days. During a stay of a fortnight in
October 1906, in a ship lying on cable ground some twenty
miles south of Sumbawa, the current flowed constantly
westward at the rate of about one and a half knots; and the
waters were carrying all sorts of drift wrack in their stream.
If this rate were to be maintained, the time of the journey
would be reduced to a half; but the near-shore currents are
subject to variations, for the region is one affected by the
monsoon. The atoll itself, however, lies without the mon-
soon area, and the steady westerly drift is apparently constant
throughout the year’s cycle.

Several creatures must have survived this perilous
journey of almost two months of clinging to some floating
thing. There are several instances of large snakes having been
killed in the atoll, and others have been picked up on the
beach, where the waves have cast their vessels. There are,
preserved in the Governor’s house, the skulls of two crocodiles,
shot after successful colonising: and these colonists must have
used this route. Of the permanent additions to the island
fauna we must probably reckon TZyphlops braminus, the one
earthworm, some of the coleoptera, and perhaps the millepedes
which live in rotting wood. Of the flora that has been assisted
by flotsam, the fungi are of most interest; and, as throwing
some light upon their means of dispersal, it is important to
note that every one of the eight species established in the

296 CORAL AND ATOLLS

atoll grows upon tree-trunks; not one of them grows upon
the ground.

Concerning those plants the seeds of which are hard, and
fitted for a sea journey, some very interesting points are to
be noted. On the island beaches are many kinds of seeds
which may be picked up any day—there are perhaps half a
dozen kinds of which one specimen could be found, almost
with certainty, in a walk along a hundred yards of beach.
Now these seeds will grow with great readiness when picked
up and planted in the earth; but as their leaves become
recognisable, you are at once astonished to see that they are of
a kind of which no representative may be found in all the atoll.

It is a strange chapter in Nature’s story, for these seeds
have been arriving at the atoll from time immemorial; they
have been cast on to its beaches by the waves, all ready
to grow, but unable to take root in the broken coral that
composes the beaches. A link in the chain is missing, for
there is no bird or beast that will move them the very short
distance to the resting-place where they could turn centuries
of failure into a successful colonising enterprise. The coco-
nut has become the dominant plant of.all atolls for the reason
that, where the ocean waves toss it, there will it grow; and
it sprouts green and fresh within the lap of the waves.
The smaller seeds have been aided in their journey by birds or
by the wind; but there is nothing that will give a helping
hand to the large “Queensland bean,’ and several others of
its kind, and assist it over the foot or two of coral beach
which cuts it off for ever from any hope of success.

The colonists that are water-borne, and that have made
the passage in ships, are again of two very different companies ;
for some, man has introduced deliberately into the islands, and
some, despite his very best intentions, have accompanied him
on his travels.

As a general rule, it may be said that those animals that
man has attempted to establish in the atoll have not thriven
exceedingly ;—whilst those that came with him against his
wish have multiplied to an unusual degree.

NATURAL COLONISATION METHODS 297

Whenever man takes a hand at assisting Nature in the
process of distribution of living forms, it is to be expected that
either his efforts will result in an inconspicuous failure or in
a conspicuous disaster. Here, the efforts have been mostly
failures.

He has turned down deer—the sambar and the kidang—
and at one time quite a numerous herd was preserved with
every care. The advent of a British man-o’-war, and the
efforts of a regrettable shooting party, ended in driving to sea
all those that were not shot—and the fate of those that took
to the barrier was no better than that of those which were
slain on land, for sharks accounted for all that the crew failed
to secure. From Singapore others were reintroduced, but the
last of the kidang passed away during my stay on the atoll.
In the same island with the deer, a number of rabbits were
liberated, but they failed to attain to the wonderful standard
of fecundity of their race, and their colony is a very struggling
one to-day.

Sheep have been imported, but sheep which, after a pro-
longed hunt, are shot with great difficulty, and then only yield
some twenty-eight pounds of tough mutton, are not reckoned
as a success, even by men who welcome anything that does not
come from tins. The only influence that the deer and the
sheep ever had in the atoll is that the former have created a
“browse limit” to the trees—a novel feature in a coral island
landscape; and that the latter have left an ever-growing
legacy of ticks—and the many mongrel dogs are their legatees.
It may be of some interest to the student of parasites to know
that these ticks, having lost their proper host, took so readily
to the dogs, and that when dogs are not at hand, they have no
respect for a far more dignified patron. It may also interest
him to know that there is an atoll distribution of parasites,
and that the dogs of one island have ticks, but are free from
fleas; while the dogs of another island do not possess ticks,
but harbour fleas innumerable ;—but these things are among
the least pleasant members of the atoll fauna.

Pigs thrive exceedingly upon coconuts, and—for pigs—

298 CORAL AND ATOLLS

lead a very clean life upon the dry coral islands:—but pigs
are not to be introduced lightly into Mohammedan commu-
nities; and though the Chinaman loves the animal in its every
phase, and tends with fond care the few that are now kept in
the islands, the pig is not destined to extend its domain beyond
the limits of its sty. Cats have been introduced, and, cat-like,
have become feral, and a nuisance. A somewhat strange in-
cident may be related about cats. On the island on which
I lived, no cat had ever been introduced ; the settlement
was some five years old, and no cat had ever belonged to
any of the settlers, so far as is known. A female kitten was
brought to the island as a pet, and as she grew up, suspicions
concerning her seemed obviously justified, though apparently
absurd. The birth of a healthy batch of kittens put all
doubts to rest. Their prevailing colour was black, and a
great black tom cat ultimately fell to a gun at dead of night,
but if he had not called to see his family, he might still
be living on that island unknown and happy. This was a
lesson to me, for the island was less than a mile long, and
about two hundred yards wide at its broadest part; [had been
hunting for the fauna of that tiny strip of land for over a
year—and twenty-eight men lived on the island—and yet
not I, nor any other person, had the least idea that a decent-
sized mammal existed on the little stretch of coral land. A
monkey and a lemur, which were lost on the island during my
stay, also managed to disappear utterly. No one ever saw the
lemur again, and, being nocturnal, that is not so strange; but
though every man’s hand was against the monkey, for he stole
our chickens, he was only sighted once in the course of very
many months. These facts show how very easy it is to live in
the tiniest of restricted areas, and yet, with every interest in
your surroundings, to fail to see even fair-sized animals that
live around you. While these things are so, we have no
necessity to create new species on the strength of their having
white whiskers instead of the usual grey.

In some extraordinary fashion, a monkey once effected his
own passage to the islands; the manner of his coming is a

PLATE XVII.

NEST OF MERULA ERY THROPLEURA, THE CHRISTMAS ISLAND
GROUND-THRUSH.

NATURAL COLONISATION METHODS 299

mystery, and he was not seen again after a cyclone had swept
over the group. His coming was mysterious, his going was
mysterious, yet his visit is well authenticated, and it was one
which lasted many years.

With the introduction of birds, man has been much more
successful, and some of his protégés are among the most
thriving of all the atoll fauna. Three species have multiplied
greatly, and have become quite permanently settled in the
islands; one has a somewhat insecure foothold; and one is by
now practically extinct as far as this atoll is concerned. Un-
fortunately it is the most valuable of all the birds that man
has attempted to bring to a new home which has now practically
ceased to exist—and this is the Pergém (Carpophaga Whartont),
the fine fruit pigeon of Christmas Island.

From Christmas Island also came the little “love-bird,”
Burung chinta, and though it exists as numerous little com-
panies upon one island, it has never spread beyond the limits
_of that island, so that it is easy to live in the atoll and to be
unaware of the existence of Zoesterops natalis. It is a wonder-
ful thing that, after colonising Christmas Island in the distant
past, it has never had the adventure of crossing the mile and
a half of sea that separates its home in the atoll from the
other islands where suitable food exists in plenty.

The Christmas Island Ground Thrush (Merula erythro-
pleura) has, however, shown far more enterprise, and has
made itself at home on every island of the atoll ring—but
always preferring those in which the vegetation is most
varied. It is one of the very tamest and most trusting of
birds ; it will come quite freely into a bungalow when it wants
a drink of water, and it seems to have no fear of man what-
ever. Asit has a song which is reminiscent of the blackbird’s
and.a familiar and attractive way in all its doings, it is a
very welcome addition to the live-stock of the atoll.

From Singapore came the familiar Java sparrow—tho
Burung glatek—and he, having all the sturdy insolence of
his more homely relations, has become, I should suppose, the
most numerous bird of the southern islands. He, too, is a

300 CORAL AND ATOLLS

distinct gain to the island community, for he is a gay fellow
in appearance, and not unattractive in his habits. In Forbes’
account of the atoll, it is said that the Jungle Fowl has been
established in the islands; but all the wild fowls to be seen
during my stay were domestic fowls run wild. In North
Keeling atollon are great numbers of these, and it is a strange
thing that of the myriad birds in that vast sea-rookery,
the domestic fowls are by far the wildest. In this little
island sea-birds breed in thousands; frigate birds of two
Species, gannets and boobies of three species, and terns of
three species, crowd all over the trees and bushes and litter
their eggs about the ground. All these birds have to be
practically pushed from their eggs, and they simply do not
know man, or regard him as in any way a peculiar product
of Nature. During my visit to this atollon, I met with no
further regard for my presence on the part of the birds than
the very solemn vomiting of large quantities of fish by the big
Abbott’s booby when it was pushed from its solitary egg.
There is something rather sad about these birds. Their big
black masks give them a pathetic appearance, and when they
have vomited their fish, and their egg has been taken from
them, they walk away with an air of dejected resignation, as
though the affair was no longer of interest to them.

The wonderful tameness of the birds on lonely islands is
sometimes difficult to credit, and the charge of exaggeration
is easy to make—yet I have myself seen the great frigates—
birds nine feet across the wing—picked up from the ground
without offering any sort of resistance. These birds were
not gorged with food, they were not sleeping, and they were
not unwell ;—they simply did not mind us walking about
among them. Yet you are lucky if you may get within thirty
yards of an old cock rooster who shows Plymouth Rock, and
Dorking, and many other familiar strains in his varied
plumage. He is in the company of three or four hens of
the same variegated hues, and in thorough farmyard style he
scratches over the turtle-grass which lies along the lagoon
shore; but you have no sooner sighted him than he is away

PLATE XVIII.

Le

THE SMALLER FRIGATE-BIRD
(FREGATA ARIEL).

NATURAL COLONISATION METHODS 301

with his hens, over the tops of the coconut palms, or through
the dense undergrowth.

Migrant birds also come to this atollon. Golden plover
walk in the shallow lagoon flats, but even the air_of peace which
the place holds, and the general trust displayed by the resi-
dent birds, cannot quite make them forget the fears which they
have learned of man in their travels—and they give him a
wide berth. It is not flattering to mankind to realise that
every animal that really knows him always takes care to
avoid him; parasites form the sole exception.

CHAPTER XXVI
SHIP-BORNE AND WIND-BORNE COLONISTS

THE creatures that have come in ships, as hangers-on to
human enterprise, are perhaps the largest contribution to
the islands’ colonists ;—they are certainly by far the most
economically important.

It is not right, always, to regard these creatures as
passengers; for as a rule they are residents on the ship, and
they came ashore, maybe, quite against their will. It is not
that a cockroach joins a ship in Singapore, and steps ashore
in the atoll; for the ship is his home, and he was born there,
and probably he only left his home by being passed ashore in
a packing-case. When a ship has traded for long in Far
Eastern waters she has a very considerable fauna of her own,
a fauna almost as rich as that of a coral island itself; and, in
the case of some ships, it is a question as to which would gain
by the exchange, if she were moored alongside a coral island
for any length of time. Rats and mice; cockroaches, of
perhaps two or three species; ants in varied armies; a scorpion
or two; some centipedes, and a host of spiders are the dwellers
on practically every ship that floats in Far Eastern waters.
Even there is such a thing as distribution on a ship, and in
the ship that I came to know best, the cockroaches and the
ants had very fairly defined areas of distribution, beyond which
they did not greatly transgress. Some creatures, which are
not permanent residents upon the ship, make their passage in
different forms of merchandise: many beetles, some moths,
and a variety of crickets are passengers of this class ;—there
is one species of cricket that appears to be the invariable
accompanier of every straw-stuffed soda-water crate that lands.

By far the most important items in the atoll fauna are
302

SHIP- AND WIND-BORNE COLONISTS 303

the rats, and they undoubtedly came through human agency.
The rats have this distinction, that they were firmly established
in the atoll before the original settlement was made by man,
some eighty years ago. Hxactly how the original rats came is
of course unknown, but people had occasionally landed in the
atoll previous to 1827, and, judging from the after-history,
many unfortunates had been wrecked upon its barrier. When
the settlers arrived, they found these rats confined to one
island, and this island they named Pulu Tikus, or Rat Island,
in consequence. It was upon this island that the original rats
landed from the wreck, or were brought by man, and there
they stayed ; for the channel that separates Pulu Tikus from
its neighbour is deep and very swiftly running. A man may
_ swim the channel at all tides, but he will have to make a long
course, for the current carries him rapidly down; and a rat
would surely go to sea in any attempt at making the passage.

After the advent of the settlers, the rats of Pulu Tikus
remained the sole representatives of their family in the atoll
for very many years. They were not permitted to extend
their domain beyond the limits of their own island, for a strict
examination was made of all the boats that went between the
different islands. They were not exterminated, for the simple
reason that the extermination of rats, by any ordinary means,
even on so. tiny a strip of land, is quite impossible.

Theoretical authorities on tropical public health are apt
to. talk somewhat lightly about exterminating two disease-
carrying pests—the rat and the mosquito. It is repeatedly
stated that wonderful good would come of its entire accom-
plishment, as though it were a practical possibility ;—and ina
great many cases very wonderful results have been attained.
If any one has a fancy that he can exterminate rats or mos-
quitoes, he might fittingly try his hand in this atoll, for it is
an ideal spot for its attempting.

Mosquitoes he will never get rid of, for they pass their
larval stages quite comfortably in salt water, and though he
may render all his fresh water nasty with petroleum, or copper
sulphate, or anything else that he chooses—he cannot hope to

304 CORAL AND ATOLLS

poison the Indian Ocean, or even that small portion of it that
lies within the lagoon.

Rats may be killed in thousands—this has been done for
years and years, and special gangs of the natives have no
other employment than rat-catching. Rats may be trapped
in astonishing numbers, they may be hunted with dogs, preyed
on by cats, shot, poisoned, and routed out, but they will not be
exterminated. Trapping merely consists of a wearying repeti-
tion of baiting traps overnight and emptying them in the
morning, dogs and cats become lazy, and prefer to hunt the
land-rail ; and, at the best, it is an altogether exceptional dog
that will organise a rat-hunt on his own account, and in my
experience a dog that can at any time catch a rat by walking
a few paces soon tires of the occupation. Shooting becomes
expensive, and is but a feeble method. Poison leads to
disastrous suicide in drinking-water tanks, to untimely deaths
amidst provisions, to horrid death scenes in your. bathroom,
to beastly sounds in your bedroom by night, and to deaths of
other and more valued animals, and is not to be recommended.
Immense numbers may be killed by turning over the piles of
coconut husks, and murdering all that bolt—but next day
the pile will have its tenants again. Danysz virus is quite
inert by the time that it can reach the atoll, the journey is too
long, and the temperatures of the voyage too varied ;—and
those rats that were fed on it, and inoculated with it, whilst
kept in cages, only exhibited a more than normal fecundity.

If other forms of virus will succeed any better I do not
know; but the experiment is being made.

In 1878 the islands received a fresh accession of rats.

The ship Robert Portner was wrecked in that year, and
from this wreck the atoll derived the greatest curse that has
ever fallen to its lot. Cyclones have visited the islands from
time to time, and have done. incalculable damage; but the
steady depredations of the rats cause far more loss than the
occasional visits of cyclones; their numbers are legion, and
their work is continuous.

The Robert Pourtner rats settled on most of the islands

SHIP- AND WIND-BORNE COLONISTS 305

to the south of Pulu Tikus, but on Pulu Tikus itself they never
gained a footing, and the “ original” rats of that island remain
quite distinct from the new-comers. The bad habits of the
invaders rapidly became notorious, and, unlike the “ original ”
rat, which did very little damage, they soon became the source
of the most real economical loss to the islands.

The rats of Pulu Tikus have always been in the habit of
climbing the coconut palms, and in some cases their offspring
are reared in the crown of a palm some seventy feet from the
ground ; and they have learned to open the ripe nuts which
have dropped from the palms—for coconut constitutes practi-
cally the whole of their dietary.

But the Robert Portner rats were not content with this,
and when they had become accustomed to climbing the palms,
they learned to attack the nuts whilst they were still green.
To-day, thousands of nuts must be ruined daily by being
nibbled, just where the stalk joins the husk, and a nut that is
attacked in this way falls from the palm, and is quite useless
for any purpose. It seems a strange thing that these new rats
should so quickly have learned this mischievous practice, when
the members of the older coleny still remain contented with
the nuts that fall naturally.

On one of the islands of the atoll ring the whole colony of
rats consists of black rats, and stragglers of this species are
found in some of the other islands; but the Robert Portner
rat is the common brown rat, and the Pulu Tikus colony is
probably composed of a variety of the same species.

Mice have become fairly abundant in some islands, and
they do a considerable amount of damage to the stores, but,
although they have no great struggle against enemies, they
-do not appear to increase much, and their importance, when
compared with that of the rats, is practically nil.

The next most conspicuous additions to the fauna made
accidentally by ships and merchandise, are the centipedes and
scorpions.

The great centipedes have become exceedingly numerous,

and they have spread to all the islands; and the scorpions exist
U

306 CORAL AND ATOLLS

in very fair numbers. The wounds inflicted by both these
creatures are painful, and a good deal dreaded, but I do not
think that they are ever fatal, and all the cases that I have
seen have done very well. Both are common species which
are widely distributed in the topics, and are the invariable
accompaniers of man’s enterprise in ships.

Spiders of several varieties have become abundant, and
although some are of obscure species which are at present
unnamed, it is likely that they are all of a wide tropical dis-
tribution. The larger species are well known as being the
hangers-on to man in all his wanderings, and the big Hetero-
poda venatoria has even made attempts to colonise England.
Probably nearly all came to the atoll in ships, but without
ship intercourse spiders very easily colonise distant pieces of
-land, and the spider fauna of the neighbouring Christmas
Island contained, ten years ago, none of the common ship
species, but a variety of kinds peculiar to the island.

The cockroach is another of the creatures which have
almost certainly taken advantage of commerce to effect their
settlement, and its success has been very great. No less than
seven species have become well established in the atoll, and
most of them exist in very large numbers. Crickets certainly
come in packing-cases, and are always to be found in numbers
after a ship has discharged her cargo of stores, but they make
themselves quite at home on the atoll, and may be found all
over the islands.

Ants are found on every ship, and their spread is easy.
Their numbers are very great in individuals, but by no means
an astonishing variety of species has made its home in the
atoll. Ants are among the most useful of tropical insects, and
their introduction is in no way to be regretted; their thread-
like journeyings are always a source of quiet entertainment,
and as scavengers they cannot be overrated. It is difficult to
say which of the sixteen species of beetles have come by
accidental tree-trunks, and which have accompanied man;
but the widespread Dermestes felinus is a follower of trade,
and the Copra beetle — Corynetes rufipes—owes its truly

SHIP- AND WIND-BORNE COLONISTS 307

~ marvellous abundance to the copra industry. Others have
probably used this route, but beetles have a far more varied
manner of colonising than is open to many orders of insects,

It is impossible to say if the two species of Geckos came
by ships, but the Gecko is no lover of ship life, and as the egg
is laid securely beneath the bark of a tree, and two months
may elapse before its hatching, it is obvious that the more
independent route is open to it. Both the species that have
made their home in the islands are widely distributed in
tropical countries.

It may be said, on the whole, that the creatures which have
accompanied man in his wanderings, and have made their
landing unbidden, are a very undesirable set. The rats and
the copra beetles are the most harmful of all the atoll’s fauna ;
and the centipede and the scorpion are the most unpleasant.
The highly desirable animals which he would have to live
around him have ceased to thrive, and the very undesirable
ruffians which have taken advantage of his enterprise have
thriven exceedingly.

The colonists that have come to the atoll by flying through
the air compose by far the most numerous group of all the
island fauna; and they are a class of the very greatest
interest.

So many of the difficulties which lie in the path of these
wind-borne creatures are commonly forgotten, or not ap-
preciated, that their presence upon the atoll often does not
create the admiration which it deserves. Two cases may be
told of wind-blown waifs in the Cocos-Keeling Islands that will
serve to make real some of the problems surrounding this
method of colonisation.

At certain times in the year—generally after the rare
north winds—the whole atoll will be swarming with dragon-
flies: the commonest species of them all is the widely dis-
tributed Pantula flavescens, but Tramea Rosenbergii and Anax
_guttatus are mostly to be found on such occasions. They come
to the islands in thousands. They hawk for flies over the
surface of the lagoon, and are to be met with sailing up and

308 CORAL AND ATOLLS

down every quiet glade of coconut palms, with all the accus-
tomed air of settled residents. Yet of all these thousands not
one is born or bred in the islands, and not one is ever able to
start a permanent race of settlers :—for there is no open fresh
water anywhere in which they may pass their larval stages,

They are quite alive to their responsibilities as pioneers,
and they do not merely spend their time in a butterfly fashion
during their somewhat short life of sunshine ;—on all sides
they may be seen making every attempt to carry on their
race. When rain falls heavily, and coliects in puddles on the
asphalt tennis court of the cable-station, the puddles are at
once exploited as likely places for laying eggs, and numerous
females will be seen busily preparing for a new generation.
But half an hour of sunshine and all their efforts are vain, for
the puddles dry up as quickly as they came.

Even salt water tempts them to lay their eggs, and the
females of 7. Rosenbergii were to be seen daily depositing
useless eggs in the salt-water fish-pond on Pulu Tikus.

This case gives some idea of the colossal waste of Nature’s
methods of colonising. Considering the fact that the atoll is
the only dry land spot in all this part of the Indian Ocean,
and that it is only eight miles across, some notion may be
gained of the extraordinary numbers of these insects which are
journeying aimlessly to a certain death. Of all the thousands
that do happen to arrive in the atoll, not one can ever hope
to carry on its race—and their fate is as useless, and their
journey as purposeless, as that of those myriads which drift to
nothing save a limitless sea.

The case is a very interesting one, for it is usually assumed
that all colonists that set off from their home, and arrive in
distant spots, are blown by winds from their native shores ;—
but this is not the case. These dragonflies go to sea of their
own accord, they are not blown from shore, and in absolute
calms of long duration, Pantula flavescens may be seen hawking
about over the sea twenty miles and more from the nearest
land. I have seen numbers of this species almost daily in a
journey through dead calms all the way from Singapore to

SHIP- AND WIND-BORNE COLONISTS — 309.

Thursday Island. I have seen them too at almost every point
between Java Head and Cocos-Keeling. They are merely
flying over the sea instead of flying over the land, and they
are certainly not in any way to be regarded as unfortunate
storm-driven exiles. Butterflies that are members of strongly
flying families do the same; they fly away from the land in a
perfectly irresponsible way. Moths come nightly to a ship's
lights when she is lying twenty miles from the shore.

I do not for one moment suggest that any of these insects
are voluntary colonists, but I am quite sure that it is wrong
to regard all insects that are met with when far from land as
being wind-borne strays, swept from their homes by gales.
Many of them start the journey of their own free will—an
aimless and irresponsible journey it is true, but not by any
means invariably an involuntary one. When once at sea,
without shelter or resting-place, they do become true waits,
and then the wind disposes of them as it will.

The case of these dragonflies, which come in such numbers
and are so conspicuous; is certainly only an easily recognised
example of what is happening perpetually to an infinity of
other creatures, and it makes real the most important fact in
Nature’s colonising—that the number of her failures must be
enormous.

In the first place, either both sexes must arrive at the
same time, or the one be ready waiting ashore for the other ;
or else a pregnant female must come prepared to undertake
the responsibilities of maternity in a new land. Vast numbers of
waifs must arrive in vain—stray solitary individuals, unmated
females, or members of one sex only. Then every creature
that has, after a perilous journey, the good fortune to effect a
landing on the atoll, still has its fate undecided, for no suitable
environment may be at hand for it and its progeny. ‘The
dragonflies have no fresh water, and therefore cannot pro-
pagate their kind. Countless other waifs, when they had
made a successful landing, all ready to carry on the reproduction
of their species, must have sought in vain for any plant which
would serve as the food for their larve, or for any nidus which

310 CORAL AND ATOLLS

would have been suitable for their earlier stages. The failures
from this cause must be beyond estimate.

The case of the dragonflies gives one aspect of the picture
of Nature’s methods, her absolute prodigality, her countless
failures, and her implicit reliance upon chance. But there is
another aspect, and this is the absolute success of her methods.
My other case illustrates this.

Some while before I left the atoll we introduced tomatoes
into our garden, and the tomatoes came as seeds in packets
from England. With the first crop of plants a swarm of. the
larvee of a moth, new to the atoll fauna, began to devour the
tomato leaves. No examples of this insect had been seen
before, and yet the suitable food plant, within a very short
time of its appearing above the ground, had received its proper
tenants. There can be no question that the insect was
certainly not introduced as eggs among the seeds, for, apart
from the general improbability of such an occurrence, it is a
species which is not found in England, although it is widely
distributed over the world. With peas the same story was
repeated, and the moth made abundant by their introduction
is again a very widely distributed one.

The history of the tomato and the pea is probably one of
universal application, for it is almost true to say that of all
the vegetable waifs which have arrived in the islands, every one
has been appropriated by the insect whose particular choice it
happens to be.

We have seen, in the case of the “ Queensland bean,’ how
uncertain are Nature’s ways of vegetable colonising, and we
have seen, in the case of the dragonflies, the enormous
chances against successful insect invasion ;—it then becomes a
matter of very real wonder that almost every chance vegetable
colonist has received, by chance, its appropriate animal client.

The question of finding a suitable food plant does not
affect all insects equally, for some are far easier to accommodate
than others. The larve of such moths as Prodenia littoralis
will eat almost any low-growing plant, and will make them-
selves at home practically wherever vegetation of any kind

SHIP- AND WIND-BORNE COLONISTS 311

exists. Such insects are the most successful of all the pioneers
of the animal world, and exist in large numbers on this isolated
spot.

But some are by no means so easily satisfied, and a
particular vegetable species—or at least a particular genus—.
is absolutely. essential for the nourishment of their larvee.

These facts give the problem a somewhat different aspect
from that which it first wears when we casually remark the
colonists and think only of the distance that they have come,
and the hardihood of their journey. Those creatures that are
settled and established are the elect, and they are appointed
out of a countless host of competitors, all of whom have had
equal adventure but have gone under in the struggle, through
no fault of their own. They are the actual colonists, the
survivors of a vast army of immigrants, every one of which
Was a potential colonist.

It is impossible to conceive the numbers of the Tost “hme
perpetually flying about over the ocean, looking for some
chance resting-place ; it is impossible to conceive the per-
petual waste going on among the flying things that have lost
touch with land. These are the waifs that come aboard ships
when far cut at sea, and, considering the chances against an
insect boarding a ship in mid-ocean, it is very astonishing how
many may be picked up by any one who cares to watch the
ship lights by night, and keep his eyes open by day.

From Suez to Penang is an unbroken steamship run of
some 15 days, for many Far Eastern boats do not touch at
Colombo ; and this run gives a very favourable opportunity for
collecting any ocean strays.

Upon this tract of ocean, in January 1905, eleven insects
came aboard the ship on which I travelled. These eleven
insects represented nine different species of five different
orders. ‘There were seven specimens of five species of Lepi-
doptera, one specimen represented the Diptera, one the
Coleoptera, one the Orthoptera, and one the Homoptera, Out
of the total of eleven insects, seven came aboard at night, and
were found flying round the ship’s lights. ~

312 CORAL AND ATOLLS

An extended series of such observations would enor-
mously amplify this list,and it would give us some idea of the
resources of Nature in her distribution of potential colonists.

Of the orders which have probably arrived by flying to the
atoll Lepidoptera is by far the best represented, and no less
than thirty-three species of butterflies and moths have made
their home in the islands. Three species of winged Hymenop-
tera, and eleven of Diptera, have also become settled, and
probably most were wind-borne.

The four species of Neuroptera, and several of the Orthop-
tera have also followed the same route.

Besides the insect fauna, other and higher orders of
animals have come to the atoll by flying.

Of the mammals only the bats may use this route, and
the Pipistrell peculiar to Christmas Island has found its way
to the Cocos-Keeling atoll, but it has never become an estab-
lished resident. Forbes has reported that the large fruit-
eating bats have been blown to the atoll—also probably
from Christmas Island—but no memory of their occurrence
remains, so far as I could ascertain. It is quite likely
that these large creatures should come as colonists, for
they commonly make journeys to sea among the Malayan
Islands, and across the Sunda and the Bali Straits they flap
every evening of their lives as they pass to their feeding-
grounds, It is strange that,if these bats have ever come into
the islands in any numbers, they have not beccme settled ;
but the supply of wild fruit is perhaps too limited, for the
Papaia is practically the only fruit that grows outside the
limits of the gardens, and its supply is very precarious, and
its season is very short. Yet it would not be surprising to
learn that these powerful fliers cross the sea from Christmas
Island, or from Java, and live upon the bananas and saohs
cultivated in the gardens of Cocos.

There is practically only one resident species of land
bird that has not been introduced by man, and this is the
Philippine land-rail, whose apparently feeble powers of flight
make its advent all the more remarkable. It is difficult even

“TIVE GNVT HUNIddVIIAd AL AO SOOH GNV LSHN WAM

“XIX GLY Td

SHIP- AND WIND-BORNE COLONISTS 313

to make this bird rise from the ground, and, unless it is hard
pressed, it never attempts to fly; yet, after the fashion of its
family, it has become a pioneer.

A widely distributed species of heron, Demiegretta sacra,
breeds in fair numbers in the islands, and a few wind-blown
migrants, swallows, wagtails, snipe and plover, occasionally
land, always in very poor condition, The bulk of the avi-
fauna of the islands of course consists of the sea-birds pure
and simple ;—the frigates, boobies, bo’swain birds and terns:
and it is remarkable that the three species of terns, which
breed in such numbers on the atoll, are not represented in the
Christmas Island fauna.

Since the atoll lies in the south-east Trade zone, and the
wind blows from the S.E., E.S.E., or E. for three hundred
days in every year, it might be thought that the place of origin
of the greater part of the fauna of the atoll would be a very
easy thing to determine. It would be presumed at once that
Australia was the fatherland of the bulk of all the living things
of the islands—and very many of the forms are Australian.
But it is a well-known fact, to which I believe Wallace first
called attention, that it is the abnormal rather than the normal
winds which are the most important in bringing waits to dis-
tant tracts of land. The dragonflies show the truth of this,
for their advent certainly follows the rare northerly and north-
easterly winds.

The terrestrial fauna is therefore not entirely Australian,
or even Austro-Malayan, for it contains a very considerable
admixture of Indo-Malayan forms, just as does that of its
neighbour Christmas Island.

CHAPTER XXVII
THE LITTORAL AND MARINE FAUNA!

THERE is another group of settlers in the atoll which has
neither come by flying nor by chance flotsam in the ocean drifts,
and this is the large army of crabs that has invaded the
land so successfully. The crabs of the atoll show a series of
transitions between those forms that live entirely in the waters
around the group and those that, living always on the land,
seek, so far as I have observed, those parts of the islands
farthest removed from their ancestral home. |

Around the sandy beaches of the lagoon live several species
of the genus Ocypoda, and they are interesting and rather
beautiful crabs of .a pale olive-grey colour, and of excessively
active habits. They run sideways on the extremities of their
long legs, and their two wonderful eyes are held above them
as watch-towers, for ever turning this way and that as danger
threatens. These crabs live in spiral burrows along the sandy
margins of the lagoon, and digging one out of his burrow
is a rather difficult proceeding, for the holes are deep and
winding. Concerning the spiral nature of the burrows a
point of interest arises, for is it always possible to tell from
without which way the spiral is wound within. At the
mouth of each burrow is a little shoot of sand, cast out by the
crab in his excavations; at times the sand-heap is to the
right, and at times to the left of the hole, and the tailing out
of the shoot leads to the hole in a curve which commences the
spiral of the whole boring.

There is no mere chance concerning the winding of these
tunnels, for the right-handedness cr left-handedness of their
spiral depends upon which of the chele of the individual is

the most developed. If the crabs are watched, it will be
, B14

THE LITTORAL AND MARINE FAUNA 3815

noticed that some carry a large right claw in front of their
bodies—as a boxer waiting to parry a blow—while others carry
a well-developed left claw instead. These holes are excavated
anew with every fall of tide, and so the crabs are always
busy ; indeed, their lives are passed in surprising activity, and
the rapidity with which one will decide whether to run for
the water or for the shelter of his hole, when danger threatens
him, shows a most extraordinary aptitude for taking in a
situation at a glance. Nothing escapes the watch-tower eyes,
and their range of vision appears to be’particularly long, for
they are aware of the presence of a man at a surprisingly
great distance; and when one has found safety in his hole, he
soon learns when danger has passed, for the two great eyes
slowly protrude from the hole and take an observation of
everything going on outside.

The crabs that have made the next step towards the land
are the quaint little “fiddlers” (native name, Kapeting delima)
of the genus G'elasmus, tiny greenish creatures with one great
pink claw as large as all the rest of the animal. The pink
claw is carried in front of the body in a most grotesque
fashion, and is practically all that is seen of the animal as one
walks over the muddy flats in which they live, for they are
very shy, and they rapidly disappear into their holes as
an intruder approaches. The melting from sight of the
great army of pink-clawed “fiddlers” is very curious, for, at
about the same distance all round, these quaint creatures
disappear, and one becomes the centre of a moving circle
whose fringe is made up of vanishing pink claws and tiny
greenish bodies. In the same mud flats—and practically all
over some of the islands—are the large burrows of the
Kapeting balong (Cardisoma hirtipes), a large and powerful
crab whose back is dull purple and whose legs are pale yellow.
This crab is a very important item in the economy of the
islands, for it has a habit of burying the husks of the coconut,
and so helping to produce the surface layer of vegetable mould
which clothes the more densely vegetated islets.

It has not completely cut itself off from the sea, and

316 CORAL AND ATOLLS

large numbers may be seen at certain times marching in armies
along the lagoon shores, some wholly in the water and some
in the wash; and they are there doubtless for the purpose of
depositing their ova.

Its burrows, too, are always made down through the loose
débris to the water-level, so that the crab when residing even
in the centre of an island is always able to get into the water
at the bottom of its burrow when it wishes to moisten its
gills.

The other land crabs of the order Brachywra, such as the
handsome red Gecarcoidea lalandci, and the members of the
genus Geograpsus, live for the most part beneath the loose coral
boulders and the roots of trees, and Gecarcoidea lalandet seems
to have cut itself off from the water very completely at all

times of the year save that in which it goes to the beach for

the purpose of depositing its ova.

It is the members of the order Anomura that have most
successfully become land animals, and the great “ robber crab ”
(Birgus latro), whose native name is Udung darat, is the most
completely terrestrial of them all.

The “robber crab” is becoming rare in the Cocos atoll,
and I only succeeded in getting possession of five in the fifteen
months I lived in Pulu Tikus: but in Keeling atollon it still
lives in abundance. It is a very remarkable creature in every
way, and no notion of its uncanny cleverness would be
imagined from a mere inspection of the animal; but the
Birgus is one of the cleverest animals living. Its strength is
quite incredible. It can nip through wire netting as easily as
can a man with cutting pliers. It can tear up tin with ease,
and break with its great pinchers the wooden bars of a cage
that would serve to imprison a large wild animal. Added to
this, it possesses an extraordinary quickness of movement, so
that the sudden and well-aimed grabs that it will make are
far more reminiscent of the actions of a great spider than of a
crab.

So far as I observed, in regard to those that I kept in
captivity, the Birgus is far more fond of dead rats or meat

gs

THE LITTORAL AND MARINE FAUNA 317

than of coconut in any form, and it was difficult to persuade
them to eat pieces of coconut when there was any animal food
to be had.

I have never seen the Birgus crab go near the water,
nor did those that I had in my possession show any desire to

Fic. 69.

Tue “ ROBBER CRAB.”

(Birgus latro.)

get into the water that was provided for them; in fact, the
creature seems to prefer the driest parts of the islands for its
home. The crab has a wonderful power of climbing, and
can clamber up anything that affords the least foothold. It
climbs trees with ease, and it could escape from any cage that
I ever tried, except an old oil-drum, the metal sides of which
afforded no foothold for its claws.

This power of climbing trees it shares with the hermit
crabs of the genus Cwnobita (native name, Umpan), of which
two species—the one bright red, and the other dull blue—
swarm everywhere in the atoll. These too are interesting
creatures, and the way they will march and march—ever
leaving behind them a curious sand-print of their trailing shell

318 CORAL AND ATOLLS

—making an ever-increasing army, the nearer they get to
some piece of sea-tossed carrion, is something to watch on
tedious coral island days. Very often in the early morning
an army will be seen on the beaches, drawn up in a circle, all
beginning to turn homewards from the feast upon the body of
some fish that has been cast ashore, for a dead shark will
collect around it half the hermit crab population of the island.
None of these crabs is used for food except the Kapeting
balong and the Birgus, and at the present time so few of this
latter species are to be found that its dietetic importance has
ceased. The hermit crabs are, however, extensively used for
bait, and the big Kapeting balong, put whole upon the hook,
makes the best and most attractive bait for the green-fish,

One crab, named Kapeting trelek, a member of the genus
Leptograpsus, is very good eating indeed, and, as it is an
abundant species upon the seaward side of the islands, it is to’
be ranked among the useful animals of the atoll.

In the calm waters of the lagoon, and in the rock pools of
the barrier, are a myriad forms of crabs, strange and grotesque
in shape and often of very beautiful colours, each marvellously
fitted to the environment in which it finds itself. In every
colony of coral is some tiny crab, always adapted most wonder-
fully to its place of residence, and often it is almost impossible
to distinguish these coral crabs from the branches of the type
of coral upon which they dwell. Many of them bore holes in
the coral rock, and these are very difficult to see, for they are
small and coloured like the surface of the rock which they
inhabit, and one is generally not aware of their presence until
they have withdrawn into their holes. Some that live in the
soft lagoon sand, as Calappa hepatica, are strangely like a
fragment of coral conglomerate, for, as they lie in the ooze,
their legs are pressed firmly against their shell and not a
joint in their whole body is to be seen.

The whole crab fauna of the lagoon and barrier is too
numerous to discuss in detail, and the names of only a very
few of the army are recorded in the appendix to this chapter ;
but one rock-pool crab deserves an especial notice. This

THE LITTORAL AND MARINE FAUNA 319

crab is a minute creature and is very similar to Melia
tessellata, described by Mr. Borradaile as living in the Male
atoll and elsewhere. The crab is rather conspicuously marked,
but it has adopted the curious habit of carrying a bouquet of
green alge in its claws, and behind this bouquet it hides its
boldly need body. In

the cases which I have Fre. 70.

examined the alga had
actually taken root upon
the claw, for it had been
held so long in position ;
but there seems to be no
doubt that in the first
place the crab plucks
the alga and carries it
in front of its body as
a kind of stalking-horse,

and as a shelter from its Sanu Crap THAT HOLDS IN ITs CLAWws
paeticg TWO SMALL Bovaurts or Livinc ALG,

It must not be for- (Enlarged to twice natural size.)
gotten that the creatures
that live in the waters of the lagoon, no less than those that
have invaded the land, are properly a part of the peculiar
fauna of the group. For those animals which spend their
_lives in the shallow waters of the coral beds there is no
feeding-ground nearer than the coast of Java, or the shores
of Christmas Island, five hundred miles away. The greater
part of all the fish about the atoll in all probability spend the
whole of their lives in its immediate vicinity, and the many
beautiful little coral-haunting species ae the lagoon belong
peculiarly to the local fauna.

The fish are necessarily among the most economically
important animals of all the fauna of coral atolls, for they
furnish an abundant supply of fresh food in these places in
which it is otherwise very scarce. It is fortunate that of all
the quaintly shaped and gaudily coloured atoll fish, there is
only one which is poisonous when used for food, and even that

320 CORAL AND ATOLLS

one may be safely eaten when properly prepared. All the fish
which live in the warm waters about the islands are not, how-
ever, either useful for food or at all desirable from any point
of view ; and chief among the undesirables are the sharks.
Sharks exist in great numbers ; a large brown species, which

Gee

A Mrxep CatcH or Cocos-KEELING FISH.

grows 10 or 12 feet long, inhabits the lagoon; a smaller black
and white species comes to the barrier in armies at certain
times of the year; and a yellow “ dog-fish,” some 4 or 5 feet
in length, lives everywhere. It is a strange fact that though
these fish are so very abundant, and so ready to seize a bait or
a fish that has been hooked, no instance of their taking a man
has ever occurred, and their only victim that I know of was a
boy who, though lacerated severely in an encounter, recovered
from it perfectly. Notwithstanding their abundance it is

THE LITTORAL AND MARINE FAUNA 321

the custom to bathe in the lagoon regardless of their presence,
and even the dogs of Pulu Tikus are accustomed to swim
about quite unmolested. It was once my misfortune to be

Fic. 72.

LANDING A SMALL BLACK AND WHITE SHARK ON THE SEAWARD
BEACH OF PuLU TIKUS.

upset in the middle of the lagoon, but the sharks did not

avail themselves of the opportunity. Yet, only a few days

before, and in the very same spot, they had been so bold in

their quest of the body of a speared devil-fish, that they seized

the blades of the oars when an attempt was made to tow the

carcass beyond their reach, The black sharks afford good
x

322 CORAL AND ATOLLS

sport when gear strong enough to hold them can be used from
the shore, and on the occasions of the killing of a pig, they,
and the dog-fish, may be easily shot or speared, for they will
dash into a few inches of water to secure a tit-bit thrown into
the wash. Although the sharks appear to be but very little
threat to life, they are in every way a nuisance, for they spoil
the fishing, and not only do they take bait and break up
fishing-tackle, but they will seize a fish that has been hooked,
before it can be drawn into the boat. All this they do with-
out even thinking to turn on their backs in order to bite.

Another fish that may be a real danger is the beautiful
sword-fish—Histiophorous gladius—for it dashes through the
water with great speed, and on more than one occasion
has made a sudden rush among a party of bathers. It is
coloured a vivid purple when alive, and presents a very
different appearance from the dried leathery specimens seen in
museums.

The sword-fish has been known to transfix thick wood
with his prolonged snout, and the Governor has found upon
the beaches a stout bamboo run through by the sharp sword
of one of these creatures, but I have never heard of a case in
which a man had been speared, although the narrow escapes
have been many. The baracouta has, however, left its mark
upon at least one man, and I have seen a Cocos native, the
whole of whose calf muscles had been torn away by the cruel
teeth of this large relation of the pike. The baracouta is an
ugly and a dangerous fish, for it is of all sea-creatures the most
difficult to see, and it has a habit of dashing upon whatever
stirs in the water. Another fish with the same habit is the
Sambar, whose name is given to it in consequence of this habit,
for it signifies the swoop as of a bird of prey upon its victim.
So quick is the Sambar that it will dash among a shoal
of smaller fish, and with the impetus of its rush be carried
clean ashore, and I have seen a Sambar in chase of mullet
dash up the beach at a man’s feet, and be kicked high and dry
beyond the reach of the water before it had recovered from its
surprise.

PLATE XX.

THE 1KAN PDANG, A SWORD-FISH NAMED HISTIOPHORUS
GLADIUS, TAKEN ON A TROLLING LINE.

THE LITTORAL AND MARINE FAUNA 323

The largest of all the atoll, fish is the Zhan parch, or “ devil-
fish ” (Dicerobatis eregoodoo), and it is to a certain extent a useful
fish, for a quantity of oil can be prepared from its liver, and

Fig. 73.

A BARACOUTA TAKEN ON A TROLLING LINE IN
THE LAGOON.

its skin makes excellent sand-paper. The Pareh is the royal
game of the islands, and the spearing of one which measured
13 feet across was the most exciting fishing expedition that I
have been privileged to take partin. The Pareh isa curious fish,
which comes to the lagoon only at certain times of the year,
and generally on a rising evening tide. At such times I have
seen as many as twenty in the sail between Pulu Tikus and Pulu

324 CORAL AND ATOLLS

Selma; but. for long periods not one will be seen anywhere in
the neighbourhood of the islands.

Fic. 74.

Tue Ikan Pareh (Dicerobatis eregoodoo), A GIANT SKATE MEASURING
13 Frnr across its Bony.

Among the fish which are useful for food, the “ Green fish ”
is one of the most important, as it is excellent of flavour and
grows to a great size. It is a fish of remarkable appearance,

THE LITTORAL AND MARINE FAUNA 325

being covered with a thick armour of very large and beautiful
green scales, but its armour is useless as protection against
the spear of the Cocos-born Malay. It may be either speared
or taken on the line, and in the latter case it is usually
killed in about 6 or 8 fathoms of water with a Kapeting
balong as the bait. In the stomachs of those Green fish that I
have opened, there have been, in addition to the legs and
shells of crabs, large numbers of the spikes of sea-urchins; and
for the proper mastication of these hard objects the fish is
provided with a remarkable grinding-mill situated far back in
its gullet. The teeth that are set round the jaws of the Green
fish are peg-top teeth, used only for picking up its food from
the crevices in the rocks ; all the grinding of the hard parts of
its spiny victims is carried on by the gullet mill, which consists
of knobs of bone studded over with enamel-covered bosses.

The Green fish grows to a great size, one of 200 pounds
in weight not being a very exceptional fish, and with increasing
size the delicacy of the flesh as an article of food does not
diminish, but rather improves.

There are several relations of the Green fish, the
Dongol being the finest of them, but this fish is not to be
taken on the line, for it lives in the tumble of the roughest
barrier water. Great shoals of them swim along the barrier
rocks and feed upon the Nullipores which clothe the boulders,
and though the natives can spear them in the breakers, I was
never successful in stopping one with a rifle bullet. The
Kakatua, however (of which there are several species all
belonging to the genus Scarus), may be shot with cased bullets
with comparative ease, and fish-shooting in the tumbling surf
of the barrier was one of my most delightful occupations in
the atoll. The many Kakatua are fine fish, splendidly coloured,
and while some species are bright blue or green, some are
blue and yellow, some blue and pink, and some are vivid
red in various combinations. Some that inhabit the lagoon
may be taken upon a line, but most must be speared or
shot, for although it is possible to fish at times upon the
barrier, the loss of fishing-tackle among the rocks makes the

326 CORAL AND ATOLLS

process an expensive one. Some fish may even be killed with
a shot-gun, and the Garfish (native name Zalang-talang) and

ING, 7/5).

THe “GREEN Fisu,” Jkan Ijou (Pseudoscarus),
WEIGHING 112 PounpDs.

mullet, which swim at the surface, are those that fall the most
easy victims. At times when the tide is out upon the barrier

THE LITTORAL AND MARINE FAUNA 327

very fine fish may be shot as they bolt across a shallow pool
to the deeper water beyond. These barrier pools are the homes
of some of the most wonderful of all the atoll fish, and it is
impossible to describe the splendid colouring, vivid blue and
intense purple, of the myriad tiny fish that spend their lives
among the branching growths of coral. The Pig-fish (Lalistes)
and a host of others have wonderful patterns picked out in
boldly contrasted colours, and there are many very curious
points in connection with the mode of coloration of these
bright tropical fish.

It is not uncommon for the two ends of a fish to be
entirely different in coloration and pattern, so that an appear-
ance is presented as though two fish had been cut in half,
and the head end of one joined to the tail end of the other.
Another very common design is for a fish to possess one patch
of colour that presents a glaring contrast to all the rest of its
body ; and in such a case a vivid spot is seen moving through
the water as the fish swims, but the whole body of the creature
becomes very difficult to define.

There is no doubt that these strange freaks of coloration
serve a very definite purpose, but we know so little of the
elements that enter into the factors of the lives of these
fish, and so little of the appearance that these things present
when viewed by other creatures in the water, that they appear
to us rather as strange and fantastic efforts of Nature than
as the useful defensive and offensive aids that they un-
doubtedly are. There is much to be learned about these
creatures, and no better place than Cocos-Keeling could be
chosen for studying the reasons for their strange ornamenta-
tions, for, in the clear water of the barrier pools and in the
lagoon, observations may be made upon them in their natural
environment under the very best conditions.

Strange methods of defence of more obvious utility than
coloration are to be seen in the many fantastic forms of fish
which carry armour-plate, and spikes around their bodies. The
Tetradonts show some of these modifications in their most
exaggerated forms, as some are shaped like little hard boxes

328 CORAL AND ATOLLS

with thick spikes projecting fore and aft, others are one mass
of prickles like hedgehogs, and some have only a few cruel
spikes situated so that they are sure to wound any animal
that attempts to seize them. One of the strangest of the
Tetradonts is the Zkan Buntal, that can inflate its whole body
at will, and so float lightly on the surface of the water, a

Fig. 76.

Tue Ikan Buntal (Tetrodon patoca).

bobbing air-ball clothed with prickles. The method of infla-
tion is truly marvellous, for a Buntal when irritated can inflate
itself in a very few seconds, and from being a creature not
much larger than your fist, swells up to the size of a football:
and all this is done when the fish is lying calmly under water.
It can inflate and deflate itself many times in rapid succession,
and can go through the whole performance on land as easily
as it can when in the water.

The Luntal is the one poiscnous fish of the atoll, and the
whole of its poison is concentrated in its gall-bladder, so that
if the gall-bladder be removed whole, the fish is said to be
excellent eating; but I never tried it. I have kept these fish

THE LITTORAL AND MARINE FAUNA 329

in ponds and tanks, and they rank among the most remarkable
of all living creatures, and certainly among the most tenacious
of life, for they can stand an extraordinary amount of exposure
to the sun and air.

Some fish that seem to have no special method of pro-
tection in the matter of coloration or armour-plate, are those
that spend their lives in the shelter of the irregular coral
masses of the deeper parts of the lagoon and island slopes ; and
they evidently depend for their preservation upon their power
to lurk among the rocks. One such fish is the “ rock cod” or
Grapu, a fine fish which offers good sport, as it is a most
voracious feeder, and is excellent eating. The rock cod is
mostly remarkable for the enormous size of its mouth ; it can
swallow any bait which even a good-sized shark could take in,
and the gape of its jaws is wonderful. In some measure it
seems to depend upon this enormous gape to resist being
pulled from its rocky haunts, for when hooked it invariably
comes up with its mouth wide open, and a fish of only a few
pounds can offer a very stubborn resistance in this way.

Around the slopes of the island ring, various fine fish may
be taken on days when it is possible to go out into the open
ocean in a small boat, and snapper, albicore, and “ king fish,”
are among the best to be had in this way, but the success of
the fisher is frequently marred by the presence of the sharks,
which steal his victim from the hook before it can be pulled
aboard the boat.

Large numbers of strange eels live around the rocky
shores. They are of many species, of which the commonest is
a large grey creature that may often be seen coiled up on the
surf-washed edge of the barrier.

The animals are named Lendong by the Malays, and are
justly held in some dread, for their bite is very severe and
they are dangerous creatures to deal with. One species is even
credited with possessing poisonous properties, and I think that
it is not unlikely that its character is a true one; and certainly
its appearance is so really repulsive that it naturally suggests
the possession of evil powers. This eel (Ophichthys colubrinus,

330 CORAL AND ATOLLS

var. annulata), which is called Ular tana by the natives, is
boldly banded with black and white, and it lives in the sand
of the lagoon: it does not appear to be a very common species
in Cocos, and I only obtained one specimen during my residence
in the atoll. A small grey eel that is very common among
the holes in the barrier rocks is an interesting creature to
watch, for it preys upon the crab named Kapeting trelek, and
the swiftness with which it will pounce upon one of these
crabs and withdraw with it into its hole is quite astonishing.
Besides eating the Kapeting trelek, these eels destroy a far
more valuable creature, for they are particularly fond of young
turtles, and turtles are none too common in the lagoon to-day.

Of the myriad other creatures that live in the clear waters
of the atoll, some have already been mentioned because of the
importance of their actual influence upon the structure of the
atoll, and the Béche-de-mer are the most important of these.
Béche-de-mer live in extraordinary numbers in this atoll for
the reason that their exportation is wisely prohibited; they
form avery valuable asset to the islands, and one which could
be easily exploited in the event of a sudden failure of the crop
of coconuts. Béche-de-mer when dried are called trepang,
and they are extensively used in China and elsewhere in the
preparation of a soup that is nearest akin to turtle soup, and
is very excellent. The demand for them is great, and very
few of the Chinese servants who come to the atoll can resist
the temptation to try to smuggle some away when they leave
the place.

The temptation is very easy to understand, for the sea-slugs
may be picked up by the thousand without any trouble, and
they need no preparation for the market save simple drying in
the sun.

Many beautiful star-fish (native name, Ular bintang)—one
being a very vivid blue—lie scattered about the rock pools;
and a great number of very quaint and beautiful sea-urchins
(native name, Zerongan) are to be found about the crevices of
the barrier rocks. The shells to be found both in the lagoon
and on the barrier are extremely fine, and one of the most

THE LITTORAL AND MARINE FAUNA 331

pleasant occupations of coral island life is walking at low
tide along the flats in the hunt for the fine polished cowries
which are to be found beneath any loose boulder that may be
turned over.

It is a delightful quest, and one of which the interest
never flags, for beneath every stone that is moved there are
a hundred forms of life, and most are of brilliant colour and
strange in shape. Splendid Nullipores clothe the rocks, bright
and delicate Nudibranchs and shining cowries (native names,
Dakon and Groos) live in the crevices; crabs and fish dart this
way and that as their home is moved from its foundations.

Octopuses (native name, Gorita) also live among these
rocks, and they are among the most wonderfully protected
of animals, for in the twinkling of an eye they assume per-
tectly the colour of the boulder to which they cling. They
are fairly easily speared, and are by no means bad to eat when
properly prepared and cooked. Their curious defensive method
of expelling their ink into the water is a very effective one as
far, at any rate, as man is concerned, for when one has been
hunted from rock to rock, and has finally obscured all the
water for many yards around with the dark brown stain, it is
as well to give up the chase, for he will not likely be found
when at length the water clears. Portions of the tentacles
of Octopuses of phenomenal size have been washed ashore
upon the atoll beaches, but none of those that I have seen alive
about the islands have been of alarming proportions. Jelly-
fishes (native name, Obor obor) come to the lagoon at certain
seasons only, and then they are to be found in great numbers,
but for many months on end not one will be seen anywhere
about the whole of that part of the ocean, The Portuguese
man-o’-war (Physalia), however, knows no season, and some are
generally to be found in rock pools damaged by the breakers
which cast them up. They are also at times encountered when
bathing, and the sting of their long tentacles is a very severe
one. ‘The sailing of these curious creatures is often to be
witnessed, and the queer bubble-like vessels may be seen
almost any day proudly navigating the ocean, or wrecked, a

332 | CORAL AND ATOLLS

blue and white mass, thrown on the shores. Besides the
Portuguese man-o’-war, there is one other harmful creature to
be met with in rock pools, and that is a worm whose sides
are fringed with an array of hairs that look as though they
were composed of fine spun glass. The hairs of this worm,
upon the very least handling, run deep into the flesh, producing
an intense irritation; and their removal is apt to be a matter
of some difficulty. |

The majority of the creatures to be found any day on a
walk along the barrier at low tide are objects of pure delight,
and nothing that I know is quite so good, or quite such a
perfect pleasure to a lover of Nature, as the hunting for the
many beautiful tropical sea-products in the warm waters about
the atoll.

APPENDICES

heath

ze Ey NED ee

DIST OF) Enna eS UiNEA:

MAMMALIA.

(1) THe Puiu Trkus Race or Mus decumanus.

During Darwin’s visit the only rat in the islands was the
Pulu Tikus race, and of it he says: “These rats are con-
sidered by Mr. Waterhouse as identical with the English kind,
but they are smaller and more brightly coloured ” (“ Naturalist’s
Voyage,’ p. 461); and Wallace (“Island Life,’ p, 275) adds :
“We have here an illustration of how soon a difference of
race is established under a constant and uniform difference
of conditions.”

In the seventy years that have passed since Mr. Water-
house described the rat, the difference appears to have become
more marked, and a description of the race is justified for the
reason that the modern introduction of rat virus into the atoll
may possibly exterminate it.

External Appearance.—The rat is a slenderly built and sleek-
looking animal; its general colour is a warm russet-brown.
The fur of the back is coarser, and some hairs stand out which
are almost black, and are 30 mm. in length; these long dark
hairs are more numerous over the hind end of the body than
over the shoulders. The general colour of the body is a rather
rich brown, the belly being but little lighter than the back in
most specimens. The shorter hairs are grey at the basé and
bright red-brown at the tips; the longer hairs are dark brown

to black. The fore limbs are somewhat lighter than the rest
335

336 CORAL AND ATOLLS

of the body, but the hind limbs are of the general rich red-
brown. The under surface is only slightly lighter than the
back, and is a light warm brown, not grey; the chin and
throat are somewhat lighter. The under surface of the
scrotum of the male is covered with long bright red-brown
hairs, brighter than those of any other part of the body; the
distal extremity of the scrotum is bald, the naked skin being
a dark purplish brown.

The vibrisse—-many of them more than 60 mm. long—
are numerous and dark-coloured. The ears are almost naked,
they are oval and prominent and average 22 by 15 mm. The
hind feet are large and long, their soles are dark, almost
purple in colour; the digits are slender, the claws are long
and much curved. The tail very slightly exceeds the length
of the head and body, it is darker than the body colour; it is
finely scaled, 12 rows to the centimetre, and sprinkled with
numerous fine black hairs. There are two pairs of pectoral
mamme, and three inguinal pairs, 2-3=10. The tibiz are
almost straight, having practically no “ bow” forwards.

The Skull—tThe skull is long and narrow, and is delicately
built. The nasals, which extend posteriorly to the anterior
margin of the orbit, are 15°5 mm. long, and 2°8 mm. wide
at their broadest part. Basal length of skull 35-38 mm.;
greatest breadth18 mm. Interorbital constriction, least breadth
6mm. Interparietal, length 5 mm., breadth 10 mm. Length
of base of anterior root of zygoma 6 mm. Palate, length 20
min.; breadth, outside m' 9 mm., inside m’ 4 mm. Palatine
foramina 8 mm., extending 1 mm. behind the anterior edge of m*
—this is a constant feature. Alveoli (back of incisors to m?*)
12 mm. Upper molar series 6°55 mm. Incisors pale yellow
in the lower jaw, orange in the upper jaw.

_ ‘The measurements in millimetres of a series of specimens
are as follow:

APPENDIX I 337

Hind Leg Front

ae Tail Foot. (Tibia). Foot. Forearm. Sex.
it 190 190 40 50 20 35 3
2 207 208 40Q 45 20 33 3
3 195 200 40 50 20 30 3
4 195 191 38 42 Dy) 25 }
5 205 215 40 52 19 34 3
6 187 185 40 50 18 32 3}
7 192 215 40 50 19 34 3
8 200 205 40 45 20 30 2
9 180 185 40 45 20 30 ©
10 163 158 39 40 22 29 2
11 150 155 38 39 22 27 g
12 147 149 36 38 18 24 2
13 215 230 42 45 22 36 g
14 198 202 39 48 20 32 g
15 187 190 39 48 18 32 S
16 170 160 35 42 20 38 g
17 120 130 30 35 17 32 2

(2) Mus DECUMANUS, typ.

This is the pest of the atoll, and does incalculable damage
to the coconut plantations.

(3) Mus RaTTUS, typ.

Is common on Pulu Gangsa and some of the small southern
islets, and I have had one specimen from Pulu Selma, where
it is not nearly so common as MW. decwmanus.

(4) Mus muscuuus, Linn., is not uncommon on Pulu Tikus.

A herd of feral deer lived for long in Pulu Luar. The
animals were introduced from Java and from Singapore, and
consisted of two species, the Sambar (Cervus hippelaphus) and the
Kedang (Cervus muntjac). All were dead when I left the islands
in 1906.

Waits and strays include bats of some small species which
did not appear during my stay in the atoll,and which are said

to be the Pipistrellus murrayi, Andrews, from Christmas Island.
.

338 CORAL AND ATOLLS

AVES.

4) {
COLUBRIFORMES.
(1) CarpopHacaA WHARTONI, Sharpe.
Native name, Pergdm.

Introduced from Christmas Island, but by 1906 it was
practically extinct in the atoll.

RALLIFORMES.

(2) RALLUS PHILIPPIENSIS, Linn.

Native name, Ayam utan.

Very abundant on all the islands, and is everywhere very
tame, it being a matter of some difficulty to make it take
wing. It fags on the shore when the tide is out, but it may
also be seen perched high in papaia trees eating the ripe fruit,
and it has a bad name for eating the eggs of domestic fowls.
It nests in September, in tufts of grass, about a foot from the
ground; it lays from two to six eggs, very like the English
corncrake’s. The young are all black when hatched, and can
run directly they are out of the egg. The call-note is a shrill
grating sound, and in the breeding-season the cock adds a
deep croak not unlike the noise made by frogs. This species
is not found in Christmas Island.

LARIFORMES.

(3) STERNA FULIGINOSA, Gmel.

Native name, Burung dali.

Breeds in Keeling Island but not in very great numbers.
It keeps very much to itself on the breeding-grounds, and lays
one egg in the sand above the beach rise, on the western side
of the island. By sailors itis cailed the whale-bird. Although
not uncommon, and often seen at sea, far from the atoll, it
does not occur on Christmas Island.

APPENDIX I 339

(4) Anous sToLipus, Linn.

Native name, Burung krok.

Not resident in the Southern atoll, but a frequent visitor
to the lagoon. In Keeling atollon it breeds in great numbers.
In June the breeding-season has almost come to an end. The
nests are made about a foot above the ground, on little collec-
tions of sea-tossed wrack. In the atollon all the “ noddies”
build close together in a rather limited area. Only one egg
was found in each nest.

(5) Gyeis canpIpA, Wagler, Isis, 1832, p. 1223.

Native name, Burung chuit-chwit (onomatopeeic).

Darwin noted the bird in 1836 (“ Naturalist’s Voyage,”
p- 462), and H. O. Forbes gave an account of it in 1879
(“ Naturalist’s Wanderings,” p. 34). The bird agrees with the
description of Wagler: “Irides nigro-cerulee: rostrum
nigrum, basi ceruleum; lingua sublata rostro brevior ; pedes
cerulei, palma alba, ungues nigri.”

In a good description of this bird (Cassin, “U.S. Explor.
Exped.,” p. 389) the feet are said to be “ pale blue, having a
deeply indented yellow membrane.” In Gould’s “ Birds of
Australia,” vol. vil., the feet are described as orange, and are
figured of that colour in the accompanying plate. In other
accounts it is stated that the feet are yellow or brown. The
bird, as I have seen it in the Cocos-Keeling group, invariably
has the feet entirely blue, the web being slightly lighter than
the toes.

The bird is a common one, and occurs in all the islands,
It breeds in the Southern group as well as in Keeling Island.

Although the bird feeds mostly on fish, it may often be
seen hovering around the papaia trees and eating the soft
fruit. It is quite arboreal in its habits, and its one egg is laid
sometimes at a height of 40 feet from the ground. Most of
the eggs were found in September, but on visiting Keeling
Island in June two eggs were found after very little searching.
The egg is almost perfectly oval (40-43 mm. in its long axis,
21-22 mm. in its short axis); it is cream-coloured, mettled

340 CORAL AND ATOLLS

and streaked with olive-brown. It is laid on the branch of
a tree (Pisonia inermis, Cordia subcordata, Guettarda speciosa)
wherever a slight irregularity in the bark suffices to hold
it steady. In all the cases that I have seen the egg was

Mies 7.

Eaa or Gygis candida, LAID ON THE BRANCH OF A
GRONGANG TREE (Cordia subcordata).

laid in the long axis of the branch, and the parent bird sat
across the branch in the process of incubation. From the
fact that three nesting sites were used twice over, it would
seem as though the suitable branches for balancing the egg
were somewhat limited. The birds take it in turn to sit, or
rather stand, on the egg, and they invariably leave it by falling
backwards off the branch, in order not to disturb it with their
feet; great caution is taken in getting into position on the

APPENDIX I 341

ege again. Incubation lasts 36 days; the newly hatched
young is buff-brown, and it does not move from the tree
until it has assumed its white feathers. A bird hatched on
September 3, 1905, did not move from the site of its hatch-

Fic. 78.

Gygis candida SITTING ON ITS Hac.

This is the same egg as is shown in Fig. 77.

ing till October 22; it then moved higher up the branch,
and remained, with slight changes of position, till Novem-
ber 12, when it took its first flight. Many eggs were watched,
and in no case did any accident befall them, but a young
bird, batched on November 17, fell from its insecure position
on the fourth day, and was killed. This bird does not occur
on Christmas Island, and is not seen far from the atoll—it is,
in fact, a sign that the atoll is near when one is seen on the
voyage from Java.

342 CORAL AND ATOLLS

ARDEIFORMES.

(6) DEMIEGRETTA SACRA, Sharpe.

Native name, Blakok.
Exists in fair numbers, especially in Pulu Atas, where as

Fic. 79:

Youne Gugis candida, SOON ABOUT TO TAKE ITS FIRST FLIGHT.

The plumage is still mottled with buff-coloured feathers.

many as twenty may be seen fishing together. Birds in the
white phase and in the grey phase mix freely together;
though it is said in the islands that white pair with white,
and grey with grey. The nest is made in the Pisonia trees,
and two pale blue eggs are laid. I obtained specimens in
January.

APPENDIX I 343

In Pulu Atas I have also seen a solitary individual of a
larger species of Heron, which, in the distance, appears to be
pure white; this individual was well known to the natives of
the atoll.

PELECANIFORMES.

Suborder FREGATI.

(7) Freeata aguiia, Linn.

Native name, Burung itam.

Does not breed in the main atoll, but nests in large
numbers in Keeling Island. The nest is a slight collection of
twigs, stolen from the nesting Gannets, and placed on the flat
tops of the Pemphis bushes. The nests are crowded together
in thousands ; one egg only is laid, which is pure white and its
surface is shining; eggs were abundant in June. The bird is
so tame in Keeling that it has to be driven from the egg, and
those birds that are sitting in the hot sun, with their wings
half spread, may easily be approached and lifted from the
ground,

(8) FREGATA ARIEL, Gould.

This species is more common than the last. Its habits are
similar, and it nests freely with the larger species. It is also
called Burung itam by the natives.

Suborder SULA,

(9) Suna suLA, Linn.

Native name, Burwng bebek.

I have never seen this bird in the main atoll, and its
numbers are somewhat limited in Keeling Island. It lays one
or two dirty white eggs on the bare ground, at the top of the
western sand-beaches.

(10) Suna AssortTi, Ridgw.

Native name, Burung gangsa.
Ts not common in Keeling, and, so far as I have observed,

344 CORAL AND ATOLLS

does not come to the Southern group. The face is black, and
the bird is considerably larger than the other two species.
One or two eggs are laid on the bare ground. The black mask
gives the sitting bird a very curious appearance, and it leaves
its egg only after having, with great solemnity, vomited large
quantities of fish ; it makes no attempt to resist being pushed
from the egg, and when the egg is taken, and the fish has
been vomited, it quietly walks away. No skin was obtained.

(11) Suna piscatrix, Linn.

Native name, Burung main main.

The commonest of the three species, and generally called
“Booby” by sailors in these seas. Exists in very great
numbers, and is seen in all the many age changes of plumage,
from a uniform light brown to white. It lays one egg, in a
nest placed, as a rule, high up in trees; but numbers build on
Pemphis bushes in Keeling. The birds commonly perch on
trees, and are often taken from the rigging of ships.

Suborder PHAETHONTES.

(12) PHAETHON RUBRICAUDA, Bodd.

Native name, Burung buntut.

Does not exist in great numbers, but still breeds in Pulu
Atas. This species, as well as P. fulvus, is to be seen every-
where between the atoll and Java Heads.

PASSERIFORMES.

Fam. PLOcEIDA.

Dr. H, O. Forbes mentions Ploceus hypoxanthus as one of
the nesting species of the atoll; but no examples of it exist
to-day.

(13) Munta oryzivora, Bp.

Native name, Burung glatek.

Introduced, and has multiplied greatly; it is now one of
the commonest birds of the atoll.

APPENDIX I B45

Fam. TURDID.

(14) MeRULA ERYTHROPLEURA, Lister.

Introduced from Christmas Island, and now very common,
especially on Pulu Tikus. The bird is remarkably tame,
coming freely into rooms to obtain water. Has a rather
pleasing song, which it utters in September and October.
The nest is built in November, and the egg is very like that
of the Kuropean Redwing.

Fam. ZOSTEROPIDA.

(15) ZosTERops NATALIS, Lister.

Introduced from Christmas Island. Confined to Pulu
Luar, where it exists in some numbers. In Christmas Island
it is known as Burung waringin, but in Cocos-Keeling it has
earned the name of Burung chinta or “ love-bird.”

Golden plover, native name Burwng bdlebis, and Curlew
are said to breed on some of the islands, but I never found
the eggs, nor did I ever obtain a bird for identification. A
flock of ducks is resident on Pulu Panjang, and on the same
island I have seen many small waders, but the wise policy of
the islands is that these birds shall not be interfered with
and I cannot name the species. I have, however, shot a stray
Snipe (Scolopax gallinago, Linn.) from a small flock on Pulu
Tikus, and on the same island obtained a Swallow, Wiruwndo
guituralis, Scop. These, and a wagtail, are the only visitors
that came to the islands in my stay of fifteen months, and the
atoll would seem to be altogether out of the track of any, save
very wind-blown, migrants.

In North Keeling atollon are large warrens, the nesting
sites of “ Mutton Birds” ; during my visit to the atollon these
birds were not present, but they evidently visit the place, for
the purpose of breeding, in very large numbers.

In all probability they are Pufinus brevicaudus or P,
sphenurus,

346 CORAL AND ATOLLS

REPTILIA.
Species determined by G. A. BouLENGER, F.R.S., V.P.Z.S.

(A) LacerTILia.

Fam. GECKONIDA.

LEPIDODACTYLUS LUGUSTRIS, D. & B.

Native name, Cheechak.

A very abundant species, inhabiting dwelling-houses, and
also living beneath the loose bark of coconut palms. Six or
seven eggs are laid at a time—as a rule beneath some con-
venient chink of bark. It has the characteristic voice of its
family, which has given rise to its name.

(2) GEHYRA MUTILATA, Wiegm. (Plate xx. Fig. 1.)

Not nearly so common as the first species, and lives
exclusively out of doors, being always found beneath the
bark of the coconut palms. It is at once distinguished
from No. 1 by its flattened tail. It lays from three to four
eggs, and the incubation period is as long as two months
(June to August).

(B) OpHIDIA.

Fam. TYPHLOPID.

(3) TYPHLOPS BRAMINUS, Daud.
Native name, Ular minyah.

This is not a common species in the islands, and its habits
lead to its being very rarely seen. It is most commonly found
beneath large coral boulders, and it is extremely active. It is
found on practically all the islands in the atoll. The average
length is 150 millimetres.

None of these species is found in Christmas Island.

XXI.

PLATE

NY
js, 2
Zw Ni)

ene Bere)

ALL trompet.

(00) IK

PLATE XxI,

Pa TUT

y

APPENDIX I 347

(C) CHELONIA.

Two species of Turtles are frequently speared in the lagoon.
They are Chelone imbricata, L., native name Pinnew sisih, and
C. mydas, L., native name Pinnew betul. Both have ceased
to use the main atoll as a breeding-place, for the native
demand for them is too great; but on Keeling atollon their
nests are common on the sandy beaches.

PISCES.

The fish of the group were not collected, and no doubt
a rich harvest awaits the investigator of the myriad smaller
coral-haunting species. The native names of those that are of
the most economical importance only are given here; and no
specific determination is possible in most cases.

(1) Ikan babi=pig-fish, Balistes, sp. (2) Ikan buntal=
inflated fish, Tetrodon patoca. (3) Ikan buntal besagi = square-
shaped inflated fish, Zetrodon, sp. (4) Ikan blanah. (5, 6)
Ikan bandang and Ikan bandang laut. (7) Ikan buntut
burih = spotted tail. (8) Ikan chuchut, generic name for the
numerous Sharks. (9) Ikan dongol, Scarus, sp., grows to a
great size. (10) Ikan grapu. (11) Ikan iju=green fish,
Pseudoscarus, sp. (12) Ikan jengot=bearded fish. (13) Ikan
jengot karang (karang=coral). (14) Ikan kakap. (15) Ikan
kakap kuning= yellow kakap. (16) Ikan kakatua iju, Scarus,
sp. (17) Ikan kakatua merah=red kakatua, also a Scarus.
(18) Ikan merah=red fish. (19) Ikan malam=night fish.
(20) Ikan menyrat. (21) Ikan padang déeh, the albicore.
(22) Ikan pdang, Histiophorus gladius. (23) Ikan pareh,
Dicerobatis eregoodoo, grows to a great size, being upwards of
13 feet across the back. (24, 25) Ikan peteh, and Ikan
peteh kuning. (26) Ikan puti=white fish. (27) Ikan palo.
(28) Ikan sambar. (29) Ikan samsi. (30) Ikan skagnol.
(31) Ikan talam talam. (32) Ikan talang. (33) Ikan
tangiri, the king fish. (34) Ikan todak, the baracouta,
(35) Ikan trompet.

348

(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)

(10)

(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
(19)
(20)
(21)
(22)
(23)
(24)
(25)

CORAL AND ATOLLS

MOLLUSCA.
Species determined by E. A. Smiru, F.Z.5S.

CypR#A HIRUNDO, Linn.

CypR@A ANNULUS, Linn.

CypR#A MONETA, Linn.

CypR#A CAPUT-SERPENTIS, Linn.
Cypr#A ARABICA, Linn.

CyPR#A LYNX, Linn.

Cypraa cAuRICA, Linn.

CypRHA ISABELLA, Linn.

CypR#A VITELLUS, Linn.

CypRHA MAURITIANA, Linn.
CypR&A TIGRIS, Linn.

CrpyR#A GLOBULUS, Linn.

NASSA GRANIFERA, Kiener.

NERITA POLITA, Linn.

NERITA ALBICILLA, Linn.

TURBO PETHOLATUS, Linn.

TuRBO LAJONKAIRII, Deshayes.
LAMPROSTOMA, sp. (young).
FossaAR TROCHLEARIS, A. Adams.
MELAMPUS LUTEUS, Quoy and Gaimard.
Meuampus castaneus, Muhlfeldt.
Metampus CAFFAR, Kiirler.
AURICULA TORNATELLIFORMIS, Petit.
MELEAGRINA MARGARITIFERA, Linn.
Pinna niGRA, Chemnitz.

The collection of the Mollusca was by no means a repre-
sentative one, for it comprised only the most conspicuous and
more abundant forms.

APPENDIX I 349

LEPIDOPTERA.

LEPIDOPTERA RHOPALOCERA,
(Native family name, Kupu kupu.)
Species determined by Mr. F. A. HEron,

(1) Precis vitiipa, Fabr.

This is the commonest butterfly in the atoll. It is most
abundant in July and September, and again in March and
April. The larva is black, and is covered with spines: food-
plant is Asystasia coromandeliana. The pupa is suspended by
the tail from the branches of its food-plant, it is mottled
brown and has no metallic markings. The pupal stage lasts
for a fortnight or three weeks.

(2) Hypotimnas Borina, Linn.

(3) HyPoLIMNAS BOLINA, form NERINA, Fabr.

An uncommon species, on the wing in April and August.
Specimens are in perfect condition, and the species is evidently
resident.

(4) Hypotimnas misrppus, Linn.

A common species on all the islands of the group. One
worn specimen was taken on Christmas Island by Dr. Andrews.
The males and females of this species are very rarely seen
together; the males flying high in the shade of the thickly
growing cocopalms, and the female being generally seen flying
low over the herbage in open spaces. It is abundant from
July to October, and again from the end of February to April.
The larva is brown and spiny, the head is reddish and bears
two spines. It feeds on a succulent weed that grows all over
the open spaces in the atoll. The pupa is brown and has no
metallic markings; it is suspended by the tail. Six days are
passed in the pupal stage. The female mimics Danais (Limnas)
chrysippus petilia, Stoll, and is nearly always seen flying in
company with that species, whose habits of fluttering over the

350 CORAL AND ATOLLS

herbage it has closely imitated. The male, on the other hand,
flies strongly and is not often seen fluttering near to the
ground.

(5) Vanessa KersHawl, McCoy. (Australian.)

Only four examples of this species were met with in fifteen
months, and since all were observed between May 15 and
17, 1906, it is probable that they were waifs. Two rather
worn specimens only were captured on Pulu Tikus.

(6) Danats (LIMNAS) CHRYSIPPUS PETILIA, Stoll. (Australian
and Christmas Island.) (Plate xx1. Fig. 2.)

Common on Pulu Luar and Pulu Tikus, but not so often
met with on the other islands. The larva is pale green,
banded with yellow and black. The food-plant is Asclepias
curassavica. The pupa is suspended by the tail, and is pale
green or buff with small golden dots. The average stay in the
pupa is only 135 hours.

LEPIDOPTERA HETEROCERA.

Species determined by Sir GEorcE F. Hampson, Bart., F.Z.S.

ARCTIADA.

(1) UTETHESA PULCHELLOIDES, Hmpsn.

A very abundant insect on all the islands. It mostly
frequents the seaward side, for there the food-plant, Touwrnefortia
argentea, Linn., lives most luxuriantly. The perfect insect, and
the larvee, are to be seen at all times of the year, and season
appears to make no difference to its numbers.

Nocruip&.
(2) CHLORIDEA OBSOLETA, Fabr.

No specimens were taken during 1905, and the insect only
became at all plentiful in June 1906.

APPENDIX I 351

(3) CHLORIDEA ASSULTA, Guen.

Appeared at the same time as the last species, and was not
seen in any numbers.

(4) CiRPHIS LOREYI, Dup.

Not an abundant insect, and strictly confined to a seasonal
appearance. ‘Taken in June 1905, and not again until the
same month in 1906.

(5) PRODENIA LITTORALIS, Boisd.

Abundant on all the islands, on the wing most plentifully
in May and June. The larva feeds on a multitude of low-
growing plants. This very widely distributed species is found
on Christmas Island.

(6) SPODOPTERA MAURITIA, Boisd.

Several specimens were taken in June 1905, but in June
1906 it did not appear again.

(7) LeocyMA sERICEATA, Hmpsn.

Not abundant; appears in June and again in November.
The larva is pale green and bears a few scattered hairs; it
feeds on the leaves of the Waru (Hibiscus tiliaceus, Linn.).
Not on Christmas Island, though its genus is represented by
L. tibialis, Fabr.

(8) OpHtusa coronaTa, Fabr.

Only one specimen taken, and a few more seen, in June
1905. Occurs on Christmas Island.

(9) OpHIUsA MELICERTA, Dru. (Plate xx1. Fig. 3.)

This is one of the most plentiful of the atoll insects,
being found wherever the bushes of Pemphis acidula (native
name Kayu burung) are growing. It is on the wing practically
all the year round, but is most common from June to Sep-
tember. The larva bears a wonderful resemblance to the
twigs of the Pemphis acidula or the Ricinus on which it feeds.
It has a habit of dropping from its twig when disturbed, but
it always manages to get a fresh hold of a lower twig, on the

352 CORAL AND ATOLLS

way down, although it appears to be falling to the earth. It
pupates in a few leaves spun together at the ends of the twigs;
the pupa is covered with a fine bloom. Fourteen days are
passed in the pupal state.

(10) Remicia FRUGALIS, Fabr.

Most abundant in May, and a few stragglers again in
September. It flies by day, and is not uncommon in the
grassy spaces, where the undergrowth is kept cleared.

(11) PuLusiA cHALYCTES, Esp.

The development of a garden on Pulu Tikus appeared to
be the cause of the abundance of this species; but since the
garden plants were introduced as seed, there appears but little
chance of their having been the agent for its introduction. It
swarmed in June 1906, and the green larva was abundant on
peas, tomatoes, and almost every plant that was grown in the
garden. The pupal stage lasts only a week.

(12) Hyprena stricata, Fabr.
One example only taken.

(13) CATEPTRIA, sp.

This is a handsome insect, boldly marked with black and
white. It appears on the wing in September. The larva feeds
on the leaves of Pemphis acidula, in company with Ophiusa
melicerta, to which larve it bears a considerable rough re-
semblance. It is mottled grey on the back, and the belly is
velvety black; at the anal extremity is an upwardly directed
tubercle with a bifid tip. A collar of magenta colour surrounds
the larva behind the head, but in all ordinary attitudes this is
hidden by a skin-fold. It pupates among the little leaves of
its food-plant; the pupa has no bloom, and the pupal stage
lasts for twenty days. I never saw an example of the perfect
insect, save those that I reared from larve; and, so far as I
could find out, no one on the island was acquainted with the
moth. The larve were by no means uncommon on a few

APPENDIX I 353

bushes in Pulu Tikus, but I did not meet with them on any
other islands.

SPHINGIDA.

(14) Macroatossa passatus, Dru.

By no means common, and more frequently seen as a larva
than as an imago. On the wing in September. Larva green,
with fine bright pink side-stripes; feeds on Morinda citrifolia,
Linn., native name Mngkudu. The pupa is a mottled yellowish
brown.

(15) CEPHONODES HyLAs, Linn. (Plate xxt. Fig. 4.)

Common on Pulu Tikus, but not often seen on the Southern
islands. Frequents the flowers of the Papaia by day, and is
evidently the source of the error that bees are common in the
islets. When hovering in front of the food-plant, in the act of
laying its eggs on the leaves, the long hairs of the extremity
of the body stand out like a fan. Most abundant in June and
September, but stragglers may be taken in almost any month.
Larva feeds on the leaves of Guettarda speciosa, Linn., native
name Melati. When first hatched they are quite black, and
they only become green in their penultimate and ultimate
skins. They are subject to endless variation, and if bred in the
dark, very striking larvee, coloured yellow and black, can be
produced ; the imagines resulting from these larve are normal.
This insect occurs on Christmas Island.

(16) Hersz convoivun, Linn. (Plate xxt. Fig. 5.)

Not a very common insect. Found most plentifully on
Pulu Selma. The larva is either green or brown, and it feeds
on Convolvulus purviflorus, Vahl. In the atoll this is the moth
specially named Rama rama, although in the Straits that word
is the general name for all moths.

(17) CHa@rocampa viciL, Guer. (Plate xxi. Fig. 6.)

The perfect insect is rarely seen, but the larva is very
abundant in September, It feeds on the leaves of Pisonia
Z

354 CORAL AND ATOLLS

wmermis, native name, Ampol; it is green, but in its last skin
it frequently becomes putty-coloured. When fed in the dark
the last skin is almost.invariably brown. A month is passed
in the pupa. This species occurs on Christmas Island.

GEOMETRIDA.
(18) CHLOROCLYSTIS TENUILINEA, Warr.

Not uncommon in June and August. The larva is green,
with darker markings, and is slightly hairy. Feeds on the
sticky flowers of Pisonia, and pupates among them.

PYRALIDZ.

(19) PYRALIS MANIHOTALIS, Guer.
Not at all plentiful.

(20) MELISSOBLAPTES, sp.
Very abundant on all the islands of the group.

(21) ZINCKERNIA FASCIALIS, Cram.

Always common. This widely distributed species also
occurs on Christmas Island.

(22) MaRAsMIA VENIHALIS, Walk.
_ One of the commonest insects. Appears in June, and is
always to be found about the station lamps.

23) CROCIDOLOMIA BINOTALIS, Zell.
Only a few examples met with in June.

(24) PACHYZANCLA LICARSISALIS, Walk.

Abundant from June to August. The larva is green, with
minute black spots; it feeds on the leaves and flowers of
Dicliptera burmanni, Nees. It pupates in the spun-up leaves,
and the pupal stage lasts for fourteen days.

(25) PAcCHYZANCLA STULTALIS, Walk.

Abundant in June on all the islands,

APPENDIX I 355
(26) GLYPHODES INDICA, Saund.
Not abundant.

PTEROPHORIDA.

(27) A brownish coloured Plume is very common in all
the islands. The larva feeds on Boerhavia diffusa, W., which
grows everywhere in the atoll.

TINEINA,
(28) The solitary species collected was a very common
one ; there are probably at least two others that are distinct.

There are therefore practically thirty species of Heterocera
on the Cocos-Keeling atoll, and there are more than twice as
many species in the fauna of Christmas Island ; yet only five,
very widely distributed species, are common to the two places.

HYMENOPTERA.
Species determined by the late Col. C. T. Brycuam, F.Z.S.

FoRMICIDA.
(Native family name, Samut.)
(J): ODONTOMACHUS HHZMATODES, Linn, 9%.

(2) PLAcroEris LoncrPES, Jerdon. ¢ .

Unfortunately this does not represent the whole of the
species of ants found on the atoll. The natives distinguish
at least three species: (1) Samut saman; (2) Samut arpi;
(3) Samut alus. The other representatives of the family are
unfortunately mislaid.

FOSSORES.
(3) STIzus, sp. (very near reversus, Smith). ¢.

(4) PIson HOSPES, Smith. ¢.

356 CORAL AND ATOLLS
(5) NOTOGONIA SUBTESSELLATA, Smith. ¢ 9.

The Fossores are not at all numerous, and only very
careful collecting produced the very few specimens obtained
in the course of fifteen months.

These five species are not represented in the fauna of
Christmas Island.

DIPTERA.
Species determined by E. E. Austen, F.Z.S..

ASILIDA.
(1) PutLopicus savaNnus, Wied.

A common insect of very voracious habits. It is generally
een sitting on the coral boulders in the hottest sunshine.
It lies in wait for smaller diptera, and it will attack and kill
even butterflies as large as Precis villida.

BoMBYLIDA.

(2) ANTHRAX, sp. Only a very few examples met with.

DoLICHOPODIDA.

(3) New genus and species (No. I.), near Psilopus.

(4) New genus and species (No. II.), near Psilopus.

Both of these are common insects, but owing to their
green colour and small size, they are not at all conspicuous.
_ They are shining green flies that are most commonly seen
running on the leaves of the Hibiscus. They are extremely

active, and are generally found during the hottest hours of
the day.

MusciIp&.

(5) SARCOPHAGA, sp. No. I.

APPENDIX I 357

(6) Sarcopnaca, sp. No. II.

Both are common species, frequenting the dead bodies of
rats. 3

(7) Rurnia Testacua, Rob.-Desv. —

Few examples seen, mostly upon the herbage of the sea-
ward side of the islands.

(8) PyCNCGSOMA FLAVICEPS, Macq: -
The commonest fly in the atoll.

(9) OPHYRA CHALCOGASTER, Wied. —

MICROPEZIDA.

(10) Nertius LINEOLATUS, Wied.

HIPPOBOSCID.
(11) PSEUDOLFERSIA SPINIFERA, Leach. |

Usually swarming over the plumage of the Frigate-birds.
On Keeling atollon, where these birds breed in thousands, the
fly is often seen crawling about the nesting bushes.

CULICIDA.
(12) The Mosquito of the islands is a species of Stegomyia
and it exists in great numbers.

COLEOPTERA.

Species determined by ec O. WateRHOousE, P.ES. ;
C. J. GAHAN, M.A., F.E.S.; and G. J. Arrow, F.E.S.

_ BRACHELYTRA.
(1) ALEocHARA, sp. This genus is not represented on
Christmas Island.

358 CORAL AND ATOLLS

CLAVICORNIA.
(2) DERMESTES FELINUS, Fabr.

A common species, found mostly in the dead bodies of
rats. Universal distribution, and common on Christmas

Island,
(38) COCCINELLA TRANSVERSALIS, Fabr.
A very abundant insect ; not present on Christmas Island.

(4) TROCHOIDEUS DESJARDINSI, Guer.
Not present on Christmas Island.

LAMELLICORNIA.
(5) ONTHOPHAGUS, sp.
Genus not represented on Christmas Island.

(6) ANOMALA, sp.
An abundant species that lives on the iron-wood (Cordia
subcordata). Freely attracted to light and most commonly

found in the lamps of the Telegraph Station. Not found on
Christmas Island.

(7) PoTATIA ACUMINATA.

The largest beetle found on the atoll. It is by no means
common, and not more than a dozen specimens were taken in
fifteen months. Not present on Christmas Island.

MALACODERMATA.

(8) CORYNETES RUFIPES, De Geer.

Native name, Kutw copra.

The most important insect of the atoll, from an economical
point of view. Called by Europeans the “Copra_ beetle.”
Exists in myriads near to the sheds where the copra is stored.
It will swarm to anything oily, and was killed in thousands
on the refrigerator engines. It is very active, it flies by day,

APPENDIX I 359
and has a habit of creeping beneath the clothing and biting
rather sharply. It does not occur on Christmas Island.

SERRICORNIA.
(9) MEGAPENTHES, sp.

Not a common insect. It does not occur on Christmas
Island, where MZ. Andrewsi, Waterh., is the representative of the
genus.

(10) MELANOXANTHUS MELANOCEPHALUS, Fabr.

One specimen only, flying in the sunshine, on March 2,
1906. Does not occur on Christmas Island, but the genus is
represented there by I. dolosus, Cdz., and M. litura, Cdz.

HETEROMERA.
(11) OparrRuM, sp. (near simplex, Fabr.).
A common insect, found mostly under the bark of the

coconut palms. Does not occur on Christmas Island, where
the genus is represented by O. dubiwm, Arrow.

(12) CEROoPRIA INDUTA, Wied.

Very abundant on all the islands. Lives in rotting wood.
The genus is not present on Christmas Island.

(13) SESSINIA, sp.

Native name, Madi.

A very common insect. Found on all the islands, and
freely attracted to light. It is the subject of great variations
in size.

Well known to the inhabitants, Malay and European, as
producing an acute dermatitis.

This species would appear to be very similar to, if not
identical with, S. Andrewsi, Arrow, from Christmas Island; and
of that insect Dr. Andrews says that it exudes an oily liquid,

360 CORAL AND ATOLLS

‘“ which is considered by the residents to have most injurious
properties ” (“ Monograph of Christmas Island,” p. 107).

LONGICORNIA.

(14) CERESIUM SIMPLEX, Gyll.

Rare. Not found on Christmas Island, but two repre-
sentatives of the genus—C. quadrimaculatum, Gahan, and
C. nigrum, Gahan—were taken there by Dr. Andrews.

(15) CALocycLUS ANNULARIS, Fabr.

Only one specimen taken, flying in the sunshine, on
March 1, 1906. Does not occur on Christmas Island.

(16) Ruopica HONESTA, Pase.

(17) Ruopica Brnorata, Gahan.
Both unique specimens. The genus is not represented on
Christmas Island.

Seventeen species of beetles therefore occur on the Cocos-
Keeling group, and only one—a cosmopolitan species—is
definitely known to be also an inhabitant of Christmas Island,
although the coleopterous fauna of that island embraces ninety-
five species.

RHYNCHOTA.
By W. L. Disrant, F.ES.

Suborder HETEROPTERA.

Fam. PENTATOMID.
Subfam. CyDNIN2E.

(1) GEOTOMUS PYGMAUS.
thus pygmeus, Dall. List Hem. 1. p. 120 (1851).

APPENDIX I 361

Geotomus pygmeus, Sign. Ann. Soc, Ent. Fr, (6), lll. P. cle
t. ui. f 160 (1883); Dist. Faun. B. I, Rhynch. i. p. 98, f. 49
(1902).

Hab.— Widely distributed : recorded from Ceylon, Bombay,
Burma, Andaman Islands, and generally distributed through-
out the Malayan Archipelago; found in China and Japan,
and recorded from New Caledonia and Hawaii.

Subfam. PENTATOMINA,

(2) NEZARA VIRIDULA.

Cimex viridula, Linn. Syst. Nat. ed. 10, p. 444 (1758).

For full synonymy ¢f. Dist. Biol. Centr, - Am,, Rhynch. 1.
p- 78 (1880).

This species is distributed throughout the Palearctic,
Nearctic, and Ethiopian regions, and over a large portion of
the Neotropical, Oriental, and Australian regions.

Subfam. ASOPINA,

(3) CHCHALIA CONSOCIALIS,

Pentalatoma consocialis, Boisd. Voy. Astrol., Ent, 11. p. 630,
Uy Salyer (QESSua))E :

Cchalia consocialis, Stal, Enum. Hem. i. p. 59 (1870);
Schont. in Wytsm. Gen. Insect.,, Fase. 52,-p. 75, t. v- f. 12
(1907).

This species is recorded from, and not uncommon in,
Australia, New Zealand, and Tasmania. It is a well- known
species in Queensland.

Fam. ARADIDA,

(4) BRAcHYRHYNCHUS MEMBRANACEUS.

Aradus membranaceus, Fabr, Syst. Rhyng. p. 118 (1803).

Brachyrhynchus membranaceus, Stal, Hem. Fabr. i. p. 96
(1868); Dist. Faun. B. I, Rhyne. i. p. 160 (1904).

Found throughout British India, Malay Peninsula, and
Malayan Archipelago,

362 CORAL AND ATOLLS

Fam. REDUVIIDA,

Subfam. NABIDINA,

(5) NABIS CAPSIFORMIS,

Nabis capsiformis, Germ, in Selb. Rev. Ent. v. p. 132
(1837); Dist. Faun. B. I, Rhynch. ii. p. 400, f. 256 (1904).

Distributed in the Nearctic, Palearctic, Ethiopian, Oriental,
and Australian regions,

Fam. CapsipD&.
(6) Lyaus, sp. ?
Two specimens of a species of this widely distributed
genus.

[None of these species can be described as at all abundant
on the atoll, and only Nezara viridula is commonly met with ;
the others are, for the most part, very seldom found. The
whole of the order collected on Christmas Island by Dr.
Andrews was not worked over at the time of publication of his
Monograph ; but of the four species described by Mr. Kirby,
there is not one that is common to Christmas Island and the
Cocos: Keeling atoll. So far as I know, none of these species
has earned a distinct native name, but the domestic repre-
sentative of the family, which is common in native houses, is
called kutw basuk = stinking insect.—F. W. J.]

Suborder HomoprERa.

Fam. FULGORIDA.

Subfam. RIcaAnrInz.

(1) NoGopina BOHEMANI.

ficania bohemani, Stal, Freg. Kugen. Resa, p. 280 (1858).

Nogodina bohemani, Melich. Ann. Hofmus. Wien, xiii. p. 305,
t. xiv. f., 9a (1898).

Originally described from the Keeling Islands.

APPENDIX I 363

. [This is not a common insect, and only a few examples
were taken, all from the leaves of the Hibiscus trees on the
seaward side of Pulu Tikus.—F. W. J.]

NEUROPTERA.
Species determined by W. F. Kirsy, F.L.S., F.E.S.

ODONATA,
(Native family name, Kachapong.)

(1) PANTULA FLAVESCENS, Fabr.

A very abundant insect ; it flies about all over the lagoon,
and is present on all the islands. In April and May of 1906
the whole atoll swarmed with dragonflies, but for some time
previous to that it had been very rare to meet with a single
specimen. In the early months of 1905 these insects were
entirely absent from the atoll.

(2) TRAMEA ROSENBERGII, Brauer.

First seen on May 16, 1906, and during the following
week it became abundant; but for nearly a year previous to
this it had not been seen, and its numbers soon diminished
afterwards.

(3) ANAX GuTTaTus, Burm.

This species also came first to the atoll in May 1906, and
then only about a dozen examples were seen.

PLANIPENNIA,
(4) HemeERosius ? sp.
Native name, Lalar you = green fly.
This is a very abundant insect ; it occurs on all the islands,
and is certainly resident. When alive it is bright green, and
its eyes are remarkably brilliant. On account of its very

364 CORAL AND ATOLLS
offensive smell when crushed, it is very well known. It is

common at all seasons of the year.

(5) Isoprera.

Native name, Gegat.

One species of “ white ant” is very abundant in the wood-
work of dwelling-houses. It is noteworthy that the natives
never include this insect under the title of Samut—the family
name for ants.

ORTHOPTERA.
Species determined by W. F. Kirpy, F.L.S., F.E.S.

FORFICULIDA.

(1) ANISOLABIS ANNULIPES, Lue.

A common species found upon all the islands. It does not
occur on Christmas Island.

BLATTipé.
(Native generic name, Kerklak.)

(2) BLATELLA GERMANICA, Linn.
(3) ALLACTA NOCTULATA, Stal.
(4) ALLACTA OBTUSATA, Brinn.
(5) LOBOPTERA, sp.

(6) Moryrria, sp. (young larva).

(7) LEUCOPH@A SURINAMENSIS, Linn.

All these species are common, and are mostly found
beneath the bark of trees, or in the dwelling-houses. Only
Leucophea surinamensis occurs on Christmas Island.

(8) PERIPLANETA AMERICANA, Linn.
Common in store-houses ; introduced by ships.

APPENDIX I 365

ACHETIDA,

(9) GRYLLODES sIGILLATUS, Walk.
Native name, Orong orong.

Most commonly taken near to dwelling-houses; not very
abundant. | |

(10) ORNEBIUS, sp.

A very abundant species. It lives in the bushes of iron-
wocd (Cordia subcordata), and passes its early stages in a rolled-
up leaf.

Neither of these species occurs on Christmas Island.

(11) GryLacris, sp. near siynifera, Stal.

Native name, Chingkrek.

This species also passes its early stages in the leaves of
the Cordia. When adult it appears to be carnivorous ; it has
exceedingly powerful jaws, and is credited with waging war on
the large centipedes. When put in a box with a centipede
it certainly fights with great vigour, and though it bites the
centipede, the fights that I have arranged ended fatally for the
Gryllacris—but I believe that this is by no means invariable.
This species does not occur on Christmas Island, but the

allied G. rufovaria takes its place. The antenne are 150 mm.
long.

(12) PHIsIS PECTINATA, Guer.

A fairly common species, usually found in the Cordia
bushes. It is a bright green when living. It is not found
on Christmas Island, but P. /isteri, which occurs there, is very
nearly allied to it.

(13) CONOCEPHALOIDES SOBRINUS, Bol.

The male is buff-coloured, and the female green during
life. The species is abundant, and commonly lives in bushes.
It is very musical. Does not occur on Christmas Island.

366 CORAL AND ATOLLS

LocusTID&.
(Native name, Blalang blalang.)

(14) AcryprvuM, sp. near japonicum, Sauss.

Very abundant on all the islands. Feeds on the fronds of
the coconut palm. It varies greatly in size when adult, and
is the subject of great seasonal changes of abundance and
rarity.

It is not found on Christmas Island.

Although fourteen species of Orthoptera are found on the
Cocos-Keeling atoll, and twenty-three occur on Christmas
Island, there is only one species—Leucophea surinamensis—
which is common to both places. Of this species Dr. Andrews
took only a single specimen.

ARACHNIDA anp MYRIAPODA.
By A. S. Hirst, F.Z8.

SCORPIONS.
(1) IsomErRus MAcuLATUS, De Geer.
Native name, Klajingking.
Cosmopolitan.

SPIDERS,
(Native family name, Laba laba.)

(2) SMERINGOPUS ELONGATUS, Vinson.
Cosmopolitan.

(3) PHysocycLus GLoBosus, Taczaon Taczanowski.
Widely distributed in the tropical regions of the world.

(4) NEPHILA IMPERATRIX, C. K. (Plate xxi. Fig. 7.)
An adult female example which apparently belongs to this

APPENDIX I 367

species. It differs from Koch’s description of the species in
having the abdomen marked ventrally with two dark medium
patches, the anterior one being square and separated from the
smaller posterior patch bya whitish line. The British Museum
possesses examples of this form from Buitenzorg, Java, and
from Australia (Keyserling Coll.). 7

(5) HereropoDA VENATORIA, Linn. (Plate xx1. Fig. 8.)
Cosmopolitan,

(6) Lycosa, n. sp.
(7) Bavia, n. sp.

MILLEPEDES.
(Native name, Kaki ribu.)

(8) TRIGONIULUS, n. sp.
(9) ORTHOMORPHA COARCTATA, Saussure.

(10) There is also a large Centipede—native name Alipan
——no specimen of which reached England.—_F. W. J.

CRUSTACEA.
(Native family name, Kapeting.)

The collection of this order was not at all a representa-
tive one; such specimens as were brought home have been
determined by Dr. W. T. Calman, F.ZS.

BRACHYURA,

(1) CaRuUPA LAVIUSCULA.
(2) LEPTODIUS SANGUINEUS,
(3) LioxANTHO PUNCTATUS.

(4) LioMERA PUBESCENS.

368
(5)
(6)
(7)
(8)

CORAL AND ATOLLS

ACTA FOSSULATA ?
CARPILIUS MACULATUS.
EURUPPELLIA ANNULIPES.

PSEUDOZIUS CAYSTRUS.

(9) ERIPHIA LAVIMANA.

(10) MELIA TESSELLATA.

(11) GELAasIMUS, sp.

(12) CARDISOMA HIRTIPES.

(13)
(14)
(15)
(16)

(1)
(2)
(3)
(4)
(9)
(6)
(7)

(1)

GECARCOIDEA LALANDEI
LIOLOPHUS PLANISSIMUS,
SCHIZOPHRYS ASPERA.

CALAPPA HEPATICA.

ANOMURA.

REMIPES TESTUDINARIUS.
BIRGUS LATRO.
C@:NOBITA CLYPEATUS.
Ca@:NOBITA RUGOSUS.
CG@NOBITA PERLATUS.
CALCINUS HERBSTII.

CLIBANARIUS CORALLINUS.

CARIDEA.
ALPHEUS STRENUUS,

APPENDIX I 369

STOMATOPODA.

(1) GoNODACTYLUS CHIRAGRA.

VERMES.

There is one species of earthworm found in all the islands,
but unfortunately no specimen reached England.

2A

Bok PB Nap Iexee

LIST OF THE FLORA

THE following list of the flora of the atoll is based upon that
compiled by Dr. H. O. Forbes from his own observations and
from those made previously by Darwin.

To this list I have added several species, and I do not
doubt that it still remains far from complete. Where it is
possible, I have added the native names of the plants, and
noted their distribution and the use to which they are put by
the natives. For the naming of the Fungi, which (with the
exception of Polyporus luridus, which I did not myself observe)
are all additions to our knowledge of the flora, I am indebted
to Miss A. Lorrain-Smith and Mr. A. Gepp, of the British
Museum.

DICOTYLEDONS.

ANONACES.

(1) Anona RETICULATA, Linn.

CRUCIFERA.

(2) Srnapis guncEA, Linn.
Distribution: Aru Islands, &e.

CAPPARIDEA,

(3) GYNANDROPSIS PENTAPHYLLA,

PEDICELLARIA PENTAPHYLLA, Schrank,
371

372 CORAL AND ATOLLS

GUTTIFERA.

(4) CALOPHYLLUM INOPHYLLUM, Linn.
Native name, NVyamplong.

Distribution: Widely distributed in tropics; wood useful
for furniture.

MALVACEA.
(5) Hipiscus TiLtaceus, Linn.
Native name, Warw.
Distribution: Common sea-coast tropical tree; fibre from
bark used in islands for cordage, &c.
(6) Hipiscus RosA-sINENSIS, Linn.
Introduced. .

(7) SUDA CARPINIFOLIA.

TILIACE.
(8) TRIUMFETTA PROCUMBENS, Forst.

LEGUMINOS&.
(9) ACACIA FARNESIANA.
Distribution: Timor, &e.

(10) PoIncIANA PULCHERRIMA, Linn.
Introduced.

(11) GureLanpina Bunpoc, Ait.
Native name, Klenchi.
Distribution: Timor.

ROSACEA.
(12) ErIoporrya, sp.
Cultivated.

(138) Rosa cENTIFOLIA, Linn.
Cultivated.

PLATE XXII.

NATIVES CLEARING AWAY THE GROWTH OF KLENCHI (GUIELANDINA
BUNDOC) ON PULU TIKUS.

APPENDIX II 373

MyYRTACEA.

(14) BarrIneGToniA RACEMOSA, Blume.

Native name, Jambu basagi.
Distribution : Malaya and Polynesia.

LYTHRACES.

(15) PEMPHIS ACIDULA, Forst.

Native name, Kayu burung.
Distribution : Tropical coasts of Old World.

PAPAYACER.

(16) Carica papaya, Linn.
Native name, Katis. Useful fruit.

CRASSALACEZ.
(17) BryopHyLLUM CALCINUM, Salisb.

PORTULACEA.

(18) PortuLaca OLERACEA, Linn.
Distribution : Timor and Lombok.

RUBIACE.

(19) GurETTaRDA speciosa, Linn.
Native name, Melati. Tree with sweet-scented flowers:

used by women to decorate their hair.
Distribution : Tropical shores of Old World.

(20) Morinpa cirriro1iA, Linn.
Native name, Ingkudu. <A useful dye bark.
Distribution: India, Australia, Polynesia, Malay Archi-

pelago.

3

4 CORAL AND ATOLLS

CoMPOSIT&.

(21) AGERATUM CONYZOIDES, Cass.
Distribution: All hot countries.

(22) SoNCHUS OLERACEUS, Linn.
Distribution: Java, Sumbawa, &e.

~ APOCYNACER.
(23) Vinca RosEA, Linn. ts

(24) ORCHROSIA PARVIFLORA, Hensl.

GOODENOVIEA.

(25) Scavoia Konic, Vahl.
Native name, Kembang kangkong.
Distribution: Tropical East Asia, Australia, Polynesia.

SAPOTACER.

(26) SIDEROXYLON suNDAICUM, Burch.

Native name, Saoh. A useful fruit.
Distribution: Sunda Islands.

ASCLEPIADIACEA.

(27) ASCLEPIAS CURASSAVICA, Linn.

Native name, Ujan mas.
Distribution: Java, Xe.

BIGNONIACEA,
(28) OroxyLum INbDIcuM, Vent.
Cultivated.

BoRAGINES.

(29) CorpbIA suBcoRDATA, Lam.

APPENDIX II 375

Native name, Grongang. A useful hard wood used for

boat- building.

Distribution: S.E. Asia, Australia, Sandwich Islands,

Tropical Africa.

(30) ToURNEFORTIA ARGENTIA, Linn.
Distribution: Ceylon, Australia, Malaya, Mauritius.

CONVOLVULACER.
(31) Iroma@a PES-cAPR&, Roth.
Distribution: Tropical sea-shores.

(32) Ipoma@a GRANDIFLORA, Lam.
Distribution: Widely spread in trupics.

SOLANACES:.

(33) PHYSALIS PERUVIANA, Linn.

-\(84) PHYSALIS ALKEKENGI.

ACANTHACEA.

(35) ASYSTASIA COROMANDELIANA, Nees.
Distribution: India, Malaya, Africa, Arabia.

(36) DictipreraA BuRMANNI, Nees.
Distribution: Java, Timor, &c.

LABIATA,
(37) Lronurus srerricus, Linn.

VERBENACEA.

(38) STACHYTARPHETA INDICA, Linn,
Distribution: Tropical Asia,

376 CORAL AND ATOLLS

NYCTAGINE.

(39) PIsONIA GRANDIS.
Native name, Ampol.
Distribution: North Australia, Polynesia.

(40) PISONIA EXCELSA, Bl.

Native name, Jumbu. <A useful fruit.
Distribution: Malay Islands.

(41) BorRHAviA birFusa, W., var.

AMARANTACEA.
(42) AMARANTUS SCHOFLEXUS.

(43) ACHYRANTHES ARGENTEA, Lam., var. villosior. —

LAURINES.

(44) HeERNANDIA PELTATA, Meissn.

Native name, K’mador. <A tall forest tree of which not
many specimens exist.
Distribution : India, Malaya, Polynesia.

EUPHORBIACEA.

(45) EuPHoRBIA PILULIFERA, Linn.

Native name, Cheplok. Produces an edible berry used in
the islands. |
Distribution: Trepics and sub-tropics.

(46) ACALYPHA INDICA.
(47) Ricinus communis, Linn.

(48) ALEURITES MOLUCCANA.

_ URTICACER.

(49) URERA GAUDICHANDIANA, Hensl.

APPENDIX II 377

MONOCOTYLEDONS.

GRAMINEA.

(50) PANICUM SANGUINALE, Linn., var.
Distribution : Timor, &c.

(51) STENOTAPHRUM LEPTUROIDE, Hensl.

(52) LEPTURUS REPENS, Forst.
Distribution: Timor, &c.

(53) ERAGROSTIS AMABILIS, Linn.
Distribution: Timor, &e.

(54) FIMBRISTYLUS GLOMERATUS, Nees.

PALMACES. —

(55) Cocos NUCIFERA.

Native name, K/apa. Exists in many-varieties.

A great number of useful articles are made from the coco-
nut palm and its products, and in the atoll full advantage is
taken of all the many uses to which the nuts can be put.

From the husk of the nut (saddt) fibre is taken, but in the
atoll no coir is manufactured, and the husks are used to spread
on the ground and assist in building up the superficial layer
of vegetable soil. The shell of the nut is used for fuel, and
water-vessels (datok) and ladles (gayong) are skilfully made
from cleaned nuts. From the flesh of the nut itself the copra
is prepared by simple sun-drying, and the copra is used in
the preparation of all the grades of coconut oil (minyah klapa).
Soap (sabon) is also made from the oil by admixture of the
ashes of burned husks.

The grated nut (ampas) is used for a variety of culinary
processes, and the expressed juice is used as a very nutritive
drink (santan).

The nut that has already begun to sprout, or is soon about

378° CORAL AND ATOLLS

to do so, provides another excellent article of diet (tombong),
which consists of a sponge-like mass that occupies the whole of
the interior of the mature nut.

An alcoholic drink (¢wak) is prepared from the juice of the
wounded spathe (manchong), and when the twak has turned
sour it makes an excellent vinegar (chukak klapa).

The central. shoot of the palm, while still white and
brittle (wmbit), makes good salad and an extremely pleasant
pickle.

The palm itself furnishes the materials for many domestic
purposes. Baskets (Kroso) are made from its interlaced leaves ;
thatch (atap) for house roofs is prepared by tying the leaves
into bundles, and the midrib of the leaf (plépa) when split
makes the well-known house wall called dinding. The wood
of the trunk is used at times for certain parts of houses, and
the central wood (nibong), which is stronger and tougher than
the outer parts, makes useful walking-sticks.

PANDANACEA.

(56) PANDANUS, sp. Though not observed by Darwin or
Forbes, is now very abundant on Pulu Selma and the larger
Southern islands—one island being actually named Pulu
Pandan.

The fibre of the leaves is used to make matting and
basket-work.

Musct.
(57) HypNnuM RUFESCENS, Hook.

FUNAGI.
(58) POLYPORUS LURIDUS.

(59) MARASMIUS, sp.

(60) Panus connatus, Berk.? This fungus grows generally
upon the ampol tree, and is good to eat.

>» APPENDIX Al 379
(61) HIiRNEOLA AURICULE-JUD&, Berk.

Distribution: Europe, W. America, Mexico, Cuba, Tas-
mania, Borneo, and Christmas Island.

(62) AURICULARIA MESENTEBICA, Fries.
(63) LenziTES NIVEA, Cooke.

(64) PoLysTICcHUS RIGENS, Sace et Cub.
(65) PortA HOMAMA, Sace.

(66) STEREUM, sp.

PLATE XXIII.

THE SPATHE OF THE COCONUT PALM.

eps; 4
tal Rae

PLATE XXIV.

THE FLOWER SPIKE OF THE COCONUT PALM AFTER THE SPATHE
HAS BEKN OPENED. SOME NUTS ARE ALREADY FORMED ON THE
BASKS OF THE STEMS, THIS STAGE IS CALLED MANGGAR.

es.

PLATE XXV.

sea
ivy

YOUNG NUTS DEVELOPING ON THE FLOWERING STEMS WHEN THE
INFERTILE FLOWERS HAVE BEEN SHED. WHEN IN THIS STAGE
THE NUT IS KNOWN BY THW NATIVE NAME Or BLULUK.

bit
41

Aa
tay pe

PLATE XXVI.

YOUNG COCONUTS IN THE STAGE IN WHICH THE CONTAINED NUT IS
BEGINNING TO BECOME HOLLOWED—STAGE CALLED KLAPA CHINKER.

Asie

PLATE XXVII.

god

GREEN COCONUTS IN THE STAGE IN WHICH BOTH HUSK AND NUT
MAY BE CUT WITH A KNIFE, AND THE CAVITY OF THE NUT
ALREADY CONTAINS “MILK,” KLAPA MUDA.

BIBLIOGRAPHY OF SOME OF THE MORE
IMPORTANT WORKS REFERRED
TO IN THE TEXT

(A) Retatine To Cocos-KEELING AND THE CLUNIES-Ross
FAMILY.

1. Clifford, Hugh, ‘Heroes of Exile: The Romance of a Scots
Family.” London, 1906.
2. Darwin, “ A Naturalist’s Voyage Round the World,” chapter xx.

3. Denys, “A Dictionary of British Malaya,” article on Cocos-
Keeling.

4, Guppy, Dr. H. B., “The Cocos-Keeling Islands,” Scottish Geo-
graphical Magazine, vol. v., 1889, Nos. 6, 9, and 11.

5. Forbes, Dr. H. O., ‘‘A Naturalist’s Wanderings in the Eastern
Archipelago.” London, 1885.

6. Straits Settlements: Papers relating to Cocos-Keeling and
Christmas Islands, 1897 (‘‘ Parliamentary Papers,” C. 8367), and

7. “Colonial Reports” annuals since 1897.

(B) Retatineg tro NEicubourine Cora JsLanDs IN THE
INDIAN OCEAN,

8. Alcock, A., “ A Naturalist in Indian Seas.” London, 1902.

9. Andrews, Dr. C. W., ‘A Monograph of Christmas Island.”
British Museum publication. 1900.

10. Andrews, Dr. C, W., Proc. Zool, Soc., 1909, p. 101.
381

382 CORAL AND ATOLLS

11. Hickson, Prof. 8. J., “A Naturalist in North Celebes.”
London, 1889.

12. Wallace, A. R., “‘The Malay Archipelago.” London, 1869.

(C) RELATING TO THE CorAL PRoBLEM—T'HE RATE oF GROWTH.

13. Dana, “‘ Corals and Coral Islands,” 1875, p. 96 e¢ seq.

14, Darwin, “On the Structure and Distribution of Coral Reefs,”
p. 96 et seg.

15. Gardiner, 8., ‘‘The Maldives and Laccadives,” 1908, vol. i,
Part III. p. 327.

16. Wood Jones, F., “ The HENS of Growth of the Reef-building
Corals.” London, 1908.

(D) RELATING TO THE VARIABILITY OF CoRAL SPECIES.

ie Daerden: J. E., National Academy of Sciences, vol vill. 7th
Memoir. Washington.

18: Hickson, Prof. 8. J., “On Millepora,” Proc. Zool. Soc.,
April 1898. ee

19. Hickson, Prof. 8. J., Proc. Zool. Soc., Noy. 1898, and other
papers by the same author.

(E) RELATING TO ATOLLS AND ATOLL FORMATION,

20. Agassiz, A., Memoirs, American Academy of Arts and ae
vol. xi. Part II. No. 1, 1885, p. 107.

21, Agassiz, A., Bull. Mus. Compan. Zool. Hop vard, vol. xiv., 1888,
22. Agassiz, A., Roy. Soc., 1908. Mature, 1903, p. 547.
23, Basset-Smith, Ann. and Mag. Nat. Hist., 1890, p. 353.
24, Bourne, G., “ Atoll of Diego Garcia,” Proc, Roy. Soc,, 1888.
(25, British Association, ‘‘ Discussion on Coral Reefs,” 1893.
26. Dana, ‘‘Corals and Coral Islands,” 1875.
27, Darwin, “ On the Structure and Distribution of Coral Reefs,”

BIBLIOGRAPHY 383

28, ‘Funafuti, the Atoll of,” Roy. Soc, 1904. Professors Judd,
Solas, David, Sweet, Sorby, &c,

29, Gardiner, S., “The Fauna and Geography of the Maldive and
Laccadive Archipelagoes,” vol. i. Part ITI., 1902,

30. Gardiner, 8., Vature, 1904, p. 371.

31. Guppy, Dr. H. B., “Coral Reefs of the Solomon Islands,” Proc.
Roy. Soc. Hdin., July 1886, p. 857.

32. Guppy, Dr. H. B., Proc. Linn. Soc. V.S.W., 1884, vol. ix.
Part 4. ;

33. Guppy, Dr, H. B., various short articles in Vature, 1888, &e.
34, Hedley, C., “* Coral Reefs of the Great Barrier Reef.” 1907.

35. Hedley, C., “The Broadening of Atoll-Islets,” Vat. Science,
vol. xii. No. 73, 1898, p. 174.

36. Hedley, C., Vature, August 4, 1904, p. 319, and other papers.

37. Langenbeck, Dr. Von R., Die Theoriem wiber die Entstehung der
Koralleninseln und Korallenriffe und thre Bedeutung fiir Geophysische
Fragen. 1890.

38. Ledenfeld, R. Von, see Vature, May 8, 1890.
39. Mann, W., Korallen und andere gefteinsbildende Tiere. 1909.

40. Murray, Sir John, “The Structure and Origin of Coral Reefs
and Islands,” Proc. Roy. Soc. Hdin., April 5, 1880.

41, Murray, Sir John, and Irvine, Nature, J une 12, 1890.
42, Reade, T’. Mellard, Vatwre, March 22, 1888.

43. Semper, Prof., “Animal Life.” English trans. International
Scientific Series, vol. xxi.

44, Wharton, Admiral Sir J. A., Vature, June 19, 1890.
45, Wharton, Admiral Sir J. A., Vatwre, Feb. 23, 1888.

INDEX.

Axsgort’s booby, 800, 3438
Abu bunchahan, 54
Agar-agar, 122
Agassiz, Prof. A., Darwin’s letters
to, 217
on formation of submarine
banks, 225
on volcanic basis of atolls, 216
Age of coral zooids, 105
Alcs yonacee, 58
Aldrich, Captain, visit. to Christmas
Island, 42
Alexander Clunies, 7
Algee, effects of, on coral growth,
113, 122, 125, 194
effects of, on silting of lagoon,
194
growth of, in lagoon, 189
Andrew Ross, 38, 42, 286
Aunexation of islands to Ceylon, 37
to Great Britain, 28
to the Straits Settlements, 40
Ants, 306, 355
Asexual reproduction of corals, 71
Atoll, description of, 135
Atollons, formation of, 269
Atolls, *‘ miniature,” in lagoon,
Australia, cable route to, 147
Australian horse ships, visits of, 24
Ayesha, wife of George Clunies-
Ross, 44

252

Banchahan, 54.

Bands, effects of, when placed round
coral branches, DG :

Banks, submarine, 221

Baracouta, 322, 547
Barometric readings, 208
Barrier, formation of, 253
outer edge of, 155
slope of, 156
structure of, 151
windward, 163
Barrier pools, coral growth in, 91
124
Barrier reefs, formation of, 278
Basin-shaped reefs, 251
Basset-Smith on bathymetrical
range of corals, 239
on deep-water corals, 245
Bathymetrical limits of corals, 239
caused by sedimentation, 242
Bats, 312, 337
Batu Dua, 195
Batu Satu, 195
Beaches, lagoon, 172
seaward, 168
Beagle, visit of, to atoll, 20
voyage of, 216
Béche-de-mer, 330
as sand-producers, 165
Beetles, 306, 357
Best man, at wedding ceremony, 52
Betrothal, formalities of, 49
Bibliography, 381
Birch, Mr. E. W., visit to atoll, 37
Birds, introduced, 299
species of, 3 38
tameness of, 300
Birgus latro, 316, 368
Birth, customs connected with, 46
Blora, visit of, to atoll, 16

385 2B

386

Boat-building in Cocos-Keeling, 38
Booby bird, 300, 343
Boring animals, effects of, on corals,
126
Borneo, Ross’s ship, 12
Bottles used to determine currents,
293
Bourne, G. C., on currents, 232
Breccia, formation of, 254
in lagoon, 173
sites of formation of, 161
Breccia platform, definition of, 154
evidences afforded by, 279
raising of, 282
Brekat, 53
Bridegroom, costume of, 50
British Christmas Island Phosphate
Co., 43
Broadening of reefs, 277
Buchanan, J. Y., on deep-water
corals, 245
Buntal (fish), 828, 347
Burial customs, 54
Burung chinta (Zosterops natalis),
299, 344
Burung glatek
299, 344
Butterflies, species of, 349
Button Islands, 262

(Java sparrow),

CaBLE ships, soundings made by,
145

Cable, submarine, 44

Calappa hepatica, 318, 368

Calm water, influence on coral
growth, 90

Calms, in region of atolls, 209

Capel, Admiral, visit of, 8

Cardiosoma, 176, 178, 315, 368

Carpophaga Whartoni, 43, 299, 338

Cats, introduction of, 298

Centipedes, 305, 367

Ceremonies connected with mar-
riage, 50

Challenger, voyage of, 227

Chamisso on atolls, 232

INDEX

Changes in ideas regarding atolls,
214
of lagoon shores, 198
of ocean-level, 199
periodic, in atoll-building, 169
Charts, early, 5
Chinese blood in natives, 21
Christmas Island, settlement of,
Al
Cilia of coral embryos, 64
Circumcision, rite of, 48
Climate of atoll, 201
Clouds over atolls, 204, 211
Cockroaches, 306, 364
Coconut, introduction of, 17
uses of, 377
Coconut oil, 35
Canopsammia, 85
Coleoptera, species of, 357
Colonising routes for fauna, 288
Coloration of corals, 93, 96
of fish, 327
Comet, H.M.S., visit of, 16
Commerce as distributer of fauna,
502
Conglomerate of beaches, 175
Corallites, form of, 76
in rough-water types, 90
Corallum, alteration of, in repair,
118
Corals, growth of, in lagoon, 190
living, 57
Corynetes rufipes, 306, 358
Crabs as producers of vegetable soil,
178, 269
species of, 314, 367
Crescent shape of islets, 177
Crickets, 306, 365
Crocodiles, 295
Crustacea, species of, 367
Currents, alteration of, 163
as carriers of fauna and flora,
293
in lagoon entrance, 198
Currents, oceanic, rate of, 295
Cyclone of 1862, 29

INDEX

Cyclone of 1876, 35
Cyclones, 205

Dana on “ Subsidence,” 234
Darkness, corals active
241
Darwin on atoll formation, 216
on fauna, 287
visit of, 20, 137
Dead corals in lagoon, 193
Death, ceremonies connected with,
54
Death-processes in corals, 121
Deep water, influence on corals,
92
Deep-water corals, pigmentation of,
97
Deer, introduction of, 297, 337
Deformities, superstition regarding,
46
Demiegretia sacra, 313, 342
Dermestes felinus, 306, 358
Desertion of Hare’s followers, 18
Destruction of corals in lagoon, 122,
192
Devi!-fish, 323
Dew, absence of, 211
Nicerobatis eregoodoo, 325, 347
Dicotyledons, species of, 371
Diptera, species of, 356
Discovery of the islands, 5
Distribution of corals, 68
in the atoll, 157
Doldrums, 201
Domestic fowls, feral, 300
Dominant apical zooid, 82
Dragonflies, 205, 307, 363
Dredging corals, 191
‘“¢ Drowned atolls,” 230
Duerden, J. E., on coral embryos,
63, 67
on nocturnal habits of corals,
241
Dutch navigators, visits of, 24
Dymoke, Elizabeth, 13
Dymoke Shoal, 195

during,

387
Earrn, carried on tree-trunks, 290
disappearance of, on coral

débris, 180

Earthworms, 179, 369

Eels, 329

Egeria, H.M.S., visit to Christmas
Island, 42

Embryos of corals, 63

Environment, changes of, 116

influence of, on corals, 79, 84

Espoir, H.M.S., visit of, 37, 40

Evil spirits as a cause of infantile
death, 47

Experiments on change of environ-
ment, 123

Exposure of corals above tide-level,
127

Fauvacies of Gardiner’s theories,
224

of Murray’s theory, 226
Fauna of the atoll, 286, 335
Feasts, ceremonial, 53
Feeding of corals, 241
Fiddler crabs, 176, 315
Fish eating coral, 130

of the atoll, 319

species of, 317
Fish-shooting on the barrier, 325
Fitzroy, Admiral, soundings by, 145
Fixation of coral larve, 68
Flat-topped Porites colonies, 247
Flies, species of, 356
Flinders on atoll formation, 213
Floating corals, 171, 291

islands, 288
Flora, list of, 571

of the group, 286
Forbes, Dr. H. O., on fauna, 287

on lagoon, 192

visit of, 36, 137
Foul water, influence on corals, 122
Fremantle, Captain, visit of, 28
Fresh water, presence of, 183
Freshening of salt water, 129
Frigate birds, 300, 542

388

Fringing reefs, formation of, 279
Fruit pigeon, 299, 338
Funafuti atoll, investigation of, 234
Funerals in the atoll, 55
Fungi, list of species of, 378

of the islands, 295.
Fungia, observations on, 243

GARDEN, arrival of insects in, 310
Gardiner, 8.,on atoll formation, 222
on bathymetrical limits of
corals, 241
on Lithothamnionide, 157
on “ reversed atolls,” 272
Gecarcoidea lalandei, 316, 368
Geckos, 307, 545
Gelasimus, 176, 815, 568
George Clunies-Ross, 32
Glendinning Shoal, 149
Globigerina ooze, 149
Governor, appointment of Ross I1.,
appointment of Ross III., 34 .
Gradient of submarine slope, 150
Graham’s Island, destruction of, 225
Grass, influence of, on reclamation
of land, 177, 187
Graves in Cocos- Keeling, 55
Green fish, 824, 547
Ground thrush, 299, 344
Growth forms of corals, 71
Guppy, Dr., references to work of, 3,
Toi, LOO; aia Lo? 208)
225, 278
Gygis candida, 3389

Hacug, J. H., on Jarvis Island, 210

Hair, superstitions regarding, Bd

Hare, Alexander, 10

Hare, John, 12

Hedley, C., on Great Barrier Reef,
235

Height of island-building, 263

Hermit crabs, 517

Hesse, Mr. F., 145

Hickson, Prof. 8., on currents, 233

INDEX

Hippomenes, 14

Histiophorus gladius, 322, 347

Holothurians as sand-pr oducers, 165,
265

Homoptera, species of, 362

Horsburgh, 4

Horseshoe islands, 269 .

Hymenoptera, species of, 350

Inin, wife of Ross IIT., 34
- death of, 44

Insects flying out to sea, 309
Introduced fauna, 297
Islands, growth of, 259

origin of, 257

sieneunne of, 176
*“ Itceby corals,” 153

J. G. C.-Ross, ship, 37
loss of, 39

Samper, 5d

Jarvis Island, 210

Java, cable route to, 148

Java sparrow, 299, 344

Jelly-fish, 331

Jetsam of reef, 290

John Clunies- Ross, 8

John George Clunies-Ross, 27

Juno, H.M.S., visit of, 28

Kaxatua, 150, 164, 264, 825
Kapeting, 176, 178, 315, 367
Karang gatal, 133
Keeling, Captain William, 3
Keeling Island, 141

development of, 271
Keyser, Mr. A., visit of, 129
Krakatua, eruption of, 290

pumice from, 170
Kramer on atolls, 256
Kunyet, 50

Lapps, Captain, 136

Lagoon, beaches of, 172
description of, 185
silting of, 186

INDEX

Lagoonlets of atoll islands, 176
Land crabs, 316
movements, 279

Larval forms of corals, 63

Le Conte on barrier reefs, 278

Legal code of Cocos-Keeling, 380

Legend concerning spirit voices,
285

Leisk, 22

Lendenfeld, R. von, on solution
theory, 228

Lepidoptera, species of, 349

Leptograpsus, crabs of genus, 318

Level of ocean, variations of, 199

Lifeless areas of lagoon, 193

Light, influence on bathymetrical
range of corals, 241

Lime, coral turned into, 191

“ Limiting line of sedimentation,”
238

Lithothamnionide, 157, 270

Littoral fauna, 314

Living corals, study of, 135

Lowness of islets, 139, 182

Luigi Raffo, ship, 39

Mactear Deep, 148
Madrepora, repair of damage in,
JOEL

Maniciana, embryos of, 67

Marine fauna, 314

Marriages, 49

Mas kawen, 49

Meandrine fission, 119, 74

Measurements of growing corals,

69

Melia tessellata, 319

Merula erythropleura, 299, 344

Meteorological records, 201

Mice, 305, 337

Migrant birds, 301

Millepore, 132, 179

Mollusca, boring in corals, 110, 126
species of, 348

Monkey, arrival of, in atoll, 298

Monocotyledons, species of, 377

389

Montipora, repair of damage in,
118
Mosquitoes, 357
breeding in salt water, 303
Moths, species of, 350
Motile coral larvee, 64
Mourning customs, 56
Murray, Sir John, 150, 221
Mus Macleari, 43
Musical rocks, 285

Name, customs of change of, 48
Native customs, 45

rising, 33
Negro element in natives, 21
“ Negro heads,” 284
Neh Basir, 20
Neuroptera, species of, 363
Nocturnal habits of corals, 241
Northerly winds, 204
Northern part of lagoon, 195
Nubia, sand formation in, 267
Nulliporce, importance of, 156

Oxpsect of original settlement, 15
Observations on living corals, 60
Obstruction to currents, autinenice
of, 257
Ocean bottom, 149
level, alterations of, 199, 282
Octopuses, 331
Ocypoda, crabs of genus, 314
Ooze, deposition on ocean bottom,
221
Ophichthys colubrinus, 329
Orbulina ooze, 149
Origin of fauna, 313
of marriage ceremonies, 51
Original settlement, 14
Orthoptera, species of, 364 |
Outer edge of barrier, 160
Outward growth of atolls, 277
of reefs, 263

Papaun blood in natives, 21
Pareh, 323, 347

390

Pergdm, 43, 299, 338
Petition of Ross I., 23
Philippine land-rail, 312, 538
Phosphates, formation of, 210, 180
in Christmas Island, 42
Physalia, 331
Piazza, Captain, 294
Pig-fish, 327, 347
Pigs, introduction of, 297
Plapen, Catherine, 8
Plate forms of coral growth, 80, 86
Pocillopora, colonies on floating log,
100
Poisoning of lagoon corals, 192
Poisonous fish, 328
Porites colonies as miniature atolls,
247
manner of growth, 106
Portuguese man-o’-war, 331
Prevalent winds, 202
Problems involved in atoll-building,
ZAG
Pulu Atas, settlement of, 19
Pulu Bras, Hare’s residence on, 19
Pulu Selma, settlement of, 19
Pumice, 170, 290

QUEENSLAND bean,
beaches, 296

Question of basis on which atolls are
built, 217

cast up on

Rasstits, introduction of, 297
Radiolarian ooze, 150
Raftles, Sir Stamford, 10
Rain from cloudless sky, 206
influence on corals, 129
Rain-clouds, splitting of, 210
Rainfall, influence of, 180
of atoll, 206
Raised beaches, evidences of, 283
Rapid growth of corals, 196
Rate of growth of corals, 69
Rats, black, 305, 337
damage done by, 30+
description of, 335

INDEX

Rats, introduction of, 303

Raymond, Joseph, 22

Reade, T. Mellard, 229

Reefs, formation of, 245

Religion of islanders, 45

Reproduction of corals, 61

Reptilia, species of, 345

Resins, cast on beaches, 171

** Reversed atolls,” production of,
271

Rhynchota, species of, 360

Robber crab, 316

Robert Portner, wreck of, 304

Rock cod, 329

Rocks cast on barrier, 284

producing sounds, 285

Rodriquez, cable route to, 148

Ross I., death of, 25

Ross II., death of, 31

Rough water, influence on corals,
86

Rujak, 50

SAMBAR, deer, 297, 337
fish, 322, 347
Sand flats, 186
influence of, in lagoon, 188
influence of, on land formation,
264
islands, 181
produced by fish, 164
Sand-kanks, development of, 257
Sandstone, beach formation, 175
Scart as sand-producers, 164, 264
habits of, 130
Scorpions, 305, 566
Sea-level, alterations of, 199
Sea-slugs as sand-producers, 165
Sediment, deposition in lagoon, 226,
228
effects of, on corals, 123, 136
effects of, on type of growth, 94
Sedimentation banks, 238
in open ocean, 237
influence on corals and reefs,
249

INDEX

Seeds cast up by waves, 171, 296
Semper, Prof., Darwin’s letters to,
217
on ‘‘ Subsidence,” 234
Serpent, H.MLS., visit of, 29
Settlers, original, 16
Sharks, 320, 347
Sheep, introduction of, 297
Ships as carriers of fauna, 296
fauna of, 302
insects flying aboard, 311
Shoaling of lagoon, 187, 275
Shores of lagoon, 172
Showers, influence of, 183
Silt cast off by corals, 248
effects of, on corals, 123, 189
Silting of lagoon, 125
Skada bumi, 53
Slamatan, 53
Slope of lagoon reef, 197
Slopes, submarine, 150
Slowness of coral growth, 193
Snakes, arrival of, in atoll, 295
Solitary corals, 59
repair of, 104
Solution, theory of, 222
Somaliland, currents to, 294
Soundings of Indian Ocean, 145
Spawning of corals, 62
Specific distinctions of corals, 83
Specific gravity of sea-water, 129
Spherical growth of corals, 72
S’pia Dupong, wife of Ross II., 31
Spicer, Rev. E. C., visit of, 215
Spiders, 306, 366
Spirits, belief in, 45
offerings to, 53
Stinging corals, 133
Stones carried on floating
290
Storms, influence of, 261, 263, 283
Submarine basis of atoll, 144
slope of atoll, 146
Subsidence, absence of signs of, 280
theory of, 216, 218
Surf, influence of, on corals, 158

logs,

391

Sword-fish, 322, 347
Symmetrical zooids, 82
Sympathy between zooids, 106

Tauzor, Mr. A. P., visit of, 40
Talus slopes, extension of, 277
Tameness of birds, 300
Telegraph ships, soundings by, 145
station, 44
Temperature of islands, 207
‘of lagoon, 200°
Tendencies of coral growth, 86
Tern, the white, 339
Tetradonts, 327, 347
Theories of atoll formation, 212
Thunder and lightning, 211
Ticks left by sheep, 297
Tide excursion, influence on height
of islands, 262
variation of, 199
Tomboro, eruption of, 201
Tombstones, form of, 55
Trade winds, 201
influence on atoll form, 256
Tiee-trunks, floating, 289
Triangular islands, 4, 138
Turk Reef, 195
Types of coral growth, alterations of,
115
distribution of, 158
in lagoon, 191
Typhlops braminus, 290, 346

Udung darat, 316

“ Undecided” forms of corals, 131

Unreliability of signs of change in
atolls, 199

Upward land movements, 281

VANDERWADL, 11
Variability of corals, 101
Vegetable soil, 179
Vegetation, influence of, on land
formation, 267
of islands, 179, 371
of lagoonlets, 177

392

Vegetative forms of corals, 71
Voleanic origin of atolls, 215

WATER-BORNE flora and fauna, 288
Waves as sand-producers, 166
depth at which felt, 238
force of, 170
Wedding ceremonies, 49
Wells of fresh water, 182, 210
Westerly winds, effects of, 196, 198
Wet days, 207
Whalers, visits of, 22
Wharton, Admiral, on soundings,
144
on submarine banks, 225
White tern, 339

INDEX

Wilson on coral larve, 67-
Wind, influence on land formation,
180, 261, 266

prevalent, 202
Wind-borne fauna, 307
Worms, boring in corals, 110
Wreck of ship Robert Portner, 304
Wrecking of settlement, 36
Writings of Ross I., 24

Zephyr, H.M.S., visit of, 37, 40
Zoantharia, 58 :
Zooanthelle in coral larvee, 66
Zooid, *‘ dominant,” 82
Zooids, methods of division of, 71
Zosterops natalis, 299, 344 é

Printed by BALLANTYNE & Co. LIMITED
Tavistock Street, Covent Garden, London

Pea eRe en
Hy Wasi !
ray

ny fh aie
Psa

Ty pete Wye

as

i r
be

Ht ‘ sy
AS a Wi
BUN chy
a

TT

#