PLATE 1

-'[

1

n\A

sg

3

,j\1

\ ■ V

<l)

/ / ■ <?

3

E

cr

3

"/ \CA

nj

7 \^

*o

5 "J

'•. ••"/

/ 0 •,

; 1 ;a /

/ §■ \

2 \\
- \"

^ **

t\

"i"> i§ /

&. N

.* / "»/

-1. \ '

o

y

_.-z.'.°>-L

-

._

_ . — ■ ■'"

i ; /2

\

.-^~~~ "

1 1 /«

/ j

s

b -— '

*■ »•' f/ §

/

0

N.

/ °

•^ • .

o

tf

X.

\ 1 1 :r

•v.

—

*•

\ 1 / /

* ! / I' ^

^

NN

Q.

3

V

^ O

\ \\l -5

^v^^ O

* I 1 // ,^o

\ C

0 \ ■ 1 //

3°

— =i— __

a

d) 3

/ 1

-0

O

.ye

/ /. ... _

(. 0
u <

'.« •' /

o

c — .
'5 .**>'

/

m

TJ

X

/

<!> C

o

or

o

/

0/ /

->: C

0:0 .

/
//

/ c
0

5/ w
/ f/W

b

/

his ° fa

^ .7 "

e

/

0 'O

• 0

3

" • _r

3

5-5 <?/

>-f

o

• .3
• O

0 / )

<• /

7 /

(*r <■

a)

C

* " O

0 /

;/

_)

' / O I'

<

*o

' / /■

c

r-^y

<

°o

»V# * !

c

Q-

«S

Si*; Sfc

*?

H

5- C2*

vk *

\ s

OS

b

<#*

*Y \\

Ml

P

-'•.?

\ \ 0

V-^^-^

o

r-^ '.•'(<«/

c

o

o

w

c

The Polynesian Wanderings

Tracks of the Migration Deduced from an Examination

of the Proto-Samoan Content of Efate and

other Languages of Melanesia

BY

WILLIAM CHURCHILL

Sometime Consul-General of the United States in Samoa and Tonga,

Member of the Polynesian Society, the Hawaiian Historical

Society, the American Philological Association

PUBLISHED BY

THE CARNEGIE INSTITUTION OF WASHINGTON
1911

CARNEGIE INSTITUTION OF WASHINGTON

Publication No. 134

PRESS OF GIBSON BROTHERS
WASHINGTON, D. C.

PREFACE.

To introduce the work contained in the succeeding pages would
be irksome; it partakes of the nature of a task, for the book must
be its own best record of the results of years spent in study in the
distant South Seas, of yet other years of intimate toil. But there
is pleasure in making a brief and prefatory record of my introduction
to these Polynesian researches.

That I owe to James D wight Dana, not the least distinguished of
the scientific staff of the United States Exploring Expedition which,
under the naval command of Lieutenant Wilkes, made that brilliant
cruise of discovery in the Pacific Oceans between the years 1838
and 1842. It was a work of supererogation on the part of my pre-
ceptor to answer my questions about the discovery of savage men
in the distant sea half around the globe; it formed no part of the
studies which I was pursuing under his direction. Yet he was ever
cordial; my questions never went unanswered. Thus gradually I
acquired a distinctly personal knowledge of the great work which
had been done in Polynesia, an intimate acquaintance with men
and scenes which serves still to supplement the formal record con-
tained in the reports of that historic voyage.

Not long thereafter, yet it was all of forty years since Dana's day
in the bright South Sea, it was granted me to cruise over many of
the inter-island courses which he had followed. That my cruises
were easy was a debt which I owed to the cartographic work of
that expedition ; that they were almost always safe was no less due
to the moral effect which |the American voyagers had impressed
upon the savages at their first discovery.

When once I landed on the island of Malolo, remotely set in the
Fiji Islands, and found the people coming to the beach to greet me
with yams and bamboo tubes of water, it seemed an interesting,
somewhat picturesque, ceremony. When I inquired into the reason
I learned that it was called an ancient custom to proffer food and
water to all visiting strangers. Yet I found that in less than half
of a stagnant century a custom had become ancient. It was at
Malolo that a boat's crew of the Wilkes expedition had been cut
off, as duly set forth in the third volume of the narrative; after
exacting punishment for this act of murder the Americans laid the

IV PREFACE.

injunction on Malolo to welcome the stranger, and thus the custom
arose.

After an interval of a dozen years I was called to take a part in
the administration of the affairs of the kingdom of Samoa in that
historic tangle in which the United States was striving to hold a
disinterested balance. The most vivacious, and to me the most
valuable, of the small group of island sages whom I gathered about
me in the windward outskirts of Apia to help in the prosecution of
my researches into the past of their people, was a very old chief
of Vaiala. It seemed that he must have lived forever, this ancient
Lauta who had long since retired from the active duties which fall
to the village chief. Like all Samoans he had no sense of the lapse
of time and no knowledge of the date of his birth. But he was able
to contribute one early landmark : when he was a mere stripling he
had gone to Tutuila to undergo his tattooing; when that painful but
socially necessary operation had been completed he returned to Apia
on the deck of "the first man-of-war." So far as related to that
brief voyage from Pagopago to 'Upolu he was a survivor of the
Wilkes expedition which had passed through his seas more than
half a century before I knew him.

In these and many other ways my participation in the study of
the South Sea has always seemed to me an inheritance from the
voyagers who sailed with Wilkes so many years ago.

Nor should I omit acknowledgment of the obligation under which
I lie to Laupepa, the last of a long line of Malietoas who had ruled
Samoa from a period which corresponds to the time in our own
reckoning when Norman William crossed the Channel and fought
down Saxon Harold on Senlac Field. We had made him king. Poor
weary soul, we could not make him royal, for we made it impossible
for him to reign. Now that he has gone beyond the sufferings of
a king, now that the line of the Malietoas has been broken off, now
that the puppet kingdom of Samoa is no more, I recall with pleasure
that he did enjoy the respite from the cares of his troubled state
in the many hours in which he delighted to communicate to me
his stores of the wisdom of the past. Few there were who could
speak Samoan with his grace of diction ; few indeed had minds so
replete with the myth and tradition in which is preserved the ancient
history of his race. In regardful memory I must not neglect to
include among those who introduced me to these studies the name
of Malietoa Laupepa, the last king of Samoa.

PREFACE. V

By rigid processes of exclusion I have sought to make the lin-
guistic material assembled in this volume tell the tale of the peopling
of so much of the Pacific as is comprehended within the range of
its extent. It will be seen that Micronesia is wholly omitted; con-
siderable material is available for the study of that equatorial region,
but it is removed by an irreducible gap from the sweep of the data
upon which these studies are based. In Melanesia this material
tells no tale of the origin of the dusky races there found; it gives
us no more than the assurance that the dark races were already
settled in their crude savagery when the migration swarms of the
brilliant Polynesians swept onward to happier homes ever eastward.
Not all of Polynesia are we to find included within the scope of
this work. The linguistic record here dealt with excludes of its own
motion the later sweep of migration, that to which I have given
the designation of Tongafiti, the adventurous voyagers who swept
onward past central Polynesia to found new races at the utmost
verge of the great South Sea.

Our material restricts us to the most ancient Polynesians, the first-
comers into the Pacific, voyagers who swept the unknown sea some
two thousand years ago. Of these Proto-Samoans we find here a
history which carries them back to their expulsion from the Asiatic
archipelago. I have essayed to plot their ocean fairways. I have
shown that in two swarms they came out from Indonesia ; that one
swarm came around the north of New Guinea and entered the Pacific
by way of Saint George's Channel and at last came to new homes in
Samoa ; that the other was driven by advancing Malayans into the
Arafura Sea and south of New Guinea through Torres Straits and
thence onward to a new home in Fiji. There in Nuclear Polynesia
the sundered kin resumed their fellowship ; thence they despatched
yet other expeditions which brought them to Hawaii, to New Zealand,
and to several spots in the distant east of the Pacific. Upon the
smaller of the accompanying charts I have plotted so much of these
Proto-Samoan voyages as I have been able to determine, and to
these I have added the voyages of the Tongafiti folk who came later,
by about a thousand years, leaving uncertain the voyage which
brought them into Nuclear Polynesia, since this material affords no
record thereof.

Nor is this all. This record points to something of wider value
than the wandering of an unimportant folk in a world of islands
which can attain but scantily to economic importance. We are

VI PREFACE.

engaged upon a group of languages of the most elemental character, a
speech wherein the parts of speech have but just begun to make their
appearance. That in itself would be matter of no great moment,
for we know many languages of the isolating type. That which is
of particular value herein is that we find ourselves engaged with a
language family in which we can discover the beginnings of human
speech. We find ourselves made witnesses of the man who can
emit a cry because he has the animal equipment of a throat and
lungs, and we see that man, with a sentient mind to give him the
impulse of progress, striving by rude and uncouth mouthings to
attain to facility in the use of the consonants which make speech.
It will be an acceptable reward of pleasant toil if it shall be found
that the Polynesian language family is capable of affording us a
true knowledge of a genesis of the speech of man.

TABLE OF CONTENTS.

Chapter I. The Problem of Melanesia i

Inosculation of the Melanesian and the Polynesian languages and the determi-
nation of values therein. — The position of Melanesia. — Viti an area of the
mingling of the two stocks.— Polynesia has charmed and Melanesia revolted
their discoverers; our acquaintance with the latter therefore falls short of
our knowledge of the former. — Islands of the Polynesian verge. — Polyne-
sian inclusions. — Languages which borrow.

Chapter II. The Dictionary of Efati 5

Most of the accessible vocabularies are very scanty. — Only three Melanesian
dictionaries, those of Mota, Efate, and Viti, are at all considerable. — The
development of Macdonald's theory of Semitic origin. — His Efate
dictionary and its false lexicography. — Some words hidden from sight. —
Clumsiness in the definition. — Introduction of polemic dishonesty. — His
false etymology. — No clear distinction of several dialects.

Chapter III. Sawaiori Migrations 13

Paucity of our knowledge of the Melanesian origins. — They may be autoch-
thons.— There are two principal theories of Polynesian migrations. — The
sieve theory and the argument of Thilenius in its behalf. — A spurious tale
of seven times seven retailed by Deeken. — The general migration theory. —
Tregear's statement that this is the commonly accepted hypothesis. — -Percy
Smith in its support. — A fallacy into which Thilenius has been led. — Per-
sisting memories of an inferior race once encountered. — The log of one of
the great voyages.

Chapter IV. The First Polynesian Home 25

A race always under an eastward momentum. — No positive statement of the
place of origin possible. — Bopp proposed an Aryan source. — Max Mtiller
connected it with the Turanian stock. — Logan regarded the Ganges valley
as the ancient home. — Macdonald's Semitic theory of a great and widely
diffused Oceanic language set forth at large.

Chapter V. Dissection of the Theory 31

The result of independent treatment of the data should be identical. — As the
other elements have long been known and carefully studied, the Melane-
sian is the critical test. — The computation of exactly what material is now
made available for study. — In what proportion the Efate vocabulary con-
tributes to the solution of the problem and the manner in which properly
it may be employed.

Chapter VI. Efate" and Viti and Polynesia 35

Comparison of the vowels of the Melanesian element of Efate and Viti. — The
establishment of consonantal variety . — Assumption of a parent speech from
which these deviate. — Viti appears to be the younger son. — The compar-
ison continued through the Sawaiori element of Viti. — Viti, Samoan, and
their parent speech, the Proto-Samoan. — Comparison of Efate" with Poly-
nesia and summation of results. — The extent to which Efatd identifications
penetrate into Polynesia. — Argument from recorded anomalies. — It is im-
probable that Efat£ received its Polynesian content through westward drift
of castaways. — Proof that this element came through the migration of the
Proto-Samoan wanderers. — Check-list of Polynesian phonetic mutations.

VII

VIII CONTENTS.

Chapter VII. Polynesian Relics throughout Melanesia 55

Check-list of the material for this section of the work. — Tables of the phonetic
relations of 81 languages of the Melanesian archipelagoes. — The several
mutations of vowels and consonants and the languages which employ
them. — Analysis of these mutations, those which are found in Polynesia
and those confined to Melanesia. — The groundwork of Polynesian muta-
tion and the Melanesian system compared therewith. — Two Melanesian
foci of Polynesian influence brought to light. — The sieve theory disproved
by this material. — Proof of the general migration theory. — -Crop colonies
and the important part they play. — Two tracks of Proto-Samoan migration
through Melanesia. — Proof that a Melanesian sojourn preceded the settle-
ment of Samoa.

Chapter VIII. Sawaiori Material in Indonesia 151

Limitation of the points of inquiry. — Check-list of the Indonesian material. —
Synoptical tables of mutation varieties. — Mutations compared with the
systems of the Pacific languages. — Character and probable place of the
contact of Indonesian and Polynesian.- — -The nature of an ethnic swarm
discussed. — The Malay advance was an affair of outposts. — Whence arose
the speech community, which after all is a matter of but a gross of words. —
The Indonesians are shown to be borrowers. — Two lines of Sawaiori escape
through the Malay archipelago lead to the two tracks identified through
Melanesia. — The designation Malayo-Polynesian should be discarded
because false.

Chapter IX. The Sawaiori Beginning Rests Unknown 175

Check-list of the Semitic words for which affinity has been sought. — Failure of
the effort to identify this material with Sawaiori stock.— The reasons lie
in false definitions and irregularity of phonetic principles. — -The Semitic
does not conform to the laws of the family. — Summation of the results of
this inquiry. — -The two Sawaiori swarms, the earlier through Melanesia,
the latter not yet discovered on the face of the trackless sea. — The double
migration track in the western Pacific. — The problem of the Melanesians
has been considered only in so far as they have been affected by the wan-
dering Sawaiori. — End of the classification which has joined Malay and
Polynesian. — The beginning of the great Polynesian race is lost in west-
ward and empty sea.

Appendix i. Data and Notes 18.5

Appendix 2. The Southern Gateway 433

Appendix 3. Bibliography 493

The Polynesian Wanderings

Tracks of the Migration Deduced from an Examination

of the Proto-Samoan Content of Efate and

other Languages of Melanesia

CHAPTER I.
THE PROBLEM OF MELANESIA.

Inosculation of the Melanesian and the Polynesian languages and the
determination of values therein — The position of Melanesia — Viti an
area of the mingling of the two stocks — Polynesia has charmed and
Melanesia revolted their discoverers ; our acquaintance with the latter
therefore falls short of our knowledge of the former — Islands of the
Polynesian verge — Polynesian inclusions — Languages which borrow.

Based upon the possession of a greater mass of material than we
have ever enjoyed for the examination of any one of the languages
of the islands of the Western Pacific, the purpose of this work is to
present such determinations of ascertainable values in the inoscula-
tion of the Melanesian and the Polynesian tongues as the present
state of our knowledge may be found to warrant. We shall find it
convenient, in due course, to list a brief bibliography of such works
as have become available in the study of this topic. These inter-
esting and valued works of my predecessors in this tangled field will
be found to lie in two classes, the record of data and the discussion
based upon such data.

The publication of Dr. Macdonald's studies in the speech of Efate,
eagerly welcomed and as warmly reprobated, has seemed to make
it incumbent upon me to engage more intimately upon the prose-
cution of the studies whose results are offered in the present volume.
His work upon Efate falls into each class. It is a considerable
vocabulary of a speech largely Melanesian; it is a labored essay to
build a structure of criticism and comment upon this material. We
shall welcome it in its former capacity as a long stride onward in
our knowledge of Melanesia; we shall find it quite as necessary to
subject its argumentative deductions to rigid scrutiny, in which our
interest is to remain cordial even though our judgment prove
adverse.

The least known of the trine division of the Pacific, Melanesia
affords the most numerous and the greatest problems which confront
those of us who have given time and have expended thought upon
the study of the life of man in the South Sea. These problems are
of two sorts. One great class consists of the problems internal to
Melanesia itself, the other has to do with the problems of the Poly-
nesian ethnic and linguistic stock. Being problems of human life,
they are by no means discrete. The closer into them our examina-
tion carries us the more intimately do we learn the interdependence
of the problems of the one sort upon the problems of the other.

2 THE POLYNESIAN WANDERINGS.

Considered geographically, Melanesia is a unit easy of definition,
save at its northern projection where it impinges upon areas in one
direction known to be Papuan and in another upon yet other areas
known to be Indonesian. The islands (to which the skin pigmen-
tation of the inhabitants, in marked contrast with that of their
neighbors eastward, has made it a simple metaphor to apply the
designation of the Black Islands) lie in a loose linking of chains a
thousand miles offshore from the northeastern coast of Australia,
and in their extent in a roughly northwest direction they closely
parallel that coast. The southern verge of this area falls little short
of the Tropic of Capricorn ; its northern limit lies almost exactly on
the Equator. These limits are, respectively, the considerable land
mass of New Caledonia and the tiny islets of the Admiralty Group.
Reckoning northward from New Caledonia, we include in the larger
subdivision of the area the Loyalty Group, the New Hebrides, the
Banks Group, the Solomon Islands, and the Bismarck Archipelago.
The designation of this last component we owe to a colonizing zeal
which has proved sufficiently potent to act upon the Germans,
geographers as well as statesmen, in blunting a sense of common
geographical propriety. The islands had borne the names of New
Britain, New Ireland, the Duke of York, and New Hanover, collec-
tively the New Britannia Archipelago, so long and so familiarly that
it was force rather than any necessity which new-named them Neu-
Pommern, Neu- Mecklenburg, Neu-Lauenburg, and dedicated them
in totality to the then Iron Chancellor. A geographical sin, unfortu-
nately a sin accomplished.

The Fijian Archipelago may or may not be included in Melanesia;
all depends upon the interpretation of certain well-defined problems
of its own. Geographically it is not necessary so to include it, for
it lies remote, out of the northwest chain, set by itself in its own
sea midway between the scarcely contaminated Melanesia of New
Caledonia and the equally uncorrupted Polynesia of Samoa and
Tonga, the region to which I have assigned the convenient designa-
tion of Nuclear Polynesia. Ethnically and philologically Viti must
be acknowledged to lie in a position of mixture of the two neighbor
stocks. I know that I go beyond many, if not all, of my fellow
workers in weighing the Polynesian element in Viti.

While with this exception we find Melanesia a well-defined unit
upon the charts we are by no means qualified to decide if this unity
extends to the ethnography and philology of the region, and that
because of our lack of consistent information. We are sadly deficient
in the necessary data, for Melanesia has received scant attention.

While Polynesia has attracted, Melanesia has repelled its dis-
coverers. About the islands of the central tract of ocean romance
has cast its charm; its power remains even in these later days.

THE PROBLEM OF MELANESIA. 6

Sensitive natures have counted the world well lost for the enjoyment
of its delights; ignorant men have yielded to the same compulsion
and have found dingy pleasure in settling down as beachcombers.
The great nations have sent brave fleets to the exploration of these
islands, have lent their most competent administrators to foster the
states of island monarchs. The people have won those who came to
seek them; they have been treated as gentlefolk.

But Melanesia is a volume whose chapters are horror upon horror.
The islands of this western area lack the charm which holds the eye
upon the atoll with its palm tiara or upon the towering summits
forest-clad and necklaced with cascades so familiar in the Polynesian
scene. The people of Melanesia have an aspect more savage than
the statuesque dignity of the Polynesian. It does no violence to
the sense of the fitness of things to look upon them as a servile crew.
It was only upon the score of our morals, not of social propriety,
that objection was raised to the labor trade in Melanesia, a form
of slavery to which the Polynesian never was subjected. Explorers
in Melanesia have relatively been few. Missionary endeavor was
slow to attack this field of crying need, and when the missions did
effect a lodgment in this dark region of the sea the pioneers were
members of a sect which affected repugnance toward all matters
which make solely for broader culture. From many the crown of
martyrdom was not withheld. Yet that blessing was, after all, indi-
vidual; we compare the information derived from the Melanesian
missionaries with the treasures of scholarship which mark the work
of their fellows in Polynesia, and we deplore the comparison.

Of such sort are the reasons wherewith we must account to our-
selves for the fact that Melanesia yet remains to the student almost
wholly in darkness and gross darkness upon the people, and that
the light which here and there in a random ray has been shed upon
its problems must first be subjected to close analysis.

In the early lines of this chapter use has been made of the term
inosculation of the Melanesian and Polynesian tongues. It is now
in order to define the nature of the approximation of these two
language series within the Melanesian area.

We have first the several islands wholly or principally inhabited
by folk of Polynesian race and speech, yet lying within the region
which geographically is classed as Melanesia. Such, among others
in a short list, are Aniwa and Fotuna,* Sikayana, Ticopia, Liueniua,t

*This spelling having been in some general use, I find it a convenient means of
differentiation from the Futuna of Nuclear Polynesia.

f "The name of this atoll as given on the chart is Leueaeuwa, but the name, I think,
is wrongly spelled, as it bears no meaning that I know of in any Polynesian language.
The proper spelling is Le ua Niua. This was certainly the way in which I wrote it
before I knew of the other spelling, and the Samoan who was with me also spelled it in
the same way." The Rev. G. Brown, D. D., "Reports of the Australasian Association
for the Advancement of Science," IX, 258.

4 THE POLYNESIAN WANDERINGS.

Tauu, Nukumanu, Nuguria, Rennel,* and Moiki. Since these islands
represent the western limit of Polynesian race and Polynesian speech,
we may name them the islands of Polynesia's western verge, or sim-
ply and briefly the Polynesian verge.

In the next class we find small communities on islands with an
otherwise Melanesian population, where Polynesian speech is spoken
by folk presenting sometimes more and sometimes less of Polynesian
race traits. Under this designation come, among others, Mae in
the New Hebrides (on the island known as Three Hills) and Fileni
in the Swallow Group. To this class of traces of Polynesian origin
we shall apply the term the Polynesian inclusions.

Necessarily the foregoing classes are limited in the number of the
instances properly to be assembled under one or the other. The
third class is more widely extended than we are yet in a position
to estimate. This class is to include all the instances in which we
find peoples of Melanesian stock and speaking languages prepon-
derantly non-Polynesian, who yet derive some portion of their
vocabulary from Polynesian loan material. For convenience we use
this term, recognizing that the subject is open to argument. In a
diagrammatic scheme of the possibilities it is as antecedently pos-
sible also that the Polynesian has borrowed from the Melanesian,
or that the common element derives in Melanesian and Polynesian
from an earlier undistributed source. Yet I have no hesitation in
anticipating the result of the argument and describing this common
matter as Polynesian loan material. It is this latter topic which
is principally to engage our attention in this work.

At this point it seems proper to invite attention to the appendix
introducing a bibliography of the published matter which has been
consulted and which has yielded more or less of assistance in the
study of this topic. This will be found on pages 493-506.

*A word of explanation may not be out of place as to the unequal dealing with the
names of the twin islands Rennel and Bellona. These are chart names and secondary
in rank to the names in use by the islanders. For Bellona we have the valuable record
of Dr. Sidney H. Ray, who has recorded its vocabulary under the name Moi-ki; the
hyphen is wrongly placed through a typographical error, as has been pointed out by
W. von Biilow; I prefer to standardize the name with other Polynesian forms and there-
fore omit the hyphen. The name of Rennel is cited by Wawn and by Thilenius as
Muava or Mungava; probably it should be Moava, but as my notes are not positive I
hesitate to adopt the form. Thilenius makes the distinct statement that Moiki has no
fixed population and is no more than a fishing station for the people of Rennel. This
is not in accord with my observation nor with that of Captain Wawn, who has written
an interesting narrative of the long search for his home in which he assisted a labor boy
who had been carried away to Queensland. The home was found at last in Moiki, which
Wawn calls Mungiki.

CHAPTER II.
THE DICTIONARY OF EFATrL

Most of the accessible vocabularies are very scanty — Only three Mel-
anesian dictionaries (those of Mota, Efatd and Viti) are at all consider-
able— The development of Macdonald's theory of Semitic origin — His
Efate dictionary and its false lexicography — Some words hidden from
sight — Clumsiness in the definition — Introduction of polemic dishon-
esty— His false etymology — No clear distinction of several dialects.

Quantitatively regarded, the lists presented in the bibliography
seem to show that we have no inconsiderable material bearing more
or less directly upon the philological problems of the islands of the
Western Pacific chain. When investigated more closely, when
measured in a qualitative analysis, the tale is far other. Many of
these vocabularies, so diligently sought and so sedulously treasured,
are mere lists of but a score or so of words and often of problematical
accuracy in reporting. In very few cases have they been subjected
to intelligent criticism. Yet they are by no means to be despised.
They are the best we have, and with them we must perforce be
content until future exploration affords better data.

We have in the Melanesian tract but three vocabularies of any
considerable magnitude. The dictionary of the Fijian is of inesti-
mable value, a storehouse of information; yet it has been found to
yield its most valuable results when associated with more strictly
Polynesian investigations. The Mota dictionary is of far lower order,
yet none the less is it a valuable implement of Melanesian study.
The Efate dictionary is by far our best contribution of data upon
which is to rest the science of Melanesian speech; and this hearty com-
mendation must be kept in mind through all the adverse criticism
which it will be necessary to pass upon it in many details.

Dr. Macdonald awakens our envy when we note the opportunity
he has enjoyed for the study of the speech and the habit of Efate\
thirty-five years spent in the search into the language and the
mind of this interesting family of Melanesians, a study directed
solely to the attainment of such knowledge as should better fit him
to become the guide of the souls to whose cure he had been sent in
the isles of the sea at the uttermost parts of the earth.

In the bibliography it will be observed that in 1882 he exhibited
his Semitic theory to the Royal Society of Victoria. In 1889 he
presented it anew and in richer development in "Oceania." At
intervals he has contributed minutely prepared papers to the Journal
of the Polynesian Society exhibiting yet further argument in behalf
of his theory. Now he has attained to such a mass of evidence,

6 THE POLYNESIAN WANDERINGS.

so satisfactory to himself, that he feels justified in entitling his
latest work "The Oceanic Languages — Origin," rather than the
dictionary of Efate, which it really is. His Semitic theory we shall
have to study in a later chapter of this work; at this point we feel
it proper to comment upon the work as dictionary alone and freed
from its speculative adornments.

An initial and serious objection to the work is that it is devoid
of even the slightest summary or sketch of the grammar of the
language. True there is an introductory section of well nigh a
hundred pages to which a hint in the title points as grammar. Yet
in the whole treatise there is not so much as a single page by read-
ing which the student may arrive at any comprehension of the
manner in which to combine into sentences the words so plentifully
listed in the vocabulary; there is no hint by which he may be
directed in the use of these many words as actual speech expressive
of the thoughts he would convey. Despite its presence where a
grammar is properly to be expected, this assemblage of introductory
chapters is the argument of a fine-spun theory, interesting to phi-
lologists, yet wholly useless lumber to the person who must rely
upon this work as his introduction to the speech of Efate.

So little has Dr. Macdonald understood the proper purpose of a
dictionary-maker that he will not even head his vocabulary for what
it really is, but prefers the polemical statement of his dear theory in
the title "The Oceanic Languages, their Material or Vocabulary Set
Forth in a Complete Dictionary Comparative and Etymological of
One of Them, the Language of Efate. " It is only as an afterthought,
deferred to the last possible moment, that the man who has spent
more than a generation in the study consents to affix its title to the
really valuable part of his work.

Such a mental attitude on the part of a lexicographer, the gran-
deur of his privilege to record that which is being so completely
obscured in his zeal to register what he fondly imagines ought to
be, is difficult of comprehension to those of us who content our-
selves through busy years of dictionary-making. Yet this attitude
is not new with Dr. Macdonald; it is well known that Webster
refused to sully the first edition of his great dictionary of the English
language by including the word "bridegroom," which merely existed
in English speech and had no right to exist, yet in the end he failed
to secure currency for "bridegoom." With this record of lexico-
graphic pertinacity we are ready to make due allowances for the
outcrop of the theory which to Dr. Macdonald means so much.

To what extent this dictionary answers the author's character-
ization of "complete" we have no means of ascertaining. I prefer
to record the fact that it contains 3,657 word entries, a fact estab-
lished by my own tally. This gives us a positive measure rand

THE DICTIONARY OF EFATE. <

establishes at once the great superiority of the Efate dictionary
over all Melanesian word collections. It is of inestimable value to
us by reason of its comprehensiveness as speech record.

When we examine carefully and in detail this work in the aspect
of that which must represent its permanent value, that which will
remain after the demolition of the theories of which it is made the
vehicle, namely, its value as a dictionary of the Efate speech, our
criticism will fall broadly into two classes, the mechanical and the
sense characteristics.

The former class is but a particularization of the postulate that
lexicography has grown into a science with no little exactitude in
its method. Without close study of the principles of the science,
it is wholly impossible for any student of language, no matter how
intimate may be his knowledge of the speech upon which his study
has been directed, to win success as a lexicographer solely by
reason of his familiarity with his language theme. Sage as Dr.
Macdonald is in all that pertains to the speech of Efate, he makes
all the typical errors of the unskilled dictionary-maker.

In a dictionary the alphabetical order of entries is necessarily
supreme. Almost any page at random in this work will exhibit
instances where the entries capsize the alphabetical order. Our
author will undoubtedly explain his inversions on the score that
thus he is able to keep together stems and derivative forms which
on the alphabetical system would be scattered. It is hardly worth
while remarking that the user of a dictionary has the right to demand
that the word of which he is in search shall be found in its proper
place. I may note that in the most assiduous use of this dictionary,
extending over many months of close examination, I have not
succeeded in conquering my annoyance at the difficulty of finding
many a word which is not in the place where it should be; in fact
that I have been able to make use of it only after the compilation
of an index — an index to a dictionary !

If this is the case in the mere arrangement of the words on the
page, where a misplacement entails no greater hardship than a
search through one or more of the neighboring pages, what shall
we say of those forms which are secreted — it may be pages away
and under a different initial — in some entry from which they may
be extracted only through a knowledge of the language far beyond
those who are likely to use this work? For the exhibition of this
blemish we may cite the noun futei, the white ant, the entry con-
taining note of the variant forms mitoi and mitei; yet neither of
these forms, though equally in use, occupies its alphabetic position
by so much as the merest tag of a cross-reference; to a student of
Efate encountering the word mitei and seeking its meaning this
dictionary would offer no help.

8 THE POLYNESIAN WANDERINGS.

In the latter class of criticism we note that many of the defini-
tions are so clumsy as to suggest that the author had very slight
knowledge of what the English name is of objects which he is describ-
ing. This is notably illustrated in the definition

sumili, a thing like india rubber in a clam shell which, when touched, causes the
shell to close.

This fairly parallels the entry in Shirley Baker's Tonga dictionary
of "balolo, a reptile much like the earthworm found in the sea."

In further showing of Dr. Macdonald's ineptitude in definition
we note the following, somewhat at random:

alo-fi, wave (with a circular and rolling motion) to him, to beckon to.

atu saki, plop up (of a turtle, also of the sound of the breath in the throat of a

man recovering from a faint or dying),
beingo, a kind of flute (coconut shell).
bisa, to utter inarticulate sounds (as those made by a coconut on the gravel

which a rat is turning about trying to get at its kernel).
nakasu nabwo na, the cartilaginous substance on the front of the throat, lit. the

stick, or tree, of the bwo (pectus).
klta roa sa, to hate turning after him (someone), as a boy sent a message meeting

another boy and (hating to do the message) turns after him to play.
libu, lebu, the middle of the lower part of the body at the upper part of the back

of the pelvis.

These are sins of ignorance, and ignorance, though stupid, is at
least innocent. In general, in my work upon these dictionaries
coming first hand from missionaries who lay no claim to skill in the
arts of lexicography, I have welcomed the naivete and have employed
the definition wherever its terms have not been too ridiculous. They
are original documents ; their simplicity is their warrant of honesty.
In some former paper, I believe, I have mentioned the care with
which I have refrained from recasting these definitions and the
reasons therefor. Such sins as have been here presented we are
glad to forgive. Our glee at their discovery carries no malice.

But when the definition begins to squint at something ulterior,
when a word is added with the sly insinuation of a purpose to link
the fact and the theory more tightly, when definition becomes
polemic, such exhibitions I do not hesitate to stamp as scientific
dishonesty; they prove an absence of conscience, without which
speculation is mere trickery held in check only by consideration of
the risk of discovery.

A very few instances here (the volume abounds with such and
they will receive detailed attention in the critical notes) will suffice
to show at this point that my characterization is not lacking in
support.

tere, the mast (of a canoe or ship), calf (column) of the leg; Arabic, sariyat, sari,
the mast of a ship, a column.

If Dr. Macdonald's compelling Semitic theory had not already pro-
duced a moral strabismus of the insight, is it to be imagined that he

THE DICTIONARY OF EFATE. 9

would have cast into a parenthesis that word "column" in explica-
tion or suggestion of an explication of the sense of calf? That
offensive and offending parenthesis was not set into the definition
innocently.

alo, to swim (wave hands).

The parenthesis is expressly for the purpose of linking it with the
verb alo-fi, which I have already cited for its clumsiness of definition.

bare, to be dirty-looking, like a sightless eye (of half-raw food) ; Hebrew 'avar,

Ethiopic 'awir, to be blind.
bwes, besii, a young pig whose mother is dead and which is brought up as a pet

and is therefore tame and gentle; also a motherless child, so called from

being deprived of the mother's milk and, as it were, arid; Arabic yabisa, to

be dry.

The same leering argument will appear in the study of ngoko (95),
where he introduces hack in definition of Hebrew hakah. If his use
of italics has any meaning at all it must be intended to mark his
obiter dictum that the name of the Teutonic ax is Semitic. And
this in a work of polemical philology!

Such childish efforts to misdirect the comprehension arouse
repugnance ; they cast upon the whole work a suspicion which really
it does not deserve. The result of such discoveries of obliquity is
that one loses confidence in each uncorroborated point in the book.
To many students it will render the work valueless. Yet so far as
my acquaintance with the speech of Efate extends, reinforced by a
considerable familiarity with other tongues of Melanesia and of
Polynesia, I have no hesitation in repeating my former statement
that this is far and away the most valuable contribution to our
knowledge of the speech of the Western Pacific.

It is not only when Dr. Macdonald espies the chance to lug in his
Semitic theory that he takes unwarranted liberties. He etymol-
ogizes generalities on materials which do not reach beyond the two
entrances to Havannah Harbor.

binauta, to be numb, devoid of feeling, as one's limb from stoppage of circulation
of the blood in it : bi, to be, nata, a person (as if the limb belonged to some
other person).

bitua na, the knee, prob. bau, the head, and tua, leg.

kuruku, the ankle is so called because the leg gathers itself, as it were, into the
knob of the joint.

To these three add tere, as presented a little earlier. It is quite
fortuitous that no less than four glaring errors have to do with the
leg; one can but wonder how it has come about that Dr. Macdonald 's
legs have proved such unruly members.

Yet another grave fault vitiates this dictionary as speech record.
The author supplies a copious store of variant forms for many words,
each ticketed with the simple notation of "d.," meaning dialectic.
Nowhere is any hint afforded us of the habitat of such dialects.

10 THE POLYNESIAN WANDERINGS.

Efate is not a large island, yet in every smallest Melanesian island
we are sure to find dialects from village to village, often amounting
to such extensive variation as to produce incomprehensibility. It
would have added clarity to this record if Dr. Macdonald had identi-
fied these dialects by their place names, if by nothing else.

As it is, we are left without the knowledge of what is the speech
which he assumes as the standard from which these dialects diverge.
It is inferential, and only inferential, that he has assumed for such
standard the speech in the community nearest his mission station;
that is to say, one of several petty villages on Havannah Harbor.
Since the seat of the administration of the New Hebrides is at Vila,
in a different bay, it is possible that the speech there in use may tend
to become standard for official communication; yet not a single
entry in connection with these comprehensively noted dialectic
variants indicates which are Vila forms.

More than this. In the examination of Efate material presented
in extenso in a succeeding chapter we have felt a grievous loss in the
inability to coordinate the several dialects in order that we might
study the several systems of vowel and consonant mutation, a
matter of vital importance. The most we can say from the study
of this dictionary is that at one spot on Efate, presumably proximate
to the mission station, the people use this word, other peoples at
one or more undefined places on Efate use this or that other and
frequently quite dissimilar word. To what a disadvantage this
necessarily puts the student appears in quantities that may be
measured in two brief Efate vocabularies,* upon which we had to
rely before Dr. Macdonald 's dictionary was published. From Hans
Conon von der Gabelentzf we learn that one of these vocabularies
comes from Mele, the other from Erakor on the south coast of the
island. The two lists contain resembling words to the number of
26, all lying within their common Polynesian content. The Mele
list, numbering 118 words, shows no less than 87 words immediately
recognizable as Polynesian; the Erakor list, numbering 121 words,
shows but 27 of Polynesian source. Mele is the language spoken
between Havannah Harbor and Vila; Erakor, on the south coast,
has the mountain center of the island between it and the mission
station which we have inferred to be the seat of Dr. Macdonald 's
standard speech. Yet the Erakor dialect, and not the nearer Mele,
most accords with this dictionary. With this suggestion of the
magnitude of the dialectic differences we shall feel great uncer-
tainty as to the results of the more intimate inspection of this
dictionary record upon which we are to enter. Dr. Macdonald'is
the one man who knows these dialects apart; it was his dutyUo

♦Turner, "Samoa a hundred years ago and long before," page 354.
fDie melanesischen Sprachen II, 1.

THE DICTIONARY OF EFATE. 11

differentiate them for the benefit of others who must rely upon his
accuracy.*

No one can regret more than I the necessity under which I lie to
pass such comments as those which have preceded. Yet let it be
well understood that they are by no means conclusions superficially
drawn. The difficulties mentioned have been encountered by me in
no mere casual glance at the work, but have arisen to hamper me
in my close study to learn this language from the material here
afforded. How close this study has been will be shown in the follow-
ing discussions, in which it will appear that the words have been
analyzed down to the very letters of which they are composed.

Despite these great drawbacks, despite a still greater fault yet to
be discussed, I would not recede from the characterization which I
have presented and which I have reiterated. Imperfect as it proves
itself to be, Dr. Macdonald's dictionary is amply the most valuable
contribution to our knowledge of any speech of Melanesia.

*Even as these pages are passing to print I am in opportune receipt of a letter from
Captain Rason, until quite recently British commissioner in the New Hebrides. This
letter affords us valuable information upon this very point and not ungracefully mani-
fests the writer's true kindness of heart.

"I wish to explain as soon as possible, if possible in time for your book, the inner
meaning of Dr. Macdonald's muddle. When the missionaries established themselves on
Efate" he was in Havannah Harbor, and natives who first became Christians left their
villages and came to the mission station for protection. Thus the language of the mis-
sion station became a medley of all the dialects around. This gradually coalesced into
a special dialect which became a lingua franca with the natives and was partially under-
stood by all. As the heathen natives died out or became Christian the mission language
was claimed as the language of the island. Then the Bible was translated into this
language and Dr. Macdonald wrote a dictionary of it as if the missionary language was
the original language of the various villages before they were Christian. The poor man
only deceived himself and is now deceiving others, but it is not wilful scientific dishonesty.
I should like him cleared of that. It is a case of self-deception."

CHAPTER III.
SAWAIORI MIGRATIONS.

Paucity of our knowledge of the Melanesian origins — They may be
autochthons — There are two principal theories of Polynesian migra-
tions— The sieve theory and the argument of Thilenius in its
behalf — A spurious tale of seven times seven retailed by Deeken —
The general migration theory — Tregear's statement that this is the
commonly accepted hypothesis — Percy Smith in its support — A
fallacy into which Thilenius has been led — Persisting memories of an
inferior race once encountered — The log of one of the great voyages.

Thus far we have considered the work in respect of its status as
an Efate dictionary, a status which it has comported with Dr.
Macdonald's lucubrations in the realm of theory to adumbrate in
the text even as he has buried it as an afterthought in the title.

Before advancing upon his main theme, the theory and effort at
substantiation of a Semitic origin for the languages of the Pacific,
we shall find it well to devote some consideration to the present
state of our opinion as to the movement in migration which has
brought to the Pacific area the peoples now spread to its remotest
isles.

So far as relates to migrations of the Melanesian peoples, we are
wholly without information. This work of Dr. Macdonald is practi-
cally the first essay toward giving any of the Melanesians a race
history anterior to their residence upon the islands which they now
inhabit. Until this or some other theory is properly established,
we can do no more than to regard them as ex hypothesi autochthons.
If future students of their life and thought succeed in bringing to
light traditions which may point to a movement over seas from an
older to their present homes, then we shall have a basis on which
to found new speculations. Until such a time arrive we shall find
it better to stand on the commonly accepted hypothesis.

But in the case of the brown Polynesian race the circumstances
are far other. We have ample traditions of migration, we have
the names of the halting-places; we find a whole race, widely sun-
dered upon the sea, looking back to the west with a single gaze to an
ancestral home. We have here and there the belief in westward
Pulotu as the abode of the dead; no mean proof, since the dead go
home. Above all we have the primordial Hawaiki across the great
sea of Kiwa, the illuminating Saba myth, than which no tradition
of men has ever had a wider extent.

13

14 THE POLYNESIAN WANDERINGS.

In so far as this element of knowledge impinges upon Melanesian
areas we have theories which have been elaborated to account for
the recognized inosculation of brown Polynesian and black Melane-
sian. Omitting mention, as here unnecessary, of movements of
convection within the general migrating mass, we find two main
theories ably propounded and stoutly argued. Each will receive
our attention to the extent in which it bears upon the phenomena
of this inosculation.

We are first to examine the theory which may be not inaptly
designated the sieve theory. In its briefest presentation it is that
the islands of Melanesia and the Polynesian verge have served as
the meshes of a net to catch the drift of castaways from islands of
central Polynesia blown away from home and set to the westward by
the concurrence of the prevailing winds and currents of that oceanic
area. Its most recent and not the least ingenious presentation is
found in the extensive studies of Dr. G. Thilenius,* professor of
anthropology and ethnology at Breslau.

His theory is brilliantly conclusive — upon the hydrographic
charts. Taking the eastern islands of Malaysia as a datum-point,
he draws the line of least resistance to migrant fleets, the line where
the current sets them on their way and where the wind blows fair.
On the chart he dots their track far to the north of New Guinea
and the Bismarck Archipelago. He gives them successive landfalls
in the southern Carolines and along the chains of the Marshalls.
Thence onward to the Gilberts the constants of the weather are
still in their favor. They make fair weather of it still further to
the Tokelaus, and at last they come to port in Samoa. Once
established in central Polynesia, the winds and currents which have
served them so well become malefic. They blow boats away from
peaceful shores and out for starving, thirsty voyages upon unfriendly
seas. If the Melanesian sieve catches any of these involuntary
wanderers their fate is that of the castaway upon inhospitable shores
of savagery. In the work cited it is very argutely presented, yet
it can scarcely be called convincing.

No one who has had occasion to recognize that the work of the
real Polynesian research is being to-day, as always, painfully prose-
cuted by enthusiastic workers in remote islands of the sea, where
one book is a treasure indeed and library privileges seem no more
substantial than the seraphic and impossible vision, will regret the
space here given to a presentation of the arguments of Dr. Thilenius.
Thus only will they reach earnest students in savage scenes who
could never have access to the rare and costly work in which these
arguments are presented. We acknowledge the pleasure in helping
these distant students, for they are the real workers.

*Ethnographische Ergebnisse aus Melanesien: Die polynesischen Inseln an der Ost-
grenze Melanesiens.

SAWAIORI MIGRATIONS. 15

After recording known instances of involuntary voyages from
central Polynesia to the safe landfall of the Polynesian verge, after
discussing the Polynesian verge as a sieve for such migrations, Dr.
Thilenius concludes his argument as follows :*

When one considers the remarkable exiguity of our islands [the Poly-
nesian verge], of whose surface, furthermore, only a fraction is habitable or
productive of food, certainly we find an adequate explanation of their
present population through these involuntary migrations. Whether at
any former time a Melanesian population of any sort lived upon these
islands is a question easy to put but hard to decide. No data of any nature
bear upon this idea. It is possible to hold the idea that there had been a
Melanesian population, but that for reasons unknown it had withdrawn
before the coming of the first wanderers, for the earliest settlers on Nuguria
and Liueniua found the land uninhabited and there is no ground on which
to set aside their testimony as false.

Therefore the migration theory needs a brief discussion. Our islands
might have been halting-places of the peoples swarming from somewhere
in the northern Moluccas toward Polynesia. These early Polynesians must
have come hither out of the northwest, fairly enough along the same course
as was traversed by the boats coming from Ninigo, Taui (?) and Kapinga-
marangi. That presupposes that our little atolls, at least 800 or 1,000
years ago, constituted a region offering an adequate supply of vegetation.
Whether this was the case may readily be questioned on the score of the
thinness of the soil-layer on the islands to-day. It is always a possibility.

The wanderers found the islands uninhabited and left behind them, on
each or on some, a company of those who were travel-weary. But these
folk, wholly ignorant of geography, in some wonderful fashion hit upon
the course to the other islands of northwest Polynesia. The knowledge of
these may have come to them from the fabulous aboriginal Melanesian
population, who surely, to have been able to give such information, con-
trary to the character of the present north Melanesians, must have con-
ducted extended voyages and thus have known one or other of the large
islands of Melanesia. The wanderers would surely have made sufficient
inquiry. Yet they prefer to follow a chain of poor atolls instead of seizing
and holding the great and fertile islands that lay close at hand. They
had sufficient numbers, for we can think of them only as the complement
of a fleet. But Buka, which lies so close to Nuguria at which they first
touched, shows no trace of Polynesians. f Then the whole aboriginal Melan-
esian population of our atoll is become a picture painted by the fancy.

None the less it is highly probable that the wanderers came to one of
the great islands, for they sailed hither against the trades, and it is more
likely that they fell to leeward than that they, ignorant of the situation of
Samoa, busied themselves to make headway against them. At the least
it is probable that wind and current set the wanderers westward. That
not a single boat, on the ten-degree stretch from Nuguria to Sikayana,
was driven westward by only so much as a degree and a half, which was
sufficient to bring one of the great islands into sight, is so much the stranger
since Kilinailau and Nisan lie west of Nuguria and approximately in the
longitude of Buka. These were originally settled by Polynesians, and in

*Op. c, page 78.

fl can not let this statement go unchallenged. In this work I have collated thirty-
one vocables from Buka, of which twenty-four are borrowed from the Polynesian, the
quality being computed at 70 per cent. — W. C.

16 THE POLYNESIAN WANDERINGS.

the sense of the migration theory simultaneously with Nuguria. This can
be understood only on the theory that a part of the fleet, at least on this
stretch, was set somewhere about two degrees westward.

Are we then to assume that a people endowed with the energy to enter
upon a voyage into the unknown were not in condition to set foot on one
of the great Melanesian islands? Did they find the whole island thickly
set with towns which beat them back? Had the weeks since they had
forsaken the homeland caused them to avoid a conflict which would have
put them into possession of a fertile island in order that on the other hand
they might follow with marvellous instinct a chain of islets which would
lead them next to Ticopia, after which they must sail a long voyage on
the open ocean? And all this against the trades! Furthermore we are
confident that the wanderers carried with them useful animals and plants.
Probably they landed some of these upon our islands. What has become
of them? Tradition and history know them as introduced by settlers or
altogether the gift of the white men. Have those originally introduced
perished, have they run out?

The more we try to establish details of the voyage in regard of local
conditions and the present character of the Polynesians, so much the more
improbable becomes the thought that strikes us at first glance that our
atolls were the halting-places of the early Polynesians.

Readily enough the possibility is suggested in accordance with which
the wanderers journeyed in close accord with the theory, used the islands
as stopping-places, etc. But among the multitude of possibilities that
suggest themselves the exact coincidence would mark this as an astounding
accident. For not only during the voyage along the Melanesian islands
must an unusually great series of accidents have been actively at work, but
in the same measure the same must have been true from the outset of the
journey.

The early Polynesian left the Malayan tract, as is properly to be assumed,
by way of the Celebes Sea and Straits of Molucca and then encountered
current and wind conditions varying from season to season. From Novem-
ber to March northwest and north winds blow with interruptions and a
current sets toward the southeast. In the latter case the Polynesians
might actually reach our islands. But that implies the beginning of the
voyage in the bad season. If one takes into consideration the remarkable
sensitiveness of the Oceanic peoples to rain, almost laughable in our sight,
he will scarcely admit the conclusion that the Polynesians intentionally
set forth in the stormy season of the rains. To this is to be added the fact
that the seasonal change of wind and current which holds in the Pacific
was unknown to them, for they are assumed to be coming out of regions in
which this change does not exist in anything like the same fashion.

Consequently there is much in support of the argument that the early
Polynesians left the Malay Archipelago during the good season, that is to
say while the southeast trades held in the eastern regions. They came
then immediately into the equatorial countercurrent, which, moreover,
flows in the northwest season, although with diminished strength ; or they
were forced into it by the south equatorial current. This area is also the
region of calms and variables. They were thus especially directed toward
the current upon which they were necessarily borne. The importance of
these equatorial currents, which attain a considerable velocity, is known
through the Spanish attempts to reach the Palaus and from the history
of boats drifted off from the Carolines. From the Palaus the boats always
drift to Samar or the southern Philippines (north equatorial current) ; on

SAWAIORI MIGRATIONS. 17

the other hand there never came folk from the Philippines to the Palaus,
but only from Celebes and the Celebes Sea (equatorial countercurrent) .
The journey against these streams is possible only with very favorable
winds; they are so considerable that even in the present time the sailing
vessels of the white men have to take them very largely into the reckoning.
The schooner in which I went to Ninigo traveled the long stretch from New
Hanover to Ninigo only with the south equatorial current, and most of the
time against the stormy but certainly light northwest wind. On the return
voyage the current was so strong against us that we had a great notion to
work up to 40 N., where we could catch the countercurrent which would
help us to an easting from which we could reach southward — that is, back
to the Gazelle Peninsula.

The early Polynesians coming out of the Celebes Sea drifted in all prob-
ability along the southern edge of the Carolines toward the east. With
this current, not exposed to strong contrary winds, they could reach the
Gilbert Islands, where local currents make their appearance ; among them
such as would set them southeast, even through the Ellice group. Accord-
ingly, on meteorological grounds, the Polynesians voyaged, not along the
Melanesian islands, but by a straight course through Micronesia, and
reached Samoa, whose Savai'i may have been the prototype of Hawaiki.
Perhaps the boats drifted still farther to Fanning Island, in order to reach
Samoa. There are many possibilities in a region where the countercurrent
has less force than between the Moluccas and the Gilberts. This is not the
place to follow out the further distribution of the wanderers after they had
once reached the present Polynesia. But that an importance attaches to
the countercurrent for migration theories, particularly in its western part,
is clear from the phenomena of the flora. Its distribution is such that the
botanical boundary incloses central Polynesia, Viti in part, next the Ellice
and Gilbert groups, finally the Carolines, as closer to India, while Melanesia
forms a province of its own. The men and the plants of Polynesia, there-
fore, must be regarded as having migrated along the same track. Would it
not be intelligible that to the Polynesian, who came from the rich Moluccas,
the atolls would be less pleasurable and after a short sojourn they swept
farther along until at last they reached again a better endowed, a moun-
tainous, and a greater island, Samoa?

Considerations such as these, of which much is lacking to the theory,
suggest themselves with divers variations. Here, before all things, should
there be but a single probability, it would be of significance for the further
fate of the early Polynesians who reached the equatorial counterstream,
and for the case that they migrated from Halmahera, in favor of which are
many good arguments.

The initial point of the wandering is of great importance for our chain
of islands, for upon that it depends with great probability that the early
Polynesians did not come to Nuguria, etc. This would not be altered by
the arrival in Liueniua of the boat from Kapingamarangi, for the conditions
of current allow us to recognize with certainty such voyages as exceptional.
We can make our account always and only with typical phenomena. Wind
and current conditions do not allow a decision at variance with the tradi-
tions of our islanders, which seem all the more credible since the industrial
products as yet met with on the islands quite confirm the essential points
of all these statements. The peopling of the northwestern Polynesian
islands quite uniformly has its origin in small beginnings, through the
coming to shore of crews of for the most part single boats, and through
infrequent raiding expeditions. The great majority of the immigrants

18 THE POLYNESIAN WANDERINGS.

came hither from the east, from Micronesian and from Polynesian groups ;
a distinctly smaller portion went out from Melanesia, a movement which
did not assume larger proportions until it came to the southern islands of
the series on account of the more peaceful relations with the southern
Solomons, with the Deni Group (Santa Cruz), and other neighbors.

It is easy to point out the easy way upon the charts and to prove
it easy. To us, in these days of veritable islands which make their
way across the seas under impulse of mighty enginery, it is easy
to feel convinced that if Polynesians ever did travel far and wide
over the stormy ocean in frail canoes held together with stitches,
they must have selected the course where the resistance was least.
So far a most excellent case has been made out.

But the Polynesians were the most hardy race of daring navi-
gators that the world has ever known. They know which way
they came, they have preserved the logs of these ancient voyages
when yet the sea was all their own, theirs alone. They were not
afraid of the sea, they fought it, and they had no charts to point
them to an easier traverse.*

Before we proceed to further examination, it seems proper to
interject a brief mention of another presentation of the theory to
whose support Dr. Thilenius has brought all the resources of his
great acumen. The presentation upon which I would animadvert
does not pretend to be a scientific statement. But since it has
found its way into print as something proved, I can not feel it time
wasted to impugn the bona fides of the author who presents it or
the source from which he derived it. This pseudo-myth, this rather
clever fabrication, is given to the world, at a by no means ungraceful
length, in the fourth chapter of Richard Deeken's " Manuia Samoa,"
a volume of travel sketches which has been somewhat severely
criticised from the economic side. With no word of credit to any
authority, with all the positiveness of statement proper to the
record of approved history, he begins his chapter with the account
of a plague in Sumatra: "Niemand kann sagen wann es war,
wahrscheinlich jedoch lange bevor Christi Geburt, als auf Sumatra
eine vernichtende Seuche wiitete? die die junges und altes Leben
in Massen hinmordete."

The story is all the more dangerous because well told. He recites
the ineffectual efforts to stay the disease culminating in the heroic

*I had written these words several months before the first of such charts was put
into the hands of the skilled navigators of the world, the Pilot Chart of the South Pacific
for the months of September, October, and November, 1909, issued by the Hydro-
graphic Bureau of the United States Navy. In the three months for which these data
are tabulated, months, as I well know, of good sailing in those seas, the currents between
the Gilberts and Samoa do not facilitate canoe voyaging, as Dr. Thilenius is so satisfied.
The average set is westward and southwestward, and the rate averages between 10 and
50 knots a day.

SAWAIORI MIGRATIONS. 19

determination of forty-two of the bravest young men and seven
of the fairest maidens (the sacred seven squared by a race to whom
four was the perfect number) to offer themselves in sacrifice to the
plague demon that he might spare the people. They put to sea
in seven canoes, and the plague ceased. The tale now follows these
victims oversea, to an island in the Philippines, where seven were
claimed by the demon. The second seven was overwhelmed in a
gale upon the high sea. The third seven was drawn to abysms of
destruction by a fish monster. The fourth seven got drunk on
toddy at Nukuoro and blasphemed the god until he slew them.
The fifth seven was killed for food on the eastward voyage. The
survivors reached Hawaii and refreshed, then sailed seventy-seven
nights to the south and came to land on Manu'a.

From what source this tale came to Lieutenant Deeken I can not
say. To me it smacks of the ability at fabrication of a half-caste
in Samoa whom I was never able to meet, but whose store of tradi-
tions was tantalizingly reported as truly remarkable. Their worth
may readily be judged from this synopsis of one of them that was
brought me at second hand, namely, that the Samoan ancestors set
sail from "Sumatala" under the leadership of their hero-chief
"Niu-sisila." If one is willing to believe in this preservation of a
recognizable name of Sumatra, what shall be said of the prophetic
instinct which gives to the voyagers a chief already named New
Zealand? Slightly pro leptic. Whatever source may be responsible
for the myth which Deeken records so positively, I should not rest
content without recording its total lack of credibility.

We are now to take up the general migration theory, a division
of our subject which may be dismissed with more summary treat-
ment since it is commonly known.

In the foregoing consideration of the sieve theory it will have
been observed that Thilenius explains the route of peopling between
termini, Malaysia as the point of departure, Samoa or Nuclear Poly-
nesia as the point of arrival. This traverse he covers by a northern
and generally equatorial route. Relative to the same termini the
general migration theory covers the traverse by a southeasterly
course, largely a coasting voyage through or on the fringes of Mela-
nesia. The final link, the distribution eastward from Nuclear Poly-
nesia, remains unaffected by the diverse views herein presented.
Similarly the Indonesian link is common to both theories.

That Malayan or Indonesian link, regarded solely as filling geo-
graphical space, we may safely assume as an antecedent probability ;
yet when we reflect that it has been regarded as a linguistic link,
although this estimate of its value rests upon high authority and
we find the names of Wilhelm von Humboldt, Franz Bopp, and
Friedrich Miiller associated therewith, we must not be carried away

20 THE POLYNESIAN WANDERINGS.

by the weight of the authority. Skilled as were these distin-
guished students, I can not avoid the conclusion that their results
are vitiated by two sources of error : Polynesia was known to them
from scanty and not always accurate information, and Melanesia
was scarcely known at all. In the course of the present work we
shall have to animadvert upon this Indonesian link in its philologic
bearings and, from the material which is about to engage our
attention and exact our best powers of analysis, we shall essay to
draw certain conclusions which will point to the need of revising
the estimate which served as the foundation for the name Malayo-
Polynesian, under which designation these languages of the ocean
have entered into the classification of linguistic systems.

In its most concise form the general migration theory could not
be stated more clearly than in the words of that master of Polynesian
lore, Edward Tregear, recently president of the Polynesian Society :

We must leave the fascinating subject of the whence of the Maori as
an open question, to be settled hereafter when more full and perfect knowl-
edge enables the student of the future to gather up the ravelled strands of
evidence and twist them into a cord that will bear the strain of scientific
investigation. In the meantime the Polynesian Society is doing much to
gather together the facts and preserve the knowledge fading fast with the
elders of the Maori people. It may be of interest to put before the reader
the hypothesis most generally accepted by Polynesian scholars as to the
advent of the Maori in the Pacific. It is as follows:

The Polynesians are a people which either originated in India or in
central Asia and passed through India. Leaving the mainland they
journeyed eastward through the Malay Archipelago, occupying perhaps
many generations in the voyages from island to island. At the time of
their passage the archipelago was not occupied by Malays, who are a sub-
sequent migration from the Mongolian seaboard. The Maori expedition
or expeditions passed by the Melanesian and Papuan islands, inhabited by
black people (New Guinea, New Caledonia, etc.), and reached the Fiji
Group, where they settled for a long time. From Fiji as a center they
colonized Samoa, Tonga, Hawaii, the Marquesas, Mangareva, and extended
their colonies even so far as Easter Island. In process of time they either
hived off or were expelled from Fiji and the waves of migration passed to
and fro among the groups of islands. (Tregear: "The Maori Race,"
page 558.)

So much in the general aspect of the case. Now let us note from
another scholar, S. Percy Smith, the present incumbent of the
presidency of the Polynesian Society, the traversing in detail of
some of the elements of the problem upon which Thilenius founds
his argument. These are cited from "Hawaiki" in its second
edition. I refer to this edition because it is intended as the defini-
tive statement of the author's position and because it will be the
edition most readily accessible. For myself I have a fond prefer-
ence for the earlier edition, the same materials treated in a different
manner. That first " Hawaiki" appeals to me with a personal note

SAWAIORI MIGRATIONS. 21

lacking to the rewritten edition; it was the intimate record of the
periplus which the distinguished author and myself made in Samoan
storm and sun in a tiny boat upon the sea, and of our wanderings
over the northern ocean and on the hot slopes of Hawaii.

Of the objection that early Polynesians could not sail against
the wind he says :

In the present state of our knowledge of the Polynesians as navigators —
about which we shall learn something further on — it is useless for some
writers to insist that the prevalence of the southeast trade winds would
form a bar to voyages made from central Polynesia to the American coast.
The number of easterly voyages on record from various parts and under all
sorts of weather conditions is so large that we must conclude these able
navigators paid little attention to the trade wind if a sufficient object
required them to face it. (Page 40.)

Whatever powers of navigation the people may have possessed prior to
their arrival at Java (Hawaiki), the vast number of islands in the archi-
pelago would induce a great extension of their voyages, and generate a
seafaring life, through which alone were they able at later periods to
traverse the great Pacific from end to end in the remarkable manner that
will be indicated. In the archipelago, where most of the islands are forest-
clad to the water's edge to this day, the water was the principal highway
and this necessitated constant use of canoes; whilst the location of the
various branches of the people on different islands with considerable spaces
of sea between would induce the building of a larger class of vessels. It
certainly seems from the very nature of the surroundings that Indonesia
was the school in which the Polynesians learned to become great navigators.
(Page 99.)

Having now presented the two opinions on this vital point, this
seems a fitting spot in which to record a flaw in the reasoning of
Dr. Thilenius which will have suggested itself already to the reader.
He is ready enough to admit the possibility of early Polynesians
navigating against a head wind from Samoa eastward, while denying
their ability to perform the same sort of navigation toward Samoa
from Indonesia. Samoa was no Annapolis for this race of seamen;
the skill of seacraft which carried them for thousands of miles over
eastward ocean was the skill which had brought them over balanced
thousands of miles of westward waters.

Where Tregear has outlined the disputed section of the route in
general, yet wholly unmistakable, terms, Percy Smith is particular,
as we shall see.

Starting from Avaiki-te-varinga, which is probably Java, the route
followed by the migrations would be via the Celebes, Ceram, and Gilolo,
where, no doubt, were colonies of their own people, to the north shores of
New Guinea. Finding this country already occupied by the Papuans
they would coast along to the southeast end, where, it would seem, a very
early migration settled, which is now represented by the Motu and cognate
tribes. This same route was probably followed by the ancestors of the
Rarotongans until they branched off past New Britain and the Solomon

22 THE POLYNESIAN WANDERINGS.

Islands on their way to Fiji, probably leaving a colony at Sikayana, or
Stewart's Island, off the coast of the Solomons, where the people speak a
dialect of Maori or Rarotongan and are Polynesians. Whether Lord
Howe's Island, or Liueniua, also called Ongtong Java, was peopled at the
same time is uncertain. It is inhabited by Polynesians, as Mr. Churchill
tells me. Possibly Nukuoro and Lukunor were also colonized at this time.
In more than one Rarotongan tradition an island or country is mentioned
named Enua-kura, or the "land of the red feathers," which is possibly New
Guinea, so called by the Rarotongans after the bird of paradise, the beau-
tiful feathers of which would be to them treasures of the highest value —
such treasures as Europeans who do not know the race can hardly believe
in; they were their jewels. (Page 113.)

The same indefatigable research supplies us yet another argument,
a persisting memory of intercourse with an inferior and servile race.
Because of its length I am forced to omit a few details :

Again, there ought to be traces of some recollection of the black or very
dark-brown negrito races of Indonesia. In the Maori traditions there are
incidental notices of an ancient people called Manahune or Manahua, who
are by some supposed to be a diminutive race and somewhat like the elves
of Old World stories. But they are not said to have lived in New Zealand.
This people is also known in Hawaii under the same name, where they are
described as somewhat like those of the Maori traditions. They appear at
one time to have been very numerous and lived in the mountains, but were
in a state of subjection to the Hawaiians. Again in Tahiti we find mention
of the same people, Manahune, who in Ellis's time formed the lower orders
of the people, but they were an ancient tribe or people. In a Paumotu
genealogy in my possession I find one of their chiefs named Tangaroa-
Manahune, who lived many generations ago ; and it is known that there was
a tribe in old times in Mangaia named Manaune. We shall find later on a
reference to them in Rarotonga history, where they are again referred to as
little people. The word manahune, both in Maori and Rarotonga, means
a scab or mark on the body. It may be that the origin of the name is due
to the people who bore it being marked with cicatrices. The vague notions
the Polynesians generally now have in regard to the Manahune — their
living in the mountains and forests, the wonderful powers of sorcery, etc.,
accredited to them — seem to point to their having been a race living in the
remote past, conquered by the Polynesians, and probably often enslaved
by them. In fact the traditions no doubt point to the Papuan or Melane-
sian race, who, it is well known, mark their flesh in gashes as an ornament,
instead of tattoo as with the Polynesians. The same Nga-Puhi tradition
goes on to state: "Some of the people of those parts were very black, a
people who smelt very strong when near, their hair was bunched out to be
stiff and appeared in tufts, and their appearance was ill-favored." This
is in brief form a fair description of a Papuan or Melanesian. (Page 103.)

I have preferred to use the words of these two great authorities
because they are authorities and because my own conclusions as
to the two theories will more properly be presented in the discussion
of the pertinent linguistic material.

As between the two theories, we must recognize that each is an
attempt to close a gap, the gap between Indonesia and Polynesia

SAWAIORI MIGRATIONS. 23

in the race history. Thilenius works it out painfully but coldly, with
every resource that may be drawn from the armamentarium of the
ocean physiographer. He neglects wholly the record of the only
folk who retain any recollection of this great ethnic movement, the
corpus of Polynesian tradition. This we shall summarily examine.
In the Rarotongan accounts of the voyages and discoveries of Ui-
te-rangioro we find the following list of new lands :

Te Ravaki.

Nu-taara.

Nu-amo.

Iti-takai-kere.

Avaiki.

Rangi-raro.

Nu-mare.

Iti-nui.

Papua.

Kuporu.

Mata-te-ra.

Nu-pango.

Iti-rai.

Tangi-te-pu.

Te Tuira

Nu-kare.

Xu-iti.

Iti-anaunau.

Rara.

Manuka.

Nu-takoto.

These last we have no difficulty in comprehending as Savai'i,
'Upolu, Tutuila, Manu'a — Samoa, in fact, plainly named. A little
earlier in the list the four which bear specific names of Iti are quite
as clearly the Viti Archipelago. These two points establish the
direction of the voyaging, a direction which in the long remainder
of the list, after the mention of Samoa, covers the eastern Pacific.
It is not improper, then, to reason back in the same direction from
our two known points and assign to the unidentified places a
position somewhere between Viti and Indonesia.

CHAPTER IV.
THE FIRST POLYNESIAN HOME.

A race always under an eastward momentum — No positive statement
of the place of origin possible — Bopp proposed an Aryan source — Max
Miiller connected it with the Turanian stock — Logan regarded the
Ganges Valley as the ancient home — Macdonald's Semitic theory of
a great and widely diffused Oceanic language set forth at large.

Noting that the Indonesian area presents its own group of prob-
lems relative to the Polynesian, which have yet by no means come
to a satisfactory solution, we pass for the present to a summary
statement of the theories which have been proposed in elucidation
of the Polynesian migration before its entrance upon the Malayan
region.

Wherever we know this race we find it under a momentum
directed eastward, under the impulse of some power behind, which
has sufficed to overcome the inertia of a race settled as autochthons.
But what that power may have been, what the place of autoch-
thonous settlement, we are without positive information.

Beyond peradventure we recognize the momentum toward eastern
Polynesia from Nuclear Polynesia. In Nuclear Polynesia we recog-
nize the momentum, some of it from Viti. In Viti we are more
and more distinctly certifying ourselves of the same momentum
exerted along intervening steps of the Melanesian archipelagoes
from the Malayan exits. In Indonesia we find the same momentum
from Sumatra toward the east. But behind Sumatra, the Malayan
entrance, while we may believe in the impulse, we are at a loss to
find the next earlier point from which dislodgment was made. Yet,
as the direction sense remains constant, we need have no hesitation
in looking toward the west.

The first great guess at a point of origin we owe to Bopp in his
classic study (1841) "Uber die Verwandschaft der malayisch-poly-
nesischen Sprachen mit den indisch-europaischen." His effort was
to establish these sea wanderers as an early, perhaps the earliest,
offshoot from the stock which, after long wanderings otherwhither,
has produced our race.

After discussion of the same material Max Miiller evolved the
theory that the sea people, by then definitely accepted as Malayo-
Polynesian, were related to the Turanian through the Thai of Siam.
Later students along the same line have reached a similar conclusion
through the Mon-Khmer.

26 THE POLYNESIAN WANDERINGS.

Percy Smith (in "Hawaiki" passim) follows the conclusions of
J. R. Logan in seeing the more probable seat of the early Polynesians
in the Ganges valley. Logan himself says, in his " Ethnology of the
Indo-Pacific Islands:"

I was especially struck with the constantly accumulating evidence of
the derivation of the leading races of the islands (Indonesia) from Ultra-
india and India, and was led to the conclusion that the basin of the Ganges
and a large portion of Ultraindia were occupied by tribes akin to the Malayo-
Polynesians before the movement of the Aryan or Indo-Germanic race into
India.

A survey of the character and distribution of the Gangetic, Ultraindian,
and Asianesian (Indonesian) peoples renders it certain that the same
Himalayo- Polynesian race was at one time spread over the Gangetic basin
and Ultraindia. As this race is allied to the Chinese and the Tibetan, it
is probable that it originally spread from Ultraindia into northeast India.

We shall now proceed to the presentation of Dr. Macdonald's
Semitic theory, premising that in his writing Oceanic race and
Oceanic language are an anticipation of the proof of his deductions.
This he explains in the following form:

These three groups of languages and dialects — the Malayan, the Poly-
nesian and the Melanesian — naming them in the order in which they have
successively become known, are, as Friedrich Muller has shown, members
or branches of the Oceanic, which is as perfectly well-defined a family of
languages as is the Semitic or the Indo-European. The Oceanic is, as its
name indicates, insular. Its habitat, which we may call Oceania, stretches
from Madagascar off the east coast of Africa, across the Indian Ocean to
the Malay Archipelago, and on through the Pacific Ocean to Easter Island.
On the north it has invaded from the island world and settled only on the
southeastern extremity of the Asiatic continent, hence called the Malay
Peninsula. On the south it has not reached the Australian continent,
though closely approaching it in New Guinea. The islanders who speak
Oceanic number about fifty millions, or one-thirtieth of the human race.

To say that the Oceanic languages are a perfectly well-defined family is to
say that they are all sprung from one mother tongue — the Oceanic mother
tongue ; and to establish the Asiatic relationship of the Oceanic is to estab-
lish that that mother tongue was originally carried by its speakers from
the Asiatic continent to the island world. * * *

It is not until we take into account the linguistic data that we get upon
the solid ground of certainty. And first of all it is to be observed that
though there was an element of negro blood in the race, due to intermixture,
the race itself, as its language proves, was not negro. What that race was
can only be determined from its language, and what that mother language
was is to be learned from an examination of its descendants and repre-
sentatives, the spoken Oceanic languages and dialects of the present day.
If the race came from the Arabian Peninsula, the Semitic motherland,
sprung from the people of the commercial empire that existed there, then
their language was Semitic. For the Phenicians, the people of that ancient
South Arabian empire and of their Abyssinian colony, and their descend-
ants now in Abyssinia and Arabia, all are Semitic speakers. If the race

THE FIRST POLYNESIAN HOME. 27

came from the Indian Peninsula one might suppose with Bopp that the
language was Indo-European; if from the Indo-Chinese Peninsula, with
Max Muller that it was Scythian or Turanian. The problem thus, as
is clear, can only be solved linguistically; and the praiseworthy efforts
of Bopp and Muller to solve it are valuable if only as having led to the
certainty that the Oceanic mother tongue was neither Indo-European nor
Turanian. Their attempts failed because made on insufficient data, and
their methods were for the same reason inadequate * * *

When we say that Arabia was the motherland of the island family of
languages, this does not mean that the primitive Oceanic tongue — of which
the multitudinous dialects of Oceania as at present spoken are the analytic
or simplified descendants, as English is of Anglo-Saxon, or the Romance
dialects of Latin — was derived from the Arabic ; but that Arabia was the
motherland of the primitive Oceanic as it is of the Ethiopic, Amharic, and
Tigre, and of the Assyrian, Phenician, Hebrew, and Aramaic. If it had
more in common with Arabic than with any other Semitic language, that
is because Arabic has more than any other preserved the features of the
primitive Semitic tongue, the common mother of all of them. The primi-
tive Oceanic must be regarded, not as a descendant of, but as a sister to
the Arabic, Himyaritic, Ethiopic, Assyrian, Phenician, Hebrew, and Ara-
maic and the Efate, Samoan, Malagasy, Malay, etc., as cousins to the Mahri,
Amharic, Tigre, Mandaitic, Modern Syriac, and vulgar Arabic dialects, due
allowance being made for the fact that these latter have always been
more or less under the conserving influence of the surrounding Semitic lit-
erature and civilization, from which the island dialects have been for ages
completely cut off, as well as completely isolated from each other.

It should be premised that Dr. Macdonald's argument, loaded with
minute details and seldom stating clear principle, is nowhere lucid.
The profundity and the breadth of his Semitic erudition will be
estimated by each student in proportion as he is fitted to pass crit-
ically upon such topics. From the confusion of statement tangled
with partial proof involving more statement with yet more proof I
have endeavored to present a simple syllabus of his argument :

A.

The gist of our author's chapter on Oceanic phonology, a plexus
of multitudinous detail, is this :

i. Letters interchange freely within their own vertical series.

2. Letters interchange freely as between series and series.

3. Letters conveniently efface themselves, whether initial, medial,
or final.

4. Initial syllables may drop off.

5. Initial syllables may be added " to lighten the pronunciation."
Granting all or most of these postulates, it will be seen that no

particular limit need be set to philological comparison. For such
proof as these positions seem to require the student will have to
follow out the intricacies of Dr. Macdonald's argument to such con-
viction as he may be able to discover.

28 THE POLYNESIAN WANDERINGS.

B.

Dr. Macdonald now plunges into the critical test, that triliteralism
of the root which has stamped the Semitic as a speech apart. It is
so vital to his contention that the thesis must be presented in his
own words:

It is now to be shown that the Oceanic primitive language had, like each
of its sister dialects, Arabic, Assyrian, etc., its share of the common stock
of purely and exclusively Semitic trilateral words (nouns and verbs) with
the purely Semitic common method of word formation or inflection by
internal vowel change and external additions (prefixed, infixed, suffixed)
and its share also of the limited common stock of purely Semitic particles.
This, if it can be shown, will be admitted to be conclusive. The particles
will be dealt with subsequently.

The ancient Semitic finite verb, with its perfect and imperfect, was
simply a verbal noun joined in a certain way with the personal pronouns,
and with it or from it other and numerous verbal nouns were formed by
vowel changes and external formative additions. The ancient finite verb
with its perfect and imperfect so formed is no longer found in the existing
broken-down Oceanic languages, though as analytic substitutes for it we
have as the finite verb, for instance, in Efatese "the verbal pronoun"
joined with these verbal nouns after the fashion of the imperfect ; as a bano
I (amor was) going, equal to I go (or I went), and in Malagasy the "pro-
nominal adjunctive" joined with these verbal nouns, after that of the
perfect, as tiaku, my loving, equal to I loved or I love. The verbal nouns
that were formed with or from that of the ancient finite verb were numerous,
and in them we have the ground forms of the modern Oceanic verb. * * *

We now proceed to compare the Oceanic triliteral words with Arabic,
Assyrian, etc., just as, for instance, we compare, say Assyrian or Himyaritic
words with Arabic, Hebrew, Syriac, or Ethiopic.

Take the simple Efate lifai, to bend round ; malibai, bent ; lofa, a thing
bent; lofai, to bend; malofa, bent; kalofa or kolofa, bent; lufa (Samoan
lavalava), a wrapper round the loins; Samoan lofa, to crouch; lofata'ina,
to cause to crouch; lave, lavelave (Arabic la/elafa, to wrap round, etc.), to
entangle; lavelavea, to be entangled; Fiji love, loved ha (Samoan lavasi), to
coil, to fold, to bend; kalove, bent; salove, flexible; Malay lipat, lampit,
lampis, lapis, a fold, to fold, to plait; Malagasy lefitra, also lufitra, folded,
bent, plaited; Arabic lafja, to be involved, intertwined, to wrap up, wrap
round (oneself, as clothing), to fold; laff, lifj, laffat, lifjat, involved, inter-
twined, etc. ; lofja, loffat, coil of turban, winding of road. In this example
the six commonest forms of the modern Oceanic verb (or noun), the ancient
verbal noun, are seen, viz :

i. lave. 3. lofa, love, lufa. 5. lipat.

2. Ufa. 4. lampit, lavasi. 6. lovedha.

The inference is irresistible that in the Oceanic primitive or mother
tongue this word was triliteral, and had the vowel changes peculiar to the
Semitic languages most fully preserved in the ancient Arabic ; and that as
a triliteral word with the middle radical doubled it underwent the usual
contractions, set forth in all Semitic grammars, of such words, as is plainly
seen by comparing it with the Arabic. These forms, originally verbal
nouns and still often used as such, formed from the ancient finite verb,
as lipat, a fold, lofa, a thing bent or bending, have become ground forms

THE FIRST POLYNESIAN HOME. 29

of the modern verb, as lipat, lipatkan, to fold, lofai, to bend ; from which
again are formed, by external additions, modern verbal nouns and derived
verb forms. Thus we have lipatan, a fold; lofaian, a bending or being
bent; lavelavea, entangled or entangling, malibai, bent; and the derived
verb forms —

Safal Fiji salove, flexible.

Mafal Malay malipat, to fold, plait; Efate malifus, bent, flexed.

Mifal Malagasy milcfitra, folded.

Tafal Fiji kalove, Efate kalofa, bent.

Manfal Malagasy mandefitra, to fold, bend.

Matafal Samoan fa'alave, to take a turn of a rope as round a pin.

As seen in this example the vowels of the ground forms of the Oceanic
verb are retained in the modern derived forms and verbal nouns. It is in
the ground forms, therefore, that we find the proof of the part played in
the ancient language (the primitive Oceanic) by internal vowel change.

To show that this is a fair specimen of modern Oceanic words, that it is
not exceptional but only one out of the mass and of a piece with the rest,
would prove conclusively that the Oceanic primitive or mother tongue
had, like each of the sister dialects, Arabic, Assyrian, etc., its share of the
purely and exclusively common stock of Semitic triliteral words with the
purely Semitic common method of word formation or inflexion by internal
vowel change and external additions. This then is what we have now
to endeavor to show.

This he follows by an example of intricate ingenuity in bringing
many insular words under one or other of eight types of triliterals ;
and thus concludes:

These examples sufficiently show that the above Oceanic word first
given (lifai, etc.) is not exceptional, but only one out of the mass and of
a piece with the rest, and this conclusively establishes, etc. [in the same
form of words as before, a protracted Q. E. F.].

C.

He next takes up the subject of prefixes, infixes, and suffixes.
Employing the freedom of phonetic treatment which he has already
postulated in his chapter of phonology, he identifies every form of
these inflectional or word-forming elements known to Semitic speech
with word components in Malayan, Melanesian and Polynesian. The
value of such identification is wholly conditioned by the adhesion
which one inclines to give to his system of sound mutation.

D.

The last chapter of the presentation of the theory rests upon
the pronouns and particles, the demonstrative elements of speech.
With remarkable patience in research, and with even more remark-
able generosity of treatment, he satisfies himself that he has detected
the same kinship in this section of the compared vocabularies.
Yet, while permitting himself the use of this material and erecting
much of the edifice of his argument thereupon, he takes the similar

30 THE POLYNESIAN WANDERINGS.

use of the same class of data by other students as the basis upon
which to combat their conclusions when they differ from his own.
He says of his predecessors in this field :

They trusted mainly if not wholly on the comparison of words, chiefly
the pronouns and numerals, in which there is always great liability to
error, and which apart from comparison of grammar and structure can
never be conclusive. As to the pronouns, for instance, Bopp, and Max
Muller following him, chose to regard the Malay kita, kami, we, and kamu,
ye, as composed of an article ki, or ka, and the pronouns ta, mi, mu. This
enabled Bopp to compare the latter with the Indo-European pronouns,
and Max Muller, it should be added, to compare them with equal prob-
ability or improbability with the Turanian ; and by this method the Oceanic
pronouns might just as well be compared with any others whatsoever. The
fact is, as the Melanesian clearly shows, that this ki, or ka, is not an article
at all, and that this comparison of Bopp, and also that of Muller, founded
on the notion that it is, is illegitimate and futile.

In this chapter it has been my aim to present the line of Dr.
Macdonald's argument as simply as possible; to avoid, wherever it
might be done, the complexity and intricacy of the detail of his
method; and in general to refrain from debating the controversial
points which arise, the latter pleasure being reserved for the critical
examination of his material, upon which now we shall enter.

CHAPTER V.
DISSECTION OF THE THEORY.

The result of independent treatment of the data should be identical — ■
As the other elements have long been known and carefully studied,
the Melanesian is the critical test — The computation of exactly
what material is now made available for study — In what pro-
portion the Efate vocabulary contributes to the solution of the
problem and the manner in which properly it may be employed.

The force of Dr. Macdonald's argument, the proof of the theory
thus summarily outlined, must rest upon the data which he has
utilized for its development. Those data we find available in the
vocabulary which he has given us as a complete dictionary of Efate.
He has made use of this material in a certain fashion, such as most
commended itself to his thought ; he has prosecuted long and pain-
fully a certain method. All this was within his prerogative. If
the theory be valid, if the method be true, the same data should
yield the same result, and no other than the same result, when
studied in accordance with such other method, being valid, as may
commend itself to another philological investigator.

This it is which is now to engage our attention. We are to take
his data, his vocabulary material assumed to be in itself accurate,
to argue it afresh and, I feel confident, without preconception or
other such prejudice, and let it lead us where it may.

If really there be an Oceanic speech family tree, with its roots in
the Hadramaut and its distal twigs in Te Pito te Henua, then we
may expect to find in the material in the course of this examination
such a mass of words showing clearly a nexus of development —
enabling us thereby to establish a distinct and probable law of the
mutation of sounds — that we may finish the investigation with the
happy satisfaction that the Oceanic tongue has proved itself.

We are not to ask that every Polynesian word shall reveal to
us through the operation of this law of mutation its primitive
triliteron in some Semitic household. We are by no means to
expect that we can take any Semitic stem and by the application
of the rule develop the succeeding forms in Indonesia, Melanesia,
and Polynesia. Not even Grimm's law will do that for us in the
Aryan family. But we do have the right to expect that, if there
prove to be a substantial base for this theory, there be a sufficiency of
examples in each direction and a consistency in their establishment.

31

32 THE POLYNESIAN WANDERINGS.

Dr. Macdonald has presented a captivating theory of an Oceanic
family wandering from a Semitic home, bold sailors over unknown
seas and far from the tents of Kedar. This theory must rest on
the data which he presents to us. If it be valid, the data must
yield the same result to independent examination.

This Oceanic speech family comprises four households. The
Polynesian is known intimately, the Malayan has been even more
extensively studied, the Semitic has engaged the attention of the
most able scholarship for centuries. The Melanesian is now just
being admitted to the family circle; it has been scarcely known
until the publication of this Efate work.

But the position of this new household is significant geographi-
cally, for it bridges the gap between Indonesia and the nearest point
of Polynesian culture. We are therefore warranted in expecting
Melanesia, on this theory, to establish its articulation with the
Malayan speech group in one direction, with the Polynesian tongues
in the other, and internally we look to find some thread of inter-
relation between Efate and such other languages of Melanesia as
are known at all. In the order of antecedent probability, a mathe-
matical deduction as generally applicable as an algebraic equation,
these associations should be arranged in the following order :

Melanesian interrelations ; the most frequent.

Indonesian and Polynesian articulations; either equal in

frequency or exhibiting a slight preponderance toward

the Indonesian.

These two terms are immediately limiting. Established cases in
which the two terms are combined will have a great weight. Similar
weight will attach to examples which are found common to Mel-
anesia, Indonesia, and the Semitic. The examples which can be
established in an indisputable chain from Polynesia through each
link back to the Semitic will be compelling evidence.

These are the matters which we are to look for in the exploration
of the material, in independent study of the Efate dictionary.

In an earlier chapter I mentioned the number of the dictionary
entries, 3,657, and I gave that as the result of a count seriatim.
The operations of arithmetic did not cease with the attainment of
that figure. I made a table of all the words for which Dr. Macdonald
claimed relationship with other households of his Oceanic family.
This table I now present, for it contains matter of note. In it I
have cast out the purely demonstrative words. Since I have shown
that Dr. Macdonald is willing to base his own theory upon them to
a certain considerable degree, but is averse from allowing their use
to other inquirers, I think it preferable to put them aside lest I be
classed with Bopp and Max Miiller.

DISSECTION OF THE THEORY.

33

In this table I have made use of the convenience of abbreviations,
M for Melanesian, V for Viti, P for Polynesian, My for Malayan,
S for Semitic. With each entry the name Efate is to be understood.
Thus the first entry in the table, M 31, signifies that Dr. Macdonald
notes 31 words in which he finds for Efate relation with some lan-
guage elsewhere in Melanesia, but without recognizing further affilia-
tions. Similarly the final entry signifies that he has found 29 words
in Efate for which he provides identification with words along the
whole chain from Polynesia to the Arabian Saba.

In such a study as this the position of the Fijian speech is anom-
alous. It is neither wholly Melanesian nor pure Polynesian. As in
part it may be claimed for each household, there seemed but one
way of securing fair treatment wherever it chances to be involved
in this investigation; namely, to set it apart and in each instance
to sift out the Melanesian or the Polynesian affiliation.

M

V

P

My

S

M

31

V

22

P

67

My 43

S 346

MV

3

MV

3

MP

9

MMy 10

MS 31

MP

9

VP

9

VP

9

VMy 7

VS 30

MMY

10

VMy

7

PMy

17

PMy 17

PS 103

MS

31

VS

30

PS

103

MyS 100

MvS 100

MVP

1

MVP

1

MVP

1

MVMv 0

MVS 6

MVMy

0

MVMy

0

MPMv

3

MPMv 3

MPS 22

MVS

6

MVS

6

MPS

22

MMyS 28

MMyS 28

MPMy

3

VPMy

9

VPMy

9

VPMy 9

VPS 25

MPS'

22

VPS

25

VPS

25

VMyS 26

VMyS 26

MMyS

28

VMyS

26

PMvS

95

PMvS 95

PMyS 95

MVPMy

4

MVPMy

4

MVPMy

4

MVPMy 4

MVPS 8

MVPS

8

MVPS

8

MVPS

8

MVMvS 6

MVMvS 6

MVMvS

6

MVMyS

6

MPMyS

24

MPMvS 24

MPMyS 24

MPMyS

24

VPMyS

44

VPMyS

44

VPMvS 44

VPMyS 44

MVPMyS

29

MVPMyS

29

MVPMyS

29

MVPMyS 29

MVPMyS 29

Total

217

229

468

443

923

P. ct. 5

■93

6

.20

12

■ 79

12 . 1 1

25-24

The total of the words for which his ingenuity has suggested
affiliations outside the island of Efate is 1,154, or 31.55 per cent
of this dictionary, which he calls complete. That is to say, for not
so much as one-third of the vocables in the language of Efate can
his liveliest fancy — how lively that may be we shall see hereafter —
find any sort of ground for his promise of "the Oceanic languages,
their material or vocabulary set forth in a complete dictionary of
one of them." Furthermore we find that an even smaller number,
in effect no more than a quarter of the whole number of entries,
covers all the suggestions of Semitic affiliation which he has ven-
tured upon. Compare his claim that no more than 29 words establish
the complete chain from Arabia to Samoa with his identification of
346 words which connect Efate with the Semitic without having
left a trace elsewhere in Melanesia or anywhere throughout the

34 THE POLYNESIAN WANDERINGS.

varying languages of Indonesia. As soon as he attempts to link
Indonesia in the backward bight of the chain, his figures fall to
ioo, and when he links in Melanesia he finds but 28 available words.
Accepted even at his own valuation, philo logically a most faulty
one, these are but small foundations for so lofty an erection of a
Babel building which is to confound Max Miiller and other names
by no means inconsiderable in systematic philology.

In these studies my position is eastward. From Polynesia I am
looking through all available linguistic material toward the west,
hoping to find some data that will establish the partition of the
Polynesian and the Melanesian elements in Viti, welcoming any-
thing that promises to enable us to comprehend the Melanesian in
areas where its proportion in speech is the more considerable,
rejoicing at the accession of new knowledge which will equip us
to give better study to the possibility of some manner of proof that
kinship does or does not exist among Polynesian, Melanesian, and
Indonesian, convinced that only as we establish these three links
of the chain in positive knowledge can we approach the yet earlier
history of these tongues with clear sight.

In Dr. Macdonald's material I see three classes of data.

1. A central area, Melanesian material for which we have not yet
established connections with either of the known areas, Polynesian
and Indonesian, respectively. We therefore want the means to
subject this to comparative study, and on this account it remains
unavailable.

2. An eastward extension, through so much of Melanesia as is
known, to the well-established knowledge of Viti and Polynesian.

3. A westward extension, through Melanesia to the equally well-
established knowledge of the Indonesian; and, yet more remote,
Dr. Macdonald's confident projection to the Semitic.

The second class is that which shall provide the instruments for
our study of these data upon which we are now to engage at much
length and with the utmost attention to the minuteness of detail.
The Indonesian languages, and in yet greater measure the Semitic
tongues, have their own enthusiastic students, and to such we may
confidently leave the prosecution of similar research from the
vantage-ground of their own knowledge, in case this ambitious
theory should seem at its western extremity as proper a subject of
debate as it does to those of us who are engaged at the eastern
extremity.

In the following detailed study I have purposely omitted many
examples, for simple inspection will show our author patently in
error. At the same time I have endeavored to afford room for all
such as seemed at least debatable. The process of elimination will
be continued in these studies.

CHAPTER VI.
EFATE AND VITI AND POLYNESIA.

Comparison of the vowels of the Melanesian element of Efate and
Viti — The establishment of consonantal variety — The assumption of
a parent speech from which these deviate — The Viti appears to be
the younger son — The comparison continued through the Sawaiori
element of Viti — Viti, Samoan and their parent speech, the Proto-
Samoan — Comparison of Efat6 with Polynesian and summation of
results— The extent to which Efate identifications penetrate into Poly-
nesia— Argument from recorded anomalies — It is improbable that
Efate received its Polynesian content through westward drift of casta-
ways— Proof that this element came through the migration of Proto-
Samoan wanderers — Check-list of Polynesian phonetic mutations.

At the eastern extremity of the language groups which it is sought
to associate into this Oceanic family we have two firm, if unsaintly,
foundations, the Polynesian (or Sawaiori of Whitmee's proposed
nomenclature, convenient even if not wholly acceptable), and the
Viti. Such element of the Viti, somewhat less than half of the
vocabulary according to my estimate, as is not identifiable with
the Sawaiori we assume to be of Melanesian origin. The larger
element we shall consider in its clear connection with the Sawaiori.
We shall examine the Efate first in its relation to the Melanesian
Viti ; thence we shall proceed to the examination of the Efate in its
relation to the Sawaiori with the inclusion of so much of the Viti
as properly pertains to that stock; last of all to the consideration
of the element of Efate which appears in the Sawaiori without
having left an impress upon the Viti.

The data 1-47 in Appendix I enable us to complete our con-
spectus of the material in Efate which is identifiable with that
element in the carefully wrought-out Viti vocabulary which has not
been identified with any of the equally familiar tongues of Polynesia.
Two explanations here are possible: the former that this is truly
a portion of the Melanesian component of Viti; the latter that this
element is Polynesian, but that it has failed of preservation in the
eastward languages. The former we adopt provisionally as by far
the more probable.

Now let us sum the observations as to the phonetic relations of
Efate and the Melanesian component of Viti, and first the vowel
system.

By far the largest mass of vocalic dissimilarities in the data under
study lies within the area of the neutral vowel. I have* already

*I7 Journal of the Polynesian Society, 87.

35

36 THE POLYNESIAN WANDERINGS.

had to deal with this matter in the consideration of the phonetics
of the more strictly Polynesian languages. The same holds true,
mutatis mutandis, of these two Melanesian tongues. I cite the
earlier discussion of the subject :

A man with a quick ear and an obedient tongue may, as the result of
long discipline, acquire almost perfect use of the Samoan consonants, but
it is most probable that no Caucasian has really mastered the art of the
Samoan vowels. It is as in their music : the intervals, the supertones, and
the fractions of the tone are developed on a system which we find it
impossible to acquire. It establishes a new group of units of vibration
of the vocal cords, for which the fundamental diapason of our own speech
is not set in unison.

With this in mind, we shall find a plain explanation of the central triangle
of the vowel changes if we regard the short a, e, o as merely so many
approximations to a primal obscure short vowel which lies centrally situ-
ated in respect of these three apical points. One congeries of the Polyne-
sian tongues may have had a vibration series and period which inclined
its use of the primal obscure vowel somewhat in the a direction ; to another
congeries the e component was the more grateful; to yet another the
tendency was in the o or labial grade. * * * Thus we have no hesitation
in taking this central triangle of d-e-o out of the group of vowel changes
in Samoan, of regarding it as no more than a doubly muffled rendering of
a single central sound, and of removing it entirely from consideration
among the criteria of vowel changes as dialectic indicia.

When we diagram upon the common alphabetic scheme the vowel
changes not in this class, we find some interesting developments.
Along the palatal strut, that which rests upon i and peaks in a,
in the data thus compared we find but a single instance of the
dialectic vowel change a-i in bila (12) to pick up Viti vili, and of
i-e in seri (22) to loose a tabu Viti sereka to untie; and the weight
of these instances is considerably lessened by the fact that they lie
in an unaccented syllable, and a terminal one at that, unaccented
terminal vowels in Viti being evanescent.

On the labial strut, from a to u, most of these vowel changes are
seen to lie. The maximum frequency, four instances, is found
between o and u, the change u-o is seen in bure (14) to wash Viti
mborea; lume a (19) to dip Viti lomotha; and suba (33) to break
Viti sovetaka; the change o-u in mako (41) offspring Viti makumbu
grandchild. A change over a slightly longer interval, o-a, is found
in trbm (38) turmeric Viti ndamu, red. A still longer gap is found
in a-u in bera (27) to crumble Viti vuruvuru; and in the reverse direc-
tion, u-a, in mutrei (5) breadfruit cake Viti mandrai.

There remain three changes which vary from the foregoing simple
system of vocalic mutation along one or the other of these struts,
for they cut across diagonally. Of the change e-u we have two
examples, in tefa ki (24) to range Viti tuva, and in bera (27) to
crumble Viti vuruvuru. The change i-o is found in bo ri (28) to
break Viti vorota; and u-i in lubwa (30) to pour out Viti livia.

EFATE AND VITI AND POLYNESIA. 37

We now proceed to the examination of the consonant scheme of
the two languages. They thus appear in the diagram:

Efate. Viti.

Semivowels y r, 1 w y r, 1 w

Nasals ng n m ng n m

«*-te:::::::::::::::::::= 7 =7

w— {ST:::::::::::::::::::*^0? ---

Miitw /Sonant g — b ngg nd mb

Mutes \Surd k t — k t —

(kw-bw-kb)
(ngm-ngw-mw)
ts
tr

The constant consonants are ng, k, I, n, s, m; the constant muta-
tion is f-v. It does not seem advisable, in the limited supply of
data, to essay the quantitative weighting of the mutations; but in
qualitative examination it seems altogether permissible to deduce
a scheme of the consonant skeleton of the common ancestor from
which derive the Efate and the Melanesian component of Viti. This
we shall designate the Efate- Viti parent speech, and for convenience
shall refer to it by initials E-V. In this diagram we present in each
triple entry the Efate consonant at the left, the parent in bold-face
type, the Viti at the right:

Efate-Parent- Viti.

r-r-r, 1
1-1-1
tr-r-ndr

ng-ng-ng n-n-n m-m-m

dh-dh-dh b-v-v

(ch) — f-f-v

g-g-ngg t-d-nd b-b-mb

k-k-k t-t-t b-p-mb

The characteristic Melanesian compounds, kw-bw-kb, ngm-ngw-mw,
and ts are left unplaced in reference to E-V by reason of the fact
that this particular group of data affords no opportunity to link them
to the known physics of the Viti consonant structure. Abundant
material for their intimate study will be afforded us later in the work.
For the present, our comments on the results thus far attained shall
be summary; the discussion will properly follow upon the similar
analysis of the remaining data, so much more voluminous.

If we examine the diagram attentively we note the particular
features in which the offspring equally favor the parent. In the
upper portion the resemblance is perfect; the nasals and sibilants
are the same; so, in effect, are the semivowels; so the sonant lin-
gual spirant, and of the surd mutes the palatal and the lingual. In

38

THE POLYNESIAN WANDERINGS.

the palatal series and again in the lingual series only one of the two
inheriting languages falls away from the ancestral estate and that
in but a single item in each series. In the labial series both lan-
guages have lost, and in equal measure, a possession of the parent.
In the comparison of the two junior languages we shall see that the
Efate* more nearly takes the property as heir-at-law, and the Viti is
struggling to keep a younger son's uncertain hold upon the family
possessions. The proof of this lies in our comprehension of the
inward significance of the reinforced or prefaced consonants.

Both languages show the impossibility of reproducing the uvular
r without a conscious effort to produce a sound for whose expression
the buccal organs lack training. The effort takes the same form;
Efate* with its tr shows a shade less exertion to be necessary than
the Viti ndr. In all other cases where Viti employs the nasal of
the same series, namely, in the three sonant mutes, it is seen that
Efate is able to take the sound without assistance. This makes it
clear that Efate, lying relatively toward the west, closer approaches
E-V than Viti, which lies so far at the eastern verge of Melanesia as to
have been overlaid deeply by Polynesian sounds, forms, and usages.

So far as these two points may determine the line, so far as the
comparison of the two tongues shows the direction along that line,
we look for E-V at some point on the line produced beyond Efate.
Now the inspection of our charts shows that the line produced
westward beyond Efate is immediately drowned in empty sea. We
may, then, look for the E-V parent either among the islands lying
south of the central New Hebrides or among those lying in the
northern area. So far we have considered but these two points;
we need new points to establish to our satisfaction what becomes
of that line when once it turns the corner at Efate. The remaining
data, which we are now to study, will tend to acquaint us with
these points. That is the definite object which the close scrutiny
of this store of newly available Melanesian material has presented
to my investigation.

Table A.

Viti.

Proto-
Samoan.

Samoan.

Viti.

Proto-
Samoan.

Samoan.

Viti.

Proto-
Samoan.

Samoan.

a

:}

e
i
i
e
u
a
o

a

(Obscure
vowel)

e

0

a

II

e
i
e
i
i

o

u
u
o
a

y

r

ndr
1, n
ng, v
n, ng
m, ng

u

y

r
r
1

ng

n

m

o
u
u
u

1

1

1

ng

n

m

h

dh

dh

s

mb

mb, v
ngg,k,dh
t, nd, k,

dh, s

mb, v

h
h

s

s

V

f

k

h

p

s
s
s

V

f

t
p

EFATE AND VITI AND POLYNESIA.

39

We shall next pass under review the comparison of Efate with
that element of Viti which has been proved to be Polynesian. The
data for this study will be found under the items 64-72, 1 21-146,
186-238, 272-365.

Before entering upon this examination we present a tabular aspect
of the mutations between Viti and Samoa as derived from the
discussion of the 465 words in the Viti dictionary which are satis-
factorily identified as Polynesian, and to collocate therewith the
Proto-Samoan as in the relation of parent. (See table A, page 38.)

Table b.

E

V

P-S

S

Data.

E

V

P-S

S

Data.

a

0

a

a

e

a

a

a

/320

\34Q

322

324

339

ng
m

ng
ng

ng
ng

ng
ng

(12 cases)
199

i

a

a

a

232

322

324

n

n

ng

ng

125

0

a

a

a

195

357

k

k

k

(28 cases)

u

a

a

a

129

ng

k

k

136 297

299

36i

a

e

a

a

70

188

307

—

ngg

k

'

188

a

u

a

a

144

202

k

ngg

k

'

191 202

224

e

e

e

e

ng

ng

k

'

135

a

e

e

e

134

203

s

s

s

s

(12 cases)

i

e

e

e

[203
1354

210

297

338

t
s

s
dh

s

s

s
s

232

( 65 143
134 1 342

298

337

0

e

e

e

189

344

a

a

e

e

121

126

s

s

—

206 339

e

i

e

e

65

s

dh

h

f 72 215

338

340

i

0

e

e

191

1352 363

i

i

i

i

— dh

h

—

274 278

e

i

i

i

204

319

f

dh

—

—

235

0

i

i

i

196

t

t

t

(34 cases)

a

a

i

i

225

t

nd

t

(208 347

348

349

a

e

i

i

237

I350 353

u

u

i

i

298

305

362

t

dh

t

70 141

e

u

i

i

133

s

s

t

133

0

0

0

0

t

s

t

238

a

0

0

0

137

133

205

VI

m

m

m

(25 cases)

e

0

0

0

349

m

ng

m

m

3i3

i

0

0

0

205

231

33i

353

b

m

m

m

191

u

0

0

0

219

}

V

V

V

296

0

u

0

0

3i4

b

V

V

V

307

u

11

u

u

}

w

V

V

291

a

u

u

u

273

b

w

V

V

281

i

u

u

u

134

284

321

34i

bw

w

V

V

189

0

u

u

u

125

286

273

f

V

(10 cases)

a

0

u

u

129

b

V

(13 cases)

0

0

u

u

138

194

u

V

206 273

363

u

0

u

u

288

—

V

360

I

I

1

I

(27 cases)

b

mb

273

r

I

1

I

(11 cases)

bw

mb

*33

I

r

r

I

[123
134 1

138

3"

3i5

b

mb

p

p

(65 125
\207 285

190

289

192

r

r

r

I

fi32
1336

136
355

225
359

327
364

bw

mb

p

p

fl24 128

1279

139

218

I

ndr

r

I

272

f

mb

p

p

134

r

ndr

r

I

334

335

b

V

p

p

45 284

286

n

n

n

n

bw

V

p

p

361

n

n

n

I

296

—

V

p

p

361

f

V

—

w

295

40

THE POLYNESIAN WANDERINGS.

The mutations properly referable to the parent neutral vowel are
very numerous. The less frequent vowel mutations are here set
down with the authority on which they rest, and in case of material
involved in this collection of data the serial number is appended for
convenience of reference.

i-e siapo-seavu. Note, the mutation e-t

is extremely common.
\-a ofanga-ovi.
\-u 'ili-kuli (294), milo-mulo, isu-uthu

(289), ilo-uloulo (334)-

u-i tulu-tiri (332).

u-o 'uli-koli (11&), lofia-luvu (253), mafu-

mavo, ufi-ovi, jangu-vango.
u-a ngungulu-ngalo , liu-lia (197)-

The less frequent of the constant mutations are these

\-n lama-nama, lamu-namu.
ng-k anga-dhaka.
ng-v lingi-livi (103).
n-ng sina-singasingau.
m-ng lima-linga (313).

k-dh
k-ng

t-k
t-s
t-dh

se'e-dhedhe.

'i-ngi.

tupoto-tumboko.

lata-lasa, uta-usa (238).

tofe-dhove, matala-madhala.

Table C.

E

P
a

Data.

E

P

Data.

a

(73 cases)

J53 95 "7 157 i63 164
\i65 256 268

m

ng

246 264

e

a

mw

ng

167

i

a

63 86 117 175 178 258

k

ng

248

0

a

95 109 161 181

g

ng

248

u

a

86 167 250

/

ng

61

e

e

(12 cases)

k

*

(21 cases)

a

e

120 163 172 249

ng

'

95 169 171 178 2

5o

i

e

87 115 152 173 261

—

258

0
u

e
e

247

87

/

s
s

(20 cases)
90 (cf 96) 169

i

i

(30 cases)

t

s

170

a

i

171 241

h

s

255

e

i

165 174 241 254

t

t

(36 cases)

0

i

241

m

m

(26 cases)

u

i

91 242 251

b

m

76

0

0

(25 cases)

bw

m

172

a

0

96 99 I72 244 255

in

—

102

e

0

88 244

mw

—

107

i

0

87 244 255

bw

V

50 84 109 242

J52 58 59 75 160 163

w

V

"7

u

0

\i84 264

1

V

152

u

u

(42 cases)

b

V

87

a

u

103 109 no 123

}

f

(17 cases)

e

u

179

l

f

J49 51 83 86 147

148

i

u

105

\i74 243 246

0

u

51 83 88 245

bw

f

245

I

1

(26 cases)

kb

f

157

r

1

(22 cases)

b

p

(15 cases)

r

158

bw

p

179 183 241

n

n

(18 cases)

f

p

173 J76 250

/83 92 93 94 J5i *54

m

p

103

ng

ng

\i77 246 264

1

Dealing now with so much of the Efate material as possesses in
the Viti a possible bridge across the gap, we construct table B,
on the preceding page. The mass of the data under consideration,
amounting to 175 entries, is sufficient to establish these results upon

EFATE AND VITI AND POLYNESIA.

41

a satisfactory basis. The Efate material is adjusted to the preceding
table of Viti-Proto-Samoan-Samoan. Corroboration is not listed
in the case of the more frequent concords or mutations.

The following mutations, which have been observed in the com-
parison of Viti and Samoan, are not involved in any of the material
for which we possess Efate data : a-u, nd-h, dh-k, k-t, v-ng, ng-n, mb-v.

The results of this collation are exhibited in the following table
showing to what extent there is an agreement persisting through
Efate- Viti-Samoa, in cases where Viti and Samoa differ to what
extent Efate agrees with Samoa and with Viti respectively, and to
what extent it differs from each. The cases of agreeing vowels have
been omitted for convenience. The spirants and mutes of the labial
series have been omitted because, owing to the absence of one or
other from each language, it has proved impracticable to establish
a comparison; the items involved in this omission amount to 52.

Vowels.

Semi-
vowels.

Pala-
tals.

Lin-
guals.

Labi-
als.

Total.

Throughout

With Samoa

With Viti

8
8

43

41

6

1 1

12

40

3
2
6

46
18

3

8

25
1

152
36
24
69

With neither

We now pass to the consideration of the data involving Efate
and Polynesian which find no bridge in Viti. (Table C, page 40.)
These will be found under the items 48-63,73-120, 147-185, 239-271.
Here again, in the possession of 160 items, we have satisfactorily
sufficient data for a comparative study.

The results of this collation are displayed in the following table.
As before, the spirants and mutes of the labial series have been
omitted because comparison is impracticable; /, however, being
comparable, is included.

Vowels. Sem!'
vowels.

Pala-
tals.

Lin-
guals.

Labi-
als.

Total.

Same

Different

182 26
/i 23

30
12

74
5

43
15

355
126

From this material we diagram the relation of Efate and Samoa
to their common parent.

Efate -Parent-Samoa.

r-r-1

1-1-1

tr-r-1

ng-ng-ng n-n-n m-m-m

bw-v-v

f-f-f

g-g-'

t-d-t

b-b-p

k-k-'

t-t-t

b-p-p

42 THE POLYNESIAN WANDERINGS.

Comparing this with the similar diagram Efate-Parent-Viti (page
37) we note the following differences:

1 . That Samoan ' has to do duty for parent g and k.

2. That Efate" and Samoan t alike are called upon to do duty

for parent d and t.

3. That Samoan p does duty for parent b and p.

4. That Efate" b does duty for parent b and p.

5. That Efate bw does duty for parent v.

In yet broader comparison we observe that all but one of these
differences is found among the mutes in their three series; that
Efate" possesses both sonant and surd of the palatal mutes; that
it accords with Samoan in the loss of the sonant lingual mute, and
that in the labial mutes it has lost, or has not yet acquired, the
surd, while Samoan has lost the sonant.

When we add to this latter item the consideration that Ef ate" has
difficulty in compassing the sonant spirant of this same series and
has to render it by the muted semivowel bw, we are led to the
conclusion that while the Efate" folk have a richer endowment of
palatals their lips are not sufficiently under nice control to do justice
to the wealth of the Proto-Samoan labials, a determination which
naturally arises from inspection of what we may deem a mouthing
process in pronunciation. We are, therefore, not at all surprised
that in the instances where Efate is comparable with Viti and
Samoan, and they are not in concord, this ruder speech of the west-
ern seas agrees with the rougher Viti in 40 per cent of such cases.

We shall next pass to the consideration of the width of these
Efate identifications with Polynesian, and we shall reckon the tale
of those that are found only in Nuclear Polynesia, the proximate
islands ; and those which have reached to the most distant verge of
Sawaiori migration. We shall be narrowly particular in specializ-
ing geographically the identifications in Nuclear Polynesia, except
for a certain important subdivision more general in regard of the
classification of identifications in the eastward Polynesia of the
later swarming. In each case we treat Nukuoro* as practically
Samoa, and the few instances drawn from Polynesian inclusions in
Melanesia or islands of the Polynesian verge t we class with Nuclear
Polynesia. Furthermore we segregate the data as involving or
omitting the Viti, a major subdivision in scope, yet one which more
complete scrutiny of distal Polynesia may quite as probably show
to be devoid of existence; it is accordingly maintained as no more
than provisional.

♦17 Journal of the Polynesian Society, 152. ^27 Amer. Journal of Philology, 370.

EFATE AND VITI AND POLYNESIA.

43

Efali- Viti-Nuclea r Polynesia .

Samoa 65 191 357

Tonga 121

Nukuoro 348

Moiki 356

Samoa -Tonga 70

Samoa-Futuna 186

Samoa-Niue 127

Futuna-Nukuoro 45

Samoa-Tonga-Niue . . .195

Samoa-Tonga-Futuna . {^p9^

Samoa- Tonga-Niue- \

Uvea/ " " * 1

f66 135 188

337

Samoa-Tonga-Niue- \

.■{225 341 (and Nu-

FutUnaqkuorof
Samoa-Tonga-Futuna) _ , . N

Uvea/217 <etc-> 22°
Samoa-Tonga-Niue- 1

Futuna-Uvea/-136 I45 3"
Samoa-Tonga-Niue-\ ("132 198 200
Futuna-Uvea-Nukuoro/ \ (etc.) 209

Samoa

Samoa-Tonga-Niue

Efatt-Nuclear Polynesia.

'54 60 77 79 84
102 105 109 115
117 119 247 251
246 (and Nukuoro)

Samoa-Tonga 56 85 156 169

Samoa-Futuna 50 82

/i57 93 (and Nu-
\kuoro)
Samoa-Tonga-Futuna[gor254 (and Nu-

Samoa-Niue-Futuna . . 1 1 1
Samoa-Niue-Uvea .... 53
Tonga-Futuna-Uvea ..112 174
Samoa-Tonga-Niue \ „

Futuna/--57 95 181

Samoa-Tonga-Niue^j 2?o (and Nukuoro)

Samoa-Tonga-Futuna) ,

Uvea/ b2
Tonga-Niue-Futuna-\ ,

Uvea/02
Tonga-Niue-Futuna- . )

Uvea/51
Samoa-Tonga-Niue-)

Futuna-Uvea/ ' '^"

255
91 178

Combining these two tables, we have noted 69 cases of identi-
fications of words common to Efate and Nuclear Polynesia and
which extend no farther along the track of the migration eastward.
As I have been at pains to indicate* that there is a possibility of
segregating the residua of the former, or Proto-Samoan, migration
and the latter, or Tongafiti, migration in Nuclear Polynesia, it is
of the highest interest to note that all but seven of these identi-
fications involve the Samoan, and that no less than eighteen reach
Samoa and go no farther. ; i

We shall next take up the tale of Efate and all Polynesia, distal
as well as nuclear, with, however, certain very important data
reserved for yet more particular note.

Efati- Viti- Polynesia.

64

67

68

72

123

125

126

133

134

138

139

141

142

143

146

187

190

192

193

196

201

203

204

206

207

208

210

211

212

213

214

215

216

218

222

223

226

227

228

229

230

231

233

234

235

238

272

273

274

275

276

277

278

279

281

282

283

284

285

286

288

290

291

292

293

294

295

296

297

298

300

301

302

304

305

306

307

308

310

312

313

3i4

3i5

317

3i8

3i9

320

321

322

323

324

325

326

327

330

33i

332

335

336

33«

339

340

342

343

344

345

346

347

349

350

35i

352

353

354

355

358

359

360

36i

362

3&3

364

EjaU-Poly

nesia .

48

58

61

63

73

74

75

76

78

80

81

87

89

90

92

94

IOI

106

107

118

120

148

149

150

151

152

153

154

158

159

161

162

163

164

166

167

168

171

172

175

177

179

183

184

185

240

241

243

244

245

248

249

253

256

257

258

260

261

263

264

266

267

268

269

271

♦17 Journal of the Polynesian Society, 214.

44 THE POLYNESIAN WANDERINGS.

From the 1 88 items of the identification of K ate material with
Polynesia, including the Nuclear Polynesian and the most remote
terminus of migration, we have withheld for particular note the
following interesting items, which, with others of similar import yet
to be developed, call for special comment.

These items are identifications made in Nuclear Polynesia and
thence traced, not to distal Polynesia in general, but to particular
termini with at least one intervening or intermediary point. We
group them as they fall into three classes as southern, eastern and
northern swarms.

c ,, (Maori-Mangareva 72 120 272(nefu)

*outhern {Maori-Tahiti 137 *55 309 328 180 250

Swarm. [Maori-Tahiti-Mangareva. . 106c 239 74 116 165 147 194 197 219

(Other intermediate points in several items are Mangaia, Rarotonga, Tongarewa, the
Paumotu.)

Eastern fTahiti-Mangareva 106 232

Swarm: \Paumotu-Mangareva 242

». ., ( Hawaii-Mangaia 332c

Northern , Ha waii.Tahiti o8 , 4o 1 60 252

Swarm. [Hawaii-Tahiti-Paumotu. .316

We have next to consider a small but well-marked group of
identifications where there is a leap from Efate to distal Polynesia
without having left a trace in Nuclear Polynesia except in so far
as that may be held to exist in the instances linked to Viti, these
being distinguished by V in the list.

General 99 176 V237

Southern swarm 96 97 V1366 V124

Eastern swarm 1060

Northern swarm 59 88 104 1 10

Finally we come to a group where identification has been estab-
lished in Nuclear Polynesia and in one, and only one, of the per-
ipheral languages. In the same classification and with the same
notation they are listed in the following table :

Swarm:

Southern .. 1066 V122 V130 V144 182 V199 V236 262 V287 V329 V3436

Eastern ... 100 108 113 ¥1390170 173 V299 V334

Northern.. 55 V 69 83 86 114 V128 V189 V224 V343<x

In arithmetical summation we find that 324 identifications in
this material may be considered quite satisfactorily established as
between Efate and Polynesia. Of these there are 69 words in which
the Efate word is found in Nuclear Polynesia alone, 191 in which
it is general to Polynesia, 1 9 in which it is associated to the northern
swarm which has found its ultimate seat in Hawaii, 12 to the
eastern swarm, 33 to the southern swarm and a Maori home.

Now what are the deductions which may justly be drawn from
these data?

EFATE AND VITI AND POLYNESIA. 45

A.

We know from our study of the Samoan record of historical
tradition and equally from our comparison of the elements and
methods of that language with its congeners in eastward Polynesia
that this group of islands, so centrally situated in its relation to the
outward islands as is the palm to its outstretching fingers, has been
the scene of at least two settlements of Sawaiori stock. To these
two periods, culminating in the victory of the earlier colonists in
the Matamatame epoch,* I have assigned the designations of Proto-
Samoan and Tongafiti, the one carrying to us its own explanation,
the other equally clear to the Samoan as his own name of the enemy
who drove him into the mountains, harried him with exactions, and
was at last expelled by the semidivine might of his great national
hero, Savea, in whom the Malietoa name began.

Our initial problem, therefore, is to deal with the incidence of the
now established Polynesian content of Efate. Is it Proto-Samoan,
is it Tongafiti, is it a mixture of both? That is to say, in another
form of statement, was it the former swarm out at the gates of the
Malay seas, or was it the latter, or was it both which bivouacked
in Efate of the New Hebrides on its way to Samoa? Or, this is
a possibility as well, did wind-driven estrays from Samoa obtain a
foothold in Efate in sufficient numbers to fix and clinch these several
items upon the Melanesian speech there existing? Was it during
the earlier Proto-Samoan period, during the succeeding Tongafiti
period, or during the modern Samoan era after Matamatame, when
the language is a mixture of these two in proportions by us as yet
undetermined, that such reverse migrations took place?

B.

The last hypothesis, the counter migration westward from Samoa
considered as an eddy current of the great eastward sweep of the
race, we shall take up first. The drift of such castaways has been
known within the historic period. I cite Codrington'sf excellent
authority upon this point :

From the limits of the Melanesian languages as defined above, the lan-
guage of the Polynesian settlements in Melanesia has to be withdrawn.
The distinction between this and the Melanesian is everywhere plain, and
there is very little distinction apparently to be made of dialect in the
speech of one settlement and another. These Polynesian outliers are to
be found in Uea, one of the Loyalty Islands; in Fate, Sandwich Island;
in some islets of the Sheppard Group, and notably in the settlement of
Mae in Three Hills ; in Tikopia, north of the Banks Islands, and in several
of the Swallow Group near Santa Cruz ; in Rennel and Bellona, south of
the Solomon Islands, and in Ongtong Java, near Ysabel. The language
of these is said, on good authority, to be substantially that of Tonga, and

*27 American Journal of Philology, 371; 30 American Journal of Philology, 171. '
f Melanesian Languages, page 7 et seq.

46 THE POLYNESIAN WANDERINGS.

the same throughout; speakers of the Maori of New Zealand can under-
stand it and make themselves understood; it has nothing to do directly
with the Melanesian languages.* The existence of these Polynesian settle-
ments, however, in the midst of Melanesia can not fail to suggest questions
of interest and importance which it is impossible to dismiss without con-
sideration. As to their origin, it is not difficult to conjecture what it has
been. Canoes accidentally drifting or blown away, or expeditions pur-
posely directed to known islands, have landed small parties of Polynesian
people either on uninhabited places or on islands occupied by Melanesians.
Some at least of such settlements may be supposed comparatively modern.
If such islands as Rennel, Bellona, or Tikopia have been reached, remote
from any large Melanesian island, the colonists naturally remain purely
Polynesian in language, habits, and physical characteristics, for there is
no admixture. If a single canoe, or a small male party, has found its
way to an inhabited Melanesian island, the Polynesian element has been
absorbed, leaving perhaps only some fairer and more straight-haired
children as an evidence of mixed blood. f In the case of such a settle-
ment as Mae the case is different. The middle part of that island, one
only about six miles long, is occupied by people whose speech is that
common to all these Polynesian settlers, but who physically are not dis-
tinguishable from their neighbors who are Melanesian both in language and
physical character. The same is the case in the Swallow Islands: the
inhabitants of islands close together speak either a language like that of
Santa Cruz or the Polynesian ; but they are all alike Melanesian in appear-
ance. The Tikopians, an isolated Polynesian settlement, are wholly unlike
Melanesians — tall, heavy, light-colored men, with straight hair. The
reason why the Polynesian-speaking people of Mae, for example, are Mel-
anesian in appearance clearly is that the Melanesian blood in them has
overborne the Polynesian element; that is to say, the Polynesian settlers
have, generation after generation, taken Melanesian wives into their villages
in which the speech was Polynesian. The speech, the descent of chiefs,
certain religious practices, have remained Polynesian, the physical aspect
has gradually lost its original character. Under such circumstances the
speech which will be permanent is the speech of the settlement ; the phys-
ical character that will prevail will be that of the blood. Hence the
Tikopian is physically and in language purely Polynesian, the Fileni man
of the Swallow Group is in speech Polynesian but physically Melanesian.
The phenomena of the case are thus explained. +

♦Some few years ago a whaler picked up in the Solomon Islands and brought down to
Norfolk Island some natives of Mae and Fate, survivors of a crew massacred in Ongtong
Java. They belonged to the Polynesian settlements, and they told me that they, the
Mae and Fate men, spoke the same language, and also understood that of the Ongtong
Java people. — Dr. Codrington's note.

fl have seen myself in Ureparapara a man and woman with a son, drifted thither from
some Polynesian island; and I have noticed straight-haired children in Saddle Island
who were known to be descendants of Polynesian castaways. — Dr. Codrington's note.

JSome fifty years ago the Banks Islands were visited in two successive years by double
canoes. The people in those canoes said they came from Tonga. They settled the first
year for a time on the islet of Qakea, close to Vanua Lava, quarrelled after a time with
their neighbors, and went off. When they returned next year they were attacked by
the natives and driven off. There were women with them. If they had settled on
Qakea there would be there now a Polynesian-speaking people, but Melanesian wives
from Vanua Lava would be continually bringing in Melanesian physical characteristics.
If Qakea had been an isolated place like Tikopia, there would have been then a small
purely Polynesian colony. — Dr. Codrington's note.

EFATE AND VITI AND POLYNESIA. 47

It remains to state another remarkable fact. In Three Hills Island,
Mae, the Polynesian settlement above mentioned is about two miles
distant from Sesake, at one end of the island, occupied by those who
may be called the aborigines. The Mae language is Polynesian, if not
purely, at least decidedly so ; the Sesake language is Melanesian decidedly,
and at any rate has nothing that makes it appear more influenced by its
Polynesian neighbor than if Sesake and Mae were in different and distant
islands. This can not be too positively stated, and the importance of the
fact is very great. It is an exemplification, in a very narrow field, of what
is also found to be the case with regard to Fiji. The Fijian group is only
some 200 miles west of the Friendly Islands, which are decidedly part
of Polynesia. There has been a considerable intercourse between the two
groups, and no doubt a great infusion of Tongan, Friendly Islands, blood
among the higher classes of Fijians. There has been also, according to
native legends, a considerable intercourse between Fiji and the purely
Polynesian Samoa. Yet the Fiji language is most decidedly Melanesian ;
it has no doubt something directly derived from Tonga, but it is no more
Polynesian than the languages of the Banks Islands, which lie far away
to the west, out of reach of any but the most casual and insignificant
intercourse with Tongans or other Polynesians. Intercourse, therefore, and
close neighborhood with Polynesians do not, as a matter of fact, materially
affect the language of Melanesians.

Yet Efate has had its Polynesian content sufficiently long to
subject it to a course of modification,* and this is not the work of
a day or of a generation. We are, therefore, to lay aside the pos-
sibility that this accretion is due to westward drift in the modern
epoch; furthermore we are yet without information on the extent
to which modern Samoan is the result of the admixture of the
speech of the earlier and of the later migrants.

It is equally impossible that the Polynesian accretion came to
Efate during the Tongafiti domination of the littoral of western
Samoa. We have segregated in this Polynesian content a consid-
erable proportion of words which, qua Polynesian, are the exclusive
possession of Nuclear Polynesia. It is absurd to hold it possible
that Tongafiti migrants should have acquired these words from the
Proto-Samoans with sufficient grip to carry them on a voyage against
the current of their race and to impress them on an alien and
resistant people, yet with entirely too feeble a hold to carry them
along the current of their further and easy migration eastward.

The data here collected must stand as proof positive that if the
Polynesian content of Efate is due to a westward drift from Nuclear
Polynesia that acquisition must have been in the Proto-Samoan
period, a period which we must consider to have ended with the
Tongafiti swarming, the epoch of the great voyages, and that was
somewhere about 1,200 of our era. Against this possibility of west-

*This is shown in examination of the data passim in this collection, and particularly
as regards a specific detail of such modification in my paper on Duplication by Dissim-
ilation, 30 American Journal of Philology, 171.

48 THE POLYNESIAN WANDERINGS.

ward drift our comparison of the Polynesian of Efate with that of
Nuclear Polynesia gives us warrant to say that it is highly improb-
able that Efate" received its Polynesian from the east; it gives us
no warrant to say that it is impossible.

C.

It may be difficult and it may remain difficult to prove that the
Polynesian in Melanesia did not come there from Sawaiori settle-
ments in the central and eastern Pacific ; the proof of a negative by
no means loses any of its logical difficulties when stated in terms
of an unlettered people. But there is all the proof that consistent
tradition can give to the migration theory which introduces the
Melanesian chain as the line along which the migrant fleets passed
from the known sojourn in Indonesia to the known occupation of
Polynesia. The presence of so large a share of Sawaiori vocables
in Efate" does not in itself prove that this island lay on the fairway
from Indonesia to Savai'i, but it is hugely confirmatory of all other
evidence upon which this theory is based. We are here really not
so much concerned with the proof of this migration theory as we
are with the determination of which swarm it was that included
Efate in its voyaging.

In the Proto-Samoan we have a distinct, and surely an early,
phase of Sawaiori speech which differs from the Tongafiti in the
possession of certain vocables which the latter has not — in the
absence of certain vocables which the latter has preserved or has
acquired since the separation of the two stems. Above all else the
Proto-Samoan is distinguished from the Tongafiti by its maintain-
ing superficially, or recoverable under the surface by the merest
scratching, the remnants of the original speech in which closed
syllables were possible.

Everywhere in this Polynesian of Efate we find the closed syllables
of the Proto-Samoan, not the Tongafiti softening away of the final
consonant. This is so common that it has not seemed necessary
to collate for this feature, it has been sufficient to insert the Proto-
Samoan radical in each item of the collected data where it has been
possible to recover it. On this score we have no hesitation in asso-
ciating the Polynesian content of Efate with the Proto-Samoan.

In consideration of the vocabulary we are dealing with 324 items.
Of these there are but 1 2 which are not identified in Nuclear Poly-
nesia; if we include Viti in Nuclear Polynesia, as seems quite
justifiable, we shall reduce the reckoning to 9. Now 9 in 324 is
readily explicable by the known habit of Polynesians to cast out
certain words under the manifold working of their tabu.* Against
these 9 is to be set the fact that of our 324 identifications 20 per

*See note 255.

EFATE AND VITI AND POLYNESIA. 49

cent, i in every 5, 69 words in all, are common to Efate and Nuclear
Polynesia and are not to be found in the languages deriving from
the Tongafiti or cadet branch of the family ; that in the close sub-
division of the Nuclear Polynesian 18 words of these 69 are to be
found in Samoa alone, and that only 7 in all fail of Samoan identi-
fication, and even of these 7 there are 2 which might readily be
identified in Samoan. The conclusion is irresistible that the Poly-
nesian content of Efate is Proto-Samoan.

Assuming the validity of the migration down along the Melanesian
chain from Indonesia to the Pacific we have no hesitation in declar-
ing as proved that the Proto-Samoan swarm had to do with Efate*
and that the Tongafiti swarm did not touch there. I have used
discretely the term "had to do" in reference to this transaction,
for we are without data upon which to base a more definite determi-
nation. The main body of the Proto-Samoans may have rested
upon Efate through such a lapse of time as to impress their better
speech upon the ruder autochthons, or a considerable unit of the
swarm may have deviated from the voyage and have colonized the
island. In either case the physical results of such commingling are
not apparent.

D.

We now come to the last items in the tabular grouping of the
data, the eastward extension of this material which we now know
to be Proto-Samoan and its discovery in territory now occupied by
Tongafiti descendants. We shall here touch upon it very lightly,
for it does not bear upon the proof as relating to Efate. We are
not to assume that the swarming impulse which set the Proto-
Samoans in motion from ancestral Hawaiki (whether that be in the
Hindu-Kushor in the Hadramaut) , which drove them into and out
of Indonesia, which pushed them through Efate* and into Nuclear
Polynesia — we have no reason to assume that this quest of the
sunrise deserted them when they reached the green hills and the
gray sands of Samoa. We read their stories but blindly if we miss
the departure of hardy voyagers in search of sea and sea and haply
land beyond. Until the Tongafiti came and drove them from the
coasts the sea was theirs for voyaging. In these memoranda we
think we find record of the ports some of them made.

There is confirmation peripheral as well. Hawaii has record* of
voyages from Samoa direct, and in New Zealand f there is a similar
tale. The Samoans say that their ancestors started on voyages
from Samoa, the Hawaiians and the Maori record that voyagers from
Samoa arrived. Here in the philological record we seem to have the

*2 Fornander "Polynesian Race," 33. fS. Percy Smith in litt.

50 THE POLYNESIAN WANDERINGS.

proof as convincing as in our civilized estate would be the production
of clearance and entry duly attested by tidewaiter and collector of
the port.

We are now brought to a point at which it will be advantageous
to offer a check-list of the mutations in the Polynesian material
contained in the fundamental data of this work. The standard
upon which the comparison is based is the Proto-Samoan, the earlier
phase which with as much certainty as ease we may recover from
the current Samoan. The figures refer to the data items as pre-
sented in the appendix , pp . 1 8 5-43 1 . The figures included within par-
entheses indicate that in the vocables thus designated the mutation
appears to run concurrently through two or more languages. The
point is important and worthy of some note, for it is quite possible
that the word stem may of its own constitution carry the mutation
into a speech whereto the mutation itself is not normal. To find
an excellent instance of this we need look no further than the first
subdivision hereunder, the a-e mutation in fanua (292) land, which
we find to persist through no less than ten of the eighteen languages
involved, yet which is the only instance of this mutation in Maori
which our material affords. To the identifying numerals the dia-
critical mark of the acute accent has been added when the mutant
vowel bears the accent in Samoan or in the speech under examina-
tion. This distinction will be found of value in all the languages
which retain the penult accent (except where otherwise indicated
in the text) normal to Polynesian speech. The Maori, it will be
recalled by those familiar with these tongues, has broken away from
the Polynesian rule and has devised an accent system of its own.
Similarly the degree mark is used to denote the cases in which the
mutation occurs in an unaccented final syllable.

a-e (170, 216, 253, 290, 292, 350)

Bukabuka. .216, 292 Moriori 73

Fakaafo .... 292 Niue 147, 397', 350

Futuna 170' Nukuoro . . . 290

Hawaii 166, 253' Rarotonga. .216

Manahiki . . . 292 Rotuma. . . . 3600

Mangareva. . 154, 253', 292, 350 Sikayana. . . 292

Maori 292 Tahiti 1830, 216, 292

Marquesas . .80, 89, 193, 258, 274, Tonga 170', 290, 350

275a, 290, 292, 322 Uvea 148', 292

Moiki 292 Viti 164'0, 188'

a-t

Moriori 350 Tahiti 2 16

Niue 290 Viti i39&°

a-o (216, 271, 292)

Hawaii 193, 292 Tonga 92, 107, 167, 216,271°,

Marquesas .75, 163, 2 16 277, 292

Moriori 214'° Uvea 271°

Niue 93', 2 16, 292

Rotuma. . . 239', 290, 291', 294', 328'
351,352'

EFATE AND VITI AND POLYNESIA.

51

a-w (144, 167, 185)

Fotuna 144', 281'

Futuna 167

Hawaii 167

Mangareva. . 167

Maori 167, 1850

a-au

Fotuna 45°

e-a (47)

Bukabuka. .327

Futuna 47'

Hawaii 210'

e-i

Aniwa 2490

Fotuna 315°, 2470, 363=

e-o (126)

Hawaii 126

Moriori 126

Rotuma. . . . 122', 272'
e-t*

Rarotonga. 257'
\-a

Fotuna 2860

Marquesas. . 1850

Niue 167, 248

Tahiti 167, 185°

Uvea 167

Viti 144, 202'

Niue 1270

Rotuma. . . .290°

Viti 47', 126, 272'

Rotuma. . . . 153'
Viti 640, 173

Mangareva. . 126
Marquesas. .126

Maori 332'/i

Nukuoro. . . .304'

Rotuma. . . .352°

Aniwa 300'

Fotuna 298', 300', 321', 362'

(This occurs in but these two islands of the verge and appears to be only initial.)

1-»I(3<>5)

Nukuoro. . . . 166'

Sikayana. . .321'

Tonga 53'

o-a

Moriori 287

Rarotonga. .257°
o-e

Rarotonga. .205'
o-i

Viti 2480

o-w (92,314)

Futuna 92

Hawaii 314'

Mangareva. .314'

Maori 314'

Niue 92

o-ui

Hawaii 1396'

Fotuna 76', 343° c

Hawaii 277'

Uvea 286'

Rotuma. . . .305'
Viti 305'

Tonga 52'

Tongarewa. .147°

Paumotu. . .314
Sikayana. . .192'

Tonga 92

Uvea 92

Viti.

I55"

Mangaia. . . . 1680
Niue 271

u-o (129)

Fotuna 129', 3280

Maori 361

Nukuoro. . . . 1390

u-au

Fotuna 151

u-ia

Tahiti 347

Tahiti 139, 216', 285'

Sikayana. . .261'

Viti 129', 1380, 194, 288'

52 THE POLYNESIAN

u-ow

Aniwa 151

ae-e

Niue 261', 315', 347'

ai-ei

Maori 268

au-a

Mangareva. .61'
au-ou (61,200,335)

Aniwa 200', 267', 301'

Mangareva. . 161', 162', 226', 335'

Maori 283'

Marquesas . . 335'
ei-i

Tonga 78'

ou-au

Fotuna 282'

ng-n (note 125)

Hawaii

Marquesas (western dialect)

ng-fc

Marquesas (eastern dialect)

ng-ngg

Viti 248'

ng-v

Viti 1540

ng—

Maori 310°

Marquesas. .151'
Niue 93'

kg

Paumotu . . . 306'

k-ng

Mangareva. .67, 214
Uvea 178

k-ngg

Viti 188, 191, 202, 224

k-t

Niue 250

k—

Hawaii

Marquesas . . 204, 261

Niue 57

Nukuoro. . . .300
Paumotu . . . 248

l-«

Hawaii 154

Nukuoro 100, 200, 297, 309, 310, 350

Tahiti 100, 154

1-ng

Moiki

1-/

Rotuma. . . .354

l-t

Mangareva. .257

l-nd

Viti 287

l-ndr

Viti 272, 307, 314. 334, 335

WANDERINGS.

Niue 61', 200'

Nukuoro. . .200'

Tonga 61', 200', 335'

Uvea 61'

Rotuma 2850

Sikayana

Viti 332'b

Tahiti

Uvea 690

Viti 135

Rotuma. . . .300, 305
Tahiti

Tonga 181

Samoa

Tonga 3".3i3

Uvea 312,313

EFATE AND VITI

I—

Hawaii 257

Marquesas

Niue 200, 230, 244, 301, 322,

327,336
n-l

Nukuoro 296

n-r

Rotuma. . . .328
n-ng

Moriori
h-h

Niue 47, 81, 206, 278

Tonga
h-s

Rotuma .... 206, 278, 340
h-th

Viti 47, 72, 215, 274, 278, 352

h-w

Hawaii 295

«-/

Futuna 239

&-h

Hawaii

Mangareva

Maori

Marquesas

Niue 91, 98

8-(f, h)

Manahiki

Nukuoro
s-sh

Moiki
s-ndr

Viti 232

s-th

Viti 65, 91, 143, 198, 233, 298,

337.338,340,341.344
s —

Hawaii 245

Mangaia. . . .204, 205

Mangareva

Maori 239

t-j (ch, tsh)

Aniwa 166, 329

Fila 166,329

t-f

Rotuma 294,324,350

t-k

Hawaii

Mangareva. .1416
Paumotu. . .216, 237

t-.r

Futuna 3326

Rotuma. . . .352

t-nd

Viti 141a, 208, 329, 350, 353

t-th

Rotuma ... 294

t—

Marquesas . . 350

AND POLYNESIA.

Tahiti 281

Nukuoro . . . 349

Tonga 244,327,356

Uvea 301

Samoa 296

Tahiti 328

Viti 342

Uvea 47.295.33i

Viti 206,331,339

53

Paumotu
Tahiti
Tonga
Tongarewa

Tonga
Uvea

Marquesas. .239

Rarotonga

Tahiti 90, 140, 239

Fotuna 133, 165, 219, 329, 355

Tonga

Tonga 355

Liueniua. . . . 350

Uvea 165, 196, 219, 325

Viti 238

Viti 1416

54

THE POLYNESIAN WANDERINGS.

m-ng

•3i3

\-h

Nukuoro. . .

281
.296

v-w

Hawaii
Maori

i-h

Hawaii
Manahiki . .
Mangareva
Maori

•259,329.36o
.201, 259, 290, 292

t-p

.86

f-v

Bukabuka .
Fotuna. .. .

.290

. 166, 329

i-w

Manahiki. . .290 J
Mangareva. .290, 293
Marquesas . . 290

Hawaii 290

i-hw

Bukabuka. .292

Maori 122, 130, 133, 147, 148, 166,

170,213,214,223, 243,273,
283,292,293,294

f-6

Tonga 86

f-mb

Viti.

273a

f-tch

Sikayana ... 259

Bukabuka
Mangaia

Nukuoro 290

Mangareva. .92, 147, 148, 213, 214, 272,
290, 292, 294

p-b

Rotuma. . . .285
Fotuna 192, 286

p-mb

p-h

p-v

Viti 64, 125, 128, 173, 190, 192

Rotuma. . . .284

Viti 103, 284, 286

Viti 313

Rotuma. . . . 122

Viti

Fotuna. . . .281

Marquesas. . 130, '133/166, 292, 293, 296

Moiki 201, .206, ,2 1 5, 2736, 329

Moriori 2 14,^272

Niue 137

Tongarewa

Uvea 86

Paumotu . . . 290

Tahiti 290

Uvea 314

Viti

Maori 290

Moriori 287, 293

Fila.

•329

Rarotonga

Rotuma. . . .206, 214, 290

Tubuai 290

Sikayana. . .285
Tonga 52, i2{

CHAPTER VII.

POLYNESIAN RELICS IN MELANESIA.

Check-list of the material for this section of the work — Tables of the
phonetic relations of 81 languages of the Melanesian archipelagoes —
The several mutations of vowels and consonants and the languages
which employ them — Analysis of these mutations, those which are
found in Polynesian and those confined to Melanesia — The ground-
work of Polynesian mutation and the Melanesian system compared
therewith — Two Melanesian foci of Polynesian influence brought to
light — The sieve theory disproved by this material — Proof of the
general migration theory — Crop colonies and the important part they
play — Two tracks of Proto-Samoan migration through Melanesia —
Proof that a Melanesian sojourn preceded the settlement of Samoa.

In an earlier chapter I have written that in accordance with Dr.
Macdonald's theory of his Oceanic language equal weight should
attach to the identification of Efat£- Polynesia and Efat£-Melanesia.
The former we have just had under rigorous review and we have
been led to certain conclusions. We are now to examine the
Melanesian material in its turn.

This is distributed irregularly through the collected data, the
distribution being based upon the tabulation of the Efate" words in
several series. To facilitate reference to the material of each of
these languages the following catalogue will be found of service.

Alite 217 239 250 251 259 265 275 278 285 291 300 301 306 309

313 317 330 337 350 351 352 356 360 361

Alo Teqel 122 123 147 190 193 214 217 239 250 251 272 275 278

294 300 301 306 307 309 316 317 318 324 328 333 335 337

343 350 360 361

Ambrym 46 147 194 212 215 239 242 252 258 265 266 274 278 284

287 290 291 297 298 300 305 312 318 324 329 336 350 352

356 357 360 361

Aneityum 124 126 147 149 154 168 173 188 191 194 198 204 205

206 207 214 216 217 228 229 240 241 256 258 260 262 264

265 266 268 269 270 271 273 274 278 279 282 283 285 288

291 293 294 298 301 306 307 309 310 312 313 315 316 317

318 319 320 321 322 323 324 326 328 329 335 337 339 340

342 344 347 349 35Q 360 361

Arag. ...122 149 152 153 160 190 193 209 212 214 215 217 239 250

258 259 265 272 275 278 284 285 290 291 292 294 300 301

306 307 309 312 313 317 318 324 328 329 330 333 335 336

352 356 357 360 361

Baki....i47 149 156 182 190 194 199 208 210 211 217 228 237 240

247 254 258 263 273 278 282 285 290 291 292 297 298 305

308 309 312 313 315 317 320 321 324 336 340 344 346 347

350 352 356 358 359 361 362 364

Baravon. 74 122 147 149 156 190 207 214 215 217 257 270 273 280

284 285 290 291 294 300 313 319 320 323 324 327 329 336

346 358 360 361

Bauro ... 47

56 THE POLYNESIAN WANDERINGS.

Belaga.. 46 147 152 203 207 211 258 284 285 290 300 317 338 339

342 343 346 352

Bierian..i48 156 199 207 210 216 217 240 252 256 258 259 260 263

270 278 282 285 290 291 292 293 294 297 298 300 301 302

305 308 309 311 312 313 3i5 3i7 3i8 320 321 322 323 324

329 336 340 344 35o 352 356 358 361 364

Bougainville 149 192 259 265 278 350

Brierly Island 190 294 317 324 340 347 361

Brumer Island 298 324 344

Bugotu..i22 147 213 214 217 239 251 257 272 275 278 284 285 290

291 298 301 305 306 309 312 313 316 317 322 328 329 330

338 339 343 346 35i 352 361 ^000

Buka.... 46 149 158 190 207 217 239 247 257 259 265 278 280 285

298 300 306 308 313 318 324 328 330 343 344 35o 352 355

360 361 363

Bululahai90 217 239 250 251 259 265 291 298 300 301 306 309 317

329 330 35o 35i 352 356 360 361

Deni 194 212 214 265 266 278 285 306 307 324 335 339 358 360

t 361

Dufaure Island 167 340

Duke of York 46 74 122 147 149 164 169 207 214 217 265 269 272

275 285 290 292 294 300 301 309 312 313 314 316 318 319

320 321 324 325 327 328 329 332 335 336 343 350 3526 358

360 361

Epi 39 44 47 75 77 186 200 202 212 265 286 330 343 363

Eromanga 194 206 217 218 240 265 270 273 274 278 282 285 290

291 294 305 312 317 324 329 333 337 338 349 350 352 360

361

Fagani..i22 147 149 190 214 217 239 250 251 257 258 259 265 272

273 278 285 290 292 294 298 300 307 309 312 313 3J6 3*7

318 324 328 329 330 342 350 351 352 356 357 358 360 361

Gog 46 122 123 147 149 *5i *54 *9<> 203 206 214 217 239 251

268 272 275 278 284 285 298 300 301 306 307 309 312 313

317 318 322 324 328 333 335 336 338 339 342 343 356 357

358 360 361

Guadalcanar 149 272 278 285 298 309 328 335

Iai 194 259 265 294 318 324

Kabadi.. 46 74 265 266 292 330 361

Kabakada 122 147 214 285 320 327

Kalil....294 296 312 322 329 352

King 47 74 147 151 196 213 217 247 258 264 266 280 294 296

309 312 313 318 319 320 321 323 324 329 336 342 343 344

349 35o 352 358 361 363

Lakon...i22 123 147 149 190 193 206 214 217 239 251 265 272 275

278 284 285 292 301 306 309 312 313 317 318 324 328 333

335 337 343 350 35 1 356 360 361

I,amassa. 47 74 151 190 213 216 247 265 266 280 285 292 294 296

309 312 313 316 317 3*S 319 324 329 342 343 344 35Q 358

361 363

Lambell. 47 74 147 *49 J5i 190 196 213 216 247 264 265 266 280

292 294 296 309 312 316 317 318 319 320 324 329 336 342

343 344 346 35o 352 358 361

Laur ... 47 151 156 190 216 217 247 257 265 266 277 292 294 295

296 309 312 313 315 3i6 317 3i8 319 324 329 333 343 344

350 352 361 363
Lemaroro 258 274

Leon 4G 122 147 149 187 193 203 214 265 292 307 318 337

Lifu 214 259 265 318 324 326 330 333 36i

Lo 46 122 147 149 152 169 190 193 195 212 214 217 239 251

265 266 272 275 278 285 290 291 300 301 305 306 307 309

312 313 316 317 3i8 321 324 328 333 335 336 339 343 35<>

35i 356 357 358 36o 361

Maewo. . 46 122 147 H9 152 153 19° *93 203 206 212 214 215 239

259 264 266 272 275 278 284 285 291 292 298 301 306 309

312 313 316 317 3i8 324 328 329 333 335 336 337 346 352

357 358 360 361

Mai 281

POLYNESIAN RELICS IN MELANESIA. 57

Makura.. 74 165 265 273 290 297 302 306 309 312 317 321 333 335
352

Malanta.312 324

Malekula.43 47 74 77 121 147 148 156 157 161 193 194 199 208

210 211 217 227 228 239 242 246 247 251 252 254 258 259

270 274 276 278 282 285 287 289 291 292 294 297 300 301

302 306 309 312 315 316 317 318 319 320 321 323 324 328

329 335 339 340 344 345 346 347 350 351 352 356 358 359

360 362 363 364

Malo. ... 44 47 121 124 147 148 190 193 194 199 203 206 207 208

217 218 228 239 247 252 254 258 266 270 282 285 289 290

292 293 297 301 302 305 306 308 309 311 312 313 316 318

319 320 321 324 328 333 335 336 337 338 339 340 343 344

346 347 35i 352 353 354 356 358 359 361 362

Marina.. 39 46 147 193 212 214 250 272 275 278 284 285 291 292

301 306 309 311 312 313 317 318 321 324 328 333 335 336

343 350 352 356 357 358 360 361

Matupit. 74 147 214 244 280 285 287 300 320 346 347 352 358

Meli 147

Merlav.. 46 122 123 147 149 151 154 190 193 197 206 207 210 214

217 239 251 265 268 275 278 284 289 291 300 301 306 307

309 312 313 316 317 318 322 324 328 333 335 337 338 339

342 357 358 360 361

Modnus. . 74 126 190 215 239 265 284 285 290 291 294 305 308 312

317 324 328 329 33o 335 345 35o 352 360 361

Mosin...i22 123 147 154 193 214 217 239 251 265 272 275 278 284

300 301 306 307 309 312 316 317 318 324 328 333 335 336

339 343 356 357 360 361

Mota 46 47 75 122 123 147 151 152 153 154 156 160 163 167

169 189 190 192 193 196 197 198 199 203 205 206 207 209

212 213 214 215 216 217 223 227 228 230 237 239 240 241

243 247 248 250 251 252 253 254 257 258 259 262 263 265

266 267 268 269 272 274 275 276 278 279 280 283 284 285

286 287 289 290 291 292 294 295 297 298 300 301 302 306

307 309 310 312 313 316 317 318 319 320 321 322 323 324

325 326 328 331 332 333 335 336 337 338 339 340 341 342

343 344 346 347 348 351 352 353 355 356 357 358 359 360

361 362 363 364

Motlav..i22 123 147 149 151 154 169 190 193 214 217 239 251 258

259 265 272 275 284 285 290 291 292 300 301 306 307 309

312 316 317 318 321 322 324 328 333 335 336 337 339 346
35o 35i 356 360 361

Motu 42 44 47 76 125 158 168 182 190 201 205 208 214 217

241 243 248 249 265 268 271 272 273 282 285 288 292 293

298 299 301 304 306 309 310 312 313 317 318 319 321 326

327 328 330 331 332 333 334 335 539 342 343 346 347 349

350 351 352 354 358 359 36o 361 362 363 364 365

Murray Island 275 301 305 309 361

Natalava 335 350 361

Nengone 122 214 245 265 274 291 300 306 309 322 330 333 335 361

New Britain 46 76 125 149 152 154 167 190 200 207 214 217 267

275 289 290 297 306 309 313 318 319 320 322 323 324 326

327 336 343 345 352 360

New Caledonia 273 330 333

New Georgia 149 190 192 217 285 298 300 301 305 312 313 324 350

35i 361

New Guinea 217 265 273 276 290 324 330 334 346 350 365

New Ireland 247 273 278 290 296 300 312 313 317 318 324 329 344

35o 35i 361 363 365

Nggao...20i 214 215 217 239 251 272 278 285 298 301 305 309 312

313 324 336 346 35i 352 361

Nggela.. 45 46 122 147 149 152 155 190 197 198 203 207 208 213

214 216 217 239 251 258 259 272 273 278 284 285 290 291

292 294 296 297 298 300 301 305 306 309 312 313 315 316

317 318 321 322 323 324 327 328 329 330 332 335 338 339

342 344 346 35o 351 352 357 358 360 361

58 THE POLYNESIAN WANDERINGS.

Nguna... 46 74 75 159 206 214 240 261 266 294 301 317 318 321

336 338 346 35o

Nifilole. .149 212 214 217 239 284 290 291 294 313 328 330 357 300

361 ,

Norbarbar 75 122 147 H9 J90 193 212 214 217 239 251 265 272

278 285 289 291 292 300 301 306 307 309 312 316 317 318

324 328 331 333 335 336 337 338 344 350 356 357 358 360

361

Omba... 42 74 122 147 149 152 153 169 190 206 207 212 214 217

239 254 275 278 284 285 290 291 292 295 300 301 306 307

309 312 313 317 3i8 324 333 335 336 338 346 352 356 358

360 361
Paama..252 274 312 329

Pak 122 123 147 149 190 193 203 214 217 239 250 251 265 270

272 275 278 292 300 301 306 307 309 312 316 317 318 324

328 333 335 337 339 35Q 356 357 360 361
Raluana.214 294 314 336 358

Retan... 46 122 123 149 193 292 312 317 318 33^ 336 344 357 35*>
Ruavatu.313 317 361

Saa 122 187 190 200 214 217 250 251 259 265 266 273 291 292

294 298 300 307 309 313 317 3i8 329 330 338 339 350 351

352 356 358 360 361

Santo... 47 153 199 206 216 217 228 239 253 255 278 285 292 300

302 305 312 313 316 318 324 336 343 346 352 355 356

Sasar.... 46 123 147 190 193 217 239 250 251 272 275 278 292 301

306 307 309 3i6 317 3i8 328 333 335 337 343 35<> 356 360

361

Savo 190 214 239 250 251 272 275 285 305 316 328 337 361

Sesake.. 39 40 45 46 147 J49 151 *53 *58 162 169 194 200 203

206 209 212 214 215 217 239 246 251 257 258 260 270 272

274 278 285 291 292 294 296 298 300 301 305 307 3io 312

317 318 322 329 330 333 335 336 337 338 339 342 35° 352

356 357 358 360 361

Solomon Islands 190 243 276 290 291 294 297 301 304 309 312 318

324 33° 361 363

Tangoan Santo 41 147 148 193 *94 206 209 217 239 247 265 278 301

302 313 316 328 333 338 340 346 358

Tanna...i47 150 168 173 190 194 200 207 216 217 218 238 240 247

256 251 258 266 273 274 285 290 291 293 306 308 312 313

317 319 321 322 324 328 333 334 337 34° 35© 352 358 360
361 363

Treasury Island 265

Uea 265 274 290

Ugi 214 250 259 265 292 300 317 318 357 361

Ulawa...i22 190 200 214 217 239 250 251 259 265 291 294 298 300

301 309 312 313 3i7 318 324 329 330 343 350 35i 352 356

360 361
Vanikoro 273 292
Vanua Lava 291 292 317

Vaturanga 122 123 147 190 200 201 206 214 215 217 251 257 258

259 272 278 284 285 294 298 301 305 306 309 313 316 317

318 324 328 329 330 335 336 338 346 352 358 360
V0I0W...122 147 149 151 154 190 212 214 217 239 251 265 268 275

278 284 285 290 291 294 300 301 306 307 309 312 316 317

318 322 324 328 333 335 336 337 338 339 35o 35* 357 358

360 361

Vuras... 46 122 123 147 149 214 217 239 251 278 284 285 292 301

306 307 309 312 316 317 318 324 328 333 335 336 339 343

346 356 357 358 360 361

Wango..i22 149 190 214 215 217 239 250 251 257 259 265 273 278

284 298 300 305 307 309 312 313 317 318 328 329 330 335

337 339 342 344 348 350 35 1 352 356 357 358 360 361

POLYNESIAN REUCS IN MELANESIA. 59

We shall next take up in a series of tables the relations of Melane-
sian tongues with the Proto-Samoan (uniformly exhibited in bold-
face type), the data upon which each of these mutations rests, and
certain other results which have a bearing on our study. These
tables are prepared for all the Melanesian languages for which a
sufficiency of data is available. Several of the groups in the pre-
ceding catalogue, while seeming to be well enough equipped with
examples, have not been included in this tabular review for the
reason that earlier compilers, to whose efforts we owe these data,
have not been precise in identifying the language. Such omitted
collections are to be seen under the headings of New Guinea, New
Britain, New Ireland, large lands in which are many distinct
languages.

Prefixed to each of these tables will be found two particular notes.
The figure following the note Polynesia expresses the number of
words in the available material in each language which are identi-
fied as of Polynesian stock. The note of Quality is a valuable index
which will be explained and discussed at length after the prelimi-
nary matter has been arranged.

The final notes in each table call for slight explanation. Under
the note Identical are assembled those words in which the Melane-
sian is the same in consonant and vowel structure as either the
Proto-Samoan or the modern Samoan, or where the vowel change
is so slight as to be explained by the difference in the system of
reducing the language to our alphabet by those missionary collec-
tors whose zeal must serve as the excuse for their lack of skill.
The second note refers to those words, otherwise identical, which
have lost their ending by abrasion. Under Consonant Identity are
listed those words of identical consonant skeleton where the vowels
have undergone change, and the converse is the case under Vowel
Identity. In the next pair of notes we record those cases where
the consonant skeleton has undergone a mutation, but where the
vowels remain constant or terminal abrasion has taken place.

60

THE POLYNESIAN WANDERINGS.

ALITE.

Polynesian 2 1
Quality 57

a a

e 00

i t u u

1 /

wg ng n / m w

h s, t

8 S, t, —

v ku

f V

k — , ng t w, — p 6

a-a 217, 239, 250, 278, 291, 300, h-s 352

309, 3i3, 3i7, 337, 35o, 360 ht 278

\-i 259, 285, 291, 300, 313, 330, s-.y 239

35o s-t 337

0-0 259, 285, 309 s— 251

u-w 239, 278, 306, 317, 330, 360 t-m 217

1 ' 309, 3i3, 35o t— 306, 350, 352

ng-ng 285, 309, 350 mm 217, 313, 317

k — 300, 301, 306 x-ku 291

k-ng 251 f-v 259, 360

n-Z 259, 317, 330 p-b 250, 285

Identical 300, 309, 3 1 3

Vowel identity 217, 300, 317, 330

Consonant mutation :

Vowel identity 259, 278, 285, 350, 360

Terminal abrasion 306

Frontal abrasion 250, 301, 306, 350, 352

Frontal accretion 239, 291

POLYNESIAN RELICS IN MELANESIA.

61

ALO TEQEL.

Polynesian 26
Quality 19

a a, e, o, u
e e, u o e
i e u u, e, i

1 /, t, ng
ng ng, n n n, t m m, ng

h
s s

v w

f V, w

k g, w t t, m, — p p

a-a 123, 147, 214, 217, 250, 309, ng-w 350

318 k-# 250, 251, 300, 301

a-e 294, 324, 337, 350 k-w 306

a-o 239, 307, 317, 328 n-n 147, 317, 343, 350

a-w 250, 307, 316 n-t 328

e-e 122, 190 s-,y 239, 251, 272, 337

e-u 272 t— 294, 306, 318, 324

i-e 300 t-t 350

0-e 123 t-tn 217

u-w 272, 306, 316 mm 217, 316, 317, 318, 324, 328

u-e 294 m-ng 316

u-i 343 v-w 307

I-/ 123, 307, 309, 350 f-v 122, 147, 214, 294

\-ng 123 f-w 272, 360

K 335 p-/> 190, 250

ng-ng 309

Identical (terminally

abraded) 190, 309

Consonant identity. . . .317, 337
Consonant mutation:

Vowel identity 147

Terminal abrasion 147, 190, 214, 217, 239, 300, 306, 317, 318, 328

Frontal accretion 301

Terminal accretion 122, 324, 335, 350, 360

62

THE POLYNESIAN WANDERINGS.

AMBRYM.

Polynesian 27
Quality 44

a a, e, i, u
e a o 0, u
i i u 11, a, i, 0

1 I, r
ng ng n n m m

h h, —
S h

V /, w
f 7;, b, h
k &, ^, h, — t t, s, dr p —

a-a 147, 215, 258, 284, 300, 318, k-k 194, 300

324 kg 297

a-e 239, 258, 324, 352 k-h 258

a-i 293, 350 k— 305

a-w 252 n-n 242, 290

e-a 297 h-h 215

i-i 194, 215, 242, 290, 297, 300, h— 352

305, 312, 350, 352 s-h 239, 252, 298

0-0 252, 336, 356, 357 t-t 258, 324, 352, 356

o-w 287, 357 t-s 329

u-w 194, 212, 298 t-dr 357

u-a 258 mm 258, 312, 318, 324

u-i 212, 239 v-/ 242

U-0 212, 284, 329 \-W 29I

1-/ I94, 212, 252, 284, 287, 297, i-V 147, 215, 290

305, 312 i-h 290, 329

1-' 336, 350 i-b 287

ng-ng 274, 336, 350 p— 284

Identical 194, 300, 305

Identical (terminally

abraded) 312, 356

Consonant identity. . . .212, 274, 284, 318, 324, 352
Consonant mutation:

Vowel identity 215

Terminal abrasion 147, 242, 287, 290, 312, 318, 329

Frontal abrasion 284, 305, 350

Frontal accretion 212, 242, 274, 298

Terminal accretion. . . .258, 336

POLYNESIAN RELICS IN MELANESIA.

63

ANEITYUM.

Polynesian

67

Quality

46

a a, c, i,

0, u

ee,

a,

i o o, i

, u

i i, a,

e

, o u
1 I, r, j,

u, e, i, o, v, p
k

ng ng, k, g, nj

n n, ny
h h, j, f

s s, h

m m

k k, g, ng

t t, s, th

V V, p, w

f /, p, u, k, w,

p p, I

h

a-a 147, 188, 198, 207, 214, 217,
229, 256, 258, 260, 262, 279,
282, 291, 293, 294, 307, 309,
312, 313, 317, 318, 320, 323,

324, 337, 339. 340, 350
SL-e 198, 216, 217, 256, 268, 278,

294
a-i 271, 274, 350
a-o 124, 294
a-w 168, 307, 316, 328
e-e 173, 240
e-a 126
e-i 191
\-i 154, 194, 204, 241, 268, 288,

291, 319, 323
i-a 154
i-e 342
i-o 269
0-0 126, 205
o-i 285
o-w 205
u-u 168, 194, 269, 273, 288, 294,

320, 326, 344
u-e 278, 306
u-i 271
u-o 329, 360
u-v 206
u-p 262
1-/ 191, 228, 241, 307, 309, 310,

316, 320, 339, 349
\-r 126, 194, 205, 260, 268, 288,

335
l-fc 312, 350

1-; 154, 312

Identical 126, 173, 194, 241, 293,

294>3i9, 323.335
Identical (ter. abra) . 198, 207, 256, 282, 317,

349
Consonant identity. . 124, 285, 301, 316, 324

Vowel identity 318

Consonant mutation:

Vowel identity. . .262, 283, 288, 291, 342
Vowel identity (ter.

abra.) 147, 188, 273, 309

ng-ng 285, 350
ng-fe 309

342, 309

274

126, 194, 269, 301, 328

188, 258, 306, 361

191, 204

147, 271, 294, 317

328

278

340

339 [344

188, 198, 206, 323, 337, 342,
204, 205

126, 207, 216, 217, 256, 258,
268, 269, 274, 282, 293, 294,
306, 324, 347, 349, 350

318, 327, 337

168

124, 191, 198, 216, 217, 256,

258, 274, 282, 294, 312, 316,

317, 318, 319, 320, 323, 324,

326, 328, 340

293, 294
147
206

214 [360

147, 271, 273, 288, 294, 329,
283

294, 310

307

291 [361

124, 173, 207, 241, 279, 285,

279

Terminal abrasion. 147, 154, 188, 193, 214,
270,273,285,306,309,
316,318,328

Frontal accretion. .147, 154, 204, 206, 217,
228, 229, 256, 268, 269,
270,273,274,285,288,

319,328,342,344
Terminal accretion 206, 274, 307, 310, 319,
324,326,339,340,347

ng-g

ng-nj

k-k

kg

k-ng

Tl-n

ti-ny

h-h

h-y

h/

s-s
s-h
U

t-s

t-th

mm

t-f

i-p

i-k

i-u
i-h

i-w
\-v
\-p
\-w

v-P
v-i

64 THE POLYNESIAN WANDERINGS.

ARAG.

Polynesian 41
Quality 75

a a, e, o

e e, a, i 00
i * u u, i, o

1 /, r

ng ng n n, I m m

h h

s h

V IV, —
f V, W

kg t t, d, h p p, v, kpw

a-a 152, 160, 193, 209, 214, 215, n-w 285, 290, 292, 309, 317, 328,

217, 250, 258, 284, 290, 291, 330

292, 294, 309, 312, 313, 317, n-l 259

318, 324, 328, 352, 356 hh 215, 278, 352

a-e 278, 300 s-h 239

a-o 357 t-t 160, 209, 217, 258, 294, 306,

c-e 122, 153, 190, 290, 318 318, 324, 352, 356

e-a 272 t-d 357

e-i 152 t-h 329

\-i 149, 193, 215, 259, 290, 291 m-w 217,258,312,313,317,318,

300, 312. 3*3, 329, 33o, 352, 324, 328

0-0 153, 160, 259, 285, 309, 326, \-w 291

356, 357 v— 152

u-w 212, 258, 272, 278, 284, 292, f-v 122, 193, 214, 215, 272, 290,

294, 306, 317, 328. 330 292, 294

u-i 212, 329 i-iv 259

u-o 239 p-p 190, 250

\l 152, 153, 212, 284, 312, 313 p-i; 284

\-r 149, 160, 335, 336 p-kpw 285
ng-ng 209, 309, 336
k-g 149, 193, 250, 258, 300, 301,

306, 357

Identical 153. 160, 190, 209, 217, 309, 312, 313, 317, 318,

324, 328, 330, 335, 336, 352, 356

Consonant identity. . . .152, 212

Consonant mutation :

Vowel identity 193, 214, 215, 258, 259, 284, 285, 290, 291, 292,

294, 306

Frontal abrasion 329

Terminal accretion. . . .250, 272
Inner abrasion 152

POLYNESIAN RELICS IN MELANESIA.

65

BAKI.

Polynesian 46
Quality 26

a a, e, i, 0, u
e e, a, i, o o 0, i
i i, o, u u u, i

1 /, j, mj
ng ng, n, g n n m m

h—
s y, k

v v, u
f v, b, mb
k k, mk t t, j, dr p b, mb

a-a 147, 240, 263, 292, 308, 309, l-j 309, 312, 313

3!3, 34o \-tnj 336

a-e 156, 254, 315, 317, 320, 340, ng-ng 285, 336, 346

346, 350, 352a ng-w 308, 350

a-i 350 ng-# 309

a-o 156, 228, 278, 292, 312, 364 k-k 149, 156, 194, 2ii, 305

a-u 290 k-mk 297

e-e 190 n-n 147, 156, 254, 290, 292, 317,

e-a 240 321

e-ii 210, 297 h — 278, 340

e-o 247, 290 &-y 182

\-i 149, 194, 2ii, 254, 273, 290, s-k 263

297, 308, 321, 346, 350, 352, t-t 247, 346, 350, 352

362 t-j 2IO, 211, 346, 358

\-o 285 t-dr 359

»« 305, 313 m-w 156, 254, 312, 313, 315, 317,

0-0 147, 263, 282, 285, 309, 336, 320, 321, 340

362 X-V 210

0-i 336 x-u 291

VL-U 156, 182, 194, 211, 228, 240, i-V 254, 290, 292

272, 278, 282, 292, 317, 320, i-b 273, 282, 290

321, 358, 359 f-mb 147

U-i 247, 359, 364 p-6 190, 285

1-/ 149, 182, 194, 228, 297, 305, p-mb 190, 247

308, 315, 320, 350, 359, 362,

364

Identical 194, 254, 362

Consonant identity. ... 156, 340, 352

Vowel identity 182, 273

Consonant mutation:

Vowel identity 147, 190, 211, 282

Frontal accretion 149, 228, 247, 320

Terminal accretion .... 305, 358
Metathesis 298, 321

66

THE POLYNESIAN WANDERINGS.

BAR AVON'.

Polynesian 30
Ouauty 66

a a, e
e e, a o 0, a, e, u

i i, u u u

I /. r, —
ng ng n n m m

h—
s

V V, w
f v, p, w
k k, — t /. ;/(/, A- p />. /», wife

a-a 74, 147, 156, 207, 214, 215, k-k 149

217, 279, 284, 290, 291, 294. k — 156, 300, 323

313, 320, 323, 324, 346 n-» 74, 147, 156, 290

a-p 360 h — 215

e-e 190, 257 t-i 74. 207, 217, 294, 324, 346,

e-a 122 358

\-i 149, 215, 290, 313, 319. 323, t-nd 329

346 t-k 270

l-M " 74 m-7;/ I56, 2 17, 257, 3I3, 319, 320,

0-0 336 323. 324- 327

o-« 147. 257 \-v 291

O-t 257 V-U' 122, 147

o -it 285 i-v 215, 329, 360

u -K 156, 207, 270, 273, 284, 320, f-/> 273

358, 360 t-W 12 2, 147, 214, 290, 294

1-/ 313, 320, 336 p-p 284

\-r 257, 327, 358 p-6 190, 285

1 — 284 p-mb 207
ng-ng 285, 336, 346

Identical 217, 313, 324, 336, 346, 358

Consonant identity. ... 74, 149, 257, 327

Vowel identity 156, 284, 320

Consonant mutation :

Vowel identity 190, 207, 214, 215

Terminal abrasion. . .273, 290, 294

Terminal abrasion 149, 270, 273, 285, 290, 294, 300

Frontal accretion 74, 215, 291, 336

Terminal accretion 285, 300, 323, 336

BAURO.

e-a 47 h-.v 47

\-i 47 t— 47

POLYNESIAN RELICS IN MELANESIA. 67

BELAGA.

Polynesian 17
Quality 100

a a, i, ai
e c, i o 0
\ i uk

1 I, V

ng ng n n m m

h h, th

S S, til

V V

f x;
kg,- t t p i', »//>

a-a 147, 152, 203, 207, 258, 284, k-g 258, 300, 338

300, 317, 338, 339, 342, 346. k— 211

352 n-n 290, 317, 342, 343

a-i 203 h-// 338

a-a?- 290 h-^ 352a 339

e-e 290, 338 s-.y 203

e-i 152 s-M 342

\-i 211, 285, 290, 3OO, 342, 346, t-t 207, 211, 258, 343, 346, 352a

352 m-m 258, 317

0-0 285 v-x' 152

u-w 207.211,258,284,317,343 i-v 147.290

1-/ 152, 284, 339 p-v 284

!-/• 203 p-m6 207, 285
ng-ng 285, 346

Identical 317, 346

Terminal abrasion 343

Consonant identity. . . . 152, 203

Vowel identity 211

Consonant mutation :

Vowel identity 207, 258, 284, 285, 290, 300, 338, 339, 342, 352

Terminal abrasion. . 147

Terminal abrasion 147, 343

Terminal accretion .... 339

68

THE POLYNESIAN WANDERINGS.

BIERIAN.

Polynesian 52
Quality 48

a a, e, o, u

e e, a, i 00
\ i, u u u, i

1 /, m, —

ng ng, m n n, I m m

h h

s h, s v u

f /, v, b

k k t t, d, s, h p b, m

a-a 156, 199, 207, 216, 217, 240, 1— 364

256, 258, 260, 263, 270, 278, ng-ng 285, 309, 336, 350

290, 291, 292, 293, 294, 300, ng-m 199

302, 309, 312, 313, 315, 317, kk 156, 258, 297, 300, 302, 305,

318, 322, 323, 324, 340, 352, 323

364 n-n 156, 292, 317, 321

a-e 320, 350 n-Z 259

a-o 148, 350 hh 278, 340, 352

a-u 252 s-.y 323

e-e 210,290,311,318 s-h 252,263,298,344

e-a 240 t-t 148, 199, 216, 217, 256, 258,

e-i 297 270, 293, 294, 302, 318, 324,

\-i 259, 270, 290, 291, 297, 298, 358

300, 302, 312, 321, 323, 329, t-d 210

35o, 352 t-s 350, 352, 356

\-u 365 t-h 329

0-0 252, 259, 263, 285, 309, 311, mm 156, 216, 217, 258, 312, 313,

336 315,317,318,320,321,322,

u-w 148, 216, 258, 270, 278, 292, 323, 324, 340

294, 298, 317, 320, 321, 344, v-w 291

358, 364 i-f 282, 290, 292

u-i 156, 329 f-v 210, 259, 293, 294, 329

\l 252,260,297,305,309,311, i-b 148

312, 315, 320, 336, 350 p-6 285

1-;;? 322 p-m 207

Identical 216, 217, 256, 258, 260, 270, 290, 292, 302, 312,

317, 318, 323, 324, 336, 340, 358
Consonant identity. . . . 156, 297, 320

Vowel identity 263

Consonant mutation:

Vowel identity 199, 210, 294

Terminal abrasion. .285

Terminal abrasion 207, 305, 356

Frontal abrasion 313

Frontal accretion 210, 256, 263, 278, 293. 297, 298, 302, 309, 311,

312, 315, 320, 336, 344, 358
Terminal accretion .... 344
Metathesis 321

POLYNESIAN RELICS IN MELANESIA. 69

BOUGAINVILLE. BRIERLY ISLAND. BRUMER ISLAND.

a-a 192, 35° a-a 294, 317, 324, a-a 324

\i 259, 350 340, 347 \-i 298

i-e 149 e-e" 190 u-u 298, 344

0-0 259 Q-i 190, 347 s-s 344

\-l 192, 350 h-6- 340 s-sh 298

I-y 149 t-/ 324, 347 t-t 324

ng-wa 35° t-k 294

k-& 149 n-n 317

n-/ 259 m-ra 317, 324, 340

t-t 192 i-p 294

t-nd 350 p-6 190

f-h 259

p-/> 192

BUGOTU.

Polynesian 33
Quality 79

a a, at, 0
e e, a o 0, a

i i u w, 0

1 /, r, t, th
ng wa n n, ng, gn m m, r

h h
s h

v m6
f v
k k, g t t, nd, th p p, mb

a-a 147, 213, 214, 217, 239, 275, ng-ng 213, 285, 309, 346

284, 309, 312, 313, 316, 317, k-k 251

322, 328, 338, 339, 346, 351, kg 275, 301, 305, 306, 338

352 n-n 147, 290, 317, 330

a-ai 290 n-ng 351

a-o 275 n-gn 328

e-e 122, 257, 290, 338 h-h 278, 298, 338, 339, 352

e-a 272 s-h 239, 251

a 285, 290, 298, 305, 312, 313, t-t 217, 306, 343 ,346, 352

329, 330, 346, 352 t-nd 329

0-0 147, 285, 309 t-th 343

o-a 257 mm 217, 257, 312, 313, 316, 317,

U-U 239, 272, 278, 284, 298, 306, 322, 328

316, 317, 328, 329, 330, 351 m-r 322

U-0 343 \-mb 291

I-J 257, 305, 312, 313, 316 f-v 122, 147, 213, 214, 272, 290,

\-r 272, 284, 322 329

\-t 339 p-p 284

\-th 309 p-mb 285

Identical 217,251,278,312,313,316,317,330,346,352

Consonant identity. . . .301, 305

Consonant mutation:

Vowel identity 147, 2 13, 2 14, 239, 284, 285, 290, 298, 306, 309, 329, 338, 339, 35 1

Frontal accretion 272, 275, 305, 328

Terminal accretion. . . . 339, 351

70

THE POLYNESIAN WANDERINGS.

BUKA.

Polynesian 24
Quality 70

a a, c, u

e e, a, 0 o 0, a, u
i i, a, e u u, a, 0

1 /, r, nd

ng nq, n n n, gn, I m ;;/

h .9

s s, h

V

f h

k — t /, d, ,? p b, mb, v, w

a-a 207, 217, 313, 318, 324, 328,

35o, 352, 360

a-e 158, 300, 308

a-M 239

e-e 190, 318, 363

e-a 190, 257

e-o 247

\-i 259, 285, 300, 308, 343, 350,

352

■-a 355

i-e 298, 313, 330

0-0 259, 285, 263

o-a 355

o-w 285

U-« 158, 207, 237, 247, 298, 306,

328, 330, 344

U-a 343

u-o 298, 360

1-/ 3i3» 350

1 ' 257, 355

\ nd 308

ng-ng 285, 308, 350

ng-w 285

k — 300, 306

tin 259, 330, 343

ti-gn 328

n-/ 259

h-J 352, 362

S-s 298, 343, 344

s-h 239

t-f 207, 217, 306, 318, 324, 350,

352, 355

t-d 350

t-s 247

mm 158, 217, 258, 313, 318, 324,

328

'-* 259, 360, 363

p-b 190, 285

p-mb 207

pv 247

p-w 1 90

Identical

Terminal abrasion
Consonant identity .
Vowel identity

Terminal abrasion .
Frontal accretion . . .
Terminal accretion .

.217, 318, 324, 328, 350, 352

•344
•330
.207, 285, 306

.285

.158, 239, 285, 298, 306, 308

•257, 308

POLYNESIAN RELICS IN MELANESIA.

71

BULULAHA.

Polynesian 19
Quality 74

e e

ng ng

a a

I /

II n

h s

s s

O O

U It,

m ;»

V

w

f

h

k—

t t, -

P

P

217,

239, 250,

291, 300,

301,

k—

250,

300,

301,

306

309,

317, 350,

351. 352,

360

n-w

259,

317.

33o

190

h-s

352

259,

291, 300,

329. 33o,

350,

S-S

239,

298,

35i

352

u

329

259,

309

t—

217,

306,

35o,

352

239.

298, 306,

317. 329,

330,

m-m

217,

317

35i,

360

V-W

291

301

f-h

259.

329.

360

309.

35o

P-P

190

a a

e-e
\-i

0-0
u-u

u-i
\l

ng «2 309, 35o

Identical 190, 300, 309, 317, 330

Consonant identity .217

Vowel identity .... 250, 306, 350, 332

Consonant mutation:

Vowel identity. .259, 291, 329, 360
Frontal abrasion . . 35 1
Frontal accretion. .239, 298

DKNI.

Polynesian i i
Quality 63

a (/, (

', ><

e

0 0

i

/'

1 /
n

u

It

ng

m in

h h, -

s

v b

f V

k k

t t, k

p mb

a-a

214.

278, 324.

339,

360

h-h

339

a-e

307,

335

h—

278

sl-h

307,

324

t-t

306,

358

\-i

194

t-k

324

0-0

285

m-;»

324

u-u

194.

2 12, 278,

306,

335- 3

58

v-b

307

1 /

194.

212, 307,

335,

339

f-v

214.

360

k-k

194.

306

p-mb

285

I den

tical

194-

212,2

78, 306,

339,

Consonant mutation :

358

Vow

el identity.

.28;

5

72

THE POLYNESIAN WANDERINGS.

DUFAURE ISLAND.

a- a

e-e
e-a

\-i

\-u

o-o
o-u
u-u

u-a
u-o
1 /

\r

a-a
u-w

167

167

167

\-n

ng-

167
167

DUKE OF YORK.

Polynesian
Quality

38
79

a a

e e,
i i, u

ng ng, n

k k, —

74, 147, 164, 169, 207, 214,
217, 275, 290, 292, 294, 300,
309, 3i3, 3i6, 318, 320, 324,
328, 35o, 352
122

272, 290

149, 169, 269, 290, 300, 312,
3i3, 3i9. 321, 327, 35o, 352
321, 329
269, 336

285, 325

207, 272, 292, 316, 320, 327,

328, 329, 332, 343, 358

325

3H

H9. 164, 309, 312, 313, 3i4,

316, 336, 350

320, 332

1 I, r
n n
h—

s

t t, nd

o o, u

u u, a, 0

m m

v

f b, w,
P ft

ng-w 74

k-k 149, 269, 275

k — 300

n-n 147, 169, 290, 292, 321, 328,

343
352
74, 164, 207, 217, 269, 294,

318, 324, 325, 343, 350, 352,
358

329

mm 217, 312, 313, 314, 316, 318,

319, 320, 321, 324, 325, 327,
328

122, 147, 169, 214, 290, 292,
294, 329

h-

U

t-nd

f-w

ng-ng 169, 285, 309, 336, 350
Identical

i-b
f—

p-b

272
360
207, 285

Terminal abrasion
Consonant identity .
Consonant mutation :

Vowel identity. . .

Terminal abrasion

Terminal abrasion

217, 269, 275, 313, 316, 320, 324, 327, 328, 336,
35o, 352, 358
74. 149, 309, 312, 318, 343

321, 325

122, 147, 169, 207, 214, 272, 292, 300

294
74, 147, 164, 169, 207, 217, 231, 275, 290, 292,

294, 300, 309, 313, 316, 318, 320, 324, 328, 350,

352

Frontal accretion 272, 332

Terminal accretion 164, 169, 300, 314, 320, 336, 358

POLYNESIAN RELICS IN MELANESIA.

73

EPI.

Polynesian 14
Quality 43

a a, u
e a, 0 o 0, a

i i u u, i

1 I, r

ng n n m m

h h

s j-

V

f V

k k t / p t, 7726

a-a 200 \-r 75

a-w 202 k-k 202

e-a 47, 186 n-n 286, 330, 343

e-o 363 h-h 47, 363

i-* 47, 200, 330 s-s 343

0-0 75, 363 t-t 47, 75

0-a 186 m-i;j 200, 202

u-u 75, 200, 212, 330, 343 f-v 363

u-t 212, 286, 330 p-v 186

1-/ 200, 212 p-mb 286

Identical 75, 200, 330, 343

Consonant identity. . . .202, 212

Terminal abrasion 286

Frontal abrasion 75

Frontal accretion 186, 202, 343

74

THE POLYNESIAN WANDERINGS.

EROMANGA.

Polynesian 26
Quality 58

a a, e, 1, 0, u
e e 00

i i, c, 0 u k, e, 0

1 /, r, h

ng ng, g n n m m

h h, s

s s

V 11

f v, p, b, s
k k, h t /, d, s p />, w/>

a a 218, 270, 274, 294, 337, 338, ng-</ 350

349, 360 k-fc 194, 305, 361

a-e 217, 278, 317, 350 k-h 338

a-i 217, 290 n-w 290, 317

a-o 350 h-/? 278

a-w 324 h-j 206, 338

e-e 240 s-j 337

\-i 194, 270, 274, 290, 305, 312, t-t 217, 270, 274, 294, 324, 349,

35o 350

\-e 206 t-d 352a

i-o 305 t-s 329

0-0 282, 285 m-w 217, 274, 312, 317, 324

U-« 194, 240, 270, 273, 282, 360 V-M 291

u-e 278 f-v 282, 290, 294

U-o 317, 329 t-p 206, 2736

1-/ 312, 349, 350 f-6 329, 360

\-r 194, 312 is 290

l-k 305 ' v-P 285

ng-n<7 274 p-m/> 218, 361

Identical 194, 218, 270, 274, 285, 337

Terminal abrasion. . .312, 349
Consonant identity. . . .217, 317
Consonant mutation:

Vowel identity 282, 350, 360

Terminal abrasion. . .273, 294

Terminal abrasion 206, 273, 290, 294, 312, 329, 338, 349

Frontal accretion 270, 273, 282, 291, 312

Terminal accretion. . . .218, 305, 317, 337

POLYNESIAN REUCS IN MELANESIA.

75

FAG AN I.

Polynesian 39
Quality 82

a a, e, i, 0

e e, a o o, a

i i, e, u u 11

1 r

ng ng n n, ng, ! m >//

h ^

s s

v /
f /, v
k &, </ t /, k, — , w p />, &, /

a-a

H7,

214,

217,

239,

250,

258,

k-k

H9.

250

278,

292,

294,

300,

307,

309,

k-g

214,

251,

258,

300,

357

312,

313,

316,

318,

324,

328,

tin

i47,

290,

292,

317.

328,

33o,

342,

350,

352,

357,

360

342

a-e

290

n-nq

35i

at

292

n-l

259

a-o

250

h-s

352£

\

e-e

190,

257,

290,

318

S-s

239,

251,

298,

342,

35 1

e-a

122,

272

t-t

258

\-i

149,

259,

273,

278,

285,

300,

t-k

329,

350

312,

329,

330,

342,

350,

352&

t—

294,

318,

352,

357,

358

i-e

290

t-w

217

\-u

298,

313

m-m

217,

257,

258,

312,

313,

316,

0-0

147.

259,

285,

309.

357

317.

318,

324,

328

o-a

257

v-/

307

U-tt

239,

258,

272,

2 73,

292,

294.

t-f

122,

H7>

214.

2 59.

290,

292,

298,

316,

317.

328,

329.

33o,

294,

329,

360

35L

358,

360

f-v

272,

2 73

1 -;■

257,
313,

272,
316,

278,
350,

307-

358

309.

312,

V-P
p-6

190

285

ng-w^

278,

285,

309,

35o

P7

250

Identical 147, 190, 239, 294, 309, 312, 316, 317, 328, 330,

342, 36o
Consonant identity- • • 122, 257, 272, 290, 292, 298, 313

Vowel identity 217

Consonant mutation :

Vowel identity 214, 258, 259, 273, 278, 285, 300, 307, 318, 329,

350, 352

Terminal abrasion 324

Frontal abrasion 351, 352

Frontal accretion 298, 351

Terminal accretion. . . .257

76

THE POLYNESIAN WANDERINGS.

GOG.

Polynesian 38
Quality 76

a a, e
e e, 0 00

i i, e, u u u, a, i, 0

1 /, r, ng, nd

ng ng, n n n, ng m m

h s

s s

X V, w

i v, w
k g, w t t, m p b, v, mb, kpw

3L-a 123, 147, 203, 214, 217, 239, ng-n 285

268, 301, 307, 309, 313, 3i7, k-0 I5i. 251, 300, 301, 33», 357

318, 322, 324, 328, 339, 356 k-w 306

a-e 324 n-n 147, 31?, 328, 343

e-e 122, 190 n-w^ 342

e-o 272 h-$ 206, 338, 339

\-i 154, 312, 313, 342 s-s 203, 239, 25i, 278, 342

i-e 300 " t-t 268, 306, 318, 324, 343, 356,

\-u 206 357, 358

0-0 147, 285, 309, 336, 356, 357 t-m 217

u-u 239, 284, 298, 306, 343, 358 m-w 217, 312, 313, 317, 3i», 322,

u-a 151 324, 328

11-* 301 v-t; 307

u-o 272 v-w 206

U 123, 154, 203, 284, 307, 309, f-v 122, 147, 214

312, 313, 339 f"w 272, 36o

\ r 268, 272, 322, 336, 358 p-r 284

\-ng 123 V-b 190

\-nd 335 p-w6 151

ng-n^ 154, 298, 309, 336 p-kpw 285

Identical 239, 309, 313, 356, 35§

Terminal abrasion. . .154, 203, 268, 307, 312, 317, 318, 328, 336, 343
Consonant identity. . . 301, 324

Vowel identity 217, 306

Consonant mutation :

Vowel identity 147, 214

Terminal abrasion. . . 190, 206, 284, 285, 338, 339, 342, 357

Terminal abrasion 151, 154. 190, 206, 268, 284, 285, 300, 306, 307,

312, 317, 318, 324, 328, 338, 342, 343

Frontal accretion 203, 272, 298, 301, 312, 339

Terminal accretion. . . .122, 336

POLYNESIAN RELICS IN MELANESIA.

77

GUADALCANAL

Polynesian 6
Quality 66

a a
e a oo

\ i u u, i

1 r, -
ng ng n n m m

h ta
s s

v

f z>
k t p b

98, 309

a-a

278, 309, 328

ng-ng 285, 2

e-a

272

n-n 328

i-i

285

h-tfz. 278

0-0

285, 309

S-i- 298

u-u

272, 278, 298, 328

m-ra 328

u-i

298

f-v 272

\r

272

p-b 285

1—

309

Identical 278, 3:

Consonant identity .... 296
Consonant mutation:

Vowel identity 285

IAI.

Polynesian 5
Quality 40

ng

k k

a-a

324

a-e

294

a-o

318

\-i

i94,

259

0-0

259

u-u

194

VL-0

294

k-k

194

a a,

e, 0
0 0

u u,

0

1 r

n n

m m

h

s

V

f v, -

t t,

k

p

\-r

194

n-n

259

U

294

t-k

318,

324

m-ra 318,

324

i-v

294

f—

259

Identical 194

Vowel identity 324

78

THE POLYNESIAN WANDERINGS.

KABADI.

Polynesian 3
Quality 33

a-a 74, 292

\-i 330

\-u 74

u-u 292, 330

Identical 33»

ng-n 74

n-n 292, 330
t-fe 74

i-v 292

KABAKADA.

Polynesian 6

Quality '50

a-a 147, 214,
e-a 122

i-i 327

0-0 147
o-« 285

u-u 320, 327

1-/ 327

320

\r

Ug-ng
n-n
m-m

f-w
p-6

^20

285

i47

320, 327
i47, 214

285

Identical

21

4-

320,

327

KAIJL.

a-a

294.

322.

1?

312

I-tt

329

0-?*

296

11-7/

329

1 /

312.

322

Polynesian

6

Quality

17

n-n

296

h-s

352

t-*

294. 352

t-cW

329

mm

-1 12 122

f-fc

294, 296,

329

Identical 312

POLYNESIAN RELICS IN MELANESIA.

79

KING.

Polynesian 32
Ouauty 53

a a, e, t, o, u
e e, a, i, 0 o o, a, i, u
i i, e, u u u, i

1 /, r

ng ng, n n n, ng m m

h s

s s

V

f v, w, p
k. k t /, (/. k, n p 6, mb

a-<7 147, 213, 217, 258, 294, 309, 1/ 309, 312, 313, 336, 349, 350

313, 318, 324, 329, 350, 352 !-/• 320, 358

a-e 151, 323 ng-ng 213, 264, 309, 336, 350

3L-i 320, 323 ng-« 74, 151

a-o 213 k-k 196

R-11 74, 266 n-n 147, 15 r, 296, 321

e-e 151 n-m7 266, 342, 343

e"« 313 hr* 47, 352, 363

e'* 47 8-J 264, 323, 342, 343, 344

e-o 247, 363 t-i 74, 196, 217, 247, 258, 266,

\ i 47, 196, 312, 319, 321, 323, 294, 318, 324, 350, 352, 358

352 t-d 47

i-e 342 t-k 349

i-u 74 t-« 329

0-0 196, 336, 363 m-m 217, 258, 312, 313, 318, 319,

0-0 264 320, 321, 323, 324

0-1 349 f-v 147, 213, 363

o-it 296 i-w 294, 329

u -it 258, 320, 321, 329, 343, 344, i-p 296

358 p-ft 247

U-/ 247, 296 p-77i^ I5I

Identical 196, 217, 321, 324, 336, 358

Terminal abrasion. . .309, 318, 344
Consonant identity- ■ 264, 312, 313, 320, 350
Consonant mutation:

Vowel identity 343, 352

Terminal abrasion . . . 294

Terminal abrasion. . . .266, 294, 309, 318, 329

Frontal abrasion 151

Frontal accretion 213, 266, 312, 318, 358

Terminal accretion. . . . 147, 213, 264, 336, 343
Metathesis 196

80 THH POLYNESIAN WANDERINGS.

LAKON.

Polynesian 32
Quality 75

a a, e, i
e a, e o o, a

i i, e u u, an

1 /, ng, dr
ng ng, n n n m m

h h
s s, h

v

f t;, w
k g, w t t, m p p, v, kpw

a-a 123, 147, 193, 214, 217, 239, ng-ng 309, 350

292, 301, 309, 313, 317, 318, ng-n 285

324, 328, 335, 337, 35o, 35i k-g 193, 275, 301

a-e 275, 324, 350 k-w 251, 306

a-i 275 n-w 147, 292, 328, 351

e-e 190 h-/i 206, 278

e-a 122 s-fc 239, 351

I-» J93, 3i3 s"^ 337

i-e 206, 312 t-* 306, 318, 343, 350, 356

0-0 147, 285, 356 t-ra 217

0-a 123 m-w 217, 312, 313, 317, 318, 328

u-u 278, 284, 292, 306, 328, 335, i-v 122, 147, 193, 214, 272, 292

343, 35i f"w 2o6> 36o

u-au 239 p-p 190

1-Z 123, 284, 309, 312, 313, 350 pv 284

\-ng 123 p-fe/>w 285

l-dr 335

Identical 214, 313, 328, 337, 351

Terminal abrasion. . . 190, 278, 309, 318, 343, 356
Consonant identity. . . .312, 317, 324, 350

Vowel identity 217

Consonant mutation :

Vowel identity 147, 285

Terminal abrasion. . . 193, 206, 284, 292, 306, 351

Terminal abrasion 190, 193, 206, 217, 278, 284, 292, 301, 306, 309,

312, 317, 318, 343, 351, 356

Frontal accretion 272, 301, 312

Terminal accretion. . . .328, 335, 350

POLYNESIAN RELICS IN MELANESIA. 81

LAMASSA.

Polynesian 29
Quality 69

a a, e, u
e e, a, o o u
i i, e, u u u, 0, i

1 /, r, nl
ng ng, n n n, ng m m

h s
s .y

v

f /, t, />, — , w
k fe t 2, fc, rt p &, w6, m

a-a 213, 216, 292, 294, 309, 313, k-k 196

316, 317, 318, 324, 350 |! n-n 292, 296, 317, 343

a-e i5*» 35o n-w<7 266, 342

a-w 74, 266 h-s 47, 280, 363

e-e 151, 190 s-j 342, 344

e-o 47 t-< 47, 74, 196, 247, 266, 294,

e-o 247 318, 324, 342, 350, 358

\-i 47, 196, 312, 313, 319, 342 t-k 216

i-e 342 t-n 329

i-w 74 m-*» 216, 312, 313, 316, 317, 318,

o-w 196, 285, 296, 363 319, 324

u-w 247, 292, 316, 329, 343, 344, f-/ 292, 294

358 i-v 280, 363

u-o 216 i-p 296

lit 317, 343 i-w 329

1-/ 309, 312, 313, 350 f— 213

\-nl 316 p-b 190, 247

\-r 358 p-mfe 151, 285

ng-ng 213, 285, 309, 350 p-w 190

ng-n 74, 151

Identical 313, 319, 324, 344, 358

Terminal abrasion. . .292, 294, 309, 312, 318
Consonant identity. . . . 196, 213, 317, 343, 350

Vowel identity 316

Consonant mutation :

Vowel identity 47, 247

Terminal abrasion . . .151, 363

Terminal abrasion 190, 266, 280, 285, 292, 294, 296, 309, 312, 318, 329,

342, 363

Frontal abrasion 151, 213, 216

Frontal accretion 216, 289, 344, 358

Terminal accretion . . . .213

82

THE POLYNESIAN WANDERINGS.

LAMBELL.

Polynesian 32
Quality 8 1

a a, e, 0, u

e e, a, o 00, u
i i, u u u, a, i, 0

1 /, r

ng ng, n n n, ng m m

h s

s s

V

f p, h

k k t /, k, n p 6, m6

a-a 74, 347, 213, 216, 292, 294, 1 >- 320, 358

309, 312, 316, 317, 318, 324, ag-ng 213, 264, 309, 336, 346, 350

346, 35o, 352 ng-n 74

a-e 151, 342 k-k 149, 196

a-o 321 n-n 147, 151, 292, 296, 317, 343

3l-u 266 n-ng 266, 342

Q-e 151, 190 h-s 47, 352

e-a 47 s-s 264, 342, 344, 346

e-o 247 U 47, 74, 196, 247, 266, 294,

i-t 47, 149, 196, 312, 319, 342, 318, 324, 343, 346, 350, 352,

346, 352 358

\-u 74 t-k 216

0-0 196, 264, 336 t-w 329

o-u 296 mm 216, 312, 316, 317, 318, 319,

u-u 247, 292, 329, 343, 344, 358 320, 324

u-a 316 f-/> 296

ui 317, 343 i-h 147, 213, 292, 294, 329

u-o 216, 320 p-b 190, 247

1-/ 149, 309, 312, 316, 336, 350 p-mb 151

Identical 196, 312, 319, 324, 336, 346, 358

Terminal abrasion. . .149, 264, 309, 318, 344
Consonant identity. ... 74, 247, 316, 317, 343, 350
Consonant mutation :

Vowel identity 47, 213, 292, 342, 352

Terminal abrasion. . .147, 151, 294

Terminal abrasion 147, 149, 264, 266, 294, 296, 318, 329

Frontal abrasion 151

Frontal accretion 190, 264, 318, 344, 358

Terminal accretion. . . .213, 336

POLYNESIAN RELICS IN MELANESIA. 83

LAUR.

Polynesian 28
Quality 75

a a, e, u

e e, a, i, 000, i, u
i i U u, i, o, h

1 I, r
ng ng, n n n, nq m m

h s
s J

V

f v, h

kg t t, n p b, mb

a-a 156, 216, 217, 292, 294, 309, 1 ; 257

313, 3i5, 3i6, 317, 318, 324, ng-no 309, 35o

35o, 352 ng-w 151

a-e 295 kg 156

a-w 266 n-w 151, 156, 292, 296, 317

e-e 151, 190 n-ng 266, 342

e-o 47 h ^ 47, 352, 363

e-* 257 s-* 343, 344

e-o 247 t-t 47, 216, 217, 292, 294, 309,

it 47, 295, 312, 313, 319, 350 313, 315, 316, 317, 318, 324,

0-0 363 350, 352

o-i 257 t-n 329

O-u 296 mm 156, 216, 217, 257, 312, 313,

u-w 156, 216, 292, 316, 329, 344 315, 316, 317, 318, 319, 324

u-i 247 f-v 363

no 343 f-h 292, 294, 296, 329

U-h 295 p-b 190, 247

1-/ 295, 309, 312, 313, 315, 316, p-mb 151

350

Identical 217, 313, 334, 344, 350

Terminal abrasion. . .309, 312, 317, 318, 319
Consonant identity. . . .257, 315, 316
Consonant mutation :

Vowel identity 47, 190, 292

Terminal abrasion. . .151, 156, 294, 352, 363

Terminal abrasion 151, 156, 266, 294, 296, 309, 312, 315, 317, 318,

3i9. 329, 343, 352, 363

Frontal abrasion 151

Frontal accretion 216, 352

Terminal accretion. . . .216, 257, 316

84

THE POLYNESIAN WANDERINGS.

a-a
i-i
u-u
\l

LEMARORO.

258, 274
274
258
2 74

LEON.

ng-na

k—

t-l

274
258
258

Polynesian 10
Quality 90

a a, e,

0

e e,

i 0

0

i i

1 /, r

U 0

ng

n n
h

s ^

m m

V w

f V

k g

t—

p

a-a 147, 214
a-e 193, 318, 337
a-o 292, 307
e-e 122
e-i 203
i-i 193

0-0

u-o

\-l

\r

k-g

n-w

187
292

307
203

i93

i47,

187, 292

s-s
t—

mm

\-w

i-v

203, 337

3i8

3i8

307

122, 147, 193,

214, 292

Identical

Terminal abrasion .
Consonant identity . .
Consonant mutation :

Vowel identity

Terminal abrasion .

. .214

187

• 203,

122,

• 147,

Terminal abrasion . . . 147, 187, 193,
203, 292, 318
318, 337 Frontal accretion 187

Terminal accretion. . . 122

214 Metathesis 139

193

LIFU.

Polynesian 7

Quality 42

a a, e

e i

0

0

i

i

1

n n
h

u u

ng

m m

s

k k

a-a
a-e
e-i
i-i

214, 361
318, 324
318
259. 33o

Identical

Terminal abrasion
Frontal abrasion . . .

0-0
u-u
k k
n-n

2 59,
324
214

t k

259

326, 330, 361

361

259, 33o

33o, 361

V

f

P P

t-k

318, 324

m-ra

318, 324, 326

v-P

361

POLYNESIAN RELICS IN MELANESIA.

85

LO.

Polynesian 37
Quality 84

a a, e, i, 0
e e, a, i 00
\ i, u u u, i

\ I, r, h
ng nq, n, g n n, ng m m, mw, ng

h j
s h

V v, p, w
f v, w, h
kg t t, nd p p, kw

a-a 152, 169, 193, 195, 214, 217, k-g 149, 169, 193, 195, 251, 300,

290, 300, 301, 312, 324, 328, 301, 305, 306, 357

339, 35o n-n 147, 169, 290, 317, 321, 328,

a-e 147, 169, 309, 313, 316, 318, 343

35i n-n^ 351

a-* 307 h-j 339

a-o 317 s-/i 169, 239, 251, 351

e-e I22 t-t 195, 306, 318, 324, 343, 356,

e-i 190, 193 357, 358

e-a 290 t-nd 350

l-» 149, 169, 290, 300, 305, 312, m-w 217, 312, 316, 317, 318, 324,

313, 35o 328

\-u 321 m-mw 313

0-0 285, 307, 336, 356, 357 m-ng 316

u-m 212, 306, 316, 343, 351, 358 y-v 152

U-i 239 y-p 291

1-/ 212, 305, 307, 309, 312, 313, \-W 307

3l6, 350 f-V 122, 147, 193, 214, 290

\-r 149, 152, 336 i-w 290

\-h 335 i-h 169

ng-w^ 309, 336, 350 p-p 190

ng-w 285 p-kw 285
ng-^ 169

Identical 214, 312, 324, 358

Terminal abrasion. . . 152, 212, 328, 336, 343, 356
Consonant identity- • • 152, 190, 217, 309, 316, 317, 318, 321
Consonant mutation :

Vowel identity 169, 193, 195, 214, 285, 300, 305, 313, 350, 351

Terminal abrasion. . . 149, 306, 357

Terminal abrasion 122, 147, 149, 152, 190, 195, 212, 306, 309, 317,

318, 321, 328, 339, 343, 356

Frontal abrasion 217, 239

Frontal accretion 290, 301, 312, 339, 351

Terminal accretion .... 305, 336

86 THE POLYNESIAN WANDERINGS.

MAEWO.

Polynesian 39
Quality 87

a a, e, 0
e e, i 00

i i, u u u, i

1 /, r, nd
ng ng, n n n, I m m

h s
s ^

v nib, —
f v, w, —
k k, g, w t t, nd p b, w, kmbw

a-a 147, 152, 193, 214, 215, 239, ng-w 285

278, 284, 292, 309, 312, 313, k-fe 149

316, 317, 318, 324. 328, 346, kg 193, 214, 301, 357

352 k-w 306

a-e 272, 291, 301, 337 n-rc 147, 292, 317, 328

a-o 335 n-l 259

e-e 153, 190, 318 h-s 206, 215, 278, 352

Q-i 152, 303 s-^ 203, 239, 298, 337, 346

i-t 149, 193, 215, 259, 291, 298, t-t 306, 318, 324, 346, 352, 357,

312, 313, 329, 346 358
\-u 206 t-nd 329

0-0 147, 153, 206, 259, 285, 309, mm 312, 313, 316, 317, 318, 324,

336, 357 328

u-w 212, 239, 272, 278, 284, 292, \-mb 291

298, 306, 316, 317, 328, 329, v— 152

335, 358 f-v H7, 193, 214, 215, 272, 292

u-i 301 f-w 206, 259

1-/ 152, 153, 212, 284, 309, 312, f— 329, 360

313, 316 p-& 190
\ r 149, 203, 336, 358 p-w 284
\-nd 335 p-few6w285
ng-ng 309, 336, 346

Identical 149, 153. 212, 239, 298, 309, 312, 313, 316, 317,

318, 324, 328, 329, 336, 346, 358
Consonant identity .... 203, 337
Consonant mutation :

Vowel identity 147, 190, 193, 214, 215, 259, 272, 278, 284, 285,

291, 292, 306
Terminal abrasion. . .352, 357

Frontal accretion 272, 298, 312

Terminal accretion. . . .212, 259, 298, 301,324, 335, 346, 358

MAI.

a-a 281 s-.? 281

u-n 281 x-b 281

\-r 281

POLYNESIAN RELICS IN MELANESIA. 87

MAKURA.

Polynesian

13

Quality

92

a a, 1

s

e

e

0

i

*',

u

I I, r

u it, i

ng n, g

n n
h h

s

m m

V

f v, -

k ngg, k

t /

P

a a

74,
335,

290, 302, 309,
352

312,

3i

7,

ng-«
ng-?

74
309

a-e

165

k-k

306

e-e

290,

297

k-ngg

297,

302

\-i

74,

165, 290, 297,

302

, 3:

12

n-n

290,

3i7, 321

\-u

321

\\-h

352

U-M

273,

32r, 335

t-t

74,

165, 302, 306, 352

u~i

306

m-iii

312,

3i7, 321

M

297,

309, 312, 335

f-v

290

\r

165

f—

273

Identical 312 Consonant mutation:

Terminal abrasion. 3 17, 352 Vowel identity. . . 74, 290, 297, 302

Consonant identity. .165, 306, 321 Terminal abrasion . 309
Vowel identity 273 Terminal abrasion. . .273, 306, 309, 317,

352
Frontal accretion. . . .321

MALANTA.

a-e 312, 324 1-/ 312

\-i 312 m-m 312, 324

MALEKULA.

Polynesian 68
Quality 51

a a, e, i, 0, it
e e, i, ou 00, i, u
i i, e, u u, u, e, i

I /, r
ng ng, n, g, m n, n, r m m

h s, j
s -y

v v, w, u, pw
f v, p, b, — , /, mbw
k k, kit, g, h, — t t, j, r p p, b, mb

88

THE POLYNESIAN WANDERINGS.

MALEKULA— Continued.

a-a 74, 147, 157, 161, 193, 199, k-k 194, 217, 297, 323

217, 239, 252, 302, 309, 312, k-kh 251

316, 324, 339, 340, 345, 350, kg 211, 301, 306

360 k-h 193, 258, 300, 302

a-e 74, 121, 156, 254, 258, 274, k— 156, 251, 301

292, 318, 323, 324, 346, 239 n-w 147, 156, 157, 217, 242, 254,

a-* 157, 208, 252, 294, 315, 317, 292, 317, 321, 328

352 ll-r 259

a-o 258, 320 h-s 47, 278, 339, 352, 363

aw 148, 328, 364 h-;' 340

e-e 47, 210 s-j- 239, 252, 323, 344

e-i 121, 297 t-t 47, 74, 148, 161, 199, 208,

e-oti 247 210, 211, 217, 247, 258, 270,

\-i 47, 74, 193, 194, 217, 242, 294, 306, 324, 345, 346, 347,

254, 259, 289, 297, 300, 302, 352, 356, 358, 359

312, 321, 35o t-j 208, 227, 289, 302, 318, 329

i-e 211,254,319 t-r 276,294,350

i-w 362 mm 156, 161, 254, 258, 312, 315,

0-0 252, 259, 287, 356, 363 316, 317, 318, 319, 320, 323,

o-i 252 324, 328, 340

O-U 285, 362 Y-pW 242

u-w 77, 156, 194, 228, 239, 246, v-w 242

270, 278, 289, 292, 306, 316, v-w 291

344, 358, 359, 360, 364 \-v 121

u-t 211, 247, 329, 359 f-/ 287, 292

u-e 239 f-v 147, 161, 193, 210, 294, 329

1-/ 252, 316 i-p 246

\r 121, 194, 228, 252, 287, 289, i-b 148, 259, 363

297, 309, 312, 315, 320, 335, f-mbw 157

339, 35o, 362, 364 f— 360

ng-ng 246, 285, 346, 350 p-p 121

ng-n 74, 274 p-b 247, 289

n&-9 3°9 p-mb 285
ng-m 199

Identical 74, 77, 194, 239, 312, 316, 324, 358, 345

Terminal abrasion . . . 344, 356
Consonant identity. . . . 156, 194, 239, 252, 254, 270, 317, 323, 328, 346, 359

Vowel identity 210, 285, 335

Consonant mutation :

Vowel identity 47, 242, 246, 289

Terminal abrasion. . 147, 306, 363, 292, 300, 309

Terminal abrasion 121, 147, 156, 161, 193, 199, 208, 217, 228, 242,

246, 258, 274, 278, 292, 294, 300, 306, 315, 317,
318, 320, 321, 323, 328, 329, 340, 344, 346, 352,
356, 363

Frontal abrasion 251, 350

Frontal accretion 72, 148, 239, 252, 274, 285, 301, 321, 360

Terminal accretion 77, 148, 239, 321, 325, 335, 347, 360

POLYNESIAN RELICS IN MELANESIA.

89

MALO.

Polynesian 6i
Quality 65

a a, \

l, 0

e e,

a, i, 0

0 0,

*

i

i, e,

U

u, a

, e,

i, c

t

I /, r

ng ng, n,

mo

n n

m m,

t

h s

s .y

V

f

u

b, u

k k, g, v, -

t *, d

, s

P

b,

v, bu

a-a

121,

124, 147, 148,

193,

199,

\r

121,

194,

203,

305,

308,

335,

203,

207, 217, 218,

228,

239,

336,

354,

358,

359

252,

254, 258, 266,

270,

290,

ng-ng

285,

309,

336,

346

292,

293, 300, 301,

302,

308,

ng-n

308

309,

312, 313, 316,

318,

320,

ng-wo

199

324,

335, 337, 338,

339,

340,

k-ife

258,

297,

3or,

302,

338,

353

346,

347, 352, 356

k-g

193

a-i

208

k-v

194

a-o

121,

203

k—

193,

305,

306

e-e

124,

190. 297, 311,

318,

338,

n-n

147,

266,

290,

292,

321

354

hs

47, 206

e-a

47,

121

s-s

203,

239,

252,

337,

338,

339,

e-i

208,

290, 347

340,

344,

346,

352

e-o

247

t-t

47,

148,

199,

217,

247,

258,

\-i

47,

193, 194, 206,

254,

266,

266,

270,

302,

306,

3i8,

324,

270,

290, 305, 312,

3X3,

321,

346,

347,

352,

356,

358,

359

346

t-d

208

i-e

297,

302, 308, 319

t-s

207,

293

\-u

305,

362

m-m

124,

217,

258,

266,

3*2,

313,

0-0

H7,
336,

206, 252, 285,
353, 356, 362

309,

3ii,

m-t

316,
254

318,

319,

320,

324,

340

O-i

353

v-«

121

U-U

148,
289,
343,

194, 228, 239,
292, 306, 316,
344, 358, 359

258,
321,

270,
335,

f-v

l-b

H7,
293
I48

193,

208,

2 54,

290,

292,

u-a

320

i-u

206

u-e

359

p-b

121,

190,

207,

218,

247,

285,

u-i

247,

301

289

U-o

266

p-bu

I24

U

228,
312,

252, 289, 297,
313, 316, 320,

309,
339,

3".
362

p-v

I90

217, 228, 2

39,25

2,258,

Consonant mutation:

270,302,306,321,337,

309, 3", 312, 313, 316,
318,324,336,343,344,
346, 347, 356

Consonant identity. 203, 266, 302, 320, 353,
358, 359

Vowel identity .... 190

Vowel identity. . 147, 193, 199, 292, 218,
„, . 254,335,354

Terminal abr . . .285

Terminal abrasion. 207, 352

Frontal abrasion. .193, 306

Frontal accretion. . 148, 203, 208, 293, 340

Terminal accretion. 194, 337, 316, 319, 335,
337

90 THE POLYNESIAN WANDERINGS.

MARINA

Polynesian 31
Quality 89

a a, e
e e, a 00

\ i u u

1 I, r

ng ng n n, I, — m m, n

h s

s

v p, t

f V

kg t t, s p p, v

a-a 147, 193, 214, 250, 278, 284, k-g 193, 214, 250, 301, 306, 357

292, 301, 309, 312, 317, 318, n-« 147, 292, 317, 328, 343

324,328,335,350,352,356 n-/ 321

a-e 291 n — 321

e-e 311, 318 h-s 278, 352

e"a 272 t-^ 306, 318, 324, 343, 352, 356,

l-t 193, 285, 291, 312, 313, 350, 357, 358

352 t-s 350

0-0 285,309,311,336,356,357 \-t 318,321,328

u-n 212, 272, 278, 284, 292, 301, m-ra 312, 313, 317, 324

306, 317, 321, 328, 335, 343, mn 291

358 \-p 291

\l 212, 284, 309, 311, 312, 313, i-v \\-j, 193, 212, 214, 272, 292,

35o 360

1-r 335, 336, 358 V-P 285

ng-ng 285, 309, 336, 350 p-v 250, 284

Identical 285, 309, 311, 31S, 328, 335, 336, 343, 356, 358

Consonant identity. . . .272
Consonant mutation :

Vowel identity 193, 214, 278, 284, 292, 301, 306, 312, 313, 317,

324, 35o, 352

Terminal abrasion. . . 147, 212, 357

Terminal abrasion 147, 212, 306, 357

Terminal accretion 328, 336, 358

POLYNESIAN RELICS IN MELANESIA.

91

MATUPIT.

Polynesian
Quality

i i
63

a a, e

1 x

ng ng

1 r
n n
h

s

O o, a, u
u u

k— t *

a-a 147, 214, 287, 320, 346, 352 ng-ng

a-e 300 k —

\-i 300, 346 n-n

0-0 244 t-t

o-a 147, 287 m-w

o-u 285 f-w

u-zt 287, 320, 358 i-p

\-r 244, 287, 320 p-b

Identical 320, 348

Vowel identity 300

Consonant mutation:

Vowel identity 147, 214

Terminal abrasion . . . 244, 285

Terminal abrasion 147, 244, 285, 352

Frontal abrasion 74

Terminal accretion .... 300, 320

m m

V

f

w,

P

P

b

284,

346

300

147

74,

346,

352,

358

320

147,

214

287

244,

285

MELE.

a-a

0-0

147
H7

n n

f-/

H7
i47

92

THE POLYNESIAN WANDERINGS.

MERLAV.

Polynesian 40
Quality 77

a a, e, o
e e, a 00

i i, u u u, a, i

1 I, r, t, nd, ng
ng ng n n, ng m m, mw

h s
S s

V T/, w6
f V, w
k fe, g, w t t, m p b, mb, v

a-a 123, 147, 193, 197, 207, 214, ng-ng 154, 197, 309

217, 239, 268, 307, 309, 313, k-fe 301

316, 317, 318, 322, 324, 328, kg 149, 151, i93. 251, 300, 338,

337, 338, 339 357

a-e 291, 301, 324 kit) 306

a-o 335 nn T47, 3i7. 328

e-e 151, 190, 210, 318, 338 n-w^ 342

e-a 122 h -^ 206, 338, 339

\-i 149, 154, 193, 291, 300, 312, s-s 239, 251, 337, 342

313, 342 t-t 207, 210, 268, 289, 306, 318,

i-u 206 c 324, 357, 358

0-0 147, 206, 357 t-ra 217

n-u 207, 239, 284, 289, 306, 316, m-m 217, 313, 316, 317, 318, 322,

328, 358 324, 328

u-a 151 m-mw 312

u* 301, 335 v"^ 307

1-/ 123, 154, 197, 284, 289, 307, \-mb 291

309, 312, 316, 313, 329 f-v 122, 147, 193, 210, 214

1-; 149, 268, 322, 358 i-w 206, 360

U 335 p-6 190

\-nd 335 p-mb 151, 207, 289

\-ng 123 p-T> 284

Identical 239, 313, 316, 318, 324, 337, 358

Terminal abrasion. . .154, 197, 307, 309, 312, 317, 328

Consonant identity. . . .268, 301

Consonant mutation:

Vowel identity 147, 193, 207, 214, 289, 338

Terminal abrasion. . . 149, 190, 210, 284, 300, 306, 339, 342, 357

Terminal abrasion 123, 149, 151, 154, 190, 193, 197, 210, 217, 268,

284, 300, 306, 307, 312, 317, 328, 339, 342

Frontal accretion 149, 197, 301, 339

Terminal accretion. . . .122, 149, 312, 324, 337

POLYNESIAN RELICS IN MELANESIA. 93

MOAN US.

Polynesian 23
Quality 83

a a, e, i

e i 00

i i, u u u

1 I

ng ng, n n n, nj m m

h s

s -y

v u

f p, mb, t, —
k k, — t t, r, ndr p p, mb, v

a-a 74- 126, 239, 291, 294, 308, tin 290, 317, 330

312, 317, 324, 328, 335, 345, n-nj 328

35o, 352, 360 h-s 352

2L-e 215, 290 s-j" 239

a-* 350 U 74, 126, 294, 305, 324, 345

e-i 190 t-r 329

i-i 215, 290, 291, 305, 312, 330, t-ndr 350, 352

350 mm 312, 317, 324, 328

\-u 305 \-u 291

0-0 126, 285 i-p 290, 294

vl-u 126, 239, 284, 317, 329, 330, i-mb 329, 360

335, 360 U 215

ng-w^r 285, 308, 350 f— 290

ng-n 74 p-p 190, 285

k-k 305 p-w& 284

k— 143 p-v 285, 290
1-/ 126, 284, 305, 308, 312, 335,

35o

Identical 126, 239, 285, 312, 317, 324, 330, 335, 345

Terminal abrasion ... 74, 308
Consonant identity. ... 190
Consonant mutation :

Vowel identity 291, 350, 352, 360

Terminal abrasion. . .284, 294, 352

Terminal abrasion 74, 284, 285, 290, 294, 308, 328, 352

Frontal accretion 190, 239

Terminal accretion. . . .305, 317, 335

94

THE POLYNESIAN WANDERINGS.

MOSIN.

Polynesian 27
Quality 62

a a, e, 1, 0, u
e e o o, e

i i, e u u, 0

1 /, r, n, ng
ng ng n n m w, mw

h j
s j-

V w
i v, w
k 9, ng, w t t, m p w

a-a 123, 147, 214, 217, 239, 309, \-n 335

317, 318, 328, 339, 357 ng-ng 154, 309, 336

a-e 193, 324 kg 193, 251, 300, 357

a-i 307 k-w# 301

a-o 316 k-w 306

a-M 272 n-w 147, 317, 328

e-e 122 h-j 339

\-i 193, 312 s-j- 239, 251

ie 154, 300 t-< 306, 318, 324, 343, 356, 357

0-0 307, 336, 356, 357 t-m 217

o-e 123 mm 217, 316, 317, 318, 324, 328

u-u 272, 306, 316, 343 m-mw 312

u-o 284 \-w 307

i-l 123, 154, 284, 307, 309, 312, f-V 122, I47, I93, 214

316, 339 i-w 272

1-;- 336 p-w 284

l-ng 123

Identical :

Terminal abrasion. .239, 309, 312, 317, 318, 328, 336, 356
Consonant identity. . . 154, 316, 324
Consonant mutation :

Vowel identity 214, 339

Terminal abrasion. . . 147, 193, 306, 343

Terminal abrasion 147, 193, 214, 217, 239, 284, 300, 306, 309, 312,

317, 318, 328, 336, 339, 343, 356

Frontal accretion 272, 339

Terminal accretion 122, 154, 272, 301, 336, 339

POLYNESIAN REUCS IN MELANESIA.

95

MOTLAV.

Polynesian 42
Quality 50

a a, e, i, 0, u
e e, a o o, e

\ i, e u u, i, o

1 /, r, ng
ng ng, n n n, no, I m m, mw

h h
s /«

v nib, w
i v, w
k g, ng t t, nd, m p b, w, kmbw

a-a 123, 147, 169, 214, 217, 239. k-<7 l69, 193. 25i, 258, 300,306

318, 324, 339 k-ng 301

a-e 258, 309, 317, 322, 328, 337, n-H 147, 169, 290, 292, 317, 321,

346, 350 328, 351

a-i 307 n-ng 151

a-o 290, 307 n-/ 259

a-w 316 h-fc 339

e-e 151, 190 s-h 169, 239, 251, 337, 35i

e-a 122 t-t 258, 306, 318, 324, 346, 356

\-i 154, 169, 290, 321 t-m 217

i-e 193, 259, 300, 312 t-nd 350

0-0 147, 259, 285, 336, 356 mm 217, 258, 316, 317, 318, 322,

o-e 123 324> 328

U-ti 316 m-mw 312

U-i 306, 351 \-mb 291

u-o 239, 284, 292 v-w 307

1-/ 123, 154, 284, 307, 309, 312, f-V 122, 147, 193, 214, 290, 292

316, 339, 35o f-w 259, 360

\-r 322, 335, 336 p-6 190 >^?C5lCj>S.

l-ng 123 p-w 284 /8^^— -OX"

ng-wa 154.309,336,346,350 p-kmbw 151, 285 ZJ /^'^Vf

"g- ^ 285 /Q3 A -^•■^ V^ 1

Identical 214, 324

Terminal abrasion. .154, 318, 321, 336, 339, 356 -**♦*< /

Consonant identity. . . .316, 317, 328, 346 S^, \4?A8%>/'v

Consonant mutation : \& w^

Vowel identity 122, 147, 169, 259, 285

Terminal abrasion. . . 190, 217, 258, 272

Terminal abrasion 151, 154, 190, 217, 258, 272, 300, 306, 309, 312,

317, 318, 328, 339, 346, 351

Frontal abrasion 151, 351

Frontal accretion 147, 193, 290, 301, 312, 339

Terminal accretion 122, 147, 336, 350

96

THE POLYNESIAN WANDERINGS.

MOTA.

a-a

a-e

a-i

a-u

e-e

e-a

e-i

e-oi

\-i

i-a
i-e
i-o
\-u
o-o

o-a
o-i

o-ai

Polynesian
Quality

123

92

a a, e, 1, u
e e, a, i, oi o 0, a, i, ai
1 i, a, e, o, u u u, a, e, i, o, v

1 /, r, ng, n
ng ng, g, n, m, n n, ng, I mm, mw

h s
s s

V V, p, w, u
f V, w, t
k g, ng, w t t, s, r p p, v, kpw

75, 123, 147, 152, 156, 160,
169, 189, 192, 193, 197, 198,
199, 203, 207, 209, 213, 214,
215, 216, 217, 228, 230, 239,
240, 250, 252, 253, 254, 258,
262, 263, 267, 268, 270, 274,
276, 279, 283, 284, 287, 290,
292, 294, 295, 300, 301, 302,
307, 309, 3io, 312, 313, 316,
317, 3*8, 320, 322, 323, 324,
328, 335, 337, 338, 339, 34Q,
342, 344, 346, 347, 348, 351,
352, 356, 357. 360, 364
291
167
167

151, 152, 153, 189, 190, 257,
276, 290, 318, 338, 347, 363

47, 122, 240, 272
203, 297
247

47, i54» l69, 193, 206, 215,
223, 227, 241, 243, 254, 259,
270, 290, 291, 295, 300, 312,
313, 342, 346, 348, 355
34i
3i9
34i
362

75, 147, 153, 160, 192, 196,
197, 206, 248, 252, 259, 263,
267, 285, 309, 325, 336, 353,
355, 356, 357, 362, 363
205, 248, 257, 287, 331

353
123

u-u 167, 207, 212, 216, 228, 239,
240, 247, 258, 262, 270, 283,
284, 286, 287, 289, 292, 294,
298, 306, 316, 317, 320, 325,
326, 328, 335, 343, 351, 358,
368

u-a 151

u-e 30 1

u-*' 156, 359
u-o 360
u-v 295

[-1 75, 123, 153, 154, 161, 189,

197, 203, 212, 228, 252, 274,
284, 289, 295, 297, 307, 309,
310, 312, 313, 316, 339, 348,
362

1-r 152, 160, 192, 205, 241, 243,
248, 257, 262, 310, 322, 331,
336, 34i, 355, 358, 359, 364

1-w 335

l-ng 123

ng-ng 167, 197, 209, 230, 248, 274,
298, 309, 346

ng-9 154, 336

ng-m 199

ng-n 285, 343

k-g 151, 156, 169, 193, 196, 214,
227, 250, 251, 297, 300, 302,

323> 353, 357
k-ng 301
k-w 306
n-w 147, 156, 169, 227, 254, 286,

290, 292, 317, 321, 328, 351
n-ng 342
n-/ 259

POLYNESIAN RELICS IN MELANESIA. 97

MOT A— Continued .

h-s 47, 206, 215, 331, 339, 340, m-mw 312, 313

363, 352 \-v 307

s-j 169, 198, 203, 205, 239, 251, v-p 291

252, 263, 323, 338, 341, 342, x-w 189, 206, 310

344, 35i, 205 v-M 310

t-t 47, 75, 160, 196, 199, 208, f-v 122, 147, 152, 193, 213, 214,

209, 217, 227, 247, 258, 267, 215, 223, 243, 254, 272, 283,

268, 270, 286, 289, 294, 302, 290, 292, 294

306, 318, 324, 325, 343, 346, i-w 360, 363, 259, 286, 287

347, 348, 353, 355, 356, 357, t-t 290

358, 359, 207 V-P l5*, J90, 192, 207, 247, 279,
t-s 298 ' 289

t-r 276 p-v 250, 284

m-*» 75, 156, 198, 217, 230, 253, p-kpw 241, 285

254, 257, 258, 316, 317, 318,

319, 322, 323, 324, 325, 326,

328, 340

Identical 152, i53> i54» l6°> l67, 190, 192, 197. J98, 207,

209, 212, 216, 217, 228, 230, 239, 240, 252, 253,
258, 262, 263, 267, 268, 270, 279, 289, 307, 309,
316, 317, 318, 320, 324, 325, 328, 336, 337, 344,
346, 347, 348, 35i, 355, 356, 358, 364
Terminal abrasion. . . 75, 163, 203, 274, 327, 343

Consonant identity 205, 247, 248, 257, 272, 319, 321, 331, 341, 362

Vowel identity 259, 276, 298

Consonant mutation :

Vowel identity 47, H7, 156, 169, 189, 193, 199, 206, 208, 213,

214, 215, 227, 241, 243, 250, 283, 284, 285, 286,
287, 290, 292, 294, 295, 300, 306, 312, 313, 336,
338, 339. 340, 342, 363
Terminal abrasion. . .154, 196, 223, 254, 294, 302, 306, 310, 323, 343,

352, 358
Terminal abrasion 75, 151, 154, 196, 203, 223, 241, 243, 254, 272,

294, 297, 321, 323, 343, 346, 352, 358, 359

Frontal accretion 189, 190, 279, 301, 310, 312, 331

Terminal accretion 189, 190, 279, 301, 310, 312, 331

f

98 THE POLYNESIAN WANDERINGS.

MOTU.

Polynesian 56

Quality 85

a a
e e, a, i, 000, u, ia
i i, e, o, u, ei u u, a, o

1 I, r, -
ng «, — n n m m

h d
s h, d

v v
f x>, /&
k k, g. — , m t t p p, b

a-a 125, 208, 214, 217, 268, 271, 1-/ 243, 268, 309, 310, 335, 339,

292, 293, 301, 309, 310, 312, 354, 362

3*3, 317, 3i8, 328, 334, 335, \-r 182,205,241,248,272,288,304,

339, 342, 346, 349, 35o, 35i, 327,33i, 332,334,359,364

352, 360, 364 1— 312, 313, 350

e-e 190, 208, 318, 354, 363 ng— 285, 309, 332, 346, 350

e-a 47, 272 ng-w 125

e-i 190 k -k 249

e-o 249 kg 299

\-i 47, 241, 243, 249, 268, 304, km 304

312, 313, 321, 327, 330, 342, k— 301, 306

346, 35o, 352 n-w 271, 292, 317, 321, 328, 330,

le 273 334, 342, 343, 351

«-° 288 h-d 47, 339, 352, 363

i-u 362 s-h 182, 205

i-ei 319 s-d 298, 342, 351

0-0 201, 285, 309, 331, 349, 362, t-t 47, 125, 168, 208, 217, 249,

363 268, 293, 306, 318, 343, 346,

0 u 205 349. 352, 358, 359

o-ta 331 m-m 76, 217, 312, 313, 317, 318,

u-w 76, 125, 168, 182, 271, 272, 319, 326, 327, 328

273, 288, 292, 298, 299, 301, \-v 310

306, 317, 321, 326, 327, 330, i-h 201, 208, 243, 271, 272, 273,

332, 334, 335, 343, 35 1, 358, 288, 292, 360, 363

359, 360, 364 f-v 214, 293

U-a 76 p-b 125, 190, 241, 285

u-o 248, 328 p-p 279

Identical 168, 214, 217, 268, 271, 279, 301, 306, 309, 317,

318, 321, 326, 327, 330, 334, 335, 343, 346, 349,

354, 359, 364
Consonant identity. ... 76, 249, 328, 331, 362

Vowel identity 304, 346

Consonant mutation :

Vowel identity 47, 125, 182, 201, 208, 241, 243, 272, 273, 285,

292, 293, 298, 299, 339, 342, 352, 360, 363

Frontal abrasion 248, 301, 306, 310, 312, 313, 332

Frontal accretion 190, 208, 241, 243, 339, 351

Terminal accretion 208, 299, 321, 343, 359

POLYNESIAN RELICS IN MELANESIA.

99

MURRAY ISLAND.

a-a

au-ai

u-u

275
301

305

\-r

k-k
kg

NATALAVA.

305
275
301,

305

a-a

SL-U

l-i

u-u

335,
35o
35o
335

350 W
\-r

ng-nd
t-k

NENGONE.

350
335
35o
35o

Polynesian i i
Quality 45

a a, e
e e 00

i i, e u u, e, 0

I I, r, n
ng ng n n

h

s s

V

f

w
w,

k k, -

t t

p

p

a-a 214, 274. 361

\-n

309

a.-e 300, 309, 322

ng-ng

274,

309

e-e 122

k-k

361

\-i 300

k—

300,

306

i-e 245

n-n

33o

0-0 309

s-s

245

u-w 330, 361

U

306

u-e 306

\-w

291

u-o 245. 306

i-w

245

U 274

f—

122,

214

l-r 335

P-P

361

Identical

.361

Consonant identity . .

• • 306,

335

Consonant mutation :

Vowel identity. . . .

• • 309,

330

Terminal abrasion .

• 274

Frontal abrasion ....

122,

214,

306

Frontal accretion ...

• 274

100

THE POLYNESIAN WANDERINGS.

NEW GEORGIA.

Polynesian 16
Quality 63

a a, o, u
e e 00

i i, e, o u u, o

I I, r
ng ng n m m m

s s

v
f

k k, h t t, s, m p p, b

a-a 217, 300, 312, 313, 324, 350, \-r 192, 305

351, 361 n%-n9 285, 298, 350

a-o 361 k-fe 305, 361

aw 192 k-h 300, 301

e-e 190 n-m 351

i-» 149, 285, 300, 312, 313, 350, s-s 351

351 t-t 324, 35o

i-e 305 t-j 298

i-o 305 t-m 217

0-0 192, 285 mm 217, 312, 313, 324

u-u 298, 351 p-/> 190, 192, 361

vl-o 361 P-6 285

1-/ 312, 313, 350

Identical 190, 312, 313, 324

Consonant identity. . . . 192, 305, 350
Consonant mutation :

Vowel identity 217, 285, 300

Frontal accretion 351

Terminal accretion 149, 190, 300, 351

POLYNESIAN RELICS IN MELANESIA. 101

NGGAO.

Polynesian 18
Quality 77

a a

& a o 0, i

i i, e, u u u

\ I, r, k

ng ng n n, m m m

h h

S s, h

V

f /, ng, kr
k k, g t t p mb

a-a 214, 215, 217, 239, 309, 312, k-k 251

313, 324, 346, 352 kg 305

e-a 272 n-w 201

i* 215, 285, 305, 313, 346, 351, n-ra 351

352 h-h 215

i-e 298, 313 s-J 251, 351

i-w 305 s-h 239, 298

0-0 201, 285, 309, 336 t-t 324

o-i 336 m-ro 217, 312, 313, 324

u-w 239, 272, 298, 351 i-f 214, 272

l-l 305, 309, 312 i-ng 215

1-r 272 i-kr 201

1-& 313 p-w6 285

ng-ng 285, 309

Identical 214, 251, 312, 324, 346, 351

Consonant identity. . . .272, 309, 336

Vowel identity 201, 239, 309

Consonant mutation:

Vowel identity 215, 285

Frontal abrasion 217

Frontal accretion 239, 272, 298, 309

Terminal accretion .... 324

102 THE POLYNESIAN WANDERINGS.

NGGELA.

Polynesian 6i
Quality 80

a a, at, o, u
e e, i oo

\ i u u

1 /, r
ng ng, g n n, ng, I m m, v, r

h h
s J, /i

V V

f v, w6

k k, g t t, nd, k p 6, roo, -y

a-a 45, H7, W 197, l9%, 203, k-fe 149, 251

207, 208, 213, 214, 217, 239, kg 258, 297, 300, 301, 305, 306,

258, 272, 278, 284, 300, 309, 323, 338

312, 313, 3i5, 316, 317, 318, n-M 290, 292, 296, 317, 321, 328,

322, 323, 324, 328, 332, 338, 33o, 342

339, 342, 346, 352, 360 n-ng 351

a-ai 290 n -I 259

a-o 45 h-A 278, 323, 338, 339, 352

a-u 350 ** 45, 203, 351

e-e 122, 208, 290, 297, 318, 338 s-h 198, 239, 251, 298, 342, 346

e-i 152, 203 t-t 207, 208, 217, 258, 294, 306,

it 149. i55, 259, 273, 285, 290, 318, 324, 346, 352, 358

297, 298, 300, 305, 312, 313, t-nd 329

321, 323, 327, 329, 33o, 332, t-k 350

342, 346, 350, 352 m-w 198, 217, 258, 312, 313, 315,

0-0 197, 259, 285, 296, 309, 357 3i6, 317, 318, 322, 324, 327,

u-w 155, 207, 239, 258, 272, 273, 328

278, 284, 296, 298, 306, 316, m-v 323

317, 321, 327, 328, 329, 33o, m-r 322

332, 35i, 358, 360 y-v 152

1-/ 152, 155, 197, 284, 297, 305, f-V 122, 147, 208, 213, 214, 259,

309, 312, 313, 315, 316, 322, 272, 273, 290, 294, 296, 329,

339, 35° 36o

\-r 203, 272, 327, 332, 358 f-mb 292

ng-ng 213, 285, 309, 332, 346 p-6 45, 190

ng-o 197 P-™b 207, 285

p-v 284

Identical 152, 155, 214, 217, 309, 312, 313, 3*7, 3*8, 321,

324, 327, 328, 330, 332, 335, 339, 346, 352

Consonant identity. . . .203, 272, 315

Vowel identity 259, 273, 316

Consonant mutation :

Vowel identity 122, 197, 198, 207, 208, 213, 239, 251, 258, 259,

273, 278, 284, 285, 290, 294, 296, 297, 298, 300,
306, 329, 338, 342, 351, 360

Terminal abrasion 147, 350

Frontal abrasion 332

Frontal accretion 190, 197, 273, 305

Terminal accretion 339

POLYNESIAN RELICS IN MELANESIA. 103

NGUNA.

Polynesian 16
Quality 93

a a

e e o 0, u

i i, a, u u u

I /, r
ng ng, n n n m m

h j
s

V

f /, V, w

k k t t p

a-a 74, 75, 214, 261, 294, 301, ng-ng 336, 346, 350

3i7, 318, 338, 346, 35o ng-w 74

e-e 261, 318, 338 k-k 214, 301, 338

\-i 206, 346, 350 n-n 317, 321

\-a 321 h-.? 206, 338

i-w 74 t-/ 74, 75, 294, 308, 346, 350

0-0 206, 336 mm 75, 317, 318

O-w 75 f-/ 294

u-w 159, 294, 301, 317, 321 i-v 214

l-l 75, 35o f-w 206

\-r 159, 261, 336

Identical 159, 261, 294, 301, 317, 318, 336, 346, 350

Consonant identity. ... 74, 75, 321
Consonant mutation :

Vowel identity 206, 214, 338

Frontal accretion 301, 321

104 THE POLYNESIAN WANDERINGS.

NIFILOLE.

Polynesian 13
Quality 66

a a, e, o, u
e 00

\ i, a u u, i, o

1 /, w
ne n n m m

h
s s

v

f i/, p,^w, n
k k t t P

a-a 214, 217, 239, 284, 360 1-w 3J3

a-o 291, 357 k-fe T49

a-e 313 nn 328, 330

a-w 217 s-j 239

\-i 149, 291, 313 t-t 217

i-a 149 m-m 217, 313, 328

0-0 357 ^ 294

u-w 212, 330, 360 i-p 284

u-i 239 f-w 214, 291

u-o 284 f-n 360

1-/ 149, 212, 284

Identical 328

Consonant identity. ... 149, 212, 217, 239, 330

Vowel identity 360

Consonant mutation :

Vowel identity 214

Terminal abrasion 294

Frontal abrasion 212

Frontal accretion 239

Terminal accretion 217, 284

POLYNESIAN RELICS IN MELANESIA. 105

NORBARBAR.

Polynesian 37
Ouauty 54

a a, e, i, 0, u
e e 00, a, e, i

\ i, e U u, i, o

1 /, r, nd

ng ng, n n n m m

h h

S s, h

v p, w

f V, w
kg, — , w t t, nd, m p />, b, mb, kmbw

a-a 147, 217, 239, 309, 317, 318, kg 193, 251, 301, 338

338 k— 300

a-e 193, 300, 312, 328, 337, 3.50 kw 306

a-i 350 nn 147, 292, 317, 328, 350

a-o 75, 292, 307, 316 h-h 331, 338

a-M 307 s-j- 239, 251, 337

e-e 122, 190 • s-h 239

\i 300, 312 t-^ 75, 306, 318, 324, 356, 358

i-e 193 t-wrf 350

0-0 75, 285, 336, 356, 357 t-m 217

o-a 147 mm 75, 217, 312, 316, 317, 318,

o-e 33i 324, 328

O-i 33i v-/> 291

u-w 212, 239, 289, 306, 316 \-w 307

u-o 272 i-v 122, 147, 193, 214, 292

VL-i 292, 358 f-w 212, 360

I-/ 75, 289, 307, 309, 312, 316 p-p 190

1-r 33i, 336 p-6 190

\-nd 335 p-m6 289

n%-ng 309, 336, 350 p-^wftw285

ng-w 285

Identical 324

Terminal abrasion. . .309, 317, 318, 336, 356
Consonant identity. ... 75, 316, 328, 331, 337, 358

Vowel identity 217

Consonant mutation :

Vowel identity 122, 212

Terminal abrasion. . . 147, 190, 289, 306, 338

Terminal abrasion 147, 190, 193, 214, 217, 239, 289, 306, 309, 317,

318, 328, 336, 338, 356, 357

Frontal abrasion 239

Frontal accretion 301, 312

Terminal accretion 122, 212, 307, 331, 336, 350

106 THE POLYNESIAN WANDERINGS.

OMB A.

Polynesian 40
Quality 75

a a, e, u
e e 00

\ i u u, e, v

1 I, r

ng ng, n n n mm

h h

s h

V V, w
f V, IV

k g, w t t, m p b, mb

a-a 74, 147, 152, 169, 207, 214, ng-ng 285, 309, 336, 346

217, 239, 254, 290, 291, 292, ng-n 74

307, 309, 3i7, 318, 324, 338, kg 169, 214, 300, 301, 338

346, 351, 356 k-w 306

a-e 278, 284, 292, 295, 300, 312, n-n 74, 147, 169, 254, 290, 292,

3i3 3i7

a-u 307 h-h 169, 206, 278, 338, 352

e-e i22, 152, 153, 190, 290, 318, s-h 239

338 t-t 74, 207, 306, 318, 346, 352,

\-i 74, 169, 206, 254, 285, 290, 356, 358

291* 295, 3oo, 312, 313, 346, t-vi 217

352 m-m 217, 254, 312, 313, 317, 318

0-0 153, 206, 285, 336, 356 x-v 152

u-w 212, 239, 278, 284, 292, 306, v-w 291, 307

317, 358 f-V 122, I47, 206, 212, 214, 254,

u-e 207 284, 290, 292

u-v 295 i-w 360

1-/ 153, 212, 284, 295, 307, 309, p-6 190

312, 313 p-mb 207, 285
\r 152, 335, 336

Identical 74, 152, 153, 309, 317, 318, 335, 336, 346, 352,

356, 358
Terminal abrasion . . . 324
Consonant identity. . . .278, 312, 313

Vowel identity 217

Consonant mutation:

Vowel identity 169, 190, 206, 214, 239, 254, 285, 290, 291, 306,

338
Terminal abrasion . . .147

Terminal abrasion 147, 324, 336

Terminal accretion 212, 313, 335, 336

POLYNESIAN REUCS IN MELANESIA. 107

PAAMA.

a-a 252, 274 u-i 329 s-j- 252

a-e 312 \l 274, 312 t-s 329

\-i 312, 329 1— 252 i-h 329

O-i 252 ng-ng 274

PAK.

Polynesian 31

Quality 45

a a, e, 0, u
e e, i, u 0 o, e
i i, e u u, o

I /, /•, ng, t
ng ng, n n n m m

h j
s J

V w

f 7,

k g, ng, w t t, — , ra p />

a-a 123, 147, 217, 250, 309, 318, U 335

339. 357 ng-ng 309

a-e 214, 317, 324, 328, 337, 350 ng-n 350

a-o 239, 292, 307 kg 251, 300, 301, 357

3L-U 250, 316 k-Mtf 30I

e-e 122, 190 k-w 306

e-t 203 n-w 292, 317, 328

e-w 272 h -s 339

i* 300 s-j 203, 239, 251, 337

»« 312 t-/ 350, 356

O-o 356, 357 t— 270, 318, 324, 357

o-e 123 t-m 217

u-u 270, 272, 306, 316 mm 217, 312, 316, 317, 318, 324,

u-o 292 328

1-/ 123, 307, 309, 312, 316, 339, \-w 307

35o f-v 122, 147, 214, 292

I > 203 p-p 190

l-ng 123

Identical

Terminal abrasion. . .190, 309
Consonant identity. . . .203, 239, 312, 316, 317, 328, 337

Vowel identity 324

Consonant mutation:

Vowel identity 122, 214

Terminal abrasion . . . 300, 339

Terminal abrasion 147, 190, 203, 214, 217, 239, 292, 300, 306, 309,

312, 317, 318, 328, 339, 356

Frontal abrasion 270

Frontal accretion 301, 312, 339

Terminal accretion. ... 122, 350, 324, 335, 350

108 THE POLYNESIAN WANDERINGS.

PALA.

Polynesian 54
Quality 80

a a, e
e e, a, o 00, a, e, u
i i, e, u u it, i, h

1 /, r, n
ng ng, n n n, ng m m

h s
s s

v b, u
f h, 11, s
k k, g, — t /, d p 6, p

a-a 74, 147, 156, 214, 2 1 6, 217, ng-n 151

253,290,294,307,312,313, k-k 196,211

317, 318, 324, 350 k-g

&-e 215, 295, 352 k— 151, 156

e-e 47, 151, 203 n-n 147, 156, 201, 290, 317, 343

e-a 47, 122, 363 n-ng 259

e-o 247 h-s 47, 215, 352, 363

\-i 215, 259, 290, 295, 312, 313, s-.y 203, 344

35o t-t 47, 74,125,196,211,216,

i-e 211 217, 247, 289, 294, 318, 324,

»« 329 343. 35o, 352, 358

0-0 201 t-d 329

o-a 363 m-m 156, 216, 217, 253, 312, 313,

o-e 259 317, 318, 324

o-u 196, 285 x-b 307

U-M 125, 151, I56, 211, 2l6, 273, V-M

289, 329, 343, 344 i-h 122, 147, 201, 215, 214, 273,

U-i 247 290, 294, 329

U-h 295 f-u 363

1-/ 203, 289, 295, 312, 313, 350 is 259

!->' 358 p-6 125, 151, 247, 285, 289

l-n 307 p-p

ng-ng 285, 350

(Note: k-g, p-p, and \-u do not appear in the material here collated,
but each has been observed in other Polynesian loan words in Pala.)

Identical 156, 216, 203, 313, 324, 358

Terminal abrasion. . .289, 317, 318, 343, 344
Consonant identity ....211
Consonant mutation :

Terminal abrasion. . . 147, 273, 294

Terminal abrasion 47, 74, 147, 151, 196, 201, 289, 290, 294, 317,

318, 343, 344, 352

Frontal accretion 74

Terminal accretion 312

Metathesis afi-iah, tina-etna.

POLYNESIAN REUCS IN MELANESIA.

109

RALUANA.

a-a 214, 294
0-0 314, 336
u-w 358

l-J 314, 336

\-r

ng-ng
t-t

f-w

358
336

294

214, 294, 314

Terminal abrasion
Terminal accretion . . .

.294
•314

RETAN.

Polynesian 14
Quality 2 1

a a, e, o
e a o o, a, e

i e u e, i

1 /, r, ng
ng ng n n mm, mw

h h
s

V

f v
kg t t p

a-a 123, 318 1-r 331, 336, 358 •

a-e 193, 292 \-ng 123

a-o 317 ng-ng 336

e-a 122 k-o 193

i-c 193, 312 n-n 292, 317

0-0 336 h-/i 331

O-a 331, 357 t-< 318, 357, 358

O-e 123 m-ra 317, 318

u-e 292 m-mw 312

U-* 358 f-T; 122, 193, 292

\-l 123, 312

Identical :

Terminal abrasion . . .318
Consonant mutation :

Terminal abrasion . . . 336

Terminal abrasion 193, 292, 317, 318, 336

Frontal accretion 312

Terminal accretion 122, 331, 336, 357

Metathesis 193

RUAVATU.

a-a 317 U 313
a-e 313 n-n 317
\-i 313 m-w 313, 317

VL-U 317

110 THE POLYNESIAN WANDERINGS.

SAA.

Polynesian 30
Quality 80

a a, e
e e 00

I i u u

\ I, r, n
ng ng n n, ng m m

h s, t
s s

v w, h
f h
k— t— /, k v P

a-a 200, 214, 217, 250, 291, 292, n-n 187, 259, 292, 317, 330

294, 307, 309, 3i7, 318, 338, n-ng 351

339, 35o, 352, 360 h-y 352

a-e 292, 300, 313, 350 h-t 338, 339

e-e 122, 187, 190, 318, 338 s-s 251, 298, 351

\-i 200, 259, 273, 291, 300, 313, t-t 329

329, 330, 35o, 35i. 352 t— 217, 294, 318, 350, 352, 358

0-0 187, 259, 309 t-k 356

u-u 200, 273, 292, 294, 298, 317, mm 200, 217, 313, 317, 318

329, 33o, 35i, 358, 360 \-w 291

1-/ 307, 309, 339, 350 \-h 307

\-r 200, 358 f-h 122, 214, 259, 273, 292, 294,

1-w 313 ' 329, 36o

ng-ng 309, 350 p-p 190, 250
k— 214, 250, 251, 300, 338

Identical 187, 190, 200, 298, 309, 317, 33°, 358

Vowel identity 214, 217, 250

Consonant mutation :

Vowel identity 259, 273, 291, 292, 294, 307, 318, 329, 350, 352,

360

Frontal abrasion 217, 250, 351, 352, 358

Frontal accretion 298, 300

Terminal accretion 351, 358

Metathesis 351

POLYNESIAN RELICS IN MELANESIA. Ill

SANTO.

Polynesian 27
Quality 8 1

a a, e, i, 0
e e, i o o, a

i i, e, u u u

1 I, r

ng n, m n n, mm

h s, —

s ^

V

f V, w

k k t t p p

a-a 199, 216, 217, 228, 292, 300, l-r 305, 336, 355

312, 313, 318, 324, 356 ng-n 285, 336, 346

a-e 216, 253, 316, 324 ng-m 199

a-i 217, 292 k-fe 300, 302, 305

a-o 239, 302, 346, 352 n-rc 292, 343

e-e 47 h-j- 47, 206, 278, 352

e-* 153, 318 h— 255

\-i 47, 206, 300, 305, 312, 346, s-j 239

352, 355 t-t 47, 199, 216, 217, 302, 318,

\e 313 324, 343, 346, 352, 355

i-u 305 mm 216, 217, 253, 255, 312, 313,

0-0 153, 206, 285, 336, 355, 356 316, 318, 324

o-a 255 f-v 292

u-w 216, 228, 239, 278, 292, 316, i-w 206
343 P-P 285

[I 153, 228, 312, 313, 316

Identical 153, 228, 300, 312, 343, 355, 356

Consonant identity. . . .216, 217, 239, 253, 302, 316, 318, 324

Vowel identity 199

Consonant mutation :

Vowel identity 47, 206, 292, 336

Terminal abrasion. . .278, 285

Terminal abrasion 278, 285

Frontal accretion 312

Terminal accretion 153, 239, 305, 313, 324, 336

112

THE POLYNESIAN WANDERINGS.

SASAR.

e e,

ng ng, n

k g, ng, w
a-a 123, 147, 217, 250, 309, 318

a-e 317, 337, 35o

a-o 239, 292, 307, 328

a-M 250, 316

e-e 190

e-u 272

0-0 356

o-e 123

u-u 272, 306, 316

u-o 292

U-* 343

1 / 123, 307, 309, 316,
\-ng 123

I-« 335
ng-wo 309

Identical:

Terminal abr 190, 309

Consonant identity . .316,317,

Vowel identity 217, 306

Consonant mutation:

Terminal abr 147

Polynesian

22

Quality

45

a a, e, 0, u

1 O 0, e

u u,
1 /, ng, t
n n
h

i, 0
m m

s s

t t, m, —

V w
f v, w
P P

35o

356

328,337

ng-w 350

kg 250, 251

k-ng 301

k-w 306

n-n 147, 292, 317, 328

s-j 239, 251, 337

t-t 306, 350, 356

t-m 217

t— 318

m-w 217, 316, 317, 318, 328

x-w 307

i-v 147, 292

f-w 272

p-p 190, 250

Terminal abrasion . 147, 190/2 17, '2 39/306,
309, 3i7.?3i8,*328,?356
Frontal accretion . . 301
Terminal accretion. 250, 272, 335, 350

SAVO.

a a

e e, a

o 0

1 a, 0

u w

1 /

ng ng

n n
h

s s, z

m m

X V

f

k k, g

t

P P>

a-a

239,

250, 316, 328, 337

k-a

250

e-e

190

n-n

328

e-a

272

S-S

337

i-o

305

s-z

239

i-a

335

mm

316, 328

0-0

285

x-v

272

u-w

239,

272, 316, 328

p-p

285

\-l

272,

316

p-v

250

ng-wa

285

p-b

190

k-Jfc

305

■v,

POLYNESIAN REUCS IN MELANESIA. 113

SESAKE.

Polynesian 57

Quality 76

a a, o, u
e e 00

i i, e u u, a, o

1 /, r, nd
ng ng n n m m

h s, v
s J

v v, u
f /, v, p, w
k k, g, w t nd, t p />, b, tnb, v

a-a 45, 147, 151, 162, 169, 200, ng-ng 151, 209, 246, 274, 285, 336,

209, 214, 215, 217, 239, 246, 350

258, 260, 270, 272, 274, 278, k-k 151, 194, 214, 251, 258, 300,

291, 292, 294, 300, 301, 307, 301, 338, 361

310, 312, 317, 318, 322, 324, k-# 169

335, 337, 338, 339, 342, 350, k-w 305

352, 356 n-w 169, 296, 317, 330, 342

a-o 203 lw 206, 215, 278, 338, 339, 352

a-u 307 h-v 169

e-e 151, 153, 318, 338 s-^ 45, 203, 239, 251, 298, 337,

\-i 149, 169, 194, 200, 206, 215, 342

270, 285, 291, 298, 300, 305, t-t 162, 217, 258, 270, 294, 318,

312, 329, 330, 342, 350, 352 324, 352, 356, 357, 358

i-e 305 t-nd 209, 329, 350

0-0 153, 206, 285, 296, 336, 356, mm 158, 200, 217, 257, 258, 312,

357 3i7, 318, 322, 324, 343
u-tt 151, 162, 200, 212, 239, 246, x-v 307

258, 270, 272, 278, 292, 294, \-u 310

296, 298, 301, 317, 330, 335, f-/ 296

358 i-v 147, 214, 246, 292, 294
u-a 329 i-p 215

u-o 194 i-w 206

1-/ 153, 212, 257, 305, 307, 310, p-p 151

312, 339, 35o p-b 361

\-r 194, 200, 203, 260, 322, 335 p-mb 285

\-nd 336 p-v 45

Identical 151, 153, 194, 200, 212, 217, 239, 258, 270, 296,

300, 301, 312, 317, 318, 322, 324, 330, 342, 356,

358

Terminal abrasion . . .357
Consonant identity. . . .203, 239, 257, 307

Vowel identity 158, 162, 291, 336

Consonant mutation :

Vowel identity 169, 206, 209, 214, 215, 246, 278, 285, 292, 294,

338, 339, 35o, 352

Terminal abrasion 147, 274, 310, 357

Frontal abrasion 329

Frontal accretion 158, 162, 260, 274, 291, 298, 310, 339

Terminal accretion 194, 239

114

THE POLYNESIAN WANDERINGS.

TANNA.

Polynesian 37
Quality 29

a a, e, 1
e e, i o e, u

i i, a, e u u, a, e, i

1 r
ng ng, n n n mm

h /
S s, r

v u

f /, v, k
k k, g, y t t, s, h p p, b, m

a-a 150, 173. 216, 218, 238, 249, ng-w 285

256, 274, 293, 317, 322, 334, kk 150, 194, 258, 361

337, 352, 360 kg 306

a-c 147, 34o, 350 k-y 251

a* 217, 313, 340 n-n H7. 3i7, 321

e-e 173, 240, 290 h-t 340

e-i 147 s-^ 25i

\-i 150, 194, 200, 274, 312, 313, S-r 337

319, 321, 350, 352 U 150, 168, 274, 306

i-e 290 t-s 207

i-o 319 t-fc 258

o-e 285 m-w 200, 258, 274, 317, 321, 322,

o-w 1 50 340

u-w 168, 173, 194, 216, 238, 240, V-u 291

247, 273, 317, 321, 358 f-/ 293

u-a 168 f-v 147, 290

u-e 306 f-fc 273

VL-i 334 P"# x73, 36i

\-r 194, 200, 322 p-b 285

ng-n^ 274 p-w 207

Identical 194, 274, *73> 293

Consonant identity. . . . 150, 168, 313, 319, 334, 35°
Consonant mutation :
Vowel identity 352

Terminal abrasion 147, 273, 306

Frontal abrasion 321

Frontal accretion 147, 150, 273, 291, 306, 340, 350, 358

Terminal accretion. . . .200, 317, 321, 340

POLYNESIAN RELICS IN MELANESIA. 115

TANGOAN SANTO.

a a
e e, o oo

[ i u u

1 I, r
ng ng n n m n

h s, ts
s s

v

f b, p, th
k k, v t t p b

&-a 147, 148, 193, 209, 217, 239, k-v 194

278, 302, 312, 316, 338, 340, n-w 147

346 h-s 206, 278, 338

e-e 338 h-ts 340

e-o 247 s-s 239

i-z 193, 194, 206, 302, 312, 346 t-t 148, 209, 217, 247, 302, 346,

0-0 147, 206 358

vl-u 148, 194, 239, 247, 278, 316, m-n 312, 316, 340

358 i-b 193, 206

1-/ 312, 316 i-p 148

\-r 194, 358 i-th 147

ng-ng 209, 346 p-6 247

k-k 193, 302, 338

UEA.
a a 274 ng-ng 274

UGI.

a- a

214.

250

e-e

318

\-i

259,

300

0-0

259,

357

u-w

292,

317

k—

214,

250

Polynesian 8

Quality 75

n /

259

t-t

357

t—

3i8

m-m

3i7

i-v

292

i-h

214,

259, 292

n-w 292, 317 p-p 250

116 THE POLYNESIAN WANDERINGS.

ULAWA.

Polynesian 28
Quauty 77

a a

e e, a 00

i i u u

I I, r, n

ng ng n n, ng m m

h 5

s s

V w

f fe

k— t t, — v P

A-a 200, 214, 217, 239, 250, 291, ng-ng 309, 343, 350

294, 300, 309, 312, 313, 317, k— 214, 250, 251, 300, 301

318, 324, 350, 352, 360 n-n 259, 317, 330

e-e 190, 318 n-ng 351

e-a 122 h-s 352

\-i 200, 259, 291, 300, 312, 313, s-^ 239, 251, 298, 351

329, 33o, 35o, 35i, 352 t-t 329, 343

0-0 259, 309 t— 217, 294, 318, 324, 350, 352

vl-u 2oo, 239, 294, 298, 317, 329, m-w 200, 217, 312, 313, 317, 318,

330, 343, 35i, 360 324
\l 309, 312, 350 \-w 291

\-r 200 i-h 122, 214, 259, 294, 329, 360

\-n 313 V-P x9o, 250

Identical 190, 200, 250, 309, 312, 317, 330, 343

Vowel identity 214, 217, 239, 294, 313, 318, 324, 350

Consonant mutation :

Vowel identity 259, 291, 300, 329, 360

Frontal abrasion 217, 250, 350, 351, 352

Frontal accretion 239, 294, 298

VANIKORO.

a-a
a-e

292
292
273

u-u

f-f

i-p

273,
292

273

292

POLYNESIAN RELICS IN MELANESIA.

117

VATURANGA.

Polynesian 39
Quality 76

a a

e e, a o 0, a

i i, a, 0 u u, o

1 /, r, k, —
ng ng n n, I m m

h s, h
s s

v

f v, h, ng
k k, h, ng t t, nd p 6, wz&, v

a-a 123, 147, 200, 214, 215, 217, ng-ng 285, 309, 336, 346

258, 278, 284, 294, 309, 313, k-fe 305, 338

316, 317, 318, 324, 328, 338, k-ng 251, 306

346, 352, 360 k-h 258, 301

e-e 122, 190, 257, 318, 338 n-n 147, 201, 317, 328, 330

e-a 272 n-l 259

\-i 200, 206, 215, 259, 285, 298, h-s 206, 215, 338, 352

313, 329, 33o, 346, 352 h-h 278

\-a 305 s-* 251, 298

i-o 305 u 217, 258, 294, 306, 318, 324,

0-0 123, 147, 201, 206, 259, 285, 346, 352, 358

309, 336 t-nd 329

o-a 257 mm 200, 217, 258, 313, 316, 317,

u-u 200, 258, 272, 278, 284, 294, 318, 324, 328

298, 306, 316, 317, 328, 329, f-v 122, 147, 206, 214, 259, 272,

330, 358, 360 294, 329

u-o 306 i-h 201

1-Z 123, 284, 309, 316 i-ng 215

l-r 200, 272, 305, 335, 336 p-o 190

1-fe 313 P-w6 285

1 — 123 p-v 284

Identical 200, 214, 217, 278, 298, 309, 317, 318, 324, 328,

330, 346, 358
Consonant identity. . . .257

Vowel identity 123, 215, 258, 259, 313

Consonant mutation :

Vowel identity 147, 190, 201, 206, 284, 285, 294, 329, 338, 352,

360

Terminal abrasion 214

Terminal accretion 258, 316, 336

118 THE POLYNESIAN WANDERINGS.

VOLOW.

Polynesian 38
Quality 37

a a, e, i, o
e e 00

\ i, e u i, e, o

1 h r, g
ng ng, n n n, ng m m, mw

h h
s //

v mb, w
f v
k g, ng t *, nd, m p 6, w, nggtnbw

a-a 147, 214, 217, 239, 294, 318, kg 251, 300, 301, 306

309. 337, 338, 339, 357 k-w? 357

a-e 268, 290, 316, 317, 322, 324, n-n 147, 317, 328

328, 350 n-ng 151, 351

a i 350 h-fc 338, 339, 35i

a-o 307 s-& 239, 251, 337

e-e 122, 151, 190 t-/ 268, 294, 306, 324, 358

i-i i54, 35i *-wrf 35o

i-e 290, 300, 312 t-m 217

0-0 285, 336, 357 mm 217, 312, 316, 317, 322,

u-e 316 324, 328

u-i 212, 306, 358 m-mw 316

u-o 239, 284 v-rao 291

1-/ 154, 212, 284, 307, 309, 312, x-w 307

3l6, 339, 35O f-V 122, I47, 2 14, 290, 294

\-r 268, 322, 335, 336 p-b 190

\-g 358 P-w 284

ng-ng 154, 3°9, 336, 350 p-nggmbw 151, 285
ng-w 285

Identical 214, 324

Terminal abrasion. 154, 309, 318, 336, 339
Consonant identity. . . .212, 268, 317, 328

Vowel identity 357

Consonant mutation :

Terminal abrasion. . . 190, 285

Terminal abrasion 151, 154, 190, 214, 268, 284, 285, 290, 294, 300,

306, 309, 3i6, 317, 3i8, 328, 336, 338, 339, 35i

Frontal abrasion 351

Frontal accretion 290, 301, 339

Terminal accretion 122, 212, 312, 350, 351

POLYNESIAN REUCS IN MELANESIA. 119

VURAS.

Polynesian 28
Quality 40

a a, e, o, u
e e o o, e

i c u u, 0

1 /, r, nd, ng
ng ng, n n n m m

h s
s s

V w
i v
k g, w t /, m p w, kpw

a-o 123, 147, 214, 217, 309, 317, ng-ng 309, 336, 346

318, 328, 339 ng-n 285

a-e 239, 316, 324, 346 kg 251, 301

a-o 292, 307 k-w 306

aw 307 n-w 147, 292, 317, 328, 343

e-e 122 h-s 339

i-e 312 s-.y 239, 251

0-0 147, 285, 336, 356, 357 t-t 306, 318, 324, 343, 346, 356,

o-e 123 358

u-w 306, 316, 343, 358 t-m 217

u-o 284, 292 m-w 217, 312, 316, 317, 318, 324,

\ I 123, 284, 307, 309, 312, 316, 328

339 v-w 307

I-r 336, 358 f-u 122, 147, 214, 292

l-nd 335 p-w 284

\-ng 123 p-kpw 285

Identical 214, 285, 318, 358

Terminal abrasion. . .309, 317, 328, 356
Consonant identity. . . .239, 316
Consonant mutation :

Vowel identity 147

Terminal abrasion . . . 357

Terminal abrasion 214, 217, 239, 284, 306, 312, 316, 317, 318, 328,

336, 339, 346, 356

Frontal accretion 123, 301, 312, 339

Terminal accretion 122, 324, 336

120

THE POLYNESIAN WANDERINGS.

WANGO.

Polynesian 36
Quality 75

a a

e e o 0, a

\ i, u u u, 0

1 I, r

ng ng n n, ng m m

h s, t

s .S-, t

v fc
f fc
kg,— t t, k, g, — p b, h

a-a 214, 215, 217, 239, 250, 284, k— 214, 250, 300, 305

300, 397, 309, 312, 313. 3i7, k-9 251

3i8, 335, 337, 339, 342, 348, n-n 317, 328, 330, 342

35o, 352, 357, 360 n-ng 351

e-e 122, 190, 257, 318 h-s 215, 352

\-i 215, 259, 273, 298, 300, 305, h-t 339

312, 313, 329, 33o, 342, 348, s-s 239, 251, 298, 342, 351
350, 352 S-2 337

i-u 305 t-t 348

0-0 259, 309, 357 *-fe 35o, 356

0-a 257 t-g 329

U-m 239, 273, 284, 298, 317, 329, t— 217, 318, 352, 358

330, 335, 35i, 358, 360 mm 217, 257, 312, 313, 317, 318,

u-o 328 328

1-/ 309, 348 V-fc 307

l-r 257, 259, 284, 305, 307, 312, f-fe 122, 214, 215, 259, 273, 329,

313, 335, 339, 35o, 358 360
ng-ng 309, 350 p-b 190, 250

p-h 284

Identical 239, 250, 298, 309, 312, 313, 317, 330, 342, 348

Consonant identity. . . .257, 328

Vowel identity 214, 217, 250, 259, 300, 318, 339, 350, 352, 357

Consonant mutation :

Vowel identity 190, 215, 273, 284, 307, 360

Frontal abrasion 217, 250, 305, 351, 352, 358

Frontal accretion 298

Terminal accretion 335, 351

POLYNESIAN RELICS IN MELANESIA.

121

In our next series of tables we present the various observed muta-
tions reduced to the Proto-Samoan original. The collation of the
liquid semivowel is designedly incomplete for the reason that we
lack sufficient and sufficiently positive data upon the occurrence of
the r grasseye and its reproduction in terms of / and r. Under the
l-r head, therefore, we note only anomalous changes.

a-i

Alo Teqel

Eromanga

Leon

Moanus

Pala

Ambrym

Fagani

Lifu

Mosin

Retan

Aneityum

Gog

Lo

Mota

Saa

Arag

Iai

Maewo

Motlav

Santo

Baki

King

Makura

Nifilole

Sasar

Baravon

Lakon

Malekula

Norbarbar

Tanna

Bierian

Lamassa

Marina

Omba

Vanikoro

Buka

Lambell

Matupit

Paama

Volow

Deni

Laur

Merlav

Pak

Vuras

Ambrym

Eromanga

Lo

Mosin

Santo

Aneityum

Fagani

Malekula

Mota

Tanna

Baki

King

Malo

Motlav

Volow

Belaga

Lakon

Moanus

Norbarbar

Alo Teqel

Eromanga

Lo

Motlav

Retan

Aneityum

Fagani

Maewo

New Georgia

Santo

Arag

Iai

Malekula

Nggela

Sasar

Baki

King

Malo

Nifilole

Sesake

Bierian

Lambell

Merlav

Norbarbar

Volow

Bugotu

Leon

Mosin

Pak

Vuras

Alo Teqel

Buka

King

Mota

Nifilole

Ambrym

Deni

Lamassa

Motlav

Omba

Aneityum

Dufaure Id

Lambell

Natalava

Pak

Baki

Epi

Laur

New Georgia

Sasar

Bierian

Eromanga

Mosin

Nggela

Sesake

Belaga

Bugotu

Nggela

Ambrym

Bugotu

Lakon

Marina

Nifilole

Aneityum

Buka

Lamassa

Merlav

Pala

Arag

Duke of York

Lambell

Mota

Retan

Baki

Epi

Laur

Motlav

Savo

Baravon

Fagani

Lo

Motu

Ulawa

Bauro

Kabakada

Malo

Nggao

Vaturanga

Bierian

King

Aneityum

Brierly Id

Lifu

Malo

Nggela

Arag

King

Lo

Moanus

Pak

Baki

Laur

Maewo

Mota

Santo

Belaga

Leon

Malekula

Motu

Tanna

Bierian

Baki

Gog

Lambell

Malo

Pala

Buka

King

Laur

Motu

Tangoan Santo

Epi

Lamassa

Alo Teqel

Pak

Sasar

122

THE POLYNESIAN WANDERINGS.

Mota

e-ou

i-M

Malekula

1

Aneityum
Buka

Mota
Nguna

Savo

Tanna

Vaturanga

Alo Teqel

Aneityum

Bougainville

Buka

Eromanga

Fagani

Gog

Kabakada

King

Lakon

Lamassa

Malekula

Malo

Mosin

Mota

Motlav

Motu

New Georgia

Nggao

Norbarbar

Pak

Pala

Retan

Santo

Sesake

Tanna

Volow

Vuras

Aneityum
Baki

Eromanga
Mota

Motu

New Georgia

Savo

Vaturanga

Baki
Baravon
Bierian
Duke of York
Fagani

Gog

Kabadi

Kalil

King

Lamassa

Lambell

Lo

Maewo

Makura

Malekula

Malo

Merlav

Moanus

Mota

Motu

Nggao

Nguna

Pala

Santo

Wango

i-ei

Motu

Baravon

Bugotu

Buka

Epi

Fagani

King

Lakon

Matupit

Mota

Norbarbar

Pala

Retan

Santo

Vaturanga

Wango

Alo Teqel

Baravon

Mosin

Motlav
Norbarbar

Pak
Pala

Retan
Sasar

Tanna
Vuras

Aneityum
Baki

King
Laur

Malekula
Malo

Mota
Nggao

Norbarbar
Paama

Ambrym
Aneityum
Baravon
Buka

Duke of York
•Kabakada
Kalil
King

Lamassa
Lambell
Laur

Malekula

Matupit

Motu

Nguna

Pala

Tanna

Mota

Motu

Ambrym

Buka

Duke of York

Gog
Lambell

Malo
Merlav

Mota
Motu

Sesake
Tanna

Alo Teqel
Aneityum

Eromanga
Malekula

Malo
Mota

Omba
Retan

Tanna
Volow

POLYNESIAN RELICS IN

Alo Teqel Bululaha Lambell

Ambrym Epi Laur

Aneityum Gog Lo

Arag Guadalcanar Maewo

Baki King Makura

Bierian Lamassa Malekula

Ambrym

Aneityum

Arag

Bugotu

Buka

Duke of York Eaur

Eromanga

Gog

Iai

Lamassa

Lambell

Leon

Malo

Mosin

Mota

Motlav

Malo

Paatna

Merlav

Pala

Mota

Retan

Motlav

Sasar

Nifilole

Tanna

Norbarbar

Volow

Motu

Sesake

New Georgia

Vaturanga

Nifilole

Volow

Pak

Vuras

Sasar

Wango

123

Lakon

U-h

Laur

u-p

Aneityum

Pala

Aneityum

Mota

Omba
AE

Nggela
Aneityum

AI

\-e Eromanga
Lo

Malekula

Motlav
AU

Norbarbar

i-ai

Arag

Murray Id

New Georgia

Nggela

au-e

l-nd

H

Norbarbar

Malekula

Mosin

Norbarbar

Alo Teqel
Baki

Bugotu

Buka

Gog

Aneityum

Baki (note 312)

Motlav

Merlav

Maewo

Pak

Pak

Merlav

Pak

Ulawa

Vuras

Sasar

Norbarbar

\-dr

Lakon

Bierian

l-tk

Bugotu

\-nl

Lamassa

124

THE POLYNESIAN WANDERINGS.

l-k

\-h

\-n

\-mj
\-ng

Aneityum

Nggao

Erotnanga

Lo

Dufaure Id
Mosin

Mota
Nengone

Baki

Alo Teqel

Gog

Lakon

Merlav
Mosin

Vaturanga (note 312)

Pala

Mota
Motlav

Saa

Pak
Retan

Ulawa

Sasar
Vuras

Baravon
Bierian

ng-rag
Alite
Alo Teqel
Ambrym
Arag
Baki
Baravon
Belaga
Bierian
Bougainville
Bugotu
Buka
Bululaha

Guadalcanar Motu

Paama

Vaturanga

ng-»

Alo Teqel

Baki

Buka

Duke of York

Gog

Kabadi

NG

Duke of York

Lo

New Georgia

Sasar

Eromanga

Maewo

Nggao

Savo

Fagani

Malekula

Nggela

Sesake

Gog

Malo

Nguna

Tangoan Santo

Guadalcanar

Marina

Norbarbar

Tanna

Kabakada

Matupit

Omba

Uea

King

Merlav

Paama

Ulawa

Lakon

Moanus

Pak

Vaturanga

Lamassa

Mosin

Raluana

Volow

Lambell

Mota

Retan

Vuras

Laur

Motlav

Saa

Wango

Lemaroro

Nengone

King

Maewo

Motlav

Pala

Lakon

Makura

Motu

Santo

Lamassa

Malekula

Nguna

Sasar

Lambell

Malo

Norbarbar

Tanna

Laur

Moanus

Omba

Volow

Lo

Mota

Pak

Vuras

ng-w;

Aneityum

ng-nd

Natalava

ng-g

Aneityum
Baki

ng-m

Bierian

ng—

k-k

Dufaure Id

Ambrym

Aneityum

Baki

Baravon

Bierian

Bougainville

Bugotu

Deni

Eromanga
Lo

Malekula

Motu

Makura
Malekula

Malo

K

Mota

Mota

Duke of York

Lifu

Motu

Epi

Maewo

Murray Id

Eromanga

Makura

Nengone

Fagani

Malekula

New Georgia

Iai

Malo

Nggao

King

Merlav

Nggela

Lamassa

Moanus

Nguna

Lambell

Nggela

Santo

Nifilole

Santo

Savo

Sesake

Tangoan Santo

Tanna

Vaturanga

POLYNESIAN RELICS IN MELANESIA.

125

k-g

Alo Teqel

Ambrym

Aneityum

Arag

Belaga

Bugotu

Fagani

Gog

Lakon

Laur

Leon

Lo

Maewo

Malekula

Malo

Marina

Merlav

Mosin

Mota

Motlav

Motu

Murray Id

Nggao

Nggela

Norbarbar

Omba

Pak

Pala

Retan

Sasar

Savo

Sesake

Tanna

Volow

Vuras

Wango

k-kh

Malekula

Tangoan Santo

k-ng

Alite
Aneityum

Mosin
Mota

Motlav
Pak

Sasar
Vaturanga

Volow

k-»gg

Eromanga

Malekula

k-h

Ambrym

Malekula

New Georgia

Vaturanga

k-m

Motu

k-mk

Baki

k-v

Malo

k-w

k-v

Alo Teqel

Lakon

Maewo

Tanna

Tangoan Santo

Merlav

Mosin

Mota

Norbarbar
Omba

Pak
Sasar

Sesake
Vuras

n-ng

Alite

Buka

Malo

Nengone

Ugi

Ambrym

Bululaha

Matupit

Norbarbar

Ulawa

Baravon

Duke of York

Moanus

Pala

Wango

Belaga

Malekula

Motu

N
Leon

Saa

Alo Teqel

Eromanga

Mota

Saa

Ambrym

Fagani

Lifu

Motlav

Santo

Aneityum

Gog

Lo

Motu

Sasar

Arag

Guadalcanar

Maewo

Nengone

Savo

Baki

Iai

Makura

Nggao

Sesake

Baravon

Kabadi

Malekula

Nggela

Tangoan]Santo

Belaga

Kabakada

Malo

Nguna

Tanna

Brierly Id

Kalil

Marina

Nifilole

Ugi

Bugotu

King

Matupit

Norbarbar

Ulawa

Buka

Lakon

Meli

Omba

Vaturanga

Bululaha

Lamassa

Merlav

Pak

Volow

Duke of York

Lambell

Modnus

Retan

Vuras

Epi

Laur

Mosin

Ruavatu

Wango

Bugotu

Lamassa

Merlav

Nggela

Ulawa

Fagani

Lambell

Mota

Pala

Volow

Gog

Laur

Motlav

Saa

Wango

King

Lo

n-gn

n-nv

Bugotu

i

Aneityum

Buka

Moanus

126

THE POLYNESIAN WANDERINGS.

n-/

Alite

Bougainville

Maewo

Motlav

Ugi

Arag

Buka

Marina

Nggela

Vaturanga

Bierian

Fagani

Mota

New Georgia Nggao

n-t

Malekula

Alo Teqel

Marina

h-h

h-s

Ambrym

Bugotu

Lakon

Nggela

Retan

Arag

Deni

Makura

Norbarbar

Vaturanga

Belaga

Epi

Motlav

Omba

Volow

Bierian

Eromanga

Nggao

Alite

Gog

Malekula

Mota

Sesake

Bauro

Kalil

Malo

Nguna

Tangoan Santo

Brierly Id

King

Marina

Pak

Ulawa

Buka

Lamassa

Merlav

Pala

Vaturanga

Bululaha

Lambell

Modnus

Saa

Vuras

Eromanga

Laur

Mosin

Santo

Wango

Fagani

Maewo

h-t

h-d

Alite

Saa

Motu

h-th

Belaga

Guadalcanar

h-ts

h-j

Tangoan Santo

Aneityum

Aneityum

Lo

h-v

Sesake

h—

Tanna

Malekula

Ambrym

Baravon

Deni

Baki

S
Maewo

Alite

Fagani

Alo Teqel

Gog

Mai

Aneityum

Guadalcanar

Malekula

Belaga

King

Malo

Bierian

Lakon

Merlav

Brumer Id

Lamassa

Moanus

Buka

Lambell

Mosin

Bululaha

Laur

Mota

Epi

Leon

Nengone

Eromanga

Wango

Duke of York Santo

New Georgia Sasar

Nggao

Savo

Nggela

Sesake

Nifilole

Tangoan Santo

Norbarbar

Tanna

Paama

Ulawa

Pak

Vaturanga

Saa

Vuras

Santo

Wango

Savo

s-sh

Brumer Id

POLYNESIAN REUCS IN MELANESIA.

127

s-/»

s-d

s-t

s-k

Ambrym

Aneityum

Arag

Motu

Alite

Baki

s-th

s-y

Belaga

Baki

Bierian
Bugotu
Buka

Wango

Lakon

Lo

Motlav

Motu

Nggao

Nggela

Norbarbar

Omba

Volow

Tanna (note 239)

Duke of York.

Alite

Santo

t-t

t-d

t-nd

t-dr

Alo Teqel

Bululaha

Laur

Motu

Saa

Ambrym

Deni

Lo

Nengone

Santo

Aneityum

Duke of York

Maewo

New Georgia

Sasar

Arag

Epi

Makura

Nggao

Sesake

Baki

Eromanga

Malekula

Nggela

Tangoan Santo

Baravon

Fagani

Malo

Nguna

Tanna

Belaga

Gog

Marina

Nifilole

Ugi

Bierian

Iai

Matupit

Norbarbar

Ulawa

Bougainville

Kalil

Merlav

Omba

Vaturanga

Brierly Id

King

Moanus

Pak

Volow

Brumer Id

Lakon

Mosin

Raluana

Vuras

Bugotu

Lamassa

Mota

Retan

Wango

Buka

Lambell

Motlav

Arag

Buka

King

Malo

Pala

Bierian

Eromanga

Kalil

Baravon

Lo

Motlav

Norbarbar

Vaturanga

Bougainville

Maewo

Nggela

Sesake

Volow

Bugotu

Ambrym

t-ndr

Modnus

t-lh

t-s

t-h

t-j

t-r

Aneityum

Baki

Bugotu

Ambrym

Aneityum

Bierian

Buka
Eromanga

Malo
Marina

Arag

Bierian

Tanna

Baki

Malekula

Lemaroro

Malekula

Moanus

Mota

New Georgia

Paama
Tanna

Mota (note 258)

128

THE POLYNESIAN WANDERINGS.

t-n

King

t-m

t-w

t-k

H

Wango

t—

Nggela
Bugotu

m-t

Lamassa

Lambell

Laur

Alite
Alo Teqel
Gog

Lakon
Merlav
Mosin

Motlav
New Georgia
Norbarbar

Omba
Pak
(note 217)

Sasar

Volow

Vuras

Fagani

Baravon
Brierly Id
Deni

Fagani

Iai

Kabadi

King

Lamassa

Lambell

Lifu
Nggela

Saa
Wango

Alite

Bululaha

Pak

Sasar

Ulawa

Alo Teqel

Fagani

Saa

Ugi

Wango

Bauro

Leon

m-rn

M

Alite

Deni

Laur

Mota

Saa

Alo Teqel

Duke of York

Leon

Motlav

Santo

Ambrym

Epi

Lifu

Motu

Sasar

Aneityum

Eromanga

Lo

New Georgia

Savo

Arag

Fagani

Maewo

Nggao

Sesake

Baki

Gog

Makura

Nggela

Tangoan Santo

Baravon

Guadalcanar

Malanta

Nguna

Tanna

Belaga

Iai

Malekula

NMlole

Ugi

Bierian

Kabakada

Malo

Norbarbar

Ulawa

Brierly Id

Kalil

Marina

Omba

Vaturanga

Brumer Id

King

Matupit

Pak

Volow

Bugotu

Lakon

Merlav

Retan

Vuras

Buka

Lamassa

Moanus

Ruavatu

Wango

Bululaha

Lambell

Mosin

m-mw

Merlav

Mota

Motlav

Retan

Volow

Mosin

m-n

Marina

Tangoan Santc

1 (note 312)

m-«g

Alo Teqel

Lo

Nggela

Malo

v-/

\-b

Baki

Baravon

Belaga

Ambrym

Deni

v-mb

Bugotu

Gog

Merlav

Omba

Sesake

Lo

Mota

Savo

Vanua Lava

Malekula

Nggela

Fagani (note 291)
Mai Pala

Maewo Merlav

Motlav

Volow

POLYNESIAN RELICS IN MELANESIA.

129

v-p

Aneityum
Lo

v-p-w

\-h

Malekula

Saa

Marina

Wango

Mota

Norbarbar

Vanua Lava

Alite
Baki
Bierian

Eromanga
Malekula

Malo
Moanus

Mota
Pala

Sesake
Tanna

v-t

Alo Teqel

Ambrym

Aneityum

Arag

Baravon

Marina

v-ku

Bululaha

Gog

Leon

Lo

Malekula

Mosin
Mota
Motlav
Nengone

Norbarbar
Omba
Pak
Saa

Sasar
Ulawa
Volow
Vuras

Alite

Arag

Maewo

f-v

Alite

Epi

Laur

Mota

Retan

Alo Teqel

Eromanga

Leon

Motlav

Santo

Ambrym

Fagani

Lo

Motu

Sasar

Arag

Gog

Maewo

Nggela

Sesake

Baki

Guadalcanar

Makura

Nguna

Tanna

Baravon

Iai

Malekula

Nifilole

Ugi

Belaga

Kabadi

Malo

Norbarbar

Vaturanga

Bierian

King

Marina

Omba

Volow

Bugotu

Lakon

Merlav

Pak

Vuras

Deni

Lamassa

Mosin

N

Bierian

Malekula

Nggao

Nguna

Tanna

Fagani

Meli

Nggela

Sesake

Vanikoro

Lamassa

f-p

Aneityum

Eromanga

Lambell

Moanus

Tangoan Santo

Baravon

King

Malekula

Nifilole

Vanikoro

Brierly Id

Lamassa

Matupit

Sesake

i-b

Ambrym

Bierian

Eromanga

Malo

Tangoan Santo

Baki

Duke of York

Malekula

f-mb

Moanus

Nggela

f-mbw

Malekula

t-k

f-kr

Aneityum
Nggao

f-n

Nifilole

f-ng

Nggao

Tanna

Vaturanga

130

THE POLYNESIAN WANDERINGS.

f-s

i-th

Eromanga

U

i-h

f-u

i-w

f—

Pala

Tangoan Santo

Moanus

Mota

Ambrym

Bululaha

Lo

Pala

Ulawa

Aneityum

Kalil

Motu

Saa

Vaturanga

Bougainville

Lam bell

Paama

Ugi

Wango

Buka

Laur

Aneityum

Malo

Pala

Alo Teqel

Kabakada

Matupit

Nguna

Santo

Aneityum

King

Merlav

Nifilole

Sasar

Arag

Lakon

Mosin

Norbarbar

Sesake

Baravon

Lamassa

Mota

Omba

Ugi

Duke of York

Lo

Motlav

Raluana

Ulawa

Gog

Maewo

Nengone

Duke of York

Lamassa

Maewo

Malekula

Nengone

Iai

Lifu

Makura

Moanus

p-p

Alo Teqel

Bugotu

Lo

Nengone

Sasar

Aneityum

Bululaha

Malekula

New Georgia

Savo

Arag

Eromanga

Marina

Norbarbar

Sesake

Baravon

Fagani

Moanus

Pak

Tanna

Belaga

Lakon

Mota

Saa

Ugi

Bougainville

Lifu

Motu

Santo

Ulawa

p-mp

Eromanga

p-b

Alite

Fagani

Laur

Motu

Sesake

Baki

Gog

Maewo

New Georgia

Tangoan Santo

Baravon

Guadalcanar

Malekula

Nggela

Tanna

Bierian

Kabakada

Malo

Norbarbar

Vaturanga

Brierly Id

King

Matupit

Omba

Volow

Buka

Lamassa

Merlav

Pala

Wango

Duke of York

Lambell

Motlav

Savo

p-mb

Baki

Deni

Lamassa

Merlav

Norbarbar

Baravon

Epi

Lambell

Moanus

Omba

Belaga

Gog

Laur

Nggao

Sesake

Bugotu

King

Malekula

Nggela

Vaturanga

Buka

p-w

Bierian

Lamassa

Tanna (note

207)

p-kpw

Arag

Lakon

Maewo

Motlav

Volow

Gog

Lo

Mota

Norbarbar

Vuras

P-/

Fagani

p-bu

Malo

p-l

Aneityum

POLYNESIAN RELICS IN MELANESIA.

131

p-v

p-w

Arag

Belaga

Buka

Epi
Gog
Lakon

Malo

Marina

Merlav

Mo&nus
Mota

Nggela

Savo

Sesake

Vaturanga

Buka
Maewo

Mosin

Motlav

Volow

Vuras

p-h

P—

Wango
Ambrym

We shall next subject these Melanesian mutations to a comparison
with those which have been observed in the Polynesian family, as
set forth in the tables beginning on page 50, confining our atten-
tion to the consonants.

The following changes are common to the two families : L to t, to nd, to dr,
to n, these in the same series ; to ng in the palatal series ; to extinction. Of
these the most widespread in the Polynesian family are the extinction and
the w-change, the others being found in but one language apiece. In the
Melanesian family the most widely extended are the extinction and the
changes to n and ng.

Distinctly Melanesian mutations are these: L to j, to m, to //;, to nl, to
k, to h, to mj. Of these, \-j identifies itself with the \-t mutation, as will
appear in the examination of T; and with this the Baki mj is probably
associable as a reinforced /. Similarly the \-th of Bugotu is seen to belong
to the \-t mutation when we observe the t-th mutation in that speech.
Lamassa nl is a reinforced /.

The purely Melanesian mutations, then, are 1 to m, to k, to //.

NG

Common to the two families are : NG to n, to k, to extinction, the more
extended in Polynesia being the first and the third, the ^-change occurring
only in the eastern Marquesas and in Viti. The ^-mutation in Melanesia
occurs but once, in Aneityum. The change to n is very frequent. Extinc-
tion is met with only in Torres Straits, at Dufaure Island and Motu, in the
latter speech forming one of the points of great resemblance with Tahiti.

Distinctively Melanesian are: NG to nj, to nd, to g, to m. The ^-muta-
tion is easily seen to be a variant upon the ^-change; it occurs in seven
languages of the New Hebrides and in one of the Solomons.

The purely Melanesian mutations are NG to m, to nj, to nd.

K

Common to the two families are : K to g, to ng, to ngg, to extinction. The
extinction is the most widely extended in Polynesia, being the rule in
Hawaii, Tahiti, and in Samoa where it is so recent that the failure of vowels
wholly to glide over the gap is represented by 4 as an alphabetic character
under the name of the catch. In Melanesia the most frequent is the
^-mutation, and the extinction comes next in frequencv.

132 THE POLYNESIAN WANDERINGS.

Distinctively Melanesian mutations are: K to kh, to mk, to h, to m, to
v, to w, to y. The M-change is of the same order as the ^-mutation, a
further move in the same series. The mk of Baki is but a prefatory rein-
forcement.

The purely Melanesian mutations are : K to h, to m, to v, to w, to y.

N

Common to the two families are : N to /, to r, to ng. None is frequent in
Polynesia ; the /-mutation occurs in Nukuoro and Samoa and in but one
word (fonu-volu 296), the r-mutationin but a single word in Rotuma and
Tahiti. In Melanesia the mutations to ng and l-r are of wide extent.

Distinctively Melanesian mutations are: N to gn, to ny, to m, to t, to
extinction. The first of these is probably an ng-form, for we recall that
the early missionaries in the Pacific adopted that spelling in the case of
Rarotogna for the indisputable Rarotonga.

The purely Melanesian mutations are: N to ny, to m, to t, to extinction.

H

The aspirate is difficult to trace in Polynesia. It is preserved as aspira-
tion only in Niue, Tonga and Uvea ; it appears as j in Rotuma and Viti,
as th in Viti, and once as w in Hawaii. In Melanesia the aspiration is
preserved quite widely, still more widely it has passed into the sibilant.
The //i-mutation appears in Belaga and Guadalcanal

The purely Melanesian mutations are : H to t, to d, to ts, to ;', to /, to v,
to extinction.

Common to the two families are : S to sh, to h, to th, to extinction. In
each family the most widely extended mutation is that from the sibilant
to the aspiration.

Distinctively Melanesian mutations are : S to z, to t, to d , to k, to r, to y,
to w.

Common to the two families are : T to nd, to /, to th, to s, to k, to extinc-
tion. In Polynesia the widest extent is measured by the t-k mutation.
In Melanesia no one mutation has a marked frequency over several others.

Distinctively Melanesian mutations are : T to d, to dr, to ndr, to h, to r,
to n, to m, to w, to g. The d-mutation is but a slight variant upon t itself,
the ^-mutation is a variant of the fe-mutation.

The purely Melanesian mutations are : T to dr, to ndr, to h, to r, to n, to
m, to w, to g.

M

In Polynesia the sole variation upon the labial nasal (m-ng) occurs in but
a single word in Maori and Viti. Similarly in Melanesia this, the only
common mutation, occurs in but a single word, not the same, however, as
the Polynesian, in Alo Teqel and Do.

POLYNESIAN REUCS IN MELANESIA. 133

Distinctively Melanesian mutations are : M to mw, to n, to ng, to v, to
r, to t. The mw-change is scarcely to be classed as a mutation; it is a
fashion in pronunciation which exists but in a restricted area of the Banks
Group in the northern New Hebrides.

The purely Melanesian mutations are : M to v, to r, to /, to n, to ng.

Common to the two families are: V to h, to w. The latter alone has
width of extent in either; x-h in Polynesia is found in but one word each
in Niue, Nukuoro, and^Rotuma; and in Melanesia in but one word common
to Saa and Wango. f j

Distinctively and wholly Melanesian are these mutations: V to/, to b,
to mb, to p, to pw, to u, to t, to ku, to extinction.

Common to the two families are : F to v, to h, to p, to b, to w, to extinc-
tion. In Polynesia the widest extent marks the mutations to //, to v, and
extinction; in Melanesia the order of frequency is v, w, p, h, extinction.

Distinctively Melanesian mutations are: F to mb, to mbw, to k, to kr,
to n, to ng, to s, to th, to t, to u. Of these the mb and mbw are reducible
to p (b), and the u to the common w.

The purely Melanesian mutations are: F to k, to kr, to n, to ng, to s,
to th, to t.

Common to the two families are : P to b {mb), to v, to h, the last occurring
in a single word common to Rotuma and Wango.

Distinctively and wholly Melanesian are: P to mp, to m, to kpw, to/, to
bu, to /, to iv, to extinction.

In a former paper in which I subjected the truly Polynesian lan-
guages to a similar detailed examination* it was pointed out that
with certain exceptions noted the wdiole play of consonant mutation
was a vertical or series matter :

One more preliminary statement: we have already said that for con-
venience we should enter upon our alphabetical conspectus the aspirate in
the neighborhood of each of the three series. The convenience is this, that
the aspirate is not palatal, not lingual, not labial, yet it lies as close to the
one as to the other. We shall find it involved in all these changes, but it
does not affect the rule which we are about to enunciate.

With the three exceptions noted (s-v, ng-n, t-k) the whole play of con-
sonant mutation in Polynesian is a matter of vertical change. When a
palatal changes it changes to another palatal, lingual modified remains
lingual still, and labial remains labial even though its play of mutation
carries it bodily into the vowel tract. But there is no horizontal move-
ment, the labial under stress of change does not become palatal or lingual.

♦"Samoan Phonetics in the Broader Relation," 17 Journal of the Polynesian
Society, 217.

134 THE POLYNESIAN WANDERINGS.

Now, vertical and horizontal are convenient terms to employ when the
conspectus is before the eye, but as terms they have no real value in nature.
That which it is of value to recognize is that which underlies this talk
of vertical mutation, of labial, lingual, palatal invariability. That all-
important underlying fact is this : no matter which of the three organs of
speech mechanism this early speaker elected to employ for the expression
of any given sense he does not change to another organ in case the result
is not satisfactory, and this holds true with his remotest descendants
wherever they may to-day be found. A novice at the trade of speaking,
he may fumble the tool he has chosen to employ, but, being man and
obstinatively progressive, he sticks to the use of that same tool until he
has learned the knack of it.

Accordingly we are to omit all such instances from our discussion
of the Melanesian dealing with Polynesian material, for inasmuch
as the mutations are found in Polynesia without exterior influence
we can not prove that similar vertical mutations when found west
of Viti are due in any degree to Melanesian influence.

The motion in each series is downward, excepting, of course, the
three mutes which stand at the foot of each column and whose
vertical motion can only be upward. Any exception, therefore, to
this general downward motion of mutation calls for attention.
There are but three such.

i. K-y. This is found only in Tanna yasuk (251) rat. As set
forth in the note upon this item the matter is too obscure to serve
as a satisfactory base for any deduction.

2. S-r. This is found only in isd (337) bad Tanna ra, a poor
language from which to draw conclusions, particularly when the
instance is unique.

3. V-u. This, with the x-w, which involves no more vital a dis-
tinction than an alphabetic symbol, is truly an upward motion on
our charts of the sounds. But under the appearance there is a
deeper principle. In the labial mutations we find such forms as
m-mw, p-pw, p-kpw, p-bu, x-ku. If we are to interpret m-mw, for
instance, as implying that m stands fast and at the same time moves
upward, we are at once engaged with the phonetics of Sir Boyle
Roche's bilocal bird. It becomes clear that the nucleus of all the
labials is the vowel- semivowel u-w. In another place I comment
upon the fact that the Melanesians have but recently begun to
acquire command of their lips, not as yet facile. Thus the pri-
mordial semivowel persists with the consonant which is evolving
therefrom, plumule and cotyledon breaking ground together to tell
the tale cf origins.

When these vertical mutations are omitted we shall find a con-
siderable number remaining which we are justified in characterizing
as Melanesian. They are the following, some effort having been
made to assort them in reasonable groups.

POLYNESIAN RELICS IN MELANESIA. 135

1. Involving the aspiration :

k-h h-v x-h h / i-h p-h

Superficially these are extra-serial mutations. If, however, my expla-
nation be valid that the aspiration should be regarded as close to each
series it will be permissible to regard it as the decay stage of each of these
columns. The f-h change, too, is frequent in Polynesia.

K-h. This rests upon five instances, one triply and two doubly sup-
ported. In mataku (258) we find Ambry m matehag Malekula
metoh Vaturanga matahuni. In kafika (193) Malekula Pangkumu
has havih, two instances in one word. In ika (300) we find New
Georgia thani Malekula na-ih. In feast (301), noting the existence
of a second stem hai, we find New Georgia hai Vaturanga hai.
There are two widely separated foci of the mutation, Ambrym-
Malekula, leeward islands in the central New Hebrides ; Vaturanga
and New Georgia, leeward in the central Solomons.

H-/. For this we have the single instance of hala (339) path Aneityum
ne-falaig. The language is not very satisfactory and no great
value may attach to this unique instance.

H-v. For this we have but a single example, Sesake vinaga (169);
and in this there is uncertainty as to whether the Proto-Samoan
was aspirate or sibilant.

\-h. This, the converse of the next preceding item, rests upon a
single instance. In lava (307) we find Saa laha and Wango raha.
The occurrence of this mutation is in the southern Solomons on
either side of the straits which part San Cristoval and Malanta,
not far from the northern k-h focus.

P-h. This rests upon the single instance of vula (284) in which we
find Wango hura. This lies within the focus of the preceding
item.

2. We have, then, two distinct and distant foci in which there is a ten-
dency to reproduce certain of the Proto-Samoan consonants by the aspira-
tion. The islands on which this occurs are large islands, with the possible
exception of Ambrym, which is near the dividing line between the large and
the small.

By far the larger group of the anomalies in mutation is that in
which there is clearly a passage from one series to the next in order.

3. Lingual to labial:

L-ra. This rests insufficiently upon the single instance of mala ma
(322) light Bierian ma ma ma.

N-m. This rests on a single doubtful instance, arms (351) to spit, in
which we find Nggao misu and New Georgia kamisu. It occurs
in the central islands of the Solomons.

T-m. If this be a valid mutation it rests upon but a solitary word,
tama (217), which in Omba, Gog, Alite and New Georgia becomes
mama, and in Merlav, Lakon, Pak, Sasar, Vuras, Mosin, Alo
Teqel, Motlav, Volow and Norbarbar mam. The argument for
this mutation will be found in the note upon this item in the
systematic study of the data. This again has two foci: one, the
Banks Group (omitting Mota), dipping down to Omba in the
northern New Hebrides; the other in the central Solomons.

136 THE POLYNESIAN WANDERINGS.

T-w. This change rests upon the single instance of the same tama,
in which we find Fagani wama. As its immediate neighbors in
the central Solomons, Ulawa, Wango, Saa and Bululaha, have the
decapitated ama this may amount to no more than an obscure
mouthing of the same form.

4. Labial to lingual :

M-r. This occurs in malama (322) light Bugotu Nggela marara,

neighbors in the central Solomons.
M-w. For this we have no less than five examples, four Marina and
one Tangoan Santo, our scanty vocabularies not allowing us to
coordinate the two languages even in a single example. These
are: lima (312) five Marina Una; lima (313) hand Marina hna±
manu (317) bird Marina nanu; mala (324) eye Marina nana; malu
(316) gentle Tangoan Santo nalum. This mutation is restricted
to the island of Espiritu Santo in the northern and leeward New
Hebrides, a large island.

M-2. Found in a single instance, manifinifi (254) thin Malo tanivinivi.

\-t. In vai (291) water, we note the doubtful case of Marina tei, an
alternative with pei.

F-n. This rests upon the single instance jua (360) fruit Nifilole nua,
and, as will be found in the systematic study of the data, a more
consistent explanation is probable.

F -s. This is based on nifo (259) tooth Pala ngise, and fafine (290)
woman Eromanga sivin. I incline to see in this, first a mutation
i-h, which is common in Polynesian and is found elsewhere in
Melanesia (Ambrym whin), then a secondary mutation h-s, which
is extremely common.

F-*. This is found as a variant Mota form in the same word fafine
Mota vavine and tavine; perhaps in fia (218) how many Moanus tje.

f-th. Occurs once, in fano (147) to go Tangoan Santo thano.

PL Such a mutation is doubtfully suggested in papa (279) a board
Aneityum apalapal, thin, fiat.

5. Lingual to palatal :

L-k. This rests upon lima (313) hand Vaturanga kima Nggao kame

Aneitvum ni-kman, and talinga (250) ear Aneityum tiknga.
L-ng. This rests doubtfully upon the single instance lalo (123).

6. Palatal to lingual :

NQ-mi. It appears solely in talinga Natalava kidinda.

The second and smaller group of these anomalies is made up of
the mutations from and to series two removes away.

7. Palatal to labial :

NG-m. This rests, abundantly supported, upon the single instance
of ngata (199) snake Santo mata Efate mwata Mota mata Malo
moaia, all probably equivalent, and a variant Efate mata is equiv-
alent with Santo mata Bierian n'mata Malekula na-mat.

K-m. If this be valid it must rest upon kill (304) Motu miri, with the
added disadvantage that Melanesia affords us almost no identi-
fications of this word. The m in kati (302) Bierian mkati follows
a line quite other, it is clearly akin to the prefacing of the mutes
in Viti ; this is found again in keli (297) Baki Bierian mkili.

POLYNESIAN RELICS IN MELANESIA. 137

K-v. For this we find the single instance of kidl (194) clog Malo vuria
Tangoan Santo vuriu. It will be observed that the area of this
leaping mutation is included in the but scantily more extended
area of the similar ng-m mutation.

K-w. This very interesting mutation, not only a leaping mutation
but as well from a consonant to a semivowel, while most firmly
validated, rests upon the single instance of kutu (306) louse Omba
(etc.) wutu Merlav (etc.) wut Pak (etc.) wu. The area of its
occurrence includes all of the ng-m k-v area of the Banks Group
and the northern New Hebrides, with a yet further extension in
Sesake to the central tract of the latter archipelago.

8. Labial to palatal :

Ni-ng. In a single instance, malum (316) Lo melunglung Alo Teqel
mulunglung. It lies in the same Banks Group area of anomalies.

F-ng. For this, too, we have but one example, fiha (215) how many
Nggao ngiha Vaturanga ngisa. It lies in an area of marked
anomalies, in fact with the k-h and n-m areas it completes a
triangle in southern Ysabel and New Georgia and northern
Guadalcanal' of the Solomons, within which lies Savo of a far
different linguistic character.

F-k. This rests as yet uncertainly on ufi (273) yam Tanna nuk, and
then only if n function as article with a substantive uk, n-uk.
We note in uncertainty ifo (206) down Aneityum suko Gog sug.
The mutation in nojo (201) to sit Nggao nokro is wholly abnormal.

9. Finally we have left for consideration a half dozen variants which

elude the foregoing effort to find order in irregularity.
NQ-nj. Noted in matangi (274) the wind Aneityum ni-mtinjop. It

is possible that m(i)tin is the survival of the stem matangi and

jop an accretion of some sort ; I have been unable to identify jop

as an independent word or in other composition in Aneityum.
N-ny. This occurs in namu (328) mosquito Aneityum inyum Moanus

njam. Buka and Bugotu have already been noted as varying

this n to gn.
S-y. In sulu (182) torch Baki yidu.
T-dr. Found in to'a (375) to subside Ambrym dro dru to abide;

tutulu (367) to leak Baki drudruli. Associated herewith is t-ndr

talinga (350) ear Moanus ndrilinga, tahi (352) sea Moanus ndras.

In this reinforced r we have elsewhere found evidence of the

effort to reproduce r grasseye.

The result of this inspection of the anomalies in mutation is that
we identify two distinct areas in which Polynesian material was
rudely subjected to purely Melanesian methods, one area somewhat
diffuse in the northern New Hebrides archipelago, including the
Banks Group and Torres Islands, the other sharply defined on the
larger islands in the mid Solomons; and that in at least the two
more numerous of these anomalous mutations there seems an inter-
relation between the two areas.

If the result of all this painful examination were no more than the
circumscription of these two small areas, interesting as that result

138 THE POLYNESIAN WANDERINGS.

might become when the material for Melanesian study is more
abundant, we should judge, and rightly, the time and labor ill spent
which brought no better returns.

But I feel confident that this material, thus handled, does unfold
to us the log of the Proto-Samoan swarm and does prove to us that
this migration, at least, followed the Melanesian course quite regard-
less of the wind-and-wave argument on which Dr. Thilenius has
expended so much attention. It will be observed that these studies
have identified Proto-Samoan elements in Melanesian, and scarcely
other than Proto-Samoan. The Tongafiti swarm does not appear.
That must be left for later study : first, its segregation in Polynesian
philologv, then its identification in whatever travel lane it may have
followed It may be that it can be identified in that mid-Pacific
track which Thilenius proposed . This much is certain, the Tongafiti
migration has left absolutely no trace of its passage in Melanesia.

On the sieve hypothesis, namely, that the Polynesian content of
Melanesia is due to drift of castaways from Nuclear Polynesia, we
should look to find such content strongly localized at those points
more immediately to leeward of the point of involuntary departure,
that is in Viti and the New Hebrides. Yet as to Viti the Samoan
record is clear. It was not drift of castaways, it was a long series
of purposeful voyages from Samoa to Viti, from Viti to Samoa,
for love and for war ; it was such a common voyage that the sisters
Tilafainga and Taema swam it. And as to the New Hebrides,
where the Polynesian content by this theory should be at its best,
we have just proved the existence there of an area in which the
Polynesian is at its poorest. Furthermore this Polynesian element
is found quite as strongly, in fact more strongly, in the Solomons
and yet more northern groups quite outside the normal course of
drift, so far as I am able to identify it upon the charts with the aid
of no merely theoretical familiarity with the winds and currents of
this western Pacific.

Let us rather examine our data in the light of what might be
expected of a great ethnic swarm, and not the feeble struggle for
life of fishermen landed in distress upon inhospitable shores. Let
us set before ourselves the manner of such voyaging.

Under the stress of some expulsive force acting upon their rear
in Indonesia, under the draft of some force leading out into the
eastward unknown, the Proto-Samoan fleets passed through some
one or more of the free channels out of the Malay seas. They were
navigators, for, as I have already had occasion to remark, we can not
deny them their ability to sail the seas to Nuclear Polynesia, while
granting, as we must, their ability to sail voyages of equal length
out of Nuclear Polynesia to yet ulterior eastern lands. Samoa was
no dockyard, it was no school of navigation; Bougainville's name

POLYNESIAN RELICS IN MELANESIA. 139

of Navigators' Islands for that group had no deeper signification
than that he found his ships surrounded by a fleet of canoes.

We know the type of these vessels. Discoverers have described
them in the accounts of their South Sea voyages ; sketches of them
there are a-plenty. I have seen the last in Samoa of the type of
double canoes with sails fit for ocean-going. We know that each
could carry its hundred or so of passengers, could eat up into the
wind and lay a course almost as close as the fore-and-afters which
are the American contribution to the marine. The one principal
defect in these vessels as the vehicles of long voyages was in the
victualling, and that defect produced a system of voyaging with-
out which we should be at an utter loss to prick their course upon
our charts.

Each of these voyages was an Odyssey. Stocked with such food
and water as they could find the means to carry they coasted wher-
ever coasts were available to follow, and thus they voyaged until
the commissariat called for replenishing. Then they landed, they
established, albeit temporarily, food colonies until the land could
yield them a crop sufficient to carry them yet farther, until the same
ventral need established them yet again in a like food colony.

These revictualling settlements are of the utmost moment in our
study. Three elements are primal in the establishing of each such
settlement. It must have a sufficient supply of water ; it must show
an encouraging area of soil fit for tilth; its autochthonous population
must be such that the voyagers might feel secure of maintaining
themselves and their families during the months of the crop period,
whether by superiority in numbers or by better skill in warlike arts
is immaterial.

In general the supply of potable water would be found ample
wherever the two other conditions were satisfied. In the whole
western Pacific area there is a wide contrast between two types
of islands mingled in close juxtaposition. The large islands are
commonly high, great masses of volcanic extrusion with forbidding
shores and little productive soil in sight save in small patches in deep
bays. A race in whom the ethnic sense had reached such a high
stage of development as to send them forth in company as these
Proto-Samoans swarmed, would naturally expect that the large
population of a large island would assemble in concert at the point
of attack to repel the invader. The small islands are commonly
low; their acreage is greater up to the visible forest, this being an
important criterion, for visible possibilities of tilth point to a neces-
sary sojourn of but one crop season ; to clear the jungle for plantation
would require three and probably more seasons. The population of
a small island, even if aggressively hostile, would be more within the
control of the adventurers of a single vessel or small squadron. We

140 THE POLYNESIAN WANDERINGS.

should look, then, to find the Polynesian element most strongly-
marked in these smaller islands rather than on the adjacent main
where Dr. Thilenius is so insistent in pointing out their absence.

Remember the Proto-Samoans are voyaging under sail.

What must a Nelson have before Trafalgar can be his day ? The
weather gage. Dreading the perils of a lee shore, does not every
sailor hug the wind ? There is in this no deep ethnic principle which
we are called upon to establish as in the possession of our Polynesian
swarm. It is sufficient to know that they were under sail upon the
sea; the rest follows as an elemental principle of the mechanics of
seamanship. Thence we shall do well to look for them to windward
in the lands along which they pass, less to leeward.

Coming from Indonesia into the Pacific, the coasting voyage along
the rugged heights of the Solomon Islands, which rise in the sea from
Buka to San Cristoval, set for these voyagers a course approximately
southeast as we now should lay it by compass, full and bye when
reduced to the rhumb of the wind prevailing during the months
which would be found most favorable for navigation. After 600
miles of navigation by landmark in this great chain which with
consistent uniformity has coincided with a full and bye, what more
natural than that, when the last landmark has sunk astern and the
open sea is to be adventured, the pilots still should follow the
course set for them by the wind? Next upon the track thus hugging
the wind lies the Santa Cruz Group. Beyond this small archipelago
lies yet another void of the sea, always the sea they must have loved,
always the constant draft of the trade wind which hitherto not only
had carried them on their course but had laid for them that success-
ful course. We must never lose sight of the fact that it was only
those who followed this course full and bye who found the chance
to survive — a few points off and the voyage was protracted in pain
and ended mutely in starvation and thirst upon an empty sea which
marks no memorial of the manhood it takes.

From Santa Cruz the intervals of the sea which we know are
uneven. Working on the wind there lie a thousand miles of all but
unbroken ocean before the next landfall. To the south — near, yet
out of eyeshot — lie the New Hebrides. With chart and compass we
can find the nearest land, to seek or to avoid as may best suit the
purpose of our voyaging. But these navigators of the Polynesian
swarm had no knowledge of what land might be in the unknown sea.
For many leagues the course set for them by the unchanging wind
had led them coastwise where land was, and when the sea grew
empty the same sailing track had led them on to yet new land.
Thus may we reasonably expect that taking their departure from the
last sight of Santa Cruz the fleets would set bravely forth upon the
course that so long had served them and so well.

POLYNESIAN RELICS IN MELANESIA. 141

I feel that I can not set too acute an accent upon this idea of course.
Without knowledge of what might lie before them, with no chart
and with no compass to guide them, even had they known what they
sought, there was but one fixed and recognizable fact in empty sea
under the cloud-flecked emptiness of sky. This fact was direction,
the angle with the wind at which their canoes were at their best
sailing speed. Where all else was uncertainty the fixity of this fact
must have kept them true upon the sea by night as well as by day,
for in the darkness, when the eye could no longer see the tremor of
the after leach of the great mat sail bellying above them, the ear
could be warned by its quivering. All else uncertainty, this alone
was fact.

The New Hebrides, therefore, lying so near would yet be distant
because out of course. Upon the shores from Norbarbar to Aneityum,
still more remote and still more to leeward, in Uea, Lifu, and Mare, at
the most remote spot to leeward beyond which lay no land whatever,
at New Caledonia, would come only the dull sailors and those who
through blast of gales had sagged down the wind. Therefore from
the point where the axis of the land masses breaks from its northwest-
southeast direction and sets off north-south we should expect
to find a difference in the Polynesian content of the indigenous
languages.

So far in these notes our attention has been given to the direction
of the Polynesian traverse through Melanesia. We may pause
briefly to consider a point of relative duration of this traverse, and
we note with surprise that Thilenius has permitted himself to write
of a measure of weeks.

Much earlier in this work I have noted the instances of Polynesian
inclusions within Melanesia. Stated in terms of the point as now
presented, these are cases where the duration of the traverse has
reached the absolute maximum in a fixed and permanent settlement,
a relinquishment of the voyage.

Remember that we have no means of determining what was the
impulse upon which these voyages were undertaken. It will, how-
ever, involve no great strain of the probabilities if we assume as
established the reasonable hypothesis which has been proposed, that
the impulse upon the Polynesians commorant in Indonesia was an
expulsive force and that it was applied upon them on their exposed
northern flank and upon their rear equally exposed to the crowding
of swarms of alien and incompatible migration from the Asiatic main-
land. Upon this assumption we may naturally draw the conclusion
that the power of expulsion had practically vanished when the
Polynesian swarm had set the great island of New Guinea behind
them. This we know of a certainty, in all the unknown ages which
have elapsed, the Malayan peoples (if it were their ancestors who

142 THE POLYNESIAN WANDERINGS.

crowded out the Sawaiori) have been able to effect no lodgment
of settlement upon that dark island, and their trade settlement has
scarcely advanced beyond the occasional raiding dash of the sea
rover and slaver.

Once in Torres Straits, for such as found the exit in the Arafura
Sea, once within the great bight of the Bismarck Archipelago, for
what seems to have been the main flight, the voyagers lacked impulse
as they certainly wanted direction. They were afloat, but they were
headed nowhere in particular. The world may have been in a sense
before them, but choice was limited to such lands as they might
chance upon. I have already spoken of the necessity of crop settle-
ment as a condition of their naval economy. Without destination
to lure onward, without force behind to drive them yet farther
along, such crop settlement under favorable conditions of soil, water,
and subduable autochthons tended inevitably to become a perma-
nent colonization.

If we note upon the charts the position of such Polynesian inclu-
sions, crop settlements become fixed colonies — the islands of the
western verge of Polynesia in Nuguria, Tauu and Liueniua, Matema,
Ticopia, Sikayana, Mai, Aniwa, Fotuna — we shall find them with-
out exception the windward islands of the archipelagoes with which
respectively they are associated in descriptive geography. Even
Rennel and Moiki, although they lie to leeward of San Cristoval
in the southern Solomons, are yet a weather ly achievement to such
voyagers as issued easterly from Torres Straits. Where we find
these fixed colonies in such number there must have been other crop
settlements similar in their beginnings which endured in the measure
of years or perhaps generations until inability to withstand the
assaults of the indigenes or the lure of some new squadron of
wanderers of their own race and speech led them to essay yet
again the great sea, never forgetting that it is inborn a character-
istic of the Polynesian to hold himself proudly the master of the
ocean.

If these considerations are to be held somewhat of more worth
than the divagations of fancy there must somewhere be some record
to give them substance. Where else, then, than in these speech
records which we have subjected to such minute analysis?

But how to make the record appear?

For Mota, for Aneityum, and for Efate we now have dictionaries
of unequal excellence. These languages we may, therefore, compare
with the Polynesian languages for which we have similar standards,
indeed regulate by so much of the comparative Polynesian phil-
ology as has been elaborated upon these data. For so much we
are thankful, yet these are but three, and in the data upon which
these studies rest we have had under intimate dissection no less than

POLYNESIAN RELICS IN MELANESIA. 143

ninety languages in Melanesia. Had we dictionaries of each of the
ninety we might easily note the exact percentage of words which
mark the Polynesian content in each. Then our comparison would
be as exact as it would be facile. Unfortunately there remain
eighty-seven languages of which our knowledge rests upon a very
few words preserved at random in many works of reference. The
exact measure we seek is impossible of application.

Yet we are by no means left without recourse. There is at
our hand a certain measure of the quality of the Polynesian inclu-
sions in these Melanesian tongues. The identifications have been
made with all the assistance which can inhere in long practical
acquaintance with the comparative Polynesian philology. I have,
accordingly, had no hesitation in definitely accepting as Polynesian
identification many a word which would completely fail of recog-
nition by the Polynesian of any one tongue. This is due to the fact
that the Melanesian, alien to the Sprachgeist of his loan material, may
deal with the Polynesian word which has come into his possession
according to the spirit of his own speech. Thus the Samoan who
says fajanga, to feed, might quite fail of recognizing his mother
tongue when the New Irelander had trimmed here and added there
to make Ivambell angan; and his familiar longo, to hear, would be
wholly inaudible to him in Marina rogotag. Yet there are words
which he could comprehend.

I have therefore taken as criteria of this measurement the words
in each Melanesian language which a Samoan, knowing only Samoan,
could comprehend if he were set down on the alien shore amid a
hostile folk under circumstances where every instinct of life would
fill him with anxious desire to know from the strange sounds what
disposition was to be made of him. Not the learned pursuits of
a philologist this, but the working of the wits of a man under the
compelling stress of elemental need. I assume that his ears thus
pricked up would gather those words which exist in common in his
language and that of his savage hosts, and there are many such,
even where the final vowel has undergone abrasion. Further I
assume that when the consonant structure of any given word remains
the same in the two languages his wits would be sharp enough to
recognize the word when the vowels had undergone modification;
thus Epi fefene would be easily comprehensible to the Samoan who
says fa fine for woman. The same will hold when the vowel structure
is constant ; thus Ulawa nimanima could not but be comprehensible
to the Samoan who knows his hand as lima. Finally, in those cases
where the vowels remain unaltered and the mutation of consonants
is not at variance with the system of mutation normal to the Polyne-
sian languages, I feel justified in the assumption that the strain of
need would awaken our Samoan to a conscious recognition of his

144 THE POLYNESIAN WANDERINGS.

own word in its new guise. In effect I set before the individual the
task which has proved easy to his race.*

For each of the languages of sufficient representation I have tabu-
lated the results of this examination and have set them down in the
foregoing tables. Employing the sum of these several elements as
dividend, and for divisor the whole number of words in the available
material which have been identified as Polynesian, we obtain a
figure which stands for the quality of the resemblance of each lan-
guage with the Proto-Samoan, and this coefficient of quality has
been inserted in its proper place in the tables.

Now for the graphic presentation of the results. Upon the chart
of the island-studded ocean between the termini of Indonesia and
Polynesia, respectively, we insert these coefficients of quality and
delimit the areas of equal resemblance.

*Not a single detail of such westward drift and its result can be lacking in interest
to us. Therefore I note the following double instance from the Rev. John Inglis's
"In the New Hebrides" (page 33) :

"On Aneityum the idols were all, like the Jewish altars, of uncarved, unhewn
stones. The only exception to this which we ever found was in the case of Tuatau,
a natmas which I found at Anauunse and took home with me. Tuatau was of wood,
a piece of a breadfruit tree. Like the idolaters mentioned by the prophet, the maker
of this idol had chosen a tree that will not rot. It was a rudely shaped, uncouth
figure, its countenance only very slightly resembling the human face divine. I was
struck with its being made of wood, and afterwards learned that it was not a native
idol— it was of foreign manufacture. It had a little history of its own, which may serve
to illustrate that of ' the image which fell down from Jupiter '(Acts xix, 35). About the
beginning of this [the nineteenth] century as nearly as native chronology supplied me
with the date on which to calculate, in the days of Tuatau, a great chief of Anauunse,
this poor idol was one morning found drifted ashore by the northeast wind. How long
it had been tossed upon the ocean nobody knew. But as it bore all the marks of a
Malay idol [Dr. Inglis thus denominates the Polynesians], and was very like the fisher-
man's god of Rarotonga, as given in Williams's ' Missionary Enterprises' and which, he
says, 'was placed on the forepart of every fishing canoe; and when the natives were
going on a fishing excursion, prior to setting off they invariably presented offerings to
the god, and invoked him to grant success' — it seemed highly probable that this idol
was a Rarotongan fisherman's god— that the canoe on which it was borne had been
wrecked— that the poor fishermen had been drowned— and that the idol had been drifted
along before the tradewinds till it was cast ashore on Aneityum. But, be that as it
may have been, its subsequent history was well enough known. Among a people
remarkablv unskilled in the pictorial arts, its faint resemblance to the human form
secured for it favor and veneration. The day on which it was found was one on which
Tuatau was making a great feast. The natmasses were always closely connected with
the feasts. It was one of the fundamental articles in the creed of heathenism on Anei-
tyum that the man who made the largest feasts and who presented the most costly
offerings to the natmasses was the man that most effectually propitiated their favor.
The sacred men all declared that the natmasses had made this image and brought it to
Tuatau; and the chief and the ignorant populace accepted the statement as readily, and
believed it as firmly, as the Asiarchs and the idolaters of Ephesus believed that the ugly
little statue, made of ebony and vine wood by Canetias, was, as the priests of Diana
affirmed it to be, 'the image that fell down from Jupiter.' The chief received it as a
token of the special favor of the natmasses, placed it within the sacred enclosure, and
thenceforth regarded it as his tutelary divinity. After the death of Tuatau the idol
received his name, and was supposed to be watching over his spirit; and it continued
to be worshipped till Christianity was accepted in Anauunse.

"Had the idol been a man — a shipwrecked sailor, or one of the poor fishermen on the
prow of whose canoe it sat conspicuous as Castor and Pollux did in the ship that carried
Paul — to a certainty he had been killed, and most probably also eaten; at least a ship-
wrecked sailor met with this sad fate at Eromanga within less than a twelvemonth of
the time when Tuatau fell into my hands; but being a block of wood, shaped so as to
have a faint resemblance to a man, it was set up and worshipped as a god."

POLYNESIAN RELICS IN MELANESIA. 145

As soon as we trace the contours connecting the islands whose
languages show approximately equal coefficients of Polynesian qual-
ity it becomes at once manifest that we are concerned with two
great areas. These are so distinct, so widely separated, that it is
practically impossible to conceive of them as having the same origin,
at least in this Melanesian tract of the Pacific with which we have
to do. We shall therefore do well to consider them in detail, each
by itself, and to reserve the discussion of their diversity of origin
until after that preliminary survey.

One of these areas embraces southern Melanesia, from the Torres
Islands to New Caledonia. In these islands, the New Hebrides form-
ing the principal and determining mass, the contours of equal quality
extend from northwest in a general direction toward the southeast.
Along the axis cutting these contours from northeast toward south-
west we find that the coefficient of quality diminishes from the east.
In the southern New Hebrides we find two instances of purely
Polynesian speech, at Aniwa and at Fotuna, these representing the
position of the contour of ioo. It is important to note that Aniwa
and Fotuna are the most weatherly of the New Hebrides archi-
pelago, that is to say they are the points to be reached by a fleet
steering full and bye, the best sailing point of canoes and the only
sailing point which gives the helmsman on unknown seas a sense
of direction for his course. The three larger islands in this section
of the archipelago show scant traces of Polynesian admixture, and
the quality coefficient is low; Eromanga 58, Aneityum 46, and Tanna
no more than 29. Still farther west the Loyalties make no better
showing: Nengone 45, and Lifu 42, while Uea records a Polynesian
content entirely of the modern epoch and known to be derived
from an involuntary voyage from Uvea in Nuclear Polynesia.

In the examination of the central and northern New Hebrides,
for except in the history of discovery it is not advisable to disso-
ciate the Banks Group, we are able to draw contours of 90, 80,
and 70 quite plainly. Along the windward face of the archipelago
appear spots which might establish the curve of 100, such being the
Polynesian settlement in Efate and Mae on the island of Three Hills.
At the north the curve of 90 is established on Vanua Lava by Leon,
the other languages of that island standing at 40 and the bush
language (Alo Teqel) at the lowest mark of 19. Working down the
weather aspect of the archipelago, Arag falls but little below this
contour and Makura, Nguna and Efate lie above it. Mota in the
extreme north shows the same influence as Leon on Vanua Lava.
The contours of 80 and 70 are satisfactorily drawn in close paral-
lelism with that of 90 and are well established by a sufficiency of
points of identification. The high value of Marina 89 in the deep
bay of the north coast of Espiritu Santo need not prove as anoma-

146 THE POLYNESIAN WANDERINGS.

lous in position as at first sight it might appear. While Vanua
Lava-Mota-Arag identify one point of migration entrance into the
archipelago, it is quite probable that Marina- Arag identify another
entrance for fleets sailing a little to leeward when making this
landfall.

We next examine the northern area. Its curve of ioo is beauti-
fully established along a great length ; from southeast to northwest
we note Anuda, Ticopia, Matema, Sikayana, Liueniua, Nukumanu,
Tauu and Nuguria, all Polynesian communities, all the most weath-
erly islands of the Santa Cruz Group, of the Solomons, and of New
Ireland respectively. At the other limit the curve of 60 may dis-
tinctly be traced along the leeward faces of the Solomons and of
the Santa Cruz Group. Between these two well-marked contours
the curves of 70 and 80 are very distinct in the Solomons and in
the strait between New Britain and New Ireland. Along the range
of the Solomons the migration track is plainly drawn along the
weather coast or in the easy channel from southern Ysabel toUlawa.

We may now examine the points at which the migration streams
which establish these two areas come into closest approximation.
These points are two, the interval between Deni and the Torres
Islands, and the interval between Moiki-Rennel and the southern
Solomons. Here, if anywhere, the two streams came their nearest
to a chance of mingling.

In this examination we must bear in mind a few clearly estab-
lished facts. These voyages were performed without chart or com-
pass. The leaders knew nothing of what land might lie before them.
Their only norm of direction was set by the constancy of the trade
wind and their only method of conforming to that norm was by
sailing closehauled, to which they had the added inducement of
seamanship in that this was the best sailing point of their canoes.
To deviate from that course where no landfall had been made and
where there was no knowledge that land might exist would be to
relinquish a purposeful voyage for merely idle cruising.

In the case of the gap south of Deni intermingling could have
taken place only from the north toward the south. It is incon-
ceivable that a fleet having made the most northerly landfall of the
New Hebrides, whether at Lo or at Vanua Lava, should leave the
new lands already in sight to beat dead to windward where no
land was known to exist. Similarly there could be no reason for
vessels taking their departure from the Santa Cruz group to leave
the one course which they knew, to set out toward the south which
was a direction they had no means of determining, and to run free
upon their least convenient and most dangerous point of sailing.
At the other point of approximation we should note that Moiki
and Rennel are invisible from the nearest islands, Guadalcanar and

POLYNESIAN RELICS IN MELANESIA. 147

San Cristoval, and that no land is visible from them. They are
islands having the speech quality coefficient of ioo, for they are
settled by Polynesians. Such islands we have already learned to
look for as the most weatherly achievements of Polynesian voyagers
in this great migration movement. In the southern Solomons the
course of migration is distinctly marked to windward of Malanta
and through the island-dotted channel north of Guadalcanar and
San Cristoval. With all these landmarks to point the way it is
inconceivable that canoes should leave the coastwise course and
head to leeward for islands far beyond their sight and wholly out
of their knowledge. Likewise for voyagers passing beyond Moiki-
Rennel the closehauled course would not carry them within sight
of San Cristoval, but would give them a more distant landfall in
the Torres Islands, and thence the land in view would deflect them
southeastward on a coasting voyage.

It is quite clear, therefore, that these two points of approach were
gaps not crossed and that the two streams of migration remained
largely distinct. The dull canoes of the northern stream, a few set
to leeward by gale or other accident, may have reached the southern
stream and have escaped notice ; but that there was any accretion to
the northern stream from the southern is wholly out of the question.

No account has yet been made of the two western points of this
identification, Moanus 83 north of New Guinea, and Motu 85 in the
Gulf of Papua on the southern coast of that great island. With
the mass of this almost continental island beween them these two
distant points of equal quality must stand apart. Each represents
the most westerly identifiable point of a migration swarm and these
two swarms must have been wholly distinct. Moanus I regard as
the first point of the stream which in part went to windward of
New Ireland and in part has left its traces in St. George's Channel,
thence has swept along the Solomons, thence past Matema and
Ticopia and onward to Rotuma, and still beyond to its lodgment
in Samoa — the Samoa Stream. Motu and Moiki likewise establish
early points on the migration track which generally parallels the
Samoa Stream, but runs some distance southward until it makes
the landfall of the northern New Hebrides and then is deflected
sharply south by the opportunity and the convenience of sailing
coastwise with its double joy of war and victual, which sets forth
once more upon empty sea from Aniwa and Fotuna and at last
enters Nuclear Polynesia by way of Fiji — the Viti Stream.

Now it comes to us to discuss briefly the relative age of the
Polynesian content identified in Melanesian possession and the same
material in Polynesia itself.

Of two forms in general, one with a final vowel and one with a
terminal consonant following the same vowel, which in all proba-

148 THE POLYNESIAN WANDERINGS.

bility is the elder? As between two great groups of speech, to one
of which the final consonant is most repugnant, to the other of
which the final consonant is so distinctly pleasurable that in many
parts of the group a stem final vowel is abraded in order to reach
a closed syllable, in antecedent probability which of these repre-
sents an earlier type?

We do not have to rest on the antecedent probability, clear as
we shall find it. That the form in the terminal consonant, the
characteristic Melanesian form, is the earlier and elder is shown us
in the Polynesian itself. Look at valu (281) to scrape taro; Efate
bar u- si Mae barusi, of the same sense, might lead us to infer that
the root form is barus; in Samoan valusanga, the derivative meaning
taro scrapings, we find a direct proof that the >? is radical, for the
formative suffix was applied early enough to protect the final con-
sonant from abrasion. Examples abound in these data; I note but
these few for the reference of such as wish to give the topic extended
study: 71, 81, 160, 162, 191, 204, 224, 227, 266, 283, 289.

This, too, plays an important part in the condemnation of the
sieve theory and the castaway drift. Which is more reasonable,
that a dozen very closely allied languages should act harmoniously
in dropping final consonants, or that ninety languages with very
scant community and no intercommunication should agree in adding
precisely the same consonant as closure to open roots brought them
in storm-driven canoes, and that the same terminals should be picked
up in several cases by the score of languages in Indonesia, a region
physically exterior to all such possibility of canoe drift? There can,
indeed, be no shadow of doubt that the Melanesians in their keeping
of the Polynesian loan material have preserved an earlier type.
Therefore the conclusion is inevitable that the Polynesians were
commorant in Melanesia at some time, and for some time, anterior
to their settlement of the unoccupied lands of the central and east-
ern Pacific.

We are dealing in these studies with the record of what the Poly-
nesians taught the Melanesians, no inconsiderable contribution in
the aggregate. But did the Melanesians teach the Polynesians
nothing? Was the gift altogether so one-sided?

In the rigid examination of the material I can find but a single
word which I suspect to have come into Polynesian possession from
Melanesian tongues. This is Samoan 'isumu (251) rat, and it calls
for no little agility to identify this word, not elsewhere Polynesian
and by no means in common Samoan use, with the word which
means rat in certain parts of Melanesia, namely in the New Hebrides,
and in the Solomons in just that area of the islands of the Malanta
channel which marks our curve of maximum quality of Polynesian
content.

POLYNESIAN REUCS IN MELANESIA. 149

The reason is by no means far to seek.

Homekeepers have ever homely wits. Our Polynesians, who had
the genius to conduct such voyages as under far more favoring
conditions have brought immortality to Quiros and Mendafia, to
La Perouse and Dumont d'Urville, to Magelhaens and Roggewein,
to Cook, to Byron and to Bligh, to Wilkes — such were a folk far in
advance of the rude autochthons of the islands they encountered in
their passage. The black man could learn much from the brown;
there was little in the liberal arts that the Melanesian could com-
municate to the bright, aye the brilliant, Polynesian. Where they
came into contact could only have been in the crop settlements,
and in such it must have been an essential condition that the black
sat in subjugation to the brave brown sea-rover. What does the
slave in any community teach to his lord which comes into the
prevailing speech? One word in these two hundred is all we can
suspect the Polynesian to have taken from the Melanesian, and that
one very doubtful. Yet who of us without painful research can
identify for so much as the five fingers of but a single hand a word
apiece which the Britons have set into the language which is ours
by right of the conquest of Britain by the Roman, the Saxon, the
Norman?

Yet one more reason is simultaneously and equally operative.
Crop settlement approximating semi-permanence may have taken
place piecemeal among ninety varying languages. Such Melanesian
elements as each settlement might have permitted itself to adopt
in the particular spot of its sojourn would be incomprehensible to all
other members of the migration swarm who had sojourned in con-
tact with each of the other eighty-nine languages. Upon their reas-
sembling in Nuclear Polynesia the alien elements comprehended but
by the company of a single vessel would be restricted in compre-
hensibility to that crew alone, and, thus from the beginning limited
in use, would tend toward disuse and Polynesia would know them
no longer; the superior language would heal its own wounds.

Before leaving this central chapter in which we have discussed in
many lights the Polynesian content of Melanesian speech I wish to
sum up the major conclusions to which we have been led.

i . In the plexus of Melanesian speech a certain element has been
proved to have a common origin with the Polynesian.

2. That this varies in quality according to the ability of its
Melanesian possessors to respect the vital principle with which it
came into their possession.

3. That this represents, wherever the data admit of deduction,
a phase more primitive than the Polynesian of the eastern archi-
pelagoes.

150 THE POLYNESIAN WANDERINGS.

4. That it can not be drift material brought by eastern castaways,
but is derived from the sojourn of Polynesian ancestors in Melanesia
when on their way from Indonesia to the mid-Pacific.

5. That from the material here involved in discussion we can lay
out one great track of the migrant swarm with much precision, and
that a second may be laid out with considerable probability.

6. That along these two tracks the only Polynesian voyagers who
have left any trace are those of the earlier Proto-Samoan swarm,
and that the wanderings of the later Tongafiti migration must be
sought elsewhere than in Melanesia.

CHAPTER VIII.
SAWAIORI MATERIAL IN INDONESIA.

Limitation of the points of inquiry — Check-list of the Indonesian
material — Synoptical tables of mutation varieties — Mutations com-
pared with the systems of the Pacific languages— Character and prob-
able place of the contact of Indonesian and Polynesian — The nature of
an ethnic swarm discussed — The Malay advance was an affair of
outposts — Whence arose the speech community, which after all is a
matter of but a gross of words — The Indonesians are shown to be
borrowers — Two lines of Sawaiori escape through the Malay Archi-
pelago lead to the two tracks identified through Melanesia — The
designation Malayo-Polynesian should be discarded because false.

The scope of this work does not include such a detailed dissection
of the Indonesian languages as that to which the Melanesian tongues
have been subjected in the foregoing chapter. This speech area
has its own diligent students, and to their researches we owe the
present advanced state of our knowledge of the multiplicity of
Malaysian speech. Thus are we spared the necessity of pathfinding.

In the present chapter we shall limit our attention to the con-
sideration of such Indonesian material as is brought into comparison
with the data from the Pacific areas here under discussion. We
shall examine it for the purpose of discovering to what extent it
may be used either in support or in disproof of the theory that the
Pacific languages have developed out of the Indonesian, or that both
derive from a common parent. In this we shall develop whatever
support such examination may give to the theory that this common
parent was Semitic. We shall be led to a rigid consideration of the
validity of the older consociation of these speech areas as the
Malayo-Polynesian family.

Beyond these several points of inquiry we shall not advance. We
shall do no more than to place these data conveniently at the service
of students of Indonesian philology.

As we have done in the earlier chapters we present a series of
tables for readiness of access to the material here assembled. For
a large amount of the Indonesian material indebtedness is gratefully
acknowledged to the industry and research of Mr. Tregear recorded
in his "Maori Comparative Dictionary."

Malay

9

10

27

29

30

31

32

33

34

36

37

45

46

47

79

170

171

173

174

175

176

177

178

179

181

183

185

215

216

218

220

221

222

223

224

225

226

227

228

229

230

232

233

236

237

239

240

241

243

244

245

246

247

248

250

251

255

256

257

258

260

261

262

264

265

266

268

270

271

272

273

274

275

276

277

278

279

281

282

283

284

286

287
151

288

152 THE POLYNESIAN WANDERINGS.

Malay... 289 290 291 292 294 296 297 298 299 300 301 302 304 305

306 307 303 309 310 311 312 315 316 317 318 321 323 324

325 327 328 329 330 331 332 333 334 335 336 337 338 339

340 341 342 343 344 345 346 347 348 349 350 351 352 353

354 356 357 358 359 360 361 362 363 364

Malagasy 28 35 37 38 46 47 169 171 172 175 177 179 184 213

216 218 219 220 221 223 228 231 232 234 235 236 238 239

242 244 246 247 249 250 253 254 255 256 257 258 259 260

267 272 273 274 277 278 281 282 283 288 290 293 294 296

297 298 301 302 307 308 311 312 314 316 317 318 319 320

321 324 325 326 327 328 329 330 331 332 333 334 335 336

337 338 339 340 341 343 344 345 346 347 348 350 353 355

357 35S 362 363 364

Ahtiago.217 290 291 300 305 306 312 317 324 329 330 333 352 360

Allor....291

Amblaw 217 265 278 291 300 312 316 317 33o 360

Amboyna 190 352

Aru 284

Awaiya.180 217 257 278 290 291 300 312 317 324 330 333 350 352

Baju ...251 257 265 278 291 301 305 306 309 312 317 324 333 335

350 360

Baliyon . 250 269 308 324 335 343

Basakrama 250 266 279 312 350

Batak. . .329

Batavia .243

Batumerah 278 290 291 300 312 313 316 317 324 330 333 350

Beu 243

Bicol....2i4 237 250 274 292 330

Binue . . .343

Bolanghitam 217 278 306 309 312 313 317 324

Borneo . . 300 309 350 360

Bouton..2i7 257 259 290 300 305 306 312 313 317 324 333 350

Brissi West 259

Bual ....278

Bugi ... 47 267 274 277 279 292 308 328 343 344 350 352 360

Buru 291 352 356 361

Catmarian 217 265 278 290 291 300 306 312 317 324 329 330 333

350 352 360

Cajeli...2i7 265 278 290 291 312 313 317 330 333 360

Ceram...278 291 329 352 356

Chamorri 350 352

Champa. 291 308 312 313

Dorey. . .291 312

Dyak.309 312 317 324 328 330 350

Ende....278

Gah 217 251 257 26s 275 278 290 300 306 312 316 317 324 329

330 352 360

Galela ..278 352

Gani 278 290 291 298 300 306 312 317 352

Gilolo...284

Goram. .291

Guaham.317 352

Ilocan 267 277 284 290 294 300 323 324 332 334 339 342 346

Java 31 171 182 224 225 251 261 265 267 270 272 273 274 276

277 278 283 285 288 289 290 295 298 299 300 306 308 312

316 321 324 330 333 334 335 336 339 342 344 346 350 352

357 360 364

Jobi 312

Kaili....278 284

Kaioa Id. 291

Kandayan 343

Kavvi ... 250 334 352

Kayan.259 263 273 275 277 278 294 308 309 312 317 322 324 328

344 346

Ki<=a 259 265 273 274 276 278 282 283 284 295 312 313 317 318

323 324

Lampong 312 343

Landa . . . 343

SAWAIORI MATERIAL IN INDONESIA. 153

Lariko..2i7 265 278 290 300 306 312 317 324 330 333 350
Loabnog.:.l84 ^ ^ 29° 291 30° 312 3'3 3'7 » 33o 333 350 360

Macassar 78 216 252 256 267 279 285 290 294 306 308 313 318 324

Madura..^ 305 IE "" ^ ^ ^ ™ 35° 352 353 355

Magindano^ 250 259 274 276 284 294 308 312 313 317 318 321 342

Mame. . .312

Manatolo 259

Massaratty 217 251 290 291 300 306 312 329 3,0 33* 350 360

Matabello 259 265 278 300 306 312 316 317 324 329 333 352

Matu....261 267 276 277 305 317 327 336 344 350 333

Mayapo..2i7 265 290 291 300 306 312 317 329 330 33, 350 360

Menado..2i7 291 300 306 309 312 313 317 324 333 360

Menankabau 350

Molucca . 284

Morella.joo 217 278 290 291 300 306 312 313 317 324 330 333 350

Mysot...2i7 278 290 291 300 306 312 324 361

Nicobar . 300 324 326

Pampangas 276 277 278 309 312 321 324 327 328 344 353

Pangasinan 329

Pani 295

Rotti . . .278 284 291

Salayer .217 275 285 290 306 312 324 330 33* 360

Salibabo 257 278 290 291 312 317 330 360

Salu 278

Sambawa 312 329 343

Sandol..278

Sanguir 217 290 306 309 312 317 329 333 ^2 360

IarPurUa.m V\l 259 29° 29' 30° 312 316 31? 324 333 335 35°

Sassac . .343

Satawal 294 7 so *52

Save... 259

Savu....294 305 317 324 335 343

Silong..29o 291 294 300 323 324 339 350

Sirang . .312

Siwa .... 346

Solor 284 29 1

Sula. ...259 275 278 306 317 350

Iunda:::352 267 290 291 294 3°° 308 3l2 3I3 3I7 324 330 335 350

Tagalog . 45 ,85 214 249 250 259 267 274 276 277 287 295 308 309

312 321 324 327 344 346 350 353

Teluti...2i7 265 290 291 300 312 316 317 324 330 3^3 360

350 360 285 29° 291 30° 3QI 3°5 3°6 3'2 3'3 317 324 ^
Ternati. .276

Tidore...278 295 300 352

Timor... 278 284 312 329

T0D0....217 278 329 330 333 352

TrnTanudS 284 ^ ^ 291 292 294 298 3°8 3'2 3l7 321 324 352

Ulea 350

Utanata 284

Vaiqueno East 259 291

Visayas..l85 274 278 287 292 310 312 317 327 344 350 361

Wahai.217 278 300 306 312 317 324 330 333 350 360

Waigiou.290 291 317 344 361 J 33° °U

154

THE POLYNESIAN WANDERINGS.

It will now be in order to tabulate as in other chapters the refer-
ences to the several mutation varieties.

A.

a-ai

Bolanghitam

309

Menado

309

a-e

Malay
Malagasy

Avvaiya

Baju

Batumerah

216
239
316
35o
35o
35o

250
250

302

Caimarian

Dyak

Lariko

Liang

Mayapo

Morella

35o

309 324

350

350

350

35o

Saparua

Satawal

Sula

Teor

Togean Ids

Wahai

350
350
350
290
292
350

a-i
Malay
Malagasy

183
246

302

312

Saparua
Teluti

290
290

Teor

3i3 324

360

a-o
Malay
Baliyon
Basakrama
Bugi

324
324
312

277

279

Dyak 317 325
Kisa 282
Macassar 350
Menankabau 350

Salibabo
Saru
Togean Ids

278
312
312

a-u
Malay
Bugi

181
35o

224

307

Java
Mysot

224 324
324

Ternati

276

ai-e

Allor
Rotti

291
291

Teluti

291

Togean Ids

291

ai-ei

Batumerah

291

Morella

291

ai-oi
Baju

291

Vaiqueno

East 291

E.

Malay

Malagasy
Ahtiago
Awaiya
Batumerah

47
247
249
290
290
290

128
272
297

132
297
338

Bugi

Caimarian

Java

Lariko

Liang

Salibabo

47
290
272
290
290
257

Saparua

Sulu

Tagalog

Teluti

Teor

Visayas

290
290
249
290
290
3io

e-ai

Matu

276

e-i

Malay
Ilocan

132

3"

338
290

261

3i5
347

290

3iS

Java
Kisa

Macassar
Madura

261 276 290
276 318 323

347
290

Magindano
Matu
Salayer
Sanguir

276
261
290
290

Java

T70
- / -

e-y

Malagasy 172 318

SAWAIORI MATERIAL IN INDONESIA.

155

\-a

Malay

Malagasy
Batavia

i74
288

355
243

^5
}29

Chamorri 350
Dorey 291

Java 225

Sanguir 329
Silong 300

Tobo 330

Malay

Avvaiya

Basakrama

Caimarian

Java

170 171 352

180 291

312

291

171

Kisa

Sanguir

Saparua

Saru

Silong

2 74
290
35o
171
291

Teluti 330

Timor 312

Ulea 350

Waigioti 291

Malay
Gah

225

352

Madura
Savu

305
305

Teor

305

Malay
Malagasy

222 223 225

362

171 362

Ahtiago

Baju

Bouton

305
305
305

Matu
Savo

305
259

\-y

Malagasy 219 346 347 350

o-a

Malay

255 257 296

Awaiya

257

Kisa

259

3" 336 353

Baju

257

Matu

336

362

Bouton

257

Salibabo

257

Malagasy

255 282 296

Gah

257

Tagalog

287

355

Kayan

259 263

Visayas

287

0-OJt

Malay

309

Tagalog

309

Malagasy 353

Malagasy
Bouton
Brissi West

33 1
259
259

353

362

Manatolo

Savo

Sula

259
259
259

Tagalog 259
Vaiqueno East

O-tt

Malay
Malagasy

181
33i
171

33i

221

357
314

325
325

Baliyon

Java

Magindano

269

285 336 354
259

Pampangas

Salayer

Tagalog

353
285
353

o-y

Malagasy

259

U

Basakrama

Java

Lariko

u-a

Malay

Malagasy

79
239
299

273

220
266
3i6

281

221

281

266
299 316

278

Matu
Samba wa
Sula

344
343
306

u-au
Malay

334

259

156

THE POLYNESIAN WANDERINGS.

Malay

Malagasy
Ahtiago

176 224
358

296 316
306 329

329

Ceram
Gani
Java
Kandayan

329

317
224

343

Mysot

Teor

Waigiou

306
301

3i7

U-0

Malay 179 328

34i
Malagasy 179 216
273 278
320 334
Amblaw 316
Batumerah 317
Bolanghitam 278
Bugi 328
Dyak 317 328

330

267
294

343

317

Gah

Ilocan

Kayan

Lariko

Liang

Macassar

Magindano

Massaratty

Matabello

Mayapo

317
294

317
317
317
216

317
306

317
306

360

328 344

306 328

Pangasinan

Pampangas

Salayer

Saparua

Sula

Sulu

Tagalog

Teor

Togean Ids

Wahai

329
321
33o
3i7
306
294
267 321

3i7
321

3i7

34i

U-0

Malay

262

u-v

Malagasy

335

u-o

Malay

226 233

262

u-we
Awaiya

330

Batumerah

330

Caimarian

330

u-wi

Amblaw
Cajeli

330
330

Massaratty
Mayapo

330
330

Morella

330

u-yo
Bicol

330

u-yu
Malay

301

NG.

n8-g

Malay

332

Ilocan

332

ng-h

Malagasy

336

ng-n

Malagasy

Batumerah

169 213 246
2 74 308 346
35o
35o

Caimarian
Lariko
Liang
Morella

350
350
350
35o

Saparua

Teor

Wahai

35o
35o
350

ng-nh
Satawal

350

ng-nj
Chamorri

350

K.

k-ch
Malay

224

kg

Malay

Bieol

178 225
297 299
214

227
302

Java
Menado

225
300

299

Nicobar
Tagalog

300

214

k-h

Malagasy

249 258
301 302

297
353

Teor

305

306

SAWAIORI MATERIAL IN INDONESIA.

157

Java

k-ng
Malay

k-ngk
Malay

224

181

171

Java

171

Ahtiago 300

Awaiya 300

Batumerah 300

Caimarian 300 306

Gani 300

\-d

Malay 260 261 272

334 335 336
364

Malagasy 297 311 327

348 350 364

H

Togeanlds 308
272

l-kl
Malay

Timor

Malay

1—
Sulu

n-/t
Kayan

Ti-kn
Kayan

284
220
35o

328
259

Lariko 300

Liang 300

Matabello 300 306

Morella 300 306

Mysot 306

L.

Baju 335

Baliyon 335

Ilocan 334

Kawi 334

Java 272

Wahai 312 350

Tagalog 350

N.

n-l

Macassar 328
296

ring

Malay

n-nj

Dyak 328

n —

Ilocan

290

h-h

Bolanghitam 278

h-d

Ilocan
Java

h-y

Malay

339
278 339

278

Sassac 343

Kisa 284

Macassar 328

H.

Matabello 352

Kaili 278

Silong 339

Tagalog 353

Saparua 300

Teluti 300

Tidore 300

Wahai 300 306

Macassar 327
Matu 261

Sulu 335

Togean Ids 284

Wahai 317

Tagalog 259

Matabello 278

158

THE POLYNESIAN WANDERINGS.

h-/

Amblaw

278

h-ndr

Malagasy

47

h-wg
Kisa

278

h-nr

Bugi

47

h-r

Malagasy

278

h-s

Malay

Ahtiago

Amboyna

Awaiya

Bugi

Buru

Caimarian

352
352
352
352

352
352
3.52

Ceram

Chamorri

Galela

Gani

Guaham

Ilocan

Java

352
352
352
352
352
323
352

Kawi

Kayan

Sanguir

Tidore

Tobo

Togean Ids

352
278
352
352
352
352

h-t

Satavval

352

h—

Macassar

352

Sunda

352

s-ch

S.

Malay

45

s-d

Malay

47

298 338

s-g
Malay

132

s-k

Malay
Malagasy

233
169

337
223

Awaiya
Java

180
342

Kayan
Saparua

263
180

Malay

34i

Malagasy

34i

s-»

Malagasy

344

s-ng

Togean Ids

298

s-r

Malagasy

298

Java

298

s-t

Malay

170

225

Malagasy

239
T.

Java

225

t-ch

Malay

237

26S

Sunda

352

t-d

Malay
Gah

357
329

35S

Java
Matabello

357
329

Timor
Tobo

329
329

t-g

Galela

352

Gani 352

Tidore

352

SAWAIORI MATERIAL IN INDONESIA.

159

t-h

Satawal

294

t-j

Malagasy

358

Macassar 352

Tobo

217

t-k

Malay
Caimarian

225
329

Java 324

Kisa 318 323 324

Teor

35o

t-m

Gah

217

Mysot 217

t-n
Java
Kayan

346
346

Massaratty 217

Mayapo

217

t-ndr

Malagasy

238

t-nch
Batak

329

t-nt

Malay 216 222 256 Macassar 329

Malagasy 256 Pangasinan 329

352

t-s

Malay 225 247 329 Bum

Malagasy 324

t-tj

Bugi

350

Samba wa 329

Sanguir 329

t-ts

Malagasy

329

t-2

Malagasy

47

t—

Malay 47
Malagasy 47
Amblaw 217
Awaiya 2 1 7
Bolanghitam 217

Bouton 217
Ahtiago 217
Bugi 47
Caimarian 2 1 7

M.

m-b

Malay

224 315 327

Malagasy 172

m-h
Tagalog

327

m-lm
Menado

313

m-p
Malay

325

m-v

Malagasy

327

m-w

Java

224

m —

Visayas

327

Cajeli

217

Lariko

217

Liang

217

Sanguir

352

160

THE POLYNESIAN WANDERINGS.

V.

Malagasy

307

Baju

291

y-k

Vaiqueno East 291

v-6

Malay

310

v-/

Malagasy

281

v-p

Malay

281

v-w

Malay

307 310

Dorey

291

Saparua

291

Ahtiago

291

Gani

291

Silong

291

Allor

291

Liang

291

Solor

291

Amblaw

291

Massaratty

291

Sulu

291

Awaiya

291

Mayapo

291

Teluti

291

Batumerah

291

Morella

291

Teor

291

Buru

291

Mysot

291

Togean Ids

291

Caimarian

291

Rotti

291

Visayas

291

Cajeli

291

Salibabo

291

Waigiou

291

Ceram

291

v —

Malay

291

Champa

291

f-/

Malagasy

233

235

254

Massaratty

290

329 360

Sulu

290

259

273

296

Matabello

259

329

(Ph)

Teor

36o

Ahtiago

329

(Ph)

Mayapo

290

329

360

Timor

2 84

Ceram

329

(Ph)

Satawal

2 94

Tobo

329

Gah

329

f-6

Malay

174

175

220

Bugi

36o

Molucca

284

222

246

272

Cajeli

36o

Salayer

290 360

273

291

292

Gah

290

Salibabo

290 360

294

360

Iloean

290

294

Sanguir

329 360

Malagasy

273

3i4

Kayan

2 94

Silong

290 294

Amblaw

360

Liang

350

Sulu

2 94

Baju

360

Macassar

290

294

Togean Ids

292 294

Bicol

292

Madura

290

Visayas

292

Borneo

360

Menado

360

Waigiou

290

Bouton

290

f-ch

Bouton

259

i-h

Malay

215

Lariko

290

Sanguir

290 (how)

Malagasy

213

329

Liang

290

Sula

259

Awaiya

290

Madura

290

Teluti

290 360

Caimarian

290

360

Manatolo

259

Timor

329

Kisa

259

Morella

290

360

Wahai

360

i-m
Bicol

214

Sulu

214

Tagalog

214

i-mb

Cajeli

290

SAWAIORI MATERIAL IN INDONESIA.

161

f-p

Malay

Beu
Bicol

t-s

f-v

Malagas)'

f-w

Bouton

Bugi

Gah

f—

Batak

170 223 245

287 288 296

329

243

214

Brissi West 259

246 273 282
288 293 294

290
292
360

329

Batumerah 290
Caimarian 329

Gani 290

Java 288

Kayan 259

Magindano 259

Mysot 290

Savo

Java
Kisa

Macassar
Salayer

259

Ahtiago 290 360

272 273 283
290 360

273 282 283

329
290

Panga^inan 329
Samba wa 329
Saparua 290
Tagalog 259 287
Visayas 287

Vaiqueno East 259

Kayan

273

Magindano 294
Savu 294

Teor 290

Saparua 259
Teluti 290

v-P

Mala)' 183 217 241
247 250 279
289

Malagasy 179 250

Amboyna 190

Baliyon 250

p-b
Malay 45 171 176

178 221 241

286
Malagasy 218
Batavia 243

P-/

Malagasy
Aru

p-mb
Malagasy

175
284

218

Basakrama 266 279

Bicol 250

Bugi 279

Java 289

Kawi 250

Macassar 279

Ilocan 284

Java 171 285

Kaili 284

Salayer 285

Lobo
Rotti

284
284

Magindano 250

Morella 190

Saru 1 7 1

Tagalog 250

Teor 285

Solor 284

Tagalog 45

Togean Ids 284
Tringanu 284

Togean Ids 284

p-v

Malay 128 284

p-w

Gilolo 284

Malagasy 171

Kisa

284

Frontal Abrasion:

Malay 47 177 258

290 291 323
Malagasy 177 325
Ahtiago 217 290

Amblaw 2 1 7
Awaiya 2 1 7

Borneo 350

Cajeli 290

Dyak 312

Magindano 284

Gah 275 290

Java 290

Kayan 344
Massaratty 350

Matu 350

Menado 217

Morella 217

Mysot 290

Nicobar 300

Utanata

284

Pampangas 328

Salayer

217

Sanguir

217

Saparua

217

Sulu

290

Teluti

217

Teor

217

Wahai

217

Waigiou

290

162

THE POLYNESIAN WANDERINGS.

Frontal Accretioi

Malay

79 170 175

Gani

312

Salibabo

3^2

176 241 278

Java

274 278 321

Sandol

278

304 321 337

342

Saparua

180

Malagasy

175

Lariko

278

vSilong

291

Ahtiago

305

Liang

278

Sula

278

Awaiya

1 So

Macassar

327

Tagalog

274

295 321

Baju

278

Magindano

274 313 321

Teor

278

Bicol

274

Menado

300

Ternati

276

Borneo

300

Morella

278

Tidore

295

300

Bolanghitar

1 217

Pampangas

321

T ogean Ids

321

Bouton

305 306 313

Pani

295

Visayas

274

Galela

278

zrminal Abrasion:

Malay

79 132 170

Gani

290 312

Kar Nicoba

r

324

175 176 270

Ilocnn

267

Sirang

312

289 299 302

Java

299 357

Tagalog

214

304 357

Jobi

312

Teor

306

35o

Baliyon

250 269

Macassar

350

Ternati

276

Bicol

214

Mayapo

329

Tobo

320

Bugi

350

Mysot

217 290 306

Waigiou

290

Dorey

312

312 324

Gah

312

erminal Accretion:

Malay

128 171 177

Borneo

300

Matabello

259

278 300

178 179 185

Bouton

217 257 300

316

3i7 324

222 223 226

305 312

Matu

267

277 305

227 228 237

Brissi West

259

336

344

239 245 247

Bual

278

Mayapo 300 317 3.50 360

248 250 255

Bugi

267 274 277

Menado

300

256 260 266

279 308 352

Molucca

284

274 278 279

Caimarian

278 291 300

Morella

190

291 300

284 291 298

306 330 350

3i3

330

300 305 308

360

Mysot

291

3" 317 323

Cajeli

217 278 290

Pampangas

277

278 327

327 328 329

3*3 317 360

328

353

33i 332 335

Ceram

278 329

Pani

295

34i 352 353

Dorey

291

Rotti

278

359 362 364

Dyak

317 324 328

Salayer

33o

Malagasy

171 177 179

Gah

278 290 300

Salibabo

257

278

223 235 238

316 317 324

Salu

278

239 250 255

329 330 352

Sambawa

343

258 260 267

Gani

291 300 317

Sandol

278

273 274 278

Gilolo

284

Sanguir

217

352 360

298 302 308

Guaham

317

Saparua

180

300 335

314 320 335

Ilocan

277 284 300

Silong

290

300 323

338 341 353

323 332 334

324

339

355 362 364

339 342

Solor

284

Ahtiago

217 300 306
317 324 329

Java

171 267 274
277 298 308

Sulu

267
335

308 317

330 360

316 339 350

Tagalog 18

5 259

267 274

Amblaw

217 278 300

352 364

277

295 308

3i7 330 360

Kawi

250 352

Teluti 2 1 ;

• 291

300 317

Amboyna

190

Kay an

259 277 278

324

330 360

Aru

284

308 317 328

Teor

278

285 300

Awaiya

257 278 291

344

305

313 317

300 317 330

Kisa

259 276 278

324

360

Baju

305 335

312 313 324

Tidore

27S

300

Baliyon

308 335

Lampong

343

Timor

278

284

Basakrama

• 250 279 350

Lariko

300

Tobo

217

27S

Batavia

243

Liang

300 313

Togean Ids

; 278

284

Batumerar

1 278 290 291

Lobo

284

Tringanu

284

300 313 316

Macassar

216 267 279

Utanata

284

324 330 350

353

Visayas

185

274 278

Beu

243

Magindanc

1 259 274 284

317

327

Bicol

274 330

30S 317 340

Wahai 278 300

306 317

Bolanghila

m 217 317

Massaratty 300 350 360

324

350 360

SAWAIORI MATERIAL IN INDONESIA.

163

Infixion:

Am blaw

300

Awaiya

278

Baju

278

Basakrama

250

Batumerah

278

Bual

278

Bugi

277

Caimarian

278

Cajeli

278

Ceram

278

Ende

278

Galela

278

Metathesis:

Malay

3i6

Malagasy

316

Ahtiago

330

Baliyon

250

Bicol

250

Gani

278

Ilocan

277

Java

267 2S3

Kisa

282 283

Lariko

278

Liang

278

Massaratty

329

Matu

277

Mayapo

329

Morella

278

Pampangas

278

Rotti

27S

Ilocan

346

Java

316

Kawi

250

Macassar

2 go

Silong

290

Salibabo

278

Salu

278

Sandol

278

Sulu

267 335

Tagalog

277

Teor

278 291

Tidore

278

Timor

278

Tobo

278

Togean Ids

278

Visayas

278

Wahai

278

Sulu

330

Tagalog

250

Teluti

330

Teor

291

Wahai

330

Next we shall compare the mutation series of the consonants in
Indonesia with the variants observed in the Pacific areas as set
forth in the tables in each of the two preceding chapters. The
results of such comparison fall under the several following classes :

Indonesi

an only.

ng-h

k—ngk

a—kn

h-ng

s-d

s-ng

t-ndr

t-ts

m—lm

m —

ng-nh

l-kl

a—nj

h-nr

s-g

t-ch

t-nch

t-2

tn—p

i-ch

k-ch

\-s

11-/

h-r

*~j

t-z

t-nt

m-/>

m— w

t—m

k-j

n-h

h-ndr

s-ch

S-tt

t-h

t-tj

m-h

k-ns

n-ng

Indonesian-Polynesian.

I nil ones ia n -Mela n est a n .

ng-g

ng-nj
k-k

H

n —

h-j

h-t

h-d

s-t t-n v-/ v—
t-d m-v v—p f-mb
t-tn v-b

Indonesian- Mela nes ian-Polyncsian.

is
P-t

D-mb
p-x

p—

ng-n

k-g

k—

\-d
\-n
1—

nl

n-ng

h-h

h-.v t-j t — V-w
h— t-s v-h i-b

s-h t-k

i-h
t-p

i-v
f-w

f—

p-b

We shall forward our understanding of whatever interrelation
may exist among these three several language groups by a more
detailed examination of each of these classes. If there be any valid-
ity in the theory that the three families are descendants of a com-
mon parent we should look for confirmatory evidence in the amount
of their concord in mutation principles. This can be made to appear
only in a detailed examination.

In Polynesia we have seen the palatals to be subject to but slight
derangement. In the case of ng the more frequent mutations are
of the form ng-n and ng — . Only the former extends through all
these families. In Polynesia it is normal in Hawaii, the western

164 THE POLYNESIAN WANDERINGS.

Marquesas, and Sikayana, sporadic in Rotuma. In Melanesia it is
not normal (with a possible exception in the case of Santo); it is
traceable through the languages of that area rather as a quality
with which a certain few loan words are endowed. In Indonesia it
is seen to be a quality of talinga (350) ; in Malagasy only is it suffi-
ciently frequent to be considered as a speech endowment ; the Mala-
gasy words in which this mutation is found are all identifiable as
Polynesian material, and several of them are carried with the same
modification through Melanesia.

In Polynesia the mutation ng-£ is normal in the eastern Mar-
quesas, sporadic in one instance in Viti. In Melanesia its place is
taken by ng-g, a word quality. In Indonesia it is found but once.

ng-n; is highly problematical in the sole instance in which it is
found in Aneityum. Its presence in the Indonesian rests upon
Tregear's citation talinga (350) ear Chamorri talanja, against which
we set talanga from Fritz's Chamorro dictionary.

ng-n/*, resting on longo (336) sound Malagasy rohona, is not cer-
tainly Sawaiori.

ng-nh in talinga (350) ear Satawal talinhe is a good identification
but the mutation is without parallel.

K.

In Polynesia the principal variation of k is to extinction, this
being normal in Samoa, Hawaii, and Tahiti, and not unknown in
several other languages. In Melanesia it is found widely extended
as a word quality; in Bululaha, Saa, Ugi, and Ulawa it appears
as a speech endowment, these being in the region of crop colonies
in the Solomons. In Indonesia it appears as a word quality of but
two vocables.

Of the k-g mutation we have record of but a single instance,
sporadic in the Paumotu. In Melanesia this is of frequent occur-
rence and in many languages it is normal. In Indonesia it is infre-
quent outside of the Malay, a speech endowment none the less
though sluggish; the Malay and Javanese words in which it occurs
are Sawaiori, those in other languages doubtful.

k-h is of rare occurrence in Melanesia. In Indonesia it is satis-
factorily identified in Teor and in some, though not in all, of its
occurrences in Malagasy.

k-ng is of rare occurrence in Polynesia, equally rare in Melanesia,
and its single appearance in the Malay is not a wholly satisfactory
identification.

k-ngk practically corresponds to the k-ngg found four times in
Viti, rarely in Melanesia. In Indonesia it is found in but two
vocables, each in two languages.

k-ch and k-j are not satisfactorily identified.

SAWAIORI MATERIAL IN INDONESIA. 165

L.

Grouping 1 and r because of the frequently mentioned impracti-
cability of establishing a stem independence in our present data,
we note that in Polynesia the most frequent mutation is the extinc-
tion. This becomes most strongly marked in the Marquesas. It
is infrequent in Melanesia. In Indonesia it appears in Sulu and
Tagalog in the same word, talinga (350) ear, and in the Togean
Islands in pula (284) to shine, at least the \-y mutation noted in
that instance is a close approximation to extinction.

The \-n mutation in Polynesia seems more a word quality, for it
appears in two languages each for data 100, 154, 312, 313; these
removed from the reckoning, we find it only five times, all in Nukuoro.
In Melanesia it is rare. In Indonesia it is found in pula (284) to
shine Timor junan the moon, twice in Wahai; these three words
are included in the list of those similarly affected in Polynesia.

The long leap from lingual semivowel to mute of the same series
is found but once in Polynesia (\-t Mangareva 287) and once in Viti
(\-nd 287). In Melanesia it is scarcely more frequent. In Indo-
nesia \-t does not occur; \-d appears sporadically in several lan-
guages, one instance (350) containing the greater number of its
Melanesian occurrences as well. This mutation is most marked in
Malay (seven times) and Malagasy (five times). Of the Malay
instances but two (260, 334) are open to any doubt, of the Malagasy
but two (348, 350) are really convincing. The \-j mutation, really
a t- variation, is found but once in Indonesia, three times in two lan-
guages of the New Hebrides, and not at all in Polynesia. \-s is noted
in a single instance in Malay and very unsatisfactory. \-kl, noted
once in Malay and Java, is rather an accretion than a mutation.

N.

The variants of the three nasals follow two cleavage planes, the
vertical in series, the horizontal across all series. In the case of
ng we have observed as well ng-n and ng-k. In Polynesia n shows
this horizontal movement only backward, n-ng, and this only in
Moriori and once in Viti. In Melanesia n-ng is quite widely spread,
affecting several vocables and most of them repeated in several lan-
guages; here is found in one instance the forward n-m mutation.
In Indonesia the n-ng mutation occurs once each in Malay, Kisa,
and Tagalog, the Kisa instance doubtful.

In Polynesia the only vertical mutation is upward in the series
to the semivowel, to / in a single word in two languages, to r in
another word in two other languages. This mutation in Melanesia
has a wide extent in the case of nifo (259) tooth, and is less widely
found in three other of our data. In Indonesia it is found once each
in Macassar, Sassac, and Wahai, the latter affecting the same vocable
as in Alite.

166 THE POLYNESIAN WANDERINGS.

The extinction of n, found once in Marina, is highly doubtful in
the single Indonesian case in fa fine (290) woman Ilocan babai.

The n-l in Kayan, n-nj in Dyak and Macassar, and n-y in Pam-
pangas are found only in the vocable (328) which in Macassar has
also afforded an instance of n-l and gives n-ny in Melanesia. The
single instance of k-kn is of uncertain nature.

H.

The foregoing tables record a considerable number of mutation
forms for the aspiration in Indonesia. For the mutation h-s, no less
than twenty-two languages afford examples; yet on closer inspec-
tion it will appear that in the latter case but three vocables are at
all involved, and that six vocables only are involved in the whole
range of mutation of the aspirate. We feel justified, therefore, in
regarding this as partaking more of word quality than of speech
endowment. In Polynesia the aspiration has all but vanished;
comparison of mutant forms, therefore, would be futile, the only
forms common to the three families being h-s and extinction. Mela-
nesia, in stiong contrast with each family, shows a marked ten-
dency to employ the aspiration in its proper function ; over an equal
and interlacing area it follows the h-s mutation. Common to Indo-
nesia and Melanesia are these, h-t, h-d, and h-j, the two latter being
variants upon the primal h-t mutation; the vocables involved in
these changes in Indonesia are similarly involved in Melanesia, which
has a few in addition of its own. Mutations peculiar to Indonesia
are the following: h-l, h — , h-ra, an upward movement in the series,
discoverable only in the two vocables recorded in our data under
numbers 47 and 278; regarding ndr, as so frequently, in the light
of an effort to compass the r grasseye, h-ndr (Malagasy 47) falls into
this group. In Kisa 278 h-ng is explicable as the upward shift h-n
(which is nowhere discoverable), upon which is added the rearward
shift n-ng; confirmation being found in our record of the latter
mutation in that speech.

S.
The sibilant is better retained in Indonesia than in Polynesia, a
facility which it shares with Melanesia. The mutation common to
the three families is s-h, normal to at least eight Polynesian lan-
guages, far less frequent in Melanesia, and in Indonesia affecting
but seven vocables in six languages. Of the mutations common to
Indonesia and Melanesia, s-r (found in Malagasy and Java 298)
rests doubtfully on Tanna 337; and s-t is found in Alite and Wango
in the same 337. Wholly Indonesian are s-ch, s-j, s-d, but they
group naturally with s-t. s-n (Malagasy 344) lacks confirmation.
s-ng (Togean 298), however, seems quite feasible, s-g (Malay 132)
is very doubtful.

SAWAIORI MATERIAL IN INDONESIA. 167

T.

Common to the three families are the mutations t-s and extinc-
tion. The former is quite rare in Polynesia, not particularly fre-
quent in Melanesia, and in Indonesia involves but five vocables in
four languages, one of these being similarly affected in four of the
Melanesian languages.

t-k is the rule in Hawaii, rapidly becoming normal in Samoa, well
established in Melanesia, and somewhat common in Indonesia, t-j
is well established in Tonga; Melanesia shows it in Baki and Male-
kula; its use in Indonesia is but slight; with it should be grouped
the t-ch of Malay, t-tj of Bugi, and t-nch of Batak.

For the extinction of t we muster but a single example in Poly-
nesia (Marquesas 350), quite a respectable number of instances in
Melanesia, but its Indonesian record is almost exclusively written
in 217, a doubtful case as will be seen in the discussion of the note
thereupon.

In the forms common to Indonesia and Melanesia t-d is the sim-
plest variant, rather widely (including t-nd) disseminated in Mela-
nesia, in Indonesia involving but three vocables in six languages.
t-m in both families rests wholly upon the doubtful 217. t-n in
Melanesia is found only in New Ireland 329; in Indonesia it rests
soundly on 346 (Java and Kayan) and doubtfully on 217.

Of forms wholly Indonesian, t-nt is not infrequent and is under-
stood as a nasal reinforcement of the mute, t-ts and t-z are met
with but once apiece, both in Malagasy, t-h (once in Satawal) may
be regarded as a secondary mutation of the t-s hitherto noted. For
t-ndr we find but one example, in the Malagasy, yet it is quite
satisfactory, t-g, a violent mutation, rests upon a single vocable
(352) in Galela, Gani, and Tidore.

M.

The mutations of m need scarcely engage our attention in this
rapid review. Polynesia affords but a single instance ; in Indonesia
but five vocables are involved ; even in Melanesian crudity of speech
m is almost constant.

V.

The v-mutations common to all three families are \-w and \-h.
In Polynesia \-w has become normal in Maori, Hawaii, and Viti;
in Melanesia it is satisfactorily established over a wide area or sev-
eral areas interlaced by this mutation; in Indonesia it rests almost
wholly on vai (291), a word quality at best. The Indonesian \-h
rests wholly upon this vai, in a single instance; it is found in Mela-
nesia in but one vocable in two languages; our Polynesian data
afford us onlv three instances.

168 THE POLYNESIAN WANDERINGS.

Of the forms common to Indonesia and Melanesia v-/ is scantily
represented in either; v-6 in a solitary Malay instance is rather more
common in Melanesia ; x-p, once in Malay, occurs in but two vocables
in Melanesia; and v extinct is represented in Melanesia by one
vocable in two languages and in Indonesia by the somewhat doubt-
ful Malay and Champa identifications of vat,

F.

In Polynesia the most common f-mutations are i-h, normal in
Hawaii, Mangareva, Maori, and Tongarewa; and f extinct, normal
in Bukakuka, Mangaia, and Raro tonga. In Indonesia i-h involves
but six vocables in fifteen languages, in Melanesia it is more frequent.
The extinction of f in Indonesia is found in four vocables in seven
languages, in Melanesia in eight vocables in eight languages.

i-b is met with but twice in Polynesia; including i-mb, it is rather
more common in Melanesia, is frequent in Malay and, with narrower
application, quite generally extended over Indonesia. Closely asso-
ciated with this is i-p, twice appearing in a single vocable in Poly-
nesia, widely extended in several Melanesian instances, frequent in
Malay, and of considerable extent in other Indonesian languages,
f-i; is normal in Viti ; in all the rest of Polynesia involves but five
vocables in eight languages ; is very common in Melanesia ; in Indo-
nesia is found several times in Malagasy and rarely in Malaysia,
f-w, once met with in two Polynesian languages, is very common
in Melanesia ; in Indonesia is found several times in Java and rarely
in seven other languages, is is common to Indonesia and Melanesia,
but is rare in each, i-ch is peculiar to Indonesia and rests upon a
single and satisfactory example, Bouton 259. The f-w mutation
(214) is most uncertain.

P.

The only mutation common to the three families is p-6. For this
Polynesia yields but few examples; including p-mb it is by far the
most frequent Melanesian variant; it is frequent in Malay and nar-
rowly extends to ten other languages of Indonesia, p-/ appears in
but a single Melanesian example; in a single vocable it appears in
four languages of the Malay Archipelago and doubtfully once in
Malagasy, p-v has a somewhat wide distribution in Melanesia, but
is rare in Indonesia, p-w appears once in Gilolo and in the same
vocable in the New Hebrides, in another in Buka. The extinction
of p involves the same vocable in three Indonesian and one Mela-
nesian languages.

In the Indonesian check-list I have distinguished in bold-faced
type the vocables which appear in Malayan tongues with the quality
factor of recognizability by the Samoan, of which I have made such

SAWAIORI MATERIAL IN INDONESIA. 169

use in treating of the Melanesian languages. Beyond this record
of quality I hesitate to tread upon a field already so well cultivated
by its own specialists; yet in so far as this area is involved in the
erection of a Malayo- Polynesian speech family it is a fair field for
the Sawaiori student. Since we have now brought one ethnic swarm
of Polynesian voyagers back to the threshold of Indonesia at two
doorways respectively, and have made it plain that the later Tonga-
fiti swarm has found its exit elsewhere than at these identified por-
tals, we are amply justified in our examination of Indonesia so long
as our search is restricted to the replevin of Sawaiori material.

The prominent numerals in the preceding list make it wholly mani-
fest that the Polynesian and the Indonesian have at some place and
at some time been in intimate contact. That this place of inter-
mingling was not the Pacific is equally manifest, for the Pacific
languages contain no elements of Indonesian origin save this scanty
collection picked out of Polynesian and Melanesian. A second line
of proof lies in the consistent tradition of the Polynesians that they
came over seas on a long voyage from the west.

Restricted, then, to Indonesia as the place of mingling, we must
first examine the history of the two races. Of the Malay race we
know, for their records of their own history show it, that Indonesia
was not the place of their origin. Upon it they are intruders; from
the Asiatic main they came by the easy path of the Malay Peninsula.
So far as the traditions of Java can be reduced to a measure of
synchronism, we can fix Java as uninhabited by Malayans about
400 b. c, and a century later settled by 20,000 families emigrating
from northwest India under Arishtan Shar and dispersing to Mala-
bar, the Maldives and Madagascar; a large emigration from the
Pan jab to the archipelago between 200 and 150 b. c. ; and by 125
b. c. the archipelago overrun with these races.

As concerns the Sawaiori we lack all but the most indefinite infor-
mation before their sojourn in Indonesia. We do not regard them
as autochthons, for their traditions call for a migration yet earlier.
But we are fairly entitled to regard them as in possession of the
islands from Sumatra to the Philippines at the time of the Malaysian
swarm. All indicia point to their retreat before the swarm advanc-
ing upon their western flank, the only lines of retreat open to them
being in the eastern quadrants.

It is difficult for us to arrive at the comprehension of the expulsion
of a race from its home. We have to go back to rude times to find
ground for the belief that such things can be, to the swarming of
the Huns upon Europe reported to us in the testimony of eye-
witnesses and sufferers at the bloody hands of Attila. Without
detail we have accepted, with dull imaginations and no great com-
prehension, the westward movement of the Aryan races.

170 THE POLYNESIAN WANDERINGS.

Such movements we denominate ethnic swarms; a convenient
designation ! no hesitation has been felt in employing it in the fore-
going chapters. But in denomination do we really describe? In
calling the movement of the Sawaiori an ethnic swarm do we get
sight of the nature of the movement?

In the hydraulic physics of geography Niagaras are rare; few
streams go tumbling over cliffs in a mass ; the course of the greatest
rivers is marked by many an eddy, many a backwater, many a pool
where motion scarcely appears. It is only in the errant fancy of
the runaway child that tired feet on a weary way will at last bring
him to a real jumping-off place. A stream of human migration can
only flow in succeeding waves; many an eddy current will bear on
a new destination.

Particularly must such have been the case with the expulsion of
the Sawaiori before the Malay. In bulk and in the end it did indeed
become a great ethnic swarm. In detail and in the performance it
could have been no more than affairs of outposts.

The genius of Polynesian culture has nowhere as yet touched the
idea of a national life. Samoa is almost singular in the possession
of a collective name for its archipelago; the Samoan, the Viti, the
Marquesan, and the Maori are alone in the possession of national
names. Civic righteousness goes not beyond the village; village
does not go to the relief of village in distress; there is even now
nowhere a nation to arise in unity to set a bold and united front
against an invader. Each tiny village makes its own defense. Vic-
torious it holds its ground until the next attack; defeated it is a
sufficiently mobile unit to set forth in search of a new home. We
must remember that such were the conditions of the Sawaiori in
the archipelago when the Malay in a similar wise were advancing
upon them.

Under such conditions there are three main possibilities. A Sawai-
ori village resists attack after attack, maintains its good defense,
retains the little islet or the pleasant bay of some larger island which
is its home, is surrounded by settlements of the alien race, in time
peaceful relations arise. A Sawaiori village realizes that resistance
is futile, puts out to sea, seeks a new home. A Sawaiori village,
debellated in some sudden onfall or by the crush of overwhelming
force, goes down in defeat, and the sole survivors, the noncombatant
women and children, are incorporated with the conquerors.

The second event accounts for our wanderers over sea, those whose
voyages we have traced through Melanesia and into their present
Polynesian homes. The first and third events are to account for
such speech community as may be found in Indonesia and Polynesia.

It is not a great community. There are very few items which
are not included in the data here assembled. See what a small basis

SAWAIORI MATERIAL IN INDONESIA. 171

it forms for the erection of a Malayo-Polynesian family. In the
Malay itself, the speech of which we have the longest record and
the fullest comprehension, there are but 75 vocables safely identified
as common in these data to the two families. Making the most gen-
erous allowance, a lavish allowance, for the vocables which evaded
compilation under the conditions of this research, we can only thus
doubtfully find a community of 150 words.

We are to consider the source of this petty common vocabulary.
In the analysis of the possibilities it may have been borrowed from
Indonesia by the Sawaiori during their joint occupancy of that area ;
it may have been delivered to each from a common source. It may
have been contributed by the Sawaiori to their Malaysian conquer-
ors, by such of the Sawaiori as persisted as inclusions in the Indo-
nesian settlement, it being well understood that sooner or later they
were absorbed in the alien culture and outside this linguistic record
have left no mark, or at best a scarcely measurable trace.

If the Malay peoples advancing upon the Sawaiori peoples whom
they found in possession of the islands of the Indian Archipelago
forced their vocables upon the folk whom it was their pleasure and
\ to their interest to scatter before them, then we are at an utter loss
to comprehend the nature of such intercourse, for it is not in one
generation nor yet in three that a people adopts any considerable
element of an alien speech. Furthermore, if the Sawaiori borrowed
directly from the Indonesian, incomprehensible as such a contin-
gency is, we should expect to find the greatest range of variety, the
widest divergence from phonetic principles, among the several divi-
sions of the borrowing race, the closest uniformity among the lenders.
Yet in the case of each of these seventy-five vocables here discussed
the Polynesian keeps the word practically without alteration; in Indo-
nesia the range of variety is enormous. The nature of the Indonesian
variety is plain to see in the phonetic tables heretofore drafted.

We note three types, and careful study of the tables will show
that all variants fall under one or other of these types.

First : phonetic variation recognizable as a Polynesian type. This
may mean that dialectic variation in the Sawaiori material existed
at the time when the two races had the word in common, a very
possible contingency. It may mean that in the word itself was a
disposition or motion toward a certain type of mutation which was
carried over with the loan, a contingency almost impossible, for we
have yet to learn of an instinct in the word ; the Sprachgeist resides
in the speaker.

Second: incorporation upon the common stem of formative ele-
ments distinctively Malaysian. This sort of thing is very common
in all speech, markedly characteristic of our own English in its word-
pilfering from every source.

172 THE POLYNESIAN WANDERINGS.

Third : modification of the common stem by mutation unusual to
the Indonesian phonetic and probably representing the resultant of
an effort to reproduce foreign sounds difficult to the borrower's vocal
organs.

Similarly we should expect to find a word delivered by the Malay-
sians to the Sawaiori diminishing in frequency along the wide eastern
extent of the migration, practically constant in Indonesia itself.
The answer thereto is patent in the check-list : Polynesia holds each
of these words practically to its utmost east ; it is Indonesia which
shows great gaps.

We can not, therefore, regard the common vocabulary as in any
sort borrowed by Sawaiori from Indonesia.

To the theory of derivation from a common parent the objections
are insuperable. A common parent would have delivered a greater
community of vocabulary, would have delivered a grammatical sys-
tem that would show some interrelation between the two branches
of the family. There is absolutely no record of a speech of man
which contains these few vocables which Polynesia and Indonesia
share, for it must be plain long ere this point is reached that the
Semitic theory has no sound base.

The third possibility is that to which alone the objections are so
few and so slight as readily to yield to the study of the problem,
namely, that the community of vocabulary is Indonesian borrowing
from Sawaiori. How this might be brought about has sufficiently
been indicated already in the consideration of the nature and extent
of the Sawaiori expulsion.

The indication of quality of such borrowed material presented in
the check-list affords more matter of interest. As in our study of
the Melanesian traverse it indicates by lines of higher quality the
lines of travel.

Assuming, and this the well-recorded traditions of Java warrant —
assuming the appulse of the advancing Malaysians as delivered upon
the islands of the archipelago by the convenient way of the Malay
Peninsula, Sumatra first would feel the shock, and Java next in
practically undiminished volume. From Java two eastward ways
lie out like a < . The northern line, following the north shore
of Borneo, leads directly to the Philippines; following the south
shore of that great island leads equally to the Philippines in their
southern extent, but affords many opportunities of deviation by
Celebes toward Gilolo northward and toward Buru and Ceram in
a southerly course, and there the flight would be stopped by the
inhospitality of New Guinea with the obstacle of its fierce and
immiscible Papuan race.

The southern line leads directly into the Arafura Sea and Torres
Straits. The halting-places are such islands as Bali, Lombok, Rotti,

SAWAIORI MATERIAL IN INDONESIA. 173

Sambawa, Timor, and lesser groups yet farther east. In the Arafura
Sea the southern line would be joined by such of the fugitives along
the northern line as deviated on the Celebes-Buru-Ceram course and
were turned still further away by New Guinea.

New Guinea is no theoretical obstacle. We find less trace of the
Sawaiori on its almost continental mass than upon any spot of land
which their canoes could have reached in the flight. Even up to
the present day the Malays, for two millenniums holding all of
Indonesia and voyaging hither and yon for the equal joys of fighting
and of trading, have succeeded in making no permanent lodgment
upon the Papuan shore.

The Sawaiori flight out of Indonesia inverts the seasonal migration
of the geese — its < opens forward. Thus parted, approximately at
Java, it is to be a long flight before those diverging lines come
together, not until Samoa is reached and a new home for the united
race, for such at least as have escaped the infinite perils of unknown
seas. It will be recalled that in the study of the migration through
Melanesia the material under examination was assembled to prove
the existence of a northern and a southern track, the Samoa and
the Viti streams respectively, one emerging from Indonesia through
an eastern portal in the Bismarck Archipelago, the other through a
southern portal in Torres Straits.

The latter is readily recognizable as the direct production of the
southern line of flight in Indonesia. If we examine such vocables
as exhibit a difference between the two streams in Melanesia we
shall find that those which characterize the northern stream find
their greatest frequency, preserve the highest quality, in the several
languages of the Philippine Archipelago, Bicol, Ilocan, Magindano,
Pampangas, Sulu, Tagalog, Visayan and others from which our
material is less complete.

It seems that now it is time to relinquish the term Malayo-Poly-
nesian. There is neither ethnic nor linguistic unity. If the Indo-
nesians in the paucity of their original speech have borrowed some
150 vocables from their victims, if they have been willing to take
even their numerals from a conquered and fugitive race, it does not
seem that they are entitled to be bracketed, and that with the
honors of first mention, with a distinct family of speech. The Poly-
nesian is an older speech than the Indonesian, one that has been
carried greater distances of sea and land than even our Aryan until
after centuries the Aryans grew bold enough to conquer the sea;
yet uncounted generations earlier the sea had been the easy path
for the Sawaiori to come to his Polynesian own.

CHAPTER IX.
THE SAWAIORI BEGINNING RESTS UNKNOWN.

Check -list of the Semitic words for which affinity has been sought —
Failure of the effort to identify this material with Sawaiori stock —
The reasons lie in false definitions and ignorance of phonetic prin-
ciples— The Semitic does not conform to the laws of the family —
Summation of the results of this inquiry — The two Sawaiori swarms,
the earlier through Melanesia, the latter not yet discovered on the
face of the trackless sea — The double migration track in the western
Pacific — The problem of the Melanesians has been considered only
in so far as they have been affected by the wandering Sawaiori — End
of the classification which has joined Malay and Polynesian — The be-
ginning of the great Polynesian race is lost in westward and empty sea.

As in the earlier chapters, we provide a check-list of the various
Semitic identifications sought to be established by Dr. Macdonald
in the data here collated. Thus those still curious in the further
examination of the theme may follow the topic back to the several
languages which he has involved in the elucidation of his theory.

Arabic . .

13

17

18

19

21

22

23

25

26

29

30

3i

32

34

35

36

37

38

42

43

44

45

46

47

80

83

84

85

86

88

89

90

9i

92

93

97

98

99

100

101

102

103

104

105

106

107

108

no

in

112

114

116

117

120

133

135

136

137

138

139

140

142

143

144

145

146

150

151

152

153

154

156

158

160

162

164

165

166

168

170

171

172

173

174

175

176

178

180

181

182

183

184

185

186

187

188

189

192

193

194

195

196

197

198

199

200

202

205

206

207

210

211

212

220

221

222

224

225

226

227

228

229

231

232

234

235

236

237

238

239

240

241

242

243

244

245

246

247

249

250

251

252

253

2 54

255

256

257

258

259

260

262

263

264

265

266

267

268

269

270

271

272

273

274

2 75

276

277

278

279

280

281

282

283

284

285

287

288

289

295

296

297

298

299

301

303

304

305

306

309

3"

312

313

314

3i5

316

317

318

319

321

322

323

324

325

326

327

328

329

330

33i

332

333

334

335

336

337

339

340

34i

342

343

344

345

346

348

349

350

35i

352

353

354

355

356

358

359

36i

363

364

365

Hebrew

. 11

H

15

19

20

21

27

28

32

38

42

46

81

82

86

90

9i

95

96

98

106

118

134

141

145

146

147

148

156

157

161

163

171

177

179

181

188

189

190

193

198

201

203

204

208

209

21 1

212

21S

219

223

224

225

233

235

236

2 39

242

248

253

269

271

276

279

282

286

288

289

291

292

2 93

294

298

300

301

303

304

307

30S

310

3"

3H

315

317

319

320

321

326

33i

333

336

33S

340

343

344

347

350

356

357

358

360

364

Ethiopic

. 1 1

12

27

44

87

98

"3

"5

119

136

149

156

160

173

189

191

230

243

248

261

264

276

287

291

294

303

308

3"

3-' 1

333

342

358

362

Syriac . .

. 16
317

94
321

101

33i

106

333

159
340

171

343

177
35o

179

iSS

248

267

282

298

302

Chaldee.

• 33

160

303

321

35o

Aramaic

• 203

208

338

360

P|'

Amharic

:.28o

329

363

k*:

Mahri . .

.167

312

/o

>8 A?

Sokotra

.312

Tigre. . .

.119

.«-,-->■

176 THE POLYNESIAN WANDERINGS.

Beyond this we may not venture. The effort has been made to
tabulate for each of these languages a scheme of consonant mutation,
and the effort has proved vain. The trouble begins even earlier than
the tabulation of variant forms.

There are cases in which there is a resemblance of form between
some Semitic vocable and some vocable in Efate and Melanesia and
Polynesia. If to the form resemblance were added correspondence
in signification, then we should not truly in any one instance have
proof that though widely separated in space they are homogenetic.
A single instance will suffice as well as a thousand to show the utter
lack of evidential value in such a case. The Hawaiian like means
to resemble, to be like, yet no one has yet arisen to predicate upon
the double identity of form and sense a theory that Hawaiian and
English derive from the same source, except in so far as Fornander's
project might be susceptible of extension to such an absurdity.

A long catena of such dual resemblances of the common element
in Melanesian and Polynesian with a Semitic parent speech would
add confirmation with each new link as welded to the chain. Yet
the weakness of any chain is in its weakest link ; that weakness meas-
ures its utmost strength. Many of these links have been vitiated
by the falsity in definition hitherto animadverted upon. We have
pointed out instance upon instance in which through ignorance, some
through manifest design, the definition of Efate vocables has been
distorted for the clear purpose of establishing a sense resemblance
where a form resemblance has already been noted, either by eyesight
or in a rather fertile imagination. Losing our faith in this link and
in that we can place no trust at all in the chain. So much for that.

We find Semitic triliterals proposed as the parents of vocables in
our Pacific islands in which the consonantal skeleton consists of but
a single consonant, or of two. We find island vocables having a
skeleton of three consonants proposed as in direct descent from
Semitic stems in which it is impossible to discover a trace of more
than two consonants, sometimes not even two. We find some Sem-
itic triliterals identified with triconsonantal stems in island languages
in which the order of the consonants is deranged. We find some
Semitic triliterals identified with triconsonantal island stems in
which the first Semitic element has vanished, the second and third
are respectively first and second in the island stem, to which is added
a third which corresponds to nothing in the Semitic; expressed dia-
grammatically it is sought to establish ABC as equal to BCD.

In vain we strive to unravel Dr. Macdonald's attempts at con-
sistent explanation. These things remain a tissue of irreconcilables,
and the explanation but serves the more to confound them.

Pretermitting as hopeless the task of comprehending these anoma-
lies so gravely proposed and so argued with what must be regarded

THE SAWAIORI BEGINNING RESTS UNKNOWN. 177

as fabrications which will stand under no law of evidence, assuming
that these obliquities can be made straight, we are to encounter a new
jungle of difficulties in the examination of phonetic mutation.

We have been able to develop for the Polynesian a very simple
system. In the primary and even in the secondary Melanesian
borrowings of Sawaiori loan material we have developed a system
that seems to hold good as far as it can be put to the test, and this
despite the confusions of the attempt of the Melanesians to catch
in unaccustomed ears and to reproduce with untrained organs of
speech, above all the lips, sounds unfamiliar. We have subjected
the diffusion of these elements in various Indonesian languages to
the same analysis.

What principle or principles of mutation have we been able to
discover in this trine examination as a factor common to the three
families which use this common stock of vocables and which possess
them in fee or as bailee? They are but few, these principles; corre-
spondingly they are simple :

i. The nasals tend toward a mutation, if any, backward in the direc-
tion of the glottis. Less frequent is variety in the series to which
each belongs. The mutation is horizontal rather than vertical.

2. The consonants in each of the series tend normally to mutation
downward in the series.

3. At the foot of each series, palatal, lingual, labial, the mutes
tend to mutation upward in the series, surd mute to sonant mute,
mutes to spirants.

4. Mutation extra seriem, horizontal mutation, is rare outside of
the nasals, most such cases being explicable as mutation to the aspi-
ration in the first instance and then secondarily from the floating
aspiration to some adjacent series.

How far do the proposed Semitic identifications conform to these
broad principles which, in the intricate detail of the study put upon
them in the foregoing chapters, have been established as perspicu-
ously as simply?

It would be idle to attempt to list all the concordances and equally
the discrepancies of the Semitic offered in identification when meas-
ured by these established principles. We note from the Arabic,
from which Dr. Macdonald has drawn most largely, the proposed
mutations in but two of the representative Polynesian consonants,
t as being central in the diagram and f as representing the labial
series and with it the maximum mutability. For each mutation we
note but a single instance in reference; it seems that there will be
no lively desire to seek out more.

Polynesian t may become in Arabic: / (107), t (44), d (38), s (35),
f (160), z (267), h (36), n (37), gr (247), rd (306), '(356).

178 THE POLYNESIAN WANDERINGS.

Polynesian f may become: / (86), b (83), m (93), n (170), w (296),

y (296).

We fail to see how such movement, even if established on far
better ground than we have here, can be brought into harmony with
the foregoing briefly stated principles.

These principles are fundamental in the three speech families
which share the possession of this common element in the vocabu-
lary. A fourth family, claiming admission to the clan yet showing
so plainly that it fails to conform to the law of the household, must
knock long at the door and long in vain.

One more point and we are done with Dr. Macdonald and his
Semitic origins. If he has proved his thesis his proof must exclude
all use of the same materials to prove some other origin. There is
another Richmond in the field, long earlier in the field, and he has
been just as substantial and just as stout in defense of his theory
that the Polynesians stem in the pre-Sanskrit Aryans. In the third
volume of Judge Fornander's "Polynesian Race" will be found
dozens of instances in which he uses the same Polynesian and the
same Indonesian material that Dr. Macdonald groups about his
Melanesian data for the proof of Semitic origin. Yet employing the
same material Fornander sees naught but Aryan source.

Let us compare the two lines of argument in but a single instance
to serve as illustration. Brevity will best be conserved by pinning
such comparison on 243. In this Dr. Macdonald carries the Samoan
fili, to plait, through Melanesian and Indonesian to the Arabic fatala
and jatV and the Ethiopic fatlat, to twist, to spin. Hear now Judge
Fornander as champion of the other cause : *

Greek, eUwf to roll up, to press together, pass to and fro, to wind, turn
round ; ^-iWw, turn round or about, roll, whirl ; eXtg, twisted, curled ; s, any-
thing of a spiral shape, twist, curl, coil; tXXut, to roll, of the eyes, to squint,
look askance; ?AA©?, squinting; Ma?, a rope, band; Uiy£, a whirlpool.

Sanskr., vel, vehl, to shake, tremble; vcllita, crooked; anu-vellita, a band-
age. To this Sanskrit vel Benfey refers the Greek sUai, the Latin volvo,
and the Gothic walojan. Liddell and Scott also incline to connect eUw and
volvo with the same root. To me it would seem as if the Sanskrit vrij,
whose "original signification," Benfey says, is "to bend," and the Sanskrit
vrit, whose "original signification," Benfey says, is "to turn," were nearer
akin to the primary form from which the Greek eUw, tXXm, and the Poly-
nesian hili, wiri, descend : that primary form being vri, now lost to the
Sanskrit, with a primary sense of to bend, twist, turn over, braid, and of
which vel, veil, or vehl, is possibly another secondary and attenuated form.
With such a Sanskrit vri, surviving in vrij and vrit, the derivation of the
Latin filum, thread, as twisted, spun; of the Latin varus, bent asunder,
parting from each other, varix, crookedness; of the Saxon wile, deceit;
of the Swedish willa, confusion, error, wilse, astray, becomes easy and
intelligible.

*3 Polynesian Race, 117.

THE SAWAIORI BEGINNING RESTS UNKNOWN. 179

It is impossible that each can be right in his deduction. To those
who have followed thus far this review of the data it will seem far
from impossible that each is in equal error.

In summation we are to consider what facts are established in our
knowledge of the earlier history of the Polynesians.

A. In Polynesia.

We have the excellent authority of concurrence of tradition, and
to those who can bring themselves into harmony with the Poly-
nesian manner of thought their tradition has the validity of history.
We shall find this history most succinctly set forth in the volume of
"Hawaiki" to which reference has been made earlier in this work,
and we shall find its several incidents as derived from diverse sources
most satisfactorily synchronized and intelligently discussed in the
same work. We have in the foregoing pages and in the data upon
which they are based a very considerable mass of language history.
Fortunately we are not under the necessity of estimating the com-
parative moment of each of these sources of information. That
would be a problem as interesting as intricate in case of conflict.
In our studies they run in confirmation and reciprocal corroboration.
For every inference to which the philological line may lead us we
find support in this fragment or that of some tradition; for every
statement set down in the tradition we find such corroboration in
the philological analysis that the legend handed down in memory
is proved to have the value of history.

Confirmed and upheld thus doubly at every point we are assured
of the following facts in Polynesia :

i. Nuclear Polynesia (Samoa the nucleus and Niue, Tonga, Viti,
describing the perimeter) was under settlement by Polynesians from
a date so remote that they had lost all direct memory of an anterior
movement thither. They held themselves autochthons, and in the
greater groups had creation myths in which land first emerged from
the tireless sea, their own the first of lands and they upon it the
first of men. These we style the Proto-Samoans. The indirect tra-
dition of a former home told no rearward tale to them. It is only
by inference and through digestion of many such traditions that we
are able to read into the consistent belief in the westward home of
the spirit a dim record of an earlier abiding-place. The dead go
home, home to a home that the living have long ceased to remem-
ber ; blessed are the dead in their direction sense.

2. Upon this Proto-Samoan settlement came a later wave of
migration of the same race. This second migration held its footing
upon Nuclear Polynesia through a period whose duration we are
quite without the data to estimate. In general the later migrants
behaved so harshly to the original inhabitants, albeit of their own

180 THE POLYNESIAN WANDERINGS.

race and almost word for word of the same speech, as to provoke
reprisals. For these later migrants we have adopted the name by
which they are known in Samoan history, the Tongafiti, it being
understood that the present names of the archipelagoes of Tonga
and Fiji (Viti or Fiti) did not supply the name, but are derived there-
from. From skirmish to pitched engagement these reprisals grew
as the Proto-Samoans, driven from the shore to inner recesses of
their islands, recovered strength in resistance. At last came the
critical battle of Matamatame, somewhere about 1 200 of our era or
a little earlier. The Tongafiti were expelled from Samoa and began
their eastward wanderings as far as Hawaii and New Zealand, the
era of the great voyages.

3. Nowhere in the present data are we able to pick up the track
of the Tongafiti prior to their descent upon Nuclear Polynesia. We
have made it clear that they did not follow the Melanesian route
between Indonesia and Polynesia. It must remain for the students
of the Tongafiti collaterals to discover their route; our concern in this
study has been to identify the migration that did sweep along the
Melanesian chain.

At this point it is profitable to add the comment of S. Percy Smith
upon a syllabus of my reasons for rejecting the Tongafiti migration
from the Melanesian area :

The Solomon-New Hebrides is not the only route open to them; they
may have stretched across from the north shore of New Guinea, or even
from the northern Solomons, to the Gilbert, Ellice or Phoenix Group and
so down to Viti. But for all that, until I see your argument, I must at
present think they came by way of the Solomon and Santa Cruz groups
to Futuna. The specimens I have seen of the dialect of Sikayana show
a close connection with Maori and Rarotongan. Just consider this case:
the Tongafiti came by the north of New Guinea following the first Samoan
route, and as they came across their own people in various places along
this route, such as in the Solomons, Santa Cruz and elsewhere, they would
learn of former migrations having gone farther south and followed without
delay, leaving none of their dialect behind among their fellow-countrymen
they fell in with en route.

The general argument has already been advanced at length. Spe-
cifically I note in comment on the foregoing interesting note, as
follows. Our data in Melanesia show a marked absence of traces of
the vocables and mutation forms which characterize the Tongafiti
in distinction from the Proto-Samoan. Furthermore the Tongafiti
have left abundant traces of their passage through Nuclear Poly-
nesia and all the more on that account is it inconceivable that they
should have quite failed to affect their congeners in Melanesia if
they had passed that way. I feel confident that a similarly careful
examination of the islands along the Line will disclose the Tongafiti
track.

THE SAWAIORI BEGINNING RESTS UNKNOWN. 181

B. In Melanesia.
From the Isle of Pines, which at the bottom of New Caledonia
sets the full stop to Melanesia, to the Admiralty Islands, which draw
the northern line of that island province around and overlapping
eastern New Guinea, we find three classes meeting our investigation.
In the first is grouped the islands of Polynesia's western verge in
which, ethnically and philo logically, we are dealing with Polynesians
as surely as if we were in Samoa or Te Pito te Henua at the eastern
limit; with this class, and because of its established Polynesian posi-
tion, we have little concern in the present series of studies.

The last class contains all those islands, if any there be, which
represent Melanesia uncontaminated by a Polynesian influence even
at second or third remove. This class also is removed from our
present study.

The central class is that with which we concern ourselves here,
those many Melanesian lands in which the language record enables
us to trace a Polynesian connection in speech, the amount and the
quality of such contamination varying largely from group to group
and from island to island and from the shore to the interior of
an island. It is upon this class that our attention is fixed in this
inquiry. It has been our task to analyze and identify each item
of the contamination in so far as we possess the record with which
to study it. It has been our duty to pass definitely upon each such
item, to reject or to admit it to Polynesian kinship as the facts may
seem to warrant. From the items so admitted we have sought to
comprehend the system of variation from the true Polynesian form to
which they have been subjected in passing into alien use. We have
massed these items to the proof that they are loan words borrowed
by Melanesia from Polynesians. We have sought to account for the
contactof thetwo races in this area. In following up and, it is hoped,
plainly establishing the overrunning of Melanesia by a Polynesian
migration swarm, we have essayed to direct attention upon two par-
allel tracks of swarming, parted far to the west and destined not to
reunite until a long eastward traverse has been concluded.

With the Melanesians themselves we have nothing to do save in so
far as we find them recording the passage of the Proto-Samoans, a
passage as to which the Polynesians have retained no direct memory.
It suffices here to state that the students best acquainted with them
regard them as a mixed race, the Polynesian admixture in blood
being more a matter of inference than a result of anthropometrical
investigation. There seems good reason to believe that under the
Polynesian admixture there is not one but several races. Up to the
present, on rather better grounds now than could be the case in even
the recent past, there seems to be a line of demarcation quite sharply
drawn between the Melanesians of the islands and the Papuans

182 THE POLYNESIAN WANDERINGS.

of New Guinea. To this conclusion point the brilliant studies of
Sidney H. Ray upon the Melanesian population of Torres Straits.
The same divergence, even within the most narrow limits, is pointed
out by Pastor Hanke as existing on the shores of Astrolabe Bay on
the northern coast of New Guinea ; the map accompanying his Bongu
dictionary, unfortunately without a scale for the measurement of
distances, shows Melanesian languages upon the offshore islands,
on the coast Melanesian languages in three areas abruptly separated
by two Papuan areas. It answers our purpose in these studies
to observe that the languages of Melanesia are not Polynesian, no
matter how much they may differ the one from the other.

In general, and this note must be understood as of equal appli-
cability to the Indonesian division of the topic, the culture plane of
the peoples is the predeterminant factor in regulation of the nature
and amount of the loan material which they may assume, the pho-
netic system of their own speech functions in the degree of the
assimilation to which they may subject the matter thus assumed.
In this work I have refrained from consideration of the former factor.
The data here presented, offering a record of probably very nearly
all the loan material, should readily provide the stuff from which
the student of manners might construct the history of the difference
in the arts which marked the Melanesian as lower than the race of
hardy and brilliant adventurers who swept past his islands these
ages ago and brought to him a glimpse of a world where achievement
ran higher.

So far as relates to the modification of the loan material in the
process of assimilation by the borrowing races I have been careful
to draft a table of such modification for every one of the languages
of which we have sufficient record, and in many cases I have supple-
mented this with a fuller discussion in the notes.

The conclusion to which I am led is that the element common to
Melanesian and Polynesian is Polynesian material directly impressed
upon the Melanesian by borrowing under stress of the lack of name
for a new object or a new idea, or by the influence of some quirk
of fashion, a principle no less operative in primitive man's mental
equipment than it remains to the highest culture attainment of the
summit races.

C. In Indonesia.

When we enter upon the island world to which New Guinea stands
as the eastern barrier we find an immediate and a great decline in
the element which has been found common to Melanesia and Poly-
nesia. Only a few of the vocables in Melanesia for which we have
discovered Polynesian affinities are found to carry that affinity back
to Indonesia. Still fewer are the words which display an affinity

THE SAWAIORI BEGINNING RESTS UNKNOWN. 183

between Polynesia and Indonesia without having left a record of
their passage through Melanesia.

The presence of these double and of these more common triple
affinities has served as the basis for the erection of the Malayo-Pcly-
nesian speech family. That family was created by great scholars
and has been supported by their followers no less great. One hesi-
tates to deviate from the conclusions upon which there is substantial
agreement of Humboldt, Bopp, Friedrich Muller, Max Muller, Whit-
ney, and the generality of students of systematic philology.

Yet now I have the less hesitation. In preceding works, where
topically, however, I was dealing only with the Polynesian, I found
myself forced to set aside the earlier estimate of the character of the
languages of the Pacific and to establish them as of the isolating
class,* a position in which I am more and more confirmed as my
studies go more deeply into the matter. Therefore I am ready to
pronounce the decree of divorce upon Malay and Polynesian. Lan-
guages of different classes, of uncoordinate syntax, of irreconcilable
vocabularies — too long have they been unequally yoked. In the
present collection of data there is not a single item which is not most
readily explicable as loan material, there is not one in which there
can be mustered any proof that its source was Indonesian.

D. The Old Home.

The material with which we have so long been engaged, and it is
hoped not without profit and interest, leads us from Polynesia back-
ward along Melanesia and to many a remote shore of the Malay
Archipelago. Did the Polynesians have no earlier history? Was
it in these warm islands that they became man and slowly acquired
that control over certain muscles susceptible of high specialization
in function which gave them speech?

A most interesting speculation. It has engaged the zeal of all such
as have felt the attraction of this least contaminated of the races
of men. Every shred of tradition has sedulously been studied for
such record as it might reveal. The interpreters of these tradition-
histories have been led back to Indonesia as distinctly as this mass
of linguistic material has led us who have been studying it together.
All paths lead to Indonesia as an early, a very early, home of the
Sawaiori.

But backward ?

Dr. Macdonald has toiled for a lifetime to prove a Semitic origin,
a yet eastern heme in the region to which the Bab el Mandeb is
indeed a gateway. I can not find that his theory stands the test
of examination.

"27 American Journal of Philology, 380.

184 THE POLYNESIAN WANDERINGS.

Myself, I have fancied that from the inner content of many of the
elemental words of the language I might reconstruct a vision of the
geophysics of the earliest home, ancestral Hawaiki in the great sea
of Kiwa. It was a pleasant speculation ; almost I could see the old
home. In casting about for a terrain which would in some sort
correspond to this artificial cloudcuckooland I was led to pitch upon
the Hadramaut, close to Dr. Macdonald's seat of origin. I have
found pleasure, better yet it has been given me to find enduring
profit in great joy, sweeping concentric circles of study upon the
languages that ring about the Hadramaut in Arabia and across the
straits in Africa. Yet nowhere have I found so much as a single
word upon which I might rely in confirmation.

Fornander, with much labor, has sought to find the origin of the
Polynesians in the origin of our own speech family. It has not
seemed to any Polynesian student necessary to enter upon a detailed
disproof of his argument ; every item sufficiently disproves itself by
the discovery of his complete lack of philological training and the
language instinct.

The school of students of Polynesian origins, that unanswered
"whence of the Maori," which has grown into enthusiastic existence
in New Zealand, has been sedulous in this study, at times almost
inspired. At present they are in general accord in regarding the
Polynesian ancestral home as somewhere in the Indian peninsula,
either in its great valleys of the Ganges and the Panjab, or else upon
the heights rising farther to the north. The best statement of this
opinion is most lucidly and most compellingly set forth in Percy
Smith's "Hawaiki."

I wish that I might take my stand with that rich scholarship which
has proved such an inspiration to me in my work. In following out
the only method which I feel sure in handling I can not go farther
than the material in my hand will lead me. At Java, or thereabout,
the last thread slips past. Up to that point I have followed the
leading toward Pulotu whither the dead go, toward Hawaiki whence
the living come, always westward with the words to go with me
from land to land — now at last the tale of the words is done.

I may go no farther. In Java I halt, and Java may be in itself
a Hawaiki. There is no further leading. Out yonder beyond my
sight, out yonder over the unended sea and the sun going down, out
and away whither my eyes tire with the strain of unavailing seeing,
somewhere lies the Hawaiki of our vain search. There I would that
I might see the canoes setting bravely forth with the rhythm of song
and the pulse of paddles, bravely out on the great sea of Kiwa, their
crews the forebears of that race of men who beyond all others made
the sea their own, even to its uttermost islands.

APPENDIX I.
DATA AND NOTES.

EFATE-VITI.

I.

baso, to pierce.

Viti: veso-ka, id.

2.

bei, preposition connecting verbs with their objects ; the final i belongs to

the pronoun of the third person.

Viti : vet, to, from ; used only before personal pronouns and personal

names.

3.
kilakila, to be shy.

Viti: kila, to be wild (of animals), suspicious.

4-
malua, niailua, malilua, malulu, to do anything gently, to be in no hurry,
to do after a time, by-and-by.
Viti : malua, to go gently, to be in no hurry, by-and-by ; vakamalua,
gently.
This may be found associable with malum 316.

5-
masere, to be torn.

lg|iN Viti: kasere, broken, loosed.

This falls more properly under note 21.

6.
mutrei (given as a variant of mitei 232), breadfruit fermented and preserved.
Viti: mandrai, bread, i. e., a cake of preserved breadfruit.

7-
sanga, senga, a crotch, a fork made by two branches; sanga-fi, to take
hold of with a crotch or forked stick.
Viti : sanga, a crotch, the thighs, a pair of tongs ; sanga-va, to take
hold of with tongs.

8.
toki, to gather up one's things or pack up preparatory to flitting.
Viti: toki, to remove one's goods and residence.

EFATE-VITI-MALAY.

9-
bara ti, to bind together ; farati, sticks fastened above and upon the rafters
of a house.
Viti: vorati, the wind beams or upper small cross-beams of a house.
Malay: barot, to gird, to bind around; baroti, rafters.

185

186 THE POLYNESIAN WANDERINGS.

10.

buta, blind (in meta-buta, lit. eye-dark).

Viti: mbutb, dark, darkness; mat ambuto, to become dizzy and fall,

lit. eye-dark.
Malay: buta, blind.

EFATE-VITI-SKMITIC.
ii.

ba, ba si, fa, va, to come, to enter, to tread (to go upon), to tread upon.
Viti: va-tha, to tread upon.
Hebrew: bo, ba. Ethiopic: bawi, to come, to enter.

This naturally suggests an association with bano (147) to go. The Efat6
ba-si implies a final ^--radical, which implication is further hinted by Viti
vaiha, for s-th is a not infrequent mutation. In bano, however, the radical n
is constant throughout Melanesia with the exception of a New Hebridean
(Mota-Ambrym-Sesake-Eromanga-Pak) and a Solomon (Nggela-Belaga)
group, which have va. In view of the constancy of n in bano and the appear-
ance of s in this item it is uncertain whether to regard this ba as a distinct
root and to transfer from 147 the two va groups to association therewith,
or to regard the latter as mere coincidences in the result of abrasion. With
this bd, Dr. Macdonald associates as a dialectic form mai to come, which
is, of course, nothing but the universal Polynesian directive.

12.

bila i, to pick up, to gather up anything, as fallen leaves, fruit, fish lying
on the ground.
Viti : vili, to pick up fallen fruit or leaves.
Ethiopic: 'araya, to gather, as fruit, herbs; to glean.

*3-
bitelo, butol, bitol, to be hungry.

Viti: vitolo, hunger, to be hungry (an unusual dialectic word).
Nggela : vitolo, to hunger. Motu : hitolo, hunger.
Arabic : talaha, tolilia, to have an empty belly.

The first classification of these data was based upon the comparative
material afforded by the Efate dictionary. Later research through other
sources of information have in several items brought to light new data
which interrupt the applicability of the class heading, as in this case. The
serial number of the items, however, had by that time been so extensively
employed in man)* calculations that it has seemed hardly worth the while
to recast the arrangement and to provide a new notation.

Peculiar interest attaches to the entry which conveniently, yet inaccu-
rately, is credited to the Motu. It is drawn from a vocabulary of 160
items collected by F. R. Barton and included in Seligmann's "Melanesians
of British New Guinea." Captain Barton designates this vocabulary by
the name "Lakatoi Language" and notes that it is "the trading language
spoken by the Elema natives and their visitors." In this remarkable
lingua franca I have been able to identify no more than this item as having
any association with the languages of our province. The lakatoi are the

DATA AND NOTES. 187

composite vessels which set out fromMotu laden with pots and sail westward
to the Papuan communities across the Gulf of Papua, there to barter their
wares at Elema and other ports for the sago of that country. This voyage,
an adventure of annual argosies, is performed with great ceremony and much
ritual, not the least of which formality is the employment of this language.

14.
bure, fure, to wash, to rub.

Viti : mbore-a, to scrape or wash the dirt off a thing, to brighten ; a

dialectic word.
Hebrew : marak, to rub, to polish, to cleanse by washing.

15.
fanau, bunu, to teach, to instruct, to preach.

Viti: vunau, to admonish, to harangue, to preach, a speech, reproof.
Hebrew: 'anah, to harangue, to proclaim, to preach, to admonish.

ka, k\ tense sign, past.
Viti: ka, id.
Syriac: ka, ga, id.

16.

17-

kilakila, knowing, sagacious.

Viti: kila, to know, to understand, to acknowledge.
Arabic: 'akala, 'akil', to be intelligent, prudent, sagacious.

18.

lele, lili, to wind, to go around, to turn, to curve; malele, to be bent or
curved, as a branch of a tree heavy with fruit.

Viti: lele, the end of a branch farthest from the body of a tree;
leletha, to bend a branch in order to gather the fruit on it.

Arabic: lawa, to wind, to bend, to turn.

19.
lume a, to wash (immerse), to dip.

Viti : lomotha, to dip, to dye, to daub the hair with ashes, to dip the

head into urine to clean or stiffen the hair.
Hebrew: saba1. Arabic: sab"a, to dip into, to immerse, to dye.

20.
saf i, bisab, bisif, to excel.

Viti: sivia, to outstrip, to exceed, to pass another, to get past or

before, to surpass; uasivi, to exceed.
Hebrew: yasaf, to add, to increase, to surpass, to excel.

21.
sere, masere, to tear.

Viti: sereka, to untie, to unloose; kasere, broken, undone.
Arabic: nasara, to tear, to rend.

22.
seri, bakaseri, to loose a tabu.
Viti: sereka, to untie.
Hebrew : sarah, to loose.

188 THE POLYNESIAN WANDERINGS.

23-
si, soi, to scrape.

Viti: soya, soi, to scrape off the skin of yams for boiling.
Arabic : saha', to scrape off.

24-
tefa ki, tefa-ngi, to put things in a series, to range.

Viti: tuva, to place in regular order, to range.

Arabic: saffa, to set or place in order in a series, to arrange.

It is at least suggestive of association with Samoan tufa, to divide, to
share out, to distribute. The Viti form readily consents.

Tonga: tufa, to divide, to portion out, to distribute. Niue: tufa-
tufa, to divide, to portion out. Uvea: tufaki, to distribute,
to divide, a partition. Futuna : tufaki, to distribute, to divide,
Fotuna: no-tufa, to give. Tahiti: tufa, to share or divide
portions. Marquesas: tuha, to divide, to apportion. Rapanui,
Mangareva: tuha, to divide, to share out. Maori: twwha, to
distribute, to apportion.

Nggela: tutuva, to distribute (food). Belaga: tuva-lisa, to

apportion. Sesake : cf . ga tova wango, I cut pig.

In note 31 2 I have suggested the recurrence of this stem in the compound
numeral of the tavalima series.

25-
tirikit, to begin to drop or spatter, of rain (kit, small).
Viti: tiri, to drop, of liquids.
Arabic: s'als'ala, to drip, to fall in drops.

This is properly included in 359.

26.
tu na, bones of fish.

Viti: ndua, bone (dialectic).

Arabic: s'a'a, to become spiky, to be rayed.

EFATE-VITI-MALAY-SEMITIC.

27-

bera, fera, to crumble, to fall to pieces.

Viti: vuruvuru, a crumb, to crumble.

Malay: author, tabur, to be scattered. Malagasy: mahavera,

mivcraberaka, to crumble.
Ethiopic: farfur, a crumb. Hebrew: parpor, a crumb; pur, par

pirper, to break in pieces. Cf. 28.

28.
bori, boriuori-si, to break.

Viti: vorota, to break, of brittle or thin things, as pots.

Malagasy : puritra, to break.

Hebrew: por, pur, to break in pieces. Cf. 27.

DATA AND NOTES. 189

29.

lita, liti, to crackle, to burst, to explode, as wood or a stone in the fire;

lita nakabu, a spark.
Viti: lindi, thalindi, to burst or explode, the report of an explosion

or bursting, as of thunder or a stone in a heated oven ; lindi ni

buka, a spark.
Malay: latok, latup, latub, to crackle, to crepitate; latum, to boom,

to give out a booming noise.
Arabic: la^ata, la"t, li"at, to crackle, as water boiling.

. u 3°-

lubwa, to pour out.

Viti: livia, talivi, to pour gently or in a small stream, to spill.

Malay: tumpah, manumpah, to spill, to shed, to pour out.

Arabic: sabba, to pour out.

In 154 the same Viti identification has been employed in a better set of

sense resemblances.

31.
sabe-li, to beat, to slap.

Viti: sambalaka, to strike in a certain way, to slap.

Malay: tampar, to slap. Java: tampel, id.

Arabic: safa'a, to beat, to slap.

32.
samit, samut, to beat, to chastise.

Viti : samuta, to beat, generally with a heavy stick.

Malay: chamiti, chamati, a whip or scourge.

Hebrew: s'amat, to strike, to smite. Arabic: s'amat, to whip.

The identification of the Malay entails the s-ch mutation, for which our
material affords no confirmation save so much as may lie in 45.

33.
siba, suba, to break.

Viti: sovetaka, to break the head to pieces; sovuta, to break a hole

in thin things.

Malay : sumba, simba, to break.

Chaldee: s'ibeb, to break in pieces; s'iba, a fragment.

We are to meet with no other instance of a v-mb mutation in Indonesia
and only one (Malay 310) of v-b; the identification, therefore, lacks con-
firmation.

34-
son, to give.

Viti: solia, to give, to grant, to permit.

Malay: sdrah, srah, to submit; sdrah kan, to give.

Arabic: s'ara'a, to submit, to give.

35-
tatalai, to warm oneself at the fire.
Viti: tatalai, id.
Malagasy: mitulu, id.
Arabic: sala, salyy' , to warm, to be warmed at the fire.

190 THE POLYNESIAN WANDERINGS.

36.
tiro, to sink, to roll down.

Viti: tiro, siro, sisiro, to descend, to go down a steep or hill.

Malay: turun, to go down.

Arabic: hadara, hudur', to descend, to put down.

37-
toto, tiso, to exude, as gum or juices from plants.

Viti: ti, titi, titiva, to ooze, to flow gently down, as gum from a tree.
Malay: titik, to exude. Malagasy: viitete, mitate, tetevana, id.
Arabic: nas's'a, to exude.

38-
trdm, torn, turmeric, a reddish curry powder.
Viti: ndamu, red, crimson, brown, dun.
Malagasy: tamutamu, turmeric; tumamutamu, yellow.
Arabic: 'adoma, to be red. Hebrew: 'adamdom, reddish.

EFATE-MELANESIAN-VITI.

39.
bati, tooth.

Viti: mbati, tooth.

Sesake: mbati, tooth. Epi: bati, id. Marina: peti, id.
f- Dr. Macdonald has considered this under two entries.yet with full appre-
ciation that his nabati is this bati with the article. It will be found fully
discussed in 259, where, however, the stem under examination is that of a
different word and of much wider dissemination. This bati is confined to
Viti and the central New Hebrides.

40.
bui na, bua na, backbone, tail, rump.

Viti: mbui, tail.

Sesake: mbuena, tail.

4i-
mako, maka, offspring.

Viti: makumbu, mokunibu, grandchild (mbu, grandmother).
Tangoan Santo: maka pi, grandchild, offshoot of pi (grandfather).

EFATE-MELANESIAN-VITI-SEMITIC.

42.
fara, a sprouting coconut.

Viti : vara, a coconut when filled with meat and ready to shoot.
Motu: vara, to grow, to be born. Omba: kmbwiri, to grow.
Hebrew: par ah, to burst forth (as the young from the womb), to
sprout; perah, a sprout, a shoot. Arabic: farh", offspring,
sprout, shoot.

loa, black, dirt. 43'

Viti: loa, black.

Malekula Pangkumu: roro, dirty.
Arabic: lo'wat, lawla', blackness.

DATA AND NOTES. 191

44.
taku na, the back.

Viti: ndaku, the back.

Epi: taka, the back. Malo: tura, id. Motu: ndolu, id.
Ethiopic: dahr, posterior part ; dahari, the last; dahara, to be after,
to be behind. Arabic: fahr', the back.

Dr. Macdonald finds kinship with Samoan tua the back. Neither that
nor the Malo and Motu identifications can be sanctioned.

EFATE-MELANESIAN-VITI-MALAY-SEMITIC.

45-
bisa, fisa, basa, to speak.

Nukuoro: pasa, to speak. Fotuna: visau, id.

Viti: nosa, to speak.

Sesake: vasa, to speak. Nggela: bosa, id.

Malay: bacha, to read, recite, chant; basa, voice, speech. Tagalog:

basa, to speak.
Arabic: nabasa, to speak, to peep or chirp; nabsat' , a word.

An interesting suite, all the more because of the sparsity of the occur-
rence of the stem in each of the provinces. The proposed Semitic affinity
contains at least one element which is not found at all in the Sawaiori.

46.

kan i, kanikani, to eat; kanien, food.

Viti: kana, kania, to eat; vakania, to feed, to cause to eat;
veikanikani, devouring one another; kani, laukana, edible.

Melanesia, all signifying to eat — Marina, Tubetube, Galavi, Boniki,
Mukawa: kani. Sesake: ganikani. Galoma: ganigani.

Belaga, Sinaugoro: gani. Rubi: gania. Gog, Merlav: gan.
Nggela: gana, ganigi. Mota: ganagana. Retan, Lo, Mota,
Maewo: gangan. Duke of York: wangan. Leon, Sasar: gen.
Vuras : gengen. Ambrym : ngene. Nguna : ngani. Ambrym :
ngene. Gog: ngongot. Buka: nan, nanni, tuanan, iana
(restricted to cannibal eating). Duke of York, Roro, Uni,
Pokau, Kabadi, Motu, Hula, Tavara, Awalama, Taupota, Wedau :
ani. Mekeo: angi. Keapara: hani. Sariba: kai. Suau,
Mabuiag : ai. Dobu : c'ai. Kabadi : ania. New Britain :
an, ian.

Malay: makan, to eat. Malagasy: hunt ana, hanina, id. Togean:
mokonie, id.

Arabic: 'akala, to eat; 'akxV , messmate. Hebrew: 'akal, to eat.

Although the alternative form nganikani points (see note 117) to a
Polynesian stock, it is impossible to link kani with the Polynesian kai to
eat. While the omission of a medial consonant sometimes takes place, and
in the tabulation of the phonetic results we have a solitary example in
inum (32 1 ) to drink Marina o'omia, yet if it were far more common it would

192 THE POLYNESIAN WANDERINGS.

still remain impracticable in this instance for the distinct reason that we
are quite unable to adduce a single word in which all Polynesia has dropped
a medial consonant which all Melanesia has retained.

Viti shows that a or i may be the final vowel. The z-forms are found in
Efate, Viti, Marina, Sesake, Belaga, Nggela, Nguna, Duke of York, Buka,
Kabadi; to which may be added an e-form in Ambrym. The a-forms
occur in Viti, Nggela, Mota. Terminal abrasion is found in Gog, Merlav,
Retan, Lo, Maewo, Duke of York, Nguna, Buka, Kabadi, New Britain,
Leon, Sasar, Vuras.

The earlier vowel of the radical is found as a in Efate, Viti, Marina,
Sesake, Belaga, Gog, Merlav, Nggela, Mota, Retan, Lo, Maewo, Duke of
York, Nguna, Buka, Kabadi, New Britain. An e-form occurs in Ambrym,
where it is associated with final e, and in Leon, Sasar and Vuras. If indeed
it be the same radical an o-form is seen in Gog.

The initial consonant k occurs in Efate, Viti, Marina, Sesake. It passes
from surd to sonant g in Sesake, Belaga, Gog, Merlav, Nggela, Mota,
Retan, Lo, Maewo, Leon, Sasar, Duke of York, Vuras, Ambrym. It goes
still higher in the palatal series to ng in Nguna, Ambrym and perhaps Gog.
It is abraded in Duke of York, Kabadi and New Britain. The second
consonant remains n without change in all these cases, save again the Gog
anomaly and the Mekeo.

We therefore diagram this radical thus:

I a I |
K > hi'e

I e I
ng J J la

The anomalies are now to be considered. With a vowel not elsewhere
found, with a change n-ng which the radical but once elsewhere exhibits,
and with the assumption of a final t which elsewhere does not appear,
Gog ngongot is not acceptable in this record. The only way by which it can
be included is to assume it to be a composite form, ngon-got, and this parting
of the palatal nasal would be wholly indefensible. The remaining anom-
alies, at the extreme north of the Samoa track, suggest an explanation by
a developmental series. Frontal abrasion is sufficiently well established in
the Melanesian handling of Polynesian loan words to admit Duke of York
ani to the kani radical. This admission will naturally carry with it Kabadi
ania (a the common verb isolating objective suffix), and by terminal
abrasion New Britain an. This an may be taken to admit New Britain
tan, and this in turn carries with it Buka iana. The other Buka forms
involve a further leap. Reverting once more to Duke of York ani, if we
can regard nanni as in some sort a reduplication (which it must be ac-
knowledged is informal) that would carry nan with it and for tuanan we
should have to postulate a composite of this nan. This is far too involved
to be satisfactory.

Our Indonesian material is based on a radical m-k-n and the identifica-
tion should be overruled. The Malagasy is at variance with Indonesian
and Melanesian alike and should be rejected. It is impossible to see that
the Semitic has anything to do with the case.

DATA AND NOTES. 193

47-

tai na, brother's brother, sister's sister.

The Proto-Samoan radical is tehi, as shown by Niue, Tonga and Uvea.

Samoa: tei, younger brother or sister. Niue: tehina, younger
brother. Tonga : tehina, younger brother or sister. Uvea :
tehina, brother, sister. Rarotonga, Paumotu, Mangareva,
Tahiti, Maori: teina, younger brother or sister. Marquesas:
teina, a younger brother. Sikayana : teina, brother. Futuna :
taina, brother's brother, sister's sister. Hawaii: kaina, the
younger of two brothers or of two sisters.

Viti : tathi, a younger brother or sister.

Mota: tasi-k, brother. Malo, Laur, Lamassa, Lambell, Kiriwina,
Dobu: tasi, id. Epi: fo/w, id. Motu : tadi, younger brother
or sister or cousin. Malekula : test, brother. King: dirsi, id.
Bauro: asi, id. Santo : tesi-na, his younger brother. Pala:
tes, tasi, brother. Keapara, Hula, Galoma: art, id. Waima:
hati, id. Kabadi: kadi, id. Misima, Panaieti: tari, id.

Sinaugoro: tali, id.

Malay: adik, brother. Bugi: ami, id. Malagasy: zandri, id.

Arabic: rasi', brother.

The initial consonant is t in Polynesia (Hawaii k), and in Melanesia, with
the single exception of Bauro asi where it is dropped ; this is our only Bauro
word, so that we are without information as to the frequency of the frontal
abrasion of this mute, but it is not unknown in Melanesia. The change to
d in King is normal.

The aspiration is found in Epi ; it undergoes the normal transition to the
sibilant in Mota, Malo, Laur, Lamassa, Lambell, Malekula, Bauro, Santo;
its mutation to th in Viti is usual; the sound in King is not quite clearly
identified, it was collected by the same German explorer who gathered
Laur, Lambell and Lamassa ; Efate alone drops the aspiration ; the Motu
d is normal to that New Guinea settlement; and elsewhere in Melanesia,
involving the surd instead of the sonant, is found in Alite, Saa and Wango.

The final vowel is constant.

Of the former vowel there are two main groups. Polynesia generally,
and Malekula and Santo have e; Melanesia generally, and Viti, Futuna and
Hawaii have a. King again stands in a class by itself with i.

In Santo the na is clearly possessive suffix, such also is the k in Mota.

The only common element in Indonesia is a-i parted by a consonant,
which is not the aspiration or a normal mutation thereof; the initial con-
sonant is absent. We can not accept this identification. The Malagasy
may be associable with Bugi, certainly not with the Proto-Samoan.

EFATE-POLYNESIAN.
48.
alo, belly, abdomen, the front, before.

Samoa: alo, the under side (as of a cloth or the belly of a fish), a
chief's belly, the seat of the affections. Tonga: alo, the

abdomen of great personages. Futuna : alo, the entrails, in the

194 THE POLYNESIAN WANDERINGS.

presence of, before. Maori: aroaro, the front, presence, face.
Tahiti, Mangaia: aro, id. Hawaii: alo, id. Marquesas: ao,
before, in front. Rarotonga: aroaro, the presence. Manga -
reva: aro, the presence, before. Paumotu: aronga, the visage.

49.

bafano, fafano, to wash the hands.

Samoa: fafano, fanofano, to wash the hands and rinse the mouth.
Tonga, Futuna, Niue, Uvea: fanofano, to wash the hands.
5o.

bwefe, fefe, oven cover made of leaves.

Samoa: veve, leaves covered over an oven to keep in the heat.
Futuna : veve, name of a mat of leaves of coconut or other plant
placed on the viands in an oven to keep them from being soiled
by the earth which covers them.
5i.
boboi, a mask, cover or disguise.

Tonga, Uvea : fufu, to hide, to secrete, to conceal. Futuna, Niue :
fufu, id.

52.

bukubukura, full of little swellings, pimples.

Samoa : po'u, pimple. Futuna : poku, id. Tonga : bokubaku, a scab.

53-
fet, a bird's nest.

Samoa: fataninga, a bird's nest. Niue: fata, id. Uvea: fatal, id.

Tonga : bununga, a nest.
Nggela: niku, a nest. Bugotu: gniku, id. Mota : nig, to build
a nest.

54-
ata, a man.

Samoa: ataWi, (li'i, small) son, lit. little man. Rapanui: atariki,
oldest son. Nuguria : atariki, son-in-law.

55-
mafa, to be broken, cracked.

Samoa: mafa, ora vaginae aperta. Tonga: mafaa, to open, to
extend; mafaafaa, split, cracked, choppy. Futuna: mafaa,
broken, cracked. Niue : mafa, mafafa, a crack, a rift. Hawaii:
maha, to make a rent or hole in. Nukuoro : mahaa, to break.
Uvea: mafaafaa, to yawn, to gape.
56.
niaia, a species of banana.

Samoa : mamae, one kind of banana. Tonga : mamae, the plantain.

57-
makinikini, to be itchy.

Samoa: ma'ini, to tingle, to smart. Tonga, Futuna: makini, id.
Niue: maeneene{9), to tingle.
The absence of the k in Niue is anomalous and points to an association
with Samoan eneene to tickle. Cf. 227.

DATA AND NOTES. 195

58.
manu, a multitude, a very great number, a thousand.

Samoa: mano, manomano, 10,000, a myriad, a great number, innu-
merable, the limit of counting. Tonga, Nuguria : mano, 10,000.
Maori, Rapanui, Tahiti: mano, 1,000. Hawaii: mano, 400,000,
multitudinous. Mangaia: mano, 2,000, innumerable. Mar-
quesas : mano, 4,000, any great number. Paumotu : manomano,
innumerable. Nukuoro: mano, 100; mano-tini, a very large
number.

59-
manubunubu, nobwanobwa, to be soft, sleek as the skin of a newborn pig
or infant.
Hawaii: nopunopu, thoroughly cooked, soft, spongy, large, plump,

fat, swelled out ; nopue, plump, round, as a well fed fat hog.
Oiun: nubanuba, soft, as cooked food. Kiviri: nunubas, id. Motu:
manokamanoka, id. Pokau : manomano, id. Kabadi :manova, id.

60.
mataisau, matakseu, a carpenter.

Samoa : mataisau, an honorific term for carpenter.

The derivation from mata as "the eye (or director or master) of cutting"
does not seem so valid as that in which I have assigned it to the Samoan
matai, one skilled as a master of craft, and sau, a particularized cutting as
shown in saupapa to cut off the outer part of a log to make it level and
smooth. The Viti matai means carpenter and then by extension a mechanic
of any craft.

61.
mauta, mautu, rising ground.

Polynesia, all signifying mountain: Samoa, Futuna, Mangaia,
Maori, Rarotonga, Manahiki, Bukabuka, Fakaafo: maunga.
Paumotu : mahunga. Hawaii : mauna. Sikayana : fakamauna.
Tahiti: maua. Mangareva: manga. Tonga, Uvea, Niue,
Rapanui: mounga. Nuguria: mauna, mouna.

The derivation suggested as from man to remain firm does not particu-
larly appeal to me, although mau-nga is in form a typical noun-making
from an attributive in which the verb sense has so strongly developed as
to call for such differentiation. I set contra the note that in the languages
in which maunga has undergone vowel change {mounga, manga) the pro-
posed radical mau as verb remains unaltered.

62.
me, name, namai, a rope, a string.

Samoa, Tonga, Futuna, Uvea: maea, a rope, cord, cable. Nuguria:
maia, a band.

63.
mitariki, the Pleiades.

Polynesia names the Pleiades thus: Samoa: matali'i. Futuna,
Tonga: mataliki. Fotuna, Nuguria, Maori, Mangaia, Ma-

ngareva: matariki. Tahiti: matarii. Hawaii: makalii.

Marquesas : mataiki.

196 THE POLYNESIAN WANDERINGS.

Micronesian names for the same constellation : Ponape : makeriker.
Lamotrek : magarigar. Yap : magirigir. Mortlocks : marikir.

EFATE-VITI-POLYNESIA.
64.
bule, a shell, lit. gleaming, shining, glittering; cf. bila 284.

Samoa : pule, a white cowry, general name for marine shells ; pulepule,
small shellfish; pulei, to be checkered, to be mixed alternately
as different colored beads in a necklace; pulepule, spotted,
striped with various colors. Tonga: bule, a cowry; bulebule,
a shellfish; bulevaka, the white cowry; bulebule, spotted.
Futuna: pule, a univalve shell; pulepule, striped, spotted
with various colors. Niue: pule, the cowry; pulepule, striped,
variegated. Uvea: pulepule, variegated. Hawaii: pulepule,
varicolored, spotted, speckled. Tahiti, Rapanui: purepure,
spotted, checkered. Mangareva: akapurepure, to color, to
variegate. Paumotu : hakapurepure, to dye, to color. Maori :
opure, pied, variegated. Rapanui, Nuguria: pure, a shell.
Viti : mbuli, the white cowry.

Mota: pule, a very dark cowry shell. Miriam: ka-(a shel\)-buli-
buli, a small univalve.

65.
fiso, an annual reed-like plant whose top is used for food.
Samoa : fiso, wild sugar cane. Proto-Samoan : f iho.
Viti: vitho, wild sugar cane.
Mota: viso, a reed with edible flower head. Motu: hido, id.

66.
ngongo, an aquatic bird, to wade.

Samoa: ngongo, the tern (Sterna longipennis) . Tonga, Niue:

ngongo, the sea gull. Futuna: ngongo, the name of a bird.
Viti: ngongo, a sea bird.
The proposed derivation through Samoan 'a'au to swim from Arabic
hamma is not convincing.

67.

kie, the plant whose leaf is baked, dried and split into threads to be woven

into mats.

Maori: kiekie, the name of a climbing plant (Freycinetia banksii),

the leaves and fiber were formerly used in making fine mats.

Tonga : kie, a mat ; kiekie, a girdle. Futuna : kie, a mat ; kiekie,

a species of liana. Rarotonga : kie, the Freycinetia banksii.

Marquesas : kiekie, moss resembling a fine beard. Tongarewa :

kie, the pandanus leaf, the mat woven therefrom. Mangareva:

ngie, small pandanus leaves used in mat-making. Samoa:

'ie, fine mats; 'ie'ie, a Freycinetia used in making fish traps.

Tahiti : ie, a sail (of matting) ; ieie, fibrous roots used in basketry.

Hawaii : ie, a vine used in basketry, material braided into mats.

Viti: kiekie, fine mats, pandanus whose leaves are used in mat

making.

DATA AND NOTES. 197

68.

langa ti, langa i, langai, to raise, to lift up.

Samoa: langa, to raise, to rise. Tonga: langa, to raise up the soil;
fakalanga, to raise up. Uvea, Futuna : langa, to raise. Niue :
langa, to rise against ; langaaki, to raise up. Nukuoro : langa,
to float. Hawaii: lana, id. Maori: ranga, to raise, to cast
up. Mangareva: ranga, to float on the surface of water.
Paumotu: fakaranga, to raise, to lift up. Tahiti: toraaraa, to
raise up. Marquesas: aha, ana, to swim on the surface.

Viti : langa, to be lifted up, said of a brandished club.

Mota: langa, to lift up, to turn up.

69.
langilangi, to be proud, uplifted.

Samoa : fa'alangilangi, to be angry because of disrespect. Tonga :
langilangi, powerful, great, applied to chiefs; fakalangilangi,
to honor, to dignify, to treat with great respect. Hawaii:
lanilani, to be proud, to show haughtiness. Uvea: fakalai,
to compliment, to adulate.
Viti: langilangi, proud.

70.
lofa i, to bend.

Samoa : lofa, to cower down, to crouch as a dog. Tonga : lofatia,

obedient, yielding, respectful. Proto-Samoan : Iofat.
Viti: lovetha, to bend.

7i.
mono-ti, munu-ti , to close, to plug, to stop, to block up ; though given as a
dialectic form of bono-ti it is probably of a different stem.
Samoa: momono, to cork, to plug; monomono, to calk. Tonga:
mono, to squeeze, to press in : monoji, to cork ; monomonoji, to
patch, to mend. Futuna: mono, to calk, to bung, to plug.
Uvea: momonono, to patch. Proto-Samoan: monot.
Viti : mononotaka, to stop up sennit or other holes in a canoe with
breadfruit gum, to calk.

72.

sakau, a reef.

Samoa: a'au, a coral reef. Tonga: hakau, a sunken reef or rock.
Maori : akau, the coast. Mangareva : akau, lowland, shoal, a
ridge of rocks. Paumotu : akau, a reef of rocks. Mangaia : akau,
a coral reef. Nuguria: agau, id. Proto-Samoan: hakau.

Viti: thakau, a reef.

EFATE-MELANESIAN-POLYNESIAN.

73-
ata na, spirit, soul, shadow, image.

Maori: ata, morning personified, a reflected image, shadow, spirit,
soul. Samoa: ata, shadow, spirit, dawn, reflected image.
Tonga : ata, a shadow, the dawn, to reflect. Niue : ata, shadow,

198 THE POLYNESIAN WANDERINGS.

reflected image. Futuna: ata, shadow, twilight. Fotuna:
ata, ghost, shadow, image. Tahiti: ata, shadow, twilight.
Marquesas: ata, shadow, likeness. Mangaia: ata, shadow, soul,
dawn. Mangareva : ata, shadow, image, twilight. Nukuoro:
ata, shadow. Fila : ata, the soul. Uvea : aata, shadow ; ata,
image. Moriori: ateata, dawn. Hawaii: aka, shadow,

likeness, dawn. Nuguria: te ata te mahina, the waning moon.
Rapanui: ata, image, picture, dawn, break of day, close of day;
ataata, close of day.
Mota: ata, soul. Bierian: ata mate, ghost.

74-
ati na, nati, nutu, child.

Maori: ati, descendant, a prefix to tribal names as descendants of
certain persons. Samoa: ati, a particle denoting a number
of chiefs of the same name or title. Tahiti : ati, patronymic
prefix grouping the name of an ancestor with the descendants.
Mangareva : ati, descendant. Also, Maori : ngati, patronymic
tribal prefix, descendants of. Mangaia: ngati, a tribe.

Omba : nati, native of. Makura : nati, son, people of. Bierian :
nati, son. Makelula: anati, netin, child. Malo, Nguna:
natu, son. Baravon: natu, son. LambeH: natu, child.

Wagawaga: natu, son, child of a woman's sister. Tubetube:
natu, child, grandchild, child of a woman's sister, child of brother
of man or woman. Motu: natu, son, daughter, brother's son.
Sinaugoro, Suau, Sariba, Panaieti, Dobu, Tavara, Awalama,
Taupota, Wedau, Galavi, Boniki, Mukawa, Kubiri, Raqa, Kiviri,
Oiun: natu, child. Kabadi, Pokau, Doura: naku, offspring,
child. Roro, Hula: nahu, child. Kiriwina: latu,\&. Keapara,
Galoma: nau, id. Mekeo: ngaunga, id. King: nutu, child.
Lamassa: nutu, child. Matupit: tu, son. Duke of York,
Pala, Moanus: nat, child. Lamassa: fandt, child.

The existence of the ati-nati in Efate and Maori ati-ngati shows that the
common Polynesian forms are the result of frontal abrasion. The pre-
sumably older complete form holds throughout Melanesia, except as noted
below, and the n undergoes no change. So, too, the second consonant
remains unaltered except that in Kabadi t-k shows a far western instance
of a change which is resistless in parts of Polynesia to-day. The former
vowel remains without alteration throughout Melanesia except for the
w-form of King and Lamassa on the New Ireland coast, yet the presence
of the a in Duke of York within eyeshot shows this a merely local variant.
The principal variation involves the final vowel. Polynesia and the central
New Hebrides (Efate, Omba, Makura and Malekula) have i. Nguna in
the same tract of the New Hebrides has u, thence we encounter no record
of the word until we find the w-form in the extreme north in Torres Straits,
Baravon, Lambell, King and Lamassa. Duke of York and Moanus nat
may be an abrasion of nati or natu, but from the fact that they lie in the
natu region the latter is probably their source. Matupit tu and Lamassa
fandt are included more on account of general suggestion of resemblance
than from established identification.

DATA AND NOTES. 199

75-
matulu, matultul, matoltol, to be swollen, thick.

Samoa: mdtolutolu, matoutou, thick (restricted to pork). Nukuoro:
matolutolu, thick. Tonga, Futuna, Niue : matolu, thick. Uvea :
matolu, thickness. Maori: matotoru, id. Fotuna, Tahiti:

matoru, thick, full-fleshed. Mangareva: matoru, fat, thick,
heavy. Rapanui : matorutoru, thick, not compact. Hawaii :
makolu, wide, thick, deep. Marquesas: motou, thickness.
Epi: torn, large. Norbarbar: motoltol, thick. Mota: matoltol, id.
Nguna: matulu, id. Malekula: metetir, id. Baki : mererolu, id.

This root is discussed under 163.

um, ubu, oven.

Samoa, Maori, Nukuoro, Niue, Tahiti, Hawaii, Mangaia, Marquesas,
Mangareva, Paumotu: umu, oven. Tonga: ngotoumu, id.
Uvea: ngutuumu, id. Futuna: umu-kai, id. Fotuna: amu,
cooking place. Rapanui: umu, oven; humu hare, cook house.

Motu : amu, oven. Mabuiag : amai, id. Miriam : ame, id. New Bri-
tain :ubu, id. Mota: um, id. Ponape: um, id. Bierian : baumo,
id. Tanna: noanumun, oven stones. Aneityum: inmunum,
oven (inmun, an opening) ; nehpanum, a large fire for cooking.

The Polynesian radical is consistently umu. Tonga and Uvea compound
with it a word which in Uvea is distinctly ngutu mouth and in Tongan we
may feel that ngutu has been specifically differentiated in this composite.
In the Futuna composite the latter element is merely kai food.

The principle of terminal abrasion is sufficient to identify with this the
Efate and Mota um, and even the remote and extralimital Ponape um.
It is no difficult task to find the identification in Motu and Fotuna, for the
u-a mutation is general. The fact that Efate has ubu as well as um serves
to link in the New Britain ubu. We lack data on which to discuss an m-b
mutation; the nearest approximation lies in a single m-v instance in
masaki (323) ill Nggela vahagi. Codrington (Mota dictionary) cites the
New Britain word as umbu. This would correspond to the system in Viti
where a b requires the preface of an m. It might be that a people who
required thus to preface a b would by attraction add a b to an m in
loan material. Then when it passed along in secondary borrowing to
others who could manage an unprefaced b the proper m would be relin-
quished in favor of the intrusive b. This is purely speculative, yet we
may cite at least one instance in which a similar principle has been active
in the borrowing. We have this on Dr. Codrington's excellent authority
(Melanesian Languages 92) :

The formation of the Fagani figti (star) deserves notice. In that place the h of Wango,
three miles off, regularly turns to /, but g represents the same letter left out, perceptibly j
with a gap in the sound, in Wango. The Fagani (Ha'ani at Wango) word figu ought \
then, to represent the Wango hi'u, and in fact it represents he'u. But it is very instruc-
tive to observe that the gap in the Wango word really means t, not g, and has been filled
up with g in the Fagani word under a misapprehension. It is plain that the Fagani
and Wango words are independent, because one comes from vitu, one from vetu. The
interest lies in the filling up the gap with g in Fagani, because the gap in Wango generally
represents g, though sometimes it is in place of t.

200 THE POLYNESIAN WANDERINGS.

77-
usu, iu, u, a reed.

Samoa: u, a reed.

Malekula: ui, a reed. Epi: yi, id.

The community of u in Samoa and Efate together with the presence of
the longer form usu in the latter argues in favor of usu as the earlier radical,
which has been preserved in Efate. We are not without examples in these
comparisons of the possibility of a double terminal abrasion, first the
abrasion of a final vowel, then the abrasion of the consonant thereby
exposed, thus presenting a new vowel final. This seems rather to have
been the method than the bodily dropping of a final syllable. This note
will be found extended in 98.

Malekula and Epi are by no means satisfactorily identified. They seem
much more distinctly to be variants of some radical of which i is the con-
stant component.

EFATE-POLYNESIAN-MALAY.

78.
ne, here, there, this, that.

Samoa: nei, this, here. Mangareva, Aniwa, Sikayana, Futuna,
Uvea, Nuguria, Niue: nei, this. Maori: nei, denoting near
position. Tahiti: nei, here, now, this. Hawaii: nei, this
(of place, time, person) . Marquesas: nei, here, now. Paumotu:
nei, here. Nukuoro: net, now. Rarotonga: ainei, this.
Tonga: ni, this.
Macassar: inni, this.

79-
tumatuma i, to knock, as at a door as a sign to open it.
Samoa: tuma, to strike with the knuckles.
Malay: antam, to knock.

EFATE-POLYNESIAN-SEMITIC.

80.
aliali, taliali, to delay, be slow.

Hawaii : alia, to wait, to stop one when doing a thing, to restrain,
used imperatively, stop ! wait ! take care ! stand aside ! Tahiti :
aria, stop ! hold ! Mangareva : karia, an interjection used to
show off a thing in the sense "there! do you see!" Samoa:
tali, to wait for; fa'atali, id. Futuna, Uvea: tali, tatali, to
wait for, to expect. Maori : tatari, to wait, to tarry. Tahiti :
tatari, to wait, to delay. Rapanui: tatari, to wait for.
Nukuoro : hakatari, to wait. Hawaii : kali, to wait, to tarry,
to stay. Tonga: tali, to wait for. Rarotonga: tatari, to
wait. Marquesas: tetai, to wait for, to stay for.

Arabic: ala (alu), alia', to delay, be slow.

Particular interest attaches to the synonymy in Efate of two forms,
each of which is found in Polynesia though never again are they brought

DATA AND NOTES. 201

together in the same speech. The existence in Polynesia of a third form
in k (Mangareva karia, but not Hawaii kali) seems to point out this as an
instance in which shades of meaning are communicable by the system of
consonantal modulants which I have elsewhere argued at length (27 Ameri-
can Journal of Philology, 392).

81.
amos i, amo, aino-taki, amo-rua, tak'amo, amoamo, to carry, to bear, to
carry on the shoulder.
Samoa, Maori, Hawaii, Marquesas, Mangaia: amo, to carry on the
shoulder. Futuna : amo, to carry a burden. Uvea : amonga,
a burden, carrying pole. Tahiti : amo, to carry on the back.
Moriori, Nuguria: amo, to carry on a pole. Aniwa: amo, to
take. Rapanui: amo, to carry; amo mat, to bring; amo, a
yoke; amonga, a burden. Tonga: haamo, to carry on the
shoulder. Niue: hahamo, to carry a burden on a pole.
Hebrew : 'amas, to bear, to carry, especially to lift up a burden and
put it on a beast.

That the Proto-Samoan radical is hamos is established as to the initial
aspiration by Tonga and Niue, as to the final consonant by Efate.

82.
bakateba, to watch, to look out for.

Samoa: tepa, to look upward. Futuna: tepa, to turn the head

or eyes in order to look.
Hebrew: sapah, to look out for, to view, to watch.

83.
beingo, baingo, a shell trumpet, a kind of flute (coconut shell).

Samoa, Futuna, Uvea: fangufangii, a flute. Tonga: fangufangu,
a flute, to blow through the nose. Uvea: fangn, to blow the
nose. Niue : fangu e ihu, id. Hawaii: hanu, to emit breath
from the lungs ; hanuhanu, to smell, as a dog following the track
of his master.
Arabic: baka, to blow a trumpet; ba'ku, ba'ko, a trumpet.

The sense of the Polynesian will be made satisfactorily clear by the note
that the flute is played at the nostrils. If the coconut shell is really used
in Efate as a musical instrument it has escaped my observation and all
record, so far as I have seen, and at any rate it would properly be classed
rather with the ocarina than with the flute.

84.
bua, to divide.

Samoa: vaevae, to divide.

Arabic: fa'a, fa'w', fa'y', to split, cleave, to be open, separated.

If there be any validity at all in this identification it must be with the
element va as meaning to have or to be a space between. It will call for
bu-v mutation. The only light which our material sheds upon such a
mutation lies in the similar pn-v in vivini (242) to crow Malekula puinpuin
to whistle. On the radical sense of va see 27 American Journal of Phil-
ology, 387.

202 THE POLYNESIAN WANDERINGS.

85-
bubu, to gargle.

Samoa: pupu, to gargle, to rinse out the mouth. Tonga: bubu,

to gargle.
Arabic: ba'ba', ba'ba'u, gurgling sound of water flowing from a
bottle.

86.

bulifulia, mabulu, mafulu, swollen here and there, fat.

Samoa: fula, dropsy of the belly, stout; fulafula, swellings on the
body ; fufula, to swell ; fulafula, fulanga'i, to be swollen. Tonga :
fula, a tumor, a hard swelling ; fufula, to swell ; bubula, a swell-
ing, protuberance, to swell, to bloat. Futuna: fula, fufula,
fulafula, pula, to swell. Niue": fufula, to swell. Uvea: fufula,
mapula, id. Hawaii: hula, a swelling, a protuberance under
the arm or on the thigh. Fotuna : no-fura, to swell ; niko-fura,
swollen.

Hebrew: 'afal, to swell up, to be tumid. Arabic: 'afila, to have a
tumor or hernia.

The Proto-Samoan radical being fulang we should look for the final con-
sonant in Efate. Since it is not found, since even the vowel is different,
we are not to accept this identification as altogether satisfactory. The
labial uncertainty, however, is perhaps critical. The hesitation as to the
employment of the spirant or the mute is carried along into Nuclear Poly-
nesia ; Tonga uses surd spirant and sonant mute ; Futuna and Uvea use
surd spirant and surd (their only) mute.

The Semitic is still farther away from the mark.

87.

buri, biri, to pierce, to stick.

Samoa: velo, to cast a spear or dart, to spear. Tonga: velo, to
dart. Futuna: velo, velosi, to lance. Uvea: velo, to cast;
impulse, incitement. Niue: velo, to throw a spear or dart.
Maori : wero, to stab, to pierce, to spear. Tahiti : veto, to dart
or throw a spear. Mangaia : vero, to pierce, to lance. Manga -
reva: vero, to lance, to throw a spear. Marquesas: veo, to
lance, to throw a spear.
Ethiopic: barara, to stick, to stab.

The Proto-Samoan radical being velos brings this under the same com-
ment as in the last preceding item.

elo, el', lolo, sweet, pleasant, agreeable.

Hawaii: olu, to be pleasant, agreeable.
Arabic: hala', halw', to be pleasant, agreeable.

Without more data we may neither wholly affirm nor quite deny this
identification as between Melanesian and Polynesian. The sole point of
Semitic resemblance lies in the possession of /.

DATA AND NOTES. 203

89.
emai, emwai, in the distance, far away,

Samoa, Nukuoro: mao, distant, far. Samoa, Uvea, Fakaafo, Vate,
Maori, Hawaii, Mangaia, Manahiki, Rarotonga, Paumotu,
Nuguria: mamao, distant, far away. Tonga: mamao, id.

Futuna: mamao, id. Niue: mamao, id. Tahiti: taumamao,
to hang out of reach. Mangareva : akamamao, to send away.
Marquesas : memao, distant, far away.
Arabic: ma1 oka, to be far off, distant; ma'k', distance.
The Polynesia mamao can scarcely be brought into association with Efate
emai, as our author suggests without any consideration of the difficulties.
To me emai seems far more likely to be the widespread mai, from, with a
verb-making prefix. In neither case can the Semitic be said to have any-
thing to do with the matter.

90.
fam i, bam i, to eat.

Samoa: samusamu, to eat the remains of food; samuti, to eat (jocu-
lar). Tonga : hamu, to eat food of one kind only. Futuna :
samuko, id. Maori : hamuhamu, to eat scraps. Tahiti : amu,
to eat; aamu, a glutton; hamu, gluttonous; aihamu, to eat
gluttonously the leavings of others. Hawaii: hamu, to eat
fragments of food. Mangareva: amu, to eat with the mouth,
not using the hands; to eat scraps or leavings.
Hebrew: pa' am, to have the mouth full, to swallow down. Arabic:
fa'ama, id.
The Proto-Samoan radical is samut.

Against the identification lie two objections. In my note 77 I have
shown how the abrasion of a final syllable may be accounted for as the
abrasion of a vowel to a new terminal in a consonant followed b)r a second
abrasion of the consonant to a new terminal in a vowel. Such operation is
sufficiently rare in our material ; far rarer would be such a case as this, the
abrasion first of consonant, then of vowel, to end upon a closed syllable
once more. The mutation s-f, s-b, must be rare indeed, for these materials
show not a single instance of at all a satisfactory nature, not a single
instance in which an s migrates to any point in the labial column.

The Semitic identification is as remote from Efate as it is from Polynesian.

9i-
fasu, fasua na, a part, portion, member of the body.

Samoa : fast, to split, a bit, a piece. Fotuna : no-fafasia, to split.
Tonga: faahi, a side, a half of anything divided. Futuna:
faasi, a side, a portion. Niue: fahi, a side, a place. Uvea:
faahi, a side, a part, to divide. Nuguria: te vahi mahina, the
crescent moon.
Viti : vathi, to cut (chiefly of yams) ; pieces of yam for planting.
Hebrew: basa', cut in pieces. Arabic: bas'a'a, to cut, to cleave;
bas''at, a part, a piece.
The vowel change i-u is not so rare as to vitiate this identification (see
17 Journal of the Polynesian Society, 85).

204 THE POLYNESIAN WANDERINGS.

92.
fonga, afo, nafo, whetstone, grinding stone, and (because used as a whet-
stone) pumice.
Tonga, Futuna, Uvea, Niue: fuanga, a grindstone. Tonga: juo-
fuanga, pumice. Samoa : foanga, grindstone. Maori: hoanga,
grindstone, whetstone. Mangareva: hoanga, oanga, a fine
volcanic stone used for whetstones, a grindstone. Hawaii : hoana,
a hone, whetstone, grindstone. Tahiti: hoaa, a whetstone.
Arabic: nasfa-t, whetstone, pumice.

The stem is fo and fu. This surely can have nothing to do with the
Semitic identification proposed. There is reason for doubt as to the defini-
tion. It will be understood that whetstone must be a very modern sense
of the word in the Pacific, for until the Europeans brought iron the islanders
had nothing to whet, since the process of putting an edge on stone imple-
ments is a tedious process of grinding. How far pumice is susceptible of
such employment in whetting iron I am unable to say, but in the grinding
of stone it can have little value. If it were not for the fact that the pumice
sense is recorded from Tonga I should incline to regard the explanation given
in the Efate case as a labored effort to produce harmony with the Arabic.

93-
ngafa, a fathom.

Samoa, Tonga, Nukuoro: ngafa, a fathom. Niue: ofa, id.

(L,amotrek: ngaf, id.)
Arabic: kamat, a fathom.
The exact agreement of Efat<§ and Nuclear Polynesia suggests no connec-
tion with the proposed Arabic source other than that of identity of meaning.

94.

ngaingai, to pant, to be out of breath.

Samoa: nga (Tutuila dialect), nga'e, to breathe hard, to pant, to
be out of breath. Tonga: nga, to pant; ngaaki, to cough.
Futuna : ngaaki, to pant, to be out of breath. Mangareva : nga,
to be hoarse. Maori: nga, to breathe; tunganga, to be out
of breath. Hawaii: na, to gasp, to half-breathe; nae, to pant.
Marquesas: nae, obstructed respiration. Rapanui: ngaengae,
shortness of breath, out of breath.
Syriac : kah, to pant.
The Proto-Samoan radical is ngak. The Efate is explicable as formed
from the already abraded stem by the verb-forming suffix.

95-
ngoko i, ngokoi, ngokei, ngokai, to scrape, to mark, to paint, to smear (all
used in reference to najona or bast cloth) ; koko, reddish juice
or paint for bast cloth made from a plant of the same name.
Samoa : 'o'ai, to mark or paint bast cloth ; 'o'a, the tree (Bischofiia
javanica) from which is obtained the coloring matter for this
use. Tonga: koka, the name of the tree ut sup. ; to paint cloth.
Futuna: koka, id. Niue : koka, name of a tree whose wood is
used for rafters.

DATA AND NOTES. 205

Hebrew : hakah, hakak, to cut into, to hack, to engrave, to carve, to
draw, to paint, to delineate.

The identification involves none but frequent mutations as between
Efate1 and Nuclear Polynesia. The verb may be derived from the tree
name; the tree may have been named from the purpose to which it is put.
Of the two equal possibilities I incline to the former, for the use of the i
in forming the Samoan verb tends to make it transitive and specific, literally
to put-&o£a-on. If this be so the sense is one of painting or daubing.
Therefore in Efate, where our author specifies that it is used only of bast
cloth as in Samoa, the sense "to scrape" can only be descriptive of the
motion of painting and without signification of removing aught. He
has evidently relied upon that sense to clinch his identification of meaning
with the Hebrew word proposed as ancestral. The general quality of his
Semitic parallels is quaintly illustrated in this Hebrew word. He has given,
in which I have duly followed him, the emphasis of italics to the definition
hack, evidently struck with the resemblance hakah-hack, and not unwilling
to let the implication stand that our hack is Yiddish. Our Germanic fore-
bears had the word long before they acquired the raw material of the
Judenhetze.

96.
kaf, to be bent.

Maori: kohu, somewhat bent, concave, warped.
Hebrew : kafaf, to be bent.
Cf. kabwe 179.

97-
kari, takan, to hasten, to go swiftly.

Maori: kari, keri, to rush along violently, as wind.
Arabic: kara, to hasten.

98.
kon, kona, kokon, ngkon, to be bitter.

Samoa : 'ona, 'o'ona, bitter, sour, acid, poisonous ; 'onasia, to be drunk,
poisoned. Tonga: kona, konahia, bitter, drunk, poisoned.
Futuna: kona, bitter, poison. Niue: kona, konahia, bitter,
acid, nauseous, drunk, poisoned. Nuguria: kona, sour. Uvea:
kona, bitter, poisoned, drunk; konahia, drunkenness. Fotuna:
kona, drunk. Hawaii: ona, drunk. Tahiti: onaona, sharp,
disagreeable.
Mota: gogona, bitter. Santo: kogona, id. Eromanga: nakan, id.
Arabic: homa-t, bitter, heat, gall, poison. Ethiopic: hama-t, id.
Hebrew: hamah, id.
The Proto-Samoan radical is konas.

The Efate forms show us still more clearly than in 77 the graduation of
the process by which a final syllable is lost, not as syllable but by successive
abrasion of its members. In kona we have the transition form after the
abrasion of the final s. In kon, kokon, ngkon we find the ultimate result
upon the abrasion of the then final vowel. Thus it is made clear that the
syllable does not drop off as a unit.

The proposed Semitic identifications accord with this only in one or in
two vowels, the consonant structure being wholly unlike.

206 THE POLYNESIAN WANDERINGS.

99-
leba, a species of earth, clay, mud, dirt.

Maori :repo, dirt. Tahiti: repo, earth, dirt, mold, dust, filth. Manga-

reva: repo, dirt, ordure. Paumotu: repo, dirt, mire. Hawaii:

lepo, general name for dirt, dust, defilement of any sort, clay.

soil, earth, dung. Marquesas: epo, dust, dirt, mire, mud, earth,

Arabic: tabi'a, dirty; tabe', taba', dirt, mud; tub' an, clay.

ioo.
Iffaru, libuis, rafalu, rifalu, a part, some.

Niue: falu, some, a little. Fotuna: efaru, some, many. Tahiti:

fanu, some. Nukuoro: hanu, a little.
Arabic: ba's'u, a part, some.

IOI.

mak, to fall, become mild, gentle, die away, as the wind; mao, maomao,
to be gentle, mild.
Samoa : mao, lull in the wind or the waves, the lull in a reef opening.
Tonga : maomao, dry, applied to the intervals between showers.
Niue : mao, fine after rain, to cease raining. Futuna, Maori,
Tahiti: mao, to cease raining. Hawaii: mao, to pass away,
as fog or cloud. Marquesas: mao, dry, as land once wet.
vSyriac: mak, to be cast down, prostrated, humble, mild. Arabic:
mahiha, to be mild.
The difficulty here is that mak, which has to do with the wind, does not
identify itself with the Polynesian mao; and that mao, which accords in
form with the Polynesian, lacks accord in sense. Because we must rest
upon the Polynesian base of form and signification we cannot accept this
identification with Efate and therefore must decline the Semitic which
our author identifies with his Efate material.

I02.

malei, to divorce.

Samoa : alei, to divorce.
Aneityum : arei, to drive away.
Arabic : hala'a, to divorce.
If Melanesia or Polynesia afforded more identifications we could pass
more confidently on this. If malei be an older form of the word we then
have to do with a case of frontal abrasion, not wholly unknown yet rare,
as will be seen in the tables in the Melanesian chapter.
But the Semitic accords with neither.

103.
maloi, buloi, a mask, cover of the face.

Samoa: pulou, a head covering, bonnet, turban. Tonga: bulou, a
veil or cover to conceal the face ; b uloa, a mask. Futuna : pulou,
to veil, to cover the head. Uvea : fakapulou, to veil. Hawaii :
pulou, to cover the head, to blindfold. Tahiti: purou, to cover
the head. Mangareva : puroku, to cover the head, as with a hat.
Rapanui: pud, to put on a hat. Nuguria: buloo, a hat.

DATA AND NOTES. 207

Viti : vulou, pulou, to cover oneself up face and all.
Arabic: barka'a, to cover the face, to veil; burka'o, burk'u, a veil.

104.
maloiloi, to be feeble, tottering from weakness.

Hawaii: loeloe, maloeloe, feeble, faint, weary.

Arabic: la'la'a, to be infirm and weak from disease and languor.

105.
masirsir, to sob, as after crying.
Samoa : masusu, to sob.
Arabic: zahara, to pant or gasp with vehemence and groaning.

The equivalence herein involved, sir-su, is extraordinary; and Dr.
Macdonald's grammar of Efate is so completely devoted to the statement
of his Semitic theory that we encounter no suggestion as to the pronuncia-
tion of sir. It may be that he intends it to be pronounced as the syllable
would be if English. In that case the vowel change is reasonable and the
identification excellent.

The Semitic, however, seems not to fit.

106.
(a) mau, true ; loamati, truth, true.

Tahiti: mau, true. Mangareva: mau, id. Rapanui: mau, certain.
(6) mau, firm, intrepid, brave.

Samoa: mau, to be firm, decided, unwavering. Tonga: mau, fast,
firm, constant, unwavering. Maori: mau, fixed, steadfast.

Nuguria: tamau, to bind.

(c) mau, to come upon, to obtain, to find.

Samoa: maua, to get, obtain, have, take. Maori: mau, to take
up, lay hold of, seize. Tahiti : mau, to hold, to seize. Tonga :
mau, to obtain, possess. Mangareva: mau, to hold, to seize.
Tongarewa: mau, to possess. Uvea: mau, to seize, to grasp,
to hold, to contain. Rapanui: mau, to hold, to accept, to
acquire; maua, to find; maoa, to hold.

(d) mau, very, indeed, continually.

Maori: mau, continuing, lasting. Samoa: mau-, very. Hawaii:
mau, constantly, continually. Tonga : mau, always, perpetually.

Bierian: lehmau, truth, true.

Hebrew: 'aman, to prop; 'amen, firm, unshaken, faithful. Arabic:
'amana, to confide in; 'amuna, to be faithful; 'amina, to trust,
to be secure. Syriac: 'eman, to persevere, to be constant;
'amen, verily, truly, certainly.

The Efate- Polynesian identifications are satisfactory in all four groups.
The Semitic is not satisfactory in form at all.

107.
mwota, motamwota, refuse, rubbish, as fallen leaves.

Samoa: otaota, rubbish, filth, ordure. Tonga: otoota, sweepings,
rubbish. Futuna: btabta, impurities in a badly strained fluid.
Niue: otaota, rubbish, refuse. Uvea: otaota, excrement, dung,
filth, dirt. Maori: ota, sawdust; otaota, weeds, litter. Tahiti:

208 THE POLYNESIAN WANDERINGS.

ota, chaff, refuse. Paumotu: ota, residue. Rapanui:
hakaotaota, to crumble. Hawaii: oka, refuse; okaoka, dust.
Arabic: "ota', rubbish, refuse.

Again we encounter what seems to be a frontal abrasion, and it involves
the same consonant as in 102. More evidence on the Melanesian side is
needed before the identification will be quite satisfactory. But if the m is
indeed radical it removes the resemblance on which the Semitic identifica-
tion is based.

108.

nai, side board of a canoe, defender of a place, fence.

Samoa: a, ai, a fence, railing. Tonga: a, fence, hedge; ai, to
surround, inclose, defend. Futuna: a, wall, hedge, fence,

anything which makes an inclosure. Uvea: a, wall, fence.

Tahiti: nanai, to put in a hedge.

Arabic : nawa', naa', to guard, to protect.

This involves a frontal abrasion, namely of n; but the recurrence of the
n in Tahiti nanai is very good evidence in favor of the identification.

109.

nobwa, ob, naob, lime (ashes of coral); nobwanobwa, to be dusty, to

become dust, to fly in the air (dust).
Samoa: navu, lime.
Pokau: avu, lime. Mekeo: apu, id. Motu : ahu, id. Hulaiabu, id.

Galoma: gabu, id. Sinaugoro: gau, id. Rubi: gou, id.

Sariba: gauarana, id. Nada : pwau, id. Kiriwina: pwak, id.

Wedau: gabubua, id. Panaieti, Misima: aru, id.
Arabic: cf. no.

The Samoan may be homogenetic with nobwa, but it is impossible to
associate it with ob and naob, unless the decrepit forms from Torres
Straits are susceptible of interpretation as transition stages. That nobwa
is not itself a modification of ob with the article is clear from the fact
that nobwa takes the article and then becomes nanobwa.

no.

noba-ni, in ma ni, tumu na, tomo na, manubunubu, to wrap in leaves with

hot stones and cook ; nobwanobwa, to be cooked, soft.
Hawaii: nopu, thoroughly cooked, soft, plump, fat, swelled out;

nopunopu, to spring or swell up (in the mind), to swell, to be

large, round, to spring up.
Arabic : tabaha, tabh', to cook, roast, ripen, to grow up, to be cooked ;

tubbah', tabih', fatness; tabW , cooked.

In the fact that there may be a satisfactory sense identification of the
Hawaiian with nobwanobwa I see no warrant to extend the identification
to the inclusion of nobani and manubunubu where the difference in significa-
tion seems prohibitive. Granted that this complete identification be valid
it can serve in no way to admit tumani and the other forms in t; yet it is
upon the t in these forms that our author hangs his Semitic identifications.
They bear no relation to the Polynesian.

DATA AND NOTES. 209

in.

nu e a, nunu ea, to wipe, to rub ; nunu, a wiper, a rubber.

Samoa: nunu, to grate down, as turmeric; nuanga, the grating of
arrowroot and turmeric. Futuna : nu, nunu, to grate arrow-
root. Niue: nu-pia, id.

Arabic: t'amma, to sweep, to rub, to wipe off.

112.

ofiofi, ofi, afi, to be near to, alongside of.

Tonga: ofi, near at hand, nearness. Futuna: ofi, ofiofi, near,

close to. Uvea: ovi, id.
Arabic: wahaja, wahf, to draw near to, to approach.

ii3-
rakei, raki, to adorn, to dress.

Samoa: la'ei, dracaena leaves tied to a stone to attract cuttlefish,
to fish with that device, to wear a train, to dress for a review
of troops; la'ei'au, to exercise troops at a review, to have all
ready for war. Tonga : lakei, leaves made fast to the stone
used in catching the catfish as an ornament, to fish therewith.
Futuna: lakei, used only of a garment, to have a long train.
Nuguria : lagei, to paint the body with turmeric. Mangareva :
rakei, to ornament, to adorn, a garland, chaplet, decoration.
Rapanui : rakei, an ornament, to adorn, to embellish.

Ethiopic : lahaya, to adorn, to dress.

114.

raku sa, raraku sa, taraku sa, taku-ti, to bind up, to remove one's things

as in a flitting, to remove, to carry away.
Samoa: la'u, to clear off, to carry away; la'u mai, to bring. Uvea:

laku, to send, to throw into. Hawaii: laulau, a bundle, bag;

a wrapper of a bundle, the netting in which food is carried;

lalau, to seize, to catch hold of.
Arabic: raka, to dig, to bind up.

"5-

ras i, tas i, to shave the head, to strip off fruit from a tree.
Samoa : lase, to scrape off warts.

Mota: ras, rasa, to scrape, to scratch, to rub. Malo: rosi, to scrape.
Ethiopic: las'aya, to shave.

116.

ruru, a cluster; rei, a band of men, a clump of trees.

Maori : ruru, to tie together. Tonga : lulu, the reeding of a house.
Tahiti: ruru, to congregate. Mangareva: rurue, to bring

together a crowd ; ruruku, to head up leaves. Paumotu: ruru,
a coop, a cage ; ruruhanga, an assembly, to collect an assembly.

Arabic: ra'a, to grow, luxuriate, be congregated; ri'at, a band, a
crowd.

210 THE POLYNESIAN WANDERINGS.

117.

siuer (siwer), suara, suuara, surata, to walk, proceed, go away; sisiuer,
susuara, to walk about.
Samoa : savali, savavali, to walk.
Arabic: safara, sifar\ to make a journey, to go away.

The identification is not wholly satisfactory nor yet to be set aside, for
we have no such chain of data as would enable us to make a sure determi-
nation. The v-u(w) mutation is sufficiently frequent to call for no remark;
thus siuer and suuara are accounted for and suara is but a slight and single
step removed. But surata is not by any means in the same line and is
impossible of association. In the reduplicated forms lies the strongest
argument against identity ; it will be observed that while Samoa duplicates
the latter member, savavali, Efate duplicates the former, sisiuer. In the
discussion of a single item in the great topic of duplication (Duplication by
Dissimilation, 30 American Journal of Philology, 173) I have presented the
argument in proof of the determination of the relative importance of the
two parts of a composite word as revealed in the duplication. It is here
seen that in Samoa the latter element, bearing the duplication, is of the
more importance. We can scarcely believe it possible that this assignment
of importance is but a modern development in Polynesia; it surely must
have been even better recognized in the earlier phase when they were migrant
through Melanesia. If siuer be a Polynesian loan word, and I have already
remarked (1. c. 180) upon the fact that wherever the words which undergo
duplication in Efatd are susceptible of identification they are uniformly of
Polynesian stock, it seems strange that in adopting the Polynesian mech-
anism of duplication these Melanesians should have misapplied it. Of the
twenty-five identifiable duplications in that paper thirteen correspond
exactly with the Samoan duplicated forms; eleven are not comparable
because of the absence of duplication in these words in Samoan ; one only
is at variance with the Samoan duplication, and in this case the Samoan
has two words of the same sound but of different sense, only one of which
is duplicated, and the duplicated Efate form corresponds with the form
of the duplicated Samoan, but with the sense of the unduplicated, and it
would seem that the Samoan had but specialized to avoid the chance of
error. We may, then, accept the general principle that Efate duplicates
the same elements as Samoan. This determination, and I must regard
it as based on good grounds, militates against this identification.

The Semitic identification proposed is altogether too good in form when
we regard the length of their separation from the putative common stock.
The Samoan has but the single sense of walking, the act of such locomotion
with no slightest suggestion of either terminus. The sense of the Arabic
bears no relation to the Samoan. It is only in Efate that the two signifi-
cations are brought together so that a transition might be possible. If
the two words are formally identifiable Efate has the walking sense.
Unfortunately our author's propensity to make out his case at any cost
is now so well comprehended that little dependence can be placed in his
ingeniously devised transition through "proceed" to "go away."

DATA AND NOTES. 211

118.

suli na, shoot (as of a banana), offspring (of man); sulia, to have shoots.

Samoa: suli, shoot of a banana, son of a chief. Tonga: huli, a
plant, a sapling, a shoot (as huli fuji, the shoot of the banana).
Niue : huli, young seed plants of the taro. Hawaii : huli, the
name of taro tops for planting. Maori : huri, seed. Rapanui:
hurt, a stalk. Mangareva: huri, banana shoot.

Viti: suli, a banana sucker.

Mota Maligo: suliu, a sucker from roots, a shoot from tubers.
Mota Veverau: sului, id., and children, offspring.

Hebrew: neser, a sprout, shoot, offspring.

119.

tafa, a hill, lit. that which goes up or is high; high, above.

Samoa : tafa, the side of a hill. Rapanui : taha, to lean, to incline ;

tahataha, boundary, frontier; hakataha, to turn aside, to decline,

to be on the side.
Tigre: dayaba, to go up. Ethiopic: diba, above.

A squinting etymology.

120.
tera, to be quick, swift.

Samoa : telea'i, to run quickly ; teletele, to be quick, step out ; televave,
tclelise, to be very quick. Futuna : telekaki, to run swiftly,
to go with speed ; televave, one who goes swiftly. Mangareva :
tere, to go well, to sail well. Maori: tere, to move swiftly, to
be quick. Sikayana: tere, to run. Fotuna: no-tere, id.

Rapanui: tere, id.; hakatere, to urge to haste.
Arabic : darra, to run vehemently or swiftly, to turn a spindle very
swiftly.

EFATE-MELANESIAN-VITI-POIATNESIAN.

121.

barab, baraf, barau, baram, long, high (as a hill).
Tonga: baleva, tall, long, overgrown.
Viti: mbalavu, long, generally of space.
Malo: barauo, long, wide. Makelula Uripiv: periv, id.

Our author's association of this with laba (307) expresses his curiosity
rather than his acumen. The identification as here set down seems excel-
lent; it certainly bears very prettily upon the Viti track and its extension
to Nuclear Polynesia in Tonga rather than in Samoa.

122.

bi, fi, reflexive preformative.

Samoa, Tonga, Futuna, Niue, Uvea : fe, reciprocal prefix. Cf . Maori

whe in wheanganga.
Viti : vet, reflexive prefix. Rotuma : hoi.

212 THE POLYNESIAN WANDERINGS.

Arag, Nggela, Vaturanga, Bugotu: vet, reflexive prefix. Wango,
Saa: hei. Ulawa: hai. Fagani: fai. Koro :bai. Kabadi:
vai. Sinaugoro, Keapara, Kabadi, Hula : ve. Galoma : be.
Pokau: vi. Duke of York : we. Motu: he. Nengone: e.
Mota, Merlav, Retan: var. Pala: har. New Britain, Bara-
von: war. Kabakada: wara. Lakon: va'. Motlav, Pak,
Leon, Vuras, Mosin, Alo Teqel, Gog, Norbarbar, Lo : ver. Volow :
vear. Omba: vui. Maewo: vagal.
We have several elements in this problem which detailed examination
will tend to simplify.

(i) The Nuclear Polynesian type, fe. In Melanesia this is found in
Efate' hi, fi; in Duke of York we; in several variants in Torres Straits;
and as an attrition fragment probably in Nengone e.

(2) The Viti type, vei. This I regard as the same as fe but representing
a later development, for the t-suffix is well marked as verb-formative when
the language is beginning to feel the need of specification in the use of the
much including attributive (27 American Journal of Philology, 378).

Its fullest form is in Viti, Arag, Nggela, Vaturanga, Bugotu, vei; thence
to Wango, Saa, hei.

A second series is Fagani fai, Ulawa hai.

With the single exception of Arag in the New Hebrides to serve as a
link these forms are all from the Solomons, with New Guinea offshoots of
Kabadi vai, and Roro bai.

Rotuma hoi is the second element in yet a third series, the spirant form
with this vowel nowhere appearing.

(3) We now have a widespread suite with a second consonant. For
this reason we must dissociate it from the foregoing. It does not come
over into Polynesia with the slightest trace of the second consonant. Yet
it is possible that it represents an original from which the foregoing derive
by abrasion.

Its fullest form is in New Guinea in Kabakada wara.

With abrasion of the final vowel we have Mota, Merlav, Retan var; New
Britain, Baravon war; Pala har. In Volow it is a characteristic peculiarity
to introduce e before a in a closed syllable; its vear serves as a transition
form to Motlav, Pak, Leon, Vuras, Mosin, Alo Teqel, Gog, Norbarbar, Lo,
ver. The Lakon va' is a degeneration of var, dependent on the language
peculiarity which Dr. Codrington records "at the end of words r is not trilled
and sometimes with abrupt pronunciation is not heard;" it may be taken
as the transition stage from the forms with two consonants to the Poly-
nesian type. This suite is found in the Banks Group, New Britain and
New Guinea.

Omba vui is not explained in our vocabularies. If it be permissible to
regard the u as functioning as semivowel w, then vwi might be regarded as
associable with Efate bi, fi, which also shows an uncertainty in striking the
exact sound of the labial. From its close neighborhood to Arag, which has
vei, Omba might be expected to show some form of this type.

The Maewo vagal seems an utter anomaly. Dr. Codrington says "it is
not clear what vagal may be." It will be seen that va-ga-l suggests the
common Banks Group type var with an infix; but we have no authority

DATA AND NOTES. 213

to assume that infixation is a formative principle so remote from Indonesia.
I note one more statement by Codrington: "the syllables are mostly open;
indeed, though it is common to close a syllable, it is hardly looked upon as
correct." If this has any bearing on the problem it may serve to indicate
a probability that vagal is loan material.

123.

lalo na, lalu na, the belly, the front, the under side (as of cloth).

Samoa, Tonga, Futuna, Niue, Uvea, Hawaii: lalo, below, beneath,
under, down. Nuguria: hakalalo, south. Maori, Tahiti,
Rarotonga, Tongarewa, Bukabuka, Mangareva, Paumotu, Sika-
yana, Aniwa, Fotuna, Nukuoro: raro, id. Rapanui: raro,
under, leeward. Aniwa: iraro,iroro, under, below. Nukuoro:
kiraro, below ; kailalopoli, mean, stingy. Marquesas : ao, below,
beneath, under, down. Moiki: ngango, id.

Viti: ira, kira, maira (ra), below.

Mota: lalangai, the under side; alalange, ilalange, talalange, under.
Motlav, Volow, Pak, Mosin, Alo Teqel: Mange, id. Lakon:
lalnga, id. Vuras: alalnge, id. Retan, Sasar: lalange, id.

Leon: lalanga, id. Merlav, Gog: lang, id. Vaturanga: ilao, id.

While I have followed our author's suggestion of identification and have
carried it out to the limit of my material I remain quite unconvinced that
there is anything valid in the Efate identification. To begin with, the Poly-
nesian material all means below, the Fijian below, the Melanesian under,
except that in Mota appears also the signification of the under side which
Dr. Macdonald ascribes to his Efate. Next, so far as concerns lalo the
belly we shall have no hesitation in bringing it into association with Samoa
alo belly in the courtesy speech, for not only have we elaborated the prob-
abilities of frontal abrasion, but in Efate we have also alo (48) belly as a
variant of lalo.

Assuming that Efate has a lalo which has so much of the "below" sense
as is suggested by "the under side (as of cloth)" we now proceed to the
examination of its identification.

The identification in that sense is satisfactory so far as relates to the
Polynesian. In Melanesia it will be found in Vaturanga lao, for an inner
/ is frequently dropped in that speech in words where Nggela retains it, a
statement by Dr. Codrington which may lead to the assumption of Nggela
lalo, although my material is empty at that point.

The Viti forms are all compounds on the base ra. This at best is but
a half of lalo, and we do not claim the identification as more than a sug-
gestion. The Viti form is not identified in Melanesia either.

The remaining Melanesian forms are all from the Banks Group and are
clearly variants of one stock. The fullest forms are Mota lalangai, lalange;
Leon lalanga; Retan, Sasar lalange. The second group is characterized by
obliteration of the second vowel — Lakon lalnga; Motlav, Pak, Volow, Mosin,
Alo Teqel lalnge; Vuras alalnge of the latter type. If we regard Merlav,
Gog lang as a decrepit form of this stock we leave this collection a com-
pact group identifiable with nothing in sight. It resembles lalo only in the
possession of two l's and the a, it nowhere suggests the 0 on which Poly-

214 THE POLYNESIAN WANDERINGS.

nesia is united, and it leaves final nga or ng unrepresented in any region
beyond, unless the ephelkustic na of Efate lalo na, which our author does
not explain, is really radical, in which case lalona represents the transition
type from lalanga to lalo.

Another explanation is equally considerable. Assume that Merlav lang
is not a broken-down form but the simple stem lal. The l-ng mutation is
represented in our material solely by this possible instance. The mutation,
however, occurs. It is found in two languages and both in this region,
although they are Polynesian inclusions or verge islands be it said, namely
in Aniwa (though not in this word) and in Moiki. The Moiki word is
ngango, as cited above. To be sure Merlav employs the ng change for but
one of its Vs. If this be considered reasonable lalanga and its devolved
forms are then reduplications. This explanation is deficient in regard of
the characteristic Polynesian o.

Neither suggestion is one to be held satisfactory. The identification
is sadly imperfect.

124.

mabwe, the chestnut tree and its fruit.

Tahiti: mape, the chestnut (Inocarpus edulis.) Maori: cf. mapau,

mapou, mapauriki, tree names.
Viti : mamba, the name of a tree, fruit edible.
Aneityum: mop(o), the chestnut. Malo: mabue, id.

125.
tobu na, grandfather, ancestor.

Samoa: tupunga, ancestor. Tonga: tubunga, tubuanga, ancestor.
Futuna: tupuna, grandparents. Tahiti: tupuna, ancestor,
grandfather. Mangareva : tupuna, grandparents, great uncle,
great aunt. Niue, Mangaia, Marquesas, Paumotu: tupuna,
ancestor. Rapanui: tupuna, tapuna, id. Hawaii: kupuna,
ancestor, grandparent. Uvea: tupuanga, parent. Fotuna:
bua, maternal grandparent ; rufeitupuna, grandfather and grand-
son. Nuguria: tipuna, mother-in-law.

Viti: tumbuna, ancestor, grandmother.

Motu: tubuna, grandparents, ancestor. New Britain: tubuna,

ancestor. Tanna: tupu(n), grandparent. Mota: tupui,
one of the second generation in the ascending or descending
line, ancestor or descendant. Malo: tubu, maternal grand-
parents, paternal grandfather. Baki: kumbuo, id. Aneityum:
etpon, grandparents, ancestor. Malekula: apu, grandparents.
Pala : tubu, grandfather. Tubetube : tubu, grandparents.

The identification is good throughout. In the Polynesian languages
which employ both ng and n the form in n is used except in Samoa, Uvea
and Tonga tupunga, etc., which are in form derivative verbal nouns from
tupu to grow but otherwise are anomalous. The same suggestion appears
in Mota, Malo, Baki, Pala and perhaps Malekula. Efate falls in with the
M-forms, provided that the na be radical.

DATA AND NOTES. 215

126.
tokalau, an easterly wind.

Samoa : to'elau, the northeast wind. Futuna, Niue, Tonga : tokelau,
the north. Nuguria: tokcrau, southeast, the trade wind.

Maori: tokcrau, eastern. Mangaia: tokerau, the northwest

wind. Bukabuka: apatokerau, north. Paumotu: tokerau,
the north; patokerau, the northeast. Rapanui: tokerau, air,
wind, fresh breeze, a squall, the noise of the wind, a season,
south ; tokerau alio, west. Tahiti : tocrau, the west or north-
west. Tonga: apatoerau, the south. Mangareva: tokorau,
the north. Moriori: tokorau, a wind name of uncertain appli-
cation. Marquesas : tokoau, the north or northeast. Hawaii:
koolau, the east. Fakaafo: Tui-Tokclau, a divinity.

Viti : tokalau, the east wind ; tokalaulutu, north or northeast ; toka-
lauvualiku, north or northwest.

Aneityum: na-tokarau, the northwest wind. Moanus: tolau, the
north wind. Santo: tokalau, northeast wind.

Our author proposes the derivation tok (toko) to remain, alau on the sea.
These respective elements have the assigned meanings in his Efate diction-
ary. From the Kfate point of view it is a definition even if it does not
particularly define. So far as I know it is the only definition, for the word
is quite incomprehensible in its Polynesian elements. On his element tok
consult 357, where it will be seen that in Polynesian it may very doubtfully
be recognized in Samoan and in a sense that would in no way fit this wind
or compass rhumb, and that in Viti it is primarily a posture in sitting.
His element a-lau, lau the sea, he identifies in Malo a lau, Epi lau and
Malay laut the sea. To which I may add that in Viti lau is the designation
of the windward islands opposed to ra (ra down) the leeward islands, on
either side of the central part of the archipelago which bears in that sense
the name Lomaiviti. The word lau in the sea sense does not occur in
Polynesia, nor do I recall it elsewhere in Melanesia.

If the derivation of the word lies in mystery, so is its use lacking what
we should call precision.

But first the mutations, the second vowel being critical.

Forms in a: Viti, Efate, Aneityum, Santo — all Melanesian.

Forms in e: Samoa, Futuna, Niue, Tonga, Maori, Mangaia, Bukabuka,

Paumotu, Tahiti, Fakaafo, Nuguria, Rapanui.
Forms in o: Mangareva, Marquesas, Moriori, Hawaii.
(Moanus irregular.)

Now we shall examine the sense, whether of wind or of direction in
general.

North quadrant (from northeast to northwest by the north) : Viti,
Samoa, Futuna, Niue, Tonga, Mangaia, Bukabuka, Paumotu,
Tahiti, Mangareva, Marquesas.
East quadrant : Maori, Efate, Hawaii.

South quadrant : Tonga, for comparison with Bukabuka and Paumotu
shows that the prefix, of wholly uncertain signification, does not
avail to establish the direction within 1800; and Rapanui.

216 THE POLYNESIAN WANDERINGS.

West quadrant : Tahiti, and Rapanui with a determinant word.
Moriori uncertain.

The onlv deduction we can draw from this is that the two extremes of
Polynesian settlement are in accord in fixing the sense in the east. Yet
the accord is only a seeming one, as we shall next see.

In all the South Sea islands we find four cardinal points : uta, shoreward ;
tai, seaward ; sake, up ; lalo, down. The latter pair are used in two dimen-
sions: in a vertical plane their directions are absolute, up and down; in
a horizontal plane their directions fluctuate, for sake is always up the wind
and therefore windward no matter how the wind may chop and change,
and lalo is leeward. The uta-tai direction may vary through opposite
semicircles beginning at any point and with a range of 1800. We may not
quite say that it adjusts itself always to the position in reference to the
nearest visible sea of the speaker in each act of speaking, although it very
frequently does so adjust itself. But for every little village community
it does establish itself with reference to its own cove and there is no com-
pass agreement on even the smallest island. The locus of the maximum
discordance would be a town built on the center of a circular island, in
which case every direction would be tai and there could be no uta. Thus
it will be seen that tai and uta may under certain conditions coincide with
sake and lalo, yet on the opposite shore of the same island they would be
in diametric disagreement. This digression is introduced to make it clear
that we have no positive direction sense by which to rate a norm for the
tokelau sense of direction.

I incline to the opinion that the solution will be found to lie somewhere
in relation to the one fixed index of direction in the tropical Pacific, the
trade wind. This is seen in the Samoan name of the fair weather season,
the Vaito'elau, the "time of tokelaus," namely, the months in which the
trade blows regularly every day long from east-southeast. Let us note in
how many instances we can see an easterly sense: Efate, Samoa, Maori,
Paumotu with a qualification, Marquesas, Hawaii. In the Marquesas, Samoa
and the Paumotu this is clearly not the trade wind but at right angles
thereto, yet such breezes are rare in the trade wind season. This makes
it appear that the direction is not the wind's eye, yet that it is in some way
associated therewith as a departure. The Maori must owe its easterly
sense to the signification brought with the word from its earlier home, for
New Zealand lies outside the trades and in the region of the westerly vari-
ables. I can not pretend to solve the problem. But I am sure that the
solution will be found to lie in the identification of the tokelau direction in
some angular displacement or departure from the prevailing trade wind in
each group — or in some point from which its people migrated with the word
already oriented — and this angle most probably has to do with the sailing
quality of their canoes, in other words, the number of points by which they
can lie the wind.

DATA AND NOTES. 217

EFATE-VITI-P0LYNESIAN-MAI,AY.

127.

alialia, insane; lala, an idiot, a fool, one demented.

Samoa: lielievale, to boast without reason. Niue: lialia pou, giddy.

Uvea: fakalialia, deformed.
Viti : lialia, foolish, absurd, crazy, out of one's mind, an idiot.
Malagasy: adala, foolish, infatuated, a lunatic, a fool.

The identification is quite satisfactory so far as it has to do with lialia
forms. I can not see that Efate lala belongs here, and certainly the mere
/ is not sufficient to establish a Malagasy relationship.

128.
bwelu ki, bwelu, beluuelii, to fold, to double.

Samoa: mapelu, to bend, to stoop, to bow down, persons stooping
with age, housebeams sagging under weight. Tonga: pelu,
bebelu, to fold, to crease. Futuna : pelu, peluki, to fold. Uvea :
pelu, id., mapelu, to bend, to bow. Hawaii: pelu, to double
over, to bend, to fold. Rapanui: peu, axe, adze.
Viti : mbeluka, kambclu, to bend, to curve.
Malay: valuna, folded, doubled.
The Proto-Samoan radical is peluk.

The Samoan mapelu is pelu with the condition prefix; pelu is found in
use in the sense of bent, crooked. Pratt indicates pelu, a sword, as intro-
duced ; I incline to the opinion that it is not the word but the specific appli-
cation which has been introduced, for my Samoan instructors told me that
the first swords ever seen were curved (pelu) on the edge and hence the
name. But as the first swords seen were undoubtedly cutlasses and not
scimitars an armorer will have to pass on the question of fact. Pelu'i,
also introduced, is wholly introduced, word as well as meaning, for it is
clearly a transliteration of the English word billhook. Samoa and Viti
lack the precise signification of pelu as it elsewhere occurs, yet not on that
account is the identification at all doubtful.

The Malay identification shows that the word had already been abraded
before the Indonesians took it from the Polynesian remnant, for only in
this state could the Malay treat it in accordance with his own methods of
word formation and add the -na.

129.
bungafunga, fungafunga, bungo ni, to be awake, to awaken.

Samoa, Tonga, Uvea, Futuna: fafangu, to rouse from sleep, to

awaken. Fotuna: no-fagona, to awaken, to be awake.
Viti: vangona, to rouse, to awaken.
Mota: vangvangov, vavangov, to waken.
Malagasy: fuha, awake; mifuha, to awaken.
In view of the fact that nowhere in the Pacific areas do we find a ng-h
mutation, and in the utter absence of possibly intermediate Indonesian
forms, we are not warranted in accepting the proposed Malagasy identi-
fication.

218 THE POLYNESIAN WANDERINGS.

130.

fai, the skate.

Samoa, Tonga, Futuna, Tahiti: fai, the sting ray (genera: Disco-
batis, Dasyatis, Taniura, Himantura, Hypolophus, sp. pp. ; the
type in Samoa is Himantura fai). Marquesas: hai, fai, the
stingray. Maori: what, id. Nukuoro: haimanu, uruhai-
pokorua, id.

Viti : vai, the sting ray.

Malay: pari, the sting ray. Tagalog: pagi, id.

We can not accept the Indonesian identification on account of the intru-
sive consonants.

The Nukuoro haimanu is especially interesting in connection with the
fact that in Polynesia passim manu means bird and beast but not fish, for
haimanu must here of marine necessity mean hai-ftsh. Dr. Codrington has
pointed out (Melanesian Languages, 56, 69) that Lakon mah means both
bird and fish and that in Maewo and Vanua Lava as masi, mes, the same
word is in use for fish but not for bird. Of course it is not to be understood
that he implies that masi is a manu variant. I note, however, Aneityum
numu, Tanna namu, Eromanga nomu, the same word and all meaning fish,
which it would be no straining of metathesis to associate with manu. Nuku-
oro has the manu-fish again in manumangamanga, starfish.

131-
fakau, f ikau, a message, messenger, ambassador, agent sent to do something
for a chief or community ; kau, to carry.
Samoa : 'au, to send ; fe'au, a message, to send for. Tonga : fekau,
a command, an order. Niue: fekau, a message. Uvea:
fekau, a servant, a messenger, to send. Fotuna : kau, to send.
Viti : kauta, to carry.

Java: panggawa, a noble, title of one of the five chief councillors of
state ; gawa, to bear, to carry, to convey, to bring.

While Java gawa does not quite accord with the Polynesian it agrees
with Viti and Efate. But panggawa seems not to be fekau, for the term is
menial in the Pacific and honorific in Java; furthermore in 122 we find no
evidence that fe- appears at all in Indonesia.

132.

seru, a comb.

Samoa, Futuna, Nukuoro: selu, a comb. Tonga, Uvea: helu, id.
Nukuoro: seru, id. Niue: hetu (anomalous), id. Fotuna:
seru, id.; ko-seserua, to comb.

Viti: seru, a comb.

Malay: sisir, garu, a comb.

NoTE — Marquesas: heu, to scratch the ground with the hands.
Mangareva : heru, to reject with hands and feet ; pahere, pahore,
a comb. Paumotu: heru, to brush with the hands. Rapanui:
heruheru, a rake. Tahiti: heru, to scratch as a hen; pahere,
pahoro, a comb. Hawaii : helu, to scratch the earth as hens.
Aneityum : ero, to scratch as a fowl.

DATA AND NOTES. 219

Tregear (Maori Comp. Diet. s. v. iveku) points out a probable inosculation
of roots in a series veku-veu-heu-heru-huru. It is to note that the oceanic
comb is not used to smooth the hair but, without disarranging the some-
what intricate coiffure, to scratch the scalp and vex its population. There
is nothing in Polynesian which sheds any light upon the /-/ mutation in
Niue, if indeed it be permissible to associate hetu with the selu stock. In
the dissection of the Melanesian material a very few instances are recorded
of l-t mutation, but they are so poorly supported as to afford little base
for the opinion that such a change has any real standing.

EFATE-VITI-POLYNESIAN-SEMITIC.

133-
bwase, bwasu, to break off with a snap or jerk.

Samoa, Futuna, Niue, Tahiti, Paumotu : fati, to break, to break off
as twigs or pieces of wood. Fotuna, Tonga : faji, id. Nukuoro,
Rapanui: hati, id. Maori: whawhati, id. Hawaii: haki, id.
Mangareva: ati, id. Mangaia: aati, id. Marquesas : fati,
hati, id.
Viti: mbasuka, mbasuraka, to break.
Arabic : fas' s' a, to break off ; fassa, to detach, shiver off.

There can be no doubt of identity in Efate and Viti, an identity which
holds in respect of each of the four root sounds.

On the other hand we have substantial unity in Polynesia and the only
sound in which there is identity with Efate-Viti is the former vowel. We
have no supporting instance to establish a mutation of the series Polynesian
/ Viti mb Efate bw or b, and little more for the mutation t-s. The identi-
fication is therefore, not cordially accepted. If the Polynesian fasi is
regarded as homogenetic with fati I do not look upon it as a t-s mutation
but rather regard fa as the stem modulated by each consonant in turn with
substantial sense agreement.

134-
fira-ni, bifira, to supplicate, to pray.

Futuna: pule, to pray. Tahiti, Hawaii, Rarotonga, Mangareva,
Paumotu: pure, to pray, worship, prayer. Maori: pure, to

utter incantations to purify.
Viti : vibure, a temple, house for the gods.
Hebrew : fatal, to supplicate, to pray.

135-
ngT, ngkl, ngiki, to creak, to squeak, to moan.

Samoa : 'i, to cry, as a fly or a bird. Rapanui : hi, to say. Tonga,
Futuna: hi, kiki, to squeak, a sharp cry or squeal. Niue:
kikii, to squeak. Uvea: hi, a cry. Fotuna: noh-ki, to

whistle (bird).
Viti: ngi, to squeak, to make a shrill noise.
Arabic: nakka, nakik', to creak.

220 THE POLYNESIAN WANDERINGS.

136.

(a) ngor i, kor i, to enclose or surround with a fence (nakoro).

Samoa: 'olo, a fort, stronghold. Tonga: kolo, a fort, a town, cloth
hung around a house in which a corpse lies. Futuna, Niue,
Uvea: kolo, a fort, tower, citadel, castle.

Viti : koro, a town, village, settlement.

Arabic: higr', hogf, a fence. Ethiopic: hagar, a town, village.
(6) koro, a halo around the moon.

Maori: aokoro, pukoro, a halo around the moon.

Viti : virikoro, a circle around the moon.

Arabic: hagar a, to have a halo.

There is complete accord from Efate through Viti to Polynesia in the
main use of this stem and in the particular use which is set to itself apart.
In Efate koro answers equally well for fence and for halo. In the marked
advance which characterizes social life in Viti and among the Maori the
need has been felt of qualifying koro in some distinctive manner when its
reference is celestial. In Viti virimbai has the meaning of putting up a
fence (mbai fence) ; viri does not appear independently in this use, but it
is undoubtedly homogenetic with Samoan vili, which has a basic meaning
of going around ; virikoro then signifies the ring-fence-that-goes-about, sc.
the moon. In the Maori, aokoro is the cloud-fence.

The Semitic here is triliteral. While the sense concord is notable, the
form resemblance involves only the second and third Semitic consonants
and we are left without explanation of the first, no shadow of which appears
in our Pacific areas.

137.
kabu, koau, the native pudding, tied up in a bundle and cooked in the oven ;
kofu sa, to wrap up or enclose (as a pudding in leaves to be put
in the oven) . The pudding koau is laid on a mass of leaves
very wide and long which are rolled up or over it all around
completely enclosing it and then tied up.
Samoa : 'ofu, food tied up in a leaf ready to cook. Tonga : kofu,
to wrap up. Niue: kohukohu, to enclose, wrap up as in a
taro leaf. Maori : kohu, to cook in an oven any article con-
tained in a hollow vessel. Tahiti: oh u, food tied up and
cooked in a bundle. Uvea : kofu, clothing. Rapanui: kahu,
clothing, cloth.
Viti : kovu, banana leaf in which puddings are done up.
Arabic: kabba, to make food into balls for cooking; kobbat, kabab',
food so cooked; kabkaba, to be wrapped up, enveloped.

Our author cites koau as a dialectic form of his theme kabu, which
hardly seems likely. The identification eastward with kovu and kofu is
satisfactory. The clothing signification in Viti and Polynesia is derivative,
and since that secondary sense does not appear in Efate it has not seemed
necessary to give it extended consideration.

The Arabic certainly shows resemblance in form and, so far as is admis-
sible in a different practice of kitchen mechanics, in sense also.

DATA AND NOTES. 221

138.
lako, loko, laku, loku, roko, nrok, to stoop, to be curved.

Samoa: lolo'u, to bend down or around. Tonga: loku, to draw
together, to gather in sewing, to pucker in a heap. Futuna,
Niue: loloku, to bend, to curve. Hawaii: lou, to bend around;
loulou, to bend down. Maori: roku, to be weighed down, to
decline.

Viti: roko, a bowing posture, bent like a bow; rokota, vakarota, to
bend a bow ; rokova, to pay respect to ; vakaroko, to bow down
with weakness, to go stooping.

Arabic: raka'a, roko', ruku', to stoop, to be curved or bent, to bow
or be bent down.

139-
(a) nab wo, nabwoa, tamo, to smell (intransitive). Cf. 221.

Samoa: ndmu, odor, to have a bad smell. Tonga: namu, to
smell; namuaa, namuku, bad in smell. Futuna: namuku, to
smell bad. Niue: namu, a smell; namud, bad smelling. Uvea:
namu, nanamu, odor; namuku, a stench. Tahiti : naminami,
stinking; namurea, savor. Nukuoro: namu, namo, a scent,
a smell. Fotuna: ehnamu, to stink.
(6) bwoa, bwon, odor, to emit odor.

Samoa: poapod, fishy smelling. Tonga: boa, the smell of fish;
tauboa, to scent the water with fish to catch others. Futuna :
poa, popoa, to smell fishy. Maori: poa, to allure by bait, to
chum. Tahiti: parupoa, to bait for fish. Hawaii : po, puia,
to emit odor. Mangareva: poa, bait, oil cast on the water
to attract fish ; akapoa, to communicate a smell of flesh to the
water to attract fish.

Viti: mboi, to smell, to yield a perfume.

Arabic : fdha, to emit odor.

In comparing bwoa and nabwoawe suspect the na to be verb-formative in
much the same use as the no thus employed in neighboring, but far more
distinctly Polynesian, Fotuna. Thus there can be no association of nabwoa
with Polynesian namu. As relates to tamo, clearly not of the same stem
as nabwoa, Nukuoro with two forms might seem to provide the transition,
but this affects only the unaccented final vowel, the least important detail.
This leaves us the far more difficult problem of establishing a t-n mutation.
This does not exactly appear in the Efate material, the nearest approach
being t-ng found in mauta-maunga (61) as this instance is solitary. So, in
the wider Melanesian field we find but a solitary instance, namu (328)
mosquito Alo Teqel torn. The identification is not established.

The formal identification of bwoa with the poa of the eastern islands is
satisfactory. We are to note that in Polynesia the sense is highly special-
ized except in Hawaii, which, with Viti, is identical with Efate.

140.
nasu na, juice, that which flows out or exudes.

Samoa: su, ngasu, wet; sua, juice. Futuna: su, watery; sua, juice.
Nukuoro: suisui, wet. Tonga: huhu, wet; hua, huhua, juice;

222 THE POLYNESIAN WANDERINGS.

huai, to pour out; tigahu, damp, moist. Niue: huhua, liquid.
Uvea: huai, to pour; huhua, sap. Nuguria: hua, coconut
water. Hawaii: hu, to overflow. Tahiti: u, to be damp, wet.

Viti: suasua, wet, moist.

Arabic : nazza, to exude ; nizu, flow, water.

While there is a close association in all the material here collected we
are to observe that in the Polynesian three forms exist, that Samoa and
Tonga possess all three, that Futuna alone has two. These three forms
are su, sua, ngasu.

su: In the sense of wet this is found in Samoa, Tonga, Futuna, Tahiti.
The Hawaiian hu is in form a variant of su, but while the sense has to do
with liquids it does not conform to the meaning of this stem as elsewhere
found. Nukuoro suisui I regard as a derivative of su by means of the
verb-formative i.

sua: Is found in Samoa, Futuna, Tonga as juice ; in Uvea specialized as sap ;
in Niue as liquid or as juice; in Nuguria still more highly specialized as
the water of the coconut. The Tonga and Uvea huai, a verb-derivative
of the sua stem, means to pour; this is comparable with the Hawaiian hu
as showing probably the coming up to the surface of a primordial general
signification which elsewhere does not break through the highly particular
sense, and this will be taken to include the Niue in the sense of liquid.

ngasu: This is confined to Samoa and Tonga and has the particular
meaning of su, wet, damp, moist.

The Viti is of the sua form but of the su sense.

The Efate is of the ngasu form but of the sua sense.

The manner of the interrelation of these three forms is by no means simple.
If su meant wetness then sua would follow in the adjectival sense of being
wet, in the common Samoan system of formation. We see here the oppo-
site movement, which is anomalous; yet it will serve to fix the position
of Viti as regular. Su is clearly the basic element of all these words which
carry the common sense. I am not familiar with any use of a prefixed
nga which has but the object of forming merely ornamental compounds,
for ngasu equals su, and in Tonga ngahu equals huhu. In Efate nasu we
note that it is impossible that it is na-su, for na is expressly not the article.
We have but one other instance (125) of the Efate n representing a Poly-
nesian ng.

The Samoan suati is the equivalent of this Tonga-Uvea huai. The form
of the Viti vakasuasuataka, to wet, gives ground for the impression that the
t is radical, vanishing somewhat unusuallv from huai. We should then
have the stem suat and by progressive degradation reach sua and su.

Provisionally we may hold nasu to be the remnant of a parent nasuat
which has undergone in general an abrasion at each end. In this view it
seems particularly difficult to understand how the Semitic, instinct with
the zeal for triliteralism, should have come to sacrifice the already existing
third consonant and then have reached exactly the same stage of demoli-
tion as in Efate and the heart of Nuclear Polynesia, and all this without
having left a trace in crowded Melanesia and Indonesia which intervene.
A resemblance rather than an identification.

DATA AND NOTES. 223

141.

(a) rat i, tat i, nrat, to loosen, to unite.

Samoa: tala, tatala, to loosen, to untie. Tonga: tatala, to tear
off, to separate what adheres, to open, to rend. Futuna:
tala, tatala, to loosen, to untie, to disunite. Hawaii: kala, id.
Maori: tatara, loose, untied. Tahiti: tatara, to loosen, to untie.
Nuguria: taraki, to open. Marquesas: taataa, separated,
loosened.

Viti : ndala, to be open (of a shellfish) ; ndalanga, to open one's
mouth.

(b) mirati, minratinrat, to be loose, untied.

Samoa: matala, to be open (as a leaf), untied, unloosed. Tonga:
matala, open, expanded (as a leaf or flower), free from restraint.
Futuna: viatala, open, untied. Niue: matala, to open, as a
flower or leaf. Hawaii : makala, to open, to untie, to unloose.
Maori: matara, untied, untwisted. Tahiti: matara, untied,
disentangled, loosened. Rapanui: matara, patara, to untie, to
acquit, to clear; patala, to let loose. Mangareva: akamakara,
to cut the first thread so as to unravel anything. Mangaia :
matara, to be loosened. Paumotu: hakamataratara, to unloose.

Viti: mathala, clear, plain, understandable, to be unfolded as a leaf.

Hebrew : nat'ar, hitlr, to loose.

The metathesis is evident.

As between the groups of words here assembled the Efate mi is the
equivalent of the general via of condition. In the Polynesian of each
group we find no further evolution of meaning than will readily be reducible
to the basic signification upon which Efate" and Samoa are in exact accord .

The Viti merits consideration. We have instances in which the Poly-
nesian t becomes the Viti nd; we have instances of the t-th mutation ; this
so far as I can recall is the only instance in which / becomes both nd and th.
From the instances presented in this work we might be led to the conclusion
that /-initial becomes nd and /-medial becomes th. That this is not the
case is doubly instanced in words not included within the essential limit
of this work; for example, tca-thea, mutu-mundu, fiti-vindi. We can but
note the anomaly.

The final vowel is weakened in Efate and tends to vanish. In the dia-
lectic forms we see a tendency toward nasal reinforcement of the l-r, which
is unusual. The dialectic form tati may indicate an l-t mutation from the
tala stem or it may be an r-t mutation from the metathetic Efate rati stem.
We have no record of such l-t mutation between Polynesia and Efat£; in
the wider Melanesian field we find it in but a solitary instance sala (339)
path Bugotu hatautu. I incline, therefore, to consider it due to merely
local variation.

142.

ror, lor, the oily juice of grated coconut used to moisten or fatten puddings.

Samoa : lolo, oil, the coconut prepared for making scented oil ; loloi,

a dish of taro and coconut juice. Tonga, Futuna, Niue, Uvea,

Nukuoro, Nuguria: lolo, oil. Hawaii: lolo, brains, marrow.

224 THE POLYNESIAN WANDERINGS.

Maori : roro, id . Tahiti : roro, the brains of mankind . Mangaia :
roro, brains. Mangareva : roro, soft, pure milk from the breast
or from coconuts, the skull, the head. Paumotu: takaroro,
headache. Rapanui: roro, brain, skull.

Viti: lolo, milk of the coconut squeezed from scraped kernel.

Arabic : rd'a, to moisten bread with fat.

In Efate and Nuclear Polynesia we find complete accord upon the par-
ticular signification of the word. In the Polynesia of the Tongafiti swarm
we find the word identical in form, but in sense in complete accord upon a
different signification, to which brain meaning the ultimate migrations have
added in Hawaiian and Maori the meaning of marrow. Along the Tonga-
fiti track the word for oil is sinu, which appears in Nuclear Polynesia con-
currently with lolo, and in the western verge in Ticopia. Now in Nuclear
Polynesia the word for brain is derived from the coconut but from another
part, the spongy substance (uto) which in an old nut occupies the space
where the water has been : Tonga, Uvea, uto ; Niue, uhoniu o he ulu (coconut
sponge of the head) : Samoa, uto, the head as a term of abuse, while for brains
yet another coconut product is employed, one advanced in manufacture,
fai'ai the cooked juice, which in Futuna faikai is restricted to the literal
sense. This uto also signifies marrow in Tonga, Uvea, Viti (utoni sui).

We next pass to the southeastern terminus of all possible migration,
Mangareva and the Paumotu. Here we pick up once more the coconut
milk in Mangareva roro and find it extended to the human breast. Con-
joined with this we find a strange variation from the Tongafiti sense, a
passage from the soft parts to the hard, from the contents to the calvaria.
Here the word uto means marrow, utuhupoko brains. Altogether a strange
fact and remote spot in which to find an inosculation of Proto-Samoan and
Tongafiti. Rapanui unites the soft and the hard parts in this word.

The lolo reappears in such parts of Nuclear Polynesia as have the animal
as a component of Samoa palolo, Tonga balolo, Viti mbalolo. I cite a note on
this subject which I wrote out for Dr. William McMichael Wood worth, who
identified the palolo as the posterior epitokal part of Eunice viridis (Gray) :

Stair's derivation from pa'a-lolo, luscious crab, is out of all consideration; it is on all
fours with the classic definition of a crab as a small red fish that walks backward, for
pa' a (paka) could not in the Samoan system of word structure undergo such a syncopa-
tion as to cut itself in two. As the bit beastie is in no sense a crab, and I must claim
for my islanders that their intelligence is sufficiently high to prevent them from putting
two such dissimilar animals together, so in turn is lolo not luscious. The organs of
sense perception by which the Samoan apperceives lolo lie, not in the peripheral nerve
endings of the tongue, but of the fingers; it is a matter of touch and not of taste such as
luscious principally connotes. I got a very instructive glimpse at this word from my
cook boy and a dish of vermicelli soup. After it had served my uses the tureen went
back to the kitchen. I found the servitor dabbling his fingers in the dish, which he
pronounced to be fa'alolo. I regard the primal signification as one of consistency, some-
what custardy, a substance partially solid that may to a certain extent be grasped in
the fingers yet which seems to slip out and elude the grasp. That, it will be noticed,
is a thread that can be run through all the significations. It applies equally to the
palolo as you feel it in the water on the great day of its appearance. In the slightly
specialized sense of slippery it applies similarly to its other two compounds in the
Samoan, ngalolo and umelolo, both being fishes and the latter a variety of Naseus
lituratus or unicornis.

DATA AND NOTES. 225

H3-
suki, to stick, to stab.

Samoa: su'i, to sew, to stitch. Nukuoro: suki, to pierce, to stab.
Tonga, Futuna, Niue, Uvea : huki, to pierce, to prick, to punc-
ture, to lance. Maori: huki, pierced. Marquesas: huki, a
small stick used to strengthen thatch. Rapanui: huki, to
transpierce. Mangareva: huki, to pierce (said of lightning);
ukiuki, piercing, lancing. Paumotu: hukihuki, to bore, to
perforate. Tahiti : hut, to pierce, to prick, to lance. Hawaii :
huiuna, a seam, a uniting by sewing together.

Viti : thuki, a digging stick.

Lambell, King, Lamassa : suki, to sew. Lambell, Lamassa : suki,
to prick. Moanus : susui, to sew.

Arabic : s'akka, to transfix with a spear.

The root suk retains its k except in Tahiti and Hawaii, where this loss
is normal, and in Moanus. Our data instance two stems in Moanus involv-
ing k; in this instance fe-medial vanishes; in 305, there initial, it remains;
of course these two instances are not sufficient to establish the usage.

144.
takutaku, to speak.

Samoa : ta'u, to tell, to mention, to announce, to certify, to acknowl-
edge. Tonga : taku, to call by, to designate ; takua, to mention,
to call by name. Rapanui: taku, to predict. Maori : takutaku, to
threaten, to recite imprecations. Fotuna : no-tukua, to confess.
Viti : tukuna, to report, to tell.
Arabic : nataka, to speak.

The Fotuna word is no more divergently specialized away from the plain
sense of Efate and Viti than is the Tonga or the Maori; it is particularly
interesting because it confirms the Viti vowel plan.

In the proposed Semitic identification a syllable is supposed to have been
worn away by frontal abrasion in passing from a common parent to the
Pacific, or the Arabic has picked up a syllable. Without confirmatory
evidence of intermediate forms this assumption is too violent.

H5-
tui, a chief.

Samoa : tui, a chief, king. Tonga : tu'i, a king, a governor. Futuna :
tui, god, supreme ruler, king (of god only). Niue: tui, a high
chief. Uvea: tui, king.
Viti : tui, king or principal chief.

Arabic: waddu, watadu, watada, to fix, stake, make firm. Hebrew:
yated, a pin, a nail, a prince.

No explanation is offered of the anomalous ' in Tonga tu'i. The char-
acter is seldom used in Tongan and is not noted at all in Shirley Baker's
grammar of the language. How inconsistent with himself he is in regard
of this character in the dictionary is exhibited in this suite : ma'u to get

226 THE POLYNESIAN WANDERINGS.

(Samoa maud), maugataa hard to get, ma'ugofie easy to get. It seems to
me that he has inserted ' into tu'i in order to differentiate that word to the
eye from five other meanings of tut. Those of us who had the pleasure of
knowing Tonga under that polemical missionary-statesman will have no
difficulty in comprehending why he should see fit to accentuate the ruling
power. Other than this slight visible but inaudible deviation the word is
identical throughout Nuclear Polynesia and in Ef ate, to which it is limited .
It can have no possible connection with Semitic words for stake.

146.

tuku, to go down, to sink down, to lower.

Among the many uses of the apparently cognate Polynesian tuku I select
only such as concord with the Efate tuku, for a close study of the Samoan
tu'u (yielding no selection for this purpose) leads to the feeling that it is
the remnant of several dissimilar stems.

Maori: tuku, to subside, to settle down. Tahiti: tuutuu, to slacken
or ease a rope. Hawaii : kuu, to let down, to slacken. Tonga :
tuku, to slacken, to let go as a rope; tukutuku, to sink in the
sea. Futuna: tuku, to put down. Niue: tuku, to bury.
Rarotonga : tuku, to let down, to let out, to drop down. Manga-
reva : tuku, to throw the fishing net or fillet. Paumotu : tuku,
to lay down. Sikayana: tuku, to put down. Nukuoro : tuku,
to permit, to allow. Manahiki, Fakaafo: tuku, to place.
Nuguria: tuku, to set. Rapanui: tuku, to give, to accord.

Viti : tukutha, to let go, to slacken a rope ; vakatukutha, to let down
in a basket.

Hebrew: s'uah, to sink down. Arabic: sah'a, sah'a, tah'a, t'ah'a, id.

F.FATE-MELANKSIAX-POLYNESIAX-SEMITIC.

147-
bano, to go, to go off or away.

Samoa : jano, passing along. Tonga : jano, to go, used in reference
to small fish going in shoals. Futuna: jano, to go, to depart.
Niue : jano, to go, to walk ; fcnonga, a journey. Tahiti : jano,
to set sail, to depart. Paumotu : jano, to set sail. Aniwa,
Sikayana, Vate: jano, to go. Nukuoro: liano atu, to go, to
depart; hano saine, to go around. Nuguria: uhano, to go.
Maori : whano, to go on. Tongarewa : hana, to go. Mangareva :
ano, to appear. Uvea: jangona (metath. janonga) road, path,
to go a journey. Fotuna: no-jano, to walk, to go.
Baki, Epi: mbano, to go. Meli, Fagani: jano, id. Maewo, Malo,
Baki, Mota Veverau, Motlav, Vuras, Merlav, Gog, Lakon, Santo,
Vaturanga, Bugotu: vano, id. Volow: vono, id. Malekula,
Marina, Omba, Arag, Mota Veverau, Volow, Leon, Sasar, Mosin,
Alo Teqel, Norbarbar, Maewo: van, id. King: va?iwin, to
come. Lo: ven, to go. Bierian Epi: mbene, id. Tanna:
(t)uven, id. Eromanga: ve, id. Mota Maligo, Ambrym,

DATA AND NOTES. 227

Sesake, Nggela, Belaga, Pak: va, id. Aneityum: apart, han, id.
Pala: han, id. Duke of York, Matupit : wan, id. Kabakada,
Matupit, Baravon : wana, id. Lambell : han, id. King : ivuan, id.
Norbarbar: vana, a going. Vuras, Gog, Lakon: vanog, id.
Motlav: navnog, id. Tangoan Santo: thano, to go.
Hebrew : panah, to turn the back, to turn to go.

See note 1 1 .

With the readily adjustable exception of the metathetic form in Uvea the
Polynesian is a unit and in perfect identification with Efate bano. Other
Efate forms are banbmai, banamai, banimai, compounds with directive mat;
banbtu, binotl, bariats, bhiats, compounds with directive atu, with which our
author most cryptically includes bin'en, baina, notu, net; ba-ki, which he
says is "contracted for" ban, bano, while ba (n) is not.

In considering the word in Melanesia we observe that there is a partition
into two markedly unequal areas according as the radical vowel is a or e,
to which is to be added the single instance of Volow which has o for a
(neutral vowel).

The a-series has a wide range among the labials from b to v, but the forms
in v are far the most widely spread . The only exception to this vertical muta-
tion is the Tangoan Santo involving a f-th or horizontal mutation, that is
from labial to lingual, a change of which we have no other example, yet in
this case it seems quite correct. Aneityum, Pala, Lambell han, according to
my principle of the aspiration, is a vertical mutation. It maybe that in this
we see a transition form to Tangoan Santo thano; a labial h reached verti-
cally would in no way, save with the resources of comparative philology,
be distinguishable from a lingual h, and for the change from this h vertically
downward to th we have abundant evidence ( 1 7 Journal of the Polynesian
Society, 160). The mutation to w is abundantly supported and is all but
vertical; it is found only in the gate by which the Samoa track leaves
Indonesia, and, disregarding the extra prefix, which is probably formative,
I incline to regard the vu of King ivuan as but a reinforced w. In Kaba-
kada, Matupit, Baravon wana we shall best regard the final vowel as an
o-a change within the triangle of the neutral vowel ; the a-final in Norbarbar
vana is formative and establishes the verbal noun of van to go, as appears
above. In Vuras, Gog, Lakon, Motlav the final g is formative of the verbal
noun; the Motlav is na-vno-g, article-stem-suffix. There remain for con-
sideration a few irregular forms. Aneityum a pan we include because our
author includes it as an identification, which it is not. Apan means to go,
apam means to come ; the merest tyro should recognize the common stem
as apa, far removed from vano and further exhibited in apahai to go land-
ward and apaahni to go everywhere. He quite overlooked, or did not know
where to look for han, which is the true identification. The King vanwin
contains the vano stem together with an extra element which in the paucity
of our material from that center of New Ireland culture we may not
comprehend.

The smaller e-series runs through its simple course of mutation and abra-
sion and calls for no more extended comment than to call attention to Tanna
(t)uven as a mixed form.

228 THE POLYNESIAN WANDERINGS.

We have now left for consideration the va and the ve. In each case
we have a record of the abrasion of final vowel and then of the consonant
left final, thus there is no reason in etymology why we should not regard
them as regular mutations of the fano stem. Since Mota shows vano in
Veverau and va in Maligo, where the differences are no more than dialectic
and of neighbor dialects at that, we are forced to the conclusion that va is
from the fano stem despite the efforts of our author to confuse the record
as shown in 1 1 .

We are not left to the compelling power of resemblance visualized to
comprehend Dr. Macdonald's indentification of the fano stem with the
Hebrew panah. Here is his extended argument in full as spread upon the
record under the classification of "triliterals doubly weak, that is, with
two weak letters or quiescents."

"Efate bano-mai or bana-mai, to come ; banats, i. e., ban ats, to go ; Maori
whanatu; Efate bano, to go; Maori whano, to verge toward, to go on,
proceeding toward; Hebrew panah, to turn, to turn oneself, to turn the
back, to turn in order to go anywhere. Thus banotu, whanatu, equals to
turn, going away, or outward, and bano-mai, bano-be, equals to turn
coming, to come." Any person who can find herein the common term
wherein to turn equals to go will have no greater difficulty in finding in fano
a trilateral, even though two of its letters are "weak or quiescent."

148.

batu, bate, to close up the roof by weaving thatch on the ridgepole.

Samoa: fatu, to commence plaiting. Tonga: fatu, to tie rafters,
to commence plaiting. Futuna : fatu, fetu, to plait. Uvea :
fetufetu, id. Niue: fatufatu, to fold; fatunga, a rafter. Tahiti:
fatu, to plait, to braid. Paumotu: pifalufatu, to fold.

Rapanui: haatu, to plait; hahatu, a plait. Hawaii : haku, to
braid a wreath; hakuhaku, to fold up. Maori: whatu, to
weave. Mangareva : atu, to fold up.

Epi : bofungo, to close up the roof by weaving thatch on the ridgepole.

Hebrew : 'abat, to interweave ; 'abot, wreathen work.

So far as it is given us to follow out the tangles of some of these definitions
it appears that Dr. Macdonald has sought to draft a statement which should
allow him to incorporate fatu the ridgepole with this batu.

The Polynesian has a word, stem fatu, which signifies to plait or braid
or weave or in some such way with deft fingers to reduce a tangle of shreds
to order. In Niue, Mangareva and the Paumotu the sense is restricted to
one of the incident operations in this broader meaning. In Hawaii we find
the two senses side by side, thus linking the two groups of meaning. With
this Efate batu is in accord so far as relates to the matter of weaving which
our author has distinguished by italics. This stem in Polynesia is nowhere
particularly applied to thatching, this roof covering being put on by imbri-
cation and not by interlacing.

But Tonga fatu to tie rafters, Niue fatunga a rafter, are distinctly roof
words. That, while they are identical in form with the weaving fatu, they
are not homogenetic therewith is apparent in the existence of Efate fatu
a ridgepole. In Samoa the roof fatu is found in fatu'ulu the thatch next

DATA AND NOTES. 229

the ridgepole, and in jatulau old thatch, and in fatunga the timbers to which
the purlins are fastened. The ridgepole Jatu is continued in the same sense
to Tangoan Santo papain, Malo uobatu, Malekula Uripiv uobut, and Bierian
botqu.

The Epi word Dr. Macdonald notes as identical in sense with batu, it
seems in no other wise related.

149.
iki, kiki, riki, small.

The following words all mean small :

Maori: riki. Samoa: li'i. Tonga: iki, liliki, likiliki. Uvea:
liliki. Nuguria: likiliki. Futuna: ikiiki, liki. Niue:
ikiiki. Tahiti : Hi. Hawaii : Hi. Rotuma : lilii. Marquesas :
iki. Mangareva: rikiriki. Paumotu: rikiriki. Mangaia:
rikiriki. Sikayana : likiliki. Samoa, Tahiti, Rarotonga,
Manahiki, Rapanui, Maori: Hi.

Duke of York: lik. Baki: teliki. Mae wo: riki. Nifilole: laki.
Motu : malaki. Kabadi : mara'i. Guadalcanar, Pokau : kiki.
Tubetube: kikiu. Nada, Kiriwana: kikita. Murua: kakiti.
Oiun: kafakiki. Galavi: berokikina. Bougainville: kekereke.
Aneityum : tintin. New Georgia : kikina. Pak, Leon : tiktik.
Vuras: netui, menet. Merlav: wirig. Mota: rig. Sasar:
wogrig. Alo Teqel : wowrig. Gog : iveskit, wesekit. Nor-
barbar: sosogut, sosogot. Retan: seget, sogot. Lo: ring.
Omba : mbiti. Ararg : tirigi. Sesake : kiki, ngiki. Fagani :
kikirii. Wango: kekerei. Nggela: kikia. New Britain:
ik. Baravon: ik,ikilik. Buka: kekereke, kikerei. Lambell:
liklik. Hula, Galoma: kirikiri. Keapara: kiri. Suau:
gagiri. Sariba: gagirini.

Ethiopic : clawik, to be small.

In the three Efate words for small the common element is ki. The
simplest form is the reduplicated kiki. With a vowel prefix we find iki,
and when that in turn is modulated by a liquid coefficient we find riki.

Two of these forms appear in Polynesian, kiki not having survived that
far:

iki: Tonga, Futuna, Niue, Moiki, Tongarewa.

liki: Maori, Samoa, Tonga, Uvea,Tahiti, Hawaii, Rotuma, Mangareva,
Paumotu, Mangaia, Sikayana.

In Melanesia we find all three forms, distributed as follows :

kiki: Guadalcanar, New Georgia, Sesake, Nggela, New Guinea.

iki: New Britain, Baravon.

liki: Maewo, Baki, Nifilole, Motu, Duke of York, Lambell, Arag, Mota,

Merlav, Lo, Sasar, Alo Teqel.
kiki and liki: Bougainville, Buka, Fagani, Wango.
iki and liki: Baravon.

Not immediately referable to the three forms in Efate our Melanesian
studies afford us the following forms:

230 THE POLYNESIAN WANDERINGS.

kit. This may be the basic ki with a persisting radical final consonant,
or the / maybe an irregular accretion ; it is impossible in the paucity of data
to come to a conclusion. If Pak, Leon tiktik be kit under metathesis this
is the simplest form. In Norbarbar and Retan, with floating vowel, we
have this stem with a prefix sc, so, common to both. In Gog I see such
resemblance to Retan that I diagram its composition as we-s-kit.

From this stem, by a common frontal abrasion, we may perhaps derive
the common word of Polynesia for small; Samoa, Tahiti, Marquesas,
Rapanui, Mangareva iti; Tonga jii (metathetic), Moiki itiiti by and by,
and Hawaii iki (a kappation of radical t and not to be confounded with
the stem iki). This would restore to Polynesia the possession of the
third Efate" stem, and it now seems to be proved by Rarotonga ngiti.

Omba mbiti we lack data to coordinate. The first impression that this
is iti with a prefixed modulant will hardly stand against the fact that */*
is an eastern form nowhere found in the west, and the other fact that we
know of no well attested instance in which a stem which has had its head
rubbed off acquires a new one by random selection.

We now are left with Vuras and Aneityum, which have the common
element net or tin according as we consider the metathesis to have been
applied to one or to the other. If we regard net as primal and as equivalent
to nit we shall find any attempt to associate it with iti by frontal accretion
blocked by the same bar as at Omba, and this somewhat considerable
array of data affords no example of a k-n mutation by which we might
account for it.

Uvea sii, with Aniwa aiihi, may open for us a small group of Melanesian
relationships in Aneityum sisi, Efate ses, sos; Epi takisi, Lakon sik, Volow
siwi, Motlav su, and thus connect with the se, so, element of the compound
forms of Norbarbar, Retan and Gog. It should be observed, however,
that Uvea sii devolves normally through Tonga jii from iti.

Dr. Codrington (Melanesian Languages, 81) comments "as is the case
with most adjectives there is but scanty agreement in the words meaning
little." I think that in working over his material quite another conclusion
has been reached and that the Polynesian content is distinctly marked
from end to end of Melanesia.

Indonesian affiliations are rare. We find Wahai kiiti, Bouton kidikidi,
Salayer kedi, and beyond these three we may scarcely venture, perhaps
not even so far.

In none of the foregoing is there aught which points to the Semitic
identification.

150.

kiat, the sticks which cross from the canoe to the outrigger.

Samoa: 'iato, the outrigger struts. Tonga, Nuguria: kiato, id.
Futuna: kiato, id. Fotuna: akiato, id. Maori: kiato, the
thwart of a canoe. Tahiti : iato, the transverse beams which
connect the outrigger to the canoe. Hawaii : iako, the arched
sticks which connect a canoe with its outrigger. Mangareva :
kiato, a large raft. Mangaia: kiato, the outrigger. Paumotu:
kiato, to pierce and cross for joining.

DATA AND NOTES. 231

Tanna : nikiatu, the outrigger bars.

Arabic : h'ata, h'iato, to sew, to join together.

151.
kubenga, a fish net.

Samoa: 'upenga, a net, not restricted to fishing. Tonga, Futuna,
Niue, Uvea, Nukuoro, Maori, Rapanui, Rarotonga: kupenga,
id. Nuguria: kupena, id. Anivva: kowpenga, id. Marquesas:
upea, tipeka, upena, id. Hawaii: upena, id. Tahiti: upea, id.
Mangareva, Paumotu : kupenga, a thread, a filament. Fotuna:
kanpenaua, the neck.

Sesake: kupenga, a net. Mota: gape, id. Merlav: gambe, id.
Gog: gamb, id. Lambell, King: mbene, id. Lamassa: mb'dne,
id. L,aur: mb'dn, id. Motlav: kmbweng, id. Volow:
nggmbweng, id. Pala: u6t'n, id.

Arabic: fc*/7a^ a net, from fca^fa, to wrap around.

In form the Polynesian word has the appearance of a verbal noun from
some verb stem kupe which we do not identify, for Tregear's suggestion of
Viti kumbe to catch hold of, to cleave or cling to, does not commend itself
in signification. Against this idea militates the fact that the word has
been preserved in Efate and Sesake, termination and all. The sense is
always and everywhere a net or meshed fabric, except in Mangareva where
it has become limited to that of which nets are made. The Fotuna form is
clearly a composite of kupenga and ua the neck, but its precise explanation
is not clear.

In Melanesia we find two groups of identifications after passing Sesake
as an unchanged loan word.

In the first group we have the languages which have assimilated a
putative stem kupe. These are Mota, Merlav, Gog.

In the second group are the languages which, while abrading the first
syllable, have used as their stem penga and have preserved more or less of
the seeming noun-formative termination. In the order of their strength
these are Motlav, Volow and the New Ireland group. The Pelion on Ossa
of the Motlav and Volow forms, it must be remembered, are each a single
sound ; therefore these do not, as might seem, reproduce the full kupenga;
it is just the Melanesian clumsiness in compassing a Polynesian p with lips
which play havoc with the fine precision of labials.

In view of the fact that the palatal nasal is most probably radical we may
scarcely accept the Semitic identification.

152.
lafi, to take up, to carry.

Samoa : lavea, to be removed, of a disease. Tonga : lavea, to bite,
to take the hook, as a fish. Futuna: lave, to comprehend,
to seize. Niue: laveaki, to convey. Rarotonga: rave, to
take, to receive. Tahiti: rave, to take. Mangareva: rave,
to take, to take hold; raveika, fisherman. Maori: raw, to
take up, to snatch. Hawaii: lawe, to take and carry in the
hand. Marquesas: ave, an expression used when the fishing
line is caught in the stones.

232 THE POLYNESIAN WANDERINGS.

Mota, Omba: rave, to catch (of fish). Lo: rav, id. New Britain:
rapa, to take by force. Nggela, Belaga : lavi, to take. Maewo :
lailai, id. Arag: lai, id.

Arabic: raja' a, to take up, to carry.

Samoan lave has a line of significations none of which is found outside
its immediate vicinage in Nuclear Polynesia. The nearest approach to the
Efate signification is in lavea, and that is not to be supported against the
first objection that may suggest itself.

The sense of taking extends from Efate to Fotuna and thence eastward
to Rarotonga, Tahiti, Mangareva, Maori. The carrying sense is found in
Nine. The two come together no nearer than Hawaii.

The specialized use in relation to fishing is absent from Efate; but its
presence in Tonga, Mangareva and the Marquesas of Polynesia, and in Mota,
Omba and Lo of Melanesia is amply suggestive of a common source. The
other Melanesian identifications call for no comment, except that in Arag
and Maewo; if this identification be acceptable, we have no other example
in all this material of the vanishing of a v. These two, therefore, remain a
little more than doubtful.

153-
leo, le, lo, lu, voice, speech, word.

Samoa: leo, voice, sound, noise. Tonga: le'o, voice. Futuna:
leo, id. Uvea, Nuguria, Niue: leo, voice, sound. Hawaii:
leo, voice, sound, speech, language. Maori, Tahiti, Mangareva:
reo, id. Rarotonga: reo, voice. Paumotu: reko, voice, speech ;
reo, the air of a song. Rapanui: reo, voice, language, air of a
song, a tale; hakareoreo, a story, to tell. Aniwa: noreo, voice.
Marquesas: co, ceo, id. Uvea: lea, to say, to speak to, to
accost, to address. Tonga : lea, speech, language. Niue : lea,
to speak. Rotuma : Ho, voice.

Sesake, Arag, Mota, Omba, Maewo: leo, the voice. Mota: lea,
speech. Maewo: leo, law. Santo: liona, the voice.

Arabic: la"a't to speak; la"w', sound, voice; lo"at, word, language,
dialect.

The Proto-Samoan stem is leo, an open form and thereby readily distin-
guishable from leo (31 0 to watch, which stems in leos.

The general sense is the sound of the voice in speech or in song, and to
this signification the word is confined in Samoa, Tonga, Futuna, Uvea,
Niue, Aniwa, Rarotonga, Paumotu and the Marquesas, that is to say all
of Nuclear Polynesia with three tongues of the eastward migration ; and the
same is true of the occurrence of the word in the New Hebrides except
Efate. If Paumotu reko be really referable to this stem the interpolation
of the k is anomalous. To this idea of the sound of the voice is added a
connotation of the product of the voice, and we find the signification of
speech, language, in Hawaii, Maori, Tahiti, Mangareva, in the Paumotu
subject to the doubt already noted as to reko, and in Efate. In the latter
we pass still farther into particulars with leo meaning a word. In Maewo
the secondary meaning ho law receives no confirmation elsewhere. On
Tonga lclo see note 145.

DATA AND NOTES. 233

In four languages, one of Melanesia and three of Polynesia, we find a
different yet appreciably similar word for the expression of this latter sense
in differentiation from the former. These are:

Uvea: leo, voice, sound; lea, to say, to speak to.
Niue: leo, voice; lea, speech.
Tonga: le'o, voice; lea, speech, language.
Mota: leo, voice; lea, speech.

We are, therefore, abundantly warranted in predicating two parallel
stems of which lea, by sense similarity and by reason of carrying its form
and sense distinction nowhere else than in an unaccented final vowel,
always a weak spot, has become assimilated to leo, this being particularly
true in the regions covered by the Tongafiti swarm. Efate we find in
possession of leo voice; it has not now lea speech; we are not justified in
the statement that it never had the latter, for the form le might survive
just as well from lea as from leo.

154-
lingi-si, malingi, malingsi, to pour out, to spill.

Samoa, Tonga: lingi, to pour. Niue: lingi, to pour in or out.
Uvea: lilingi, to sprinkle. Futuna, Nuguria : lilingi, to pour
out. Maori iringi, id. Raro tonga : riringi, to pour. Paumotu:
riringi, to pour from one vessel into another. Tahiti: ninii,
to pour out. Hawaii: nini, to spill, to pour out. Rapanui:
nininini, to pour, to shed ; hakanininini, to water. Marquesas :
iki, to pour out; iniu, teat.

Samoa, Tonga: malingi, spilled. Rarotonga: maringi, to spill.
Paumotu: maringi, to suppurate. Mangareva: meringi, to
trickle, to flow. Hawaii: manini, to spill or spatter out.
Tahiti : manii, to overflow, to be spilling.

Viti: livi, to pour gently or in a small stream.

New Britain: ligire, to pour out. Motlav, Volow, Merlav, Gog:
ling, id. Mosin: lenglengir, fluid. Mota: ligligira, ligiu, id.
Aneityum: aijangjing, to pour out.

Arabic : raka, to pour out.

The Proto-Samoan stem is lingis.

Efate has the simple stem and at least the form of the conditional deriv-
ative in ma. These two there will be no difficulty in tracing through the
Polynesian.

I have given room to Viti livi because of sense identity. Superficially
it is in form a three-quarters identification. In all my close study of the
Polynesian content of Viti this is the only case in which the leaping muta-
tion ng-v at all suggests itself, and there is not a single confirmatory instance
in this study of the broader field of Melanesia.

In the unlocalized New Britain instance collected by Tregear we are
uncertain whether this g is really g or represents ng, a common device in the
writing of South Sea languages. However this may be, in Mota there can
be no doubt that g is g and represents a not infrequent ng-g mutation. In
the other New Hebridean languages where the word occurs it has undergone

234 THE POLYNESIAN WANDERINGS.

no change more serious than terminal vowel abrasion, except for a slight
vowel change in Mosin. Aneityum has aijangjing to pour and ajaingjanse
to pour in ; this speech, so remote from either of the Proto-Samoan migra-
tion streams that it falls almost into the lowest class of quality of its Poly-
nesian content, has been too little studied to warrant our acceptance of
this identification. But we can sense a Polynesian ghost in its j-ng-s, all
the more since the l-j mutation plainly appears in lima (312) five Aneityum
ni-jman.

155-

liu, liliu, lilia, ler, bilui, bilu, to turn back, return, go or come back.

Samoa: liu, liliu, to turn, to go backward and forward. Tonga:
liu, liliu, to return. Futuna: liliu, to return, to go or come
back. Niue: liu, liliu, to turn, change, return. Uvea: liliu,
to turn, to return. Maori: ririu, to pass by. Tahiti: riuriur
to go around in a circle. Mangareva : akariu, to come and go-

Viti: Ha, to transform, to metamorphose.

Nggela : liliu, to change, to turn away.

The identification of liu is complete through all these notes. With the
dialectic occurrence of Efate lilia there need be no hesitation about admit-
ting Viti lia.

156.
manuka, a wound.

Samoa: manu'a, a wound. Tonga: manuka, manukaia, to kill,
murder, applied to chiefs. Fotuna : manuka, a sore, an ulcer.
Nuguria: manua, to wound.
Mota Maligo: maniga, wound. Mota Veverau: manuga, id.

Bierian: manika, a sore, an ulcer. Baki: menuko, id. Laur:
manug, a sore, an ulcer, an abscess. Baravon, Pala : manua,
wound. Malekula: menu, a sore, an ulcer.
Arabic : naka', to wound. Hebrew : nakah, id. Ethiopic : nakaya, id.

It is interesting to note that while on the one hand this word is confined
to Nuclear Polynesia it occurs on the other not only in the New Hebrides
but even so far back along the Samoa track as the east Indonesian gate.
Tonga makes far less use of the courtesy speech, the Polynesian Basakrama,
than does Samoa. In Samoa manu'a is an open word, the wound of a chief
is mdsoe. It would appear that Tonga manuka was an importation from
Samoa and was set aside for courtesy use as being a foreign novelty. Any
note is valuable which tends toward the elucidation of the differentiation
of Proto-Samoan and Tongafiti in Tonga.

157-

nakbe, a hollowed log used as a drum.

Samoa, Tonga, Uvea, Niue: nafa, a drum.

Malekula Uripiv: nambwi, a drum.

Hebrew: nekeb, a hollowed thing used as a musical instrument.

DATA AND NOTES. 235

158.
ruma, the breast, bosom.

Maori : uma, the breast, bosom. Samoa : uma, awide chest. Tonga,
Futuna, Uvea, Niue: uma, the shoulder. Nuguria: uma,

breast of a man. Marquesas: uma, the breast. Hawaii:
umauma, id. Tahiti: ouma, id. Rapanui: huma, breast,
chest.

Sesake: ruma, the breast. Dobu: rumaruma, id. Buka: nunume,
id. Motu: geme, id. Galoma: komakoma, id.

Arabic: ha'zum', the breast, the bosom.

Frontal abrasion is so well established that we need have no hesitation
in accepting the identification of Melanesian ruma and Polynesian uma.
The r-n mutation which will admit Buka nunume is the most frequent l-r
change in Polynesia, and in the Solomon Islands finds support in lima (313)
hand Saa ninime. Motu geme is much more remote and uncertain, and does
not well accord with the high quality of the identifications in that Torres
Straits migration station ; the r-g mutation finds a measure of support in
the l-k mutation lima (313) hand Vaturanga kima Nggao kame, where,
however, Motu has the abraded ami. Galoma with its k form adds con-
firmation.

159-
ruru, to tremble, an earthquake.

Samoa : lulu, lue, to shake. Tonga : luelue, to roll ; lulu, to shake.
Futuna: lulii, to tremble, to shake, to agitate. Niue: luelue,
to shake; lulu, to shake, to be shaken. Nuguria: ruhe, motion
of the hands in dancing; luhe henua, an earthquake. Uvea,
Hawaii : lu, lulu, lululu, to shake, to tremble, to flap. Fotuna :
no-ruruia, to shake. Maori: ru, ruru, to shake, an earth-
quake. Tahiti, Rarotonga, Rapanui, Paumotu : ruru, to shake,
to tremble. Mangareva: ru, to tremble; ruru, to shake.

Marquesas : uu, to shake the head in negation ; uuuu, to shake
up. Uvea: wet, to shake; ueue, to move. Rapanui: ueue,
to shake.

Nguna: ruru, a trembling. Mota: rir, to quake, an earthquake.

Syriac: rlel, to tremble.

Here we seem to have a common stem in two states.

The simpler is lu. This is found in its bare state in Uvea, Hawaii, Maori
and Mangareva. With a persistent reduplication, lulu, it is found in all
the above and in Samoa, Futuna, Fotuna, Tonga, Niue, Tahiti, Rarotonga,
Paumotu, Rapanui and the Marquesas, Efate, Ngunaand Mota. In Hawaii
it appears in triplication, and in the Marquesas even in quadruplication.

The modified stem is lue. This is nowhere found to the exclusion of
lulu, but with it is found in Samoa, Tonga, Niue and Uvea ; that is, it is
a Nuclear Polynesian type.

While there can be no difficulty in form or sense involved in the Mota
identification we notice the confirmatory agreement of Mota and Maori in
the particularized seismic meaning.

236 THE POLYNESIAN WANDERINGS.

1 60.

taru-si, taro-si, tarotaro, to pray.

Samoa: tatalo, to pray; talosanga, a prayer. Futuna: tatalo, to
imprecate, to desire. Tahiti: tarotaro, to pray. Rapanui:
tarotaro, a malediction, to curse. Hawaii: kalokalo, to pray
to the gods, to supplicate favors. Nukuoro: tarotaronga, a
prayer.

Mota, Arag: tataro, a prayer.

Aravic: sala', to pray. Ethiopic: salaya, id. Chaldee: sela, id.

The Proto-Samoan stem is talos.

The word has passed along its course with little variety in form or sig-
nification. Tregear, with such recognition of doubt as the cf. note may
carry, associates Avith this stem the Maori tarotaro to cut the hair. His
explanation "that the cutting of hair among Polynesians was generally
accompanied by a solemn and religious ceremony," while unimpeachable
as a statement of manners and customs, seems, for the purposes of philolog-
ical comparison, an impossible exaltation of the incident over the essential.

161.

taumafa, taumofa, to invoke or pray while sacrificing or giving an offering.

Samoa: taumafa, to eat, to drink (of a chief or a chief's pigeon).
Tonga : taumafa, to eat, applied to the Tuitonga but now used
to high chiefs. Futuna : taumafa, a thank-offering to the gods
or to a chief. Niue: taumafa, to eat, used to chiefs only.
Uvea: taumafa, a religious feast, to eat. Fotuna: taumafa,
an offering to the gods. Nukuoro : taumaha, taamaha, a feast,
a sacrificial feast; hakataumaha, to forbid. Maori: taumaha, a
thank-offering to the gods ; whakataumaha, to offer in sacrifice.
Tahiti: taumaha, an offering of food to the gods. Hawaii:
kaumaha, a sacrifice, to offer in sacrifice, to kill a victim for
sacrifice. Rarotonga: tau maa, to curse. Mangareva : toumaha,
a prayer offered up before a feast or a meal, to offer first fruits
to a god.

Malekula Pangkumu: tomav, to offer in sacrifice.

Hebrew: habhabim, offerings.

The Proto-Samoan stem is taumafat.

The identifications show remarkable accord in form and sense. It is
only in Samoa, Tonga, Niue and Futuna, strictly Nuclear Polynesia, that
the word is applied at all to mortal men. It might be taken as a piece of
gross flattery to the chiefs to assign to their eating the word which belongs
to the great gods. In my understanding of the spiritual ideasof thiscentral
area of Polynesia, which holds the most primal concepts of the race, it seems
more reasonable to regard the word as originally belonging to the great
chiefs (Polynesian Basakrama) and thence extended to the divine essences
when the Polynesians had learned to make gods in their own image.

Dr. Macdonald finds no difficulty in accounting for the word as a com-
posite of materials now in Efate, tau to pray or invoke, ma fa (mo fa) giving
or offering. The latter element he thus explains: "when the blood of men

DATA AND NOTES. 237

or animals has been shed and forms a pool on the ground, one feeling the
smell of it, or of any similar thing, says i nabwo mofa, it smells mofa," and
identifies it with Arabic ma'habat a small pool, wahaba to give, to make an
offering. Yet in the foregoing there nowhere appears for mofa the sense of
giving or offering which without hesitation he assigns when dealing with
it in the taumafa composition. The only sense which may be derived from
the narrative, not definition, of mofa is to say that when the smell of blood
is felt, if indeed that be possible in sense perception, and one says it smells
mofa the word is meant to describe the apperception of shed blood by the
nose. Not in the least inclined to accept the circuitous Arabic identifi-
cation I see in mofa no suggestion of the radical t.

Working on different materials Tregear has isolated tau to pray in Tahiti
and the mafa of mamafa heavy. The latter at least is Proto-Samoan maf at.
But tau nowhere else in Polynesian means to pray. Bishop Jaussen gives
tau in the prayer sense but altogether omits tarotaro (160), which Tregear
must have obtained from English missionary sources. Assuming the sub-
stantial accuracy of the bishop's dictionary it is possible that Tahiti tau
by the not infrequent loss of / (e. g., valu eight Tahiti vau) may be associable
with talo to pray. The association with mafat is excellent in the preserva-
tion of the radical t, but the correlation of sense leaves much to be desired.

162.
tauri, to bind, to be bound firmly to, to marry (a woman), to tie firmly to
(as a boat to a ship to be towed) ; taura ki, to be fixed or bound
firmly to one, bringing out one as from bondage or from her
relations, to redeem, to marry; bitauri, to be bound, tied,
attached firmly to each other, to be married.
Samoa : tau, taulia, to be anchored, to be fixed (as colors in cloth) ;
taula, an anchor; taulanga, anchorage; taula'i, to anchor with
or to; tauvale, to marry beneath one. Tonga: taulanga,

anchorage ; taufau, to tie. Futuna : taula, an anchor, a cable ;
taulanga, anchorage. Niue : taula, an anchor. Uvea : taula,
an anchor; taulanga, anchorage. Nukuoro: taura, a rope.
Maori: tow, to lie at anchor or moorings; taura, a rope; tauranga,
anchorage; tatau, to tie; taunga, a support, bond, tie, a bond
of connection between families; taumau, betrothed. Tahiti:
tau, an anchor; taura, a rope; tauri, to be intermixed as a
family in a house ; tardea, to rescue, to deliver. Rarotonga :
taura, a rope. Marquesas : tau, a rope ; katau, atau, an anchor.
Mangareva: tauri, to tie together; tour a, a cord. Hawaii:
kau, to tie on ; kaula, a rope ; kaupili, to unite, as man and wife.
Sesake: matau, an anchor. Mota: taur, to hold.
Arabic: sabara, sabr', to bind, to be bound to.
The Proto-Samoan stem is taul.

The Polynesian tau is protean in its shifts of meaning. I have here segre-
gated only such as are associable with Efate tauri. Other senses will be
found in 236 and 267.

In the Melanesian identifications Mota is satisfactory. We lack the data
to establish the value of ma in Sesake matau which seems to be a composite
of this tau.

238 THE POLYNESIAN WANDERINGS.

163.

telatela, matulu, mutultul, matoltol, large, swollen.

Samoa: tele, latele, vatele, large. Mangareva: terc, to be fat, to

swell out. Maori: tetere, large, swollen. Hawaii: kelekele,

fat, plump, large.
Samoa : matolutolu, matontou, thick (of pork only) . Tonga, Futuna,

Nukuoro: matolutolu, thick. Niue: matolu, id. Uvea: matolu,

thickness. Hawaii : viakolu, wide, thick. Maori: matotoru,

thick. Fotuna, Tahiti: matoru, id. Mangareva: matoru,

fat, thick. Marquesas: motou, thickness.
Mota: matoltol, thick. Epi: torn, large. Norbarbar: motoltol,

thick. Nguna: matulu, id. Baki: mererolu, id. Malekula:

metetir, id.
Hebrew: W*r, large, great.

Dr. Macdonald includes in his identifications Samoa telatela, which, if
valid, would avail to connect the two groups which follow. Unfortunately
this identification is form resemblance only, for telatela in Samoan is the
clitoris and in no way associable with the signification of these stems. Yet
without this there is abundant reason to consider matolu a conditional
derivative of tele. If this be the fact, the conditional, which remains practi-
cally uniform, must be held to preserve a primitive which since then, in its
unsupported state, has developed other vowels.

In the Melanesian identifications all is plain except Baki and Malekula.
They seem to have me as the conditional element. The primitive, then,
will appear in Baki as rerolu and in Malekula as tetir, evidently reduplica-
tions. In view of the plural preduplication of Samoa tele tetcle Malekula
tetir seems explicable in form and interesting as exhibiting the much desired
transition form between tele and matolu. Baki rerolu, apart from the irregu-
larity of vowel change in duplication, would argue a t-r mutation. This is
unusual but not impossible. These data afford four examples, which are
here presented for consideration : futi (329) banana Moanus mbur; ate (276)
liver Malekula ere Efate are; fatu (294) a stone Malekula var; talinga (350)
ear Malekula riring. The last I withdraw for reasons which will be offered
sub voce. The Baki form may pass.

164.

tere, teretere, (a) the comb of a cock ; (b) the eaves of a house.

(a) Samoa: tala, a thorn, the barb of a spear, the spur of a cock;
talatala, prickly, thorny, rough. Tonga: tala, a thorn, the
pricking fin on the back of a fish ; talatala, tala, thorny, prickly.
Futuna : tala, thorn, horn of an animal, fin of a fish ; talatala,
prickly, thorny . Niue : talatala, prickly. Uvea : tala, thorn,
horn ; talatala, prickly. Nuguria, Nukuoro : tara, a thorn.

Maori : tara, a point, as a spear point, spines in the dorsal fin of
a fish, membrum virile, clitoris ; taratara, a spine, spike, prickly,
rough. Tahiti: tara, a horn, thorn, cock's spur; taratara,
prickly, thorny. Rapanui: tara, a horn, spine, thorn. Hawaii.
kakala, rough with sharp points ; kalakala, thorny. Mangareva :

DATA AND NOTES. 239

tara, a horn, a spine, crest of a bird ; taratara, prickly, thorny,
spiny. Moriori: hokotara, to sharpen. Marquesas: taa, a
thorn, spike, point.
(b) Samoa: tala, the round ends of a house. Maori: tara, the side
wall of a house. Tahiti: tara, the corner or end of a house;
fautarajare, the bend of the round part of a house. Hawaii :
kala, the ends of a house in distinction from the sides.

Viti : teretere, combs of some birds, crest of serpents.

Duke of York: talaglagano, thorny.

Arabic : torra, crest, comb of a bird ; torrat, extremity, end of anything.

There is no reason to suspect any closer association between these two
groups than is involved in identity of spelling, which, despite change of
vowel in Polynesian, extends to their last occurrence. The Viti carries
the Efate vowel scheme within the borders of Nuclear Polynesia and serves
as the connective for the thorn sense.

In the architectural tala, Samoa, Tahiti and Hawaii agree upon the ends
of the house, Samoa and Tahiti upon their being curvilinear; with a differ-
ent plan of structure the Maori assign it to the side wall. Of the Efate word
the most that can be said is that it is architectural even if widely different
from any Polynesian use.

The only Melanesian resemblance beyond Efate, Duke of York talaglag-
ono, in the east gate of Indonesia, suggests a stem talak, of which no trace
survives elsewhere and the gap in Melanesia precludes confirmation.

165.
tere, the mast of a ship, calf (column) of the leg.

Samoa: tila, the sprit of a sail. Futuna: tila, the long boom or

sprit to which a sail is bent. Tonga: jila, yard of a canoe.

Fotuna: jira, mast. Maori: tira, mast of a canoe. Tahiti,

Rarotonga, Mangareva: tira, a mast. Uvea: sila, yard.

Makura: no-tire, mast. Tanna: tila, id. Bierian: n'dalin, id.

Arabic: sariyat, sari, the mast of a ship, a column.

In devising his second definition Dr. Macdonald has squinted so strabis-
mically at his Arabian entertainment that a single degree added to his angle
of vision would have turned him into a One- Eyed Calender.

In the Polynesian group the t-s mutation in Uvea sila is sufficiently
common.

The Melanesian identifications are simple, except Bierian n'dalin. Omit-
ting the final n, for which we have no explanation other than that it may
be noun-formative, we may reduce the word to tali, and this, it is at once
apparent, is metathetic for tila.

The variety of sense as between boom, sprit, yard, and mast is more
apparent than real. It is due to an attempt to establish as a fixed concrete
a term which is undoubtedly abstract and descriptive. In the navigation
where the triangular sail has an enormous sprit or boom and a short mast
tila goes with the longer spar; where the mast is high and the sprit, boom,
yard or gaff is subordinate, tila still goes with the longer spar. If we had
the data whereby to analyze tila down to its elements we should probably
find that it meant no more than long spar.

240 THE POLYNESIAN WANDERINGS.

1 66.
tifai, thunder, ti, article, and fai.

The following words all mean thunder and those in which the fai element
is indisputable are grouped :

Samoa, Fakaafo: faititili. Tonga: Jaijijili. Manahiki: faititiri.
Tahiti: patiri. Tongarewa: hatitiri. Maori: whatitiri,

whaitiri. Bukabuka: watitiri. Paumotu: fatitiri. Nukuoro:
haturi. Rapanui, Mangareva: atutiri. Hawaii: hekili.

Nuguria: hetuturi. Marquesas: hatiitii, fatutii, hatutii.

Aniwa, Fila: te-fachiri. Fotuna: vajiri. Vat6: vatshiri.

(Malay: tilir, to make a noise, a noise which gives alarm.)

Arabic: bahh', hoarse, used of thunder.

Dr. Macdonald's simple attempt at etymology in respect of his designa-
tion of ti as article must have reference to the Tongafiti te, the weak demon-
strative functioning as article at times, for his dictionary does not identify ti
in Efate" in any sense. The common Polynesian words for thunder involve
the two elements fai and tili. The Efate alone in Melanesia contains ele-
ments in the thunder word which resemble the Polynesian. As to the fai
there can be no doubt ; ti may readily be an abraded form of tili. I know
of no instance in which the Melanesian has inverted the order of the ele-
ments in borrowing a Polynesian composite, yet in default of more definite
information this seems not unlikely in this case.

The former element in Polynesia has the following forms:

fai: Samoa, Fakaafo, Tonga, Mana- wa: Bukabuka.

hiki. wha: Maori.
whai: Maori. ha: Tongarewa, Nukuoro, Mar-
fa: Paumotu, Aniwa, Fila, Mar- quesas.

quesas. he: Hawaii.

va: Fotuna, Vate. a: Mangareva.
pa: Tahiti.

The latter element falls into groups, which yet may not have diagnostic
significance, according to the duplication of the stem, as follow:

tili: Tahiti, Maori, Hawaii, Aniwa, Fila, Fotuna, Vate" (Nukuoro).
titili: Samoa, Fakaafo, Tonga, Manahiki, Tongarewa, Maori, Buka-
buka, Paumotu (Mangareva, Rapanui).
tilitili: Marquesas.

Mangareva atutiri, Marquesas fatutii, hatutii, suggest a fatu former ele-
ment. Yet the Nukuoro haturi, which can scarcely be hatu-ri, seems to
argue a remoter tutiri duplication through an ignorance of the stem vowel
which Nukuoro reveals. The first of the three Marquesas forms is recorded
by Tregear, but is not found in Bishop Dordillon's modern dictionary.

If we had none other than the /ai-composites we should say that the
former element signifies "to make." Not one of the other forms of this
element offers in any of the languages a sense which might explain the
composite, but not one of the languages has anything which would contra-
indicate this sense. Provisionally, therefore, we may assume that faititili
means "to make-tilt intensively."

DATA AND NOTES. 241

Nowhere in Polynesia can I find a meaning of till which would shed any
light on this composite, and I am forced to the conclusion that this element
has passed out of independent existence. The presence of the composite
in Aniwa, Fotuna, and Fila argues its great age as a compound. I am very
loth to accept an Indonesian identification except it be supported by a
satisfactory chain of evidence. To present such a resemblance based upon
a single instance goes against my practice in the present work. Yet with
this deprecation I note the Macassar djili the lightning flash. Faitili then
might mean that which makes the lightning. It may be that we shall find
a trifle in geographical support of this suggestion. Macassar is the region
in Celebes which lies in closest proximity to the island of Salayer, a name
which seems not remotely preserved in the honorific fa'alupenga of Samoa
in the high phrase for all Tutuila.

Tuloutia a 'oe le motu o Salaia. Saving thy grace, island of Salaia.

167.
ulumwa, a pillow for the head.

Samoa : alunga, alulunga, a soft pillow. Nuguria : aruna, headrest.
Tonga : olunga, pillow. Futuna, Uvea, Niue : ulunga, id . Maori,
Mangareva: urunga, id. Hawaii: uluna, a pillow, to sleep on
a pillow, to tie up a bundle for a pillow. Paumotu : r urunga, a
pillow. Tahiti: urua, turua, id. Rapanui: rangua, id.
Dufaure Island (New Guinea) : unua, a pillow. New Britain : ulula-
lag, id. Mota Maligo : ilinga, head-rest, pillow. Mota Veverau :
ulunga, id.
Mahri: here, haroh, hare, eres' , the head.
Dr. Macdonald repeats and again repeats the hint that this is a derivative
of ulu the head. Tregear notes that the word is probably connected with
uru the head, turn, and runga.

This can only apply in those instances which have initial u, Futuna, Uvea,
Niue, Maori, Mangareva, Hawaii, Tahiti, Mota Veverau. It leaves unsatis-
fied Samoa, Tonga, Paumotu and Mota Maligo. The Samoan plural is
proof that the word is a composite of which the latter element is lunga.
The former element is most commonly u, but a, o, and possibly i occur. The
Paumotu rurunga is accordingly a preduplication of the second element and
the common first element does not appear in it at all. If, then, lunga inde-
pendently or in composition carries the sense of headrest it will be seen
that the stem can not be ulu the head. The alternative Tahiti turua is
anomalous on either theory of the word. The Rapanui form is susceptible
of explanation by double metathesis.

Mota Veverau ulunga is of the general type. I incline to regard Mota
Maligo ilinga as a mere variant of local dialect. The only way by which
Dufaure unua can be brought into correlation is by establishing the loss
of n and the l-n mutation. Each of these changes is abundantly manifested
in Melanesia and Polynesia; whether they are current in Dufaure Island
the only other word we have from that speech (ama (340) outrigger Dufaure
sarima — unsatisfactory) does not allow us to judge. The New Britain
ululalag can be correlated only as a compound of ulu head and lalag equiv-
alent to lunga with reduplication, and there is no evidence forthcoming to
support the latter, while the former is distinctly contraindicated.

242 THE POLYNESIAN WANDERINGS.

168.

utu, ut i, to fill (by immersing) a water vessel.

vSamoa : utu, to draw water, to fill a bottle, to charge a gun. Tonga :
v.tu, to draw water; utufia, to pour out, to run as water from
a vessel or tears from the eyes. Futuna : utu, to draw water,
to fill with a liquid ; uku, to plunge into the water. Uvea : utu,
uutu, to draw water, to pour into. Niue : utu, to draw water.
Nuguria: utu, to fill. Maori: utu, to dip up water, to fill
with water. Tahiti: utuhi, to dip into the water, to rinse.
Mangareva : utuhi, to draw water. Hawaii : ukuhi, to pour,
as water into a cask, to fill a vessel with any fluid. Mangaia:
uti, to draw water. Fotuna: no-citu, id. Rapanui: uutu,
to fill up ; ootu, to draw water.

Motu: utu, a flood, to draw water. Tanna: atu, id. Aneityum:
athun wai, id.

Arabic: "ata ("a'tu) to immerse.

The Proto-Samoan stem is utuf.

Rather more frequently than is the case in general the influence of the
radical / is found not only in Nuclear Polynesia but persists to distant
extensions of the Tongafiti swarm. The Mangaia uti is singular in Poly-
nesia but accords with an alternative Efate form. The irregularity of the
initial vowel in Fotuna looks toward the Tanna and Aneityum forms, and
they are its close neighbors.

Dr. Macdonald in his definition squints as usual quite obliquely at the
Arabic. It is only by the accident of the container, the common water
vessel of the South Sea being the coconut pierced at the eyes, that it is
more convenient to fill by immersing. That the immersion in no sense
inheres in the word is shown by the Samoan use of utu to charge a gun or
to cram tobacco into a pipe, and by the general use of the word to signify
pouring out from the container.

EFATE-MELANESIAN-POLYNESIAN-MALAY.

169.
finanga, food.

Samoa : sina'aiunga, grayheaded from eating the hermit crab (unga),
old but foolish. Tonga: hinakaiunga, grayheaded as a punish-
ment for eating the hermit crab.
Mota, Marina: sinaga, food. Malo: sinaca, id. Motlav, Omba:
hinaga, id. Lo: hinega, id. Sesake: vinaga, id. Duke
of York: ivinangan, id.
Malagasy: hinana, food.

There is a possibility that the Efate contains a misprint, for in Dr.
Macdonald's alphabet the ng differs from the g by no more than a dot above.
The mutation g-ng is by no means impossible, but it is strange that Efate
is the only speech except the distant Duke of York in which this word
varies from the standard.

The two words in Nuclear Polynesia have hitherto been a puzzle unsolved.
The interpretations offered in the dictionaries of Pratt and Baker respec-

DATA AND NOTES. 243

tively are just such philology as the islanders invent for the student who
attempts to go under the surface of their speech. It is of a piece with the
Samoan etymology of tangata man from ta to strike and ngata snake, with
Genesis iii, 15, cited as decisive. We now reduce the word to sinaka food
in a verbal use as shown by verb-formative *, and unga, the particular sense
of which does not yet appear.

The Melanesian forms are all in accord, even including Duke of York
winangan, for in this island we find s-w mutation although no other example
thereof is included in these data ; and the Sesake vinaga is transitional.

EFATE-POLYXESIAN-MALAY-SEMITIC.
170.

afis i, afit i, afin i, afan i, afen i, to put or carry under the arm or arms
held between the arm and the side, to cover with the wings as
a bird its young, clasping them between the wing and the side ;
afini na, the armpit; afili na, the armpit, the groin.

Samoa: afisi, to carry under the arm, to carry a child astride the
hip, afisinga, a load carried under the arm. Tonga: efi, to
carry under the arm. Futuna: efi, to carry under the arm,
the under side of a fin. Mangareva: ahi, when used after
a word signifying a load or a bundle it means a load carried
on the chest or in the arms. Maori : awhi, to embrace ; awhe,
to measure a tree by embracing it. Tahiti: ee, armpit.

Malay: kapet, manga pet, to carry under the arm.

Arabic: "abana, to put under the armpit; ma" bin , groin, armpit.

The Proto-Samoan stem is afis.

This appears only in Samoa and in Efate, where variants are found in
/, n and /. The efi of Tonga and Futuna marks the transition in respect
of the former vowel, and Maori awhe the transition form for the latter;
together they substantiate Tahiti ee, the / normally vanishing in that
language.

171.

bangobango, to be crooked.

Samoa : pi'opi'o, crooked, wrong. Futuna, Tongarewa, Marquesas,
Mangareva, Paumotu: piko, bent, crooked, awry, twisted.
Nukuoro : piko piko, crooked. Niue : piko, to think erroneously ;
piko piko, to speak falsely. Uvea: piko, pipiko, crooked, slug-
gish. Maori: piko, to bend, to stoop, to be curved. Rapanui:
hakapiko, pliant, to bend, to make crooked. Tonga: biko,
crooked. Tahiti: pio, bent, crooked, wrong. Hawaii: pio,
bent, crooked.

Malay: bengkok, crooked. Malagasy : vitkuka, id. Java: bengkong, id.
Saru: pekok, id.

Hebrew: hafak, to turn. Syriac: hpak, id. Arabic: 'apaka, id.
Hebrew: hapakpak, crooked, twisted.

244 THE POLYNESIAN WANDERINGS.

172.
bwala, to be smooth.

Samoa, Tonga, Futuna, Niue, Uvea: molemole, smooth. Nukuoro:
molemole, soft. Hawaii: molemole, smooth, bald. Tahiti:
moremore, id. Maori: more, bare, plain, bald. Mangareva:
akamore, to decapitate, to cut off horns. Paumotu : moremore,
without hair on the body, polished.
Malagasy: bory, deprived of, shorn, cropped, polled.
Arabic : mara, maur', to fall off or pluck out, as wool or hair.

173-
biau, beau, biaufiau, biafiau, a wave; biaufiau, to be raised in waves,
rough sea.
Samoa, Tonga, Futuna, Fotuna, Niue, Marquesas, Nuguria: peau,

a wave. Mangareva: peau, peahu, id.
Viti: mbiau, a wave.

Tanna: peau, a wave. Aneityum: ne-peau, id.
Malay : ombak, a wave.

Ethiopic: ababi, waves. Arabic: 'ubab, id.

174.
bule, fule, complete (adverbial).

Tonga : fuli, fulibe, all, everyone. Futuna : fuli, all, universal ; fuliai,

all without exception. Uvea: fuli, all, universal, the whole.
Malay: bulah, the whole.
Arabic: bala'la, bulu", to complete, to go through to the end.

175-
busf, to blow, to spout as a whale.

Samoa : pusa, to send up a smoke (applied also to dust, spray, vapor),
Nukuoro: pusa, steam. Fotuna: noh-pusa, to rise (of dust).
Tonga: buhi, to spit; bubuhi, to spout as the whale, to blow
anything from the mouth. Futuna : pupusi, to blow, a blast
of wind. Niue: puhi, to spurt out. Uvea: pupuhi, to blow.
Maori: puhi, to blow. Tahiti: pupuhi, to blow the fire, to
blow out a candle; puhipuhi, to blow with the mouth, with a
fan, with bellows. Hawaii: puhi, to blow (as wind), to blow
(the fire, a shell), to puff, to spout. Marquesas : puhi, to blow,
to breathe. Mangareva: puhi, to blow. Paumotu: puhipuhi,
to blow. Rarotonga: pupui, id.
Malay: ambus, ambusi, to blow; ambusan, bellows. Malagasy:

mifufutra, to blow the bellows.
Arabic: nafat'a, nafah'a, to blow out, to puff, to eject spittle.
I do not take cordially to Dr. Macdonald's inclusion of Samoan pusa in
his identification. This word is found also in Nukuoro and Fotuna and
nowhere else in Polynesia. The accord of all other instances upon pusi,
evidently pus advanced in manufacture by the verb-formative i, renders it
well-nigh impossible that this rare and rigidly Proto-Samoan pusa can be
associable therewith. The only difference between the two pusi senses is
that to spout is blowing made visible; this sense in Polynesia appears in
Tonga, Niue, and Hawaii.

DATA AND NOTES. 245

176.

butu, futu, butafuta, futfut, to spring up or out, as water from a spring or
smoke from a fire; butu-raki, buti-raki, to appear, to come in
sight.

Maori : puta, to gush out, to spurt, to come in sight, to pass through,
to pass in or out. Hawaii: puka, to enter in or pass out;
hoopuka, to appear in sight when at a distance. Marquesas:
puta, to arrive. Mangareva: puta, to go out from.

Malay : tarbit, to issue, to come out, to emanate, to appear.

Arabic: nabata, to spring up or out as water, to appear, to go or
come forth, to come in sight.

If it were not that the Maori and Hawaiian comprehend the two senses
in one word I should scarcely consider buta and buturaki associable. The
latter, despite our author's division of the word by the hyphen, seems to
be butur-aki, btdir-aki, in which the hi which he describes as transitive par-
ticle seems better comprehended as aki which is verb-formative all through
Polynesia. In this case the stem is closed with a liquid. The absence of
this closing radical and the difference of vowel in buta argue against its
association with the butur stem. The Polynesian forms preserve the a and
therefore have a formal identification with buta, but in the sense there is a
mixture with significations of butur. It will be observed that this is one of
the infrequent identifications with the Tongafiti swarm having no trace in
Nuclear Polynesia ; therefore it should not be accepted without a clearing
up of these difficulties far beyond our present power.

The buturaki, butiraki, sense is found in Nuclear Polynesia in Samoa fotu,
Tonga fotu, fotui, fotuaki, Futuna fotu, Viti votu, all of identical meaning.
Yet there is nowhere any trace of the radical final consonant. The Tonga
fotuaki would appear to negative the possibility that modern fotu stems in
fotur, but this is more apparent than real, for Tonga seems not to have
adopted these inflected forms until the final consonants of closed roots had
vanished. In Samoa, where closed roots are normally preserved in com-
posite (inflectional) forms, we have no record of any form which might give
evidence upon this point. Accordingly this is not offered as an identifica-
tion, but to record a resemblance sufficient to suggest identity. It is to
note that Samoan fotu could not undergo any known variation and become
Maori puta.

177.
ngal i, kal i, al i, to stir water around.

Samoa, Tonga, Futuna: ngalu, a wave, a breaker. Fotuna,

Maori, Rarotonga: ngaru, id. Mangareva: ngaru, scum,

froth. Paumotu: puhingaru, a bubble of water. Hawaii:
nalu, a wave. Tahiti: aru, id. Marquesas, kau, nail tai, id.
Malay: alun, a wave. Malagasy: aluna, id.
Hebrew: galal, to roll; gal, fountain, well, waves. Syriac: galo',
a wave.

In addition to the identification with ngalu Dr. Macdonald, led by abraded
forms of his ngal, adds the Hawaiian ale to well up, which is negligible.
Both he and Tregear link with ngalu a wave ngalue to shake. In this case

246 THE POLYNESIAN WANDERINGS.

the nga is a formative prefix, and the stem lite has already been studied
under 159.

The identification ngali-ngalu satisfies in form as little as in sense. In
the Polynesian the word has a distinct and concrete meaning. The Efate,
if it be regarded as concrete, means something quite different from ngalu;
and if it be regarded as a loose and abstract term it has nothing in it to
suggest a wave of the sea. Furthermore we find in these data almost no
instance in which we are satisfied with the identification of a Polynesian
concrete with a Melanesian abstract.

178.
ngiringiri, bright, shining, brilliant, polished.

Samoa: 'i'ila, to shine, to glisten, to glitter. Tonga: kikila, to
shine, to glare; ngingila, bright, shining. Futuna: kikila,
bright, shining; ngingila, brilliant, resplendent. Niue: kikila,
brightness, to shine. Uvea : ngingila, bright, shining, brilliant,
fiery, to beam ; pakila, a ray, a beam.
Malay: gilang, to shine, to glitter, to dazzle, bright, brilliant.
Arabic: gala', galiyy' ', clear, bright, shining, polished.
The concurrence of ng and k forms in Tonga, Futuna and Uvea, while
Samoa and Niue have but the k form, seems to point to two channels by
which the word reached these several parts of Nuclear Polynesia. Of these
the migration which reached Samoa- Niue seems to present an earlier stage
of the root, if the Malay identification prove to be as good as it seems. Then
Efate-Uvea-Futuna-Tonga are finger-posts of a later migration overlaid
upon the earlier. We shall again and again notice evidences of two waves
in the Proto-Samoan migration stream. The consociation of Samoa and
Niue is supportable by Niue tradition.

179.
kabwe, a small basket.

Samoa: 'apu, a cup or dish made of a leaf. Tonga: kabu, the
banana leaf so folded as to hold water; kabu, the banana leaf
tied at both ends to hold water. Mangareva : kapu, a cup, a
leaf dish. Rarotonga: kapu, a cup. Marquesas: kapu,

a handful; curved, rounded. Tahiti: apu, the shell of a nut
or gourd. Hawaii : apu, a dish or cup ; aapu, a concave vessel.
All the Polynesian words are associated with the curving hollow
of the hand, Maori: kapu, id.

Malay: kabok, goblet. Malagasy: kapoaka, a cup, a goblet.

Syriac : kapo', a goblet. Hebrew : kaf, kap, the hollow of the hand ;
plural a hollow vessel, pan or bowl; kafaf, to bend, to curve.

Cf. kaf 96.

180.

las, lasi, big, large, great, sufficient, to suffice.

Samoa: lasi, many. Futuna: lasi, many, great. Tonga: lahi,
large, many, abundant. Niue : lahi, great. Uvea : lahi, many,
great. Maori, Tahiti : rahi, many, great. Manahiki: rahi, great.
Rarotonga: rai, large.

Saparua: ilahil, great. Awaiya: ildhe, id.

Arabic : 'arus'a, to be wide, large.

DATA AND NOTES. 247

There should be a limit to the confidence with which we are to follow
our author. Our limit is reached at this word "suffice." For it is thus
that Dr. Macdonald proceeds: "to meet, i. e., to suffice, be sufficient for, as
nafinanga i lasingita, the ford is sufficient for (meets) us and you, tilasi, id.,
also to meet, come upon, come across (a person) i tilasinauii nabua, he met
us — them on the way."

So far as las, last, means great the Polynesian identification is perfect.

The Indonesian suggestions are derived from Turner; they are presented
as resemblances; there seems no other relation.

soko, masoko, true, exact, to the point.

Samoa: sa'o, sdsa'o, correct, right. Futuna: sako, right, correct.

Tonga: sao, pleasing, agreeing with what is right and just.

Niue: hako, straight; hakohako, perfect, upright, just.
Malay: sung"nh, true.
Arabic: sadaka, sadk', true. Hebrew: sadak, to be straight, right,

just.

The extinction in Tongan of Proto-Samoan k is so unusual as to cause
the thought that perhaps Tonga sao is really Samoa sao perfect, without
fault or blemish, and not properly to be identified with sako.

182.

sulu, a torch; sulu e, to scorch with the flame, to illuminate with a torch.

Samoa: sulu, a torch; to light by a torch; sulusulu, to carry a
torch; susulu, to shine (used of the heavenly bodies and of
fire). Futuna: susulu, the brightness of the moon. Tonga:
huluaki, huluia, huluhulu, to light, to enlighten; fakahuhulu,
to shine; tuhulu, a torch or flambeau, to light with a torch.
Niue : hulu , a torch ; huhulu, to shine (as the moon) . Maori :
hum, the glow of the sun before rising, the glow of fire.

Baki: yulu, a torch. Motu: hururu, id.

Java : stduh, a torch.

Arabic: s'a'ala, to light, kindle a fire, torch; s'u'ulu, flame of fire;
mas'iaV, torch.

The Efate gives us the torch sense ; this runs through Samoa, Tonga, Niue,
Baki, Motu, Java. In Polynesia we encounter the sense of shining, which
may be taken to mean that the torch is the shining thing. The abstract
sense is found with the torch name conjointly in Samoa, Tonga and Niue,
and exclusive of the torch in Futuna and Maori. The Samoan susulu, which
expressly points out that fire and the heavenly bodies shine in the same
word, is the link which joins the two significations, and the Maori would be
exactly as valid if it but included a sulu torch. The celestial shining in
Samoa and Tonga is general, in Niue and Futuna is moonshine, in Maori is
the sun in the dawn glow. Of the Paumotu huru color, height, figure,
shape, only the first sense can be in the least referable to sulu to shine, and
then only to dismiss it.

The Motu identification will pass muster. If the Baki is to stand it
introduces a new mutation, s-y, one which rests upon this single instance ;

248 THE POLYNESIAN WANDERINGS.

and our Baki material is so scanty, this being the only case involving s,

that we can neither prove nor dispute it.

The Java suluh seems good, for in cases where form and sense are in

perfect accord we need not exact corroboration from other Indonesian

sources.

183.

tabwa n, tauba na, side, shore.

Samoa: tapa, the uncolored border of a bast cloth. Tonga: tapa,
the border or edge of bast cloth or of anything; kapa, the
corners and edges of anything. Niue: tapa, side. Rapanui:
tapa, edge, border, fringe, cloth, clothing. Maori: tapa, the
margin, the edge, the brim of a vessel. Mangareva: tapa,
bast cloth, the border of cloth. Marquesas: tapa, bast cloth.
Tahiti: tape, a fragment of cloth; tapemoana, the edge of the
deep water. Hawaii: kapa, bast cloth, bank, shore, side.

Malay: tapi, edge, border.

Arabic: taff , side, shore.

The sense of side, without particular restriction, is found in Efate, Niue,
Maori. The specialized sense of the border of the bast cloth, which is
absent from Efate, is found to the exclusion of the general sense in Samoa
and Mangareva, and inclusive of the general sense in Tonga. In Mangareva
the word means not only the border but the cloth, and in the Marquesas
and Hawaii the border idea has vanished and the word is applied to the
cloth in general, while in Tahiti it designates any fragment of the cloth.
The shore sense takes a long leap from Efate to Tahiti and Hawaii.

184.

taii-ni, tao-ni, to cook, to bake in the oven; tao, leaves for cooking which
are put into the oven along with the food to be cooked.

Samoa: tao, to bake; taofono, taona'i, to bake food the day before
it is used ; tan, the leaves used to cover an oven. Tonga : tao,
to cook food in an oven, to bake. Futuna: tab, to put in an
oven, to cook. Niue: tao, to bake. Uvea: tao, to cook, to
bake. Maori, Rapanui : tao, to bake or cook in a native oven,
properly to steam, to boil with steam. Tahiti: tao, the rocks
and leaves with which a pig is covered when cooking; baked,
boiled, cooked. Marquesas, Mangareva. Mangaia, Tongarewa :
tao, to bake in an oven.

Malagasy: tatao, the rice, milk, and honey cooked at the annual
feast.

Arabic: taha, tahw', to cook.

The Proto-Samoan stem is taon.

The word refers to the specific manner of cookery which involves the
pit oven. The suggestion in the Maori, therefore, does not mean a different
method ; it is but an attempt more precisely to describe the kitchen method,
a very tasty cookery, be it said. The suggestion of boiling is found only
in Tahiti, yet in his dictionary Bishop Jaussen does not record it under
the word bouillir; boiling was little known to the Polynesians before the
European introduction of pottery and other fire-resisting utensils. The

DATA AND NOTES. 249

sense of the noun tad is found in Tahiti, and in Samoa where it is distin-
guished from the verb by the form tau, and this corresponds to taunt, which
nowhere else appears in Polynesia.

The Malagasy affiliation is suggested by Tregear with the apologetic
note cf. It does not seem good, for rice and milk with honey scarcely lend
themselves to this cookery. The failure to recognize Polynesian kitchen
conditions in regard of rice once mislaid some charity of excellent intention.
The Apia hurricane of 1889, in addition to the destruction of two fleets, left
the Samoans on the verge of starvation by the uprooting of their planta-
tions and the snapping off of the crowns of coconut trees. As soon as their
plight became known in Australia the warm-hearted Colonials despatched
a cargo of rice to keep the poor islanders alive. Unfortunately the Samoans
had neither stoves nor pots, they could not bake rice in a pit oven, and they
were as badly off as ever until food supplies better adapted to their con-
ditions reached them.

185.
uta na, uta i, to pay for, repay, give in payment for.

Niue : uta, to pay, to render. Maori : utu, an equivalent, a recom-
pense, the price paid, to pay for, to compensate. Marquesas:
utu, wages. Tahiti: utua, compensation, reward, wages.

Hawaii: uku, to pay, to remunerate, wages.
Malay: utang, a debt. Tagalog, Visayas: utang, id.
Arabic: 'ada', to pay for, repay.

The Niue identification is a satisfactory showing that the word has made
an entrance into Nuclear Polynesia. The utu forms are all in the Tongafiti
migration, and that is so much later as to have allowed the word to change,
or it may have been a dialectic variant ab initio. These words are in accord
among themselves and vary from Efate only in form and that upon the weak
point of the unaccented final vowel. They are, therefore, acceptable.

The Malayan identifications are imperfectly satisfactory. Polynesia has
the sense of something that is paid to a person who has the right to
receive it; Indonesia has the sense of what a person has to pay out. The
distinction is one that would have found Wilkins Micawber, Esq., at his
best, to the rest of us a tragedy. Yet in form the Indonesian words agree
in a terminal consonant, which is also in Efate utana, if I am correct in thus
reading the word.

EFATE-VITI-MELANESIAX-POLYNESIAN-SEMITIC.
186.
afa, afafa, ofa, ofaofa, to swim.

vSamoa, Futuna : opeope, to float.
Viti : nawa, to float.
Epi: viava, mia, to swim.

Arabic: 'ama, to float, to swim (said of a man), to go (of a camel),
to dispose in sheaves or bundles ; 'amat, a bundle, float or raft
for carrying things across water.

If the following were but an exercise of Dr. Macdonald's reasoning I
should have left it in his volume where the curious might find it, yet on

250 THE POLYNESIAN WANDERINGS.

account of what seems a definite and positive ascription to a "native" I
must give it place :

The first meaning (to swim) seems not connected with the second (to carry), to a
European, but a native connects them thus: a man afa natas, swims or floats on the sea,
the sea afa natamole bears or carries the man; so a man afa ki nakasu, swims holding a
floating stick, but if he gets on to the stick and lets it float him ashore the stick is said
to afa i carry him. The sea or the stick carries him thus, hence afa, v. t., denotes carry
a man on one's back, then to carry anything on the back: and as a man so carried clasps
with his arms the carrier round the chest, the head of an axe is said to afa its handle,
and as one carrying a basket on his back holds the string of it over his shoulder, so a
man drawing a log by a string thus over his shoulder is said to afa it, and a tug steamer
is said to afa or tow a ship. A dog afa a piece of meat, carrying it off firmly held by
its teeth, and a man afa a pipe or a twig, i. e., carries it held by his teeth. A messenger
afa, carries his message, a horse its rider, and a warrior afa, carries, i. e., leads his troop;
also a person afa narongitesan, bears a disease or infirmity or trouble. In the Arabic
word there is the idea of connection together (as things in a bundle). In afai, carry
him as a floating stick carries a man in the water, or a horse carries him on land, the
transitive preposition i gives the verb its transitive force, make to swim, to go, i. e., carry.

This identification seems very doubtful at every point. In the matter
of signification, floating is by the islanders quite distinguished from swim-
ming. Through of a the Efate afa approximates ope in form, but the sense
difficulty remains. Efate and Samoa are the forms which lack an initial con-
sonant. Between Viti and Epi there is sense dissimilarity, but there is a
certain formal resemblance except for the initial consonant. It might be
argued that the fact that there is such dissimilarity shows that this con-
sonant is in a state of flux and therefore the more readily tends to vanish.
The mutation is not recognized in any of our Epi material. Regarded as
n-m mutation, we have but three instances of its occurrence in our data,
and that no nearer than Marina (312-3, 317, 324). If regarded as m-n
mutation, our only support will lie in the equally distant Nggao and New
Georgia in the single word 351.

The Semitic requires all Dr. Macdonald's reasoning above cited, and
even then wholly fails of giving satisfaction.

187.
aran, oran, arain, oraone, on, uen (in on and uen the radical r is changed
to n), sand.
Samoa, Fakaafo, Tonga, Futuna, Niue, Uvea, Nukuoro, Aniwa,
Fotuna, Maori, Tahiti, Hawaii, Rarotonga, Manahiki, Mar-
quesas : one, oneone, sand. Nuguria : one, sand ; oneone, beach.
Mangareva: one, the soil; onepatapata, sand. Rapanui: oone,
sand, gravel, clay, dirt, filth. Vate: ngone, sand.
Viti: Oneata, the name of an island.

Mota, Santo, Malo, Keapara, Galoma, Saa: one, sand. Leon: leon,

on the sand. Baki: iono, sand. Bierian: eniono, id. Santo

Wulua: parono, id. Malekula: dambanaiin, id. Motu, Sinau-

goro, Hula, Rubi: hone, id. Kiriwina: kana-kenua. Boniki:

gonugonu. Oiun: ganas.

Arabic: horr', horron, sand; from harm, to be hot.

Through inability to use his critical apparatus Dr. Macdonald has solved

his quite unnecessary difficulty by his note as to the change of radical r-n.

A different arrangement of the Efate material will develop the true radical

DATA AND NOTES. 251

quite simply, on, uen for the simple stem; oraone, arain, aran, oran for
the composite. The stem is one and the Viti, Polynesian, and much of
the Melanesian identification is exact.

The form oraone shows that one is compounded with an element ora, and
this may be the same as ara. In these compounds the one member through
in is degraded into a mere n. This ora or ara may have somewhat to do
with the per of Santo Wulua perono. When unsupported one loses its final
vowel in on, and in uen seems to undergo such vowel shift as is found in
arain.

In Melanesia, so far as is not included in the general identification, we find
the loss of the final vowel in le-on, and probably in Malekula dambana-un
we have the same loss together with vowel shift. The form ono is found
in composites in Baki and Bierian and in Santo Wulua.

Having thus removed the ara from the Efate words there is left to the
Arabic identification not even a resemblance.

188.
aso, to burn, to be scorched.

Samoa: 'a'asa, glowing hot. Tonga, Uvea: kakaha, hot, fiery,
painful. Futuna: kaka, fiery, reddened by fire. Niue:
kaka, hot, red-hot.
Viti : nggesa, burnt or scorched in cooking.
Aneityum: acas, cas, to burn, hot, burning; ecescas, burnt.
Arabic: wakada, to burn, to be kindled. Hebrew: yakad, id.
Syriac: ikad, id.
The Proto-Samoan stem is kas.

The vanishing of radical k in Efate finds support in mataku (258) to fear
Efat£ mitaku, where also we have mitan. This granted we have an excellent
identification in sense and form from Melanesia through Viti to Nuclear
Polynesia. The Niue and Futuna kaka are the reduplication of the root
after abrasion of its final consonant. One of the Aneityum forms points
the way very neatly to the Viti vowel change.

189.
bwalo, to be empty, vain, null and void, to no purpose or effect.

Samoa: vale, inactive, needless, worthless. Tonga: vale, in vain,
in ukiuki-vale, to inquire in vain. Hawaii: wale, gratuitously,
idly, without reward.
Viti: wale, uselessly, idly, only, for nothing, gratis.
Mota: tuwale, only.

Arabic: batala, but/, bot/, to no end, in vain, for nothing, idle.
Hebrew: batal, to be empty, vacant, idle. Ethiopic: .batala,
to be empty, vain.

190.
bebe, a butterfly.

Samoa: pepe, a butterfly, a moth, to flutter about. Nukuoro,
Futuna, Niue, Uvea, Fotuna, Nuguria, Tahiti, Marquesas : pepe,
a butterfly. Maori : pepe, a grub, a moth ; pepepepe, a butterfly ;
pepeatua, a species of butterfly. Tonga : bebe, a butterfly.
Viti: mbembe, a butterfly. Rotuma: pep, id.

252 THE POLYNESIAN WANDERINGS.

Arag, Ulawa, Saa, Fagani, Bululaha, Hula, Keapara, Galoma : pepe,
butterfly. Epi: lepepe, id. Moanus: ndraplpi, id. New
Georgia: pepele, id. Omba, Maewo, Wango, Vaturanga, Buka,
Baravon, New Britain, Solomon Islands, Rubi, Suau, Sariba:
bebe, id. Laur: bdbd, id. Mugula, Tubetube, Tagula,

Murua: bebi, id. Savo: bebeula, id. Motu, Sinaugoro:

kaubebe, id. Panaieti, Misima: bebebi, id. Pokau: ebebelo, id.
Roro: peropero, id. Nada, Kiriwina: beba, id. Boniki:

bebabeba, id. Mukawa, Raqa: arabebemta, id. Kwagila:
karabibim, id. Taupota, Wedau: bebeu, id. Awalama:

kapeu, id. Tavara: gw/w, id. Dobu: pepega, id. Mekeo:
/e/e, id. Kiviri, Oiun: /e/e&, id. Lakon, Pak, Sasar, Alo
Teqel: pep, id. Lo: pip, id. Norbarbar: peb, id. Volow,
Merlav, Gog, Motlav: beb, id. Lamassa: 6aw, id. Lambell:
'hamb'd, id. Mota: pepe, rupe, id. Nggela: uleulebe, id.
Buka: 6awe, id. Brierly Island: bebi, a moth. Malo: vebe,
a butterfly. Baki: bembe, id. Tanna: paubauuk, id.
Morella: pepeue, butterfly. Amboyna: pepeul, id.
Hebrew : 'w/>, '«6'e6, to flutter.
The identification is so complete that we need no more than simple
inspection until we reach the forms which are either more or less than pepe.
The pep, by abrasion of the final vowel, finds itself in Rotuma, Lakon,
Pak, Sasar, Alo Teqel, Lo, Norbarbar, Volow, Merlav, Gog, Motlav. Still
further abrasion, or what amounts to the same thing, the unduplicated stem
pe, does not, with perhaps a single exception, appear unsupported ; but as a
composition member it is found in Lambell 'hambd, Mota rupe, and Nggela
uleulebe. Our possible exception is Lamassa bam. Laur, Lamassa, and
Lambell we have through a single channel, very German and the most com-
petent piece of ethnography I know. From this we see that bd is close to
be. In Lambell the bd is compounded ; in Lamassa, but a few miles away,
bam may be regarded as but bd nasalized, as is sometimes the Melanesian
case; or the m may be a mutant from p, as we shall discuss in note 207.
Variants of pepe are easily traced in Malo vebe, Baki bembe; and Tanna pau-
bauuk, while remote from the standard, as Tanna lies remote from the free
track of migration, seems struggling to preserve the same stem. In Indo-
nesia the pepe stem is unmistakable in Morella and Amboyna.

We now turn to the pepe composites. Lambell 'hambd we have already
noted. Not by any means identifiable with 7*aw,but associable because at
the northern verge, we note Motu kaubebi and Moanus ndraplpi. There is
a liquid component which so frequently associates itself with pepe as to
attract notice. Its simplest form is le, initial in Epi lepepe, final in New
Georgia pepele. There is a stronger form, yet I incline to regard it identical :
ula final in Savo bebeula, Amboyna pepeul, Morella pepeue; initial as tde in
Nggela uleulebe. This may appear in oriori butterfly in the Lakatoi language.
This consistent reducibility to elemental pe, and Buka bawe is so reduc-
ible as well, removes all possibility of identification with the Hebrew word
offered.

We should not pass unnoticed a group of forms which Dr. Ray has col-
lected on the New Guinea shore of Torres Straits. The meaning is wing,

DATA AND NOTES. 253

but that is not particularly remote from the butterfly signification. At
least the comparison is worth making. Starting with pepe and ending
with ani, these forms show a remarkable process of dilapidation of the
Polynesian stem in the custody of Melanesians or worse, yet there is a
perfect chain from pepe to ani. The form pepe is found in Mugula,Tube-
tube, and Panaieti. The slightly variant pape is found in Dobu, Awalama,
Taupota, Wedau, and Kubiri ; and Tavara apape is a modification of the
same. The simple, unduplicated stem, pe, is recognizable in mape of
Galavi, Boniki, and Mukawa; in mabe of Mugula; and in peapea of Sariba.
Recurring to the pape stem we have no difficulty in following it to Oiun
baben and Kiviri fafen. At this point we have acquired a final n, which
thenceforward dominates the stem. Its simple series runs in this order:
Sinaugoro pane, Mekeo pant, Galoma bane, Uni bani, Keapara vane, Pokau
vani, Motu hani, Rubi ani. More involved forms are, Misima bpeni, Nada
papane, Murua pinpene, Kiriwina pinipanela.

191.
bilikit, to peel.

Samoa : mele'i, to husk coconuts.

Viti: longgata, to peel: meleka, to break off a small piece of food.

Aneityum : milinga, to peel.

Ethiopic : lahasa, to peel.

The Proto-Samoan stem is melek.

We can find this in Samoa, Viti and Aneityum. The Efate m-b mutation
we have already been compelled to recognize in umu (76) oven Efate" uba.
These two instances are very satisfactory, yet there are no other except that
similar m-v mutation in masaki (323) sick Nggela vahagi. The Viti longgata
in which our author sees identity is no more than a partial resemblance.

192.
boro, the coconut leaf; a basket made thereof; the leaf plaited for thatch-
ing houses.

Samoa: pola, a plaited coconut leaf used to inclose the sides of a
house. Tonga: pola, the nut leaf plaited for thatch and other
purposes ; bolobola, a large basket made of the nut leaf. Futuna :
pola, plaited coconut leaves. Niue : pola, a coconut-leaf mat.
Maori: pora, a kind of mat. Tahiti: farepora, a small, neatly
thatched house on a double canoe; haapora, a sort of long
basket. Mangareva : pora, a general name formats. Paumotu :
kaporapora, a mat. Fotuna: borabora, a coconut leaf basket.
Marquesas: poa, coconut leaves. Sikayana: pur a, thatch.

Viti : mbola, a coconut leaf plaited for thatching, a basket.

Mota : pora, a basket. New Georgia : poru, a mat. Bougainville :
polta, sl mat. Aneityum : naburabura, a coarse basket.

Arabic: fara', to split, to rend, to slit.

The identification is perfect in our territory, the only variants from the
pola type being Sikayana pur a and Aneityum burabura, in New Georgia
poru, and the unaccountable injection of t in Bougainville polta. This,
however, may be a syncopation of polata, the pola screens which serve for
Venetian blinds about the islander's house. As usual our author : "bworai

254 THE POLYNESIAN WANDERINGS.

to split open ; bora a basket woven out of the frond of a coconut palm whose
stalk is split asunder." He is looking not only at bworai, hut at the Arabic
he is about to suggest. The slitting of the stalk would appear to the
islander, as to us, the first and least important of the operations whereby
the basket is produced.

193-
kafika, the rose apple.

Samoa: nonufi'afi'a, the Malay apple. Futuna: kafika, a fruit tree.
Niue: kafika, a lofty tree. Fotuna: kafika, the rose apple.
Tahiti: ahid, Malay apple. Maori: kahika, the white pine.
Marquesas: kehika, kehia, ehia, tree names. Hawaii: ohia, a
tree name.
Viti : kavika, Malay apple.

Mota, Lo, Maewo, Arag, Marina: gaviga, the rose apple. Tangoan
Santo: kabika, khabika, id. Santo Wulua: keviga, id. Malo:
avica, id. Merlav, Lakon: gavig, id. Mosin: gevig, id.
Leon: vegig, id. Motlav: na-gveg, id. Norbarbar: geve, id.
Re tan: vege, id. Malekula Pangkumu: havih, id. Pak:
marag, id. Sasar: merag, id. AloTeqel: mereg, id. Tanna:
ni-gauvug, id.
Hebrew: tapuah, an apple. Arabic: toffah, id.
The Proto-Samoan is kafik.

The modern Samoan has reduced this stem by the initial syllable, but in
compensation it has incorporated the name of the custard apple (Morinda
citrifolia). With mere color justification this nonu passes in Samoan over
to varieties of the kafika {Eugenia malaccensis) as nonuui the white and
nonu'ula the red variety. It will, therefore, be not out of place to introduce
a brief record of nonu, all the more since it is one of the very few words in
Polynesia which points at all clearly to a Sanskrit source, nona in that
language being the custard apple.

nonu: Samoa, Nukuoro, Gilberts, British New Guinea, Tonga, Niue,
Futuna, Marquesas (noil).

nono: Tahiti, Mangareva.

nino: Mortlocks, Marshalls, Tagalog, Pampangas.

nunu: Viti.

nona: Sanskrit, Malay.
In reducing the element of the composite which pertains to Eugenia
malaccensis Samoa with its fi'a differs from every language in the Polynesian
system ; yet this vanishing tendency of the first syllable shows itself in the
initial abrasion which gives us Tahiti ahid, Marquesas ehia and Hawaii ohia.
In Melanesia the trisyllabic form is maintained intact in Mota, Lo, Maewo,
Arag, Marina, Tangoan Santo, Santo Wulua, and in Malo with the abrasion
of the initial k as in the extremes of eastern Polynesia. Our next group
includes the dissyllables which now appear as closed stems through abrasion
of the final vowel ; these are Merlav, Lakon, Mosin, Leon, Motlav, Malekula
Pangkumu, and Tanna. We find yet a third group, dissyllables of open
stem through the abrasion of the final consonant of the last preceding group ;
these are Norbarbar and Re tan.

DATA AND NOTES. 255

The change in the initial syllable from a to e, which we find in the remote
Marquesas, is here observed in Santo Wulua, Mosin, Leon, Norbarbar, and
Retan. Metathesis is observed in Leon from Mosin, and in Re tan from
Norbarbar. Retan vege might be fika, after the manner of the loss in
Samoan ; but as Leon vegig must be Mosin gcvig, which is kafik, we prefer
to consider Retan in accord with its neighbor. Motlav na-gveg I have
classed in the foregoing with the dissyllables of closed stem. The initial n
is the shadow of the article, just as ni in Tanna, but Motlav is a very diffi-
cult language to write because the article attracts the nearest vowel of the
stem, in this case gaveg; therefore na is article for this word and in attract-
ing has also subtracted.

Pak, Sasar, and Alo Teqel are a separate and unassociable group.

The Semitic proposed by Dr. Macdonald is clearly irrelevant.

194.
kori, kuri, oria, kuria, a dog, a brave, a warrior.

Samoa: 'nil, a dog. Tonga, Futuna, Niue, Uvea, Fotuna, Aniwa :

kuli, id. Maori, Rarotonga, Manga reva, Paumotu, Sikayana:

kuri, id. Tahiti: un, id.
Viti: kali, a dog.
Baki, Epi, Ambrym, Santa Cruz, Deni: kuli, a dog. Iai, Tanna,

Eromanga, Malekula, Aneityum : kuri, id. Epi : koria, kuliu, id.

Sesake: koriia, id. Malo: vuria, id. Tangoan Santo : vuriu, id.

Bierian: kuliu, id. Santo: wurin, id.
Arabic: gorw', a young dog; gariyy' , brave.

The identifications are satisfactory and the stem shows no variations
until we reach Malo and Santo with the k-v mutation, which is found only
in these two languages and in this word alone. The closely similar k-w
mutation spreads over a wider area of speech, yet also is found in but a
single word. The Santo word wurin derived from Dr. Macdonald's gram-
mar of that language is so close to Tangoan Santo vuriu that I must regard
the final n, otherwise incomprehensible, as a fault of the press.

The Arabic does combine in similar words the Efate significations, yet
not even on that account is it convincing in the absence of confirmatory
evidence to bridge the gap.

195-
koto bolo, a basket (bolo, basket).

Samoa: 'ato, a basket. Tonga, Fotuna, Uvea: kato, a basket.

Niue : kato, a basket, a bag.
Viti: kato, a basket.

Lo: gat, a bag. Malekula: na-cat, a basket. Malo: cete, id.
Tanna: katum, id. Aneityum: in-cat, id.; nefehcat, a large

basket.
Arabic: ka'tat, a basket for carrying dates.
As koto means a basket and bolo means a basket, the use of synonyms
in composition is an interesting example in Melanesia of the principle of
determinant compounds, the existence of which in Samoan I have else-
where established (14 Journal Polynesian Society, 40). Koto is readily
identifiable with kato in Nuclear Polynesia and Melanesia and no forms call

256 THE POLYNESIAN WANDERINGS.

for explanation except the as yet inexplicable m-final in Tanna katum. The
Malo cete is here included. It is probably a member of the kete series, which
it will at least serve to introduce here. The consonant skeleton being the
same in kete and kato, we may venture to look upon them as variants of
very early introduction.

Maori, Rarotonga, Futuna, Marquesas : kete, a basket. Tonga, Uvea,
Viti : kete, the belly. Mangareva : akaketekete, to grow big (said
of adolescent girls). Rapanui: kete, keete, basket, sack.

Nuguria : kete, fishtrap, kiddle, food basket. Samoa : 'ete, basket,
bag. Tahiti, Nukuoro : ete, id. Hawaii : eke, a bag, a small sack.

Mota: gete, a basket. Malo: cete, id. Aneityum: in-ced, in-cet
(in composition), id.

The Aneityum with in-cat {kato) and in-cet {kete), Samoa 'ato and (ete,
establish satisfactorily the transition forms. As covering the sense
passage from basket to Tonga, Uvea, and Viti belly, we may instance that
in English of the Marquis of Queensberry rhetoric the bread-basket is a
parallel case.

196.

koto-fi, kote-fi, to cut, to cut off, to break off.

Samoa : 'oti, to cut, to clip (as the hair) ; 'oto, 'otofia, to pluck one.
here and another there. Tonga: koji, to cut with scissors;
koto, to crop, to shorten shrubs; koso, to cut. Futuna: koti,
to cut; koto, to pluck a leaf from its branch. Niue: kotikoti,
to notch. Maori, Rarotonga, Paumotu: koti, to cut. Mar-
quesas, Mangareva: kokoti, id. Rapanui: kokoti, to cut off.
Uvea: kosi, to clip, to shear. Tahiti: oti, id. Hawaii: oki, id.

Viti: koti, shears; kotiva, to clip or shear; kosova, to cut across.

Lambell: koti, to cut. ~Kmg: toki, id. Lamassa: knti, id. Mota:
got, id. ; igot, a cutter; goso, to stab. Pala : kut, to cut through.

Arabic : kata'a, cut, cut off, separate.

The Proto-Samoan stems are kotif and kotof.

We here assemble three forms. Koti and koto are manifestly associable.
As to the final radical consonant we establish kotif on Viti and Efate, kotof
on Samoa and Efate, and kosof on Viti. Of these forms kotif and kotof are
identical in respect of the complete consonantal skeleton and differ only in
the unaccented and weak vowel ; kotof and kosof are in vowel agreement and
differ only in the t-s mutation, which has been proved a normal Polynesian
change. With such strong concords and with such simple differences I
regard the three as ancient variants of a parent stem.

Let us now examine the distribution of these forms.

kotif: Samoa, Viti, Efate, Tonga, Futuna, Niue, Maori, Rarotonga,
Paumotu, Marquesas, Mangareva, Tahiti, Hawaii — in other words
it belongs equally to the Proto-Samoan and the Tongafiti swarms.

kotof: Samoa, Efate, Tonga, Futuna — purely Proto-Samoan.

kosof: Tonga, Viti, and Uvea an interesting transition form.

Our Melanesian material except in Efate does not include kotof, the Mota
/-stem being omitted from the consideration. In King we have a simple

DATA AND NOTES. 257

metathetic variant. Kotif occurs in Efate, in three of the New Ireland
languages at the eastern Indonesian gateway, and probably here are to be
classed the abraded Mota forms. Kosof is not positively but quite likely
identified in Mota goso.

Tonga alone employs the three forms.

197.
lago, to prop, the wooden pins whose sharpened ends are driven into the
sama (outrigger) of a canoe and whose upper ends crossed hold
and bear up the nakiat (struts).
Samoa : lango, props to rest a canoe on, to raise on supports. Tonga :
lango, a fulcrum, to raise by logs or pieces of wood, blocks of
wood on which anything is raised. Niue: lango, a support.
Maori : rango, the skid or roller over which logs, canoes, and the
like are dragged. Mangareva: rango, floor joists. Rapanui:
rango, bed, scaffold, ladder; rangorango, stool. Rarotonga:
tirango, threshold. Paumotu: tirangorango , joists. Tahiti:
rao, a block or roller under a boat, sleepers under a floor.
Viti: lango, a threshold, pieces of wood on which anything is set.
Mota: lango, to put rollers under; ilango, a roller; taplagolago,
cylindrical. Nggela : tapalagolago (from Mota) , a cart, a wheel.
Merlav: geilang, a roller for a canoe. Santo: lako, props of
a canoe.
Arabic: rakaha, to prop.
The Proto-Samoan stem is langom.

The word runs its course across our area with only the minor ng-g
mutation in Efate and Nggela and partly in Mota, and the equivalent ng-k
mutation in Santo.

198.
mat', ebb, low water.

Samoa: masa, taimasa, low tide; masamasa, to be making ebb.
Futuna: masa, dry, waterless, empty. Nukuoro: masa,

dry. Tonga: mahaifo,mahamaha, mamaha, ebb; maha, empty.
Niue : maha, empty. Uvea : maha, dry. Rapanui : taimaha,
low tide. Nuguria: umasa, low; tai kumaha, lagoon is dry.

Viti: matt, to ebb; matha, empty, dry.

Mota: mamasa, dry; mamasaiga, parched. Nggela: mamaha, dry
(of land). Aneityum: mese, dry; in-mas, ebb tide.

Arabic : mat' a, to macerate and dissolve. Hebrew : masas, to melt,
to flow down, to waste.

We have here two words : one, Efate mat1 and Viti mati, ebb tide, with
no previous history; the other, masa, becomes ebb tide only in a resultant
fashion. Masa has the basic signification of dry; this is the only sense in
Uvea and Nukuoro, in Mota and Nggela ; it is associated with another sense
in Futuna, Viti and Aneityum. The first secondary sense, empty, is the
only signification in Niue, and in association with another sense it occurs
in Viti and Futuna. The third signification is dryness in a restricted sense,
the beach or reef from which the water has receded, that is to say, ebb or
low water. This is the only sense in which the word is used in Samoa ; in
Tonga it is associated with the empty, and in Aneityum with the dry sense.

258 THE POLYNESIAN WANDERINGS.

InViti we find matha of the mas a stem; this renders it quite unlikely that
mati has any relation thereto. If Viti mati fails, then Efate maV is left un-
supported. We are therefore justified in rejecting this identification, so far
as it enters Polynesia; we can only recognize an Efate- Viti identification
to be added to the list of words established in Viti as of Melanesian source.

Of course even without these considerations the proposed Semitic iden-
tification fails utterly on the ground of sense, for it can have no relation
whatever to words of the meaning here involved.

199.
mata, mwata, a snake.

Samoa, Tonga, Futuna, Niue, Uvea : nga ta, a snake. Maori : ngata,
a snail, a slug, a leech, a looper caterpillar. Fotuna: ta-ngata,
a snake.

Viti: ngata, a snake.

Malo, Rubi, Dobu : moata, a snake. Mota, Mugula, Sariba, Misima :
mata, id. Suau, Tavara, Awalama, Taupota, Wedau, Galavi,
Boniki, Mukawa, Raqa, Kiviri, Oiun: mota, id. Santo: mata,
maura, id. Bierian: n'mata, id. Malekula: na-mat, id.

Baki: maro, id. Murua: mateta, id. Nada: moteta, id.

Arabic: 'it' at', 'at'a', a snake.

The parent of these forms is a problem. It can scarcely have been ngata,
for ng is a consonant which in all the Melanesian languages here adduced
is susceptible of ready reproduction as ng or n. Still less can it have been
mata, for in one of the Efate forms, in Malo, and in Mota (Codrington dis-
tinguishes it as mata) we see a positive effort to express not the pure m
which they have and use but something somehow different, a suggestion
of mb. We have no means of determining or of representing a parent
sound which shall class as its children ng-mw-m.

The identifications of this ngata are in other respects distinctly satisfac-
tory. I can not include therewith the Baki maro and its congener Santo
maura. These we shall consider true Melanesian material. The absence
of the word from the Tongafiti migration (except Maori with its modified
sense) is a zoological rather than a linguistic lacuna; it is conditioned by
the fact that east of the strait which parts German from American Samoa
there are no snakes.

The Semitic words mean snake; they possess an a and a t apiece, but
these are insufficient proof of identity.

200.
mauri, to live; maurian, life.

Samoa : maui, a Manu'a salutation, 'ua mdui mat. Tonga, Niue :
moui, alive, life, to live. Futuna, Uvea: mauli, id. Vat6:
mauri, to live. Aniwa: mouri. Nukuoro: mouni, to live.
Fotuna: no-mauri, id.; ta-mauri, life: emauri, to feel well.

Viti: maurimu, a word of blessing used by the priests when people
take a thing to the house of the god ; conjecturally, mauri, life,
and -raw, possessive suffix of the second singular, "thy life," or,
"may thou live." Rotuma: mauri, to live; amauri, to make,
to live; amauringa, savior.

DATA AND NOTES. 259

Sesake, Ulawa, Waima, Kabadi, Galoma, Motu: manri, to live.
Saa: manri, meuri, id.; maurihe, life, safety. Santo: meuri,
to live. Bierian: maidi, id.; ni mauliana, life. Baki: meouli,
to live; meoulian, life. Malo: mauru, life, to live. Tangoan
Santo: nauri, to live. Mota: maur, to live. Malekula:

maur, id.; mauran, life. Ponape: maur, to live. Marshalls:
mour, id. Vaturanga: maumauri, to live; maurisali, life. New
Britain: wamanrpa, to make to live. Tanna: murif, life. Laka-
toi language : makuri, life, alive. Keapara, Hula : maguli, life.
Arabic: 'a^'a, 'a'isn, ma'as", ma'is'', ma'is' at, to live.
Before discussing the stem in its greater extension I would comment on
my identification in Samoa and Viti. When we observe the general use
of mauri, to live, we should expect to find it certainly in Samoa and very
likely in Viti. In neither does it appear on the surface; Samoa has ola to
live, a word wholly of the Tongafiti migration, and Viti in mbula to live
employs a word of Melanesian stock which enters Samoan only in the inter-
jection apuld in congratulation upon safe delivery from a sneeze.

The Viti maurimu is susceptible of no other explanation based on exist-
ing Viti material, and the explanation which I have proposed has no objec-
tion other than that mauri is not elsewhere used in Viti as signifying to live ;
therefore I have no hesitation in believing this to be a valid explanation.
It is interesting to observe that in wishing long life in ceremonial phrase
the Viti uses the Samoan word, the Samoan the Viti. Yet we have a par-
allel example in our own speech. Those of us who salute the sneezer are
more than likely to ejaculate Prosit! Dieu vous benisse! Gesundheit! rather
than English. Piety has always been prone to find somewhat esoteric in
the foreign: "Mesopotamia, blessed word!"

In our Samoan authorities, and George Pratt was a marvel of recondite
information, no explanation is offered of the pleasant Manu'a phrase 'ua
maui mai. To give it sense by identifying it with mauli to live needs but
to establish the evanescence of the inner liquid. In my work on Samoan
phonetics ( 1 7 Journal of the Polynesian Society, 1 54) I have proved not
only that this is a Polynesian change, but that it holds in Samoa in partic-
ular. In this material it needs but the comparison of Futuna with Tonga.
The l-n mutation is peculiarly persistent in Nukuoro.

Thus far we have found the word Nuclear Polynesian and signifying to
live. But it is found in Tongafiti possession and meaning native, indige-
nous rather than foreign. There is no gradation of signification. We are
thus led to discriminate between the two swarms. The Proto-Samoan
swarm left the common center with mauli meaning to live ; the Tongafiti
did not follow until it had popularized ola in the sense to live and had set
mauli into the background of a secondary sense. Meaning native, it is in
the following series : Maori, Tahiti, Mangareva, Paumotu : maori. Hawaii :
maoli. Marquesas: maoi.

201.
nono, ne, noi, to dwell or be beside some one, to abide.

Samoa, Tonga, Futuna : nofo, to sit, to dwell, to live with. Niue,
Uvea: nofo, to sit, to dwell. Fotuna: no-nofo, to dwell, to
remain. Fakaafo, Aniwa, Vate: nofo, to sit. Maori, Tahiti:

260 THE POLYNESIAN WANDERINGS.

noho, to sit, to dwell, to live with. Hawaii, Marquesas,
Paumotu : noho, to sit, to dwell. Mangareva : noho, to remain,
to dwell. Sikayana, Tongarewa : noho, to sit, to stay. Mana-
hiki, Nukuoro :noho, to sit. Bukabuka : noo, to be. Raro tonga :
noo, to sit, to dwell. Moiki: noho, id. Rapanui: noho, to
sit, to dwell, a bench. Nuguria: nofo, to sit: unoho, to dwell;
nahoa, bench.

Viti: no, to lie (of things) ; nbnb, a place to lie on. Rotuma: noh,
to sit, noho, a place.

Motu: noho, to dwell, to stay. Vaturanga: noho (are mate noho,
they are dead, have died: "this is, no doubt, the Maori noho to
sit" — Codrington). Nggao: nokro, to sit. Pala: noh, to lie.

Hebrew: navah, naah, to sit down, to rest, to dwell.

The Polynesian stem is unalterably a dissyllable, the only change which
it exhibits being the mutation /-/i-extinction, the last being found in Raro-
tonga and Bukabuka. In Rotuma noh (i.e., noho under terminal abrasion)
we find a transition form to no in Viti and Efate, thence by reduplication
nono. The principal objection to this chain of forms lies in the fact that
while Viti and Efate agree upon wo-forms Motu and Vaturanga, at a greater
backward distance from Polynesia, have noho. In Nggao f-kr gives us an
anomalous mutation, of which this is the sole instance, for in all our colla-
tion of /-variants we find nothing approximating this wholly irregular kr.
The passage from ^-sense to 6e-sense seems far greater to us than to
the islanders. They have not arrived at the need of a word to express
being in the abstract. The occurrence of this signification in Vaturanga
far to the west balances the extremeness of the easting of Bukabuka, in
which it is again found.

202.
rakum, rakoma, a crab.

Samoa : 'ama'ama, a crab (Grapsidae) found in the rocks. Tonga :
kamakama, a rock crab. Niue : kamakama, a crab. Futuna :
kamakama, a little black crab that lives in chinks of the rocks.
Viti : nggumunggumu, a kind of crab.
Epi : lakum, a crab.
Arabic : h'umh'um', a crab.

We lack data on which to establish through transition forms the identity
which seems probable between the kama of Nuclear Polynesia and the Efate
and Epi ra-kum.

203.

sar i, sari, to saw ; seri, to cut with a sawing motion ; sara, a saw.

Samoa : sala, to lop, to cut off ; sele, a bamboo knife, to cut or clip
(as the hair) ; selei, to cut, to shear, to slash. Futuna, Uvea :
sele, a knife, to cut off, to cut. Nukuoro : selesele, to carve, to
engrave. Fotuna: no-sere, to cut; seria, a saw. Tonga:
hele, a knife, to cut, to lacerate; hehele, to cut. Niue: hele,
hehele, to cut up, to reap, to castrate. Hawaii: mahele, to
divide, to cut in pieces. Maori : here, a spear for killing birds.
Paumotu : kohere, to cleave, to split.

DATA AND NOTES. 261

Viti: sele, a bamboo knife; seleva, to cut with a knife, to castrate.

Nggela, Belaga: sari, to cut. Malo: sarosaro, to saw; isaro, a saw.
Mota : sal, to cut with a drawing motion ; gasal, a knife ; sir, to
shave. Gog: gasal, a knife. Maewo: siri, to shave; siriva,
to shave off something. Pak, Leon : sir, to shear. Sesake :
soro, a saw; soroa, to saw. Pala: sele, a bush knife.

Hebrew: nas'ar, to saw. Aramaic: nsar, id.

Throughout the data assembled in this item there is manifest a tendency
to particularize the manner of cutting. Thus Efate, Fotuna, Malo, Sesake,
all specify the saw ; Samoa, Pak, Leon have the shearing and clipping sense ;
Mota and Maewo shave ; many of these languages indicate the knife ; Nuku-
oro suggests the burin. From this presentation we might infer an equip-
ment of saws, scissors, razors, lancets and all manner of edged tools. Far
otherwise is the fact. The silex-edged bamboo splinter, the lip of a shell
ground sharp, the tooth of a shark — these are the cutting tools of the Pacific
islands, east as well as west. The definition is at fault which gives the sense
of saw to any word of a people which has no tool that cuts by notches in a
blade, of shear to men who know no scissors. It would be idle to seek to
differentiate these words by the tool employed, for a rude knife is all there
is. The most that we may venture upon is to segregate the data by the
former vowel.

E. sele: Samoa, Futuna, Uvea, Nukuoro, Fotuna, Tonga, Niue, Hawaii,
Maori, Paumotu, Viti — altogether Polynesian.
seli: Fotuna, Efate — Melanesian, and wholly so if the Fotuna form
be considered due to neighboring New Hebrides influence.
I. sili: Mota, Maewo, Pak, Leon — all northern New Hebrides.
A. sala: Samoa, Efate. salo: Malo.

sali: Efate, Nggela, Belaga. sal: Efate, Mota, Gog.
0. solo: Sesake.

Viti and Maewo seem to indicate a root closed in v which does not else-
where appear, which had been quite lost at the time when Samoa erected
the verb selei upon the noun sele.

204.
sik e, sek e, saki, to raise; sike-ti, to grasp with tongs.

Samoa: si'i, to lift. Futuna: siki, to lift, to raise, to remove.
Uvea: fakasiisii, to raise; hiki, to lift up. Tonga: hiki, to lift,
to remove. Niue: hiki, to land, as fish from a canoe. Maori:
hiki, to lift up, to carry, to nurse. Mangareva : hiki, to hold
a child in the arms or on the knees. Paumotu : hiki, to fondle.
Mangaia: iki, to nurse a child in the arms. Marquesas: hiki,
hii, to nurse a child. Tahiti: hii, to nurse, to dandle, to take
a child in the arms. Hawaii : hii, to lift up, to carry, to nurse.
Viti: sikita, to raise, to lift up.
Aneityum : ahieng, to drag, to draw up.
Hebrew : hazak, to hold fast, to take hold of, to seize.
The Proto-Samoan stem is sikit.

The Efate siketi is associable by reason of its form, but the sense is too
remote and too particular for a satisfactory identification.

262 THE POLYNESIAN WANDERINGS.

In Polynesia are a primary and a secondary sense. The sense of to raise
is exclusively Nuclear Polynesian including Efate. In the Tongafiti migra-
tion the sense is exclusively to nurse a child, a particular instance of lifting
up. I have been willing to mark the sharpness of this division by using in
the foregoing sentence the word exclusively, yet with the full knowledge
that Hawaii and Maori seem to contradict. I have already pointed out the
evidence of single words showing a direct migration from Samoa to Hawaii
by the north and to New Zealand by the south. This combination of the
two sharply marked senses in Hawaii and Maori seems to me to be associable
therewith.

Mr. Tregear's suggestion that Aneityum ahieng shows kinship with this
stem does not commend itself to me.

205.

sila i, sela i, sol i, to rub, as to rub oneself with oil.

Samoa : soloi, a towel, to wipe ; solo, a towel, to wipe as after bathing.
Futuna: solo, to wipe. Tonga: holo, a towel, to wipe, to dry;
holoi, to wipe, to dry off, to rub off. Niue : holoholo, a hand-
kerchief, towel, napkin, to wash the hands or body. Uvea :
holo, holoi, holoholoi, to rub, to wipe, to clean. Hawaii: holoi,
to wash, to wipe, to brush, to cleanse. Rapanui : horoi, to clean,
to wipe out ; horohoro, to brush. Maori, Tahiti : horoi, to wash.
Mangareva: horoi, oroi, to wipe, to wash. Mangaia: oroi,
to wash. Rarotonga: orei, to wipe, to wash. Marquesas:
hooi, id.

Viti : solota, to rub, to dry oneself after bathing.

Motu: huria, to wash, to scrub. Mota: sarag, to wipe; sarav, to
rub; saravag, to brush. Aneityum: ruhoi, to rub.

Arabic: "asala, "usul', to wash.

This solo stem is not clearly distinguishable from olo (331) to rub. The
motion of the hands being the same, solo may be a particular instance of
olo. For this reason it has not seemed desirable to attempt rigid accuracy
in assigning the Melanesian examples.

In general we may note a distinction in sense between Nuclear Polynesia
and the Tongafiti eastern extension, Nuclear Polynesia referring to the
drying after bathing, eastern Polynesia to the bathing. The only exception
to this in Nuclear Polynesia is Niue which has both senses. Maori, Tahiti
and Mangaia are restricted to the washing sense ; Hawaii, Mangareva, the
Marquesas, and Rarotonga have both. In Melanesia Motu has the Tonga-
fiti washing sense, Mota and Aneityum accord with Nuclear Polynesia.

Aneityum ruhoi is metathetic.

206.

siua, siuo, si wo, suwa, sua, to descend.

The following words signify down, to be down, to come down :

Tonga, Niue: hifo. Samoa, Futuna, Uvea, Fotuna, Nuguria,
Aniwa: ifo. Maori, Tahiti, Hawaii, Marquesas, Rapanui,

DATA AND NOTES. 263

Mangareva, Moiki: iho. Paumotu: ihoiho. Nukuoro: hakaiho,
to go down.
Viti: sivo, debased, dethroned, put out of office, a tropical sense
germane to that in Samoa ifonga. Rotuma: sio, down.

These words signify down :

Vaturanga: sivo. Tangoan Santo : sibo. Omba: hivo. Nguna,
Mota, Santo, Sesake: siwo. Malo: siuo. Merlav, Maewo,
Sesake : suwo. Gog:suw,sug. Aneityum : asuol, suko. Lakon:
hew. Eromanga : sep. Roro : tivo. Pokau : divo. Mekeo : kipo.
Wedau: ipu. Mukawa: sipu. Tavara: ho pit. Wedau: opu.
Galoma: ribo. Keapara: rigo. Motu: diho. Tubetube: siio.
Raro: zi, azi.

Arabic: safala, suful', sifl', to be low, to descend.

With unchanging vowels the Polynesian identifications move in a regular
course of mutation of the two consonants, ^-//-extinction, f(v) -//-extinction.
We observe in this a tendency, not without exceptions, to maintain the
labial strong when the lingual is preserved, and to obscure the labial in
the aspiration of its series where the lingual has become extinct.

The Melanesian affiliates call for more detailed examination.

The sibilant is retained by all except that in Lakon it merges in the
aspiration, and here the initial aspirate is explosive, as if hv or vh; Omba,
too, employs the aspirate. We do not find a single instance in which the
labial is of the same value as in Samoa. In two languages it has been
strengthened to a mute, in Tangoan Santo to the sonant, in Eromanga to
the surd. In these languages the surd spirant is replaced by the sonant,
v: Vaturanga, Omba. In the greater number the reduction has progressed
as far as the semivowel of this series, w, or to the nearest vowel, u, the
distinction not being made clear in the absence of the scientific alphabet :
Efate, Nguna, Mota, Santo, Sesake, Merlav, Maewo, Gog, Lakon, Malo. In
two instances there occurs the rare leap from labial to palatal, to k in
Aneityum and to g in Gog. The former vowel remains at i in about half
of Melanesia. It changes to u in Efate, Merlav, Maewo, Sesake, Gog and
Aneityum, Efate and Sesake having the i-form as well. It changes to e
in Lakon and Eromanga. The latter vowel remains fixed at o except in
Efate and the languages which admit terminal abrasion, Gog, Lakon, and
Eromanga. The Aneityum asuol I include because of the sense, but I
can not identify the beginning and the ending, yet embraced therein is an
element which, if independent, could be referred to sifo.

207.

tabu, tab, bakatabtabu, to be forbidden, prohibited, sacred.

Tonga: tabu, forbidden, consecrated, sacred. Samoa,' Futuna,
Niue, Uvea, Maori, Tahiti, Rarotonga, Rapanui, Marquesas,
Mangareva, Fotuna, Aniwa: tapu, id. Paumotu: tapu, an
oath, to swear. Nukuoro : tapu, prohibition. Hawaii : kapu,
forbidden, consecrated, sacred.

Viti: tambu, unlawful, forbidden, sacred.

264 THE POLYNESIAN WANDERINGS.

Buka, Baravon, Nggela, Belaga: tambu, sacred, holy, prohibited.
Merlav: vatambu, to make holy. Omba: tambetambe, to worship.
Dukeof York : tabu (watabunabulu) , sacred. Dobu, New Britain :
tabu, sacred, holy, prohibited ; watabu, to hallow. Suau : tabuna,
sacred. Pokau: kabukabu, id. Mekeo: afu, id. Motu: koau-
aku, id. Aneityum : in-tap, a sacred place ; itap, sacred, holy, for-
bidden ; imiitap, to hallow, to make sacred. Mota : tapu, sacred.
Malo: sab, saburu, id. Tanna: asim, id. Bierian: ham, id.
Motu: tabu, a very important feast, a species of mythical beings.
Arabic : dabba, dabbu, to prohibit.
An interest exterior to our present study pertains to this word in the
fact that it is one of the two words, tattoo the other, which the English
has borrowed from the Polynesian.

In scarcely altered skeleton of vowels and consonants tapu occurs in
Polynesia and in Melanesia. In the western island world it undergoes less
variation than almost any of the Polynesian loan material. In Omba the
final u becomes e, in Efate and Aneityum it undergoes a terminal abrasion.
We need give particular attention only to the following forms.
Malo sab (of a man) saburu (of a woman). This needs but the establish-
ment of the t-s mutation to prove its identity. I have already (17 Journal
of the Polynesian Society, 212) proved it to exist as between Polynesia and
Viti; in the material now under examination it clearly appears in talinga
(350) ear Marina salinga, buto (247) navel Buka vuso, mate (318) to die
Aneityum mas.

Next upon this follows Tanna asim . Taking the s as now accounted for,
we have to consider a p-m mutation. In note 190 I have found a case
which is susceptible of this interpretation, and taking that with this there
seems to me a probability of this change. We have abundant proof of a
nasal reinforcement of p and b, mp and mb. It is quite possible that when
such a nasalized consonant was transmitted at second hand to others to
whom the double consonant was harsh they might excise the radical mem-
ber of the composite and in their ignorance of the true form retain the
purely accidental nasal. Bierian ham, then, would derive from Malo sab
in the vowel and from Tanna asim in the p-m mutation, while the change
from s to h is so common as to call for no note.

208.
tafe, tabe, to flow out, to go out.

Samoa, Tonga, Futuna, Niue, Uvea: tafe, to flow, to run. Fotuna:
no -tafe, id. Tahiti, Mangareva: tahe, id. Paumotu: take, a
river. Nukuoro: tahea, to drift along. Maori: whakatahe, to
clear from obstruction, as a watercourse or channel. Hawaii :
kahe, to run, to flow. Marquesas: tahe, to flow, to gush, to
stream, to trickle. Rapanui: tahe, to flow. Nuguria: tahe,
a current.

Viti : ndave, to flow ; ndavendave, the channel in which liquids flow,
or the source of them ; ndaveta, a passage through a reef.

Nggela: t ave, to flow. Motu: atahedid, to overflow. Malekula:
jivjiv, to run (nose). Malo: madividivi, id. Baki: jevi, id.

Aramaic: dub, to flow out. Hebrew: zub, id.

DATA AND NOTES. 265

There is no need to comment upon the identifications within the Pacific
area except in respect of one particular. Viti ndaveta has a form resem-
blance to ndave but the sense does not so clearly hang. I have included
inhere because it seems to suggest a radical tafet, which may appear also
in Motu atahedid. Yet in Viti the passive is ndavena, in which the sug-
gestion of final n is equally worthy of consideration. In Polynesia there is
no evidence to warrant the idea that tafe was ever other than an open root.

209.

tanga, tonga, tronga, rong, a basket, the stomach.

Samoa: tanga, a basket. Futuna: tanga, a sack. Niue, Uvea,
Rapanui, Nukuoro: tanga, a bag. Tonga: tangai, a narrow
bag, a sack. Nuguria: tana-mimi, the bladder.

Viti: tanga, a bag, a pocket, a purse.

Arag: tanga, a bag. Sesake: ndanga, id. Motu: tanga, a bag,
a deep basket. Tangoan Santo : tanga, a basket.

Hebrew: tene', a basket.

210.

tef i, tefi, tetefi, to cut, to circumcise.

Samoa: tefe, to cut, as circumcising. Fotuna: tefe, to circumcise.
Maori: tehe, the glans penis left uncovered by the prepuce as
if circumcised. Tahiti: tehe, to castrate animals, to slit the
prepuce above. Mangareva, Marquesas: tehe, to circumcise,
to castrate. Rapanui : tehe, to split ; tehetehe, a notch. Hawaii :
kahe, to cut or slit longitudinally, to castrate; kaheomaka,
kaheule, to circumcise after the Hawaiian fashion.

Viti: teve, to circumcise.

Mota : teve, to cut with a drawing motion ; teveteve, a knife. Merlav :
tevtev, a knife. Malekula: teve, to circumcise, to cut with a
bamboo knife. Bierian: mdeve, to circumcise. Baki: jivi, id.

Arabic : 'as' aba, to cut.

Little calls for note in the Pacific identification. In Hawaii we observe
the e-a change which does not elsewhere appear. The Bierian mdeve is a
variant of the t-nd characteristic of that Epi dialect. The Baki jivi and
tafe (208) to flow Malekula jivjiv to run at the nose exhibit the t-j mutation
which is rare in Melanesia but is the normal change from Samoa to Tonga
before e and *'.

The word appears in Nuclear Polynesia only in Samoa and Viti, and is
there restricted to circumcision. In its occurrence in eastern Polynesia it
combines with the circumcision sense that of castration. In Samoa it is
defined as "the operation equivalent to circumcision," in Hawaii "to cir-
cumcise after the Hawaiian fashion," in Tahiti "to slit the prepuce above."
The operation is so singular that in almost all the versions of the Bible in
the Pacific the word circumcision has been rendered by peritome, a trans-
literation from the Greek, rather than by the employment of this word
from the vernacular.

266 THE POLYNESIAN WANDERINGS.

The operation is surgically described by Dr. Kramer in "Samoa," ii, 6r :

Die Beschneidung, 'o le tefenga (tefe beschneiden) , bei der es sich nicht um Circum-
cision wie die Englander sagen, handelt, sondern um einfache Spaltung der Vorhaut,
also Einschneidung am oberen Rande, wie allgemein in Indonesien iiblich, wird so
ausgefiihrt,dassmaneinenSpatelunter die Vorhaut schiebt und diese durch einen Schlag
mit einem scharfen Gegenstand als Haifischzahn, Muschel, Bambusmesser, neuerdings
natiirlich mit Eisenmesser durchtrennt. Sie wird bei den samoanischen Jiinglingen,
ahnlich wie bei den Mohammedanern, bei Eintritt der Mannbarkeit, stets zwischen dem
7. und 15. Lebensjahre ausgefiihrt. Religiose Gebrauche wie bei andern Volkern, z. B.
auf Fidji, wo die Operation auf den Nangaplatzen geschah und wo richtige Circumcision
geiibt zu werden scheint, scheinen auf Samoa nie mit der Operation, die meist von
einem darin Erfahrenen ausgeiibt wurde, verbunden gewesen zu sein. Das leitende
Motiv scheint fur Samoa nur in der Reinlichkeit zu liegen, indem gesagt wird, dass kein
samoanisches Madchen mit einem unbeschnittenen Tiingling schlafen wurde. Deshalb
nennen die Samoaner die Bliite der Amorphophallus-Pflanze (teve) welche dem mann-
lichen Gliede nicht unahnlich sieht und einen fotiden Geruch verbreitet, wie ich mich
selbst zu iiberzeugen Gelegenheit hatte, tajao, und ebenso nennen sie einen unbe-
schnittenen Jiinglingspenis. DerArztweiss, dassdiesnichtohneGrund geschieht. Wenn
mann nun behauptet, dass die Beschneidung bei den Juden an Stelle unserer Taufe
(am 8. Tage) ausgeiibt, rein rituell sei, und dass Reinlichkeitsgedanken ferae lagen, so
mag das sekundar so geworden sein, ursprunglich wird man aber die bei den oriental-
ischen Volkern ausgeiibte Beschneidung auf den Reinlichkeitsgedanken, der in Samoa
nach langst eingefiihrtem und allgemein ausgebreitetem Christentum heute allein noch
diese Sitte aufrecht zu erhalten im stande ist, zuriickfiihren miissen.

211.

tuk i, tuki, to strike, to beat, to pound.

Tonga: tuki, to strike, to drive, to drub. Futuna: tuki, to beat,
to strike, to hit with a hammer, rock or fist. Uvea: tuki, to
strike, to beat. Niue: tuki, to knock. Nukuoro: tuki,

to beat, to strike, to pound. Maori: tuki, to ram, to butt, to
strike endwise, to beat. Rarotonga: tuki, to strike, to beat.
Rapanui: tukituki, to bray. Marquesas: tuki, to beat poi, to
bruise, to strike, to bra}' with a pestle. Mangareva: tuki, to
pound, to bray with a pestle. Paumotu: tukituki, to hit
against, to strike, to pound. Nuguria: tukituki, to beat; a
breadfruit beater. Fotuna : no-tukia, to strike with the fist.

Samoa: tu'i, to thump, to beat, to pound, to strike with
the fist. Tahiti: tui, to butt, to strike, to pound. Hawaii:
kui, to pound with the end of a thing, to smite, to hammer.

Viti: tuki, to beat or knock with the fist, to hammer.

Belaga : tutui, to pound. Baki : Juki, to strike with the fist. Male-
kula: tice, id. Pala: tuke, to beat.

Hebrew: duk, dakak, to beat, to pound. Arabic: dakka, dakka, id.

There is nothing in these identifications involving any principle with
which we have not already become familiar.

There is, however, an interesting specialization in the sense. It is, of
course, well understood that peoples on the speech plane in which we find
the languages of this study have a large number of distinct terms whereby
to describe an act as performed in a certain manner or through the employ-
ment of a certain implement in cases where the languages of superior culture
make use of a general term for the act limited by an adverbial modifier

DATA AND NOTES. 267

detailing the manner or the implement. Thus, in the terms of the present
instance, we should say "to strike with A, B, C, . . . X, Y, Z," covering
every possible thing with which a blow could be inflicted from the simplest
pugilism through fustigation to the fulmination of the bolt from the blue
ineluctable. But our islander would maintain a series of distinct verbs
from "to strike with A" clean through to the crisscross row if the tale of
his striking machinery should extend so far.

In this instance we shall find that we are dealing with material collected
by just short of a score of observers, each independent of the other, each
under the control of his own personal equation of observation, apprecia-
tion, and ability to reproduce in one language the idiom of another. For
this reason we shall all the more highly regard the evidences of special sense
in tuki. Maori gives us "to strike endwise," Hawaii "to pound with the
end of a thing," the Marquesas and Mangareva the same idea in particular
terms as "to bray with a pestle." Next we find a group "to strike with
the fist," Futuna, Samoa, Viti, Baki, Malekula. From sight of many such
encounters I can aver that Pacific pugilism has none of the niceties of the
straight-arm jab, the left hook to the jaw, and the other fine touches of the
diction of the prize ring. A blow in such combats is delivered by regard-
ing arm and fist as helve and head respectively of a hammer wherewith
to belabor the head and shoulders of the opponent. The motion of such
a blow is exactly that of using the pestle or the hammer or the rock in the
hand. The motion can be traced still farther by the curious in the defini-
tions here assembled.

212.

ulua, to put forth leaves, to grow up (of plants and hair) ; uluulua, to be
full of leaves, to be hairy.
Samoa: uluulu, to be umbrageous (of trees), to be bushy (of the
beard) . Tonga : ulu, thick, bushy (as a dress of leaves) . Niue :
ulu, hair. Nuguria: rauulu, id. Hawaii : ulu, to grow ; uluulu,
to grow thick. Maori: uru, a single hair. Mangaia: uruuru,
coarse hair. Mangareva: uru, feathers, hair on the body.
Viti : vakaulu, having a large head of hair or a wig.

The following words signify hair:

Lo: ul. Deni: ulu. Mota, Maewo: ului. Sesake: ululu. Kpi,

Arag: ilu. Ambrym, Hi, wolu. Marina: vul. Omba:

vulugi. Norbarbar: wulugi. Volow: iligi. Nifilole: lu.

Suau: uru. Tubetube: hulu. Mugula, Sariba: kuru. Nada,

Kiriwina: kulu. Tagula: wuluwulia.
Arabic: lala, 'aluw', to go up; 'ilawat, the head. Hebrew: 'alah,

to go up; 'aleh, leaf, leaves; 'oleh, sprouting forth, growing up.

lulu, hair (of the head, face or body).
The following all mean hair, feather :

Samoa, Uvea, Futuna, Tonga, Niue: julu. Maori, Tahiti : huruhuru.
Mangareva: huru. Rapanui: huhuru. Nuguria: hahuru.
Hawaii: hulu. Marquesas: huu. Rarotonga: uru.
Viti: vidua, hair about the pubes. Rotuma: leav, hair.

268 THE POLYNESIAN WANDERINGS.

In connection with the words here assembled it is quite impossible to
dissociate the two stems ulu and fulu. It is probable that in my theory
of word formation by consonantal coefficients the explanation will be found.

Before proceeding to our task of tracing out the intricacies here presented
I wish to call attention to the existence in Polynesia of yet another word
for hair, lauulu. Where a distinction is made fulu is the hair of the body,
lauulu that of the head inclusive of the beard, yet frequently accompanied
by a specific term for the beard.

Tregear interprets this as lau leaf and ulu head ; such also is the interpre-
tation given by the islanders, the value of their etymologies having been
mentioned in note 169. This is very simple, very obvious. Yet the form
of lauulu used in Tahiti is rouru, and in that language ro does not mean
leaf at all ; nor yet does it in the Viti dialect which employs ro ni vulu for
ndrau ni ulu in the same sense. Furthermore, the vulu in the Viti Levu
form is not ulu head but vulu hair, and ndrau itself means hair as well as
leaf. Thus we have found that hair as ' ' the leaves of the head " is not such
a simple explanation as it appears.

Having already established the nature and employment of determinant
compounds, I recognize in this composite lauulu two words of one identical
sense among others, lau hair and ulu hair; their employment together
determines for the composite the sense of hair beyond any doubt. Cod-
rington (Melanesian Languages, 73) seems to have felt some suspicion
about the leaves of the head explanation, but, the determinant compound
not having come within his knowledge, he was unable to carry on his note
to a satisfactory issue.

We shall now examine the interlacing of these two stems in the area of
their greatest intricacy, Melanesia ; and shall rearrange the material in the
order of a developmental series. It should be noted that such termina-
tions as gi, ge, and i are merely local means to indicate a noun as absolute.

fulu series. ulu series.

vulu: Gog, Malo, Omba, Mosin, ulu: Deni, Malo, Sesake, Mota,

Vuras, Eromanga. Maewo.

wulu: Norbarbar. uli: Merlav.

wolu: Ambrym. ilu: Epi, Arag.

holu: Buka. Hi: Ambrym, Volow, Motlav.

weu: Duke of York, ul: Lo.

houi: Motu. lu: Nifilole.

vul: Marina, Lakon. kalu: New Georgia.

vili: Makura, Bierian, Pak, Sasar,

Alo Teqel.
viji: Baki.

The two series inosculate in vulu-ulu, vili-ili, thus showing that the
intimacy of their interrelation is not fortuitous. The bare simplicity of
Nifilole lu is repeated in the Efate lulu. The Eromanga novlimpu becomes
clearer by indicating the several members of the composite, no-vli-mpu.
Baki viji, so close to its neighbor Bierian vili, clearly establishes an /-/
mutation; we find confirmation in sola (339) path Bugotu hatautu, and
(l-nd) langi (308) sky Buka indengid. Finding, then, in Melanesia this form

DATA AND NOTES. 269

in t we may incline toward a warmer reception of Indonesian /-forms where
the vowel is truer to stem than in Baki. Such are Tidore, Galela : hutu;
Menado: uta; Sanguir: utan; Gah: uka; Matabello: ud; Batumerah: hud;
Wahai : hue. Other Indonesian forms in which a resemblance appears are
these : Mayapo : folo; Massaratty : olofolo; Cajeli : buloni; Baju : buli tokolo;
Malagasy: volo; Bouton: bulwa; Ahtiago: ulujuim; Tobo: ulvu; Salayer:
uhu; Teluti: keulo; Morella: keiule; Liang: kaiola; Caimarian: keori.

Rotuma leav and Tobo ulvu and Ahtiago ulujuim show interesting vari-
eties of metathesis. Assigning position, we mark vulu 1234; then Rotuma
represents 3412, Tobo 4312, and Ahtiago probably 4321, or less likely 2341.

EFATE-MELANESIAN-VITI-POLYNESIAN-MALAY.

213.
banga, bangan i, fanga, to feed, to charge, to fill.

Samoa, Tonga, Uvea: fafanga, to feed. Tonga: fafangai, id.
Futuna, Niue, Paumotu : fangai, id. Tahiti: faaai, id. Rapa-
nui: hangai, id. Hawaii: hanai, id. Maori: whangai, to feed,
to nurture. Mangareva, Rarotonga: angai, to feed, to nourish.
Marquesas: hakai, to feed.
Viti: vdkania, to feed.

Mota: vangan, to feed. Nggela, Bugotu: vanga, food. King:

ivangon, to eat. Lambell: hangdn, id. Lamassa: angdn, id.

Malagasy: mamahana, to feed, to load a gun; ma causative and

fahana.

The Proto-Samoan stem is probably fangan. Yet the final n nowhere

appears in Polynesia, while it is plainly to be seen in Efate, Mota and our

three New Ireland languages.

Dr. Macdonald's identification with Viti vdkania at first seemed to me
wholly superficial. On closer examination it is found to be worse; it
is an attempt to wrest the record. The only place in which the word
appears in the Viti dictionary is not in its alphabetic place, but as a note
under the word kana to eat, of which it is clearly stated to be a causative.
That our author had this entry before his eyes is shown by the fact that he
cites the word as vdkani-a, just as Hazlewood has printed it. He forgets
that he identified it correctly in 46. Since va is the causative prefix and
kana means to eat there can be not the remotest relation with the fangan
stem. The discussion of these two Viti stems will be found under 214
and 46 respectively.

In Polynesia generally the word is transitive, yet there are uses in which
evidence appears of the intransitive as well. Regarding the primary sense
of the word as transitive, we find in its use as signifying the charging of a
gun or the filling of a pipe in Efate no great deviation from an elemental
signification of putting something into the mouth.

214.
baka, faka, causative prefix.

Tonga, Paumotu, Niue, Futuna, Uvea, Fakaafo, Fotuna, Sikayana,
Aniwa: faka. Marquesas, Paumotu, Nukuoro, Rapanui,

Tongarewa: haka. Maori: whaka. Samoa: fa' a. Tahiti,

270 THE POLYNESIAN WANDERINGS.

Uvea, Tonga : faa. Tahiti, Hawaii, Marquesas, Nukuoro : haa.
Moriori: hoko. Hawaii: hoo. Paumotu: fa. Hawaii: ha.
Maori: wha. Hawaii: ho. Rarotonga, Mangareva, Buka-
buka: aka. Mangareva: anga. Rapanui, Paumotu: a.

Viti: vaka, va. Rotuma: faka, jak, a.

vSesake, Kiriwina, Santo, Nguna: vaka. Malo: vaca. Marina:
Nggela, Belaga, Sinaugoro, Omba, Maewo, Mota : vaga. Fagani :
faga. Tangoan Santo : thaka. Keapara: vaha. Ulawa.Wango,
Saa, Ugi : ha' a. Sesake, Nguna : paka. Bierian : baka. Sesake,
Marina, Arag, Merlav, Gog, Lakon, Mota, Motlav, Volow, Lo,
Deni, Vaturanga, Nggela, Bugotu, Motu, Pak, Leon, Vuras,
Mosin, Baki, Kabadi, Hula, Nguna: va. Nifilole, Duke of
York, Raluana, Kabakada, Matupit, Baravon : wa. Aneityum :
ua. Nggao: fa. Roro, Pokau: ba. Mekeo, Panaieti, Nguna:
pa. Motu, Pala : ha. Nengone, Lifu, Motu, Panaieti, Dobu : a.
Motlav, Pak, Leon: ve. Vuras: vi. Norbarbar, AloTeqel: v-.
Savo: au. New Britain: wara. Malekula Pangkumu: vaha.

Sulu: mak, maka. Tagalog, Bicol: mag, pag.

faka: Futuna, Efate, Tonga, Pau- fa: Uvea, Nggao, Paumotu, Efate.

motu, Uvea, Rotuma, Samoa. faa (fa): Tonga.

faga: Fagani. va: Viti, Sesake, etc.

faa: Tahiti, Uvea. wha: Maori.

fak: Rotuma. ba: Efate.

vaka: Nguna, Viti, Sesake, Malo, ve: Motlav, Pak, Leon.

Santo. vi: Vuras.

vaga: Marina, Maewo, Omba, Mota, v — : Norbarbar, Alo Teqel.

Nggela, Belaga. b — : Efate.

vaha: Malekula Pangkumu. f — : Efate.

baka: Efate, Bierian. pa: Nguna.

paka: Nguna, Sesake. ha: Hawaii.

whaka: Maori. ho: Hawaii.

haka: Marquesas, Paumotu, Nuku- wa: Nifilole, Duke of York, Raluana,

oro, Tongarewa. Kabakada, Matupit, Baravon.

thaka: Tangoan Santo. ua: Aneityum.

haa: Hawaii, Ulawa, Tahiti, Marque- a: Rotuma, Nengone, Lifu, Pau-

sas, Wango, Saa, Ugi, Nukuoro. motu.

hoko: Moriori. au: Savo.

hoo: Hawaii. wara: New Britain.
aka: Rarotonga, Mangareva, Bukabuka.
anga: Mangareva.

It is manifest that two forms are here involved and that the sense is
identical. I have given close attention to the examination of those lan-
guages in which the two forms are in use simultaneously, and except for
one slight discrimination I have failed to discover any principle in the selec-
tion of the form which shall be used. It is not euphonic, for the long form
is used before consonants as well as before vowels and the same is equally
true of the short form. The only discrimination which I have satisfied
myself to exist consistently is in Uvea, where faka is causative and fa (faa)
is used of resemblance.

DATA AND NOTES.

271

The stems are faka and fa . Rotuma fak is the only vestige of the transi-
tion form.

To show the devolution I have rearranged the forms in two lists. We
need comment on but few facts. The forms in o seem to have developed
sporadically and, of course, independently in Hawaii at the extreme north-
ern limit of the Tongafiti swarm and in Moriori at the extreme south and
apparently earlier than the Tongafiti movement. Not a vestige of the o
has been retained along the migration tracks. In Savo au and New Britain
wara we find apparently irreducible forms, possibly heterogenetic.

215-

bia, bisa, fia, fisa, how many.
In the same sense —

Tonga, Niue: fiha. Samoa, Futuna, Uvea, Nukuoro, Aniwa,

Nuguria, Sikayana: fia. Maori, Moiki, Tahiti, Marquesas,
Mangareva, Rapanui, Paumotu: hia. Mangaia: ia (eia).

Viti: vitha.

Mota, Maewo: visa. Sesake: pisa. Vaturanga: ngisa. Pala:
hise. Wango: siha. Arag, Ambrym: viha. Nggao: ngiha.
Baravon: aivia. Moanus: tje, id tjt.

Malay: hia, what.

The Proto-Samoan stem is fiha.

Dr. Macdonald disposes of the matter with the declaration "the final
part of bia and blsa, namely a or sa, is the interrogative pronoun." Refer-
ence to these several words in their proper places in the dictionary confirms
his statement — so far as relates to Efate. He offers no explanation for the
former element of such a composite, and I have none to suggest. But:

Language.

How many.

What.

Language.

How many.

What.

Mota

Maewo

Vaturanga.. .

visa
visa
pisa
ngisa

sava

sava

sa

na hua

Wango

Arag

Ambrym. . . .
Nggao

siha
viha
viha
ngiha

e taha

hava, havanau

ha

na no

It is tantalizingly close, yet with such instances as Vaturanga and Nggao
we are debarred from accepting the explanation until we know enough of
the /i-element to enable us to account for the word as a whole.

216.

matua, to be old, mature, large, great, wise.

Maori, Manahiki, Tongarewa: matua, parent. Rarotonga, Buka-
buka: metua, id. Samoa: matua, parent, mature, elder.

Tonga: matua, parents, old people; motua, old, mature, ripe.
Futuna: matua, old, parents, mature, ripe. Niue: matua,

parent; motua, old. Uvea: matua, parents, old, mature.

Rapanui: matua tamaroa, father; matua tamaahine, mother.
Tahiti : -matua, old ; mitua, metia, parent. Marquesas: motua,
father. Mangareva: motua, father; -matua, old. Nuguria,

272 THE POLYNESIAN WANDERINGS.

Sikayana: matua, old. Aniwa: tomatua, to be able. Hawaii,
Paumotu : makua, parent, old, mature. Fotuna: mahtua, old.
Viti: matua, ripe, mature.

Nggela : kukua, ancestors. Laur : imdtuk, ripe. Lambell : makos,
id. Lamassa: imakos, id. Bierian: matua, old; tamatua, old
man. Tanna : matu, ripe. Santo: metu a, ripe. Mota: matua,
ripe, full-grown. Aneityum : meto, ripe. Pala: matua, uncle.
Malagasy: matoa, eldest son or daughter. Malay: mdntuwah, a
parent-in-law ; mentua, mother. Macassar : matowang, father-
in-law.
Three senses are involved herein, to be old, to be ripe, parent. I should
like to see my way to the idea that age is the central idea, but the mate-
rial does not warrant this conclusion — or, in fact, any. The three senses
(assuming the mature of the dictionaries to cover ripe) appear concurrently
in Tonga, Uvea, Futuna, Hawaii, Paumotu. In Samoa matua does not
mean ripe and that language is not included in this category. Concurrence
of the two senses old and parent obtains in Samoa, Niue, Tahiti, Mangareva.
Differentiated forms (matua, motua) are found in Tonga, Niue, Mangareva ;
and in Tahiti matua, mitua, and metua. In Tonga and Niue motua is old
and matua is parent, which is its sole signification in Maori, Manahiki, and
Tongarewa. But in Tahiti and Mangareva the usage is opposite. In the
Marquesas motua is the only form and its only sense is father.

In our Melanesian area Nggela kukua ancestors is included only as sug-
gesting a partial resemblance. Efate is the only language which gives the
stem an extended range of meaning. Bierian shares with it the common
Polynesian signification of old. All the other forms have the sole meaning
of ripe and the languages employ other words for age and parent.

217.
tema na, tama na, father.
In the same sense —

Samoa, Fakaafo: tama. Aniwa: tama. Tonga, Uvea: tamai.
Futuna, Sikayana, Fotuna, Nuguria, Nukuoro: tamana.

Viti: tama, father.

Pala, New Britain : tama. Redscar Bay (N.G.) : tamaa. Aneityum :
etma. Eromanga: temi. Mota: tamai. Nifilole: tumai.
Mota, Duke of York, Buka, Baravon, Nggela, Laur, King,
vSesake, Malo, Bierian, Tangoan Santo, Arag, Vaturanga,Bugotu,
Motu, Sinaugoro, Rubi, Suau, Sariba, Tubetube, Panaieti,
Misima, Nada, Murua, Kiriwina, Dobu, Mukawa, Kubiri, Raqa,
Kiviri: tama. Oiun: tame. Tanna: timi(n). Santo: tima.
Pokau, Doura : kama. Ulawa, Wango, Uni: 'ama. Fagani:
wama. Saa, Bululaha, Wagawaga, Mekeo, Hula, Keapara,
Galoma, Tavara, Awalama, Taupota,Wedau : ama. Roro : hama.
Nggao, Lo : ma. Mota, Omba, Gog, Alite, New Georgia, Koita,
Motu : mama. Boniki, Galavi : mamai. Merlav, Lakon, Pak,
Sasar, Vuras, Mosin, Alo Teqel, Motlav, Volow, Norbarbar : mam.
Malekula, Tangoan Santo : tata. Nengone: chacha. Baki: ka-
rama. Panaieti, Misima : nam. Tagula : rama. Raqa: dada.

DATA AND NOTES. 273

Salayer, Liang, Lariko, Saparua, Awaiya, Caimarian, Wahai, Teor:
ama. Morella :a'ma. Cajeli: a'mam. Ahtiago : amdi. Teluti:
amaeolo. Amblaw : amao. Bouton : amana. Menado : iama.
Sanguir : yaman. Tobo : jaman. Bolanghitam : kiamat. Gah:
mama. Mysot: mam. Mayapo: nama. Massaratty: ndama.

In the Polynesian this is distinguished from tdma child by the accent
iama or by the addition of a final syllable which automatically secures the
same incidence of the accent, tamdi, tamdna. Diacritical marks have been
but sparingly used in our Melanesian vocabularies and for that reason we
lack a sure guide as to the accenting of the western affiliates. We find but
the cases of New Britain, and Pala in which the accent is printed, and Mota,
Nifilole, Redscar Bay, and Aneityum in which it is inferential. In not one
of these languages have we any evidence of the use of tdma child, therefore
the accent is not a differential in their own material, but has carried its
ictus from the source whence the tarn a father has been borrowed.

We shall first examine the languages which retain the t-m consonant
skeleton. There is a long series in which the two vowels remain as in
Polynesian ; they are therefore identical words except that the accent may
vary, and on that point we are without information. The final a is almost
wholly permanent, not only in the t-m series but in the m series which will
come up for our later consideration. The solitary exceptions are Tanna
timiin) and Eromanga temi. The former a in this case becomes e in Ero-
manga and i in Tanna. The i- mutation also appears in Santo lima. In
Nifilole we find u in tumai.

Our next variant of tama involves frontal abrasion affecting the t. This
we find in two discrete areas in the Solomons, respectively, north and south
of Malanta and separated by an area of greater degradation. With the
ama of this abraded type I include Fagani wama.

This middle type gives us a suite of occurrences of the transition form
by which we arrive at forms in which not only the initial t but the vowel a
thus become initial have been subjected to frontal abrasion. The simplest
form is ma in Nggao and Lo. By reduplication of ma we may more logically
account for mama than by attempting to establish a t-m mutation. By final
abrasion of mama we arrive at mam, a form very widespread in the Banks
Group, the northern subdivision of the New Hebrides.

Few forms lie outside this chain. Aneityum etman is only superficially
irregular, for by punctuating apart its formative elements we find in
e-tma-n our tama theme with loss of the former vowel, which is character-
istic of many of the Polynesian homogenies in that language. The lata of
Malekula and Tangoan Santo and the chacha of Nengone appear to be the
result of a terminal abrasion involving the final syllable and then a redup-
lication. There is a priori no reason why such a course should not have
been followed by the former syllable as well as the later; we must note,
however, that we find no evidence such as transition forms would afford in
support thereof. If Baki karama stems with the ma type of tama it
involves a component kara as to which we lack information.

The Indonesian homogenetic forms are most largely of the transitional
ama type, only Tobo jaman suggesting a tama possibility and in our igno-
rance of the source of this record jaman may be but a variant transcription

274 THE POLYNESIAN WANDERINGS.

of Sanguir yaman. The still more dilapidated forms of ma, mama, and
mam also appear.

From the Micronesian Pacific we record Gilbert Islands tdma father, and
in passing note that this archipelago has borrowed much from Samoa and
in somewhat recent times ; and Ponape jam father.

EFATE-VITI-POLYx\ESIAN-MALAY-SEMITIC.

218.

bwaka, a fence of stone or wood made for protection or fortification in war.

Samoa, Tonga, Niue, Tahiti, Marquesas, Moriori: pa, a wall, fence,
hedge. Maori: pa, a barricade, an obstruction, a fort, a
stockade. Paumotu: pa, a rampart, bulwark. Rapanui:
pa, a wall, to inclose. Hawaii, Mangareva: pa, a wall and its
inclosure. Rarotonga: pa, an inclosure. Fotuna: pa,

kaupa, a fence. Samoa : 'aupa, a wall, fence, hedge, bulwark.

Viti: mba, a fish fence ; mbai, a fence around a garden or a town, but
not around a house.

Malo: baba, a fence. Tanna: kaupa, id. Eromanga: nim-pa-t, id.
Mabuiag: pa (plural pal), a fence for a garden, a stockade.

Malay: pagar, a fence, a railing; palang, a bar, a piece of wood laid
crosswise in obstruction. Malagasy: bako, a pinfold; bamba,
a wall or fence in fortification.

Hebrew : ma'akeh, a parapet (surrounding a fiat roof) to hinder one
from falling off ; 'akah, to hold back, to hinder, to impede.

Our Polynesian, Viti, and Melanesian identifications deal only with a
simple pa stem which exhibits but the slight normal ^-variations. Efate
bwaka involves a new element which we are unable to identify, and the
same is true of the Indonesian.

219.

bakauti, buti, to make an end, to finish.

Niue : oti, all, entirely ; jakaoti, together, to destroy utterly, to make
a clean sweep. Maori: oti, finished, ended. Tahiti: oti, to
be done, finished. Mangareva: oti, the end, finished, all over.
Rarotonga: oti, finished. Paumotu: jakaoti, to finish, to con-
clude. Nukuoro: hakaoti, to end, to finish. Moriori: hokoti,
to cause to cease. Nuguria: huoti, to finish. Sikayana:
oti, all, to finish. Samoa: talafa'aoti, to tell all. Tonga:
oji, to be finished, to be done, all gone. Fotuna : oji, all, the
whole. Uvea: fakaosi, to finish.

Viti: otia, vakaotia, to finish, to bring to an end, to complete, to
perfect; oti, finished, done, destroyed, utterly ruined.

Aiieityum: oti, gone, done, finished.

Malagasy: cf. oty, picked off, gathered (of fruits), weaned.

Hebrew : kaseh, an end ; kasah, to finish.

In Efate buti I find an instance of the most degraded faka-fa stem,
b-uti, paralleled by the similar v — shown in 214.

The identifications here offered are so patent that they need not detain us.

DATA AND NOTES. 275

A particular interest of an ethnographic nature will attend our exami-
nation of the areas in which this homogeneity is found traceable. OH
seems to be a word of the Tongafiti swarm. It is found at the remotest
beaches upon which broke that wave of migration. In the Nuclear Pacific
we find it in Samoa, Tonga, Nukuoro, Uvea, Viti, and Niue. In Samoa it
has found a lodgment in but the one word cited. In the islands of the
Western Verge we find it in Sikayana and Fotuna. In Melanesia it appears
only in Efate and Aneityum. Compare with this any of these records
showing a word of the Proto-Samoan baggage ; Melanesia is speckled with
its occurrences. In this case we are at no loss to account for the Tongafiti
word in Nuclear Polynesia, for we know that to have been a halting-place
for the later swarm in the permanent home of the former. We know, too,
that Samoa by a mighty effort cast off the invaders, and we are therefore
not surprised to find so slight a remnant of the enemy's speech. The
presence of this Tongafiti word in two of the islands of the Western Verge
and in two of the New Hebrides calls for attention. The absence of Tonga-
fiti homogenies in Melanesia indicates for that migration a different course
in general, but such instances as this go to show that, while our conclusion
is in the main true, now and then a small squadron may have found its
way down the ancient track, or that when the second swarm was expelled
from Nuclear Polynesia some of its fleets may have gone westward to
homes where the chances of settlement were but slight.

220.

balo-ni, bano-li, balo-si, bilo-si, bulo-si, bulu-ngi, bulu-ni, bunu-li, to wash
anything, to wash by rubbing. Cf. bafano 49.

Samoa : fufulu, fulua, to rub, to wash, to wipe ; fulunga, the rubbing
of a thing. Nukuoro : fufulu, fulua, to wash. Tonga, Futuna,
Uvea: fufulu, to wash, to cleanse. Fotuna: no-furuna, to
wipe.

Viti: vuluvulu, to wash the hands.

Malay: basuh, to wash. Malagasy: uza, id.

Arabic: masa, maus', to wash, to rub with the hands.

These Efate" forms are in a snarl which needs disentangling before we can
give them precise study. We shall first examine the forms which exhibit
the skeleton b-l-n. These are baloni, buluni, each being accompanied by
a metathetic form, banoli and biinuli respectively. With the b-l-n skeleton
I must include the slightly variant bulungi.

Another skeleton, b-l-s, occurs in balosi, bilosi, bulosi.

Dr. Macdonald's crossing of reference to bafano can only apply to the
forms which I have preferred to regard as metathetic. If he regards them
as principal, his identification with fufuluhas no standing; and if he regards
baloni as principal, his reference to bafano is irrelevant.

But neither b-l-n forms nor b-l-s forms can be properly identified with
the fulu stem of Nuclear Polynesia, for that is an open stem. Yet in
Fotuna no-furuna we find an n as to which we have no explanation in the
jejunity of our only account of the grammar of that speech. It may be a
formative suffix, even as we know the no to be a formative prefix. It may
be the sole Polynesian survival of a fulun stem. It may be grafted upon

276 THE POLYNESIAN WANDERINGS.

the open stem by attraction from a New Hebridean neighbor. If fulun be
the Proto-Samoan stem its affiliates will be the b-l-n forms of Efate. To
associate therewith the b-l-s forms entails an n-s mutation of which we
can not find a single trace in this material.

The Malay and the Semitic identifications do not come up for consider-
ation at all.

221.
bwoa, nabwo, tamo, to emit odor; bwon, odor. Cf. 139.

Samoa : poa, a yam having a fragrant odor ; poapoa, fishy smelling ;
fa'apoa, to feed young children with fish. Futuna : poa, popoa,
to smell fishy ; poa tai, odor of the sea. Tonga : boa, the name
of a species of yam, the smell of fish ; fakaboa, to scent anything
with fish, to smell of fish ; tauboa, to scent the water with fish
to catch others.
Viti : mboi, to emit an odor.
Malay: bua, odor. Malagasy: fufuna, id.
Arabic : faha, fall' a, to emit odor.

222.
but i, buti, futi, to pluck (as a fowl), to pluck out or up (as weeds) ; mafuti,
to be plucked.

Samoa : futi, to pluck feathers or hair, to pull up weeds ; fufuti, to
haul in the fishing line. Futuna : futi, to pull out feathers or
hair ; futitaula, to raise the anchor. Niue : futi, to draw up
(as a fish on a line), to hoist (as a flag), to pluck (as a hair).
Nukuoro : futi, to pick, to pluck. Sikayana : ufuti, to pull or
haul. Tonga: fufuji, to pull, to stretch out; fuji, to pull, to
pluck, to deplume. Maori: huti, huhuti, to hoist, to pull up
out of the ground ; huti-ika, to pull up a fish. Tahiti : huti,
to pull or draw up a fishing line, to hoist a flag ; huhuti, to pluck
feathers, hair, grass. Marquesas : huhuti, to pull one another
by the hair; hutihuti, to pull out the feathers of a bird, to pull
the hair. Mangareva: huhuti, to pull up as by the roots;
hutihuti, to pull up herbs, to pull out feathers ; uhuti, to pull up
by the roots. Paumotu : hutihuti, to denude the body of hair.
Rapanui : huhuti, to weed ; hutihuti, to pluck feathers. Hawaii :
huki, to draw, to pull ; uhuki, to pull up as grass or weeds.

Viti: vutia, to pluck feathers or hair, to pull up grass or weeds.

Mota : cf . pit, to take up or off with the tips of the fingers, to pick,
to pluck.

Malay: bantun, to pluck, pull out.

Arabic: namasa, to pluck out, as hairs.

Because of its signification I collate Mota pit, yet with some hesitation
because of the fact that in the study of some 1 50 Mota words homogenetic
with Samoa this is the only instance of the f-p mutation. This objection
rests upon this case rather than the broader knowledge of consonant move-
ments in these languages, for f-p is a sufficiently well-established mutation
in the true Polynesian and appears frequently in the Melanesian material
here collated.

DATA AND NOTES. 277

The Malay proposed in identification assumes without confirmation a
stem of three consonants. The Arabic shows not the slightest resemblance
save in meaning.

223.

fis i, fisi, fifis i, fif i, to bind around or about, to twine around or twist;
fifi, to twine or go around, a fillet.

Samoa : fisi, to entwine, as a vine around a stick ; fa'afisi, to entwine,
to coil; fifi, the small intestines; aft, to do up in a bundle.
Tonga : fihi, thick, bushy, entangled ; fihifihi, curled in the grain,
linked one into another, inextricable ; fifihi, one who in wrestling
is dexterous with his limbs locking the limbs of his antagonist ;
fi, to plait, to twist, to curl; fifii, to enclose fish in plaited
coconut leaf. Nuguria: afii, to wrap up. Futuna: fifi, a
bundle of cooked fish. Niue : fi-ika, a bunch of fish. Tahiti:
fifi, entangled; faafifi, to entangle. Maori: whiwhi, twisted
together; whakawhiwhi, to wind around; whiwhiwhi, the fat
covering the intestines. Hawaii: hihi, the twining of vines;
hoohihia, to entangle.

Viti: ww, to roll up or around, to coil around.

Mota: viv, vivis, to wind around, to bind around and around.

Malay: pusing, to turn around, to twist. Malagasy: fihina, grasp,
seizure, fihitra, a clutch, a grasp.

Hebrew : habas', to bind, to bind on, to bind about.

In"this group we are dealing with a stem fis, which may be expanded to
fisi and which may be abraded to fi; see note 243.

In Polynesia we find only fisi and fi, in Viti only fi, in Efat£ all three
forms, and in Mota only fi and fis.

The Indonesian suggestions involve the difficulty of a third radical con-
sonant which has left no vestige in our area; see also note 245.

The Hebrew likewise entails a third consonant in the root, differing from
the Indonesian in being applied frontally.

224.

katnut i/ngamut i, kami, to take, to grasp with the fingers, to nip, to nip

or cut with scissors ; kam, native tongs (a split stick for grasping

hot oven stones and lifting them) ; kamkam, scissors.
Tonga : kamu, to cut off anything round. Futuna: kamu, to cut,

to shorten. Samoa: 'amu, to cut off, as part of a beam.

Hawaii : amu, to shear or shave the hair from the head, to trim

the hair.
Viti : nggamu, pincers, vise ; nggamuta, to take hold of or hold with

pincers or between the teeth.
Malay : cubit, chubit, to nip, to pinch. Java : juwit, id. Malay :

angkub, tongs, nippers.
Hebrew : kamas, to squeeze together, to take with the hand ; kamat,

to hold fast with the hand, to seize firmly; kafas, to contract,

to shut (as the mouth) ; kabas, to take or grasp with the hands.

Arabic : kabasa, to take with the tips of the fingers ; kabas' a, to

grasp with the hand.

278 THE POLYNESIAN WANDERINGS.

The Proto-Samoan stem is manifestly kamut, yet the material is so scanty
in Nuclear Polynesia that we can establish it only from Viti, and possibly
from Hawaii amuku to cut short, which, however, Andrews associates with
muku to cut, without pointing out or explaining the difficulties which such
derivation entails.

225.
kar i, ngar i, karu-ti, ngaru-ti, to scratch; karo i, ngaro i, to scratch, to
scrape, to shave, to seize; tangaru, to seize, to grasp.
vSamoa : 'Hi, a rasp, a file, a saw. Tonga : kili, a saw ; kilijakamata, a
file ; kiliji, to saw. Futuna : kiliti, to file ; kili, a rasp, a file.
Uvea: kili, id; kilisi, to file. Niue: kili, a file, a saw.
Viti: kari, to scrape.

Melanesian : all meaning to scratch or scrape — Galavi : giri, lagiri.
Nada: qiri. Raqa: kairi. Wedau: giai. Mukawa: giagiai.
Kiriwina : kuriqari. Panaieti : ktirikuri. Kubiri, Kiviri : gagara.
Oiun: kakakara. Motu: hekagalo. Taupota: karakaroi.
Wedau: kakaroi. Boniki: kelologi. Mota, Nggela: karu.
Bugotu: g'ag'aru. Wango: karohi. Malo: garasi.
Malay: garis, to scratch, to score; garut, to scratch, to scrape, to

claw; garok, to scrape. Java: garit, to scratch, to score.
Arabic: garra, to drag, to snatch, to sweep, to seize. Hebrew:
garar, to scrape, to sweep; gara', to scratch, to scrape.
Dr. Macdonald has amassed a number of similar forms, there are many
more than these in his dictionary, and we shall have to disentangle them.
The Proto-Samoan stem is kilit, and by a normal Polynesian mutation
this becomes kilis in Uvea. This permits us to identify karuti and ngaruti.
Thence we may pass to karo and ngaro, kar and ngar, so long as they carry
the meaning to scratch, to scrape. Here we find the Viti kari to fit into
the scheme ; its form shows that it was not derived from the neighboring
Polynesian kili, but came from the stem by way of Efate. The Indonesian
forms are in closer accord with Efatd and Viti than with the Nuclear Poly-
nesia and give the impression of an earlier type.

It will be observed in the Efate that while the scratch sense runs through
all the forms the grasp sense is absent from those which exhibit the
radical t. We may, therefore, judge that there are two stems interlaced,
karut-karo-kar to scratch, and karo-kar to grasp, and that the reduction
forms of karut have become involved with the karo-kar forms of the other
stem. This will remove the Arabic identification entirely from considera-
tion and will leave to the Hebrew but a partial resemblance.

226.
kau, a collection, bunch, herd.

Tonga : kau, plural sign. Futuna : kau, a multitude, a troop. Niue :
kau, a troop, a company. Uvea: kau, a company, herd.
Mangareva: kou, a multitude (kouika, a shoal of fish). Samoa :
'an, a troop, a gang, a bunch, a cluster. Tahiti, Hawaii : au,
collective plural.
Viti : kau, a bunch.
Malay : kawan, a herd, a troop.
Arabic: gam', gama'a, a collection.

DATA AND NOTES. 279

In view of the fact that kau retains its place in Nuclear Polynesia and
has scantily entered the Tongafiti languages it will be interesting to cite
Tregear's note (Maori Comparative Dictionary s.v. tckau) :

It is evident that there was an original Polynesian word kau, a troop of persons, a
cluster of things, etc. The Tongan kauvaka, a crew; kaugane, fellow-workmen; kau-
mea, a companion; the Samoan 'au, a bunch of bananas; a troop of warriors; 'aujale,
women living together in a house; the Tahitian auono, a large fleet or company of
travelers; autahua, a company of priests; the Mangaian kaunuku, in groups, etc., all
point to a word signifying collection, assemblage.

I can not see that the Arabic has aught to do with it.

227.
kinit i, nginit i, ngunut i, to nip with the fingers ; nakini na, fingers (nippers),
toes. Cf. 57.
Tonga : kini, to strike, to cut the hair short, to let blood ; kiniji, to
strike with anything light, to hit with a whip. Futuna: kini,
kikini, to beat, to strike, to whip. Niue : kini, to beat down (as
bushes) . Uvea : kini, to whip. Maori : kini, to nip, to pinch.
Samoa : 'ini, to take hold of with the nails, to pinch, to pull up
small weeds. Hawaii: iniki, to pinch with thumb and finger.
Viti: kinita, to nip, to pinch between finger and thumb.
Mota : gin, ginit, to nip, to pinch. Malekula Pangkumu : kinji, id .
Malay : gantas, to break off, to nip off, to snap off.
Arabic: karasa, to nip with the fingers, to pinch, to snip off.
The Proto-Samoan stem is kinit.

The distribution of this word is striking. The nipping sense is found in
Mota, Malekula, Efate\ Viti and Samoa, thence far north to Hawaii, far
south to New Zealand ; yet in Tonga, Futuna, Niue and Uvea we find
the word completely devoid of this sense and charged with a wholly different
meaning which nowhere else appears. I find it quite inexplicable.
The Semitic has no relation other than that of sense.

228.
lua, le, lai, to vomit, to put out the tongue (or anything), to flow out.

Samoa: lua'i, to spit out; lulua, to be sick, to vomit, to puke.
Tonga : lua, lulua, to vomit, to disgorge ; luaki, to be sick with ;
jakalua, to nauseate. Futuna : lulua, luaki, to vomit ; fakalulua,
to nauseate. Niue: lua, to vomit; jakalua, id.; fakalue, to
spew out. Uvea: lua, to vomit. Fotuna: noh-lua, id.

Rapanui : rua, seasick, to vomit. Maori, Rarotonga, Nukuoro :
ruaki, to vomit. Paumotu : ruaki, to vomit, toeructate, to belch.
Tahiti: ruai, to vomit. Mangareva: aruai, akaruta, id.
Hawaii: luai, id. Marquesas: iid, akaruta, id.
Viti : lu, to run or leak out ; lua, to vomit ; loloa, qualmish, seasick.
Mota: lua, to spew. Malo: lua, id. Santo: lulua, id. Baki:
mjuluo, id. Malekula: ru, id. Aneityum: a-lo, id; aluo,
aluun, to put out the tongue.
Malay: luwat, luai, to vomit; luwar, luar, out, away; luwari, luwar-
kan, to put out, to expel. Malagasy : lua, to vomit ; mandua,
to vomit ; luatra, over and above, taken up, put out ; manduatra,
to take out or up.
Arabic: t'a't, ta'a, tai'at, t'a'at, to vomit.

280 THE POLYNESIAN WANDERINGS.

The Proto-Samoan stem is lua. The Mangarevan akaruta is quite
singular, and, while not succeeding in identifying its primitive ruta, I can
not admit it as a lua derivative. The same word akaruta cited by Tregear
from the Marquesas is not found in Bishop Dordillon's dictionary.

The le and lai of Efate are profoundly degenerate forms if they be really
of the lua stem. The most degenerate form in Melanesia is the ru of Male-
kula and the ro as a component of Aneityum a-lo. The discovery of these
two forms in the sense to vomit gives a certain probability to the Viti lu
in a sense which, while different, is not wholly irreconcilable with this stem.
The Efate sense "to put out the tongue (or anything) " — and I am free to
acknowledge that I can not imagine what Dr. Macdonald means within the
parentheses — finds support in Aneityum aluo and aluun, to put out the
tongue, to thrust out the tongue, respectively. But the sense is only
remotely, if at all, related to the stem sense of lua.

The Indonesian carries lua identifications with much extraneous matter.
The value of the Semitic identification wholly fails to appear.

229.
ma i, to chew (softening food for an infant).

Samoa, Tonga, Futuna, Tahiti, Rapanui, Marquesas : mama, to chew.
Niue: mama, a mouthful, that which is chewed. Hawaii:
mama, to chew with a view to spit out of the mouth. Man-
gareva : mama, to chew, to bruise with the teeth. Nukuoro :
manga, to chew. Uvea: maanga, a morsel.
Viti : mama, to chew, used chiefly of the kava root.
Aneityum : a-mai, to chew (kava or any bark) .
Malay: mam ah, to chew.

Arabic: ma"ma"a, to chew meat, but not wholly.

230.
manga, maka, to open out, to gape, to wonder, to speak, to open the jaws ;
tnangamanga, to gape often and rapidly, to pant, manga, a
part of the names of places, as gorges and valleys, and especially
of the abysses of Hades below Bokas.
Samoa: manga, a branch of anything forked, as a tree, river, road
or fishhook ; tnangamanga, branched, forked ; fa'amanga, to open
the mouth, to gape ; fa'amangai, to set astride. Tonga : manga,
a branch of a tree, road, fishhook, stream, open, forked, spread-
ing ; mangamanga, to branch off, to spread open ; mamanga, to
stride, to extend the legs; fakamanga, to open, to gape; faka-
mangamanga, to barb, to jag, to make forked. Futuna : manga,
a branch, a fork ; fakamanga, to have the legs spread out. Niue :
manga, forked ; mangaua, cloven ; fakamanga, to open the mouth,
to gape; fakamamanga, to straddle, to open the mouth, to
spread out, to extend. Maori: manga, the branch of a tree
or a river. Mangareva: manga, the branch of a tree, forked,
cloven. Paumotu: manga, a branch, a division. Rapanui:
mangamanga, a branch ; mangamanga rima, a finger. Hawaii :
mana, a branch, a limb, to branch out, to be divided. Mar-
quesas: mana, a branch, as of a river. Tahiti: maa, cloven,
divided; amaa, the branch of a tree.

DATA AND NOTES. 281

Viti: manga, pudendum rauliebre.

Mota: manga, an opening with lips, mouth, to open, to gape.

Malay : manga, open ; mangah, to pant, to palpitate ; nganga, to gape.

Ethiopic :nakaKa, to gape, to yawn, to be rent, parted, sundered ;nka'at,

an opening, gap, fissure. Arabic: naka'a, to rend asunder.

This forms an excellent and consistent series from Polynesia to Indonesia.

The Semitic identifications proposed by our author have not the least

connection with the Polynesian m-ng root.

231.
mina, pleasant, nice.

Samoa; momona, fat, rich (of pigeons and fish). Tonga: momona,
fat (as shellfish). Futuna: momona, flesh of sea food. Maori:
momona, fat, rich, fertile. Hawaii: mona, fat, rich, fertile,
round, plump, the fat of an animal. Tahiti: mona, momona,
sweet, delicious. Marquesas : momona, delicious, good to taste,
fat part of an animal. Mangareva: momona, grease, fat.
Paumotu: momona, odor, savor.
Viti: mona, brains.

Malagasy: monamonany, fat, plump, of a child or young animal.
Arabic: 'anik\ pleasant, nice.
The Marquesas combines the two significations of fat meat and deli-
cious. The meat sense runs through Samoa, Tonga, Futuna, Maori, Hawaii,
Mangareva. In 142 we have seen a word beginning in oil and ending in
brains; the Viti attains the same end from an equally greasy beginning.
The Paumotu momona has undergone a particular and independent special-
ization. Taking a fresh start from the Marquesas and examining the sense
of delicious we find that to be the only sense in Tahiti and Ef ate ; possibly
it may be inferred in the rich of Samoa, Maori and Hawaii.
Again Dr. Macdonald's Semitic is not even a resemblance.

232.
mitei, breadfruit fermented and preserved. Cf. mutrei 6.

Samoa : mast, fermented breadfruit ; matt, stale (of water, coconuts,
kava) . Futuna : mast, breadfruit or bananas fermented ; matt,
wilted and yellow, of leaves of tobacco and taro. Nukuoro:
mast, bad-smelling. Tonga: mahi, sour, acid; maji, sour,
decayed, as the nut when kept too long. Tahiti : mahi, bread-
fruit fermented. Mangareva: cf. mahimahi, cooked food kept
until the next day to make it better.
Viti : masimasia, the breadfruit in a certain state.
Malay: masin, salt. Malagasy: masimasina, saltish.
Arabic: mast', salt (of water).
The alternative form mutrei I have (note 6) carried out to its Viti con-
gener mandrai.

This form mitei we may adopt as related to the Polynesian mast, as set
forth above. I have noted the Polynesian mati as being nearer mitei
in form, but it clearly has no connection in sense. Mati carries the impli-
cation of being unfit to eat, and no Polynesian would think so ill of his mast,
even though the odor is overpowering to Europeans and suggests the reflec-
tion that stale would be but a weak description. I am not quite confident

282 THE POLYNESIAN WANDERINGS.

enough of the Nukuoro vocabulary to accept bad-smelling as the definition
of mast, though it is fact none the less.

There is no physical fact in the fermentation of pounded breadfruit to
enable us to connect it with the Indonesian and Semitic here proposed. The
Malay masin finds its near relative in Nuclear Polynesian masima salt.

233-
sau, gentle breeze, cold air, as in the morning and evening.

Tonga : haua, to be exposed to the wind, to blow to and fro. Niue :
hahau, hauhau, cool. Maori: hau, wind, to be borne on the
wind: hauhau, cool. Tahiti: hahau, to go aslant or beat in,
as the rain driven by the wind into a house ; haumoe, the cold
night breezes of the valleys ; mehau, wind ; puihauhau, to blow
gently, as a small breeze; haumaru, cool, grateful. Hawaii:
hau, the name of the land breeze that blows at night, any cool
breeze ; hauhau, cool ; kehau, the mountain breeze in the morn-
ing, a cold fine rain or mist. Marquesas : tohau, a gentle wind.
Mangareva: hau, to blow gently. Rapanui: hou, breeze;

hahau, hauhau, air, breeze; hakahahau, to expose to the air.

Viti: thauthau, the land breeze.

Malay: hawa, wind.

Hebrew : nas'af, to blow ; ncs'ef, the evening twilight when a colder
gale blows, the morning twilight.

In discussing the sau stems which are involved in this and the next
two items Mr. Tregear with truth remarks that it is difficult to seg-
regate the several senses under proper headings since the significations all
pass into one another. The only contribution toward the classification
of the matter which I can offer is the note that sau dew belongs to a stem
which is either saut or saum in Proto-Samoan, and that saxi to cut derives
from a stem in sauf . True, this will not avail much in the present stage
of Polynesian with open stems, but it will serve as a safe guide so far as it
may go in tracing out earlier affinities. The sau of wind and temperature
seems to be from a primitive sau.

The Semitic identification offered is distinctly triliteral. Even were we
to grant that the Hebrew / might become Polynesian u we are still left with
an initial n upon our hands and unaccounted for.

234-
sau, the dew.

Samoa: sau, the dew; sasau, heavy dew; fa'asau, to bedew. Fu-
tuna: sau, the dew; fakasau, to expose to dew. Tonga: hahau,
dew, mist; haujia, hauhau, wet with dew. Niue: hahau, dew;
haumia, bedewed. Uvea: hahau, dew. Maori: hau, haurutu,
hauku, haunui, dew; haumaku, hautaku, bedewed, wet.
Tahiti: hau, dew; tahau, to bleach clothes in the dew of the
morning ; tiohau, to bleach in the dew ; toehaumi, soft or damp
as by dew. Hawaii: hau, cold dew. Mangareva: hau, an,
dew. Rapanui, Marquesas: hau, id. Rarotonga: au, dew.

Viti: sasau, dew.

Malagasy: andu, dew.

Arabic: nada' (for nadau), dew.

DATA AND NOTES. 283

The Proto-Samoan saut rests upon protected forms in Samoa, Tonga, and
Maori. The stem saum in the same sense is preserved in Niue, Tahiti,
and Maori. I can find no evidence of the interchange of t and m except
the apparent instance in 2 17, which, as there shown, is susceptible of a far
simpler explanation.

The Viti sasau is a dialect form, the Bau using tengu and mbite, but with
sau appearing in Efate it is not necessary to consider sasau a recent acqui-
sition from the Tongan.

The Malagasy and Arabic might better be offered as homogenetic with
dew itself than with saum and satit.

235-
sau fi, to scoop or shave the surface off water; to cut or shave off the sur-
face of wood ; sau- baba, an adze, to strip off, to peel off (as clothes) .
Samoa: sasau, a large axe; saupapa, saupapa, to cut off the outer
part of a log of wood to make it level and even ; saujono, to cut
planks for a canoe ; sautasi, one wide plank of a canoe. Tonga :
hahau, to adze, to chip logs of wood square. Maori: hau,
hauhau, hahau, to hew, to chop. Tahiti: hauhau, to take off
the first chips in hollowing a tree. Hawaii : hahau, to hew stones.
Viti: sautha, to cut (as bamboos, reeds), to break a coconut for

drinking (by cracking off a piece at the tip).
Malagasy: sauka, saufina, to scoop out (of water), to draw water.
Hebrew : s'a'ab, to draw water. The primary idea lies in the raking
off the surface, sahaf, to sweep, to scrape off, hasaf, to strip off.
Arabic: sahaf a, to scrape, peel or rub off, to shave.
My recognition of a Proto-Samoan sauf is based upon the Efate, for our
Polynesian is scanty and affords us no protected forms from which the stem
might be revealed. I recognize that this stem does not seem to apply to
the Viti, but that is in sense a doubtful identification.

The first of Dr. Macdonald's definitions of saufi is a treasure; it surely
was first drafted by one of the Danaids in her aquatic employment. The
reason for its presence in the Efate dictionary will, of course, be found in
the Semitic suggestion. The true signification is to hew with the adze,
and it is only in the Maori that this sense is not made to appear in the
definition. This stamps the Viti identification as inconsistent.

The Malagasy has not the Polynesian sense, but it does accord with the
Efate scooping of water.

236.
tau-ngi, to grasp firmly with the hand, to pluck off with the hand (as fruit).
Samoa: tau, to pluck fruit with the hand; tau, to press out (as
juice), to milk. Tonga : tau, to squeeze or wring out. Futuna :
tau, ta-tau, to squeeze, to express. Niue: tau, to gather
gardenias; tatau, to wring, to strain, to press out. Uvea:
tatau, to press; taui, to pluck. Maori: tatau, to squeeze, to
express juice. Fotuna: ko-tauia, to wring, to express.
Viti: taura, to take hold of.

Malay: sambut, to lay hold of. Malagasy: sambutra, id.
Hebrew: sabat, to grasp, to lay hold of firmly, to seize, to pluck.
Arabic: s'abata, s'abat'a, id.

284 THE POLYNESIAN WANDERINGS.

It is impossible to see how the Indonesian and the Semitic, though there
is a superficial resemblance between them, can have anything to do with
the Efat6 and Polynesian of this item.

237-
tula, earwax.

Maori, Tahiti : taturi, earwax. Mangareva: teturi, id. Paumotu:
katuri, id. Hawaii: kokuli, id. Fotuna: turituri, id. Mar-
quesas: tetui, id.
Viti: tule, ndule, earwax.
Baki: tiro, earwax. Mota: tul, id.

Malay: chulik, to clear the ears of wax. Bicol: tuli, earwax.
Arabic: salah, deafness.

Because of the fact that the island diagnosticians regard cerumen as
the sole cause of deafness and in their practice of medicine not infrequently
produce deafness by their exploratory excavation of the ear, and because
of the intimate association of this stem with the most common word for
deafness, I include the latter for the extension of this record.

Samoa: tuli, talingatuli, deaf. Tonga: tuli, id. Futuna: tuli-
tuli, id. Uvea: tuli, id. Niue: talingatuli, id. Hawaii:
kuli, id. Tahiti : turi, taturi, id. Maori: turi, id. Marquesas:
tui, to disobey, to turn a deaf ear ; hadtetui, to turn a deaf ear ;
putui, deaf, disobedient ; hadputui, to turn a deaf ear. Rarotonga :
turi, deaf. Mangareva: turi, noise. Paumotu: taringaturi,
disobedient. Fotuna: eturitura, deaf. Nuguria: tarina-

turi, id.

Viti: ndalingatule, deaf.

Motu : tuia, to quiet.

Malay: tuli, deaf. Matu: turang, id.

From this it apppears that in the very center of Nuclear Polynesia tuli
means deaf, yet that the sense is more precisely conveyed by joining with
it the organ affected, talinga being, of course, the outer ear, which is as far
as their knowledge of aural anatomy goes. This composite is the only
means of recording deafness in Niue, which has not retained the tuli-stem
in independent existence. It is the only means in Polynesian Viti (for it
has a Melanesian term, ndindivara), which retains tule {ndule) for the ceru-
men. In Samoa both talingatuli and tuli exist side by side in the same
sense. In the remotest Polynesia of all, the Paumotu, an archipelago of
linguistic problem, the word exists in a tropical sense only. But the rest
of Polynesia expresses its deafness satisfactorily by tuli, and we find the
word in Indonesia and possibly in Motu.

Now if we regard the employment of tuli for cerumen we shall note that
such use extends from Indonesia into Viti. Of Nuclear Polynesia we can
not speak with greater precision than to say that all our dictionaries have
omitted this sense in defining tuli; at the same time they have neglected
to define cerumen at all. But in all eastward Polynesia, the lands of the
Tongafiti swarm, it is necessary to reinforce tuli with another element in
differentiation from the sense of deafness. This element stems the same
throughout this second migration, subject to the normal variation. It is ta

DATA AND NOTES. 285

in Maori and Tahiti, ka in the Paumotu, ko in Hawaii, te in Mangareva and
the Marquesas. That it was brought by this migration from its western
home is shown by this uniformity, and it is in use wherever that migra-
tion has reached. It is so ancient that no trace of its original significance
can now be discovered.

By combining these two records we find :

(i) The Proto-Samoans used tuli to mean the visible physical obstruction
of the ear, and to convey the sense of deafness they employed a locution
signifying waxed-ear.

(2) The Tongafiti when their migration swarmed had reached a stage
in which tuli had lost its primal sense, was altogether used of deafness, and
to convey the cerumen meaning a compound was necessary.

(3) Nuclear Polynesia, being a meeting-ground of the two migrations,
shows the record of the earlier overlaid by the later.

The use in Fotuna of the /w/i-cerumen and tuli-deai is one of the inter-
esting pieces of evidence to show that the squadron which settled that verge
island had been in the area where the Proto-Samoan and Tongafiti swarms
had been in conjunction, and that the period of its voyage must have been
subsequent to the coming of the Tongafiti fleets to Nuclear Polynesia.

238.
uta i, uta ki, to load (to make sink, to immerse) a canoe ; uta, a canoe load,
a cargo.
Samoa: uta, the cargo, the load of a boat, ship or canoe. Tonga:
uta, the cargo or freight of a vessel. Uvea: uuta, to fill up.
Futuna: uta, cargo, lading. Maori: uta, to put on board a
canoe, to freight, to load ; utanga, cargo, lading. Mangareva :
uta, to carry by sea to land or by sea to another country ; utanga,
a big loading ;'or freight. Marquesas: uta, to carry, to

transport; utatina, utaia, cargo. Paumotu: utanga, bag-

gage, burden, freight, the lading of a ship. Rapanui:
hakauta, to give passage. Tahiti: utaa, the burden or load of
vessel. Hawaii : ukana, baggage on or to be put on a canoe
or vessel; hoouka, to freight, to put aboard a canoe. Fotuna:
auta, cargo; fakaute, to load.
Viti: usana, usa, to convey a cargo; usana, usausa, a cargo.
Tanna: (t)auuta, to load; nauuta, cargo,
Malagasy: undrana, to load a canoe.
Arabic: "ata ("a'tu), to immerse.
There is no evidence to show that this is other than an open stem, yet
Dr. Macdonald joins it with utu (168), of which the stem is demonstrably
utuf, and derives both from a Semitic parent of an Arabic word meaning to
immerse. In clinching the identification he so defines this uta as to lose
all sight of common sense. The one aim of the Efate stevedore and of
the lumpers aboard the greatest cargo tramp that ever steamed away from
the Broomielaw is so to stow the lading of his canoe, of their whaleback,
as to preclude all risk that he shall "make it to sink or immerse." Does
not the man know that in the days when shipmen had the piety that fears
the sea all ships' papers, after reciting the cargo, wound up with the prayer,
at least the formula, "and so may God send the good ship safe deliverance" ?

286 THE POLYNESIAN WANDERINGS.

"To load, "he says, "to make sink, to immerse," and all because the Arabians
of the nights use for immerse a word which he thinks to look like uta.

The word, belonging to both migration swarms, is practically one from
Melanesia to the Paumotu, both in form and in sense. Very little, there-
fore, calls for comment. In the Marquesas utatina involves the termination
Una peculiar to that language, but which has the same function as the
common nga or na wherewith nouns are formed from verbs. Our Uvea
dictionary is very jejune, but there can be no doubt that uuta has the char-
acteristic cargo sense. In Viti we encounter the t-s mutation, which is by
no means unusual in that speech and will be found elsewhere in Melanesia.
The Malagasy has the usual distortion of the resemblance of the words in
that language which stem upon the Polynesian.

EFATE-MELANESIAN-POLYNESIAN-MALAY-SEMITIC.

239-
asua, to smoke.

Samoa: asu, smoke. Nuguria: ohu, thin smoke ; aw, thick smoke.
Nukuoro: asu, smoke. Rotuma: aasu, osu, id. Tonga,

Niue, Uvea: ahu, id. Futuna : afu, id. Maori: au, auahi, id.
Tahiti: au, smoke, vapor; au auahi, smoke; auahi, fire. Ma-
ngareva : aw, cloudy mist on the ocean ; ahu, mist, cloud ; auahi,
smoke. Marquesas: auahi, smoke, vapor. Rapanui: au,
auahi, steam, vapor. Hawaii: uahi, a cloud, a vaporous
appearance. Rarotonga: auai, smoke.
Moanus: kdsu, kdsumoan {moan, fire), smoke. Sariba, Panaieti,
Mukawa: kasu, id. Awalama: bogahu, id. Malekula Aulua :
basua, id. Galavi: basu, id. Taupota: bahubahu, id.

Wedau, Boniki : bau, id. Motu : qalahu, id. Nifilole : nggasi, id.
Ulawa, Bululaha : sasu, id. Alite: r asu, id. Malo, Tangoan
Santo, Maewo, Merlav, Gog, Mota, Wango, Fagani, Suau, Dobu :
asu, id. Sesake: asua, id. Savo: azuazu, id. Mosin: as, id.
Mafoor (New Guinea) : aas, id. Malekula Pangkumu: ese, id.
Ambrym: walehi, id. Vuras: es, id. Sasar, Alo Teqel, Pak:
os, id. Santo: osun, id. Nggao: ngganggahu, id. Bugotu,
Nggela,Omba: ahu,id. Motlav, Volow, Arag: aho, id. Lakon:
ahauav, id. Lo: hiev, id. Roro: hiavu, id. Norbarbar: ah,
suio, id. Buka: uruhu, orru, id. Nada, Murua: museu, id.
Kiriwina: umseu, id.
Malay: asap, smoke. Malagasy: ctuna, id.
Hebrew: 'as' en, 'as'an, to smoke. Arabic: 'at' ana, id.
As between the two migrations we observe that the Tongafiti has under-
gone a loss of the consonant element. Our Melanesian homogenetic forms
are numerous and widely placed; we shall have to examine several con-
siderable changes, but in not a single instance do we find this central con-
sonant obliterated.

When we look more closely into the Tongafiti languages we shall see that
au has not only lost the radical s; it has scarcely succeeded in retaining the
recollection of its own meaning. Mangareva is the only Tongafiti speech
which recalls the consonant, in ahu, and this is paralleled by au; and both
have gone so far away from the brands of the fire as to mean no more than

DATA AND NOTES. 287

any vapor. Maori and Tahiti are the only languages in which au means
smoke; in each it means vapor in general, each employs more freely a
definitive compound, and Maori has several other words for smoke. Hav-
ing expanded in signification to mean vapor in general, the only way in
which it was possible to make it plain when smoke alone was meant by au
was to make the definitive compound fire -vapor auahi. This we find in
Maori, in Mangareva, in Rarotonga to mean smoke. In the Marquesas this
too means vapor as well. By an odd mischance in Tahiti auahi passed
from the smoke to the fire itself, and to designate smoke it became neces-
sary to cry back once more to au and devise au auahi. We have seen that
in the Marquesas auahi means vapor as well. Passing thence to Hawaii
we find that auahi has lost even the memory of the fire and it has broken
down in form to uahi, used of any vapor. Judge Andrews, ignorant of the
life history of the word, etymologizes this — no other description so fits —
as u ooze or milk, and ahi fire ; unfortunately the author of the very respect-
able Hawaiian dictionary was no philologist. It is proper to note that in
Emerson's English-Hawaiian dictionary (1845) smoke is defined as uahi.

In Nuclear Polynesia asu carries the vapor idea as well. Samoa differ-
entiates by the use of the primitive for smoke and the conduplicate asuasu
for haze and mist ; as is commonly the sense of conduplication forms, it is
intensive. Futuna employs afu for both. In this area no matter of form
change need engage our attention except this of Futuna. It is a particu-
larly interesting case, one with a story to tell.

We have in the material here assembled no record of an s-f mutation;
in my more extended studies in Polynesia I have encountered no other
instance and but two which are at all near it. We know that the Samoan
has the primal form asu. The normal mutation is to h, and thus we have
ahu in Tonga, Niue, and Uvea. But Futuna has no aspiration and it does
possess the sibilant. Now if Samoan users of the word had carried it to
Futuna, Uvea would have learned to say asu. From the Futuna afu we
see that the word came to them in the ahu form. Our charts will show
us that of the three languages which use ahu Uvea lies almost within sound .
But Uvea uses both s and //. If asu had come to Uvea from Samoa asu
it would have remained; that it is ahu shows that it came to Uvea from
some language in which ahu was the form, and, having the aspirate, Uvea
was under no compulsion to change the form which it received. But
Futuna had to provide some way of dealing with the inconvenient or dis-
agreeable h. In Polynesia h is a mutation result from s and from /. For
some reason Futuna felt the impulse to work back from h to / instead of
to s, which was all wrong, but, like all error, the more picturesque it is the
better does it teach. Now the line is sharply drawn between the s-h and
the f-h mutations; s-h is general in Polynesia; f-h occurs only between the
Proto-Samoan and certain of the Tongafiti tongues. If, then, Futuna inter-
prets the h which is brought to it in terms of / it can only be that the Futuna
folk have the Tongafiti Sprachgeist, in other words that they are a Tongafiti
folk left behind on their exiguous two islands when the swarm of their
fellows swept along to distant discovery. At the beginning of this excursus
I mentioned two instances of a near mutation. They involve Samoan salo
to rasp (Tonga-Niue halu)Viti varo, Samoan scle to snare Tonga heleViti vcre.

288 THE POLYNESIAN WANDERINGS.

We have already seen that the Tongafiti folk have found it necessary
to particularize smoke as fire-vapor. The same usage is found, though far
less commonly, in Melanesia. We have in Moanus kasumoan, in which
moan is fire although by no means of our ahi-fire stem ; Lakon ahau av and
Lo hiev are clearly homogenetic with auahi; but these are all.

The stem asu we find intact in a considerable group of languages in the
New Hebrides and two in the southern Solomons. This stem is altered by
frontal accretion of three palatals in Moanus, Nifilole, and Nggao, severally
too remote to be regarded as local interinfluence ; of two linguals in Ulawa-
Bululaha and Alite, which only once more in all our material, sa (337) bad
Tanna ra, are found interchanging; of one labial. I note Codrington's
opinion that "Nggao ngganggahu is the Mota gagavu thick, clouded." But
is it ? Moanus kasu is nearer phonologically and far nearer in the migration
track, for a derivation from Mota to the Solomons is an upstream movement
and against the current. I do not incline to regard this frontal accretion
as a Melanesian device; it seems more reasonable to look upon it as the
remnant of an earlier stem form, but what that earlier and now vanished
radical initial may have been we may not seek to know.

Savo shows asu with the slight change from surd to sonant. Two lan-
guages in the Solomons and one in the New Hebrides have asu with equally
slight change from sibilant to aspirate. Motlav, Volow and Arag, all in
the New Hebrides, carry the change a small step farther by the alteration
of the final vowel. A step beyond in the change of the final vowel brings
us to Lakon. Efate and Sesake have the asu stem unchanged save for the
terminal accretion of the vowel a.

Thus far we have kept close to the primal form. Now, while holding
fast to the s, we are to find a greater vowel change in Malekula Pangkumu
ese and one which sacrifices the differentiation of the two vowels which has
hitherto been found to persist. To this branch of the stem belongs the
Vuras es. With it I have included Ambrym walehi provisionally ; the ehi
can not be a mutant of afi fire, for that is av in Ambrym ; I have thought
it possible that ehi is a further mutant upon ese, and wal is as yet unac-
countable. Santo osun with a terminal accretion introduces the o-branch
which will be noticed in Rotuma as well and in three of the Vanua Lava
languages with terminal abrasion. Norbarbar shows in one of its forms a
terminal abrasion deriving from either ahu or alio, but probably the former.
Lo suggests in hi a frontal abrasion, and that principle may also account
for a part of the second Norbarbar form suio. One Buka form, uruhu, quite
plainly suggests in uhu another vowel change of asu, leaving ur unaccount-
able yet somewhat smacking of wal in Ambrym walehi, or perhaps in these
words ruhu and lehi are akin to Alite rasu ; the second Buka form is mani-
festly a devolution form from the former.

In the Malay asap we recur to the definitive compound of the Tongafiti,
as-ap, in which the latter component is again afi fire. The Malagasy seems
like an asu derivative.

In the Semitic there is seen the initial palatal which appears here in
Melanesian, the former vowel is the same, the s is present, and the final n
is found once at least in Melanesia. So far as this one stem is concerned
the resemblance is striking.

DATA AND NOTES. 289

240.

aue, interjection of surprise, commiseration.

Samoa: aue, alas, oh (wonder). Niue: aue, alas. Maori: aue,

alas (surprise). Hawaii: aue, auwe, alas. Mangareva: aue,

alas (surprise and grief) . Tahiti, Raro tonga, Paumotu : aue, alas.

Tongarewa: awai, id. Tonga: oiaue, alas. Aniwa: kawe, alas.

Futuna: uei, alas (indignation, surprise). Marquesas: aue, ue

(surprise). Fotuna: awe, alas. Rapanui: aue, aueue, ue, alas.
Aneityum: awe, auwe, iyauwe, alas. Nguna: ai, id. Mota:

awa, id. Eromanga: uwe, id. Tanna Weasisi: awe, id.

Bierian, Baki: awa, id.
Malay: ahi, ayi, ayue, alas.
Arabic: awwi, alas.

It is a sure identification from the Paumotu through all Polynesia,
through all Melanesia, through all Indonesia, and to the Semitic if one so
will. And to the cow, if one so will, for I have caught it in the calf-call of
lowing kine. And why should it not be identifiable, for the word lacks
all the elements of strength which part the word from the cry? There is
not a consonant anywhere in it, for the w is no more than a matter of tran-
scription ; it might just as well be u. Only in Aniwa do we find a consonant,
the deepest palatal and initial at that. To me it seems no more present
than is the initial m in the moo of cattle, a mere appulse.

What is this cry which our primitive islanders share with the animals?
Look at its elements, all full-throated. First we have a, the sound of
mouth open, fauces open, lungs full of air. As air expires the sound recedes
in the mouth toward the palate and we find the u. Last comes the con-
scious finish of the utterance, the muscles begin to retract, the sound-mak-
ing point is forced forward and the sound is e. If the man had but a few
more cubic centimeters of lung capacity he could attain cow volume for
his cry, or interjection, since it amounts to the same thing.

241.

bwala, bwela, bwola, to incline to, to be close to, to be stuck.

Samoa : pili, to be near, to approach, to be caught, entangled. Niue :
pili, to stick together. Hawaii : pili, near, close, to stick
together. Maori : piri, to stick ; pipiri, to be close together.
Tahiti: piri, to be squeezed or confined close. Rarotonga:
piri, to stick together ; akapiripiri, to get near. Mangareva :
piri, to stick together; akapiri, to patch, to glue. Paumotu:
pxripiri, glue ; fakapiri, to adhere. Rapanui : piri iho, to devote
oneself to; pipiri, glue, gum, sticky; hakapiri, to join to.
Marquesas: pit, to be joined together, to stick.

Motu : hebirihebiri, to sit or stand close together, as trees standing
close together; hebirimatemate , to be squeezed, crowded. Mota :
kpwir, to be close together; vakpwirkpwir, to crowd together.
Aneityum: bili, promiscuously.

Malay: ampiri, to bring near to; ambir, near, nigh.

Arabic: mala, mayl', to incline, bend, or lean to, to be close or near.

290 THE POLYNESIAN WANDERINGS.

The identifications here in Polynesia, Melanesia, and Indonesia are closer
than the Efate, yet that is not so far away as to be excluded. It will be
noticed that the places in which these Melanesian identifications appear —
Motu, Mota, New Hebrides — are those which go to establish the Viti track
which I have proposed as one of the migration courses.

In the Malay words the pili stem is recognizable in company with a
formative element. The m is manifestly not associable with piri-bir.

It must be upon these Malay words as transition forms that our author
relies in passing to his Arabic. This lacks permissibility.

242.
bwinu, bwin, bonge, to whistle.

Samoa: vivini, to crow. Futuna: vini, vinivini, to utter a cry of

alarm, of distress or of joy. Mangareva : vinivini, a soft sound,

musical, music. Paumotu: vinivini, the cry of a baby, to

chirp, to warble; hakavinivini, to whistle, to hiss at.

Malekula Pangkumu and Uripiv: puinpnin, ivinivin, to whistle.

Ambrym: mo fin, id.
Malagasy: enu, ncnu, manena, to sound, to crow, to ring, to sing.
Arabic: ma'anu, a singer; "aniya, to sing, to coo; "ina, song.
Hebrew: 'anah, to sing.
As to the homology of v-bm-pu see note 84 dealing with Efate bu. The
dialectic form bonge does not seem such that we may include it.

The word is singularly rare in its distribution. The Malagasy can not
properly be included in the identification, and the Semitic with its deepest
guttural for an initial is far remote from the island stem.

243.
bir i, to plait a string or rope.

Samoa, Futuna, Niue : fili, to plait, to braid, to twist. Tahiti : firi, id.

Maori: whiri, id. Mangareva: hiri, id. Hawaii: hili, id.

Mangaia: iri, id. Tonga: fi, id. Marquesas: fix, to twist.

vSolomon Islands: fili, a rope. Motu: hilia, to twist round and

round. Mota : vir, to twist, to wring, to squeeze with a twist,

to plait.

Malay: pintal, to twist. Beu: pilin, id. Batavia: bilan, id.

Arabic: fatala, fatV, to twist, to spin. Ethiopic: fatlat, id.

In the Tongan fi here included because of identity of signification we

encounter a form anomalous in the vanishing of the second of the stem

syllables. In fisi (223) it will be observed that a similar loss is observed in

Futuna, Niue, Viti, and the Tongafiti identifications. If in this light we

examine the Mota identification of that group, vivi, vivis, it may be proper

to separate them, vivis to go with the fisi stem, and viv to go (as abraded

vivi) with a fi stem. To that would also belong the Efate fif i and probably

the Samoan.

The remaining identifications are satisfactory, excluding Malay pintal,
until we reach the Semitic. This involves the triliteral ftl. To link this
with the fili stem requires that we account for the loss of I without having
left a transition form anywhere along the line, all the more difficult to
account for since, being centrally situated, it is protected from the common
casualties.

DATA AND NOTES. 291

244.

boro-aki, biro-aki, bero-aki, baro-aki, to bequeath to or order to do (by-
will when dying), to commission (one to do something), to give
orders to.

Samoa: poloa'i, to send a message to, to command a person at a
distance, to leave commands, as when going a journey or dying.
Futuna : poloaki, to order, to command, to bid farewell. (Niue :
poaki, pnaki, to command. Tonga : boboaki, to send a message.)
Hawaii : poloai, to send orders for one to come. Maori : poroaki,
to leave instructions when departing, to take leave. Ma-
ngareva : poroaki, to command, to order. Marquesas : pooai,
podhaki, to command, to entreat. Tahiti: poroi, a direction
given, a charge, to take leave. Mangaia: poro, last words.

Matupit : bor, to shout at, to scold harshly.

Malay: pasan, to commission, to enjoin. Malagasy: haf aha , a will
or testament, order.

Arabic: wasa', to bequeath by will, to command, to enjoin.

The stem is polo.

This is seen in a radical signification in Mangaia, the only case in which
the noun has been preserved. This sense of the last words, or, in the verb,
to impart dying injunctions, has as much importance to the Polynesians as
police attach to ante-mortem statements, a fetish in the performance of
their peculiar functions. He dies well in the Pacific who, upon fine mats
and with the members of his family seated gravely about him, can divide
among his kin not his possessions, for those are largely communal, but the
functions in life upon the successful performance of which he can look back
with pleased gratification.

This is so concrete an act that I incline to the belief that it is the gravely
underlying sense of the word. It is found as noun in Mangaia, as verb in
Samoa, where, however, the particular sense is in modern usage more defi-
nitely expressed in fa'amavaenga or parting words. A secondary sense, as
the word broke down, was to cover the farewell of any parting; this in
Samoa, Futuna, Maori and Tahiti. Next, with the idea still persisting of
the inexorability of the death-bed injunction, the word weakens still further
into the sense of a command ; this in Futuna, Mangareva, Tahiti and the
Marquesas, and, disregarding now the form anomaly, in Tonga and Niue.
All these languages have other words meaning to command. That this
command differs from other orders is felt in all these tongues; it is defi-
nitely expressed in Samoa "to command a person at a distance" and the
tantamount Hawaiian.

Except for Efate we do not find the stem in Melanesia. I have included
the Matupit form from New Britain because of its form resemblance; the
sense is so far awry as to make the identification quite doubtful.

In Efate and Polynesia, excepting Tahiti and Mangaia, we find that the
word has spread in its most highly finished form. In general the Polyne-
sian verbs in -aki impress me as a most modern development. The diffu-
sion of this one shows that it must have existed in this form at the time of
the expulsion of the Tongafiti swarm from Samoa. The form in Niue and

292 THE POLYNESIAN WANDERINGS.

Tonga is anomalous; we do not often find the excision in Polynesian of a
whole root syllable ; yet the sense and the identity of the remnant seem to
prove these forms of the polo stem.

Backward of Efate, save the Matupit resemblance, I recognize no identi-
fication.

245«
bwosi, to twist (a rope).

Samoa: fust, a belt, a girdle, to tie, to bind, to gird, to clasp, to
embrace. Tonga: fuhi, to fasten on. Niue: fuhi, a bunch,
to tie together. Uvea : fihu, to tie. Hawaii : hut, a bunch,
a cluster. Mangareva: huhui, a tied-up bundle of fruit, a
cluster, a bunch.

Viti: vusi, to suspend by a loop, to fasten, to tie up.

Nengone: wose, to bind; nawose, a band.

Malay: pusing, to twist, to turn round.

Arabic : 'afasa, to twist.

Because Dr. Macdonald has identified bwosi with fusi I have here expanded
the ramifications of that stem, the more particularly as it has Melanesian
affiliations. Yet I do not by any means accept the initial identification.
Principally from the sense I regard bwosi as homogenetic with fisi (223);
and the form variety is no more than we find in other Efate identifications
which we are quite ready to accept as valid.

Of Polynesian fusi we are to note that it is wholly Nuclear Polynesian,
the Uvea variant being readily resolvable as metathetic of the 1432 type.
This fusi we find in Viti, Nengone, and Malay. The Arabic is a triliteral,
*fs, and we can not account for the sacrifice of the initial.

The Hawaiian and Mangareva forms I have expressly included in order
that I might clarify a Niue peculiarity. They are not of the fusi stem;
their stem is fui, so found in Samoa and in the Polynesian of each migration
swarm. Niue fuhi is of the fusi stem in the sense of tying; to the cluster
(fui) sense it has assigned the fusi stem. This I can not regard as a transi-
tion form between the two stems, for neither s nor h undergoes extinction
in passing from Samoa to Niue. I take it as simply a Niue blunder. That
such blunders may happen I have already made sufficiently clear in Futuna
afu (239).

Dr. Macdonald equally identifies the Malay pusing with fisi (223).

246.
buma, funga, to flower, to blossom.

Samoa: funga, a flower, a blossom. Nukuoro: hunga, a flower.
Sesake : vunga, a blossom, a berry. Malekula Pangkumu : pung,

to blossom.
Malay: bunga, flower, blossom. Malagasy: vuni, flower; mamuni,

to blossom.
Arabic: fukah, flower.

It is impossible to avoid the impression that there lies somewhere a con-
nection between this relatively infrequent word and the widespread fua
(360) fruit. The consideration of this stem, therefore, will be involved,
so far as may be fitting, in the discussion of that item.

DATA AND NOTES. 293

247.
buto, the navel.

Samoa : pute, the navel.

Tangoan Santo, Lambell, Lamassa : buto, the navel. New Ireland

(Carteret Harbor) : buta, id. New Ireland (Duffield) : ambu-

tang, id. Laur, Pala, King, Malo: bito, id. Malekula

Pangkumu: bitou, id. Baki: burimbito, id. Buka: vuso,

bussusse, id. Mota: putoi, id. Tanna: nu-puti, id. Tagula:

bibido. Roro : botoa. Kabadi : puko. Mekeo : fuko. Motu : udo.

Pokau : mudo. Mukawa : puso. Suau : uso. Tubetube : pusua.

Awalama: buhoho. Wedau: buo. Nada, Murua: poso.

Sariba: post. Mugula: poasi. Panaieti, Misima: pohu.

Hula : buro. Rubi, Keapara, Galoma : bulo. Sinaugoro : ulo.

Malay: pusat, the navel. Malagasy: fuitra, id.

Arabic: bugrat, the navel, a knob.

With this should be included the general Polynesian pito navel. Samoa,

which has pute for navel, uses pito only in the sense of the end of a thing

and \pitopito for anus, this apparently for modesty's sake in preference

over pu.

Maori : pito, navel, end. Tahiti : pito, navel, navel string; pitoraoere,

the ends of a leaf surround in fishing. Marquesas : pito, navel,

navel string, end. Mangaia : pito, navel string. Mangareva :

pito, navel, end; pitopito, button. Paumotu: pito, navel;

pito pito, button. Tonga: bito, navel. Niue, Rapanui, Nukuoro,

Futuna: pito, navel. Hawaii, piko, navel, end.

I have no hesitation in associating the two as probable offshoots of

the same stem and as certainly alternate forms. I can not yet produce

any one of these languages in which the two forms appear side by side, yet

it is possible to present evidence but little inferior. We have four languages

recorded from a sixty-mile stretch of southern coast on the west face of New

Ireland, and three of these are so intimately associated that the differences

are but dialect variations. In fact before the casting of this note into its

final form the Pala speech has become accessible and shall stand as a fifth

consenting speech. In these five we find buto in Lambell and Lamassa, bito

in King, Laur, and Pala, all in the sense of navel. The conclusion is not

to be gainsaid that buto and bito mean navel and that Samoan is the only

Polynesian speech which has retained the buto form and thus avoids

confusion with pito the end, a confusion which exists in other Polynesian

speech, particularly of the Tongafiti swarm.

The possession of the 5-forms in Buka and New Guinea leads the way to
the acceptance of the Malay pusat, regarding the final t as due to local needs.
But we have no evidence for the Malagasy.
The Arabic is scarcely a resemblance.

248.
goro, koro, to snore.

Maori: ngongoro, to snore. Mangareva: ngoro, id. Paumotu:
ngooro, id. Hawaii: nonolo, id. Tahiti: ooro, id. Niue:
tungolo, id. Samoa, Tonga: t&ngulu, id.

294 THE POLYNESIAN WANDERINGS.

Viti : nggonggori, a man that a god enters when asleep, indicated by

a singular kind of snore.
Motu : uru, stertorous breathing. Mota : ngora, to grunt, to snort,

to snore.
Malay: ngorok, to snore.
Hebrew : nahar, to snort, to breathe hard through the nose. Syriac :

nhar, id. Ethiopic: nehera, to snore.

In Nuclear Polynesia we find the stem in association with an element
whose function we can not explain, ta in Samoa and Tonga, tu in Niue.
The remaining identifications, including the Malay, are free of difficulty.
The Paumotu ngooro is a rare form of duplication ; that it is a case of pre-
duplication is established by comparison with the Maori.

The Semitic exhibits the usual difficulty in that it calls for the reduction
of a triliteral, nhr, by internal loss.

249.

kita, to divine; kikita, ngkita, to perceive or feel with the eye or mind.

Samoa: 'i'ite, to predict, to foretell; fe'ite'itea'i, to see indistinctly
as in the twilight, to be just distinguishable. Tonga : kite, to
appear, to see at a distance when at sea; kikite, fakakikite, to
divine, to foretell, to augur, to prophesy; jekitengaki, to be in
sight of each other; jakakite, to look anxiously and narrowly
at anything, as at what may be reported (land at sea or a vessel
in sight) ; fakakitekite, anything new or strange done or said by
a person just before his decease and afterwards referred to as a
prognostication. Futuna: kite, to appear; kikite, to predict,
to foresee, clairvoyant. Niue : kitekite, to look at, to examine ;
jakakite, to make known, to show; fakakiteanga, a vision,
Uvea: kikite, fakakikite, to augur, to divine, to prophesy; kite.
to appear. Maori : kite, to see, to know, to perceive ; whakakite,
to reveal, to disclose. Mangareva : kite, to see, to understand,
to perceive. Rarotonga : kite, to see. Marquesas : kite, to
see, to know ; haakitea, to appear. Paumotu : kite, to perceive,
to know; jakakite, to show, to presage. Tongarewa: kite,
kikite, to see ; hakakikite, to cause to see. Nukuoro : kite, to see ;
hakakitea, to show ; matakite, a soothsayer, a prophet. Manahiki :
kitea, to know. Sikayana: kite, to see. Aniwa: citi, id.
Tahiti : ite, to know, to understand, to perceive ; jaaite, to teach.
Hawaii: ike, to see, to understand, to perceive; hoike, to show.
Motu: kito, to spy, to watch for.

The following mean to see:

Panaieti: kite. Sariba: kita. Galavi, Boniki, Mukawa: kitai.

Sariba : gita. Sinaugoro, Tubetube : gitai. Motu, Rubi, Kubiri,

Kiviri: itai. Doura: ikai. Suau, Dobu: ita. Roro: itana.

Raqa: iti. Oiun: itin. Pokau: ikala. Hula, Keapara: gia.

Galoma: ia. Kabadi: is'ana.
Tagalog: quita, to see. Malagasy: hita, mahita, id.
Arabic: wagada, to find with the eye or the mind, to perceive.

DATA AND NOTES. 295

I suggest an identification in Viti ndike to look at, to scrutinize. This
is permissible if we regard this as metathetic and in the 3214 type, a rarity
in metathesis, yet exactly paralleled in the same language in uila-liva.

If it were not for its occurrence in Efate and in Motu I should class this
word as of the Tongafiti migration, for in very few instances do we find a
stem so well nigh universal in Polynesia which has left vestiges so rare and
so widely scattered in Melanesia. It seems quite as rare in Indonesia.

The Semitic involves Dr. Macdonald's favorite principle of whittling
down a triliteral (wgd) in any way which will fit it to his proof, in this
instance by excision of the first syllable.

250.
kufangufa, to fly, to flap the wings, to flutter.

Tonga: kapakau, wing; kapakapa, the side fins of sharks, to flicker,
to flutter, to hover on the wing. Futuna: kapakau, wing;
kapakapatau, the movement of birds about to perch. Niue :
tapakau, wing, fin ; kapakapa, to flap, to flutter. Uvea : kapa-
kau, wing. Maori, Mangaia: kakapa, kapakapa, to flap, to
flutter. Samoa: 'apa'au, wing; 'apa'apa, fin; 'apata, to clap
the wings. Nuguria: kapaukau, upper arm. Tahiti: apaapa,
to flap. Fotuna, pahkau, wing. Rotuma: papau, id.
Ulawa, Saa, Bululaha, Ugi : apaapa, wing. Wango, Alite : abaaba, id.
Mota : gava, to fly with flapping wings. Fagani : kakafo, wing,
Marina: gave, id. Arag: gapaun, id. Pak, Sasar, Alo
Teqel: gapugi, id. Savo: gavara, id.
Magindano: kapakapa, a fan. Malagasy: kepakcpaka, flounced in
the wind. Malay: kapak, to fly with flapping wings; kepak,
wing. Kawi: paksa, bird. Basakrama: paksi, id. Magin-
dano: papak, a wing. Baliyon: papak, a bird. Tagalog,
Bicol: pacpac, wing.
Arabic: h'afaka, to fly, to flap the wings.
In Polynesia gliding flight is expressed by lele, flight on flapping wing
by kapa. In Nuclear Polynesia kapa does not pass into the wing sense
except through the aid of a composition member kau. In Samoan 'au we
find this to mean a stalk, a handle; in reference to the body its sense as
that of some projecting member is exhibited in 'aualuma (the 'au in front)
as a very delicate euphemism for the penis. So 'apa'au would mean liter-
ally the projecting member that flaps. We encounter kapakau in Tonga,
Futuna, Uvea, and in Samoan 'apa'au. In Niue we find a change that can
only mean failure to comprehend and to preserve the primal signification
of the composite, for, while kapakapa is to flutter, tapakau shows a modi-
fication that has no reason either in sense or in phonology. I am no little
doubtful as to Rotuma papau. But as that interesting language has many
irregularities which as yet quite elude systematic reduction I deem it
advisable to class papau with kapakau; it does seem an echo form.

Fotuna pahkau shows a frontal abrasion of kapakau. This is character-
istic of the Tongafiti word for wing.

Maori: pakau, wing. Moriori: pakau, arm. Tahiti: peheu, fin.
Marquesas: peeheu, pekeheu, wing. Mangareva: pehau, wing.
Hawaii: peheu, wing, fin.

296 THE POLYNESIAN WANDERINGS.

There is a uniformity about these Tongafiti words which indicates a
reason acting upon them all which was not in action upon the Proto-
Samoan swarm, and we see that the change is progressive in the Tongafiti.
As I feel it this reason is that by the time the Tongafiti were ready to leave
Indonesia they had lost all recollection of kapakau as the flapping member,
and being a mere vocable used uncomprehendingly as a label for a natural
object it was subject to the humors of speakers in making changes. Thus
we may comprehend the changes in Indonesia. Malay, Magindano and
Malagasy alone retain the first syllable. All the others received the word
when that syllable was gone, received it at the same time probably, received
it certainly from the Tongafiti after they had reduced the word. The
Magindano papak wing and Baliyon papak bird smack more than a little
of Rotuma papau.

Next, when we turn to Melanesia, we find forms variant only of the kapa
stem. Mota, Marina, Arag, Pak, Sasar, Alo Teqel, Savo, and Fagani all
retain the initial palatal. It is only in the Solomons that we find this
vanishing, and the mere dropping of the k is yet a long way from the
dropping of the first syllable entirely.

The Arabic has more than we can digest for our plain kapa, and even
were we to admit the propriety of Dr. Macdonald's method of obtaining
pleasing results by dropping out inconvenient elements the consonants of
the Arabic stem, hfk, which might be identifiable with kp, appear in the
inverse order.

251-
kusue, kusuiie, rat, mouse.

The following words all signify rat :

Samoa: 'isumu. Nuguria: kisumu.

Sesake: kusuwe. Mota: gasuwc. Vaturanga: ngasuve. Fagani:
gasufe. Wango: gasuhe. Lo: gahuwa. Merlav: gasuw.
Sasar, Vuras, Alo Teqel, Norbarbar : gosow. Pak : gosog. Gog :
gosug. Tanna : yasuk. Mosin : gusuw. Motla v, Volow : gohow.
Lakon: wohow. Alite: nguaua. Ulawa, Saa, Bululaha:
asuhe. Malekula Pangkumu : asua. Nggao : kusi. Nggela,
Bugotu : kuhi. Savo: kuzi. Malekula: khasup, akasu.

Malay, Java, Baju: tikus. Massaratty: tikuti. Gah: karufei.

Arabic: kutrub' .

This is a peculiarly interesting word, for, if Polynesian, it seems to be a
form that in the travel from Indonesia was widely disseminated by the way
and yet not carried the whole distance.

The Samoan 'isumu differs in one element from kusuwe, and nowhere in
the many forms in Melanesia do we find a trace of this m. Yet the two
earlier syllables in Samoan 'isumu and Nuguria kisumu strongly indicate a
connection with the Melanesian in the backward past. Samoa has three
rat names. It has this 'isumu; it retains, but little uses, the Tongafiti
'iole; it employs most frequently the word 'imoa, which is identifiable only
in Nukunau.

Codrington (Melanesian Languages 87, note) says that "the old black Fiji
rat is ngatho," a clear equivalent of kusuwe. This is not in Hazlewood's

DATA AND NOTES. 297

dictionary, but Dr. Codrington had the benefit of notes on Viti by Lorimer
Fison, a most competent authority; therefore this does not contravene but
supplements the earlier dictionary.

In the order in which I have arranged the Melanesian material it will
be seen that there is a simple and easily followed sequence down to Alite,
except for three items which need comment. In Gog -g suffixed to a stem
determines its use as a noun, the same holds true of Pak; the two taken
together argue the same explanation of the Tanna form. These, therefore,
are not to be taken for a mutation to the final palatal. So little is known
of Alite that we are not in a position to judge nguaua satisfactorily. I
have examined all my material carefully with a view of identifying con-
sonant mutations between Alite (and its neighbors Saa and Bululaha) on
Malanta, and Vaturanga across the strait, and to no result. It will be
plain that ngua echoes Vaturanga ngasuve, thus ng(a)(s)u(v)a; but this
wholesale amputation of vowel and consonants too much resembles the
freehand proof of Semitic origins to meet with cordial approval. Malekula
khasap and akasu belong somewhere in this sequence which retains the
initial consonant. So, too, do Nggao, Nggela, Bugotu, and Savo, in which
the final syllable has been abraded. We find a small group in which the
initial consonant has been abraded, Ulawa, Saa, Bululaha, and Malekula
Pangkumu; and in the last the third consonant has vanished, although
under the protection of a final vowel, but compare fua (360) Malekula
Pangkumu mi uan for a vanishing /.

The Indonesian offered by Macdonald and by Codrington does not seem
susceptible of coordination. The Arabic is, of course, out of the question.

252.
laso, the testicles.

Samoa: laso, the scrotum. Tonga, Niue: laho, id. Hawaii,

Nuguria : laho, the testes in man and animal. Fotuna : raso, id.

Tahiti: raho, pudendum muliebre. Maori: raho, the testicles.
Ambrym: luho, testicles. Paama: as'i, id. Malekula: list,

erasi, raso n, id. Bierian: loho, id. Malo: laso, id. Mota:

lasoi, the male genitalia.
Macassar: laso, penis.
Arabic: h'isy', h'usy', h'usyat, h'usa', testicles.

It is quite uncertain what was the primal sense of this stem. In Samoa,
Tonga, and Niue it distinguishes the scrotum, and in Samoa the testes are
designated by the name fua (360) fruit, to which Codrington assigns the
root sense of anything globular. This use of fua extends into eastern
Polynesia. Tahiti: hud, testes. Marquesas: hud, genitalia in general.
Hawaii: hua, testes. Mangareva: ua, genitalia.

In the area of Tongafiti colonization (Futuna therein inclusible) and in
Melanesia wherever the word is identified, laso has passed definitely from
the scrotum to its contents. In the latter subdivision Mota exhibits for
laso the same comprehensiveness that fua exhibits in the Marquesas and
Mangareva, if it be not too violent an interpretation of our vocabulary
definitions to assume this to apply only to the male parts. The switch of
sex in Tahiti raho is unexpected ; the entry is derived from Tregear's com-

298 THE POLYNESIAN wanderings.

parative dictionary without opportunity to check his record by consul-
tation of his original source; in Bishop Jaussen's dictionary the word finds
no entry.

In Macassar the word passes to yet another part.

The only resemblance which the Arabic bears to laso is that its forms
also contain a modification of the radical s as the second consonant, no
sufficient proof of identity.

253-
ma, me, with, and.

Samoa: ma, with, and. Tonga, Rotuma : ma, and. Futuna: mo,

with. Niue: ma, and; mo, and, with. Uvea: mo, with.

Tahiti: ma, and, with. Hawaii: me, with; a me, and. Rapa-

nui, Marquesas : ma, and (in a particular numeral use) ; me, with.

Mangareva : me, with, and . Paumotu : ma, together with ; me,

with. Rarotonga: ma, and. Maori: ma, and; me, with, and.

Nuguria, Aniwa: ma, and. Fotuna: ma, for, with, along with.
Mota: ma, me, with, and. Santo: we, with: wo, and. Pala: ma,

and, with.
Malagasy: amana, with.
Hebrew: 'im, with, and. Arabic: ma', id.

I can not do better in comment upon this item than to repeat the ana-
lytic conclusions which I reached in an earlier study of root reducibility
(27 American Journal of Philology, 389) :

Let us now look at the root ma. In its paradeictic function we find it serving as a
connective; it is the spoke that joins tire and hub into the effective unit of the wheel.
It is the conjunction "and," yet its development is in a dual sense incomplete; it is
available to connect words of the same grammatical function; it has not yet become
sufficiently conjunctive to link clause with clause. At the same time another function,
that which we know and employ as prepositional, exhibits in the sense of "with," "for"
(for the sake of), thence differentiating to "from" and "on account of"; these different
uses we in analytic speech find it necessary to distinguish by varying words; to the
Samoan it is sufficiently clear to use ma and trust to inference from existing conditions
to elucidate the character of the relation the existence of which is thereby indicated.
Stated in terms coordinate with those employed in the preceding particulars of this
series of roots, we may say of ma that it points to the non-ego and not-here and links it
to the central concept of that which is active and present.

In the elaboration of my theory of explanation in the paper from which
the foregoing is extracted I dissected the central signification of the a and
sketched out the coefficient value of the several consonantal modulants
which might be prefixed thereto. In dealing according to that theory with
the other forms associated in this item, me and mo, we should hold the
consonantal value as carrying the linking, conjunctive, associating sense; the
shade of variety in meaning would be found to exist as the nucleus of the
e and of the 0 respectively.

254-
manifenife, to be thin.

Samoa, Nukuoro: manifinifi, thin. Tonga, Futuna : manifi,manifi-

nifi, id. Uvea: manifi, id. Fotuna: mafinfini, id.
Omba : m anivinivi, thin . Mota : mavinvin, id . Baki : m enivinivi, id .
Malekula :mcn iven iv, id. Malo: tanivinivi, id. Roro: nivinivi,

DATA AND NOTES. 299

id. Mekeo: mangipina, id. Galoma: magipi, id. Keapara:
magivi, id. Kabadi: kevekeve, id. Galavi, Boniki: kavaka-
vana, id. Taupota: wogevagevana, id. Awalama: wogewoge-
wana, id. Wedau: avavana, id.

Malagasy: manifi,thm.

Arabic: nahifa, nahafat, nahif,' manhuf,' thin, slender.

The word exists for us only as a composite, for the element ma is most
clearly the prefix of condition which so frequently occurs ; the stem nifi is
nowhere apparent, either independently or in other composition. Simple
metathesis occurs in the Fotuna and Mota forms. The word is confined
to the Proto-Samoan. The Malagasy identification is far better than is
commonly the case with that speech. Macdonald's Semitic identification
entails laparotomy of a syllable protected by its inner position.

255-
maso, main', to be cooked, done.

Futuna : moso, cooked. Tonga : moho, cooked, ready to be eaten,
rotten ; momoho, ripe, brown in color, ready to burst. Niue :
moho, cooked, done; momoho, ripe, mature. Uvea: moho,
cooked.
Santo: mda, cooked.

Malay: masak, cooked. Malagasy: masaka, id.
Arabic: nas'iga, to be ripe, cooked.
The Proto-Samoan stem is absent in Samoa, for the reason that Moso is
the name of one of the great terrestrial gods. The mere suggestion of
cookery, the plaiting in coconut leaves and the slinging on a pole like a
pig ready for the oven, added the pitch of indignity to the Samoan ifonga
or solemn rite of submission. It must therefore be clear that moso would
never be permitted in the land where Moso was revered. How little Moso
would put up with any such disgraceful kitchen verb in the very accents
of his name may be inferred from one of the prayers addressed to him:
"Oh, Moso, make haste, show thy power, send down to down-below-here,
sweep them away like a flood, may they never see the light of another
day!" Oh, no, it's not strange that moso does not mean cooked in Samoa.
I can not accept Macdonald's Santo identification. This involves the
dropping of s from the inner and protected position. This is not only
objectionable in itself, but is contraindicated by asu (239) smoke Santo osun,
and except as to position by hifo (206) down Santo siwo.

The Malay and Malagasy may be sound ; in the dearth of other instances
in that area we may not accept them definitely.

Dr.Macdonald undoubtedly relies upon mahi, a dialectic form in Efate and
nowhere else found, and upon the k in the Malay and Malagasy, to bolster
up his Semitic identification, which furthermore involves the m-n mutation.
This is too much to rest upon premises so insufficiently established.

256.
meta, raw, unripe, crude, green.

Samoa, Tonga, Futuna, Niue, Maori, Mangaia, Tongarewa, Rapanui,
Mangareva: mata, raw, unripe. Hawaii: maka, id. Fotuna:
mata, raw. Nuguria: oimata, raw; koimoto, unripe.

300 THE POLYNESIAN WANDERINGS.

Aneityum: emetmat, raw, not dry or seasoned; mat, new, raw.
Bierian: nmata, raw. Tanna: (t)e'mta, id.

Macassar: mala, raw, unripe. Malay: matang, mantah, id. Mala-
gasy: mania, id.

Arabic : 'anut'a, to be raw.

The dissimilant duplication of Aneityum emetmat establishes perfectly
the unity of Efate meta and the mata which belongs to the Polynesian of
both migrations. The other Melanesian forms are readily recognizable, and
in Indonesia the series is confirmatory inter se. The Arabic is, of course,
entitled to no consideration.

257-
miel, mimiel, red.

Samoa: melomelo, memelo, red. Tonga: melo, melomelo, brown,

ripe. Futuna: memelo, red. Hawaii: meomeo, omeomeo.

red, orange, blushing. Mangareva: metometo, yellow, orange.

Rarotonga : muramura, red. Bukabuka : kura melo, light red.
Mota: mera, red light in morning or evening sky. Vaturanga:

mera, yellow. Wango : meramera, red. Fagani : merameraga, id.

Baravon: mera mere, id. Bugotu: mela, id. Buka: mar ara, id.

Laur: mirik, id. Sesake: miala, id.
Malay, Gah, Baju: merah, red, bay. Bouton: merai, red. Awaiya:

meranate, id. Salibabo: maramutah, id. Malagasy: mena, id.
Arabic: ma"ir', reddish; 'am"aru, of the color of red clay.

In the Polynesian the stem ismelo. Its appearance as meto in Mangareva
is unusual, but not without the precedent of an l-t mutation. The Raro-
tonga mula is a valuable transition form in respect of the final vowel, to
link the Polynesian and the Melanesian. In the Bukabuka locution we find
melo used in limitation of the more widely diffused word for red, kura.

In Melanesia the stem is mera, with which mar a in Buka mar ara is readily
associable; while Baravon merarnere naturally leads to Laur mirik.

In Indonesia also the stem is mera, both independently and in composi-
tion, and mara also appears. Malagasy mena is more remote yet not
improbable.

We have yet to discuss two puzzling forms, Efate miel and Sesake miala,
which are evidently in close consociation. If it were not for the intrusive i
they would fall into ready alignment with the mela-mala stem of Melanesia,
but in the present stage of our knowledge we are quite at a loss to account
for the intrusion of the vowel. We may feel that these anomalous forms
derive from the common stem, but we have no proof which we may bring
to bear. The preduplicated Efate mimiel shows that the vowel i is no mere
accident, but is regarded as an essential part of the syllable mi, and therefore
structural.

To bring our author's Semitic into alignment calls for the evisceration of
an interior consonant, the strongest palatal at that and perfectly supported.
Even if that were permissible, which is open to grave doubt, the Arabic
mal'ir would be brought only imperfectly into likeness with no more stand-
ard a form than the miel-miala which we have just seen to be wholly
unaccountable deviations from regular stem forms.

DATA AND NOTES. 301

258.
mitaku, mataku, mitau, matau, to fear, to be afraid of.

Niue: matakutaku, to fear; matakumataku, dreaded, inspiring fear.
Maori: mataku, to fear, to be fearful, inspiring fear. Uvea,
Fakaafo, Nukuoro, Aniwa, Fotuna,Vate, Rarotonga, Mangareva,
Bukabuka, Manahiki: mataku, to fear. Rapanui: mataku,
alarm, dread, fear. Samoa: mata'u, to fear, to be afraid.
Tahiti: matau, fear, dread, to be in terror. Hawaii: makau, id.
Marquesas: hadmetau, id.
Sesake, Bierian: mataku, to be afraid of. Malo: matacu, id. Arag:
matagu, id. Nggela : matagu, fear, to be afraid of; mamataguga,
fearful. Belaga: matagu, to fear; mamataguga, fearful. Mota
Maligo: matagut, to be afraid of; tagut, to be startled. Fagani:
maguta, afraid. Mota Maligo: matagtag, to fear. Motlav:
metegteg, fear. Ambrym : matehag, to be frightened at. Vatu-
ranga: matahuni, to fear; matahu, fear. Malekula: metoh, id.
Aneityum: i-mtac, afraid, timid, cowardly; imiimtac, to fear,
to reverence. Suau, Tubetube : matausi, id. Dobu, Wedau :
matauta,id. Oiun: matautei, id. Sariba: matousi, id. Tavara,
Awalama : matouta, id. Pokau : maka'u, id. Kabadi : mekau, id.
King: matiit, to fear. Tanna: meheker, id. Lemaroro : matau,
afraid. Baki: merou, to fear. Tagula: marode, id.
Malay: takut, fear. Malagasy: tahutra, fear; matahutra, to fear,

to be afraid.
Arabic: taka', to fear; takiyyat, fear, caution.
The Proto-Samoan stem is matakut, and it is not only in form a com-
posite of the ma of condition with takut, but we have the stem preserved
in independent existence in Mota Maligo, in Malay, and in Malagasy. In
the Mota tagut to be startled, taken in conjunction with Uvea mataku a
trembling with fear, we may find the primal sense of the stem, the quivering
of the body in expectant poise to seek refuge in flight when some unwonted
noise in the forest home has at last become identified with danger too
overpowering to face.

The final t has vanished except in Samoa objective aspect mata'utia, where
it is protected, in Mota Maligo matagut, and in the Indonesian forms. The
stem vowel u remains in Mota Maligo, Sesake, Bierian, Malo, Arag, Nggela,
Belaga, Fagani, Vaturanga. By abrasion of that vowel we find a new
closed form in which the final stem syllable has disappeared ; Mota Maligo
matagtag, Motlav, Malekula, Aneityum. The following irregularities call
for comment. In Nggela and Belaga occur forms involving the duplication
of the conditional ma; this is of very rare occurrence and may perhaps
argue that the mataku stem was so archaic that the recollection was lost
that this was the ma of condition. This mamatagu in these two Solomon
Island languages may be understood as introducing a g in the stem /-place ;
Ambrym matehag seems to do the same ; Vaturanga matahuni has an n in
that place. We have not sufficient information as to these languages to
warrant venturing an opinion as to whether these are mutations of the
stem consonant or local devices of word formation. I have refrained from
including Aneityum imiimtac with mamatagu as duplication of the condi-

302 THE POLYNESIAN WANDERINGS.

tional prefix, for we recognize in imi a verb-formative prefix of a causative
value. Fagani maguta is metathetic from the neighboring Bugotu matagu.
To associate King matut with the matakut stem involves the extirpation of
an inner and therefore abundantly protected syllable, a syllable of a type
(ak) to class which as a syllable would be doing violence to the whole
structure of the Polynesian word ; furthermore I can find in the brief King
vocabulary no other word of Polynesian resemblance which depends upon
such a procedure. Against these valid objections we raise in support of
matut only its resemblance. Tanna meheker is another case of resemblance.

These resemblances are by no means without their value. On the theory
that the Polynesian content of Melanesian is loan material we should expect
to find for many words no more than the ghost and echo. On a far higher
culture plane it is notorious that Hodge, when he marries according to the
ordinances of the Church of England as by law established, solemnly avers
"with all my worldly goods I thee and thou," the echo of a word of his
native language.

To establish Lemaroro viarau and Baki merou as even a resemblance it
will be necessary to maintain the t-r mutation ; this is found in our material
four times: in fatu (294) stone Malekula var, talinga (350) ear Malekula
Pangkumu riringa, ate (276) liver Malekula Uripiv ere, futi (329) banana
Moanus mbur. If this be considered sufficiently well established we find
in marau a variant of the Efate matau, a reduced form involving the loss
of the k, which nowhere else appears until we encounter it ultimately in
extreme eastern Polynesia.

The Indonesian identifications have already been mentioned as preserv-
ing the primal stem.

In the Semitic offered for our consideration the verb has a partial resem-
blance to the takut stem, the t being represented by a palatal. The resem-
blance of the noun is more specious than real, for it is hard to see why, when
there is already a verb or bare stem, we are to accept an almost inflected
form as a noun to enter upon a new course of activity in the Pacific as a

verb.

259.

nabati, tooth.

vSamoa, Tonga, Niue, Futuna, Uvea, Fakaafo, Fotuna, Vate, Moiki:
nifo, a tooth. Nukuoro, Aniwa, Maori, Tahiti, Hawaii, Mar-
quesas, Mangareva, Manahiki, Rapanui, Paumotu: niho, id.
Sikayana: nitcho, id. Mangaia, Rarotonga: nio, id. Nuguria:
ngiho, id.
Ulawa, Saa, Bululaha : niho, tooth. Buka : niho, liho, lihon,uliho, id.
Lifu: nyo, id. Iai: niou, id. Fagani: lijo, id. Alite,

Vaturanga, Nggela, Bierian, Epi: livo, id. Arag: liwo, id.
Mota, Maewo: liwoi, id. Motlav: lewo, nc-lwo, id. Ugi,
Bougainville: liho, id. Wango: riho, id. Mukawa: nibo, id.
Awalama: niwo, id. Tavara: niuwo, id. Taupota, Wedau:
ivo, id. Roro: nihena, id. Kabadi: nise,id. Pala: ngise, id.
Motu: hise, id. Mekeo: ni'e, id. Pokau: nike, id. Doura:
ike, id. Uni: igeo, id. Galavi, Boniki: oke, id. Panaieti:
ni, id. Tubetube, Misima: nini, id. Tagula: nungi, id.
Makekula: riho, id.

DATA AND NOTES. 303

Matabello: nifoa, tooth. Saparua: nio, id. Malagasy: nify, id.
Guam: nifin, id. Manatolo, Sula : nihi, id. Bouton: racfti, id.
East Vaiqueno: nissy, id. West Brissi: nissin, id. Savo:
nuhsi, id. Kisa: nihan, id. Kayan: knipan, id. Magindano:
nipun, id. Tagalog: ngipin, id.
Arabic: 7ta6', nnbiib', tooth.
Dr. Macdonald does record that this is &a/i (39) with the article, and he
does set it down that Polynesian nifo is "another word for tooth, teeth,"
and he includes his Semitic writhin the same brackets as the lower set of
these teeth. But why does he collate the nifo material with this nabati if,
despite his protestations, he did not think it added to his Semitic scheme?
The w-form which holds without exception through Polynesia is found in
the western Pacific only in Ulawa, Saa, Bululaha, and Buka, stages in the
northern Solomons on the Samoa track, and in Lifu and Iai at the extreme
south, the terminus of a migration concerning which Codrington notes
(Melanesian Languages, 17) "more archaic they well may be, belonging to an
earlier movement of population, carried forwards by an earlier wave of
speech passing onwards among the islands, but having somewhere a com-
mon origin with those which have since and successively passed among
them." Yet in the instance of this word we observe as a curious fact that
they most closely resemble Mangaia and Rarotonga of the Tongafiti swarm,
known to be the most modern of the successive waves of the speech.

Buka with its niho and liho is the sure identification of the transition
phase of the passage to the /-forms which are so characteristic of Melanesia.
These forms differ in the second consonant from the primal / in Fagani lifo
and the immediately proximate h in Ugi and Bougainville liho and Wango
riho, all these in the Solomon Island crop colonies of the Samoa track ; up
the labial column to v in Alite, Vaturanga, Nggela, Bierian and Epi (if this
be not a reference to the Bierian) ; thence to the semivowel approximating
the labials, w, in Mota, Motlav, and Maewo; finally down the column to
b in Malekula ribo. So far, in all Polynesia and in all Melanesia, the two
vowels have remained unaltered in quality and fixed in position.

When we examine the Indonesian retention of the nifo stem we shall find
but one of the four elements of the word which has been treated with care,
namely, the former vowel, which preserves its place unaltered in all except
Savo. Matabello alone has nifo and has added a decoration of its own;
Saparua has the form which we have seen in the Loyalties and in Mangaia
and Rarotonga. There are here no /-forms; the n is found intact in Mata-
bello, Saparua, Malagasy, Guam, Manatolo, Sula, Bouton, East Vaiqueno,
West Brissi, Savo, Kisa, Magindano ; in Kayan it is prefaced by a palatal,
and in Tagalog it has become the palatal of its own group. The second
consonant is kept as / in Matabello, Malagasy, Guam. It becomes h only
in Manatolo, Sula, and Kisa, and in Bouton nichi it follows the same change
as in Sikayana nitcho. It passes irregularly to s in East Vaiqueno and West
Brissi, and to a gruffer sibilant in Savo. This mutation we explain as
meaning that the first change is to an aspirate lying close to the labial posi-
tion of the buccal organs; then that the mouth can not hold in this position
so amorphous a sound as this aspirate; that the aspirates lying near all
three series tend to assume the position of the central, or lingual, aspirate;

304 THE POLYNESIAN WANDERINGS.

thence easily pass into the sibilant, regardless of the source from which the
aspirate derived. In eastern Indonesia the mutation down the column to
the p of Kayan, Magindano and Tagalog correlates the b form in Malekula.
The final vowel, o in Matabello and Saparua only, is i in Malagasy, Guam,
and all the rest save only Kisa and Kayan, in which it becomes a, and in
Magindano which has u. An ephelkustic n is found in Guam, West Brissi,
Kisa, Kayan, Magindano, and Tagalog.

To bring the Arabic in nab' into identity with the nifo stem requires but
the establishment of the following laws of mutation:

(i) That a shall pass into i without a vestige of transition form.

(2) That b shall represent / against the universal practice of the Poly-
nesian users of the word, and the occurrence of this change among the less
careful Melanesians and Indonesians only in a single instance among the
former and but three times among the latter.

(3) That a final palatal, a strong guttural, shall drop off unnoticed.

(4) That an o strong enough to hold throughout the Pacific area and to
be represented by some substitute in all of Indonesia shall be acquired
somewhere and somehow.

The initial n, however, withstands all hostile assault ; it is there in the

Semitic.

260.
ra na, a branch.

Samoa: la, a branch of a tree; laid, small branches. Nukuoro:

la 0 te manu, a branch. Niue: la, laid (plural), branch of a

tree. Hawaii: lala, limb or branch. Futuna: lad, branch.

Maori: rara, a twig, a small branch. Tahiti: rara, a branch;

ara, small twigs or branches. Rarotonga, Mangareva: rara,

branch. Fotuna: ra, id.

Sesake: ndara, a branch. Aneityum: in-ran, id. Bierian: la, id.

Malay: daan, a branch. Malagasy: rahana, rahaka, id.

Arabic: s'agnat, s'agan', a branch.

The Polynesian stem is found in Efate, Aneityum and Bierian, and in

Malagasy. In all this range we find no transition forms to account for the

d which occurs in Sesake and in Malay. Yet an l-t mutation is by no means

unknown ; in fact it occurs in the next item.

I fail to see on what score, even of mere resemblance, it is sought to

include this Semitic.

rai, re, the forehead, aspect, face.

Samoa: lae, the part between lip and chin without hair; ta'alaelae,
a wide or bald forehead, a beardless chin. Tonga, Futuna,
Uvea, Nuguria, Hawaii : lae, the brow, the forehead . Sikayana :
moalae, id. Niue: le, matale, muale, id. Maori, Tahiti,
Paumotu, Rarotonga: rae, id. Mangareva: raemata, the face,
the countenance ; akarae, to cut the hair on the forehead ; korae,
to cut the hair of women on the forehead ; akakorae, to cut the
ends of the hair short behind. Rapanui: korae, the brow.

Nguna: rae, the front. Malo: rai, forehead.

Malay, Matu: dai, the brow, the forehead. Java: rai, id.

Ethiopic : rey, the sight, aspect.

DATA AND NOTES. 305

The stem signification relates to something which, by being clear, comes
into prominence from out of its surroundings. This we see from the refer-
ences to the hair in Samoa and in Mangareva. In Niue matale and Ma-
ngareva raemata we have the same two elements ; the meaning is the clear
spot of the face, and in a people so bushily bearded as are the men of Niue
this must be localized upon the brow so long as the primal significations per-
sist. In Niue and Sikayana we find in muale and moalae composites which
define the forehead as the bare spot in front. The town green, malae (315),
is in form a conditional derivative of this stem. The sense is equally
clear, for the malae is distinctly a place clear of growth and of the habita-
tions of men; it must be clear in order to fit it to be the stage of the public
activities of the simple island life.

The Melanesian identifications are satisfactory, but in the brevity of our
vocabulary material we have no means of knowing if aught of the primary
sense here persists. The alternative Efate re finds a parallel in Niue le,
the change of ae to e being notably common in the latter tongue (17 Journal
of the Polynesian Society, 91).

The Indonesian identifications, though few, are found in the strongest
languages of that area and are acceptable, continuing from the preceding
item the note as to l-t mutation.

The Semitic proposed is at least in form a resemblance, but the sense is
widely remote from the stem signification of lae.

262.

rau, rarau, to grope for with the hand, to seize, to snatch out or away.
Tonga: lau, lalau, lauji, to pinch with the fingers, to nip. Hawaii:

lau, to feel after a thing; lalau, to extend (as the hand), to

seize, to catch hold of.
Mota: rau, to thrust the hand into a bag; raun, to thrust in the

hand and take out something. Aneityum: rap, raprap, to

grope for.
Malay : raba, to feel for, to grope ; rawa, to handle.
Arabic : lamaa, to feel for, to grope, to take away.

It is by no means certain that Tonga lau is herewith correlated, for in that
the sense is particularly to nip and in the Efate, Mota, and Hawaii to grope
for and take hold of. The Tonga stem is laut, and this does not elsewhere
appear. In Mota raun it is explained that n is a suffix to make verbs defin-
itely transitive; yet, as no less than seven of the Mota consonants are
stated to be thus employed indifferently, I incline to view this n as radical,
the stem being laun.

Aneityum rap involves a most infrequent mutation from the vowel u to
the labial mute. The sense is satisfactory, but this change is not to be
accepted until a better knowledge of this crabbed tongue shows this muta-
tion to be normal though violent.

The same consideration casts grave doubt upon the validity of the Malay
raba.

Of course the Arabic lamaa is too remote for attention.

306 THE POLYNESIAN WANDERINGS.

263.
soa (saisai), so, companion, follower, especially of the opposite sex.

Samoa, Futuna: soa, friend, fellow, companion. Ticopia: soa,
friend. Sikayana: tosoa, id. Nuguria: haisoa, id. Maori:
hoa, a friend, a mate ; hoahoa, a spouse. Tonga, Tahiti, Hawaii,
Marquesas, Rapanui, Paumotu: hoa, friend, fellow, companion.
Mangareva: hoa, oa, id. Rarotonga: oa, id. Fotuna: soa,
brother's brother, sister's sister. Nukuoro : soa, soka, a friend.
Bierian: ohoa, wife, husband. Baki: koa, id. Mota : soai, member,

component part of an organic whole.
Kayan : hawa, a wife.
Arabic : s'ai1, an associate, follower.
Dr. Macdonald associates sod and so with saisai to assemble, with less
attention to the eytmology of Efate than to the Semitic with which he
wishes to affiliate it.

The sex idea is widely extended ; in Efate, Maori, Bierian, Baki, and
Kayan the opposite sexes, distinct statement of the same sexes only in
Fotuna. In Nukuoro the intrusion of k in the alternative form soka is not
understood. In Baki koa the change from h to k is so violent, particularly
when it is recalled that the small island of Epi is the home both of Baki
and Bierian, that we suspect the not unnatural error of a printer in reading
foreign manuscript.

Without cognate forms in Indonesia we may neither quite accept nor
wholly reject the Kayan word.

The Arabic is greatly in need of proof before it can be accepted.

264.
song i, sung i, sum i, to kiss.

Samoa : songi, to rub noses, to salute ; songisongi, to smell. Futuna,
Nukuoro: songi, to kiss. Tonga: hongi, to smell, to sniff.
Niue: hongi, hohongi, honungi, id. Uvea: hongi, hohongi, id.
Rapanui: hongi, to kiss, to smell; hongihongi, joy. Paumotu,
Tongarewa: hongi, to rub noses, to kiss. Marquesas: honi,
hoki, to kiss, to smell. Hawaii: honi, id. Rarotonga,
Mangareva: ongi, to smell, to kiss, to salute by rubbing noses.
Tahiti: hoi, hohoi, id.
Lambell : isong, to smell. King : sangopi, id. Aneityum : aijnmnyi,
to kiss. Maewo: cf. mbunimhuni, mbunimbunisi, to smell,
to kiss.
Malay: chyum, to smell, to kiss.

Ethiopic: sa'ama, to kiss. Arabic: s'amma, to smell (there is no
trace of this meaning in Efate).
The Proto-Samoan stem is songit.

Outside of Efate we do not recognize this stem in Melanesia until we
reach the New Ireland coast at the east gate of migration. The Maewo
forms noted are quite irreconcilable with this stem even in abrasion, and
the Aneityum word is more a puzzle than a comparable form. The Lambell
isong is clearly homogenetic. King is in geography and speech so close to
Lambell that we have no doubt in the recognition of sang- in sangopi, but

DATA AND NOTES. 307

the language is yet so little known that we can not explain the change of
vowel or the latter member of the composite.

The Malay is not at all to be considered . It seems to me that Dr. Macdonald
has brought in the Aneityum and the Malay to establish a chain running
sungi-sumi-aijumnyi-chyum-s'amma. Inasmuch as there is no strength in
the inner links the chain scarcely supports the Semitic load.

265.

suma, uma, himwa, house.

Maori: ruma, a room, an apartment.

The following words mean house :

Duke of York, Motu, Uni, Doura, Kabadi : ruma. Wango : rumwa.
Lambell, Lamassa : rumai. New Guinea, Laur : rum. Buka :
luma, aluma, lumu. Bougainville, Alite, Ugi, Uni, Pokau:
luma. Iai, Kpi, Rubi: uma. Lifu, Uea: uuma. Lakon:
umwa. Waigiou, Moanus: urn. Tubetube: yuma, id. Treas-
ury Island, Sinaugoro, Hula, Keapara, Galoma, Suau, Sariba,
Tavara, Awalama, Taupota, Wedau, Galavi, Boniki: numa.
Saa: numwe. Tubetube: numi, id. Fagani: rima. Ulawa,
Bululaha : nimwa. Tangoan Santo, Ambrym : ima. Eromanga :
imo. Aneityum: im. Mota, Arag: imwa. Deni, Nengone:
mwa. Merlav, Mosin : imw. ho:emwa. Motlav, Volow : e mw.
Norbarbar, Pak, Leon : eng. Makura : na-ingma. Aneityum:
neom, niitn.
Malay, Baju, Liang, Lariko: rumah. Matabello: oruma. Kisa:
rome. Amblaw : lumah. Cajeli, Caimarian : luma. Q&hJume.
Mayapo: humah. Java: humah, uma. Teluti, Nikunau : uma.
Arabic: h'a'mat, h'im', house.
Tregear is by no means convinced that Maori ruma is not the borrowing
of the English room. Since the word does not anywhere else enter the
Polynesian area I prefer to let our best Melanesian authority argue the
case for his client. I cite herewith Codrington's highly interesting note
(Melanesian Languages, 77) :

This is an interesting and important word. The very wide range of the word, which
in Malay is ruma, and the great variety of its forms point to the great antiquity of this
as a common possession of these languages. As is the case with the very widely pre-
vailing name for a canoe, we may argue that a word which has spread so far and changed
so much goes to show that the thing which it names was known to the undivided people
whose dispersion spread the word so widely abroad. If the presence of certain common
words in Aryan languages shows that the Aryans did not separate till certain arts were
known and practised by the common ancestors, so we may argue that the ocean lan-
guages testify that the ancient speakers made canoes, built houses, cultivated gardens,
before the time came when their posterity branched off on their way to Madagascar
and Fiji.

The word now immediately in view as the name of a house ranges from the Malay
Peninsula, through the islands of the Indian Archipelago, to the very extremity of
Melanesia in the Loyalty Islands. It has not a continuous range, it appears and dis-
appears at intervals, but in that line and chain of islands it is never absent long. It
appears in Mafoor at the northwest of New Guinea, and in Motu at the southeast, and
in the Marshall Islands of Micronesia. In Polynesian languages it does not appear;
in the Kingsmill it is im. The fact that the word has in this way established itself

308 THE POLYNESIAN WANDERINGS.

generally, but not universally, at intervals, and not in a continuous line, shows that
it is not one which can be traced to one center, from whence it may be thought to have
been introduced by commerce or modern intercourse. The same conclusion is enforced
by the consideration of the great variety of the form of the word, which ranges from
ruma to eng. If a word appearing in its full form in Malay were to appear corrupted
and changed as it receded in distance from the region in which Malay is spoken, we might
well suppose the Malay the original. But when the changes in form bear no certain
relation to the distance from Malayan regions, and the variations are local and discon-
nected, it is not so; some center there must have been, but it can not now be pointed out.
The geographical range of the word must be observed by comparing the vocabularies
with the map. The variation of the form can be seen in the vocabularies. In Mr.
Wallace's list the Malay rumah and the Javanese umah give at once typical forms, one
with and the other without an initial consonant. Of the first type there are also luma
and huma, of the second um and probably om. (The Ceram word jeiom, used by
Alfuros, is probably om, a form of uma, with the collective prefix fei, Fiji vei.) Out of
thirty-three words twenty-two are forms of these types. The variety of forms in
Melanesia is greater, but the types are the same; ruma is in Duke of York and San
Cristoval, uma in Api and Lakona. The vowel also changes, and ruma, with changes
of initial consonant and vowel, becomes luma, nume, huma, rima, nima. By similar
change uma becomes ima, ema, and dropping the vowels at the beginning or end, 'ma,
im, eom, em, eng. To account for this last change it is enough to say that, in the
neighborhood where it is made at any rate, the m is the nasal one which, as mentioned
above, regularly changes into ng; ima, im, makes eng, as lima "a hand" makes Fiji
linga, Maori ringa. This m in Nengone is written 'm, and the Nengone 'ma is identical
with the Santa Cruz mwa.

It will be noticed that Efate has three forms, all more or less remote from
the primal stem. Suma is found nowhere else, this being the only language
in which the word has initial s. Himwa, so far as the initial is concerned,
seems to be a transition form leading to the general imwa-ima of the north-
ern New Hebrides. Uma is found in Epi and the distant Loyalties.

As to the Arabic I venture the suggestion that it taxes philological prin-
ciples far less to present its affiliation with the German Heimath and Heim.

266.
tanu mi, tanu-maki, to cover with earth, to put into the ground; tan i,
tun i, to earth it, to cover with earth and then with anything ;
tano, tan, earth of any kind, soil, clay, ground.
Samoa : tanu, to bury, to pave ; tanufale, to cover a house with coco-
nut leaves in a storm; tanuma'i, to cover up with. Tonga:
tanu, tanua, to bury, to conceal, to hide ; tanuanga, a burial place ;
tanuma, to bury the dead by numbers; tanumaki, to earth up
any plant or tree, to hoe, to cover up; tanumia, to be buried
by falling of earth debris ; tanutanu, to bury, to cover over with
earth ; tano, a burial place. Futuna : tanu, to bury, to inhume,
to cover with earth; tanuma, a tomb, ditch, trench. Nuguria :
tanu, to bury; taruma (? tanuma), a grave. Niue: tanu, to
bury, to cover up. Uvea: tanu, to bury; tanuma, tano, ceme-
tery, sepulcher; tanumanga, a pit. Maori: tanu, to bury, to
plant, to fill up. Tahiti : tanu, to plant, to bury. Mangaia :
tanu, to plant. Mauke: tanu, to bury. Rapanui: tanu, to
hide, to conceal, to plant ; tangata tanu kai, a cultivator. Man-
gareva : tanu, to plant, to bury, to inhume. Paumotu : tanu,
to cultivate. Fotuna : no-tanu, no-tanumia, to bury. Hawaii :
kanu, to bury, to cover up in the earth.

DATA AND NOTES. 309

Mota: tano, ground; tanu, to bury with earth. Maewo, Nguna:
tano, ground. Malo: tanomia, to bury. Kabadi: kano,
land. Deni: ndano, ground. Saa: 'ano, id. Tanna: tana,
land; (t)anum, to bury. Tarawa: tan, id. Ambrym: tan,
ground. Aneityum: in-tan, id. Lo: ten, id. Laur, Lambell,
Lamassa: tun, a grave. King: nutun, id.

Malay: tanam, tanuman, to bury, to inter, to plant. Basakrama:
tanam, id. Malay: tanah, land.

Arabic: tana, to cover with clay; tino, tano, earth, clay.

After defining tanumi and tan as burial with a covering of earth Dr.
Macdonald says boldly "hence" and proceeds to join therewith tano-tan,
earth, soil, ground.

We have clearly two distinct stems mingled here.

That which means the ground does not at all appear in Polynesia. In
Melanesia it is tano, ndano, kano, 'ano, tana, tan, ten. In Indonesia we find
Malay tanah land. With this stem the Arabic certainly has an identity of
form and of sense.

The stem signifying to bury is in the Proto-Samoan tanum, and the radi-
cal m holds in Polynesia almost without exception when under the protec-
tion of a succeeding vowel. This radical m is found in two instances in
Melanesia and it persists in Indonesia. The stem in use in Polynesia is
tanu, but in Tonga and Uvea tano is found along with tanu. Similarly
in Malo we find tano, but with the radical m to fix the word. It would seem,
therefore, that in these instances abraded tanum and tano have been con-
fused. The Efate tun to bury is confirmatory of the New Ireland grave
words tun and nutun. With the tanum stem the Semitic proposed is in
no great accord.

This word has been exhaustively examined in the inner record of its
structure in my paper on root reducibility (27 American Journal of Phi-
lology, 383).

267.

tau, a season, time, year.

Samoa : tau, a season, a year; tausanga, a season, a year of six months.
Tonga: talu, a season, a crop, a year; ta'ube, annual. Futuna:
tau, season for planting and particularly for yams. Uvea: tau,
era; tau, year. Niue: tau, a year, a season. Maori: tau, a
year. Tahiti, Rarotonga: tau, a season. Marquesas: tau,
a year (of ten months). Rapanui: tau, a year, a season.
Mangareva : tau, a year, the season of the breadfruit. Paumotu :
tau, a season, a period. Nukuoro : tau, ngatau, a year. Aniwa :
tou, a year. Hawaii: kau, a season, the summer or warm
season, a lifetime.

Mota: tau, a season. New Britain: taun, a year.

Java, Sulu: tahun, a year. Matu: ta'un, id. Ilocan: taoen, id.
Tagalog: taon, id. Bugi: ta ung, id. Macassar: taoeng, id.
Malagasy: taona, a year, a time, a season.

Arabic: zaman, a year. Syriac: ziban, id. Modern Syriac:
zona, id.

310 THE POLYNESIAN WANDERINGS.

The Samoan tausanga seems to suggest a taus stem. This can not further
be recognized in Polynesia; therefore we are not justified in putting it
forward adversely to the taun stem which appears in all the Indonesian
identifications and in Melanesia is found in New Britain.

The Semitic is impossible except Modern Syriac zona, and that is but a
partial resemblance.

268.

telei, talai, the ancient axe or adze-like axe (a shell) .

Samoa : talai, to adze. Tonga : talai, to smooth off rough edges.
Futuna: talai, to cut off knots or thorns. Niue: toki talai,
an adze. Maori: tarai, tarei, to chop or smooth as with an
adze. Tahiti: tarai, to chop or adze a piece of timber; toi
tarai, an adze. Mangaia : tarai, to adze, to hew. Mangareva :
tarai, to rough hew. Paumotu : tarai, to cut, to hew. Hawaii :
kalai, id. Marquesas : tadi, to smooth with an axe or tool.

Motu: talai, to chop. Mota: tara, to hew, chop, cut. Merlav:
tara, to cut. Massim: tara, to cut off. Gog: tar, to cut.
Volow: ter, id. Aneityum: inpas aterei, adze.

Malay: charai, to part, to separate.

Arabic: s'araha, to cut, to slice, to carve, to dissect.

This stem is readily identified as homogenetic in Polynesia and Mela-
nesia, and it preserves the verb-formative i as far away as Motu. The
undoubtedly earlier verb tara exists in Mota and Merlav, and has undergone
terminal abrasion in Gog and Volow. In the Aneityum locution aterei has
altogether the appearance of identity with talai; but Inglis in his vocabu-
lary of that language defines the word: aterei, crooked, bent; inpas aterei,
an adze. Inference should not be allowed to outbalance the record of the
reporter; therefore we leave the matter open pending further research in
the field.

In the Malay charai we find a change of initial t which is not wholly
without precedent, but of far greater moment is the change of sense ; the
identification is, therefore, doubtful.

The Semitic is remote from talai both in form and in sense.

269.
toki, an axe; tok, violence, force.

Tonga, Futuna, Niue, Uvea, Nukuoro, Fotuna, Mangaia, Marquesas :
toki, an axe, a hatchet. Maori, Mangareva: toki, an axe, an
adze, or similar tool. Tongarewa: toki, an adze. Rapanui:
toki, a stone axe. Nuguria: toki, a shell axe. Paumotu:
toki, to hit, strike, drive in, the edge of tools. Samoa: to'i,
a hatchet. Tahiti : toi, a hatchet or tomahawk. Hawaii :
koi, a small adze.

Duke of York: torki, toki, to cut or lance, to cut out a spear point.
Mota: toto, totogag, to chop. Aneityum: cf. etuko, to split
wood.

Baliyon: tuk, to chop.

Arabic: takka, to cut. Hebrew: tok, violence.

DATA AND NOTES. 311

The Efate tok violence is visible nowhere within the horizon of our studies,
and is suspicious in view of its positive resemblance to Hebrew tok violence.

In Polynesia we find no variation from the axe signification except in
the Paumotu, which in many particulars represents a very primitive type
of the language. The Paumotu meanings make it possible to accept with-
out hesitation the Duke of York signification, which is not readily associable
with a hatchet sense.

The second form in Mota totogag, of which ag is recognized as a verb
definitive suffix, yields us an abraded form tog, which undergoes yet further
abrasion into toto. The Aneityum etuko is, like most of the identifications
in that obscure speech, by no means distinctly established.

We meet the stem but once in Indonesia, an abraded form as in Mota
and with a vowel change like that in Aneityum.

The Arabic is a resemblance; more, however, to the eye than to the ear,
for when spoken the triliteral is at once apparent.

270.

tuai, tuei, old, ancient, long ago, a long time hereafter; bakatuai, to pro-
long, to put off, to delay.

Samoa: tuai, former, olden, to be a long time; tuatuai, somewhat
long, to delay; fa'atuai, to prolong, to put off, to defer, to
delay. Tonga: tuai, delay, procrastination, slow, dilatory,
to be long, to defer; fakatuai, fakatuotuai, to linger, to delay,
to defer, to procrastinate, to protract. Niue : tuai, old, ancient ;
fakatuai, slow. Uvea : tuai, to delay, to loiter. Nukuoro :
tuai, ancient. Aniwa, Vate: tuai, old. Rapanui: tuhai,
old, ancient.

Eromanga: itetuai, of old; etuai, some time ago. Sesake: tuai,
formerly. Mota: tuai, old times. Bierian: tuai, long ago.
Malo: tuai, old. Malekula: tue, old. Aneityum: itu, old,
former. Baravon: kua, old. Pak: 'ue, id.

Malay: tuwah, old. Java: tuwa, id.

Arabic: 'adiyy', old, ancient.

It seems that here we are dealing with a tua stem of which, with conditional
ma, matua (216) may be a derivative. That the stem really is tua, and i
but the definitive suffix, appears in the duplicated Tonga faka-tuotuai.

In tracing the stem in Melanesia we find in Eromanga and Aneityum pre-
fixes of uncertain value. Most of the identifications in the western Pacific
maintain the full tuai form. Absence of the definitive suffix is found in
Baravon, if we accept the t-k mutation despite the fact that this is its only
appearance among eight cases of t found in our material, and in Malekula.
In this group belongs Pak 'ue, for it is characteristic of that tiny speech
to throw out t. The last stage of abrasion is found in Aneityum i-tu.

The Indonesian identifications are scanty but valid.

The Polynesian has a t in tuai; a Semitic word meaning old has been
found which has a d in it somewhere. Another link has been forged in the
chain of Semitic origins.

312 THE POLYNESIAN WANDERINGS.

271.

un, a fish scale.

Samoa : una, a scale of a fish, a plate of tortoise shell ; unafi, to scale
a fish. Futuna : una, tortoise shell ; unafi, fish scale. Niue :
una, tortoise shell; hinafi, fish scale. Maori, Marquesas,

Rapanui, Mangareva: unahi, fish scale. Tahiti, Marquesas,
Paumotu: unahi, to scale a fish. Hawaii: una, tortoise shell;
unahi, fish scale. Tonga: uno, fish scale, tortoise shell.
Uvea: uno, fish scale. Fotuna: ano-na-unafi, id. Nuguria:
unafi, id.

Motu: una, fish scale; unahia, to scale a fish. Aneityum: ninihen,
fish scale.

Malay: unus, to pull out.

Hebrew: halas, to pull out, to pull off. Arabic: h'ala'a, id.

The Proto-Samoan stem is unaf . In general these closed stems are better
preserved in the Proto-Samoan migration than in the Tongafiti; here,
however, the strong stem oddly appears where least it would be expected.
Unafi is in form a verb ; the verb-formative suffix i is added to the simple
stem. In Samoa and Motu unafi is a verb and distinct from the noun, in
which the final consonant has been abraded ; in Maori, Mangareva, Futuna,
Niue, Fotuna, and Hawaii unafi is the noun and una is either absent or else
specialized to mean one of the plates in a head of turtle ; in the Marquesas
it is both noun and verb.

The only phonetic variants in the Polynesian are these: in Niue the
frontal accretion by an aspirate, this seeming to be sporadic, since a primal
aspiration in the stem would have been represented in other languages
of the stock in some mutation form ; in the same speech the change of u to
i, a change which we shall find again only in Aneityum ; in Tonga and Uvea
the change of a to 0 in the second vowel.

In Motu, that very remote colony of the swarm which left Indonesia by
the southern gate, we find noun and verb used exactly as in Samoa. The
only other identification which Melanesia has afforded us is n-inihe-n of
Aneityum. In this the initial n is clearly the article ; the final n is observed
upon a great many noun stems which end in a vowel. For the initial vowel
of inihe we have already observed a precedent in Niue. The mutation f-h
is common in Polynesia and we have several instances of its occurrence in

Aneityum.

The Malay identification is not only presented as fact by Macdonald,but
has also received very respectful consideration by Tregear, who is by no
means cordially disposed to the Malayo- Polynesian theory. Yet to accept
this identification requires us to assume the equivalence of unaf and unus,
an assumption far too violent to be satisfactory ; and then in addition to
disregard the fact that in the sense of the Malay unus to pull out Samoa has
unusi, Tonga has unuhi and so have several of the Tongafiti languages.

About the only point of identity upon which the Semitic could be hung
is the final s in Malay unus. Now that that is removed from consideration
the Hebrew can show no resemblance.

DATA AND NOTES. 313

EFATE-MELANESIAN-VITI-POLYNESIAN-MALAY-SEMITIC.

272.

abu, afu, au, ashes; abuabu, afuafu, to be dusty, to fly in the air (dust);
libu, lifu, to be ashy, ash-colored, dirty or covered with ashes,
as in mourning for the dead; mafu, a thick vapor like dust,
uncleanness (ritual) which makes the sight dim.
efu.

Samoa: efu, efuefu, dust, to become dust. Tonga: efu, dust;
efuefu, ashes; efuhia, dusty, covered with dust; fakaefu, to
raise a dust; maefu, dust; ngaehu, muddy, turbid. Futuna:
efu, dust in general. Niue : efu, efuefu, dust; efuefu-afi, ashes.
Uvea: efu, ashes; efuefu, dust. Marquesas: ehu, fragments,
to fall in particles ; ehu ahi, ashes ; ehuehu, twilight of morning
and evening. Maori : ehu, turbid, mist. Tahiti : ehu, dis-
colored, muddy. Hawaii : ehu, spray, steam ; ehuehu, darkness
arising from dust, fog or vapor; kuehu, to shake the dust from
a mat. Mangareva: ehu, dust, ashes. Rapanui: ehu, fire-
brand, ashes.

lefu.

Samoa: lefu, ashes; lelefu, to be burnt to ashes; fa'alelefu, to reduce
to ashes. Niue: polefu, mist, fog. Futuna: lefu, ashes
slacked in water for dressing the hair. Uvea: lefulefu, ashes.
Nukuoro: lefu, ashes; rehu, dust. Nuguria: lehulehu, ashes.
Hawaii: lehu, ashes; lelehu, to see with difficulty, to become
blind. Maori: rehu, mist; pungarehu, ashes; rehutai, sea

spray; turehu, indistinctly seen. Tahiti: rehu, ashes, soot,
powder in general; rehurehu, the dusk. Rapanui: rehu, dust.
Mangareva : rehu, reu, ashes ; rehurehu, the morning soon after
sunrise; reureu, morning. Sikayana: rehu, lime. Rarotonga:
reu, ashes; reureu, dark. Moriori: purungehu, ashes.
nefu.

Samoa: nefu, to be stirred up, turbid, muddy; nenefu, to be dim,
indistinct. Tonga: nenefu, twilight, dimness, dim, dull,

uncertain. Futuna: nenefu, dimness of vision. Uvea:

nefunefu, darkness, gloom, mist, vapor, turbid, dusty. Maori:
nehu, dust, steam; nehunehu, dusky. Mangareva: panehu,
to dry up, to wither; paneu, gray, covered with dust.

Viti : ndravu, ashes ; ndravundravua, ashy ; ndravukasi, dust, dusty ;
ndravusa, ashes, to rub the head with ashes. Rotuma : roh,
ashes.

Marina : avuavu, ashes. Arag: taniavu, id. Lakon: tangehav, id.
Maewo : ndigevu, id. Fagani, Vaturanga, Nggela, Guadalcanar :
ravu, id. Bugotu : pindaravu, id. Nggao: parafu, id. Motu:
rahurahu, ashes, a fireplace. Doura : kokorahu, ashes. Kabadi :
rauna, id. Savo: lavu, id. Roro: r abu, id. Vni : labu, id.
Wedau, Galavi, Boniki : lapukare, id. Duke of York : kabu, id.
Taupota: gabuwari, id. Wedau: ai-gabuwari, id. Sesake:

314

THE POLYNESIAN WANDERINGS.

tano au, ashes. Mota: rav, dusk; ravrav, dusk of evening;
malurav, id. ; marav, dim, misty. Pak, Sasar, Alo Teqel : uwtis,
ashes. Mosin: tuwus, id. Mota, Gog: tarowo, id. Vuras,
Motlav: wowo, id. Lo: ivowa, id. Norbarbar: powo, puio, id.
Keapara: abu, id. Mekeo: ae-apu, id. Motu: ^a/m, id.
Tavara: gahue, id. Awalama: gahuwe, id. Sariba: gavara, id.

Malay: abu,dabu, labu, dust, ashes; kalabu, klabu, ashy, ash-colored.
Java: aww, fc/ituitt, dust, ashes. Malagasy: vuvuka, id.

Arabic: fca&a (/*a&w), to rise, to float in the air (dust), to become
like dust, ashes; habwat, dust; habut, dust, dust mixed with
ashes; hebwa, fine dust, powder; mutahabbi, weak in sight.

There can be no doubt that an intimate relationship obtains between
efu, lefu, and nefu. It is too early to establish the relation, yet from
the form I am hopeful that further inquiry along the lines which I have
already discussed in the presentation of my theory of development by
consonantal modulants will lay bare the life history of this interesting
group. The only backward glimpse which our material affords is that
which Tonga efuhia gives us of a Proto-Samoan stem efus, and this is found
again to the westward.

As the stems indicate relationship by their form, so do the several signifi-
cations show that the stems are so near that it is possible for a certain
meaning to appear first in one and then in another. The following tabu-
lation will make that clear :

Dust.

Samoa

Tonga

Niue

Uvea

Futuna . . . .
Nukuoro. . .

Maori

Moriori . . . .

Tahiti

Marquesas .
Rarotonga .
Mangareva
Hawaii. . . ,

efu

efu

efu

efu, nefu

efu

rehu

nehu

rehu
ehu

ehu, neu
ehu

Ashes.

Vapor.

lefu

efu

efu

efu, lefu

lefu

lefu

rehu. . .

rehu

rehu

ehu

reu

ehu, rehu

lehu

lefu
nefu

I ehu
nehu
rehu

ehu

Darkness. Twilight. Muddy

nefu
nefu

nefu
nefu

rehu, nehu

reu
ehu, lehu

nefu

reh u
ehu

nefu
ehu

nefu

ehu
ehu

A glance down these columns will show how the sense plays from one
stem to each of its associates ; there is but one column that does not carry
all three stems, and the Maori employs all three to express the signification
vapor. We are, therefore, wholly justified in dealing with the three as a
unit in examining their exterior relations.

In the three Polynesian lists there is little which calls for explanation,
for we scarcely need pause here to dissect the few simple composites. The
Moriori purungehu ashes is readily reducible as a metathesis upon the
Maori pungarehu. The crepuscular sense is plain in Tonga, Tahiti, and the
Marquesas as an immediate development out of the yet more general
meaning of darkness and indistinctness of vision ; yet in Mangareva rehurehu
we find the word transferred from the gloaming before the dawn to a posi-

DATA AND NOTES. 315

tion after sunrise, when, in the clear light of day, the primal sense is lost.
The vSikayana rehu lime, a specialized powder, finds a far-eastern parallel
in Tahiti rehu, which is distinctly applicable to any powder.

In Efate we find efu {abu, afu, au) and lefu (libu, lifu); mafu is not to be
correlated with any of the Polynesian stems.

Viti ndravu, both dust and ashes, recalls the vowel found in Efate' abu,
and at the same time the ndr is evidential that the Proto-Samoan stem
was ref u with the r grasseye (Samoan Phonetics, 1 7 Journal of the Poly-
nesian Society, 152).

In Melanesia we shall institute our search for the simple stem. I have
already remarked that this was efus. This enables us at once to pick up
theVanua Lava forms, the uwus of Pak, Sasar, and Alo Teqel; the s, in fact
the us, is clear at first sight, the f-w mutation is confirmed by a widely
extended range of examples. Mosin tuwus is this identified uwus with
frontal accretion of some sort. Vuras is but a few miles from Sasar along
the same beach, therefore wowo is readily acceptable as a dialectic variant ;
the terminal ^ has dropped off and the then final wu has undergone modifi-
cation to wo and later duplication. Motlav has the same form, and Lo
wowa but slightly differs. Having now established wo as the vestige of efu
we find it a composition member of puwo of Norbarbar and tarowo of Gog
and Mota. Starting afresh with efu, we find it in Efate afu, abu; in Marina
avuavu; as a composition member in Arag, Lakon and Maewo forms ; having
lost its labial we have au in Efate and Sesake, this identification being
open to such doubt as our lack of cordiality to internal loss may warrant.

We next examine the lefu stem. In Efat6 lifu and libu are clear. Ravu
is well distributed, both independently and in composition, rafu, rahu, lavu,
and in Mota rav.

The Duke of York kabu ashes is to be rejected, but I have left it with
this material because of its reappearance in Viti kambu vapor, which is one
of the senses pertaining to this protean efu stem.

Our Indonesian material is scanty, but efu and lefu stems are recognizable.

In the Semitic offered for our consideration it must be acknowledged
that there is a certain resemblance, but the disclosure of a final t where our
Polynesian has the ^-terminal is enough to stamp this as not a family
resemblance.

273-

(a) 5fa ki, 5fa ki, to bury.

Samoa : iifi, a lid, a cover, to cover, to conceal ; ufita'i, to cover with.
Tonga : uufi, a cover, to cover, to overspread ; ufiufi, to cover,
to conceal. Futuna: uufi, to cover, to hide, to conceal, to
disguise ; ufia, covered ; ufiufi, to cover things without envelop-
ing them. Uvea: ufi, uufi, to cover. Niue: iifi, a covering,
to cover up, to conceal. Maori: uhi, uwhi, to cover, a cover-
ing. Hawaii : uhi, a covering, a veil, to cover over a thing so as
to hide it. Mangareva : uhiuhi, to hide, to cover.

Viti: umbia, to cover over; umbi, the top, lid, cover, a quilt.

Malagasy : afina, to be concealed ; manafina, to conceal, to bury.

Arabic: uaba, to conceal, to be concealed, to bury.

316 THE POLYNESIAN WANDERINGS.

(b) ui, uui (uwi), the yam.

Samoa, Tonga, Futuna, Niue, Uvea, Aniwa, Sikayana : ufi. Vat6 :
uf. Maori: uwhi, uhi. Nukuoro, Moiki, Tahiti, Hawaii,
Mangareva, Rapanui, Paumotu: uhi. Marquesas: buauhi.
Mangaia, Rarotonga: ui. Fotuna: aufi.

Viti: uvi. Rotuma: uk.

British New Guinea, Murua, Kiriwina, Galavi, Boniki: kuvi, yam.
Nggela: kuikuvi, a cover, to cover; kuvihia, to cover; kuikuvi,
a leaf used as an umbrella. Mukawa: kubi, yam. Nada:
kuva, id. Fagani: uvi, id. Pokau: veu, id. Wango, Saa:
uhi, id. Motu: uhe, the end of the yam, which is kept for
planting. Aneityum: n-uh, yam. Tanna: n-uk, id. New
Caledonia: ubi, id. Vanikoro: upie, id. Baki: yubi, id.
Eromanga: n-up, id. Baravon: up, id. Makura: na-u, id.
New Ireland (Carteret Harbor): u, id. Pala: uh, id.

Malay: ubi, yam. Kayan: uvi, id. Java, Kisa: uwi, id.

Malagasy: ubi, ovy, id.

Arabic: "ayab, roots (so called because buried in the ground).

The cause of this collocation is the satisfaction of Macdonald's Semitic
theory in his naif interpretation that roots are things buried in the ground.

A.

It is highly problematical whether there is any identity between afa ki and
ufi. The Polynesian and the Viti make no reference to burial ; the dominant
sense is to cover by setting the covering agent upon the object to be con-
cealed. It is not until we reach the Malagasy, and a secondary form at
that, that we find any recognition of the burial sense. In general I am
opposed to the acceptance of any Malagasy identification where there is
any but the closest resemblance of form and meaning, except where we find
it upheld by a satisfactory suite of transition forms in Indonesia. The
Arabic is quite as imperfectly correlated with Polynesian ufi as is Efat6
afa-ki, in fact more so, for we should need evidence as to the abrasion of
so strong a palatal as ". Furthermore the Proto-Samoan stem is uf it. Its
only Melanesian appearance is in Nggela kuvihia, which bespeaks a kuvis
stem, not wholly to be accepted as to ^-accretion and t-s mutation in the
absence of transition forms, although we have some evidence tending to
establish this mutation.

B.

The absence of the inner labial spirant in the Efate forms finds a true
Polynesian parallel in Mangaia and Rarotonga; in Melanesia it is found
only in the most degenerate forms, in which naught survives of the word
but its initial vowel.

The Polynesian life-history of the word is plain ; it is written about the
successive mutations of the central consonant, /-z>-/*-extinction.

In Melanesia we shall examine the identifications by the same criterion.
The form kuvi is anomalous in the frontal fe-accretion and it is loosely
accredited to a territory too wide and heterogeneous to admit of more pre-
cise identification. The ufi form is absent from this area. The first stage

DATA AND NOTES. 317

of the upward progression therefrom, uvi, is found in Fagani, and probably
in Pokau if we regard veu as a 231 metathetic form. The next stage, the
passage to the aspiration, is found in Wango and Saa uhi, Motu uhe, and
in the abraded uJi of Aneityum and Pala. A further step is represented
by the uwi of Efate. Last of all we find that the consonant has vanished
to extinction in ui, also of Efate. In the downward progression the first
step rests upon the ubi of New Caledonia and the Baki yubi; the next upon
the abraded up of Eromanga and Baravon. The last degeneration is shown
in the u of Makura and New Ireland. In Tanna n-uk shows a mutation,
labial to palatal, which we should scarcely be able to support if it were not
that Rotuma, far more intimately within the Polynesian sphere, has also
the form uk.

In Indonesia we find five identifications of the most satisfactory character.

The Arabic "ayab offers several difficulties in form besides its manifest
lack of connection in sense, even after the pleasing parenthetical informa-
tion. The guttural "might readily pass into k as in kuvi, but that is an
uncertain form at best and in the wide range of the stem the initial k is
nowhere else discoverable. The duplicated u in Tonga, Uvea, and Futuna
is in no sense a parallel to argue a double vowel sound persisting after the
dropping of the y in "ayab, for the uu in these three languages is no more
than a device to express the long vowel in a font of mission type which
lacked the macron.

274.

angiengi, the air, breeze; in, the wind, the air; langi, the wind.

Samoa : angi, to blow (of the wind) ; angiangi, to blow gently.
Tonga : angi, to blow ; angiangi, to begin to blow. Futuna :
angi, angiangi, to blow. Niue: aw^', id. Uvea: angiangi, id.
Nukuoro: angi, the wind; angiangi, a fan. Maori: angi, a
gentle breeze; hengi, to blow gently. Mangareva: angi, a
light wind ; angiangi, to blow gently. Moriori : hokaangi, to
shake in the wind. Hawaii: ant, aneane, to blow gently.
Marquesas : ani, the air.

Samoa, Tonga, Fakaafo, Futuna, Niue, Uvea, Maori, Mangaia, Mana-
hiki, Rarotonga, Mangareva: matangi, the wind. Paumotu:
matangi, the air. Rapanui: matangi, the air, a squall.
Sikayana, Nuguria, Liueniua: matani, the wind. Hawaii:
makani, id. Tahiti: matai, id. Marquesas: metani, meiaki, id.
Aniwa: tu-mtangi, id. Fotuna: mtangi, id.

Viti: thangi, the wind, the atmosphere; thangina, to be blown away
by the wind. Rotuma: lang, the wind.

Tanna, Eromanga: matangi, the wind. Tanna: mtangi, id. Anei-
tyum: nimtinjop, id. Lemaroro: langi, id. Sesake, Paama,
Mota: lang, id. Nengone: 'n lang, id. Ambrym: leng, ling,
ying, id. JJeaiang, id. Malekula: ni-en, id. Tobi: yang, id.

Malay: angin, the wind, air, atmosphere. Java, Visayas, Tagalog,
Magindano: hangin, the wind. Bicol: hagnin, id. Bugi:
anging, id. Kisa: ange, id. Malagasy: anina, id.

Arabic : nasama, to blow gently ; nasam',na'sam, nasim', a light wind.

318 THE POLYNESIAN WANDERINGS.

In Polynesia angi is generally the verb ; the noun matangi is a composite
of angi in its earlier attributive value with mala (324). Only in Nukuoro
and in Maori, Mangareva, and the Marquesas does angi function as noun.
The Proto-Samoan stem is hangi, this being the true light aspiration which
has wholly vanished in the current phase of the languages and is recoverable
only from Viti thangi. With this angi is affiliated the angiengi of Efate.

So, too, in an abraded and degraded form is, in my opinion, Efate in.
In its first stage, the simple abrasion, we find Uea (Loyalties) ang; parallel
with this, but in the remoter Micronesian area, is Tobi yang. Thus we
reach Ambrym ying, from which it is but a simple step to in. Another
short step thence leads us to ni-en of Malekula.

The common Polynesian matangi appears intact in two of the languages
of the New Hebrides, of which Tanna has a shortened form mtangi, which
also appears in the Polynesian verge islands Aniwa and Fotuna. A yet
further deviation from this stem is the rude ni-mtinjop of Aneityum.

Interwoven with these wind words we find langi. I feel that this is the
Polynesian langi sky borrowed with only a general regard for its sense.
We find it in the wind meaning and in its full langi form in Efate and
Lemaroro; abraded in Sesake, Paama, Mota and Nengone; abraded and
subjected to vowel deterioration in Ambrym leng, ling. Of these Nengone
and Mota express the sky by Melanesian words. The Melanesian lang wind
occurs in Rotuma, which uses langi for the sky.

In Indonesia the primal aspiration is retained in Java, the Visayas, Tag-
alog, Magindano and Bicol ; and in these and in the other instances cited
the identification is wholly satisfactory except for the apparent inversion
gn in Bicol.

Between hangi and Semitic triliterals nsm it is impossible to see any

relation.

275-

(a) aka, koa, stringy, fibrous ; akoa na, ako ana, root (literal and figurative) ;

aka na, ek, eka na, a relative, family connection (considered
as root or offshoot from) .

Samoa : a' a, fibers of a root, family connections, a plant whose root
is sometimes eaten. Tonga: aka, the root of trees, to take
root in the earth. Futuna : aka, root in general, a vine with an
edible root. Niue : vaka, a root ; aka, a soft-leaved creeping
plant whose root is eaten in time of scarcity. Uvea : aka,

root. Maori : aka, long fibrous roots, a climbing plant. Tahiti :
aa, roots. Hawaii: aa, small roots. Marquesas: aka, a
root; eka, young roots of trees from which native cloth is made.
Rarotonga, Nukuoro, Rapanui, Paumotu : aka, root. Nuguria :
haka, id.

Viti: yaka, a creeper whose root is edible.

(b) kaka naniu, the fibrous substance, like coarse cloth, that grows around

the stem of the coconut.
Samoa : 'a'a, lau'a'a, the fibrous stipule of the coconut leaf. Tonga :
kaka, thin fibrous substance found around the young coconut.
Futuna: MM, species of tissue which grows on the coconut.
Maori : kaka, anything fibrous or stringy. Marquesas : kaka,

DATA AND NOTES.

319

web or cloth covering the leaves of coconut trees, a sack.
Mangareva: kaka, the envelope of coconut leaves. Tahiti:
aa, the fibrous substance which grows on the coconut, the husk
or covering on the young breadfruit branches. Hawaii: aa,
the cloth-like covering near the roots of coconut leaves, a coarse
kind of cloth.

Viti: waka, the fibers or roots of a tree. Rotuma: va'a, root.
These all signify root :

Murray Island : sip kak. Duke of York : akar. New Britain :
okor. Bugotu: oga. Savo: ogni. Arag: garo. Alite:
kalokalo. Lo: gurah. Merlav: gari. Gog: gerin. Mota:
gariu. Motlav: goren. Sasar: gorgi. Pak, Alo Teqel:
gergi. Volow : girigi. Mosin : sigrigi. Omba : goarigi.
Maewo : goarii. Lakon : gegi. Marina : goe.

Malay: akar, the roots of a plant, a scandent plant, the parts of a
plant that climb ; akar, root, origin, principle, foundation (this
last word is said by Crawfurd to be Arabic). Kayan: aka,
root. Salayer, Gah: akar, id. Sula: kao-akar, id.

Arabic : 'akka, 'akak, to be split, fissured ; 'akko, a fissure ; 'akikat' , a
bag; lawako, small sprouts shooting from the upper part of a
palm; 'ikkano, shoots sprouting from the roots of palms and
vines; 'akka, to send forth such shoots from the roots.

A.
In aka we find two words alike in form, aka a root and aka the vine. In
general these two words have exactly the same form, but in Niue and Viti
they are distinguished and in practically the same way. The root signifi-
cation in Niue is expressed by vaka (vaka is also canoe) ; in Viti waka (canoe
being wangga), and Rotuma va'a (vaka being canoe) are the same. If this
^-initial were radical we should expect its recurrence somewhere along the
line ; the frontal accretion of Polynesian stems is so unusual that we are
without data for its intelligent study. The plant name in Niue is aka, as
in the rest of Polynesia. In Viti it is yaka. This Viti ^-initial is of great
interest and therefore I give in the following list all the instances in which
I have identified the word in which it is used with a Polynesian stem.

Viti.

Samoa.

Meaning.

Viti.

Samoa.

Meaning.

yambia

pia

arrowroot

yate

ate

liver

yandra

ala

awake

yatu

atu

a row in series

yaka

aka

a plant

I yava

vae

foot

yane

arte

a moth

i yava

pa

barren, of women

yam

ane

directive

yavato

ajato

a maggot

yasi

asi

sandalwood

yawa

ava

a fish

In yambia and in both identifications of yava it will be seen that it is not
merely y-initial but the syllable ya. There is nothing in the life-history of
these words (cf. 276 for the only other instance included in these data) to
indicate that the Viti y-initial retains the impress of something radical.
Yet one must regard the impression unavoidably acquired that Viti yaka,
e. g., retains some delicate intonation of vowel which characterized in the
Proto-Samoan the initials of these words.

320 THE POLYNESIAN WANDERINGS.

Efate aka of relationship is repeated only in Samoa. The word for root,
akoa, despite its extra syllable, may be found to come into concord with
Polynesian aka through the equivalence of aka-koa fibers. The type of
root embraced in this word is the fibrous roots, as reappears here and there
in these definitions.

In Melanesia the identification is intricate. The Murray Island locution
seems to include a form of this word. The Duke of York akar is a certain
identification. New Britain okor is clearly the same word, and Bugotu oga
comes readily into line, and Savo og-ni is recognizable. If now we accept
an abraded form ka, not found independently, we shall find a chain of
development as follows :

gar Savo, Merlav, Mota, Alite. gor Motlav, Sasar, Omba, Maewo.
ger Gog, Pak, Alo Teqel. gur Lo.

gir Volow, Mosin.

Lakon gegi and Marina goe are not to be omitted, but they are highly
irregular.

The Indonesian identifications are very clear.

Unfortunately for the Semitic identifications offered their correlation
depends on the harmonizing of aka and kaka. Even then they have no
significations which mean fiber, root, or this specialized stipule.

B.

It is not improbable that aka and kaka are homogenetic, but the lan-
guages have not yet been sufficiently studied after my theory of coefficient
modulants to enable us to present any sort of proof. Ex hypothesi we
should have a stem ka with a broad meaning of fiber ; the element a should
determine the ka fiber in some such manner as to limit it to root fibers;
the reduplicated kaka with its intensive force should picture the interlacing
fibers of this natural product. Unfortunately we can not yet prove a single
one of these points. Our kaka is limited to Polynesia and restricted to the
coconut fabric except that in Tahiti it extends to the breadfruit, the bam-
boo, and the sugarcane, and that among the Maori, who have not in their
colder land the coconut, it has possibly reverted to the primordial sense of
anything fibrous.

276.

ate, the liver (of a shark), the spleen; are, the liver; uateam, the kidneys.
Samoa, Tonga, Futuna, Niue, Uvea, Fotuna, Nuguria, Maori, Tahiti,

Marquesas, Mangareva, Rarotonga: ate, the liver. Rapanui:

ate, the liver, the lungs. Hawaii: ake, id.
Viti: yate, the liver.
Solomon Islands: ati, the chest. Malekula Uripiv: ere, the liver.

New Guinea (Maclay Coast) : arre, the liver. Mota : varai,

vare, the chest, the liver. Rubi, Suau, Dobu: ate, the liver.

Wedau : ate, the gall bladder. Motu : ase, aze, the liver. Roro :

ahe, id. Mekeo: ae, id. Nada: ata, id. Sariba: kate, id.

Mukawa: katekate, id. Kiriwina: kata, id. Murua: katu, id.

DATA AND NOTES. 321

Sinaugoro: gase, id. Galoma: gae, id. Hula: aie, id. Awa-

lama: ade, id.
Malay: ati, the liver, the heart (morally). Java: ati, the heart.

Kisa: akin, id. Magindano: ati, the liver. Pampangas:

ate, id. Ternati : fatf, id. Matu: atai, id. Tagalog: atay, id. ;

afo', the middle.
Arabic: fcaW, kabid', the liver. Hebrew: fea&ed, id. Ethiopic:

kabde, id.

In the Efate are several puzzles. The general name for the liver in all
Polynesia is here the spleen of men, and it should be borne in mind that
in an anthropophagous society the works of one's own kind are far more
familiar objects than in communities of a wider dietary. The word does
appear as the liver, but of the shark ; the human liver is are.

It is possible that ate may become are, this depending on the validity of
the proof establishing the t-r mutation presented in note 258. A re is readily
identifiable as ere in Malekula Uripiv, aire in New Guinea. Mota vare may
indicate : (1) that there is a stronger stem from which are develops by frontal
abrasion — in which case the ate-are identification is scarcely probable ; (2)
that Mota has acquired v by frontal accretion, a process which is by no
means satisfactorily established ; (3) Mota vare in conjunction with Viti yate
(see note 275) may indicate some light initial sound in Proto-Samoan, too
light to be caught in the general scheme of Polynesian consonant mutation,
but attracting the ear of the Melanesians as a sound to them difficult and
unusual, therefore requiring special effort to preserve, thus accounting for
the variance of the result in Mota and Viti. My own opinion inclines to
the third view. The fact that in Mota the liver word includes the chest
leads naturally to the inclusion of Solomon Island ati chest and Tagalog
ati the middle in these identifications.

The Indonesian citations are perfectly satisfactory.

The Semitic is the triliteral kbd. No trace of the d is found in the
three eastern areas. The subject of an initial sound has already been dis-
cussed. We know it only as astronomers know invisible celestial bodies
by the regularity of certain perturbations in their computations. We do
not certainly know what this echo of an initial sound may have been, but we
do know beyond any doubt that it could not have been k, for the mutation
scheme of that palatal is abundantly understood.

There remains to us the consideration of Efate tiateam. I cite Dr.
Macdonald in extenso:

In Efate uateam' (d. uateau) the kidneys is ua ate am', lit., fruit of the liver (or inside)
of the belly {am' the belly); ua-nate-natuo or ua-nate-tuo, the calf of the leg, in one
dialect is denoted by uateau natore, lit., kidneys of the shin (i. e., the leg from the knee
to the foot), and uateau laso denotes kidneys of the scrotum; ua, fruit, is used because
the parts spoken of are round or fruit-shaped.

The validity of this determination of uateam' as pertaining to the ate
liver stem rests upon ua-ate, fruit of the liver, becoming uate through crasis.
In my judgment it is simpler to avoid all reference to ate liver and to find
in uate an identification with fuata, the stronger form from fuat fruit, as
to which see note 360.

322 THE POLYNESIAN WANDERINGS.

277.
aiia (awa), naiia, ua, veins, muscles.

Samoa: ua, sinews, tendons, veins, arteries. Tonga: uoua, sinew,
tendon; uouatanu, deep, out of sight, the veins of the arms.
Futuna: ua, vein, sinew, muscle; uaica, veins which show.
Uvea: ua, vein. Maori : uaua, a sinew, vein, artery. Tahiti:
uaua, sinew, tendon, ligament, vein. Marquesas: uaua, vein,
artery, nerve, tendon. Mangareva: ua, tendon, vein, nerve.
Rapanui: ua, sinew, tendon; uaua, vein. Paumotu: tareua,
tendon. Moriori: uau, artery. Nuguria: nauka, vein.

Hawaii: aa, vein.
Viti: ua, a vein, muscle; vakaua, uauana, muscular, strong.
Laur: urat, vein.

Malay: urat, nerve, sinew. Matu: tirat, vein, sinew. Ilocan:
urat, vein. Tagalog: ugat, id. Pampangas: uy at, id. Kayan:
uat, sinew. Java: wad (through uhat, uat), vein, muscle.
Bugi: urdk, vein. Malagasy: uzatra, huzatra, vein.
Arabic: 'irk', 'araka, veins.
Excising the article a(na) the Efate ua accords with the ua of Polynesia
in form and general signification. In the crude anatomical knowledge of
these races it is easy to see what zia really is, the cordlike bodies in the flesh
which appear under the skin. Thus vein and tendon are the same thing
and one word describes them. It is not until we reach Hawaiian aa that
we find any variant of ua.

Our only Melanesian identification is Laur urat. This is a form which
appears exactly in three Malayan languages. Laur is right in the east gate
in a region where it is possible to suspect a late Malayan source of the word
after the Polynesian migration had passed through.

The Indonesian forms exhibit a remarkable treatment of the ua stem, so
remarkable that we may entertain grave doubts of their identity. There
is everywhere one consonant unaccountable and generally two, the Polyne-
sian stem nowhere exhibiting the slightest trace of any consonant. Kayan
uat is the nearest approach to ua; from that to Malay urat we have a steadily
strengthening chain. These two consonants are such as the Polynesian
would have had no difficulty in preserving had they been primordial.

The Semitic here offered is yet farther removed from ua; it is even
stronger than urat, which only imperfectly does it resemble.

278.
bwa, ua, boua, to rain.

Samoa, Fakaafo, Futuna, Uvea, Fotuna, Nuguria, Maori, Tahiti,
Hawaii, Rarotonga, Marquesas, Mangareva, Rapanui, Manahiki :
ua, rain. Aniwa: ua, towa, id. Sikayana: oua, id. Nukuoro:
mata-ua, raindrop. Tonga, Niue: uha, rain.
Viti: utha, rain. Rotuma: ttas, id.

The following signify rain :

Sesake, Marina, Tangoan Santo, Suau, Maewo: tisa. Dobu, Kubiri:
usana. Kiviri: usan. Sariba: kuse. Vaturanga, Bugotu,
Nggela: uha. Deni: ua. Epi: ua, yua. Baki: yuo.

DATA AND NOTES. 323

Bierian : nihua. Arag, Omba : uhe. Aneityum, Eromanga : ehe.
Guadalcanar : utha. Alite : uta. Santo, Vate, Malekula : us.
Lakon: uh. Ambrym: o. New Ireland (Duffield) : eyus, eus.
Rubi : ura. Buka : urata, uroto, uruotta. Bougainville : urata.
Gog: urei. Arag, Merlav: reu. Pak: wat. Sasar, Alo Teqel:
wet. Lo : weta. Volow : wend. Norbarbar, Vuras, Mosin : wen.
Mota: wena. Wango, Fagani: rangi. Nggao: hani.
Kayan: usan. Togean Islands: udjan. Rotti: udan. Matabello:
udama. Kaili : uda. Malay, Sandol : hujan. Java : hudan.
Gani, Wahai, Salu, Timor, Visayas: ulan. Tobo: u'lan.

Cajeli, Caimarian: ulani. Bual: ulanu. Awaiya: uldne.
Liang, Morella: hulan. Batumerah: hulani. Amblaw : ulah.
Ende : ura. Tidore, Pampangas : uran. Ceram : urana. Sali-
babo: urong. Kisa: ungang. Gah: uan. Malagasy: orana.
Galela: hura. Baju: kron. Teor: hurani. Lariko: haran.
Sula: huya. Bolanghitam : oka. Mysot : golim.
Arabic: ba'a, to rain continuously; ba'a'a, rain, rain water.
The Proto-Samoan stem is uha. This is shown by Viti utha, which might
derive with equal facility from uha or usa, but if the latter were radical
Samoa would have preserved it as usa, and it would have appeared as uha
in the long list of sister languages which represent a Proto-Samoan sibilant
by the aspiration. It does appear as uha in Tonga and Niue, which nor-
mally preserve the Proto-Samoan h. In Rotuma it has become usa and
has then undergone metathesis to comply with the local idiosyncrasy for
closed stems. In Efate the prefixing of b is unexplained.

The Proto-Samoan uha is preserved in Vaturanga, Bugotu, Nggela; with
one vowel change in Arag and Omba, with alteration of both vowels in
Aneityum, and with terminal abrasion in Lakon. Guadalcanar repeats the
Viti utha. The mutation to s is seen in the usa of Sesake, Marina, Tangoan
Santo, and Maewo; the us by abrasion in Santo, Vate, and Malekula; prob-
ably in Duffield's New Ireland eyus, eus. The ua of the modern phase of
Polynesian is found in Deni and Epi, and in the latter island both Baki
and Bierian preface it with weak sounds. The uta of Alite is a degeneration
form of Guadalcanar utha arising from a second borrowing of loan material
and without knowledge of the true form of the original. This is no merely
theoretical deduction. In Fiji I have observed that Melanesians in learning
the Viti almost uniformly reproduce the dh of that speech with t, d, nt, nd,
where the sound (though the Fijians had no knowledge of the fact) was the
preservation of Polynesian stems containing the aspiration. The o of Am-
brym is explicable as a degeneration form to which we are led by Lakon uh.
We must postpone further consideration of the Melanesian material until
we have examined the Indonesian record. We shall compare these forms
with the primal and successive forms hereinbefore developed.
uha. This appears in but a single instance, Bolanghitam oha.
usa. This, too, is found but once, Kayan usan, in which appears the
final Indonesian n which is so prevalent in this area and will not
be further noticed here.
ua. In Gah uan and in Sula huya with the prefix of the aspiration
which is found in Bierian ni-hua and freely in Indonesia.

324 THE POLYNESIAN WANDERINGS.

We now enter upon a great group of forms in which the primal aspiration
has undergone greater mutation. The majority of these mutations are
in the lingual series, a proof that the original is the h near the lingual
closures. In this series we shall consider the weakening mutations, those
upward in the vowel direction, disregarding the initial accretion of h and
the formative terminal n.

ula. Gani, Wahai, Salu, Timor, Visayas, Tobo, Cajeli, Caimarian,
Bual, Awaiya, Liang, Morella, Batumerah, Amblaw. Mysot
golim may be considered in this series, but with no great insistence.
ura. Ende, Tidore, Pampangas, Ceram, Salibabo, Malagasy, Galela,
Baju, Teor, Lariko.
We next examine the strengthening mutations, the downward movement.
The first of these is usa, already presented as a normal variation.
udja. Togean Islands, Malay, Sandol.
uda. Rotti, Matabello, Kaili, Java.
We have a single example in which occurs h-ng mutation ; this represents
a resultant of the first weakening variation to ula and then the horizontal
strengthening by movement to the palate, l-ng being a mutation of which
the tabular view will supply examples in Melanesia. These are to be
regarded as specifically Indonesian mutations; they are not found at all
in Polynesia, and h-j and h-d very sparingly in Melanesia outside this stem.
We shall now resume the consideration of the Melanesian material. The
ula-ura which has occupied so large a space in Indonesia may with consider-
able probability be identified in Buka and Bougainville urata and variants.
In case this identification be considered acceptable I would ascribe it to a
Post-Polynesian period, a raid along the ancient track conducted by Indo-
nesian peoples, for these two northern outliers of the Solomons are within
reach of the prahus. In note 277lhave already introduced this explanation.
But Gog urei and Arag-Merlav reu, metathetically derived therefrom, lie
geographically too remote to be susceptible of this explanation in view of
the fact that no intervening record appears. But is such the case?

In Buka uruotta, readily seen to be a variant of urata, we find the pos-
sibility of a transition form, a stem in which the final / is established for
this area. It is possible that in further borrowings uruotta, losing its first
two letters (not a true syllable), might, as uotta, lead us to weta of Lo, wet
of Sasar and Alo Teqel, wat of Pak. From wet to Volow wend is easier than
it might seem, for wet varying toward wed would encounter the idiosyncrasy
that Volow and several other Melanesian languages, as well as Viti, can not
essay the d without the supporting preface of the nasal of the same series.
From wend to wen and wena is very simple on the explanation of secondary
borrowing, as already pointed out in the case of h-th-t, the reinforced con-
sonant being objectionable to later borrowers, and on that score they reject
the final one of two linked sounds in their ignorance that this it is which
represents the primal sound. Thus we may bring the second group of
Melanesian identifications into line, representative of a Post- Polynesian
movement.

The third Melanesian group, Wango-Fagani rangi and Nggao hangi, is
the Polynesian langi the sky, which we have already seen (274) pressed into

DATA AND NOTES. 325

Melanesian service to denominate the wind. The reason is not far to seek,

for it's the wind and the rain and the sky above us all, for it's all one and

naught lies beyond but the great, the pitiless gods.

And this Semitic, foolishly treading on the heels of so wide a unity, has

but this plea to make, that Efate has added to uha a b, which nowhere else

appears to have left a trace.

279.

bwabwa, a hollow, channel or bed of a stream dry except after heavy rains,
an opening through the jungle, a board ; bwala, level.
Samoa: Papapapa, the name of a rocky path among cascades in
the bed of a stream just south of the tuasivi on the alasopo
from Apia to Safata; papa, a rock, a board, plane, level, flat;
papapapa, level, as a rocky road. Tonga: papa, a board,
plane, even (as a road much trodden) ; papapapa, smooth.
Uvea, Futuna : papa, a board. Niue : laupapa, a board, a floor.
Nuguria: papa, rocks. Maori: papa, anything broad, flat and
hard, to lie flat, a flat rock, a slab, a board, a door, a shutter.
Tahiti: papa, a board, a flat rock. Hawaii: papa, applied to
many substances having a hard flat surface, a board. Mangaia :
papa, a base, a foundation. Mangareva: papa, a foundation,
a plank. Marquesas : papa, stones on the shore ; papahua, a
board ; papapoho, a plank, a gate, a door. Paumotu : papa, a
rock ; tipapa, lying flat. Nukuoro : papapa, flat ; papapapa, low.
Viti: mbamba, a board.

Motti : papapapa, a flat rock. Mota : taptapapa, flat-sided ; tapa, a

board ; papalak, papalaota, thin. Aneityum : apalapal, thin, flat.

Malay, Basakrama: papan, a board. Macassar: papang, id. Bugi:

papon, id.
Arabic: baba-t', surface, board, table, slab; bib', channel; bab', door,
gate, hall ; baba, to dig a hole. Hebrew : babah, a gate.
A particularly interesting coincidence of specialized significations lies in
Efate bwabwa and the Samoan Papapapa.

In the Polvnesian which we can mass this stem appears as a duplication
of pa. But Efate bwala, paralleled by Aneityum apalapal and Mota papa-
lak and papalaota, very strongly suggests a stronger closed stem pal. The
other Mota forms seem to be composites of ta and pa, for taptapapa is
explained by Codrington as a duplication form, although highly irregular
when referred to Polynesian duplication mechanics.

If this pal stem be acceptable it interposes difficulty in the way of Semitic
identification through form resemblance.

280.
balu-sa, to paddle, row; balusa sa, paddle or row with it (a paddle or oar).
Epi dd. mbeluo ka,mbahuakin,v . t., Aneityum aheleth, to paddle,
to row, to sail, Ambrym fuloh, to paddle, Fijian ai vothe, an
oar, vothe, to paddle, to row, vothe-taka, v. t. ( = balu-saki, to
paddle a canoe, to row a boat), Paama palusa, Malekula d.
masu, Malekula Aulua sua, Malo mo sua, Tanna asua, Futuna
sua, Malagasy vui, act of rowing ; mivui, to row ; vuizina, rowed ;
fivui, an oar; Malay dayung, an oar; dayung, bdrdayung, to row.

326 THE POLYNESIAN WANDERINGS.

Note: Balu-saki is the same as vothe-taka. The verb " to row " is balu, vothe, (m)beluo,
(m)bahua, vui, masu, and without the preformative b', (v' , m'), asua, sua, dayung, and
the I in balu, th in vothe, h in mbahua, s in sua, d in dayung, all are variations of the
same original consonant which is elided in vui. The word for oar, ai vothe, fivui, is in
Efate uose, d. uohe (wose, woke), Futuna joi. In Futuna the connection between sua,
to paddle, and foi, an oar or paddle, is not so apparent as that between Malekula Pang-
kumu su, to paddle, and bos, a paddle, because in joi, as in vui (Viti vothe) the s has
been elided; and the connection between Efate balu, to paddle, and uose, a paddle, is
not so apparent as that between Epi mbahua, to paddle, and voho, a paddle, Epi d.
bahua, to paddle, and boho, a paddle. Arabic gadaja, kadafa, (or 'at'afa) , Amharic kaza}
(or 'azaf), to propel with oars, to row, Modern Arabic kaddaf, or ' addaf, part, mo' addij
(anc. mo'addif, or mo'azzij, cf. vothe, bose, uose, vui, foi). Sua is without the preforma-
tive, cf. 'azafa, 'addaf: balu seems to have the same prefix as Samoa pale, to row, with-
out which is Samoa alo (ps. alofia), and alo-fa'i, to paddle, row, and with another verb,
Samoa taualo, to row, to keep on rowing.

The foregoing is cited from Dr. Macdonald's work, an excellent example
of his dictionary method : I subjoin a few notes from the Melanesian tract
as a slight addition to the record.

Mota: sua, to paddle; suava, a paddling. Buka: sschue, soa, to
row. Matupit: walua, to row. Baravon : wahie-vue, to row
off. Lambell: valiso, to paddle. King: vulusii, id. Lamassa:
iavds, id. Moiki: ango, to row, is Polynesian, Samoan alo,
under the influence of the local change of / to ng.
281.
baro-si, baru-si, to rub, to grate ; farofaro, a thing which rasps.

Samoa: valu, to scratch, to scrape out (as coconuts) ; valu, to scrape
(as taro); valusanga, scrapings (as of taro). Tonga: valu, to
scrape. Futuna: valu, to rub, to rasp, to scrape. Nuguria:
valuvalu, to peel taro. Tahiti: varu, van, to shave, to bark
a tree, to scratch. Mangareva: varu, to scrape fruit, to cut
the hair. Rapanui : varuvaru, to scrape, to rasp, to shave, to
plane, to peel; hauhau, to scrape. Marquesas: vau, to shave;
vavau, to scrape cooked breadfruit. Maori: warn, to scrape,
to shave, to cut hair quite close; waruwaru, peeled. Hawaii:
walu, to scratch, to rub, to rasp, to polish. Niue: halu, to
scrape, to peel. Fotuna : wurusia, to scrape.
Viti : voaluya, to rub or scrape pandanus leaves to render them pliant
for mat-making ; wandrutha, to clear the nose of mucus, to wipe
dirt off a thing with the hand; varota, to saw, to file, to rasp.
Mai : barusi, to scrape.
Malay: paras, to shave, to pare close to the surface. Malagasy:

fara, to scrape, to scratch, to make smooth.
Arabic : faraka, to rub, to grate.
The Proto-Samoan stem is varus, which reappears in Efate, Viti {wandru-
tha), Fotuna, Mai and Malay. While the two Samoan words are akin in
sense it would seem that they have different stems, or perhaps that valu
is a modification by vowel prolongation entered upon when the closing con-
sonant of the stem had dropped from memory; the evidence for this lies
in the objective aspect of the two verbs, of valu valusia, of valu valua. On
the other hand the noun of action derived from valu, taro scrapings, is
recorded by Pratt in the two forms valunga (without the macron) and valu-

DATA AND NOTES. 327

sanga . Yet as this distinction of special use by quantity is not carried along
even into Nuclear Polynesia we may be justified in considering this an error
on the part of Pratt, the more since Pere Violette marks diacritically no
such distinction.

The fact that the valu stem carries the specific action of the rasp (a sprig
of coral being the common implement) through Futuna as far as Hawaii
removes any obstacle which might exist as to the affiliation of Efate faro-
faro and Viti varo with valu. It is interesting to find in Viti three phases
of the same stem. The varying treatment of the radical I suggests that
Viti borrowed and twice repeated the borrowing from different sources.
The varota points to a loan when the stem held its r, and Efate farofaro
hints at the loan coming along the migration track which I have segregated
as the Viti stream, but in this the closing radical s had vanished. The
wandrutha points to a loan from a source which had rgrasseye and the closed
stem, in local terms to Samoa. Then walunga indicates a source wherein
the stem was open and the r had passed definitely into /. In our knowl-
edge of the Polynesian waves over Viti this corresponds to theTongan.

The Semitic is triliteral frk. There is a sense resemblance, but stated
as a triliteron the Polynesian is vrs, the difference in the third element
being considerable.

282.
bau, bao, fau, fao, new.

Tonga: foou, foofoou, fofoou, new. Futuna: fo'ou, new, recent.
Uvea: foou, id. Samoa: fou, new, recent; fa'afou, to make
new. Niue: fou, new, young, recent. Aniwa, Vate: fou,
new. Fotuna: fau, id. Maori, Nuguria, Hawaii: hou, new,
fresh, recent. Mangareva, Tahiti, Nukuoro: hou, new. Mar-
quesas: hou, new, recent. Rapanui, Paumotu: hou, young,
new. Rarotonga: ou, id.
Viti: vou, new; vovou, young. Rotuma: fo'ou, new.
Eromanga: ite-vou,new. Malo: baro, id. Malekula: meaner, id.
Motu: matamata, id. Baki: bou, id. Bierian: feu, id.

Aneityum: mat, new, raw. Taupota, Wedau, Galavi: vou,
new. Kubiri, Raqa: baubau, id. Kiviri, Oiun: bobu, id.
Malagasy: vau (havauzana), new. Malay: baham, id. Kisa:

wohruwohru, id.
Arabic: mahdut' (Jiadat'a), new. Hebrew: hadas', id. Syriac:
hdat', id. Ethiopic: hadas, to renew.
No explanation is offered in the Efate dictionary as to the use of the
diaeresis on the former vowel. Yet we may be justified in assuming that
it is clumsily an indication of long quality, for in Tonga, Futuna, Uvea, and
Rotuma the doubled 0 is used for the same purpose.

This stem is soundly identified throughout Polynesia, in Efate, in Ero-
manga, in both the Baki and Bierian of Epi, in Malagasy, and less clearly
in Malay. Tregear's inclusion of the Kisa -wohruwohru seems somewhat to
reflect Malo baro; but in the great facility with which our Sawaiori retain
l-r we are not warranted in admitting this to the company of fou. The
Malekula mermer may be a variant of baro, yet that is hardly likely, since
in note 207 we have seen that p-m is only a rare possibility.

328 THE POLYNESIAN WANDERINGS.

In Motu and Aneityum we find the intrusion of another stem, yet even
this intrusion has both interest and value. Aneityum mat new, raw, sup-
plies a sense link between Motu matamata new and Polynesian mata (256)
raw. The position of these two intrusions is very significant. Motu in
Torres Straits establishes for us one point of the migration through the
south gate; Aneityum is in that more southern region which is a second
determining point in the Viti stream.

The Semitic stems in the triliteron hdt (hds) and it is impossible to
bring into association therewith a stem whose only consonant is a sliding
labial centered at/.

283.

bau-si, fau si, bau-fau, to fasten together, to plait a mat.

Samoa : fau, to tie together, to fasten by tying, the tree {Hibiscus
tiliaceus) whose bast is used for cord, the kava strainer made
therefrom, strings in various uses; fafau, to lash on, to fasten
with sennit; faufau, to fasten on, to tie together. Tonga:
fau, to fasten up the hair, the name of the hibiscus, the kava
strainer made therefrom ; faufau, to fasten on the outriggers of
small canoes; hau, to fasten to; fehaiiaki, to tie. Futuna:
fau, the hibiscus, the kava strainer ; fah, fafah, faiifait, to attach,
to tie. Niue : fau, fafau, to make by tying. Fotuna : no-fausia,
to tie, to fasten. Tahiti: fau, the hibiscus; fafau, to tie

together. Paumotu: fau, the hibiscus. Nuguria: hau, id.
Maori : hou, to bind, to fasten together ; whan, a shrub ; whauwhau,
to tie. Hawaii: hau, name of a tree with a practicable bark.
Marquesas: hau, the hibiscus. Mangareva: hau, id.; hahau,
to join or tie with cords. Nukuoro: hau, the hibiscus, a
garland. Mangaia: aw, the hibiscus.

Viti: vau, the hibiscus; vautha, to bind together.

Aneityum: in-wau, a creeper, a vine. Mota: vau, a pandanus;
to mat, to plait, to weave.

Malay: baru, the hibiscus. Java: waru, id. Kisa: warau, id.
Malagasy: fehi, fehizana, to tie, to knot.

Arabic: habaka, habikaf, to weave, to bind, to intenveave.

The Proto-Samoan stem is faus.

In the utter absence of perspective in which these languages appear
before us it would be idle to engage upon the attempt to discover whether
in sense the tree or the act of using its bast is primordial. In the records
before us the stem carries the tree sense without the verb in the Paumotu,
the Marquesas, Nukuoro, and Aneityum ; nowhere the verb where the noun
does not designate a plant which yields a string. Niue and Fotuna are the
only instances of the verb without the noun, and these may be negligible
since the vocabularies on which we rely do not attempt to name this tree.
For comparison I add the following hibiscus names from Micronesia on
the authority of Mr. Christian's researches.

Ponape: kaVau. Mortlocks: kili-fau. Ruk: sili-fau.

Pulawat: kini-fau. Satawal: kini-fau. Lamotrek: gilt- fau.

Sonsorol: giri-fai. Uluthi: gili-fai. St. David's: gini-fai.

DATA AND NOTES. 329

In Indonesia we encounter a stem with an intrusive consonant and, as
in 282, we can not accept the identification. The consonant skeleton of
Malagasy fehi, fh, readily associates with the Polynesian fs. But this
should not lead us astray into the idea that we have an identification of
stems. If ati were diphthongal and e a very long vowel it would be a great
task to entertain the possibility of a vowel mutation au-e. The Malagasy
requires even more than this: it requires the extirpation of an inner and
protected vowel, for fau is not diphthongal and the proof thereof lies in
the preduplication fafau in Samoa, Futuna, Niue, Tahiti, Mangareva.

The Semitic hbk is certainly far away from the fs skeleton.

284.

bila, bilafila, to shine, to lighten, to gleam, to flash, to appear. Cf.
fill 295.

Samoa: pula, pupula, to shine. Futuna: pula, to fix the eyes
on, to regard fixedly. Niue: pupula, to shine (of the eyes),
the new moon. Uvea: pupula, to shine, to glow, to gleam,
to beam. Maori: kapura, mapura, fire. Tahiti: pura, a
flash of light or fire, to flash, to blaze. Mangareva: pura,
a spark, to shine, to glow. Rapanui: pupura, to shine, to
be bright; hakapura, to kindle, to light. Paumotu: pura,
phosphorescent. Nukuoro: purapura, bright. Marquesas:
pupua, phosphorescent, shining.

Viti: vula, white; vula, the moon. Rotuma: hula, hual, the moon.

Nifilole: polao, the light. Bugotu, Nggela : pura, white. Santo:
pula, light. Baravon: puapua. id. Mota: vula, white.

Nggela, Belaga, Marina, Arag, Mota, Vaturanga, Bugotu: vula,
the moon. Pokau: vula, id. Uni: bulo, id. Pokau: vuia, id.
Uni: buia, id. Doura: huia, id. Motu: hua, id. Keapara,
Galoma: hue, id. Rubi: wui, id. Omba: vule, id. Merlav,
Gog, Lakon: vul, id. Moanus: mbul, id. Maewo: wula, id.
Vuras, Mosin, Motlav, Volow: wol, id. Wango: hura, id.

Ambrym : ola, id. Aneityum : laav, lav, to shine (of the moon).

Malay, Ilocan: bulan, the moon. Tringanu: bulang, id. Solor:
burang, id. Molucca: bulam, id. Kaili: bula, id. Gilolo,
Solor: wulan, id. Aru: fulan, id. Rotti: fula, id. Lobo:
furan, id. Timor: funan, id. Togean Islands: fuya, id.;
buya, white. Kisa: ulang, moon. Magindano: ulan, id.
Utanata: uran, id.

Arabic : barak, bara', to shine, to gleam, to flash, to glitter, to appear,
to lighten, to open the eyes, to glance at.

The Proto-Samoan stem is pulaf.

Dr. Macdonald's identification of this and the lightning stem is, of
course, disposed of by the comparison of this material with that assembled
in item 295.

In terms of color psychology we are here dealing with the unreduced
ray of light, the union of the spectrum colors in white.

Furthermore we are dealing with a singularly elemental people, but little
advanced in the arts, not yet emerged from the stone age. The back-

330 THE POLYNESIAN WANDERINGS.

ground of their color life is brown — brown mats to sleep upon, brown thatch
to shield them from the gods and from the rain, these the browns of mar-
cescent leaves, brown tapering trunks of coconuts, the brown of seasoned
hard wood in the club without which in hand life is not long to be lived,
brown skins to look upon in love or hate and to live within. Not merely
the deductions of philology these remarks ; the eye trained in color values has
looked upon these scenes, and those who have gazed upon John LaFarge's
sketches from the South Sea have seen a suite of studies in brown.

Their simple arts rarely produce a white; I recall but the white of siapo
made from the bast of Broussonetia papyrifera or Pipturus incanus when
retted in still waters and bleached in the sun. It is beyond the resources
of their few pigments to produce a synthetic white. The red of a few
mineral oxides, a purple from a sea mollusc, the yellows of the turmeric, the
black of soot, are all that so advanced a people as the Samoans might spread
upon a palette. Mix these and brown residts.

Bathed in white light they see little white in nature. Their eyes no
more than ours may rest upon the sun undazzled by its glory long enough
to separate from its heat and glare any true sense of color. But there is
one object which never fails to yield a white, the moon. Therefore we are
not surprised to note in what number of tongues of these remote islands
the word which means to shine, to be bright with the undissolved light
beam, should be used to designate the moon. There are fewer exceptions
than would appear. Niue alone among the Polynesian tongues in this
record has the moon in this pula shine. But a glance farther along at the
notes upon item 342 will show that these languages which here omit the
moon present it in that record as masina, again the shining one, the most
conspicuous object in the perfect light, composite white.

Few of these forms in our three nesiote areas but are self-explanatory.
In Rotuma we find the metathesis so characteristic of that language. This,
too, I am convinced is found in Aneityum laav-lav.

The Semitic triliteron is brk. While the resemblance in sense is very
close and there is a resemblance in form covering two elements of the
consonant skeleton, I hesitate before accepting so violent a mutation as
that involved, f-k, when we put alongside of brk the Proto-Samoan pit.

285.

bong, black, dark ; bongi, bong, darkness, night, day in calendar reckoning ;

bongien, darkness ; bong, a dark black powder used in painting.

Fakaafo, Niue, Uvea, Vate, Fotuna, Tahiti, Manahiki: po, night.

Samoa : po, night, day in calendar, blind ; pongipongi, twilight ;

pongisd, darkness. Futuna : po, night, day in calendar; pongia,

benighted. Maori: po, night, a season; pongia, benighted.

Hawaii, Mangaia : po, night, darkness. Marquesas : po, night,

day in calendar, darkness. Mangareva : po, night, darkness,

obscurity. Rapanui : po, night, late ; po rah, day in calendar.

Nukuoro: po, pongi, night. Paumotu: matapo, blind (as

elsewhere in Polynesia) ; potangotango, darkness. Aniwa : kopo,

night; pouri, dark. Bukabuka: popo, black. Tonga: bo, night.

Nuguria, Sikayana: bo, po, night, dark.

Viti: mbongi, night; mbombo, blind. Rotuma: boni, night.

DATA AND NOTES. 331

Nggao, Belaga, Nggela, Vaturanga, Bugotu, Omba, Sesake: mbongi,
night. Fagani, Alite, New Georgia, Guadalcanal-, Epi : bongi, id.
Sinaugoro, Galoma, Kiriwina : bogi, id. Marina, Southeast Epi :
pongi, id. Keapara, Galoma, Rubi: pogi, id. Western Epi:
ombongi, id. Bierian: im-bong, id. Malo: bong, id. Savo:
/>0W<7, id. Duke of York, Matupit, Kabakada, Pala, Baravon:
bung, id. Malekula: ambling, id. Tanna: la-ben, id. Santo:
/>ow, id. Santo Wulua: />om', id. Maewo, Gog, Motlav,

Norbarbar: kmbwon, id. Arag, Vuras, Lakon: kpwon, id.
Mota: mpwon, id. Lo: kwon, id. Volow: nggmbwon, id.
Aneityum: />twg, night; aping, black; poing, dark. Motu,
Rubi: 6oi, night. Dobu: boiboi, id. Vaturanga: fcora, black.
Baki: bongian, night. Deni: m&o, black; w&w, night. Nggela:
pungi, darkness. Bugotu: puni, darkness. Bugotu: puni,
dark. Buka: abung, evening; boni, night. Pokau, Sariba,
Tubetube : boni, night. Baravon : bobotoi, darkness. Lamassa :
mbung, dark. Moanus : pong, vong, day in calendar. Santo :
pon, id. Baki: bongo, id. Bierian: bong, id. Eromanga:
po-arap, evening. Murua: bogu, night. Kiviri: pom, id.
Mukawa: pomai, id.

Java, Salayer: bungi, night. Macassar: boetta, dark; boeta, blind.
Teor: pogaragara, night.

Arabic: fahuma, to be black; fuhum, black; fahma-t, night.

(Darkness — Samoa: pouliuli. Niue, Futuna, Nuguria: pouli.

Tonga : bouli. Rarotonga, Manahiki : pouri. Tahiti : poiri.
Aniwa: ouri (vide sup.). Hawaii: poeleele. Marquesas:
potana (vide sup.). Fakaafo, Vate, Maori: po.)

The Proto-Samoan stem is pongis, which appears in Samoan pongisa.
By a continuous process of terminal abrasion we reach successive operative
stems in pongi, pong and po, all of which will readily be found in this record,
and all except the closed stem pong in Polynesia.

The primal sense is the absence of light, just as the common name for
day, aso, is mediately a sun term and ultimately a word for light. This
absence of light is most conspicuous in life by the night ; a very easy tropical
change establishes the night word out of the dark word.

It marks an advance in capacity for thinking in analogies to recognize
in black color and in the black pigment, in which color knowledge first
becomes practical, the absence of color, therefore the absence of light.
Physiology drawing upon optical physics and the psychology of the color
sense in equal drafts is but lately beginning to discover that some rod and
cone structure of the retina is responsive to color stimulus and other to the
varying stimulus of the amount of light. Yet certain of our islanders
have seen the analogy ; Efate bong is the black pigment.

To all comes alike the darkness after the day. In the crudity of island
life there may come to almost anyone the accident of a life all warfare
which may bring night upon the sight, the sunset forever of the blind. To
far too many helpless innocents comes ophthalmia neonatorum. Hence po
darkness (more commonly in the composite matapo eye-dark) means also
blind.

332 THE POLYNESIAN WANDERINGS.

Daytime is life time, its joys are joys only when the eye can see clearly.
When the shadows snap suddenly upon the swiftly fading glow of the sunset
the cruel gods are abroad, man in black terror cowers with his kin in a
gloomy home about the fat smoke and dull glow of his string of kindled
candlenuts. From the dusk of evening until the eastern gleam he will not
venture from his shelter except that, under the pricking of his scatomantic
dread, in some middle hour he slinks out under the wheeling stars and to
the beach where in the darkness and the swirl of the tide not even the
whistling gods may see that which might work him harm. Man may live
heedless days, but when it comes to the reckoning it is the nights he counts,
the nights he has lived through. Here, there, and well nigh everywhere in
the Pacific world, po as night means calendar day. The Maori goes yet
farther: po may mean to him a season, it may mean the obscurity of
eternity.

We now pass to the consideration of the forms of this stem beyond the
Polynesian limits. Our records are too scanty to reveal in Melanesia the
existence of the pongis stem, but pongi is very common, pong is of equal
frequency, po is rare.

We observe a group of forms in which the initial consonant is subjected
to a wide yet systematic variation. This peculiarity is known as the Mela-
nesian q. It is a composite of k and b and w; in this composite k may
become ngg, b may become mb or p. It is not to be interpreted as the effort
on the part of Melanesians to compass an unwonted Polynesian sound, for
it is of far greater frequency in Melanesian words for which we can find no
affiliation with Polynesia. Rather are we to regard it as showing the
struggle in sound evolution by a primitive people in the genesis of their
speech who are coming into first possession of a labial mute and whose
untrained buccal muscles reveal to us the wrestling. And if we can thus
look upon the birth of one sound in human speech may we not indulge
ourselves in the fancy that man went through a similar struggle to acquire
each of the consonant sounds in his phonetic system? At last across ages,
uncountable ages, we see the first of speaking men lifting himself above the
crying animal when he is teaching the muscles of his tongue to move its
tip here and there within the mouth to give him some clumsy / or fugitive
;■ which is to give him speech, and upon speech knowledge, and perhaps to
some chosen ones of his remote descendants the promise of intelligence.

It is not without interest that we are in a position to observe the Mela-
nesian rounding out his phonetic range by this last gift of the labial mute,
even though it strike our ears as anything but sweet concord of sounds.
In the Polynesian material here under discussion we encounter the Mela-
nesian q but twice besides its presence in the variation of this stem, namely :
naja (157) a drum Malekula Uripiv nambivi; and 'upenga (151) a net
Motlav kmbweng Volow nggmbiveng. An index of its frequency in Mela-
nesian will be seen in the fact that in Codrington's Mota dictionary there
are no less than 337 principal entries under ^-initial, and yet Mota has
already attained to the possession of v and p.

The Aneityum forms give in poing (perhaps a 1 243 metathesis of pongi)
the transition to ping, the only wide variant from the radical 0. Motu boi is
bongi with elision of ng; this is found so uniformly (309, 332, 346, 350) that

DATA AND NOTES. 333

we have but a single instance (151 ng-n) in this language of any other
treatment of the palatal nasal, and that a doubtful one. Vaturanga bora,
Baravon bobotoi, and Eromanga po-arap are evident composites on the po
stem. Deni is the only language embraced within our inquiry in which the
attempt has been made to differentiate po sense variety by vowel variation.

In Indonesia we find the pongi stem in Java and Salayer, the other identi-
fications are po composites.

Revert now to our Proto-Samoan stem pongis, skeletonized to pngs.
What can Semitic f hm have to do therewith ?

286.
bon, bono, to be shut, closed, stopped ; bono-ti, bonu-ti, bunu-ti, to shut,
to close, to plug, to stop, to block up.

Samoa: puni, a place inclosed to catch fish; pupuni, to shut, to
inclose; punitia, to be shut up, inclosed. Niue: pupuni, to
shut in, to inclose; ponoti, to cork, to calk, to stop up, to
close. Uvea: pupunu, to shut in, to inclose. Nuguria:

hakapunu, to stop up. Maori: puni, stopped up. Tahiti:
puni, to be inclosed, to hide. Hawaii: puni, to inclose.

Mangaia : puni, to hide. Mangareva: pupuni, id. Paumotu:
punipuni, to take shelter. Marquesas : pupuni, punipuni, to
hide oneself, to squat. Nukuoro: pono, to shut. Tonga:
buni, closed, shut; bubuni, bunibuni, bubunu, bunubunu, to
close, to shut. Fotuna: buna, to block up.

Viti: vuni, hid, concealed; mbonota, vonota, to stop up water by a
dam; vono, a canoe plank.

Epi: mbin, to block up. Mota: wono, ivonot, to close, to fill up,
to make solid, to oppress.

Malay: buni, to hide, to conceal.

Hebrew : baham, bahan, to shut, to cover, to conceal.

The Proto-Samoan stem is punit.

The insecurity of the vowel in the Polynesian is of no little importance.
Niue ponoti is Samoan fonoti. Uvea gives us punu, Tonga bunu for puni.
Nukuoro pono recalls Niue. Fotuna buna is a yet wider variant. In Viti
we find not only vuni for puni, but in the homogenetic of fono we have vono
and mbono, which again recalls Niue. Epi mbin is identifiable.

Efate has not the exact analogue of puni, which would be buni. But
its bono recalls Niue, its bunu recalls Tonga.

It is not until we examine the Mota words that we begin to see the reason
for all this shifting; wono, wonot has the fono form but the puni sense. It
functions as a transition form to account for all this variation heretofore
noted ; it points to a time when punit and fonot were themselves divided
by so slight a distinction as to escape alien notice. The Polynesian has
yet another root meaning to plug, to stop up, monot (71). These three
have the common factor not, or nit as a slight variant. The conclusion
is irresistible that the general sense of closing inheres in not, that the other
and variable elements are in some sense limiting. Now since we find pu
in the sense of a self-existent hole, pu-nit is hole-closing when the hole is
always there. The fo-stem we find in foa, to chip as a hole in an eggshell,

334 THE POLYNESIAN WANDERINGS.

to break the head; fo then is a hole not self-existent, but the result of acci-
dent or design ; fo-not is hole-closing when the hole has been made where
it does not belong, therefore fono is to patch. The root mo we find in moa
and more specifically in moalili, the soft flesh in the round hole which
appears when the operculum is prized out of the shell of Turbo petholatus,
momo'ulu clitoris (cf. Maori horn a loop, bight, or fold) ; mo-not is the closing
of a round hole, putting a plug into the open eye of a coconut as the com-
mon water-container.

Malay bum, in form close to the Polynesian punt, has a sense, a particular
result of closing upon the object included, which reappears not in Proto-
Samoan but in the Tongafiti migration in Tahiti, the Marquesas, Mangaia,
Mangareva, the Paumotu.

If we confine ourselves to the Proto-Samoan stem punit, pnt is in no
agreement with Semitic bhm, far less is the not root which has just been
evolved from these several forms.

287.
borau, to ride, to be carried (on a canoe, ship, horse, vehicle or other thing),
to voyage.
Samoa: folau, a voyage, a ship, to go a voyage, to die. Tonga:
folau, a voyage, a fleet, a voyager, to sail. Futuna: folau,
a navigator, to navigate, to go a voyage. Niue: folau, to
commit suicide by jumping into the sea. Uvea: folau, to
navigate. Fotuna: ko-forau, to go a voyage. Moriori:
wharau, a ship. Rapanui: horau, to hasten, to run to.
Viti : vondo, to embark, to go on board, to ride.
Matupit: parau, a ship. Ambrym: bulbul, a boat. Malekula:
foro, to navigate. Mota: walawalau, to paddle all together;
walaua, to collect things for a voyage.
Tagalog, Visayas: parau, a boat. Malay: prahu, prau, id.
Arabic: markab', a ship, a vessel. Ethiopic: markab, id.
The sequence of the Efate definition is, of course, inverted, the voyaging
sense being primal, and the extension of the definition as "to be carried
. . . on a horse, a vehicle" can have had significance only since the imme-
diately modern introduction by Europeans of horses and their concomitants.
A strange admixture is twice found in Nuclear Polynesia. The word in
vSamoan which means to sail means also to die. In Niue the only signification
of folau is to commit suicide by jumping into the sea. The vocable is of the
most obscure ; probably a distinct word in all but form is involved, for we
can not imagine any seafaring race so pessimistic as to see death the goal
of their voyaging and thus to develop such a secondary significance.

In Melanesia we find two distinct epochs of the stem. The earlier and
more worn phase is shown in Malekula foro. This forms an easy transition
to Viti vondo so far as the vowel element is concerned, and the l-nd muta-
tion is not without warrant in our table of consonant mutation. Anei-
tyum bulbul is a form which to abrasion adds vowel degeneration. The
Mota walau is nearer to the folau stem in form, its two senses are easily
recognizable particulars of the general meaning, and, so far as the general
theory of duplication may be considered to hold in Mota, the duplicated
walawalau shows the division of the stem elements as fola-u. Matupit

DATA AND NOTES. 335

parau is of most modern type ; in form and meaning it is exactly Indonesian ;
I have no hesitation in assigning it to the Post- Polynesian period and to
Malaysian rovers, all the more since Matupit lies in the jaws of the eastern
gateway. The explanation of Moriori wharau eludes our search. With
this exception the word is Nuclear Polynesian, and Moriori is a curious sur-
vival of an uncomprehended migration earlier than the current Maori, before
whose coming it was driven onward to the chill of the Chatham Islands.

It is impossible to reconcile with this stem of fl the Semitic mrkb. Not
only is the latter of a far more complicated nature, but the only point in
which it suggests fl is in the r and the b, and these are not only in an order
the reverse of fl but they are parted as well by k, one of the most permanent
of consonants as it is one of the first acquired.

288.
bulo ki, bulosi, bulusi, bulisi, to turn, to twist ; tafolo, to be turned, twisted ;
tafulus, to be turned; bolonga, tafllonga, to turn itself (as a
thing sinking in water) ; bulora, filora, twisted, confused (as a
lot of things turned or twisted about).
Samoa : full, to turn over, to capsize ; fulifuli, to roll over and over ;
mafuli, to be turned over, upset ; tafuli, to turn over. Tonga :
fulihi, to turn over, to upset, to reverse ; mafuli, to be capsized ;
tafuli, to move around. Futuna: fuli, fulisi, to turn, to over-
turn, to reverse ; mafuli, id. Niue : fuluhi, to turn, to overturn ;
mafuluhi, turned over. Uvea: fuli, fulihi, to turn over, to
overthrow. Hawaii : huli, to turn in general, to turn over and
about. Maori: huri, to turn around, to overturn, to roll over;
hurihuri, to turn over in the mind, to ponder. Tahiti: huri,
to turn over, to roll. Sikayana : huri, to turn over. Rapanui :
huri, hakahuri, to turn over; tahuri, to pirouette. Rarotonga:
uri, to turn over, to roll over. Marquesas: hut, to turn over,
to roll.
Viti : voli, to go around ; volivoli, to revolve.

Aneityum : uhuri, to dig or root as a pig. Motu : huro, a grindstone.
Malay: pulas, to wring, to wrist, to turn, to turn round, to turn aside
out of the way. Java: pulis, id. Malagasy: vorivory, round,
circular.
Hebrew: palas, to turn round, to twist, to spin. Arabic: falakat,
a spindle.
The Proto-Samoan stem is fulis.

We are therefore in no doubt as to the Efate forms involving the sibilant.
The remaining forms may with varying degrees of certainty be connected
with forms in s.

In Polynesian, as is quite common, the fulis stem is readily discoverable
in Nuclear Polynesia. The Tongafiti languages use the abraded form and
betray no recollection of the earlier closed stem.

The Viti voli is not quite satisfactory as an identification. Its definite
transitive is voli-ta. This may suggest a stem volit, yet in the Viti verb we
can never be quite certain that ta may not have acquired sufficient identity
as a mere termination to be applied to any open stem without consideration
of what former closing consonant may have been abraded.

336 THE POLYNESIAN WANDERINGS.

Our Melanesian material is so scanty and so widely separated, in form
and in sense so uncertain, that we include Aneityum and Motu without
argument, but solely on the score of resemblance.

In Indonesia, on the contrary, we find in Malay and in Java the most
satisfactory identifications, and Malagasy quite probable.

The Semitic pis and Polynesian fls are distinctly coordinated through
Malay pis. They may therefore be accepted as at least resemblances.

bulu-ti, to plaster, to overspread with some sticky substance (as oil, lime,
paint, pitch), to cover with a plaster or poultice (as a wound) ;
nabulu, a plaster; bubulu, bulubulut, buloki, mabulu, to be
sticky (as a plaster).

Samoa : pulu, gum, breadfruit gum, glue, coconut husk, resin ; puluti,
to glue, to pitch ; pupulu, to apply gum, to interpose, to mediate
(cf . puluvanga) ; pulupulu, a large cloth about the whole body ;
pulupuluta'i, to cover up so as to take care of. Tonga : bulu,
a gum used in calking canoes, coconut husk ; buluji, birdlime,
paste, plaster, to paste; bulubulu, a shawl, a cloak, to cover the
back and shoulders ; bubulu, slimy, adhesive. Futuna : pulu,
coconut husk, gum, resin, pitch, all sticky stuff; puluti, to
cement, to plaster, to pitch; pupulu, pitch; pulupulu, cloak,
wrapper. Niue: pulu, coconut husk, fiber, sennit. Uvea:
pulu, pitch. Hawaii: pupulu, adhesive, soft; pulu, scales

from fern fronds used as plugging or stuffing material. Maori :
puru, a plug, a cork, to plug up, to stuff up, to calk. Tahiti :
puru, coconut husk. Rarotonga: puru, coconut fiber used
as a plug for calking, anything used to calk with. Mangareva :
puru, coconut husk, the almond of the pandanus. Rapanui :
puru, a covering, to cover. Marquesas: puu, a band of tow
made of coconut fiber.

Viti : mbulu, an external application, a thing that covers or buries,
coconut husk; mbuluta, to cover with earth, to bury, to repair
an injury, to apply an external remedy; bulubulu, a grave,
that which covers or buries, a peace offering. Rotuma : pul,
gum ; pulpul, gummy.

New Britain: pulpul, a cloth wrapped round the body. Mota:
pul, pulu, gum, to stick ; pulut, to stick together ; gapulut, glue,
paint. Pala: bulitt, to smear. Merlav: mbulut, to make to
stick ; gambulut, glue. Norbarbar : -vambulmbul, to make to stick.
Malekula : burji, to plaster; buruj, sticky. Malo : bulla, to stick.

Java: pulut, gum, birdlime. Malay: pulur, the pith of plants,
farina.

Hebrew: kapar, to cover, to cover over; koper, pitch. Arabic:
"afara, to cover over, to pardon (sin).

The Efate words are all in satisfactory accord with one or other of
several stems which in Polynesia interlace within the abraded form pulu.
These we shall examine with the aid of the Proto-Samoan stem where that
is discoverable.

DATA AND NOTES. 337

pulut, gum, pitch, sticky, to stick. Samoa, Tonga, Futuna, Uvea,

Viti (?), Rotuma, Efate, Mota, Merlav, Norbarbar, Malekula,

Malo, Hawaii.
pulut (this ascription resting solely on Samoa pulupuluta'i), a large

article of clothing. Samoa, Tonga, Futuna, Viti (?), New

Britain.
puluv, to interpose, to mediate. Samoa.
pulu, coconut husk. Samoa, Tonga, Futuna, Niue, Viti, Maori,

Tahiti, Rarotonga, Mangareva, Marquesas, Hawaii.

The pulut stem is thus seen to belong to the Proto-Samoan migration.
If the identification in Hawaii is good we have this stem to add to our
fragments of evidence showing a Proto-Samoan migration from Samoa to
Hawaii.

The pulu stem is common to both migrations.

In Melanesia the pulut stem preserves its t except in Malo.

In Indonesia the Java pulut is exact in form and sense. Malay pulur
may be connected with the pulu stem, but the sense is almost impossible
to reconcile.

The Semitic kpr can not be said to stand in strong likeness to Poly-
nesian pit.

290.

fafine, bite, fite, matu, woman, female.

Samoa: fafine, ma fine, a woman, wife, female; afafine, daughter;
tama fa fine, id.; tua fa fine, sister. Tonga : feflne, fafine, woman ;
ofefine, daughter; tuofefine, sister. Fakaafo, Futuna, Uvea,
Sikayana, Vate, Moiki: fafine, woman, female. Liueniua:
fafini, id. Niue : fifine, id. Nuguria : tahine, wife. Maori,
Hawaii: wahine, woman, female. Tahiti, Manahiki: vahine, id.
Marquesas: vahine, vehine, id. Rarotonga, Tubuai, Bukabuka:
vaine, id. Tongarewa: oahine, id. Nukuoro: ahine, feflne, id.
Mangareva: veine, ahine, aine, id. Aniwa: tafine, fine, id.
Fotuna: fine, woman, female. Paumotu: vahine, a wife;

mohine, a woman. Uvea : finematua, mistress. Tahiti: mahine,
daughter. Tonga: fine, women. Mangareva: mohine,

daughter. Maori: hine, girl. Rapanui: nuehine, old woman;
tamaahine, daughter.

Viti: yalewa, alewa, woman. Rotuma: hoina, id.

Solomon Islands : fafini, a sister. Fagani : fefene, woman. New
Ireland (Duffield) : fifine, feefeen, id. Makura, Arag, Omba, Uni,
Kabadi, Sinaugoro, Hula, Keapara: vavine, id. Mota: vavine,
tavine, id. Nggela, Bugotu: vaivine, id. Belaga: vaivine,
vinekama, id . Duke of York, New Britain : wawina, id . Tavara,
Wagawaga, Awalama: wawine, id. Taupota, Wedau, Galavi,
Boniki: wavine, id. Dobu: waine, id. Suau: waihin, id.
Mukawa: wasike, id. Baravon: wawin, id. Roro, Uni,
Galoma: babine, id. Oiun: babin, id. Mekeo: papie, id.

Rubi : gaiine, id. Nada : ilna, id. Murua : vine, id. Kiriwina :
vivila, id. Port Moresby, Motu: hahine, id. Redscar Bay:

338 THE POLYNESIAN WANDERINGS.

ahine, id. Ambrym: vihin, id. Suau, Sariba, Tube tube
sine, id. Dobu: sinesine, id. Mugula: sina, id. Dobu
ine, id. Eromanga: sivin, id. Moanus: pein, id. Uea
in, a girl. Nifilole: singenda, woman. Pala: hahin, id
Volow: linggmbweven, id. Motlav: Ikmbwovin, id. Lo

lakwavina, id. Mota: wte, feminine. Malo: othwm', woman,
female. Malekula : ne-ven, old woman, young woman ; febin,
female ; nev-seven, woman, female ; febinin, man's sister. Tanna :
nau vene(n), man's sister. Baki: buvino, female; kovivine,
man's sister. Bierian: fafine, woman, female; fefine, man's
sister; buvino, female.
Sulu: fina, a woman. Massaratty: fineh, id. Mayapo: gefineh, id.
Ahtiago: vina, id. Teor: mewina, id. Bouton: bawine, id.
Java: winih, the female of animals. Sanguir: mahoweni,

woman. Waigiou Alf uros : bin, id. Malay: bini, id. Gah:
binei, id. Cajeli: umbinei, id. Salibabo: babineh, id. Mysot,
Waigiou: />&», id. Gani: mapin, id. Saparua: pipina, id.
Liang, Morella, Lariko, Awaiya, Caimarian : mahina, id. Teluti :
ihina, id. Madura: bahini, id. Salayer: baini, id. Batu-
merah: mainai, id. Silong: benaing, id. Macassar: banie,
daughter. Ilocan: babai, woman. Malagasy: tow, female;
vehivavy, woman.

Of the Efate forms /o/?we is in full accord with the Polynesian. Our
author proposes fite as a mutant of fine; the material shows but a single
instance of n-t mutation, namn (328) mosquito Alo Teqel torn; fite and
bite, if coordinated with fine, stand by themselves in a class apart in the
handling of this stem.

The primal stem is fine, and this it is which carries the sex sense. We find
it in independent existence in Tonga, Uvea, Maori, Fotuna, Aniwa, Uea,
Mota, Tanna, Murua, and in composition away from the type forms in
Belaga vinekama and in Moanus.

The most common form in which this stem appears is fafine. Neglecting
the range of variation of the /, fafine is found throughout Polynesia and
Melanesia, less frequently in Indonesia. Upon this as a secondary stem
several compounds are erected.

Mafine is a secondary stem which is found in Samoa, the Paumotu,
Tahiti, and Mangareva.

T a fine is found in Aniwa and Mota.

Sifine is found in Malekula nev-seven, Eromanga sivin, and metathetic
in Ambrym vihin.

In Volow, Motlav and Lo we find a composite of fine with another stem
which it is impossible to identify. Nifilole singenda seems scarcely possible
of correlation, and Viti alevoa is also anomalous.

The Indonesian forms are readily reducible to the foregoing forms, fine,
the most common, and mafine more frequent in reference to fafine than in
the Pacific.

Macdonald's explanation is that fafine is a composite wherein the first
member is human being and the second characterizes it as feminine. We
may not yet isolate the particular sense of fa, ma, ta, which we find in our

DATA AND NOTES. 339

three island areas, but of the three only one and that the least frequent
one, ta, is susceptible of association with any word meaning man, tangata
of the human being, while in tane we have the sex antithesis of fine. Dr.
Macdonald's effort is labored but unconvincing.

291.
fai, vai, nlfai, noai, n'uai, n'wai, nai, n'ai, ai, water.

Samoa, Fakaafo, Tonga, Futuna, Niue, Uvea, Nukuoro, Tahiti,
Rarotonga,Tongarewa, Marquesas, Mangareva,Fotuna, Nuguria,
Vdit€: vai, water. Rapanui: vai, juice, liquid, water. Aniwa:
vai, tavai, water. Maori, Hawaii: ivai, id. Sikayana: wai,
vuuai, id.

Viti: wai, water. Rotuma: vai, voi, id.

Motu : vai, river. Omba, Ulawa, San Cristoval, Mwala, Saa, Bulu-
laha, Arag: wai, water. Sinaugoro, Hula, Keapara, Galoma:
wa^, river. Moanus : uai, water. Alite: kuai, id. Aneityum:
in-wai, id. Nifilole: woi, id. Tubetube: waiila, id. Sariba,
Massim: waira, id. Taupota, Wedau: waira, water, river.
Kiriwina: waia, river. Misima: weil, river; weweil, water.

Panaieti: wel, river; wewel, water. Tubetube: wawei, water.
Mugula, Suau, Awalama, Kwagila: goila, id. Tavara: goila,
water, river. Sesake, Southeast Epi : noai, water. Bierian :
nuai, id. Epi, Baki: ue, id. Ambrym, Uni: we, id. Nengone,
Duaru: wi, id. Yengin, Balade: ue, id. Murare:gm, id.
Nikete: que, id. Malekula: n-ue, id. Tanna: nui, id.

Eromanga: nu, id. Motlav, Volow : mbe, id. Maewo, Merlav:
mbei, id. Roro: bei, id. Bugotu: mbea, id. Nggela: beti, id.
Roro, Mekeo, Uni, Pokau: vei, id. Doura: vei, water, river.
Kabadi : veina, water. Vanua Lava, Norbarbar, Lo : pe, water.
Mota: pei, id. Marina: pei, tei, id. Baravon: tava, id.
Nada, Murua: rai, river.

Buru, Ceram, Salibabo, Cajeli, Mayapo, Massaratty, Amblaw, Sapa-
rua, Ahtiago, Solor, Sulu : wai, water. Waigiou Alfuros : ue,
we, id. Allor: we, id. Togean Islands: ue, id. Rotti:
ue (oee), id. Baju : voi, id. Vaiqueno East: hoi, id. Waigiou:
wayer, id. Gani: waiyr, id. Mysot: wayr, id. Liang:
wehr, id. Teor: weha, id. Batumerah, Morella: weyl, id.
Awaiya, Caimarian: w'deli, id. Teluti: welo, id. Dorey: waar,
id . Kaioa Island : woya, id . Malay : ayer, id . Menado : aer, id .
Champa: aya, id. Goram: arr, id. Silong: awaen, id.

Hebrew: me", water. Ethiopic: mai, id.

In general I am most reluctant to admit the possibility of diphthongs
in Polynesia, at least in Nuclear Polynesia. In the case of this word, vai
has always been received by my ear as two vowel sounds and in perfect
distinction from vae the foot. Yet without receding from my position in
regard of diphthongs we shall find it a great simplification of our analysis
of the vai mutants to adopt a working hypothesis that, even if not properly
a diphthong in Polynesian speech, the ai of vai has been accepted by its
Melanesian borrowers as of diphthongal value. The need for this assump-

340 THE POLYNESIAN WANDERINGS.

tion is that vai undergoes a series of vowel changes which would not be
possible unless its ai were regarded as a vocalic unit and susceptible of
variation as such.

The Efate forms are reducible to four types, vai, jai, ivai, ai; we shall
discuss our collected material in relation to these types, this discussion
referring only to the initial consonant.

vai. This is the most common type in Polynesia, is absent from Mela-
nesia, and is found but once in Indonesia, Baju voi.

fai. Efate alone shows this type. In general it is to be said that the
spirant rarely alters from sonant to surd, our only examples being lava (307)
very Fagani raja, and vivini (242) to crow Ambrym nofin to whistle. The
change from surd to sonant is of great frequency.

wai. This is found in Polynesia only in Maori, Hawaii, Viti, Sikayana.
It is the most common type in Melanesia and Indonesia. In the former
it is found in Omba, Ulawa, San Cristoval, Saa, Arag, Bululaha, Moanus,
Aneityum, Nifilole, Baki, Ambrym, Nengone, Duaru, Yengin, Balade.

A secondary Efate form is noai, n'uai, n'wai, the n probably functioning
as article. The same prefix, and probably of the same value, we find with
wai in Sesake, Bierian, Southeast Epi (probably the Bierian), Malekula,
Tanna, and Eromanga. These n-forms are found nowhere outside of
Melanesia.

ai. This type is exclusively Efate. It is normal to find the labials when
in mutation so extensively — we should interpolate the aspiration form of
this series in Vaiqueno, hoi — progressing to extinction ; yet this Efate type
is our only instance in this stem. Over against this we set for comparison
Eromanga, which has lost its radical vowel and exists only by the complete
alteration through the w semivowel to vocalic u; the chain is a complete
one in the links Bierian, Malekula, Eromanga.

We have already examined a secondary wai type in u. We find yet
another secondary wai type in k, a prefix whose purpose does not appear.
It occurs in Alite and in New Caledonia in the languages of Murare and
Nikete. This, too, does not pass outside of Melanesia.

Before entering upon the discussion of forms consonantly less obvious
it will be proper here to engage upon the consideration of the vowel forms.

ai. Polynesia altogether, Viti, Rotuma, Efate, much of Melanesia, most

of Indonesia.
oi. Rotuma, Nifilole; Baju, Vaiqueno East, Kaioa Island.
e. Baki, Ambrym, Yengin, Balade, Nikete, Malekula; Waigiou Alfuros,

Allor, Togean Islands, Rotti.
i. Nengone, Duaru, Murare, Tanna.

Having thus acquainted ourselves with the vowel variability we may
recur to the consideration of a second group of Melanesian homogenetics.
In these the vowel has changed to e, or to ei which is the same in effect.
The consonantal change is from spirant to mute, to b, mb, p. The table
of variations will afford abundant store of examples of this change. We
may, then, having confirmed both consonant and vowel mutation, accept
Mota, Marina, pei; Maewo, Merlav, mbei; Vanua Lava, Norbarbar, Lo,
pe; Motlav, Volow, mbe; Bugotu, mbea. Marina has not only pei but a

DATA AND NOTES. 341

synonym tei as well ; we are without evidence to support a mutation from
labial to lingual among the mutes and will not venture to suggest that tei
is homogenetic with pei, but it certainly means the same thing. This
may explain the Nggela beti, a determinant compound of the two stems
which, among others, have one like meaning (14 Journal of the Polynesian
Society, 41). We are left with Baravon tava; it is permissible to see in
va an abraded derivative of vai; the ta may be associable with Marina
tei, with Nggela beti and, even more remote, with Aniwa tavai, of course
with the possibility that in the last form the ta may prove to be the weak
demonstrative functioning as article.

We have already examined such of the Indonesian variants of the vai
stem as have been found in concord with the general mutation. There yet
remain for us distinctively Indonesian types of mutation. These involve
assumption of a final liquid. Based on the wai type we may follow this
from Waigiou wayer through Gani, Mysot, Batumerah, Morella, Awaiya,
Caimarian, Teluti, Dorey, Liang, Teor. A shorter series based on the ai
type begins in Malay ayer and runs through Menado, Champa, and Goram,
with Silong awaen in some way connected.

The initial consonant of the vai stem has run a long course, even to extinc-
tion, in our three island areas, but nowhere has it approached an m which
might give warrant for accepting the proposed Semitic identifications.

292.
fanua, inhabited country, land.

Samoa: fanua, the land, the earth, the ground. Aniwa: fanua,
the whole earth, the land, the soil. Fotuna: fanua, land,
country. Futuna : fenua, people, race, nation, land, country.
Uvea, Tahiti, Sikayana, Moiki, Fakaafo, Vate: fenua, land,
country. Nuguria: henua, id. Marquesas: fenua, henua, id.
Rapanui: heenua, id.; puheenua, soil. Paumotu, Manahiki:
henua, land, country. Maori: whenua, the whole earth, a
country, land, the ground, the soil. Bukabuka: whenua,
land. Rarotonga, Bukabuka: enua, id. Mangareva: enua,
land, used to denote shallow places in the sea. Tonga : fonua,
the whole earth, a country, the land, the soil. Niue: fonua,
land, soil, mold. Hawaii: honua, id.
Viti: vanua, a land, a region. Rotuma: hanua, land.
Arag, Vanua Lava, Marina, Maewo: vanua, land. Sesake: vanua,
a place, a village. Malo: vanua, land, house. Mota: vanua,
a place, island, land, village. Kabadi: vanua, village. Pokau:
vanua, land, village. Uni: vanua, land. Omba: vanue,
land. Lakon: vanu, id. Baki: venuo, land, country. Duke
of York: wanna, id. Galoma: banua, village. Mekeo:

pangua, land, village. Panaieti: pangua, land. Ugi: vanua,
hanua, id. Laur: hanua, a place. Lambell: 'hanua, id.;
hanua, a town. Motu: hanua, a village. Tubetube: ianua,
land, village. Suau: eanua, village. Lamassa: fdnu, a
place; fanu, a town. Bierian: fanua, land, country. Roro:
anua, land. Rubi: anu, village. Epi: venua, land. Santo:
venua, a house, a village. Malekula: fenu, land. Retan:

342 THE POLYNESIAN WANDERINGS.

vene, island. Saa : hernia, henue, a village. Vanikoro : jenua,
land . Fagani : finua, a place. Nggela : mbona, id . Norbarbar :
vonio, island. Pak, Sasar, Vuras: vono, land. Leon: vono,
country. Motlav: na-vno, land. Baki: vonua, land. Sinau-
goro, Hula, Keapara: vanuga, village.

Malay: banua, a land, a country. Bicol: banua, a town. Visayas:
banua, a village. Bugi: wanna, land. Malay: benua, a
region. Togean Islands: benua, a house.

Hebrew: banah, to build, as a house; banu', built; binyaha, building.

Dr. Macdonald makes this note:

The Santo word has best preserved the primary meaning "house" or "building" ; then
a country, district or land is called bdntia, or fanua, because, like a house or village
(or building), it is the dwelling-place of men, or place of buildings.

And all this because Hebrew banah is to build! As to the primary
meaning, circumspice.

In Polynesia the structural idea finds no place whatever; fanua means
the land, from the mold at one's feet (Samoa, Aniwa, Maori, Tonga, Niue,
Hawaii) to the land in which one lives (Samoa, Aniwa, Fotuna, Futuna,
Uvea, Tahiti, Sikayana, Moiki, Fakaafo, Vate, Marquesas, Paumotu, Mana-
hiki, Maori, Bukabuka, Rarotonga, Tonga, Viti, Rotuma, Niue, Hawaii),
to the whole world of many lands (Aniwa, Maori, Mangareva, Tonga).
And what has the Hebrew builder to do with even the least of these?

In our Polynesian material the variations affect but the first syllable.
Its radical vowel is a and it varies to e in eastern Polynesia and in the
islands of the Western Verge ; to o in Tonga, Niue, and Hawaii, again a
memorandum of the direct migration from Nuclear Polynesia to the north-
ern archipelago. The consonantal variations are all in the lingual series
upward, from / to v to h to wh to extinction.

In our examination of the Melanesian material we shall observe the
same three vowels; a is very common; e is found in Baki, Epi, Santo,
Malekula, Retan, Saa, Vanikoro; o is found in Nggela and the Banks
Islands. Fagani gives us finua, the only occurrence of i in all our material.
The consonantal variation of the first syllable gives us / in Lamassa,
Bierian, Malekula, Vanikoro, Efate, Fagani. The common form is in*»,
found in Arag, Vanua Lava, Marina, Maewo, Sesake, Malo, Mota, Kabadi,
Omba, Lakon, Baki, Ugi, Epi, Santo, Retan, Norbarbar, Pak, Sasar, Vuras,
Leon, Motlav. The h form is in Ugi, Laur, Lambell, Motu, Saa. The w
form is found only in the Duke of York group. Nggela gives us a down-
ward stronger variant in mbona.

Polynesia admits its variability only in the first syllable, so in general
does Melanesia. But there are instances where other parts of the word
are affected. Final a becomes e in Omba, Saa, and vanishes in Lakon and
Lamassa. In Baki final a becomes o in venuo; Norbarbar vonio establishes
an alteration of u in the penult, which proceeds to extinction in vono of Pak,
Sasar, Vuras, Leon, Motlav, and indicates the same loss in Nggela mbona.

The Indonesian identifications modify only the first syllable, the vowels
are a, e, o, the consonants b, v, w.

In conclusion we are to note upon what slight ground rests the statement
that the primary signification is house. This meaning is found only in

DATA AND NOTES. 343

Malo, Santo, and the Togean Islands. Knowing this to be a loan word we
are better warranted in regarding it as a misunderstanding or an inten-
tional misapplication on the part of the borrowers, three among fifty-
odd. I do not regard the use of fanua for town or village to arise
from the rare house sense. The village sense is found in Sesake, Mota,
Kabadi, Laur, Lambell, Motu, Santo, Saa, Fagani, Bicol, and the Visayas.
Rather does it seem to me to record the fact that our connotation of land
is wider by far than is within the scope of these poor islanders. When
one lives at the edge of the jungle and has to wage a steady fight against
the voracity of the advancing timber the dread forest is not a land or a
country, it is the dripping abode of the gods ever vengeful. Land is only
the place of human habitation, hence a village or a town, the designation
varying as foreign observers may choose to record their impression on a
volumetric scale which is always loose.

293-
fata, uenr\ uere, uete, kofeta, a bench, shelf, stand, platform.

Samoa: fata, a raised house in which to store yams, a shelf, a hand-
barrow, a bier, a litter, an altar, to carry on a litter; fatamanu,
a scaffold. Tongan: fata, a loft, a bier, a handbarrow, to
carry on a bier; fataki, a platform. Futuna: fata, a barrow,
a loft; fatataki, two sticks or canes attached to each other at
each side of a house post to serve as a shelf. Niue : fata, a
cage, a handbarrow, a shelf, a stage, (sometimes) the upper
story of a house. Uvea : fata, a barrow, a bier. Fotuna : fata,
a stage. Tahiti: fata, an altar, a scaffold, a piece of wood
put up to hang baskets of food on ; afata, a chest, a box, a coop,
a raft, a scaffold; pafata, a cage, a box; ahata, a box; ihata, a
box, a cage, a scaffold. Paumotu: fata, a heap; afata, a box,
a chest. Maori: whata, a platform or raised storehouse for
food, an altar, to elevate, to support. Moriori : whata, a raft.
Marquesas: fata, hata, hatad, shelves. Rapanui: ha ta, a table.
Hawaii: haka, a ladder, an artificial henroost; alahaka, a ladder.
Mangaia: ata, a shelf; atamoa, a ladder; atarau, an altar.
Mangareva : avata, a coffer, a box.

Viti: vata, a loft, a shelf; tavata, a bier.

Aneityum: ne-fata, a press, a shelf; noforofata, a ladder, scaffolding.
Motu : vatavata, a ladder. Tanna : nafatafata, a stage. Bierian :
kovata, id. Malo: ivasa, id.

Malagasy: vata, vatra, a box, a trunk, a coffer.

Hebrew: 'omed, a platform, a place; 'cmdah, a lodging-place.

The Samoan fata is a pair of light timbers pointed at the ends and tied
across the center posts of the house, one in front, the other behind the line
of posts ; rolls of mats and bales of sennit may be laid across these timbers ;
baskets of reserved victuals may be hung on the ends. The litter and the
barrow are two light poles with small slats lashed across at intervals. The
Marquesan fata is a stout stem of a sapling with the stumps of several
branches, a hat tree in shape, though found among a barehead folk. These
illustrations are sufficient to show what is the common element in all these

344 THE POLYNESIAN WANDERINGS.

fata identifications, light cross-pieces spaced at intervals. With this for a
primal signification it is easy to see how a ladder, a raft, a henroost, an altar
come under the same stem for designation. Perhaps Samoan fata fata
the breast obtains the name by reason of the ribs ; it would be convincing
were it not that the plumpness of most Samoans leaves the ribs a matter
of anatomical inference.

In our three island areas the fata stem is unmistakable. Malo ivasa
probably belongs with fata, for tapu (207) forbid Malo sab exhibits the t-s
mutation in this speech and it is found in other Melanesian tongues. This
form ivasa finds its parallel in Tahiti ihata. Bierian kovata and Efate
kofeta are the same /ato-composite. Efate uete may be a derivative from
fata, for f-w has a wide range and a-e mutation is so frequent as to call for
no comment except that in all our homogenetics we have found both the
a's permanent. Uete may produce uere (on the t-r mutation see note 258).
Having once accepted uere we can accept uenr', strange as nr seems.

But the Semitic md can have no relation with a stem in ft.

294.

fatu, a stone.

Samoa : fatu, a stone, the stone of fruits, seed, heart, gizzard. Tonga :
fatu, gizzard; fatukala, a black pebble-stone; matafatu, hard,
not easily made to cry (stone-face). Tahiti : faturei, fatuumuti,
stones in special uses. Fakaafo, Niue, Futuna, Aniwa, Sika-
yana, Vate, Manahiki: fatu, a stone. Maori: whatu, a stone,
the stone of fruits, the pupil of the eye, a testicle. Nuguria,
Nukuoro: hatu, stone. Hawaii: haku, a hard lump of any-
thing, the eyeball, adze-stones; pohaku, rock. Mangaia: atu,
kernel of fruit ; koatu, a stone. Mangareva : atu, a round fruit
stone, any round form; atumata, pupil of the eye; atutaha, a
stone used in turtle fishing. Fotuna: tafatu, stone.

Viti: vahi, a stone, a rock. Rotuma: hof, -both, hathu, id.

Nggela, Vaturanga, Sesake, Bierian, Mota, Arag: vatu, a stone.
Bierian: votu, id. Eromanga: ni-vat, id. Mota: vat, id.
Nguna: fatu, id. Fagani: fa'u, id. Lamassa : fat, id . Volow:
veat, id. AloTeqel: ve'e, id. Nifilole: ve, id. lai: veto, id.
Aneityum : na-fetumanava, na-fotumanava, the heart (cf . Samoa :
fatumanava, id.). Duke of York: wat, watwat, id. Raluana:
wat, id. King: wat, id. Baravon: watwat, id. Solomon
Islands: patu, id. Moanus: pat, id. Aneityum: ne-pat, a
whinstone. Brierly Island: pak, stone. Kalil: 'hat, id.
Laur: 7m/, id. Pala, Lambell: hat, id. Aneityum: n-hat, id.
Saa: /jaw, id. Ulawa: hoihau, id. Malekula: var, ni-vit, id.
Kabadi: vakuna, id. Kiriwina: daku, gaku, id. Tagula: varu,
id. Mekeo : fan, id. Hula, Keapara : vau, id. Galoma : bau, id.
Tubetube : veku, id. Sariba : weku, id. Mugula, Suau : veu, id.

Malay, Kayan, Silong, Macassar, Togean Islands: batu, a stone.
Magindano: watu, id. Savu: wawatu, id. Ilocan, Sulu:
bato, id. Malagasy: vato, id. Satawal: fahou, id.

Hebrew: eben, stone. Ethiopic: ebdn, id.

DATA AND NOTES. 345

In the Polynesian identification of jatu the variants all follow well-com-
prehended laws until we reach Rotuma hof, both, haiku. In the preceding
Polynesian we have seen the f-h mutation; in the succeeding Melanesian
we shall see yet more of it. The mutation t-tk in hathu and hoth is found
in the passage from Proto-Samoan to Viti, and yet other instances are found
in Aneityum and Bugotu. The still further mutation t-f, found nowhere
else in our present material, is common in Rotuma, as in these instances,
talinga (350) ear Rotuma faliang, fetu star Rotuma hefu.

We shall examine our Melanesian material first in reference to the initial
consonant.

fatu. Nguna, with elision of t in Fagani fa'u, abraded fat in Lamassa.
Fetu and fotu are found as composition members, the latter with
Bierian votu and Rotuma hoth and hof being the only occurrence
of that stem vowel.
vatu. Nggela, Vaturanga, Sesake, Bierian, Mota, Arag. Bierian votu,
an alternative form, has just received comment. Abraded vat is
found in Eromanga and Mota. The value of the evidence for the
t-r mutation appears in note 258; if this be considered valid
Malekula var belongs in this group. With vowel change we find
in series Iai veto, Volow veat, Alo Teqel ve'e, Nifilole ve. With
change to yet another vowel we have abraded Malekula vit.
hatu. Rotuma hath u. With elision of t, Saa and Ulawa hau. Abraded

hat in Aneityum, Kalil, Laur, Lambell.
watu. Abraded wat, Duke of York, Raluana, King; duplicated watwat,
Duke of York, Baravon.
These have been ascending variations. Descent in the series is found in :
patu. Solomon Islands (?). Abraded pat, Moanus, Aneityum and
probably Brierly Island pah.
The Indonesian identifications, though few, are satisfactory.
It is impossible to find any resemblance, to say nothing of more intimate
relation, in the Semitic proposed in identification.

k P* ^ 295.

fill, fila, bfla, bile, file, lightning.

Samoa, Fakaafo, Futuna, Sikayana, Nuguria, Rarotonga: uila,
lightning. Tonga, Niue: uhila, id. Uvea: uhila, lightning;
hila, to lighten. Hawaii : uila, uwila, wila, lightning. Tahiti,
Mangaia, Nukuoro, Rapanui, Manahiki: uira, id. Maori:
uira, lightning, to gleam, to flash. Moriori: rauira, lightning.
Marquesas: uia, id.
Viti: liva, lightning.
Mota: vila, lightning. Omba: vile, id. Pala: hile, id. Laur:

hille, id. Malo: uila, id.
Kisa: uila, lightning. Pani: kuilat, id. Tagalog: kuirlat, id.

Java: chalirit, id. Tidore: kila, id.

Arabic : barak, bara', to gleam, to flash, to lighten ; bark', the lightning.

Macdonald derives his lightning words from bilafila (284) to shine. If

these words be kin to Polynesian uila this derivation can not stand, for the

Proto-Samoan stem of bilafila has been shown to be pulaf, and of uila we

have probably a stem uhila.

346 THE POLYNESIAN WANDERINGS.

It is advisedly that I use the adverb probably. The true aspirate is so
gentle a breathing that we have no very sufficient evidence for its existence
in Proto-Samoan. From a careful examination of all the available material
I have been led to the conclusion that Proto-Samoan possessed two aspi-
rations, one so weak that it has failed of influence upon Samoan in its
modern phase, but is retained in such other tongues of Nuclear Polynesia as
employ the aspirate; the other, probably of a stouter intonation, which
has passed into the sibilant in modern Samoan and remains an aspiration
in Nuclear Polynesia.

The uila of Samoa (uhila in Tonga, Niue, Uvea) involves the former or
weak aspiration. The verb hila in Uvea goes far to prove that the aspirate
is structural and no mere grace note of speech. The presence in Hawaii of
the form uwila is another finger-board to the direct migration from Samoa
to that group, and the verb wila is confirmatory of Uvea hila.

Viti liva is metathetic, closer to Melanesian consonantal forms than to
the Polynesian initial vowel.

In Melanesia Malo has the modern Samoan form. Laur hille, Pala hlle
are akin to the verb form in Uvea. Mota, Omba and Efate forms also
derive from the verb.

Indonesian forms are in accord with the later Tongafiti uila, except per-
haps Tidore kila in kinship with the verb hila, and in their own cycle of
development they have undergone accretion by prefixes and suffixes which
correspond to nothing in Polynesia.

Having segregated an effective stem hila, it needs but to put the Semitic
by the side of it to show that kinship is out of the question.

296.
fonu, turtle, tortoise.

Tonga: fonu, turtle, tortoise. Futuna, Niue, Uvea, Fotuna: fonu,
turtle. Tahiti, Hawaii, Mangareva, Rapanui, Tongarewa:
honu, id. Marquesas: honu, id. Samoa: volu, tortoise.
Nukuoro: holu, turtle.
Viti: vonu, turtle.

Nggela : vonu, turtle. Sesake: fonu, id. New Ireland (Duffield) :
kauk-foon, id. (kauk, fish). Kalil, Laur: 'hun, id. Lamassa,
Lambell: pun, id. King: puni, id.
Malay: panu, turtle. Malagasy: fani, id.
Arabic: 'dwinat, 'ayinat, turtle, tortoise.
Two matters engage our attention at the outset of the examination of this
material. Nukuoro and Samoa are the only languages which present the
n-l mutation, a mutation of wide extent in Polynesia and of considerable
frequency in the Melanesian area, as the table will exhibit. This is but one
of many evidences which go to show that Samoa in its modern phase and
Nukuoro form a binary system in the general sphere of Polynesian affinities.
The other point is that while the variation in Melanesia is generally
upward, we find the downward mutation in Lamassa, Lambell, and King,
and the extreme of upward mutation in Kalil and Laur; yet these five
languages are comprehended within 61 miles of New Ireland littoral.
Thus we see that, considered by itself, the direction of mutation lacks
diagnostic value.

DATA AND NOTES. 347

The Indonesian identifications are satisfactory.

In the Arabic offering there is far too much, and of what does exhibit
some resemblance, 'd-win-at, the alternative form sacrifices a half. Further,
the change is not in series, which we have come to regard as essential in our
speech family, but in w-y, a two-column leap from labial proximity to palatal
proximity. No matter were the superficial resemblance even greater, we
could not accept an identification which so far violates the instinct of
Polynesian phonetics.

297.

ngil i, kil i, kili, to dig.

Tonga: keli, to dig, to sink, a dyke, a trench, a ditch; jekeli, to
paddle quickly. Niue, Uvea: keli, to dig. Samoa: 'eli, to
dig, to pull hard in paddling. Hawaii: eli, to dig in the
ground. Maori: keri, to dig, to rush along violently. Manga -
reva, Nuguria, Rapanui, Paumotu: keri, to dig. Marquesas:
kei, id. Nukuoro: keni, id. Tahiti: eri, to undermine; fieri,
to dig a hole, as a rat or a crab ; ari, to scoop out the earth
with both hands.

Viti: keli, a ditch; kelia, to dig a hole.

Santo: keli, to dig. Kiriwina: kelikeli, id. Sariba: keri, id.
Malo: cele, id. Nggela: gelt, id. Motu : gei, id. Solomon
Islands: eli, id. Makura: nggili, id. New Britain: kir,
kire, id. Baki, Bierian: mkili, id. Mota:^z7, id. Malekula:
kiri, id. Keapara: gia, id. Misima: giar, id. Ambrym:
gali, id. Kwagila: karo, id. Taupota : garai, id . Wedau,
Galavi: g'arai, id. Awalama: haraia, id. Tavara: halaia, id.
Dobu, Kiviri, Oiun: sara, id.

Malay: #a/t, to dig. Malagasy: hadi, id.

Arabic: kara', karw', to dig.

In the Polynesian section Nukuoro keni (l-n mutation) is a common
variation.

In the broader aspect the variants in the three island areas are the usual
changes of k and /. The former vowel is somewhat critical ; in Polynesia
and Viti it is e throughout, in Indonesia (one sound identification, one
debatable) it is a; in Melanesia the characteristic form is in i. Variants
from the Melanesian kili stem are : in the Polynesian direction Nggela, Santo,
Malo and the Solomon Islands (?); in the Indonesian direction Ambrym,
Efat6, and several New Guinea languages.

It would be better if we had transition forms to assist in the Malagasy
identification, but its changes are in line with Polynesian phonetics and are
no bar to acceptance of the form.

The Semitic here is at least to be accepted as a resemblance, for in the klh
triliteron the final consonant might readily pass from a Polynesian lan-
guage, yet it must be said that it would more probably be retained as s.
There is no evidence to show that the Proto-Samoan stem was ever other
than keli, that is the root kel with the suffix of the verb-formative i. Even
when reduced to kel the Semitic resemblance is not abolished.

Item 303 should be included with this.

348 THE POLYNESIAN WANDERINGS.

298.
ngore, usu, ngusu, the nostrils, nose.

Samoa: isu, nose, snout, bill. Futuna, Fakaafo, Aniwa, Manahiki :
isu, the nose. Nuguria: kaisu, id. Fotuna: eisu, id. Moiki:
ishu, id. Tonga, Niue, Uvea, Maori, Tahiti, Hawaii, Marquesas,
Mangareva, Paumotu, Rapanui, Tongarewa, Nukuoro: ihu, id.
Rarotonga: putaiu, id. Vate: tus, id.
Viti: uthu, nose. Rotuma: isu, id.

Vaturanga, Suau, Sariba, Tubetube: isu, the nose. Mugula:
is'u, id. ~Roro : izu, id. Nggela, Bugotu: ihu, id. Sesake,
Epi: ngisu, id. Bierian: kinihu, id. Baki: sunu, id. Brumer
Island: ishuda, id. Maewo: Usui, id. Wango: barisu, id.
Roro, Kabadi: #«, id. Pokau: idu, id. Sinaugoro, Keapara:
i/w, id. Hula, Galoma, Rubi: iru, id. Raqa, Oiun: iu, id.
Awalama, Taupota, Mukawa: niu, id. Nggao: nehu, id.
Buka: wem, uies, osu, id. Ulawa, Bululaha: palusu, id.

Saa: pwalusu, id. Fagani: barusu, id. Ambrym: gw/m, id.
Motu: wtfw, id. Aneityum: in-gidjin, id.
Gani: usnut, the nose. Malay: idung, id. Java: irung, id.

Togean Islands: iri^w, id. Malagasy: uruna, id.
Hebrew: nhiraim, the nostrils. Syriac : nhiro', the nose. Arabic:
noh'rat, the aperture of the nose.
In the Efate the form ngore is not to be found in Polynesia nor is it con-
spicuous in our Melanesian material. It may be that it reappears in Alite
ngongora and Duke of York gigiro.

The other words, usu and ngusu, serve as transition forms, usu pointing
to isu the nose in Polynesia and ngusu to ngutu the mouth, which is very
near, nearer yet when we bear in mind that ngutu the mouth is snout as
well and that isu the nose is snout too. We shall examine our material
first for the discrimination of the isu homogenetics.

The Polynesian languages have isu firmly differentiated and this form
goes over into Rotuma. In the foregoing list Vate tus rests on Turner's col-
lation of two languages from Efate, Meli and Erakor, differing widely from
the unnamed language which Macdonald records, and not distinguished by
name in Turner's work.

In Melanesia isu is found unmistakably in Vaturanga, Nggela, Bugotu,
Brumer Island, and probably Buka (wesu, uies). In Indonesia it is possible
that the Togean Island preserves it ; the Malay and Javanese are remote
and beyond proof.

Next we shall correlate the usu forms, restricting these to such as have
u for the former vowel and are devoid of a consonantal modulant. This
is found in Efate, Viti, Buka (osu), and Motu udu. It is recalled by Gani
usnut, and perhaps by Malagasy.

Efate identifies for us usu and ngusu. The latter is not common, Am-
brym guhu is the only affiliated form. More common is the wg-modulant
upon the isu stem: Sesake, Epi, ngisu; Bierian, ki-nihu; Nggao, nehu;
Baki, sunu (metathetic).

We now encounter another group of forms, the Z-modulant upon the isu
and the usu stem. Simply it is Usui in Maewo. Beyond this point it does

DATA AND NOTES. 349

not appear independently, but only in composition with another element,
ba-pa-pwa: Wango, barisu; Fagani, barusu; Ulawa, Bululaha, palusu; Saa,
pwalusu. The value of this prefatory member is at present unknown. As
regards the lisu-lusu member of the composite we are by no means certain.
We may choose to regard it as a mutant form of ngusu, but it should be
noted that neither in Polynesia nor in Melanesia have we confirmatory
evidence of ng-l mutation.

We have already seen the interchangeability of isu and usu. The Mela-
nesian material shows also the interchangeability of ngisu and ngusu, Usu
and lusu. As to the former of these pairs Efate shows that usu and ngusu
are not only broadly interchangeable, but that they may exist as alternates
in the same language. Now ngusu of Efate is clearly Polynesian ngutu.
Accordingly we are now to pass the homogenetics of that stem under review,
not stopping to distinguish in sense between mouth, snout, lips, beak, bill,
all of which are found :

Samoa, Tonga, Futuna,Uvea, Niue, Fakaafo, Moiki, Maori, Rarotonga,
Mangareva,Rapanui,Paumotu: ngutu. Hawaii :nuku. Tahiti:
utu. Fotuna : rangutu. Marquesas : kiikutu, kiinutu. Aniwa :
tangutu. Nuguria: launutu, launuhi.

Viti: ngusu. Rotuma: nuchu.

Mota Maligo : ngusiu. MotaVeverau, Merlav: ngusui. Gog: ngus,
ngusun. New Georgia : ngusu. Baki : sunu. Guadalcanar :
ngisu. Pokau: nutu. Motu: udu.

Saparua: nuku.

In this latter collation it will be seen that the Polynesian is altogether of
the ngutu type, the Melanesian of the ngusu type, except that Motu with
the very same which reappears in Tahiti serves as the transition phase.

We may lay out this material in a degenerative series, ngutu-ngusu-usu-
isu. Yet it seems to me far more consonant with that we now know of the
spirit of Polynesian to present them in the other as monuments of progress
from a broadly diffuse primitive su root in the direction of higher speciali-
zation in use by modulant coefficients, u, i, ngu, ngi and li. In that view
of the case we need recognize no difficulty in the fact that Usu does not
readily derive from ngisu, lusu from ngusu; they do not have to, they stand
on their same plane of specialized evolution and they stand independent.

And with this primal root su the Semitic presents not the slightest
resemblance.

299.

ngum i, urn i, ngu i, ngw i, mw i, to seize, to grasp, to catch, to hold (with
or in the hand).
Tonga : kuku, to clench the fist, to hold fast in the hand ; kukukuku,
to hold and carry in the hand. Futuna: kukumi, kuku, to
clasp in the hand. Niue: kuku, to hold fast, to grasp. Uvea:
kumi, to grasp after; kukumi, to throttle; nima kuku, the fist.
Mangareva: kukumu, to close the fist. Samoa: 'u'u, to take
hold of, to grasp, to clutch. Fotuna: no-kumia, to grasp.
Viti: kukuva, to hold a thing fast; nggunggu, clinched; nggumi, to
clench; nggunggutha, to hold in the hand.

350 THE POLYNESIAN WANDERINGS.

Motu : guguba, to hold tightly, to squeeze with tight fingers.
Malay: gangam, to clutch, to clench, the clenched hand, the fist.

Java: gagcim, id.
Arabic: kamkama, to collect, to seize or catch with the hand, to

take.

The Proto-Samoan stem is kum. This is apparent in Efate nguvi and um.

It is purely a Nuclear Polynesian stem except for the extreme eastward
extension to Mangareva. That it is identified only in Motu, Efate, Viti, and
Nuclear Polynesia shows it to have been brought by the migration stream
through the south gate and along the Viti stream ; and its sole Malay identi-
fications lie along the channel which that stream must have followed before
leaving Indonesia.

Our Viti forms are of particular interest. They reduce to stems kum,
kuv, and kuh. Of these the first is the stem common to Nuclear Polynesia,
Efate, and Indonesia ; the second accords with Motu, far back in the Torres
Straits fairway; the third nowhere else appears. The sense is so close for
all three as to show their unity in essentials. This is exceptional, but it
falls readily into accord with the hypothesis of seed as well as root in this
primordial language family.

In brief, this is that out of a seed of speech roots become such by the
addition of consonantal modulants to differentiate specific particular mani-
festations of the primordial act or state. I have worked out the value of a
number of these consonants when prefixed as coefficients. Ex hypothesi
we should look for a similar range of differentiations with consonants
suffixed.

In this instance we have ku as a root thus developed. Its primal sense
is clearly something to do with the hand and its elemental activity. What
then is the elemental activity of the human hand ? Our ergographs might
not register it at a milligramme, but the curling baby fingers are in the
position of grasping, the instinct to clutch which Dr. Louis Robinson has
said embodies whole ages of comprehension of the first need of arboreal
life. In its primal sense ku is the expression of the primordial clutch of
this race of primitive speech inventors. Such variety as succeeding conso-
nants may affix to the primordial clutch may express the manner or the
degree of such grasping.

In our three Viti stems we find a certain order of degree ; just to hold in
clutching hand, kuh; to clench the hand, kum; to clench and hold fast, kuv.
In even degree h is one of the weakest of consonants, merely a breath ; m
is so little removed from the vowels that it is among the very first of the
consonant acquisitions; v is one of the latest acquisitions, a strong sound
for men who have long since learned to speak and to speak strong as
becomes men.

300.
ika, a fish.

Fakaafo, Tonga, Futuna, Niue, Uvea, Moiki, Nuguria, Sikayana,
Vate, Maori, Marquesas, Mangareva, Mangaia, Paumotu, To-
ngarewa, Rarotonga, Manahiki: ika, fish. Aniwa, Fotuna:
eika, id. Samoa: i'a, id. Nukuoro, Tahiti, Hawaii: ia, id.
Rapanui: ika, fish, animal.

DATA AND NOTES. 351

Viti: ika, fish. Rotuma: i'a, id.

Epi: yika, fish. Sesake, Ambrym, Santo, Southeast Epi: ika, id.
Bierian: ne-ika, id. Mota, Lo, Fagani, Belaga, Nggela: iga, id.
Arag, Omba: ige, id. Wango, Bululaha: i'a, id. Ugi, Ulawa,
Alite: ia, id. Roro: maia, id. Norbarbar: ie, id. Nengone:
wa ie, id. Saa, Tubetube: He, id. New Georgia: ihana, id.
Dobu, Tavara, Awalama, Taupota, Wedau, Boniki, Mukawa,
Kubiri: iana, id. Kabadi: veana, id. Duke of York: ian, id.
Kiviri : iun, id. Kiriwina : iena, id. Murua : iini, id . Matupit :
ten, id. Baravon: en, id. Buka: eina, aenna, aienne, aiena, id.
Merlav, Pak: ig, id. Motlav, Gog, Mosin, Alo Teqel, Volow:
eg, id. Malekula: na-ih, id. New Ireland (Duffield) : kauk, id.

North Borneo: jikan, fish. Malay, Massaratty, Teor, Ilocan,

Mayapo, Gah : ikan, id . Bouton : ikani, id . Amblaw : ikiani, id.
Menado: maranigan, id. Silong: ackan, id. Sulu: m?a, id.
Kar Nicobar: &a, id. Central Nicobar: ga, id. Lariko,
Wahai, Gani, Saparua, Ahtiago, Matabello: ian, id. Liang,
Morella: iyan, id. Tidore: nyan, id. Batumerah, Awaiya,
Caimarian: iani, id. Mysot: «'n, id. Java: iwa, id. Teluti:
yano, id.

Hebrew: da#, d&fo dagah, degath, fish.

In the data here assembled we have two working stems, which, from the
regions of their greatest frequency, are to be described as ika the Polynesian
stem, and ikan the Indonesian stem.

We find ika as the stem in all but five of the Polynesian languages ; and
the only change affects the consonant; in Samoa and Rotuma it has so
lately vanished that enunciation still marks the gap ; in Nukuoro, Tahiti,
and Hawaii it has been forgotten that a consonant ever intervened.

This stem is of the most common occurrence in Melanesia. In various
stages of dilapidation we may trace it from Nengone in the far south to
the central Solomons and with possibly one sporadic instance farther north
in the eastern gateway. Epi shows in yika the ika stem with a slight semi-
vowel reinforcement, possibly repeated in North Borneo jikan, though the
value of this j is not distinctly set forth. The normal ika is found in the
New Hebrides, and with a variant not possible in Polynesia, as iga and
ige in the New Hebrides and the central Solomons. The change of the
final vowel from a to e is so slight as not to call for detailed consideration.
Wango and Bululaha reproduce Samoa and Rotuma i'a. Ia is found in a
part of the Solomons, but not restrictively, for ie carries it along to Norbar-
bar and Nengone. We are last to consider a variant which would be im-
practicable in Polynesia, terminal abrasion to a closed stem, ig and eg in the
Banks Group, ih in Malekula, perhaps in the New Ireland composite ka-uk.

Modifications of the ikan stem are found in the northern part of Mela-
nesia, the southernmost occurrence as well as the least altered being New
Georgia ihana. The presence of the n serves to identify ikan for us, despite
the dropped k, in Duke of York, Matupit, Buka and Baravon. These are all
within the range of Post- Polynesian Malayan voyaging, recourse to which
explanation has been had in jolau (287).

352 THE POLYNESIAN WANDERINGS.

Regarding the suffixed n as characteristic of Indonesia and affixed upon
Polynesian loan material in accordance with an idiosyncrasy foreign to the
source of the loan, we may dispose of many of these forms by a list which
exhibits their progress in variation from the standard type: jikan, ikan,
ikani, ikiani, maran-igan, ackan, iani, yano, nyan, iyan, ian, ein. Sulu isda
is impossible of coordination. Nicobar ka and ga would need proof that ika
could sacrifice that former syllable which only in Silong ackan has admitted
of any alteration whatever. Java iwa is last to be noted ; it should not be
understood as involving a k-w mutation, but rather does it represent some-
thing very much like Samoa i'a, and the semivowel w is but the attempt
to hold the two vowels apart, as is accomplished in Samoa by the '. This
is a particularly interesting identification, for it is the only language in
Indonesia which has not added n to the Proto-Samoan stem.

Concerning the Semitic Dr. Macdonald writes : "It is possible that ika is the
same by the elision of the d." Equally is it possible that darksome is
irksome "by the elision of the d." A simple philology.

"I'a is the general name for fishes," Pratt notes in his Samoan dictionary,
"except the bonito and shellfish (mollusca and Crustacea)." We may for-
give the inaccuracy of the biology in our gratitude for the former note. The
bonito is not a fish, the bonito is a gentleman, and not for worlds would
Samoa offend against his state. The Samoan in his 'upu fa'aaloalo has his
own Basakrama, the language of courtesy to be used to them of high degree,
to chiefs and bonitos.

One does not say that he goes to the towns which are favorably situated
for the bonito fishery; he says rather that (funa'i) he goes into seclusion,
he withdraws himself. He finds that the fleet which is to chase the bonito
has an honorable name for this use, that the chief fisher has a name that
he never uses ashore. He will not in so many words say that he is going
to fish for bonito, he says that he is going out paddling in the courtesy
language (alo) ; he even avoids all chance of offending this gentleman of his
seas by saying, instead of the blunt vulgarity of the word fishing, rather that
he is headed in some other direction (fa'asanga'ese). He does not paddle
with the common word but with that (pale) which he uses in compliment
to his chief's canoe. He will not so much as speak the word which means
canoe; he calls it by another word (tafanga), which may mean the turning
away to one side. In this unmentioned canoe he may not carry water by
its common name, he must call it (malu) the cool stuff. He will not men-
tion his eyes in the canoe; he calls his visor (taulauifi) the shield for his
chestnut leaves. Even the word for large becomes something else (sumalie)
in this great game. The hook must be tied with ritual care; it is called
(pa) out of the common name for hook ; no bonito will take a hook which
has not been properly tied ; the fastening is veiled under the name (fanua)
for the land. There are many rules to observe; their disregard is called
(sopoliu) the stepping over the bilges, from the most unfortunate thing that
the fisher can do. He may hail the bonito by his name (atu), or he may
call him affectionately or coaxingly (pa'umasunu) old singed-skin. If he
has the fortune to hook his bonito he must raise the shout of triumph,
Tu! Tu! Tu e!, not his whole name but one of its syllables; he triumphs
as over a foe honorably slain in combat, but he avoids hurting the feelings

DATA AND NOTES. 353

of the other gentlemen of the sea. The first bonito caught in a new canoe
he calls (ola) life ; the first bonito caught in any season bears a special name
(ngatongia), of uncertain signification, and he presents it to his chief. His
catch he reckons by a special notation ; to his numerals he adds the word
(tino) body; he counts them as one-body, two-body, three-body. Parts of
the gentleman have specific names of their own; his fins (asa) and his
entrails (fe'afe'a) are called in terms nowhere else employed; the tidbit of
the belly part, which the fisher must give to his chief, is called (ma'alo) by
the honorific title of the chief's abdomen. And if the rites were not duly
observed, if the hook was not rightly tied, if the fisher was so incautious
as to mention his eyes, if one of a hundred faults was committed and the
fishing was in vain, then the fisher acknowledged his ill success abjectly by
saying that (malod) he was conquered.

Such is the language Samoans use to the gentleman of the seas, and he
is not i'a.

301.
kasu, kas, kau, tree, wood.

Samoa: la'au, a tree, a plant, wood, timber; 'au'auli, 'auali'i, names
of trees. Futuna : ladkau, a tree, plant, wood ; kauasi, sandal-
wood. Niue : lakau, a shrub, a tree, wood; akau, a tree, wood,
timber; kauhuhu, the name of a shrub. Fakaafo, Manahiki:
lakau, a tree. Nuguria: lakau, rakau, id. Hawaii: laau,
a tree, wood, timber. Maori: rakau, id; kauere, kauka, kauri,
names of trees. Rarotonga, Tongarewa, Fotuna, Mangareva :
rakau, a tree. Paumotu : rakau, a tree, a plant. Sikayana :
rakau, wood. Nukuoro: rakau, a tree; ururakau, medicine;
haiururakau, a doctor. Rapanui: rakau, medicine. Vate:
rakau, a tree, a club. Tahiti: raau, a tree, wood, timber.
Aniwa: foirakou, a tree; rakao, a club. Tonga, Uvea, Mar-

quesas: akau, a tree, wood, timber. Moiki: ngakau, tree.
Viti: kau, kai, a tree, wood. Rotuma: oi, a tree.
Nggao: gazu, a tree. Sesake, Hpi: kau, id. Marina: gau, id.
Nguna: nakau, id. Motu: du, a tree, firewood. Aneityum:
nelcau-un, a rafter; cat, a tree. Solomon Islands: au, a tree.
Omba, Arag, Nggela, Bugotu : gai, id. Murray Island : gair, id.
Maewo : geiga, id. Tangoan Santo : tagai, id. Mota : tangae, id.
Gog: regai, id. Lakon: rega, id. Lo: raga, id. Bierian:
lakai, tree; leke, wood. Malo: wu-cai, tree. Tanna: ni-gi, id.
Epi: lakai, yesi, id. Merlav: tankei, id. Volow, Motlav,
Norbarbar: tenge, id. Vuras: retenge, id. Mosin: rekenge, id.
Pak, Sasar, Alo Teqel: enge, id. Malekula: ni-ge, n-ai, id.
Ulawa, Bululaha, Alite: ai, id. Vaturanga, New Georgia:
hat, id. Duke of York: diwai, id.
Malay, Baju: kayu, tree, wood. Teor: kai, id. Malagasy : hazu, id.
Hebrew: 'es, tree, wood; from Hebrew, 'asah, Arabic, 'asa', to be
hard, firm.

The stem kau does not appear independently in any language of Poly-
nesia proper. For tree and for timber we have the composite lakau in
various stages of transformation. But kau will also be found as an initial

354 THE POLYNESIAN WANDERINGS.

component of various tree names. It is in Yiti first that we find it in
free existence. In Melanesia this form is rare. It occurs as kau in Efate,
Sesake, Epi, Nguna, and perhaps may be preserved in Aneityum ; as gau in
Marina ; as au in Motu and somewhere in the Solomon Islands.

The triplicity of the Efate forms suggests a possible transition. Kasu
and kas are easy to be correlated, kasu and kau less easy. They might be
linked by the assumption of a parent form kahu, from which each might
derive. This would appear in modern Samoan as kau; but I have found it
the rule that even the mildest aspirate in Proto-Samoan becoming extinct
in modern Samoan is yet retained as aspiration in Nuclear Polynesia and
as th in Viti, none of which mutations is found on this record.

With the statement of this contrary argument, let us adopt this as a
working hypothesis. With kasu we identify Nggao gazu and Malagasy hazu.
With a parent kahu we may associate Malay and Baju kayu. This is as
far as this unsupported hypothesis will carry us, although kayu invites
enticingly as a bridge to kau.

We next employ the comparison of Polynesian lakau and Bierian lakai to
give us a transformation phase by which we may cross from kau to kai, and
in Viti we find kai occurring as a dialectic variant of kau. This secondary
stem kai (as kai, cai, gai, gae) is discoverable in the free state or in com-
position over a wide Melanesian range, and in Teor in Indonesia. In Vatu-
ranga and New Georgia hai we find the transition to Malekula n-ai and to
at of Ulawa, Bululaha, and Alite, probably including Duke of York diwai
as a composite.

For the next deterioration phase we have the benefitof a similar transition
member. As we were able to link lakau and lakai, so may we link lakai and
Lo raga, Lakon rega, and Bierian leke. Thus we obtain a tertiary stem ka,
ke. The ka (ga) form occurs only in Lakon and Lo. The ke (ge) form we
list from Norbarbar, Vuras, Mosin, Pak, Sasar, Alo Teqel, Malekula ; and
the gi from Tanna.

Let us not neglect to observe that in many of these Melanesian languages
kau is able to stand free, as in Viti and Efate ; omitting the combined article
n variously vocalized, we list free forms in Nggao, Sesake, Epi, Marina,
Nguna, Motu, Aneityum, Solomon Islands, Omba, Arag, Nggela, Bugotu,
Murray Island, Tanna, Malekula, Ulawa, Bululaha, Alite, Vaturanga, and
New Georgia.

In the Polynesian area we have noted that the prior composition element
is la, variously modified to ra, nga, a. We find this recurring in Melanesia.
First as la: Aneityum, ne-lcau-un. Then as e, corresponding to a in Tonga,
etc. ; Pak, Sasar, Alo Teqel, further establishing a variant stem nge out of
the tertiary ge. We now pass from la to ta, a mutation ill supported ; it
may be better to regard it as an independent member of the composite.
As ta we note its occurrence in Tangoan Santo, Mota, Merlav; as te, in
Volow, Motlav, Norbarbar. It may even be di in Duke of York diwai.
The form tenge illuminates Vuras retenge as a double composite, the com-
mon la stem prosthetic upon the te-nge compound ; and upon retenge hangs
Mosin rekenge. The prior element in Malo wu-cai is not elsewhere found.
Maewo geiga eludes analysis, except upon the scheme that it is a duplicate
of slightly variant kai derivatives.

DATA AND NOTES. 355

This is all very intricate, but the proof needs just this link by link elabo-
ration. The only point as to which we may retain proper doubt is as to
the kasu group of forms ; they are certainly an anomaly. Yet it is just
upon the kasu group, this irreconcilable kasu, that the whole of the pro-
posed Semitic origin rests.

302.
kat i, katikati, ngat i, to bite; kakat, a bite; fikit, to bite each other.

Futuna : katikati, to gnaw off the bark of sugar cane with the teeth ;
kakati, to corrode, to pierce, to eat in. Niue: kakati, to bite,
to chew. Maori: kakati, to eat into, to gnaw through, to
corrode; katikati, to nibble. Rarotonga: kati, kakati, to bite.
Paumotu: kakati, id. Nukuoro: kati, id. Rapanui: katikati,
to scratch. Samoa: 'ati, to eat in, to corrode; 'a'ati, to eat
in, to corrode, to gnaw off, to pierce (as the teeth of a dog) ;
'atimotu, to bite through. Tahiti: ati, to bite; aati, to bite,
to gnaw, to tear with the teeth. Hawaii: aki, aaki, akiaki,
to bite, to nibble.
Viti : katia, to bite.

Malo: cate, to bite. Tangoan Santo: kati, id. Makura: nggati, id.
Bierian: mkati, id. Santo: kotkot, id. Malekula: haji, id.
Mota: ngit, id.; gat, to chew. Tagula: gadu, to bite. Nada:
gad, id. Kiriwina: gadi, id. Tubetube: letai, id. Suau:
retai, id. Murua: gedi, id. Kubiri: gitaboni, id. Dobu:
go'i, id. Boniki: kutai, id. Mukawa: kutakibai, id. Taupota,
Wedau: utai, id.
Malay: gigit, to bite; gigitan, a bite. Malagasy: hehitra, to hold,
to grasp, to seize, clutch, bite; kaikitra, a bite; manaikitra,
to bite.
Syriac: nkat, to bite.

At the time of the earliest, the Proto-Samoan, migration the effective
stem was kati, this, however, being formed upon the root kat by the verb-
formative *, The bare root must have been carried down into Melanesia by
the Proto-Samoans, for it is found preserved in Santo and Mota. Fur-
thermore, in these two languages reproducing the archaic form we find the
only deviations from the radical a. One of these, represented by the Mota
ngit, is the vowel of the Malay identifications.

The Syriac stands as a resemblance, form and sense.

303-
kil i, kill, to dig; kali, a digging stick; kfli, a current (as in the sand; lit.
that which digs). Cf. ngil i (297) to which this note is

supplemental.
Ethiopic: karaya, to dig. Arabic: kam', id. Hebrew: karah, id.
Chaldee: kera', id.

304-
kirikiri, gravel, pebbles.

Tonga: kilikili, small stones placed on graves. Futuna, Niue:
kilikili, gravel. Samoa: 'Hi' Hi, gravel, pebbles, small stones;
'iliti, to be pained by walking over sharp stones ; ta'ili, stony,

356 THE POLYNESIAN WANDERINGS.

gravelly. Maori : kirikiri, gravel. Mangaia : kirikiri, flints,
small stones. Moriori: kiripohatu, gravel. Rapanui: kirikiri,
polished stone. Mangareva: kirikiri, flints, small stones.
Paumotu: kirikiri, stony, pebbly, gravelly; huakiri, gravel,
stony. Sikayana: kirikiri, shingle. Tahiti: iriiri, gravel.
Hawaii : Hi, iliili, small stones worn smooth by water. Nukuoro :
kerikeri, pebbles. Marquesas: kiikii, gravel.
Viti : kilitha, to turn up stones, to turn a thing up and look under it.
Solomon Islands: kilifela, flint; pokiri, a fishing sinker. Motu:

miri, gravel.
Malay: krikil, karikil, karikil, gravel, pebbles; batu-kelikir, gravel.
Arabic: girgir', a berry; garaV, gravel. Hebrew: garger, a berry.
The Proto-Samoan stem is kilit.

To the significations assigned in the Polynesian section should be added
that of fragments of broken coral such as are used for flooring houses where
shingle of volcanic rock is not available. I am very doubtful of the miner-
alogy which assigns flint to Mangaia, Mangareva and the Solomon Islands.
The Viti langgere gravel may perhaps be a composite of this stem. While
kilitha very clearly reproduces the kilit stem its signification is far from
satisfactory.

So far as we are in a position to judge Motu miri amounts to no more
than a partial resemblance, for in all this material we have no evidence
confirmatory of a k-m mutation, and Motu in four instances disposes of k
thus: twice by extinction (301, 306), once preserves it (249), once alters it
to g (299).

The Malay forms are coordinate with the modern abraded form of the
stem.

The Semitic has a general resemblance, but it has the stem grh, of which
the last member throws it out of accord with the Polynesian skeleton kit.

305-
kuli na, uili na, the skin, the bark.

Tonga, Futuna, Niue, Uvea: kili, the skin, bark, husk. Nukuoro:
kili, kiri, the skin. Fotuna, Maori, Mangaia, Mangareva: kiri,
skin. Paumotu: kiri, the bark. Rapanui: kiri, iri, the skin.
Marquesas : kii, skin, hide, leather. Samoa : 'iliola, the outer
skin; 'iliasina, light-colored, as the skin; fa'a'ililua, to injure
down to the second skin by scratching. Tahiti: iri, skin,
hide, bark. Hawaii : Hi, the skin, the bark.

Viti: kuli, skin. Rotuma: uli, id.

Ambrym: Hi, skin, bark. Sesake: weli, id. Moanus: kulit, id.
ho: gilit, id. ~Epi: kulu, id. Murray Island : egur, id. Nggao:
guli, id. Bugotu, Nggela: guiguili, id. New Georgia:

korekore, id. Savo: korakora, id. Vaturanga: kokora, id.
Wango: uriuri, id. Malo: uri, id. Baki: kulukuti, id.

Bierian : kul, id. Eromanga : no-kohitan, id. Santo : kurina, id.

Malay: kulit, skin, hide, leather, bark, rind, husk, shell. Baju:
kulit, skin, bark. Matu: kulit, id. Ahtiago:ikulit,id. Bouton:
okulit, id. Teor : holit, id. Madura : koli, id. Savu : kori, bark.

Arabic: gilid, skin, bark.

DATA AND NOTES. 357

The Polynesian word varies but slightly, and that normally, from kilt.
As it is in use solely with the value of a noun and hence needs no formative
affixes which might protect a radical terminal consonant as in verb-deriva-
tives, we have no means of proving a kilit stem. Indonesia exhibits a stem
which differs in the former vowel and preserves a final consonant, kulit.
Melanesia sometimes approaches the Polynesian, other times the Indonesian.

Of the Polynesian type we note Ambrym and Sesake, both of which have
abraded the initial k, and in neither is it usual to drop that letter, Lo pre-
serves the vowel in gilit and at the same time has the Indonesian /-final.
Codrington (Melanesian Languages 90, note) suggests as a possibility that
Mota wil to peel is homogenetic ; in his Mota dictionary, however, he defines
with great precision "to turn round horizontally, to peel, turning the fruit
over in peeling," this in sense and in form bringing the word into close
association with vili to turn round ; the k-w mutation in Mota is suggested
but once, kutu (306) a louse Mota wut.

The two forms in Efate are not wholly satisfactory as transition phases,
for they would involve the unsupported k-w mutation. Still we do find
kuli, a frequent Melanesian and altogether Indonesian form, alternative
with uili, which seems to offer some possibility of a connection with the
Polynesian kill.

The kuli form without consonant termination is found in Viti, Rotuma,
Epi (Bierian and possibly Baki), Murray Island, Nggao, Bugotu, Nggela,
New Georgia, Savo, Vaturanga, Wango, Malo, and Santo. There is a con-
siderable variation in these forms, but they are to be connected with this
stem. In Baki kulukuti we may have a composite of which kulu alone
pertains to this group. In Moanus we have a distinctly Indonesian form;
geographically it is within the feasible limit of Post- Polynesian Malayan
intercourse ; yet Lo, almost too far south for this explanation to be consid-
ered valid, has the /-final in gilit; and Eromanga no-kohit-an is suggestive,
although we have no other instance in that language of l-h mutation.

All the Indonesian identifications are of the kulit type, except that
Madura and Savu have either lost or have not acquired the /-final.

The Arabic shows resemblance to the Indonesian type as to consonants,
to the Polynesian as to vowels.

306.
kutu, the louse.

Tonga, Futuna, Uvea, Fotuna, Marquesas, Mangareva, Rarotonga,
Rapanui, Nukuoro: kutu, the louse. Niue: kutu, the louse,
the flea. Maori: kutu, the louse, human vermin in general.
Samoa: 'w/w, the louse; 'utufiti, the flea. Paumotu: gutu, the
louse. Tahiti: utu, id. Hawaii: uku, a small insect.

Viti: kutu, the louse.

Deni, Nada, Kiriwina, Mukawa, Raqa: kutu, the louse. Murua:
kuti, id. Arag, Nggela, Bugotu, Sinaugoro: gutu, id. Kiviri,
Oiun: guta, id. Vaturanga: ngotu, id. Marina, Lo, Malekula
Pangkumu: gut, id. Hula, Keapara: gu, id. Aneityum:
get, id. Tanna : kiget, id. Motlav, Volow: git, id. Makura:
na-kit, id. Nengone: ote, id. Buka: autu, id. Motu,
New Britain, Malo, Rubi, Dobu, Awalama, Taupota, Wedau,

358 THE POLYNESIAN WANDERINGS.

Kubiri: utu, id. Pokau, Doura: uku, id. Kow.uhu, id.
Bululaha: u'u, id. Galoma: uu, id. Alite, Mekeo, Galoma :
u, id. Omba, Maewo, Mota: wutu, id. Merlav, Mota, Gog,
Lakon, Vuras, Mosin, Norbarbar: wut, id. Pak, Sasar, Alo
Teqel: wu, id.

Malay, Java, Salayer, Menado, Bolanghitam, Sanguir, Gani, Lariko,
Gah, Baju: katu, the louse. Bouton: okutu, id. Sula:
kota, id. Mayapo, Massaratty : koto, id. Macassar: fettfo, id.
Ahtiago Alfuros: kutim, id. Teor: /m*, id. Morella, Mata-
bello: utu, id. Caimarian: utua, id. Wahai: utun, id.

Mysot: w/, wit, id.

Arabic: kurdw, the louse.

Pediculus seems from the uniformity of the occurrence of this stem to
have been an ancestral possession of the Pacific races in their voyaging.
That other source of uneasiness, the flea, would appear a later acquisition,
since it is rather commonly described in kutu composites as the jumping
louse. In Efate the flea was unknown, the people say, until it was intro-
duced by white visitors.

While the Melanesian forms show a wide range of variation, they may
readily be set forth in two series, according as they keep or lose fe-initial.

i. kutu, gutu, ngotu; gut, get, git, kit.
2. utu, ote, uku, uhu, u'u, u.

There remains a group of three forms characteristic of the northern
New Hebrides. This seems to involve, in its several forms wutu, wut, wu,
the k-w mutation which nowhere else appears nor is it exhibited in any
other Polynesian loan word in these languages except in the very doubtful
kusue (251) rat Lakon wohow. I incline to look upon this w as not at all
a fe-mutant, but rather as an accretion after the k has been dropped, a
hint that something has vanished. I have suggested the same in ika (300)
to account for the Javanese form.

In Indonesia we find the same double series, with and without the k.

The Arabic krd< contains two elements foreign to the Polynesian, of
which the protected r could not have been dropped arbitrarily.

307-
laba, to be much, many; laba, leba, lafulafu, to be or become or grow big;
leb, indeed, very; barab, baraf, barau, long, cf. barab, 121.

Samoa: lava, to be enough, to complete, the whole, indeed, very.
Nukuoro: lava, enough. Nuguria: hulava, id. Rarotonga:
rava, much. Rapanui: rava, to be able, capable; hakarava,
to make large. Maori: rawa, numerous, many, quite, very,
at all. Hawaii: lawa, enough, a sufficiency. Niue: leva
(in o leva), very, quite.

Viti : levu, great ; ndrawa, a thing that fills up.

Mota: lava, greatly; liwoa, large. Sesake: lavulavu, very large.
Merlav, Gog : lav, large. Lo : lilav, large. Omba: law ua, id.
Arag: gaivua, id. Fagani: rafa, id. Wango: raha, id.

Saa: laha, id. Deni: lebu, id. Mosin, Motlav: liwo, id.

DATA AND NOTES. 359

Lo: liliwo, id. Leon, Sasar, Pak, Volow: lowo, id. Vuras,
Lo, AloTeqel: luwo, id. Norbarbar: luwoa, id. Aneityum:
lu pas, id. Pala: nabd, very; /a&a, great.
Malagasy: /ova, long, tall, continuing long. Malay: luwas, wide,

extensive, large, ample.
Hebrew: rabab, to become much or many; rabah, to be or become
great, to grow up.
There seem to be involved two stems here, one characterized by a-forms,
the other by some weaker vowel. Yet the consonant structure exhibits a
degree of connection between the two.

The a-form is the rule in Polynesia, with the exception of Niue. It
occurs in Efate\ It appears in Mota and Viti together with the weaker
form, but especiallydiff erentiated therefrom. It is found in Sesake, Merlav,
Gog, Lo, Omba, Arag (?), Fagani, Wango, Saa, Bululaha, Pala.

Of the weaker type we find c-forms in Niue, Viti, Efate and Deni. An
t'-form occurs in Mosin, Motlav, Mota, Lo ; and lilav in the last-named may
serve as a transition phase between weak and strong types. An o-form is
found in Leon, Sasar, Pak and Volow. Forms in u come to light in Vuras,
Lo, Alo Teqel, Norbarbar and Aneityum.

In Indonesia we are able to identify one instance of the strong and one
of the weak form.

The Semitic has at least a resemblance.

308.
langi, the sky, heaven, above.

Samoa, Fakaafo, Tonga, Futuna, Niue, Uvea, Manahiki: langi, the
sky. Nukuoro: langi, id.; para-o-te-langi, rain; rang, a day;
huarangi, noon. Rapanui: rangi, sky, heaven, blue. Maori,
Rarotonga, Mangareva, Aniwa, Fotuna, Paumotu, Tongarewa:
rangi, the sky. Nuguria, Hawaii: lani, id. Tahiti: rai, id.
Marquesas: aki, ani, id. Vate: rang, id.
Viti: langi, the sky. Rotuma: langi, id.

Moanus: lang, the sky; rang, day (the lighted hours). Buka:
indengid, the sky. Tanna: neai, id. Malo: tukaelange, id.
Baki: lani, day. Bierian : ligian, day. Tanna : lenyan, day.
Malo : rane, light, day. Pokau : lani, daybreak, light. Kabadi :
rani, id. Roro: rani-ne-rere, daybreak. Roro: lani, wind,
time. Waima: la unwind. Wedau: VaneVanene, id. Galoma:
lagi, id. Dobu, Kiriwina: iagila, id. Wedau: Vagi'na, id.
Galoma: gagi, id. Murua: yagi, id. Hula, Keapara: agi, id.
Misima : Ian, daybreak. Panaieti: ran, id. Motu, Nada : lai,
wind. Nada: laina, id.
Malay, Kayan, Java, Tagalog, Magindano, Sulu: langit, sky.
Baliyon: langid, id. Bugi, Champa, Macassar: langi, id.

Togean Islands: janggie, id. Malagasy: lanitra, id.

Hebrew: rum, ram, to be high, to raise, to lift up. Ethiopic:
aryam, heaven.
In this very smooth series of identifications only a few merit comment.
The general signification is the sky, in some obscure fashion conceived
of as disposed in concentric layers; for in Samoan legend Tangaloa, self-

360 THE POLYNESIAN WANDERINGS.

existent and heedless creator of all things, rests in the eighth heaven or
the ninth. In Nukuoro, where the Samoan influence is altogether para-
mount, and in Moanus, the first identifiable halting-place of the migra-
tion swarm through the eastern gateway, we find the word used for the
bright half of the day; this usage occurs in Baki and Bierian of Epi, in
Malo, and probably in Tanna. In Moanus we have positive record (po 285),
and in Nukuoro we are warranted in assuming the common Polynesian
practice, that the calendar days are counted by the lapse of darkness. We
note with surprise that in these two languages the day sense is differentiated
from the sky sense by the use of r for the former and / for the latter. We
are not justified in drawing a conclusion that on this account is there any
particular association between these two languages. Moanus marks the
exit from Indonesia, Nukuoro can only be understood as a very modern
counter current from Samoa ; the same differentiation holds in Malo. We
have already had occasion {matangi 274) to discuss yet another langi sense,
that of the wind. In San Cristoval rangi is the rain.

Buka indengid seems resolvable into in-dengi-d, the former element
functioning as article. We have no other example from Buka of the l-nd
mutation, but in the table it will be seen that l-nd, or its practical equiv-
alent l-t,\$ not unknown. The d-final points to the possibility of acquisition
from Malay influence, and Buka is within the range of such cruises. Malo
tukae-lange is clearly a composite, and the former member is not explic-
able from our scanty memorials of that speech ; but the latter element is
confirmed in yet another composite, tae-lage cloud. Tanna neat may not
improbably be ne-(l)a(ng)i.

The Indonesian identifications are satisfactory, noting only that the
J-final as a formative element is generally idiosyncratic of this language

SrollP-

The only way in which the Semitic can be brought into association with

langi sky lies in the Efate preposition (?) elangi above. If that becomes

"above" by reason of a development of a secondary sense out of the literal

"in the sky," then it has nothing to with this Semitic. If, on the other

hand, its primal sense is "above" (possibly akin to Polynesian lunga), then

it has nothing to do with langi the sky. In either way the Semitic is

irrelevant even before we consider the amount of its resemblance.

309-
lango, a fly.

Samoa, Tonga, Futuna, Niue, Uvea: lango, the common house-fly.
Nuguria : lano, id. Fotuna, Maori, Rarotonga, Paumotu : rango,
id. Nukuoro: nango, id. Tahiti: rao, id. Hawaii: nalo, id.

Viti: lango, a fly. Rotuma: lang, id.

Arag, Omba, Bululaha, Alite, Marina, Maewo, Gog, Mota, Malo,
Ulawa.Wango, Saa, Vaturanga, Nggela: lango, a fly. Fagani:
rango, id. Bugotu: thango, id. Nggao: glango, id. Nengone:
nengo, id. Bierian: alago, id. Motu: lao, id. Guadalcanar:
ango, id. Baki: jago, id. Sasar, Vuras, Mosin, Alo Teqel,
Laur, Lambell, King, Lamassa, Duke of York, Merlav, Lakon,
Pak, Volow, Norbarbar : lang, id. Motlav, Lo : leng, id. New
Britain: laga, id. Makura: na-lag, id. Malekula: ne-rag, id.

DATA AND NOTES. 361

Aneityum: inlag, id. Murray Island: narger, id. Solomon
Islands: lau-au, id. Pokau: lalo-maka, id. Motu:/ao, id.
Doura : lao-maka, id. Mekeo : angu-ma, id. Roro : au-maha, id.
Kabadi: ao-kama, id. Panaieti: nagunagu, id. Nada:

nigunagu, id . Murua : niga u-wari, id . Kiriwina : nigonagula, id .
Dobu: nene-wara, id.
Kayan, Sanguir: lango, a fly. Pampangas: lango, id. Menado,
Bolanghitam: raingo, id. Baju: langow, id. North Borneo :
lalangou, id. Tagalog: langau, id. Dyak: lengeau, id.

Malay: langau, a large fly, a bluebottle.
Arabic: lakka'u, a fly.
With the exception of the /-w- mutation in Nukuoro and the plain meta-
thesis in Hawaii we encounter nothing calling for note until we have
advanced a considerable step into the Melanesian area. By far the greater
part of this western region falls somewhere along the regular series of
progression downward: lango, rango, lang, leng, loo, lag, rag.

Bugotu thango is unique, yet the sense and resemblance may be counted
on to carry the unusual l-th mutation. The same may be remarked of
the prosthetic g in Nggao glango. Nengone nengo and probably Murray
Island narger are carried in the company of Nukuoro nango, the /-n mutation
appearing in no less than five Polynesian languages, and in Melanesia
several times recurring in lima (313). The Epi forms are out of step, for
Bierian without explanation acquires an initial vowel alago, and Baki jago
presents an /-;' mutation which is supported only by lingi (154) to pour
Aneityum aijangjing, and lima (312) five Epi jimo Aneityum ni-jman.
Motu lao, as is a by no means uncommon accord between these two widely
separated languages, is Tahiti rao. The dropping of the / in Guadalcanar
ango appears nowhere else in the life-history of the word; the form is
recorded by Tregear without statement of its source; it is clearly from
a speech other than Vaturanga. The same author preserves the Solomon
Island lau-au, but with no record of what one of the many languages of
that archipelago employs it ; there is such a resemblance to Malay and Dyak
forms that I have little hesitation in assigning it to the Post- Polynesian
period.

The Indonesian forms exhibit the usual local peculiarities in dealing
with loan material, but the sense is plain.

The Arabic most resembles the Indonesian, differing therefrom most
markedly in the intrusion of a palatal between a and win the final syllable,
and this is the point where the Indonesian forms least resemble the Proto-
Samoan. We shall, therefore, have to regard the Semitic as yet farther
removed from suggestion of a common origin with the Polynesian and
Melanesian stem.

310.

lau, kalau, kolau, kalai, a spiderweb.

Samoa: apungaleveleve, apongaleveleve, a spider, a web. Tonga:

kaleveleve, a large spider. Futuna : kaleveleve, a spider, a web.

Niue: kaleveleve, a cobweb. Nukuoro: halaneveneve, a spider.
Uvea: kavelevele, a spider. Mangareva: pungaverevere, a spider.

Paumotu: pungaverevere, cloth. Mangaia: pungaverevere, a

362 THE POLYNESIAN WANDERINGS.

cobweb. Tahiti: puaverevere, id. Maori: pungawerewere,

puawerewere, puwerewere, a spider. Hawaii: punawelewele,

a spider, a web. Marquesas: pukavecvee, punaveevee, id.
Viti: lawa, a fishing net; viriialawalawa, a cobweb; butalawalawa ,

a spider.
Sesake: kalau, a spiderweb. Mota: talau, id.; marawa, a spider.

Motu: valavala, a cobweb. Aneityum: nilva, a cobweb;

nilvanilva, caulfat of a pig.
Malay: labalaba, lawalawa, a spider; sarang-lawalawa, a cobweb.

Visayas : toa, a cobweb.
Hebrew : 'araft, to weave.

This word has provided the theme of a curious paper by Mr. Tregear
(32 Transactions of the New Zealand Institute, 298).

There is an enormous metathesis here, and for that reason I have ordered
my Polynesian material in two groups, leve and vele. It will be seen at
once that the distinction is practically that of Proto-Samoan and Tongafiti,
specifically older and later forms. With the Proto-Samoan go the forms
in Viti, Efat6, Sesake, Mota, and perhaps Aneityum. Motu is the only
Melanesian form in accord with the Tongafiti, and it will be recalled that
several times, and no farther away than the last item, I have had occasion
to remark upon the resemblance between Motu and Tahiti.

Nor is this matter of the order 1-v or v-I the only point of divergence.
The stem is found free only in Viti, Efate, Motu, and perhaps Aneityum.
With it in composition we have two distinct elements. The simpler
is ka, and probably ta is the same ; Nukuoro hala differs, yet we assign it
provisionally with this element, for it certainly is not associable with the
other. We find ka in Tonga, Futuna, Niue, Uvea, Efat6, Sesake ; ta we find
in Viti and Mota, and in the latter we find a ma distinguishing marawa the
spider from talau the web. These are all Proto-Samoan, and they are all
found as compounds upon the Proto-Samoan stem leve, except Uvea which
shows a mixture, Proto-Samoan ka and Tongafiti vele.

The other compounding stem is, in series, apunga, punga, puna, puka,
pua, pu. These are compounded upon Tongafiti vele, except Samoa which
shows a mixture balancing contrarily the Uvea mixture, for Samoa has
Tongafiti punga and Proto-Samoan leve.

The lau of Efat£, Sesake, and Mota will readily be recognized as a degen-
eration form of a parallel Proto-Samoan stem lava, lav, lau. Aneityum
nilva seems to be ni-liva, of which n functions as article and according to
the common usage of that language extracts the nearest stem vowel; but
nilvanilva puzzles me ; I can hardly conceive it possible for a word to con-
duplicate article-stem ; this casts a doubt upon the otherwise plain inter-
pretation of nilva.

The Indonesian here recorded preserves the Proto-Samoan stem, whereas
more commonly it resembles Tongafiti forms.

As to the Semitic ; if Mr. Tregear and the Indonesian authorities can find
their parent stem in the Sanskrit, and if Dr. Macdonald reads his title to a
mansion in Semitic skies, the student of Polynesian may be justified in
regarding each with equal doubt.

DATA AND NOTES. 363

3"-
le, leo, Io, to see; lele, to look for; leo, Io, to watch, to look.

Samoa: leo, to watch. Tonga: leo, leohi, to watch, to guard.

Futuna: leo, leosi, to watch, to be vigilant. Niue: leoleo, to

guard, to watch over, to protect. Uvea: leoleohi, to guard.

Viti: rat, seeing, to look; raitha, to look at. Rotuma: io, to look.

Marina: leleo, to see. Malo: leo, to look; leleo, to open the eyes.

Bierian : mleo, to look, to open the eyes.
Malay: Hat, to see, to look. Malagasy: hiratra, sight, seeing.
Hebrew: ra'ah, to see. Arabic: ra'a', id. Ethiopic: re'ya, id.
The Proto-Samoan stem is Ieos.

Our Melanesian forms are in exact accord with the Nuclear Polynesian
and with the Efat£ leo. In Efate- we may easily see in le a still further
degradation of stem, but lo with an internal abrasion is anomalous.

The position of Viti rai, raitha, is baffling. It suggests a stem rais as a
parent. If we look upon Efate as suggesting an origin for rai, raitha from
rais raises an insuperable obstacle, for in the series leos-leo-le it will be seen
that le is twice removed from the least hint of s, and rais would need to
have les in Efate, which neither is nor could be found.

The Malay may stand as a good identification, the Malagasy not. Simi-
larly the Semitic has a palatal to prevent identification with leos.

312.
lima, five.

Samoa, Futuna, Niue, Nukuoro, Nuguria, Sikayana, Hawaii: lima,
five. Maori, Tahiti, Rarotonga, Rapanui, Moriori, Aniwa,

Fotuna, Mangareva: rima, id. Tonga, Uvea: nima, id.

Moiki: ngima, id. Marquesas: ima, id.

Viti: lima, five. Rotuma: Ham, id.

Pala: liman, five. Epi, Solomon Islands, Sesake, Arag, Makura,
Malo, Santo, Nggela, Bugotu, Nggao, New Georgia, Lambell,
Moanus, Kiriwina: lima, id. Nada: aqai-lima, id. Tagula:
go-lima, id. Brierly Island : paihe-lima, id. Bierian : ilima, id.
Santo Wulua : kilima, id. Paama, Omba, Malanta, Likkilikki :
lime, id. Yela: limi, id. Epi: Hmo, id. Baki: jimo, id.
Tangoan Santo, Marina: Una, id. Duke of York, Laur,

Lamassa, New Ireland (Port Praslin), Ambrym: Urn, id. Ero-
manga : (Sie) siklim, id. ; (Utaha) sukrim, id. ; (Ura) suorem, id.
King: telim, id. Malekula Pangkumu : erim, id. Malekula,
Wango, Fagani : rima, id. Kubiri, Panaieti : nima, id. Misima,
Panaieti: nima-ma-panuna, id. Mukawa: nima-masiana, id.
Taupota: nima-ruag' a-i-tutu, ten. Galavi: ko-ma-nima-sago,
six. Kiviri: nim, five. Murua: qei-nim, id. Tavara:
nim-i- tutu, id. Espiritu Santo : lum, id. Tanna: (Weasisi)
karilum, id.; (Kwamera) karirum, id.; (Naviliag) kadilum, id.;
(Numerat, N erokwag) kilkilep, id.; (Ra'na.)kerkcrep, id. Waima,
Roro, Mekeo, Uni, Pokau, Doura, Kabadi, Motu, Domara,
Mailu: ima, id. Boniki: ima-i-kove, id. Sinaugoro, Hula,
Keapara, Galoma, Keakalo, Rubi: imaima, id. Galavi:
ma-i-kove, id. Lo: tevelima, id. Mota: tavelimwa, id. Gog:

364

THE POLYNESIAN WANDERINGS.

tavalima, id. Aneityum: ni-jman, ni-kman, id. Maewo:
tevelim, id. Norbarbar: teveliem, id. Merlav: tavalimw, id.
Mosin: tevelimw, id. Vuras, Motlav, Volow: tevelem, id.

Lakon : tivilem, id. Sasar, Pak : 'evelem, id. Leon : 'evelimw, id.
Retan: tavalemw, id.

Malay, Java, Cajeli, Morella, Batumerah, Teor, Magindano, Champa,
Sulu, Sambawa, Visa}ras, Tagalog, Pampangas, Kayan, Mame,
Salayer, Sanguir, Mayapo, Massaratty, Amblaw, Awaiya,
Caimarian, Teluti, Ahtiago, Baju: lima, five. Lampong:

limah, id. Sam: limoh, id. Togean Islands: limo, id.
Kisa: liman, id. Bouton: limanu, id. Salibabo: delima, id.
Timor: lema, id. Basakrama: leww, id. Sirang, Gah, Mysot :
Zrw, id. Gani: leplim, id. Menado, Liang, Bolanghitam,
Lariko, Saparua, Matabello: nwa, id. Jobi, Dorey: rim, id.
Wahai: nima, id. Malagasy: ctoray, id. Dyak: ma, id.

Arabic: h'amsat, h'ams, rive. Mahri: khomo, id. Sokotra:

khemah, id.

The numeral five being in many cases the same word as that for the hand
on which the tale is reckoned we shall first note the cases in which the sense
differentiation is indicated by a form modification. By comparison with
3 1 3 it will be seen that such distinction is found as follows :

Five.

Hand.

Remarks.

Maori

Viti

Moriori

Marquesas . . .

Vv'ango

Ulawa

Fagani

Tanna

Baki

Duke of York

Laur

Lamassa

Omba

Cajeli

Morella

Batumerah . .

Teor

Liang

Maewo

Merlav

Mota

Lo

Magindano. . .

Bouton

Menado

Nggao

Bierian

King

rima

lima

rima

imi

rima

lima

rima

kari-lum

jimo

Urn

lim

Urn

lime

lima

lima

lima

lima

rima

tavalima

tavalimw

tavelimwa

tevelima

lima

limanu

rima

lima

ilima

telim

ringa
linga

ririma

iimi

rimarima

nima nima

rum a

n el' I imi

juma

lima

lima

lima

limegi

limamo

limaka

limawah

limin

rimak

lima

lima

limwai

limwe

a lima

olima

rilma

kamc

ma

lama

m-mutant

preduplication

eonduplication

and /-mutant
t'-mutant

(inverted order)
" anda-mutant
abbreviated five

final accretion (gi noun formative)

(a-mutant)
frontal accretion of five

frontal accretion of hand
interior intrusion

Which of these is the primal signification? Is the numeral called by the
hand name because of its five digits? Is the hand called by the numeral
because its digits tot up five? So far as I have seen, all my predecessors

DATA AND NOTES. 365

in these studies have assumed that there is first the hand and then by the
accuracy of digital mathematics the numeral has followed.

It has not been found necessary to call the numeral one after some object
which is a visible unit in nature: one is not the word for nose, for an
instance; nor is two the word for eyes or ears, which as pairs upon the
primitive mathematician are surely as visible, tangible, obvious as the five
fingers of one hand. Three is found to be independent of any such concrete
presentation ; four also. Why, then, must five be considered a secondary
sense of hand ?

As to our English five we might see a beautiful reasonableness in naming
it from the fingers of our own mathematical hands. We stick up our fingers
in reckoning ; the first task of our nursemaid mathematicians at school is to
teach the child that sums are no longer to be done on the fingers but on
slates with pencils. Thereafter follows mental arithmetic with a new series
of tortures all its own.

But in the islands of our study fingers go not up but down for the count.
The hand with its digits displayed coram publico is zero, cipher, naught.
It is the clenched fist which counts most, it reckons five ; a usage paralleled,
to be sure, in our idiom of that noble art of defending usually most ignoble
selves, "I put my 'bunch of fives' in his — " mug, was it? Or peeper? Or
possibly breadbasket, this being before the days when solar plexus had
given to the ring the dignity of astrological anatomy. The five of the
clenched fist I recall from many an island race.

Let me, however, confirm my testimony from an authority who believes
that five is the hand, Dr. Codrington (Melanesian Languages 222, notei):

The way of reckoning on the fingers differs in various islands. In Nengone the fingers
are turned up and brought together at five. In the Banks Islands the fingers are turned
down. This is often done with the spoken numerals, often without the use of words.
The practice of turning down the fingers, contrary to our practice, deserves notice, as
perhaps explaining why sometimes savages are reported to be unable to count above
four. The European holds up one finger, which he counts, the native counts those that
are down and says "four." Two fingers held up, the native counting those that are
down, calls three; and so on until the white man, holding up five fingers, gives the
native none turned down to count. The native is nonplussed, and the enquirer reports
that savages can not count above four.

It may well be that I shall seem heterodox in placing five as the parent
sense, yet if there be any meaning at all in the foregoing table I am resting
upon demonstrable facts and not upon mere fancy. Many of the languages
collated in this and the next item show no difference in form between the
two senses, the word is the same. Such instances are negative, they lack
evidential value, they prove nothing, they disprove nothing.

This table which I have compiled is far from complete ; in many languages
I have the five word and the hand word has not been recorded in the scanti-
ness of most of this vocabulary material. Here I have assembled twenty-
eight languages, all that I have been able to bring into comparison, in which
the five word and the hand word are homogenetic yet variant. I might
have extended the list by the inclusion of languages in which the five word
is a lima form and the hand word is heterogenetic. These I omit, for, while
they are not without their value in pointing to a primal ftve-lima, they
would beyond that complicate the problem by the addition of incommen-
surable factors.

366 THE POLYNESIAN WANDERINGS.

We shall then examine these eight-and-twenty languages. Four lan-
guages seem to point one way, two dozen point the other. The twenty-four
congruent languages have the five- lima in its clearest form, and hand-lima
has undergone some form of secondary mutation, the nature of which is
indicated in the remarks.

The four opposing languages are now to be examined for the ascertain-
ment of the value of their testimony. Tanna is an island remote from either
of our migration streams, the quality of its Polynesian content is absolutely
the lowest in my scale. The five-lima and the hand-/iwa are alike obscured
by alien admixture, so much so that they are barely recognizable. The
testimony of Tanna, therefore, should not for one moment count against
the agreement of twenty-four.

The Duke of York has not only Urn for five, but also lima. The posses-
sion of the latter is sufficient to remove this speech from this consideration ;
it belongs in the outer ring in which the five word and the hand word are
the same lima.

Laur and Lamassa (and King, despite another variation element, may be
associated therewith) are really the only evidence in opposition. What do
these witnesses amount to against a principle established in our triple area,
in Polynesia, in Melanesia, in Indonesia? They are three languages, dia-
lectic variety at best, spoken on the New Ireland coast in the very jaws of
the eastern gateway. The source from which I derive the vocabularies con-
tains also a social register, a city directory, of two of these abodes, or huts,
of culture : Lamassa has a population of 23 men, 28 women married to them,
and 40 single persons including children ; King is peopled by 38 women and
girls, 30 men and boys. Not many in the census are these who reverse the
system of the whole Pacific.

On the whole we may disregard such exceptions entirely. Therefore I
am willing to aver that in every case where five-/tmo and hand-lima differ
it is the latter that is secondary in form ; therefore five it is which is the
primordial sense; the hand is lima simply by virtue of its possession of
five enumerable fingers.

The following notes have to do with the varieties in form of the five-lima.

In Baki and in some other unnamed dialect of Epi the o of jimo, limo, is
explicable on my theory of the neutral vowel; this vowel change is found
in Basakrama, Saru, and the Togean Islands in Indonesia. The l-j muta-
tion occurs also in Aneityum ni-jman. The m-n mutation in Marina and
Tangoan Santo Una is found again in manu (317) bird Marina nanu, and
mata (324) eye Marina nata. Eromanga shows a /iw-composite with the
element sik-suk-suo. Motu ima finds its only congener in the extreme
east, not in Tahiti this time but in theadjacent archipelago of the Marquesas.
Aneityum, with n-article and attraction of the nearest vowel, yields two
forms, jiman and kiman, each with the n suffix. The former has already
been discussed ; the l-k mutation is found only once again in lima (313) hand
Nggao kame Vaturanga kima. We now find in the northern New Hebrides
a group of lima forms prefaced by tava in no less than four vowel variants
and a fifth in which the initial t has been abraded. Not a word of expla-
nation is offered in any of Codrington's disquisitions upon these numerals.
I venture the suggestion that the added element may find some illumination

DATA AND NOTES. 367

in Efate tefa (24) to arrange in a row, as one who should tally one-two-
three-four, and five-in-a-row, this being as plain to view when the fingers
are down as when they are up. The prefix in Bierian ilima and Malekula
Pangkumu erim is probably not a formative accretion ; it is very close to
such a truly Polynesian use as e lima for five, the e being the visible sign
that lima functions as verb. Thus, also, may Santo Wulua and King be
accounted for. In Tanna we have (Weasisi, Kwamera, Naviliag) three
composites of a lima form {lum, rum) with the initial element kari. In two
other Tanna forms (Numerat, Nerokwag, Ra'na) we see fccm-variants
(kit, her) capable of carrying the five sense without lum. Therefore I
regard the karilum forms as determinant composites in which each member
has the same meaning.

In Indonesia the variations are less extreme. We note composites in
Salibabo and Gani, but neither de- nor Up- is at all associable with the
composition members which have been found in Melanesia. Wahai nima
reproduces Tonga, Uvea, and Moiki forms, yet the only trace thereof in
Melanesia is found in lima (313) hand Saa ninime New Ireland nemdn
Nifilole nime Ulawa nimanima, for which we have no record of a five-Ztwa.
Dyak ma finds a transition phase in Motu ima and yet more positive
support in Bierian ma hand.

It is quite impossible to see in Arabic Warns a form homogenetic with
lima.

3*3-
lima, nalima na, the hand.

Samoa, Fakaafo, Nukuoro, Futuna, Niue, Sikayana, Hawaii, Mana-
hiki: lima, hand, arm. Liueniua: makalima, id. Tahiti,
Rarotonga, Nuguria, Aniwa, Fotuna, Rapanui, Mangareya,
Paumotu: rima, id. Moriori: ririma, id. Maori: ringa, id.
Tonga, Uvea: nima, id. Moiki: ngima, id. Marquesas:
iima, ima, id.
Viti: linga, hand, arm.

Epi, Maewo, Merlav, Gog, Nggela, Arag, Malo, Mota, Santo, Lakon,
Bugotu, Buka, Duke of York, New Georgia, New Britain,
Baravon, Pala, Laur, Lamassa : lima, the hand, arm. Alite :
limalima, id. New Ireland (Carteret Harbor) : limak, bralima, id.
Ruavatu: limenamanu, wing of a bird. Omba: limegi, hand,
arm. Buka: tenia, id. Mota: limwat, id. ho: limwe, id.
Kiriwina: iamila, id. King: lama, id. Marina: Una, id.
Wango: rimarima, id. Mugula, Tavara, Taupota, Galavi,
Kubiri, Kiviri: nima, id. Suau, Sariba, Tubetube, Panaieti,
Tagula, Nada, Dobu, Wedau, Mukawa: nima, hand, arm.
Ulawa: nimanim a, hand. Nifilole : nime, id. Saa: ninime, id.
New Ireland (Duffield) : nemdn, id. Epi : sima, id. Vaturanga :
kima, id. Hula, Keapara: gima, hand, arm. Sinaugoro:
gima, hand. Nggao: kame, id. Epi: yima, id. Roro,
Motu: ima, arm. Mekeo, Pokau, Galoma, Rubi: ima, hand,
arm. Uni, Doura, Kabadi, Sinaugoro, Galavi, Boniki, Kwa-
gila: ima, hand. Kabadi: imana, arm. Roro: imana, hand.
Fagani: ruma, id. Epi (Baki) : juma, id. Oiun: uma, id.

368 THE POLYNESIAN WANDERINGS.

Epi (Bierian) : ma, id. Aneityum: ni-jman, ni-kman, id.

Tanna: neVlimiin), id. Santo Wulua: lemantra, id.

Macassar, Champa, Sulu: lima, hand, arm. Kisa: liman, id.

Cajeli: limamo, id. Morella: limaka, id. Batumerah:

limawah, id. Magindano: alima, id. Bouton: olima, id.

Teor: limin, id. Bolanghitam: rima, id. Liang: rimak, id.

Menado: rilma, id.
Arabic : alh'ams, the fingers.
Most of the form varieties here presented have received consideration
in the preceding item. Note should be made of the fact that the meaning
includes the whole member from fingertip to shoulder, along the inner
aspect as far as the axilla. Two forms, sima and yima, assigned to Epi
without particularization of dialect, are entirely anomalous, yet their der-
ivation from lima is not to be doubted. Santo Wulua lema-ntra introduces
an unexplained element ; the same is true of the Tanna composite and of
the bralima of Carteret Harbor in New Ireland.

3H-
lolofa, lulum, lumu, to be wet, moistened ; Iuma, to sink, to dip.

Samoa: lolo, lofia, to flood, to overflow; lolo, to be wet (as the
clothes). Tonga : lolo, to rain in torrents; lomaki, flood, deluge.
Futuna : lofia, inundated, submerged, inundation ; lomaki, deluge,
inundation. Niue: lofia, overflowed. Uvea: lolo, to flow;
lovai, deluge; lomaki, id. Maori: roma, a stream; rumaki,
to duck in the water. Hawaii : lu, to dive or plunge in the
water; lama, lumai, to put to death by putting the head under
water. Mangareva: akarumakimaki, to dive often, to inun-
date. Paumotu : rumaki, to sink in the water.
Viti : luvu, to sink in the water, to drown ; luvutha, to flood, to over-
flow; luvuraka, to put under the water, to press a thing down
under the water; ndrondro, a current, chiefly of the sea.
Raluana : lowon, a stream ; lolonga, to flow. Duke of York : lomon,

a flood. Kiviri: loloro, a. river.
Malagasy: rubuka, plunged, dipped, soaked.

Hebrew: seba', to dip into; 'istaba', to be wet, moistened. Arabic,
Chaklee, id.
We find two stems, lof and lom, each showing a tendency to vowel shift
and appearing as luf, lum.

The common factor is lo or lu, the common signification of water and a
motion, one of water in motion, the other of motion into water. These
senses are not restrictive in the languages which have both stems, and we
are not warranted in suggesting more than that there may have been a
distinctive value to each of the terminal consonant modulants, but that
in the drift along the ages and across the seas the distinction has in some
places become obscured.

Stem lof. This is found in Efate, Samoa, Futuna, Uvea, Niue, Raluana,
with the o-radical. This is accordingly seen to be Nuclear Polynesian ;
the omission of Tonga from the habitat of this stem is due to the fact that
our dictionary authority affords no instance which distinctly proves a lof
form, yet Tonga records lolo, which in Samoa is seen to belong to lof.

DATA AND NOTES. 369

Stem lom. This occurs in Tonga, Futuna, Uvea, Niue, Maori, and Duke
of York with the o-radical ; and with the ^-radical in Efate, Maori, Hawaii,
Mangareva, and Paumotu. If it were not for the ^-radical in Efate we
should be wholly justified in the statement that Iof and lom are Proto-
Samoan, lum Tongafiti; and this is indeed probable, for Maori roma is
explicable as carried by the direct migration which our material shows to
have passed from Nuclear Polynesia to New Zealand. This would afford
us an explanation of malu water, the bonito fishing euphemism. It might
be a conditional form of the lum stem, adopted for this purpose as a slightly
alien word and therefore incomprehensible since the bonito understands
Samoan perfectly. This would not be the only instance in which the
honorific speech has drawn from a foreign source.

In the Viti ndrondro is clearly the Proto- Samoan lolo (Iof stem) and the
reinforcement of the r points out that the primal stem was rof with r
grasseye. Luvu would seem to be a vowel mutation on the luf stem
derived from a source in which the r had become /. Against this is to be
set the fact of luvutha and luvuraka. The former implies a stem luvuh,
luvus, or luvut; in the latter we may choose to regard, and with much
reason, the r of raka as non-radical and that this termination is applied
evenly to all open stems.

Without support in transition forms we are not at liberty to admit the
Malagasy, the extraneous k being the obstacle.

The Semitic is, of course, beyond the range of possibility; the sole point
of resemblance being the b, and this is not sufficient to carry the load of
the former consonants.

315-
mala, malala, the cleared place in each village where the ceremonial drums
are set up; a place or part (as of a garden); malmal, a small
place or part.
Samoa : malae, the town green. Nukuoro : malae, a cleared space,
an open place, a plantation. Tonga : malae, a green, a grass
plot. Futuna, Uvea: malae, the public place in front of the
houses. Hawaii: malae, smooth (as a plain) . Niue: male,
an open space, plaza of a village. Fotuna: marai, public
house. Maori: marae, an inclosed place in front of a house.
Tahiti: marae, the sacred place of worship. Mangaia: marae,
the sacred inclosure of sacrifice. Tongarewa : marae, a sacred
inclosure. Paumotu: marae, a temple. Mangareva: marae,
a temple. Nuguria: marai, an open meeting-place.
Viti: mora, a burying-place.

Nggela: male, malei, a place. Laur: malar, a town, a place.

Bierian: ka-mali, public house, village. Baki: ko-meli, id.
Malekula: he-mir, id.
Malay: balai, an audience hall, a reception room.
Arabic: 'ardf, 'ara', 'arat, an open place. Hebrew: 'arah, ma'ar,
sl naked space ; ma'drah, a plain or field devoid of trees.

In note 261 I have advanced the opinion that malae is in form a condi-
tional derivative of lae. This holds of the signification found in Nuclear
Polynesia. The secondary sense which the Tongafiti carried to eastern

370 THE POLYNESIAN WANDERINGS.

Polynesia has obscured the lae element ; but the sacrosanct content of the
viarae in the four-godded theology of eastern Polynesian is after all but a
logical outgrowth of the Nuclear Polynesian malae as the civic center of
social life where god is sole, supreme — and Lucretian. We note with
interest that the Maori marae is Nuclear Polynesian rather than Tongafiti —
again that Samoan voyage due south.

In the Viti vara is the town green. Mara, therefore, shows the condi-
tional form of this abbreviated ra-stem.

In Efate mala is the homotype of Viti vara, and the subduplication
malala is clearly a duplication of stem la under the conditional prefix. It
is clear, then, that malmal, duplicated mal, can have nothing to do with
mala, and there is nothing in the signification to require its association
therewith.

In the remainder of our Melanesian material we shall have no difficulty
in following out the signification by metonymy, its most conspicuous area
standing for the town which surrounds it. The word public house in Epi
and Malekula is awkward as suggesting the tavern idea with its connotation
in the English mind of the traffic in liquors. Island hospitality maintains
the largest house in every hamlet for the reception of visitors, and as this is
always at the point of honor on the town green it might reasonably assume
the designation thereof. Aneityum has inmaleom, inmaliyum, inmalyum,
town or city. At first sight, after dissociating the article in, they look
like malae homogenetics. They are really composites of in-mal a collection
of objects and eom (265) house.

Without supporting evidence the Malay balai (Tregear's suggestion) is
no more than an interesting resemblance.

The Semitic stems in 'rh. The resemblance to the Polynesian lies only
with respect to the I of the simple /a-stem, and this is not enough to serve
as a basis for further deduction.

316.

malum, weak, faint, soft. Cf. malua 4.

Samoa: main, gentle, easy, soft. Tonga: main, loose, soft, mild,
easy. Uvea, Nukuoro: malu, tender, soft. Hawaii: malu,
quiet. Futuna: malu, tender. Nuguria: mam, soft. Tahiti:
maru, soft, gentle, easy. Paumotu: hakamaru, to grow milder.
Rapanui : maruaki, to decay.

Viti: malumu, weak, faint, sick, soft.

Nggela, Savo, Bugotu: malumu, soft. Duke of York: malumalum,
to faint with hunger; galom, soft. Malekula: malum, soft,
meek, gentle. Laur: malmalungana, weak, feeble. Maewo,
Merlav, Malo, Mota: malumlwm, soft. Mota: mala, id. Vuras,
Santo: melumlum, id. Mosin, Norbarbar: molumlum, id.

Pak, Sasar, Motlav: mulumlum, id. Aneityum: mulmul, id.
Vaturanga: maluka, id. Lo: melunglung, id. Alo Teqel:
mulunglung, id. Volow: melemwlemw , id. Lambell: mala,
weak, feeble. Tangoan Santo: nalum, soft, meek, gentle.
Fagani: marumurumu, soft. Lamassa: martin, calm (of the
wind. Mukawa: merumeruna, soft. Kubiri: m em eruna, id.

DATA AND NOTES. 371

Saparua, Teluti : main, soft. Matabello, Gah : maluis, id. Amblaw :
maloh, id. Batumerah: mahita, id. Malagasy: lemi, soft-
ness, meekness; malemi, soft, meek, gentle. Malay: lamah,
soft, flexible, weak. Java: lamas, id.
Arabic: haluma, halim', to be gentle, weak.

The stem is malum or malumu. This is in form a conditional of lumu,
a primitive not yet as such identified.

In the Polynesian we have the utmost abrasion in main. This may per-
haps be identified with Lambell mala. I have associated herewith Lamassa
manlu; we are not yet sufficiently acquainted with this recently reported
language to know whether n is such an infix as is common in the not distant
Indonesia, or if man is a local form of the common wa-conditional.

The malum form appears in Efate and Malekula. Vaturanga maluka is
not properly in this family ; it forms a small group with Eromanga molok-
loku, Sesake manukunuku, Mota manoga, and Motu manoka. On the m-n
mutation in Tangoan Santo malum see note 312. The Duke of York galom
may not be malum, but there can be no doubting the identity of lorn.

The subduplicated malumlum occupies an interesting position in the New
Hebrides and is illuminative of the manner in which loan material is broken
in foruse byan alien race. Itwillmore convenientlybe studied by dissection.
The ma-conditional retains its a in Mae wo, Merlav, Malo, Mota. It becomes
me in Vuras, Santo, Lo and Volow. Mosin and Norbarbar have altered it
to mo. Pak, Sasar, Motlav, and Alo Teqel have allowed it to degenerate
to mu. If we regard this last as an attraction to the vowel of the lum
element we shall find the same principle of attraction operative in Volow
wielemwlemw, but not in Lo melunglung. The lum-stem remains unaltered
over the greater part of this subdistrict ; in Lo and Alo Teqel it becomes
lung; in Volow it not only changes to the difficult mw, but accompanies it
with a vowel change to e.

Aneityum mulmul offers a problem. Codrington (Melanesian Languages
91) suggests the probability that it is metathetic for lumlum; metathesis is
rarely employed in Aneityum, but see pula (284) to shine Aneityum laav,
lav, and it would be the solitary instance in which we have identified the
primal stem lum. On the other hand Aneityum has taken such liberties
with its Polynesian loan material that if we regard mulmul as a degraded
mulumulu the foregoing note as to the impracticability of Efate malmal
would have far less application. The duplication anomalies of Duke of
York m alum alum and Laur malmalungana are less considerable when we
record that the remote languages of the eastern portal have developed their
duplication mechanics along lines quite other than those which I have
established for Polynesia.

In Indonesia the best identification, despite wide vowel diversity, is the
Malagasy. Inverting the apparently metathetic Java and Malay forms we
find stems in s, malas, malah, akin to Amblaw maloh, probably to Matabello
and Gah maluis, and by a frequent change akin to Batumerah maluta. For
this reason the malu of Saparua and Teluti, despite its present identity with
Polynesian malu, probably is an abraded mains.

In the Arabic haluma we should have a close resemblance if it were
possible to establish the identity of ha as a conditional and homogenetic
with ma.

372 THE POLYNESIAN WANDERINGS.

3*7-
manu, a bird.

Samoa, Tonga, Tahiti, Mangareva, Rarotonga, Mangaia: manu,
birds and animals in general. Futuna, Niue, Uvea, Fotuna,
Bukabuka : manu, animals. Maori, Hawaii, Nnguria, Rapanui,
Marquesas, Tongarewa : manu, birds. Paumotu: manu, birds;
manumanu, animals, insects.

Viti: manumanu, birds and animals. Rotuma: manman, birds.

New Ireland (Carteret Harbor) : manuk, bird. Tanna: manug, id.
Motu, Sesake, Epi, Bierian, Maewo, Mota, Ulawa, Wango,
Bululaha, Fagani, Saa, Arag, Omba, Vaturanga, Bugotu, Nggela,
Nguna, Ruavatu, Ugi, Belaga, Pokau, Kabadi, Sinaugoro,
Hula, Keapara, Galoma, Rubi, Suau, Nada, Awalama, Taupota,
Mukawa, Wedau: manu, id. Uni: manumanu, id. Lambell,
Lamassa, Tubetube, New Ireland (Carteret Harbor): mani, id.
Kiriwina, Dobu: manua, id. Moanus: manual, id. Merlav,
Mota, Norbarbar, Gog, Vuras, Mosin, Tubetube, Murua, Pala,
Laur: man, id. Brierly Island: maan, id. Aneityum:

in-man, id. Makura:na-waw,id. Eromanga: menok, id. Baki,
West Epi: menu, id. Pak, Motlav, Norbarbar, Sasar, Volow:
men, id. Malekula: ni-min, id. Vanua Lava, Lo, Alo Teqel,
Retan: mow, id. Alite: malu, id. Marina :nanu, id. Kubiri,
Raqa, Oiun: mamu, id. Lakon: mah, id. Tagula : ma, id.

Malay, Sulu, Visayas: manuk, bird. Guaham: manug, id. vSavu,
Kisa, Menado, Sanguir, Sula, Morella, Caimarian, Baju, Sali-
babo: manu, id. Togean Islands: manu, the domestic fowl
(tonji, bird). Bouton: manumanu, bird. Amblaw, Awaiya:
manue, id. Cajeli: manui, id. Mayapo: manuti, id. Teluti:
manuo, id. Ahtiago: manuwan, id. Kayan, Magindano,
Matu, Gah, Matabello, Teor: manok, id. Bolanghitam:

manoko, id. Saparua, Lariko, Liang, Batumerah: raawo, id.
Gani : manik, id. Waigiou Alfuros : mani, id. Dyak : monok, id.
Wahai: malok, id. Malay: burung, id. Malagasy : vuruna, id.
Hebrew: />arafr, to fly. Syriac: parah, id.; par oh to, birds. Arabic:
farhu, young of birds. Hebrew: efroah, id.

Several of the early missionaries comment with a fine sense of humor upon
the mistake the islanders made in calling the cow when first seen a bird.
This is the word which led the good missionaries into the error of their own
ignorance.

Manu is as wholesale in its signification as our word animal, it is generic.
In the paucity of brute mammalia the first missionaries found this general
term most frequently used of birds, and it was their and not a Polynesian
mistake to translate manu into bird. In the material here collected it will
be seen that the significations animal and bird are widely extended. In
the Paumotu insects are included ; the same is true of Mota, where manu
signifies beetle as well as bird. Nor is its applicability restricted to earth
and air; it reaches into the sea as well. Samoa uses ilamanu (fish-animal)
for the whale, and see note 130 for a discussion of manu in the fish sense.
Lakon mah means bird and fish. If this stood by itself we might accept it

DATA AND NOTES. 373

as manu in an advanced stage of dilapidation ; but since it appears as mes
in Vuras and as mast in Maewo we see that it can not be from the waww-stem.
Aside from this Lakon mah the Melanesian identifications are sufficiently
satisfactory to pass without detailed comment. We find a final palatal in
New Ireland, Tanna, and Eromanga. The former might be ascribed to a
Post- Polynesian Indonesian source, but Tanna and Eromanga lie outside
the limits which we may conveniently assign to the raids of Dyak prahus.
On the other hand, if we ascribe the fe-final to a stem manuk we shall be
forced to regard that as earlier than manu, both from the internal motion of
the language and from the fact that these two languages in general preserve
primitive forms. Two objections arise in opposition to the view that manuk
is primordial : one that the fe-final has not been preserved in those Melane-
sian languages which have an idiosyncratic disposition toward closed stems,
and these are the languages in which the word appears as man and variants ;
the other that Indonesia in general preserves the later rather the earlier
Polynesian forms, Tongafiti rather than Proto-Samoan. The Moanus
manuai may be a local suggestion of inflection; yet since that language is
the propylon of the eastern gateway, the nearest to Indonesia, we need
have no hesitation in recognizing its association with such forms as Ahtiago
manuwan, Teluti manuo, Amblaw and Awaiya marine, Cajeli manui. These
occur in Ceram and Buru, south of Gilolo, and one might expect to find their
influence most prominently felt through the southern gateway; but our
earliest records from New Guinea distinctly mention Ceram as the source
of the Malay raids along the north shores of that great island, therefore in
the direction of Moanus.

The preponderance of manuk-stems is manifest in the Indonesian record.
Wahai malok shows the same n-l mutation as Alite main. Malay burung
and Malagasy vuruna are clearly homogenetic inter se; equally they are dis-
sociated from the common word of the three island areas.

If the Semitic has even a resemblance it can only be to the Malay and
Malagasy, therefore none with the manu or manuk of this study.

318.
mate, to die ; matemate, to be quiet, soft, gentle ; matian, death ; matingo,
the grave.
Samoa: mate, to die (used of beasts; oti used of mankind). Tonga,
Fakaafo, Futuna, Niue, Uvea, Vate, Maori, Tahiti, Rarotonga,
Marquesas, Mangareva, Bukabuka, Manahiki, Paumotu: mate,
to die. Nukuoro: mate, id.; hakamate, to kill. Rapanui:
mate, ill, dead, to die; hakamate, to murder. Tongarewa: mate,
death. Nuguria: umate, dead. Aniwa: komate, to be dead.
Hawaii : make, to die. Fotuna : kono-mate, to die ; tah-mate, death.
Viti: mate, to die.

Merlav, Maewo, Omba, Arag, Marina, Malo, Bierian, Sesake, Nggela,
Vaturanga, Nguna, Motu, Solomon Islands, Suau, Mota: mate,
to die. Tubetube: unui-ia-mate, to kill. Fagani, Wango,
Mekeo,Ulawa; ma'e, todie. Doura: make, id. Kiriwina: katu-
mata, to kill. Dobu : loe-masa, id. Saa : ma'e, ma'esie, to die;
ha'ama'esi, to kill. Ugi : mae, to die ; haamaesi, to kill. Santo :
mati, id. Nada, Massim, Murua : matt, to die. Mosin, Gog,

374 THE POLYNESIAN WANDERINGS.

Lakon, Norbarbar, Retan, Pala, Laur, Lamassa, Duke of York,
Motlav, Volow : mat, id. New Britain: matmat, makmak, sick,
faint. Lambell, King: imat, to die. Buka: nimat, matte,
maten, sick; mat, matte, amatte, dead. Pak, Alo Teqel; ma',
to die. Sasar: ma, id. Ambrym: mar, id. Sinaugoro:
mase, id. Motu: maze, id. Kiriwina: masisi, id. Aneityum:
mas, id. Vuras: meat, id. Lo: met, id. Malekula : mej, id.
New Ireland (Duffield) : oumet, id. Lifu: ram, id. Leon:
me', id. Iai: mofe, id.
Motlav, Vuras, Lakon, Retan, Marina, Vaturanga: mate, death.
Merlav, Malo, Mota, Maewo: matea, id. Omba, Bierian, Arag,
Nggela : matcana, id . Fagani: ma'efa, id. Wango (upland) :
ma'ha, id. Ulawa: ma'enga, id. Mosin: wa/a, id. Pak,
Sasar: ma' a, id. Volow, Norbarbar: mete, id. Alo Teqel:
rae'e, id. Lo: miji, id. Malekula: mejan, id.
Macassar : mate, dead. Malay : mati, to die. Malagasy : maty, dead ;
matimaty, lukewarm. Kisa: maki, dead. Magindano: to die.
Arabic : mata, to die, to become calm (of wind), to soften by cooking
The employment of oti in Samoa for the death of a man is not so much
euphemism as the paying of proper respect to the superior animal, for oti
(219) means finished. To the high-minded Samoan there can be no sym-
pathy with Ecclesiastes : "the sons of men are a chance and the beasts are
a chance, and one thing befalleth them ; as the one dieth, so dieth the other ;
yea, they have all one spirit; and man hath no preeminence above the
beasts."

Not the meanest Samoan sinks to the level of the brute, even in the
article of death. The beast dies, the humblest of men at least finishes.

For them of high degree there is yet more compliment ; the head of the
house dies not, but goes to the council (usufono), the chief goes on in majesty
(maliu, afio), the heavens are rent in twain (inasaesaelelangi) , and the rain
beats down (timuto). Certain families die with circumstance. Death comes
not to Tangaloa, he has but gone far afishing (ta'atin) ; and similarly on
Asiata's demise the stools of bamboo are swept clean away (tafealetau'ofe),
such a fishing has he gone upon. When Tuala goes the very winds hold
their peace (matangitongaina); over the funeral mats of Fiame the moon
comes tumbling out of the sky (pa'Tdemasina) ; when Te'o's eyes are closed
darkness palls the land (polenu'u), and when Mata'afa leaves the earth the
heavens are turned upside down (mafulilelangi) . If this be euphemism it
is of the nth power.

In Melanesia we find a considerable area in which the full form of the
mate-stem is preserved, and within it is a district at the extreme south of
the Solomon Islands in which the t vanishes. Rudely and at a considerable
distance Lifu meet may be considered to preserve the full stem, yet with
such a t-k mutation as is found near by in Iai mok, in one of Tregear's
unprecisely localized New Britain forms, and in Hawaii and in the most
modern Samoan. Interlaced with the mate-area is another equally consid-
erable in which we find the abraded mat. Aneityum mas rests upon the
t-s mutation, which is rather well established; it appears in Ugi haamaesi
and occurs again in sa (337) bad Aneityum has. Vuras meat probably

DATA AND NOTES. 375

results from the insertion of e before a and i before e in a closed syllable
which Codrington (Melanesian Languages, 322) notes as characteristic of
Volow; there is, however, a possibility that it is rudely metathetic. Abra-
sion proceeds still further to ma in Pak, Alo Teqel, and Sasar, and with
vowel change to me in Leon. The noun forms follow the same course.

In Indonesia the stem is rare, but the identifications here presented are
satisfactory.

The Semitic bears a strong resemblance.

3i9-
me, meme, urine; me, mea, to make water, to flow, to wet.

Samoa: muni, to make water; mianga, urine. Tonga, Futuna,
Niue, Tahiti, Rarotonga, Marquesas, Mangareva: mimi, urine,
to urinate. Paumotu, Nukuoro: mimi, to urinate. Nuguria,
Rapanui: mimi, urine. Maori: mi, to urinate; mimi, urine,
to urinate; mianga, urination, to urinate. Hawaii: mi, mimi,
mia, to urinate ; mii, miana, the place for urinating, the member
employed; mimi, urine.

Viti: mi, mimi, urine, to urinate.

Lamassa, Lambell, New Britain: mimi, urine, to urinate. Laur:
mim, urine. Aneityum: ami, to urinate. Tanna: (t)ami, id.;
n-ami, urine. Baravon, Duke of York: minimi, to urinate.
King: minime, urine. Motu: mei, urine. Mota: meme,
urine, to urinate. Malekula: meme, to urinate ; ne- me, urine.
Malo : meremerc, urine, to urinate.

Malagasy: amani, urine; mamani, to urinate.

Arabic: maha, to have water (of a well), to leak (of a ship), to
pour water. Hebrew: me, euphemism for urine.

The Proto-Samoan stem is clearly mi, for had it been a closed stem the
commonly abraded consonant would have been preserved in the Samoan
derivative, whereas we find it mianga.

This stem is preserved through Melanesia with but few variations. For
Malo meremcre we can suggest no principle upon which to account for the
r. Our other variants look toward Malagasy in two ways. Aneityum and
Tanna have its introductory a, which is inexplicable. In the eastern gate-
way we find the pure Polynesian mi, yet within sight of New Britain and
New Ireland we find minimi on the Duke of York Group and even in New
Britain itself in Baravon; and in New Ireland King minime marches with
Laur, Lambell, and Lamassa mimi, mim. This introduces the Malagasy n.
I prefer to regard it as an infix and ascribe it to Post-Polynesian influence
from Indonesia.

As we have no supporting evidence in Indonesia the Malagasy forms may
not wholly be accepted nor yet wholly denied.

The Hebrew me is a resemblance, the Arabic maha is foreign in sense and
not easily to be reconciled in form.

320.
melu, shade, protection; melu, melumelu, to be shady.

Samoa, Tonga, Futuna, Niue, Uvea: mala, a shade, a protection,
to be sheltered, to be shaded. Hawaii: main, a shade, a

376 THE: POLYNESIAN WANDERINGS.

shadow. Tahiti, Raro tonga, Mangareva: maru, id. Rapanui:

hakamaru, to cover with shade; marumaru, a shade, dark;

koona (place) marumaru, a shelter. Maori: maru, to be

shaded, sheltered. Fotuna: marumaru, to be clouded, shade.

Nukuoro: maru, a shadow. Paumotu: hakamaru, to shadow.

Marquesas: man, a shade, a shadow.
Viti: malumalu, shade.
New Britain : malur, shade. Baravon: malmalur, shady. Duke

of York, Kabakada : marum, night. Matupit : marum, darkness.

Uambell: morrom, dark. King: mirrum, id. Mota: malu,

shade. Malo: mala, id. Malekula: na-mor, id. Baki:

vamelu, id. Bierian: fomelu, id.
Malagasy: malomaloka, shady, cool, gloomy.
Hebrew: 'afel, 'afal, obscure, dark; 'amel, 'amel, to languish, to

droop, to hang the head.

The Proto-Samoan stem is malu, in contradistinction from malu cool of
the malung-stem.

The Polynesian and Efate are in sense agreement. The vowel a of
Samoan malu would readily be represented in Efate by e; the e is not
explained.

In other of the New Hebrides the identifications may be followed readily
as noted above. For comparison I include here Aneityum nalmui, nalmun,
picture, shadow ; removing the n-article and restoring the attracted vowel
to its proper position we find a stem lamu which might be admitted as
metathetic malu. In the eastern gateway we find what seems to be a very
probable group of identifications ; in sense the passage from shade to dark-
ness and thence to night is a rational chain, and there is no more form
difficulty in malu-malur-marum-mirrum.

The Malagasy is probably homogenetic, but in the absence of central
Indonesian forms we may not be positive.

Where the Hebrew agrees in meaning it is remote in form; where the
form seems to come into some suggestion of resemblance the sense is remote.

321.

mini -ngi, minu, minu ngi, munu, munu-ngi, munuina, to drink.

Samoa, Fakaafo, Nukuoro, Tonga, Futuna, Niue, Uvea, Nuguria,
Maori, Tahiti, Hawaii, Marquesas, Mangareva, Mangaia, Tonga-
rewa, Rarotonga, Manahiki : inn, to drink. Aniwa : inumia, id.
Rapanui : unu, to drink ; hakaunuora, to wrater, to make to drink.
Sikayana: unu, id. Fotuna: no-cinu, id.

Viti: ngunuva, unuma, to drink. Rotuma: inu, id.

Malo, Roro, Mekeo, Motu, Waima: inu, to drink. Motu, Pokau:
inua, id. Nggela: inu, inuvia, id. King: iniim, id. Kabadi:
inura, id. Uni: inui, id. Hula: niu, id. Sinaugoro,
Keapara, Galoma, Rubi: inua, id. Uni: bibinu, id. Duke
of York: inim, id. Motlav: in, nin, id. Malekula: min, id.
Mota : tin, unuv, ima, id. Panaieti : im, id. Misima : nai-im, id.
Lo: m-un, id. Nguna: manu, id. Makura: munum, id.
Epi (Baki, Bierian): muni, id. Marina: o'omia, ulumia, id.

DATA AND NOTES. 377

Sariba, Tubetube, Dobu: numa, id. Tanna: numi, id. Suau:

nom, id. Tavara: uma, id. Taupota, Wedau, Galavi,

Boniki: umai, id. Awalama: umaia, id. Kiriwina : mum, id.

Tanna: (t)amaniim, id. Kiviri: toman, id. Raqa: tanuma, id.

Kubiri: foraa, id. Oiun: torn, id. Murua: amomu, id.

Aneityum: umnyi, id.
Malay: minum, pinum, to drink. Pampangas: minom, id. Magin-

dano: ominum, id. Tagalog: ominom, id. Malagasy: ram-

nuna, id. Java: nginum, id. Togean Islands : mangino, id.
Syriac: mistuta, to drink. Chaldee: .y'/'a', }is't'o\ id. Hebrew:

.y'a/'a, id. Ethiopic: sataya, id. Hebrew: s'akah, id. Arabic:

saka\ id. Ethiopic: sakaya, id.

It is by no means easy to determine the primal stem. We shall best
consider the series of forms, paying no regard here to the vowel change.

inum. Polynesia ubique, Viti, Rotuma, Malo, Motu, King, Duke of

York, Motlav, Mota, Tanna, Marina.
minum. Efate, Malekula, Lo, Nguna, Makura, Baki, Bierian, Tanna,

Malay, Pampangas, Magindano, Tagalog, Java.
minung. Efate, Malagasy, Togean Islands.
inuv. Nggela, Mota.
nginuv. Viti.

The vowel interchange appears not to be critical, for most of the t-forms
are paralleled by similar u -forms.

The common factor of all these stems is inn; this may be modulated by
a final m, ng or v, and by a frontal m, ng. If we take the position that m
and ng may stem in a common parent we have simplified the problem in
but a small degree ; we still have to consider an inn which may be intro-
duced by a nasal and which may close in a nasal or a labial spirant.

After the preceding stem classification we need comment on but a few
more irregular phases. In Motlav nin we find an uncoordinate form; in
may be abraded inu, nin may perhaps present the central nasal n, which
should stand in the position of the parent of initial ng-m, or it may be some
obscure duplication. Mota un, nnuv, falls systematically into the scheme
of stems ; but the Maligo ima seems impossible to place. The Marina forms
suggest affiliation with the mwm-stem ; the n-l mutation is well established
over a wide area, but this is the only case in which it is found in Marina ;
and the excision in o'omia is left unexplained. Tanna numi and Epi muni
are most readily disposed of as metathesis, of the types 2341 and 4321,
respectively. Aneityum is very obscure, yet it does seem to preserve a trace
of minu.

There is, as we have seen, abundance of confusion in the island areas;
but not so much as one confounded form comes into the slightest resem-
blance with the Semitic st root.

322.

mirama-ni, merama, to be light, to shine; e meromina, in the light (as
opposed to abokas, the dark and gloomy underworld of Hades),
in the world, the world.

378 THE POLYNESIAN WANDERINGS.

Samoa: malama, the moon, a light, a lamp or torch, to be light;
malamalama, to be light. Tonga: malama, brightness, to
shine. Futuna : malama, the world, the universe, a lamp or
light, brilliancy. Uvea: malama, a light, a lamp, to light up.
Hawaii: malama, light; malamalama, a light, to shine. Sika-
yana: malama, the moon. Manahiki: malamalama, light.

Fotuna: marama, light; mrama, the moon. Nuguria: marama,
light. Maori: marama, the moon, light, to be light, to be
bright. Tahiti: marama, the moon ;maramarama, the light, to
be light. Mangareva : marama, daylight, the moon. Rapanui :
marama, brightness, light, day, to be bright; hakamarama,
to shine. Rarotonga: marama, the moon, the light, bright,
shining. Nukuoro: marama, the moon, bright. Niue:

maama, light, to light up. Marquesas: maama, light of day;
meama, the moon. Aniwa: umrama, the months. The month
sense is found in Tahiti, Marquesas, Rarotonga and Maori asso-
ciated with the moon signification, and in Hawaii is specificallv
dissociated therefrom to characterize a solar month.

Viti: malamalawa, the early part of the morning before daylight;
rarama, light, rdmaka, to cast light upon; ramaka, shining from
a distance.

Sesake: marama, to be light. Gog: marmaran, to be lighted.
Motlav: memreren, to become light. Volow: merercn, id.

Tanna: mararen, light. Mota: maran, light, morning. Merlav:
maran, light ; mamaraniga, lightsome. Bugotu, Nggela : marara,
light. New Britain: malana, light. Kalil: mdlau, bright.
Nengone: nerenc, to shine; ncreneni, to light. Aneityum:

cf. alauma, to blaze. Bierian: mamama, light.

Kayan: mala, a light, flame.

Arabic : lama1 a, to shine.

The Proto-Samoan stem is malamang, as shown in malamalamangia.
The duplication form shows this to be a composite of the two stems mala
and mang; thus it must be kept apart from forms of lama, which in Poly-
nesia has the lighting sense only as secondary to its primal signification
of a torch.

The presence of n in Efate mirama-ni and eromina serves to associate
these forms with the malamang-stem. Sesake marama is plainly Polyne-
sian malama.

Beyond these two points we seem forever on the verge of establishing
an identification and as steadily the light proves to be but will-o-the-wisp
glimmer. In Viti malamalawa is almost Samoan malamalama; this would
require an m-w mutation; but m is the least variable of Polynesian conso-
nants, our material not affording one instance of this hypothetical mutation ;
the nearest is m-v in the solitary instance of ma'i (323) sick Nggela vahagi.
The other Viti examples stem in lama, and that is contraindicated.

The nearest approach to malama in the doubtful Melanesia is New Britain
(not specified) malana; on the m-n mutation see note 312. The marara of
Bugotu and Nggela requires an m-r mutation, for which we find no support.
In the New Hebrides we find various phases of a stem which at its simplest

DATA AND NOTES. 379

is found as maran in Mota and Merlav. It is quite plain that if maran is
to be associated with malamang at all, it can only be with the composition
element mala; this maran may be recognized in New Britain malana in its
full strength and, after abrasion, as a subduplication form in marara. From
maran the duplication of mar gives us Gog marmaran, yet side by side with
this we find in Merlav mamaraniga a duplication of ma and not mar. Tanna
mararen and Volow mereren require a peculiar duplication of ran. Motlav
mcmreren joins to this anomalous duplication of ran an equally irregular
duplication of ma. Kalil mdlau may derive from the abraded form of
maran. Nengone ncrene, if we may accept the initial m-n mutation, brings
us to a yet more archaic form of maran as an open stem marane, and that
the ne is radical is made manifest by the subduplicated nereneni.

Kisa mala is in kinship with this maran.

The Semitic Im' skeleton is alien.

323-

misaki, masaki, to be sick, to be ill, to have fever ; misaki, misakia, sickness.

Futuna: masaki, sick, illness. Tonga: mahaki, id.; mahamahaki,
feeble. Niue: mahaki, very great. Uvea: mahaki, sick, ill.
Maori: mahaki, a cutaneous disease. Marquesas: maki, a
wound. Rapanui: maki, a wound, plague. Mangaia : maki ,
sick, sickness. Mangareva, Nuguria, Paumotu: maki, sick,
ill. Fotuna: maki, makinga, ill. Samoa: ma'i, sickness,
to be ill; very. Tahiti: mai, disease. Hawaii: mai, sickness in
general. Rapanui: mai, to be ill; a boil.

Viti : mathake, specifically aphthae or thrush.

Malekula Pangkumu: mcsek, sick, sickness. Epi: msaki, miei, id.
King: miseit, sick. Saa: mated, id. Nggela: vahagi, id.
New Britain: maki, makmak, matmat, id. Baravon: mail, id.
Aneityum: mehe, sick; masaki, leprosy. Mota: masag, ague.

Ilocan: masaquit, sick. Silong: makit, id. Kisa: maki, dead.
Malay: sakit, sick.

Arabic: s'aka' , to afflict one with a disease; s'akat,disease^,mas'kuw7v,,
afflicted with a disease.

In Polynesia the line of demarcation between masaki and maki is the
classic division between the Proto-Samoan and the Tongafiti migrations,
masaki pertaining to the older stock. That Samoa has the Tongafiti form
need cause no surprise, for in Samoa the Proto-Samoans were held under
Tongafiti subjugation until the historic battle of Matamatame. Similarly
the presence of Proto-Samoan masaki in the Maori is but one more incident
of that direct migration from Nuclear Polynesia to New Zealand which we
have already isolated, and this reading is corroborated by the special use
of mahaki as the name of one disease in particular. We observe a like
particularization in Rapanui, Viti, Aneityum, and Mota.

In Niue the expletive mahaki finds its parallel in Samoa in the same use
of ma'i. "A plague on both your houses."

Efate preserves the true masaki form, and Epi (Bierian) msaki is but
slightly altered therefrom. Nggela vahagi would be reducible to masaki by

380 THE POLYNESIAN WANDERINGS.

applying an m-v mutation. Mota masag and Malekula Pangkumu mesek
are abraded forms. Aneityumme/^ is a further abrasion of mesek. Saadrops
all its k's; therefore we can restore matai to mataki, hut we have no instance
of s-t mutation in that language and perforce must deny ourselves this
identification. King miseit needs but the restoration of k to come into
agreement with Indonesian forms which have final t, Ilocan masaquit and
Malay sakit. Baravon mait by the same restoration comes into agreement
with Silong makit of the Tonganti stem. Furthermore New Britain maki
is the same as Kisa maki. These three forms may be ascribed to Post-
Polynesian intercourse with Indonesia.

In form masaki sick is a conditional and implies a primal noun saki illness.
This saki is found perhaps in Maori hakihaki the itch, and then not in quite
the signification demanded ; again in the Malay with the same falling short
in sense. Thus we see that, while the Arabic with its s'akat and mas'ktiww'
offers a specious resemblance, it is after all a resemblance to something
which does not really exist in our island languages.

324-
mita, meta, bakamita, to look at, to watch, to observe, to view; mita na,
the eye, the beginning, a bud or shoot, a window or door or
other opening.

Samoa: mata, matamata, to look at, to see, to view; mala, the eye,
the face, the point, the edge, a mesh, the source or origin.
Fakaafo, Moiki: mata, the eye. Nukuoro: mata, the eye;

matakite, soothsayer; mata, the point; matakapupu, blunt;
matapunou, blunt; mata, the face, looks; matapaupau, ugly;
matatcnua, pretty; mata, any small thing; mataua, raindrops.
Tonga: mata, to be seen; mamata, to see; mata, the eye, the
face, the edge, a mesh. Futuna: mata, the eye, the face, the
point of a lance; mamata, matamata, to see, to look at, to view.
Niue: mata, mamata, to look at; mata, the eye, the face, an
edge, a blade, a point of land. Uvea: mata, the eye, the
face, a point; mata, matamata, to look like. Nuguria: mata,
the face, to see; anomota, the eye. Maori: mata, the eye, the
face, the edge, the point, the mesh; matamata, the point,
extremity, source, a headland. Manahiki: mata, the eye.
Tahiti: mata, the eye, the face, the beginning, the edge. Raro-
tonga: mata, the eye, the face, the beginning. Marquesas:
mata, the eye, the face. Mangareva : mata, the eye, physiog-
nomy, the front, the point. Paumotu: mata, the appearance
of a person. Rapanui: mata, the eye, face, visage, aspect.
Tongarewa: mata, the eye; matamata, beads. Fotuna, Aniwa
foimata, the eyes. Sikayana: karimata, the eye; lofimata,
the face. Hawaii: maka, the eye, the face, the point, the
edge, a bud.

Viti: mata, the eye, the face, the front, the point, the source, the
origin. Rotuma: maf, the face; maja, eye.

Vaturanga, Pala, Kiriwina, Nggela: mata, the face. New Georgia:
mata, isumata, the face. Southeast Epi, Bierian, Sesake,

DATA AND NOTES. 381

Malekula, Arag, Maewo, Malo, Redscar Bay, Motu, Sinaugoro,
Rubi, Mugula, Sariba, Tubetube, Misima, Nada, Murua, Dobu,
Awalama, Taupota, Wedau, Galavi, Boniki, Mukawa, Kwagila,
Kubiri, Raqa, Kiviri, Ohm, Port Moresby, Buka, Lambell,
King, Lamassa, Moanus, New Ireland (Port Praslin), New
Britain, Duke of York, Solomon Islands : mata, the eye. Suau,
Tavara : mata, the eye, face. Lo : mata, a spear. Norbarbar :
matah, id. Maewo : mataso, id. Merlav: matas, id. Mota:
mata, the eye; matag, to see; matarag, to gaze at. Motlav:
mata, na-mtege, the eye. Laur: mata, the eye; mat mata, the
face. Baravon: mata, the eye. Volow: mata, the eye ; met,
to see. Merlav: mete, eye; mata, to see. New Ireland
(Carteret Harbor) : matak, the eye. Nggao : matata, the face.
Brumer Islands : matada, the eye. Brierly Island : matara, id.
Vuras: matai, id. Gog: mate, eye; matarag, to gaze at.

Motlav, Vuras, Lakon : mate, eye. Kiriwina: mati, id. Iai:
emakang, eye ; nimakan, the face. Uni, Pokau, Doura, Kabadi :
maka, the eye. Aneityum: nes-ngani-mtan, ni-mtan, id.

Panaieti: matan, id. Deni: maku, the face. New Ireland:
mala, the eye. Tagula: mara, id. Pak: ma' an, id. Roro:
maha, id. Mekeo: raa'a, the eye, face. Ulawa: waa, the
face. Fagani, San Cristoval, Malanta: ma, the face. Hula,
Keapara, Galoma: ma, the eye. Ambrym: meta, the eye.
Santo: metana, id. Malekula: metan, id. Mosin: m«fe, id.
Vuras, Mosin: mete, eye; meteg, to see. Alo Teqel: me'e,

me'egi, the eye. Pak: me'ei, id. Duba: wa/, id. Gog:
met, id. Volow: «!d, to see. Lifu: 7wefe, the eye. Maewo:
ete, to see. Norbarbar: et, id. Tanna: nanime,nugane-mti(n),
the eye; nupugane-mti, the face. Eromanga: nipmi, ni-
mtu(m), id. Marina: nata, id. Baki: rat'ra, eye, face.

Tubetube: mani, the face; manipo, the eye. Panaieti, Dobu:
martini, the face. Misima: maneni, id. Murua, Kubiri, ma^',
id. Nada, Kiriwina: ra^z, id. Galavi, Boniki: mag'ig'i, id.
Kayan, Sulu, Savu, Ilocan, Tagalog, Pampangas, Bouton, Sanguir,
Liang, Wahai, Baju, Togean Islands, Salayer, Menado, Bolang-
hitam, Morella, Lariko, Saparua, Caimarian: mata, the eye.
Malay: mata, the eye, the blade, the edge, the mesh, the source,
the origin. Macassar: mata, the point, a mesh, a spring, the
source. Awaiya : mata mo, the eye. Nicobar Central : matsha,
the face. Matabello: matada, the eye. Silong: matat, id.
Ahtiago Alfuros: matara, id. Wahai: matalalin, the face.
Gah: matanirta, the eye. Teluti : matacolo, id . Batumerah:
ma tava, id. Baliyon: matoh, id. Ahtiago: matan, id. Dyak:
maten, id. Teor: matin, id.; matinotin, the face. Kar

Nicobar: wa^, the eye. Kisa: makan, id. Tagalog: mucha,
the face. Java: muka, id. Ilocan : muguing, id . Madura:
m«a, id. Mysot: wwi morolu, the eye. Malagasy: wmso,
the eye.
Arabic: 'ana, to emanate (water), to see, to look at.

382

THE POLYNESIAN WANDERINGS.

There is a wide range of significations in this stem. It will serve to
express an opening as small as the mesh of a net or as large as the door of a
house; it will serve to designate globular objects as large as the eye or as
small as the bud on a twig or the drop of rain, and designating a pointed
object it answers with equal facility for the sharpened tip of a lance or the
acres of a headland ; it describes as well the edge of a paddle or the source
from which a thing originates. As the islanders find no need to distinguish
these and yet other senses no obligation rests upon us to seek to establish
an artificial classification. We shall, therefore, in these notes consider only
the varieties of form.

In Polynesia the only variants are Hawaii maka, a modern kappation,
and Rotuma maja under the influence of a local idiosyncrasy.

It will be found making for greater simplicity to list the Melanesian forms
and the Indonesian irregularities.

mata. Vaturanga, Nggela, New

Georgia, Bierian, Se-

sake, Malekula, Maewo,

Malo, Arag, Buka,

Lambell, King, L,a-

massa, Laur, Moanus,

New Ireland (Carteret

Harbor, Port Praslin),

New Britain, Baravon,

Duke of York, Solomon

Islands, New Guinea

(Port Moresby, Redscar

Bay), Lo, Mota, Mot-
lav, Merlav.
matah. Norbarbar.
matas. Merlav.
mataso. Maewo.
rnatak. New Ireland (Carteret

Harbor).
matata. Nggao; Silong.
matada. Brumer Islands ; Matabello.

This is a startling picture of degradation, yet at no point may we halt
for any non plus ultra check to the dilapidation, for each step represents a
sufficient transition phase between that which has gone before and that
which is to come. Deni maku might suggest kappation, but even then the
final u remains out of accord. Iai emakang and nimakan are clearly kappa-
tion forms. In Tanna nanime and Eromanga nipmi it is possible that mi
or me represents a yet more degraded phase than appears in Fagani, San
Cristoval, and Malanta ma. Baki mira and New Ireland mala are to be
grouped in development plane, but we have no evidence to establish their
connection with mata.

In Indonesia mata has equal predominance, but there are many variants.
Of these, four are to be found in Melanesia: matada, matata, matara and mat.
Many of these Indonesian variants are composites. The principal stem

matara.

Brierly Island; Ahtiago

Alfuros.

matai.

Vuras.

mat.

Omba; KarNicobar.

mala.

New Ireland.

ma'an.

Pak.

maa.

Ulawa.

ma.

Fagani, San Cristoval,

Malanta.

mate.

Gog, Motlav, Vuras, Lakon.

meta.

Efate, Ambrym, Santo,

Malekula.

mita.

Efate.

mete.

Mosin, Vuras, Merlav.

me'e.

Alo Teqel, Pak.

met.

Gog, Volow.

mek.

Lifu.

mt.

Aneityum, Tanna, Ero-

manga, Motlav.

ete.

Maewo.

et.

Norbarbar.

DATA AND NOTES. 383

diversity is in the occurrence of »-forms in muka, mucha, muguing, mua,
and mut, this deviation not being found in the Pacific areas.
|# The1 Semitic 'ana resembles mata only in the possession of two a's, not
at all a basis of association.

325-
mot, motu, island, place, district; lit., that which is broken off.

Samoa: motu, an islet, a district; motu, to be broken off, snapped
asunder, severed. Tonga: motu, an island. Futuna: motu,
an islet. Niue: motu, land, island, a country, (?) a clump of
forest. Uvea: motu, an island; motu, to cut off. Nuguria:
motu, an island, to break. Maori: motu, anything isolated,
as an island, a clump of trees. Tahiti: motu, a low islet.
Marquesas: motu, an island. Mangareva: motu, an elevated
island. Rapanui: motu, an island, to cut, to break. Raro-
tonga: motu, an island, a grove. Paumotu: motu-puhere, an
island. Fotuna: tanga-motu, forest; motu tangata, multitude.
Viti: Moturiki, name of an islet near Ovalau.

Mota : motumotu, an island ; mot, bush, uncleared ground, land grown

over with trees; motu, to break, as string; mot, to cut, to break,

to stop short off. Duke of York: mutamuta, crumbs.

Malay: put us, to break. Malagasy: maitu, broken asunder,

snapped; maituitu, broken in pieces; utusana, being cut, broken,

^ snapped.

Arabic: makta', a place. (Deriving this from kata'a, to cut off, Dr.
MacDonald provides a common source for motu and koto-fi).

The Proto-Samoan stem is motus.

The island sense is particularly well marked in Polynesia, and the forest
sense is closely parallel. In Niue I have felt it right to query the forest
sense for the reason that it does not appear in the vocabulary, though it is
clearly inferential from PercySmith's sketch map of the island. In Viti the
word is found in one place name, Moturiki or little island, yet it is clearly
established since this is the only interpretation that the word can have. In
Melanesia it has not been recorded save in Efate and Mota. The Duke
of York mutamuta may be related, but it lacks confirmatory support.

The Indonesian is uncertain. The Malay requires evidence to uphold the
m-p mutation. In the Malagasy it is quite unlikely that the composite
ma-itu can have anything to do with motu, and utusana calls for an explana-
tion of the abrasion of m-radical.

In the Semitic if makta1 stems in a kt< root it can have no relation with
motus.

326.
mu, fu, to coo as a dove.

Samoa: mui, to murmur; mumu, words spoken in private conver-
sation; mumu, to be in swarms; langomumu, langofufu, the
carpenter bee; muna, to grumble; musumusu, to whisper.
Tonga: muhu, the sound as of persons talking together; muhu-
muhu, to speak quietly together; mumu, to collect together;
muna, to talk nonsense; langomu, a large fly. Futuna: mui,
to collect together; muna, to speak, to murmur; musu, to

384 THE POLYNESIAN WANDERINGS.

whisper. Niue: mumu, to crowd, to make a noise; langofufu,
the carpenter wasp. Uvea: munamuna, to murmur. Maori:
mu, a gentle noise; mumu, to murmur, to hum; muna, to tell
privately. Tahiti: mu, a buzz; mumu, to make a confused
noise, as a number of persons talking together; langomumu,
the carpenter bee; omuhumuhu, to whisper to one's discredit.
Hawaii: tomtom, to hum. Mangaia: mu, to sigh. Marquesas:
mumu, a confused noise. Mangareva: amui, to assemble, as
flies. Paumotu : muhumuhu, a dull confused noise ; muhimuhi,
to murmur. Fotuna : mu, to buzz ; mumu, to whisper. Moiki :
muna, to speak, to say. Nukuoro : fatumuna, to lie ; munaputonu,
true; silengamuna, to interpret. Rapanui: huhu, buzzing.
Viti : mumu, to go in troops or swarms.

Motu: mu, to coo. Loyalty Islands: mumu, the pitcher wasp.
Aneitvum: Hmu, to low (as cattle); ilmurilmu, to mumble.
New Britain: mukumuku, to whisper. Mota: mum, to make
an indistinct hum.
Nicobar Central: mumu, a dove. Malagasy: moimoy, a low mur-
mur, a hum; monomonona, a grumbling.
Hebrew: hamah, to coo, to hum (as a multitude). Arabic:
hamhamah, to murmur.
I have here collected the principal stems having raw-initial for purposes
of comparison. Connected with the mu of cooing there is a mu in Polynesia
and beyond, and it is with this that we shall concern ourselves. In Samoa
langomumu and langofufu it shows the same alternation as in Efate mu, fu.
This fu is found but once more, Niue langofufu. Tonga exhibits this ele-
mental mu only in the composite langomu. In Futuna this mu is absent;
for mux, which is the same in form as Samoa mui to murmur, is here formed
upon that mu to be a crowd which is found in Samoa, Tonga, and Niue
mumu, and which is probably the same in Mangareva amui. In Niue mumu
fails to observe the distinction between mu to buzz and mu to be a crowd.
Fotuna, Maori, Tahiti, Marquesas and Hawaii have preserved this mu, and
generally without duplication.

Viti mumu is the mu to be a crowd.

Motu has the same mu of sound, Aneityum ilmurilmu seems to be some
sort of a ww-composite, Mota mum is readily seen to be mu reduplicated
and then abraded.

Nicobar mumu a dove may quite safely be identified with this mu of
sound. But in the Malagasy moimoy, although the sense is satisfactory,
the vowel anomaly is too great to be overcome ; and the monomonona could
be associated only with muna.

The Semitic involves elements not present in our mu; in fact the only
point of form likeness is the m, and that is insufficient.

327-
muri na, the after part of a thing.

Samoa, Futuna : muli, the end, hind part, bottom, rump, to follow
after. Uvea: muli, the end, the last, to follow. Nukuoro:
muli, after, behind, in future; mule, slow. Nuguria: muri,
aft. Rapanui: muri, the stern; i muri, hi muri, afterward,

DATA AND NOTES. 385

behind. Maori: muri, the rear, hind part, afterward. Tahiti:
muri, behind, afterward. Rarotonga: muri, behind, in the
rear of, after. Mangareva: muri, behind, after. Paumotu:
muri, the rear, behind, after. Bukabuka : tainamuri, younger.
Fotuna: no-muri, to follow; muri, behind. Aniwa: wamuri,
behind, after. Tonga : mui, the tail (of a bird), the end, hind
part, to follow. Niue: mui, the last, to follow; muimui, the
stern, hind part; muli, to follow. Marquesas: mui, after, behind.
Viti: muria, to follow, to go behind.

Nggela: muri, behind; tumuri, to follow. Motu: murimuri, out-
side; muritai, the younger. Duke of York: murimuri, to
follow; murumuru, to be behind. Kabakada: muru, back;
mule, again, away. Baravon: muru, murmur, the last. New
Britain: mulumulu, to follow after. Pala: mur, id.
Matu: muli, to return. Pampangas: mulin, the stern. Macassar:
kamudi, rudder, helm. Malay: burit, the fundament; buritau,
the back, hind part, stern ; kamudi, rudder. Tagalog : huli, the
stern. Visayas : uling, id. Malagasy : vudi, the posteriors, stern.
Arabic: 'ah'h'ara, to be behind; mouh'ir, stern, hind part; 'eh'ir,
end.
Efate" muri is in exact accord with the Polynesian muli, muri, mui. The
excision of the / in Tonga and Niue is not a regular mutation in those lan-
guages, yet there are other instances. The fact that Niue has both muli
and mui leads to the recognition of an attempt, not yet completed when
the language was set by reduction to writing, to differentiate muli to follow
and mui last. In Bukabuka tainamuri we can recognize tei (47) younger
brother ; and in Motu the same elements in a different order, this being still
one more instance of the concord of that Torres Straits station with the
extreme east of Polynesia. Nukuoro mule slow is not a muli form ; it does
not appear in Samoa, a rare instance of the failure to find in the most
modern Samoan the source of Nukuoro vocables, but in Niue mule a long
time, the source is plain. Since we know Niue to have been under direct
obligations to Samoa we may regard Niue and Nukuoro as preserving a
word which in Samoa went into disuse before record was made of the
language.

In Melanesia the muri identifications are few and widely scattered. The
Duke of York muri and muru forms afford a transition phase by which we
are enabled to accept muru despite the vowel modification, and to localize
it in the eastern gateway and adjacent New Guinea. The Kabakada mule
has no such signification as would serve to establish its affinity with the
Nukuoro-Niue mule just discussed, or with muli.

In Indonesia we encounter great variety of consonant mutation with
absolute fixity of the u-i vowels. Pampangas mulin the stern is in form
and in sense a perfect identification, after noting the ephelkustic n of this
area. Matu muli to return is exact in form, but the sense is not so satis-
factory. For I we have mutations to r and d; the former so common as
not to need discussion, the latter found in Samoa and Sikayana of the
Polynesian group and existing in a few traces (/-/, l-nd, l-j) in Melanesia.
For m we have mutations to h, to v, to b, and to extinction. Not one of

386 THE POLYNESIAN WANDERINGS.

these occurs in Melanesia save m-v in the solitary instance of ma'i (323)
sick Nggela vahagi. In Polynesia none is represented, save as it may be
considered to exist in the rare m-p exhibited in Samoa tumua'i the crown
of the head Maori tumuaki and tupuaki Tahiti and Paumotu tupuaki.
Acceptance of these forms, therefore, should be postponed.

The Arabic mouh'ir is clearly a secondary form from stem 'hr, which
shows no kinship with any form in which viuli appears.

328.

namu, mamamami, the mosquito.

Samoa, Tonga, Futuna, Niue, Uvea, Sikayana, Nuguria, Nukuoro,
Mangaia, Paumotu: namu, the mosquito. Maori: namu, the
sandfly; namunamu, a blistering plaster. Marquesas: namu,
a reddish gnat. Tahiti : namu, ramu, the mosquito. Fotuna :
namo, id.

Viti: namu, the mosquito. Rotuma: rom, id.

Arag, Maewo, Duke of York, Mota, Nifilole, Fagani, Vaturanga,
Nggela, Guadalcanar: namu, mosquito. Savo: namunamu, id.
Marina: namugi, id. Lakon: namug, id. Bugotu, Buka:
gnamu, id. Motu, Raqa, Wango: namo, id. Mukawa:

namonamo, the fly. Galavi, Boniki: namonamo, the fly;

namokiri, mosquito. Kubiri: nanamo, mosquito. Kiviri:
nanam, id. Mekeo: anguma, the fly. Roro: aumaha, id.
Merlav, Mota, Gog, Lo, Vuras, Mosin: nam, mosquito. Sinau-
goro, Hula, Keapara, Galoma, Rubi : nemo, id . Dobu : nema, id.
Sariba: nimai, id. Kiriwina: nim, id. Norbarbar, Pak,
Motlav, Volow: nem, id. Sasar: nom, id. Malekula Uripiv :
num, id. Moanus: njam, id. Aneityum: inyum, id. Alo
Teqel: torn, id. Tangoan Santo: moke, id. Malo: mohe, id.
Panaieti: gumu, id. Tanna: kumug, id. Misima: kimu, id.
Taupota: himokini, id. Awalama: himoiodi, id. Nada:

simunika, id. Murua: sumoniku, id. Wedau: imokim, id.
Tavara: moeoti, id.

Malay, Bugi: namok, mosquito. Dyak: njamok, id. Macassar:
njamo, lamu, id. Kayan: hamok, id. Pampangas: yamuc, id.
Malagasy: muka, id.

Arabic: namus, mosquito.

It is very difficult to understand the second Efate form mamamami. It
does not appear in its proper place in the dictionary nor are any of its parts
identifiable in the material which Dr. Macdonald affords. It is quite pos-
sible that it would not have found entry under namu as an alternative had
not our author misread his Samoan and entered the first of his affinities as
"Sa. mamu." There is no such word in Samoa; the mosquito there, as
generally, is namu. The only variants in Polynesia are Fotuna namo, a
very slight change and possibly a fault of dulness of ear on the part of the
recorder in catching the light sound of the unaccented final syllable; and
Rotuma rom. In the latter the n-r mutation is by no means uncommon
in Polynesia, is well represented in our Melanesian material; and rom, in
fact, reappears as Tahiti ramu.

DATA AND NOTES. 387

In Motu and Wango we find a recurrence of the namo already noted in
Fotuna. Marina gi and Lakon g are recognized as noun-formative termi-
nations; the peculiar strengthening of the initial in Bugotu and Buka is
local to the Solomon Islands. With these notes we shall have no difficulty
in following through Melanesia a sequence of namu-nam-ncm-nom-num.
The Moanus njam, as in several other vocables noted from that first station
of the eastern gateway, may be assigned to Post- Polynesian influence from
Indonesia, noting particularly Dyak njamok and Macassar njamo ; still there
is Aneityum inyum of probably equivalent phonetic value and I can not
trace Dyak raids so far to the south. In Alo Teqel the n-t mutation is a
local idiosyncrasy.

In Indonesia we have no difficulty in identifying the Malay, Bugi, Dyak,
and Macassar forms, but beyond that we shall find complications. Kayan
hamok, Pampangas yamuc seem to be in accord with Tanna kumug. MaXo.-
gasymw&a,Tangoan Santo wofcf, and Malo mohe seem to form another group.
The relation between the two groups seems to depend upon the assumption
or loss of the prior syllable. Assuming a common stem in mok I have
searched the Pacific material for some variant of that stem used in designa-
tion of an insect or other flying animal, but with no success. While I have
twice grouped Melanesia with Indonesia in these obscurities it has been
solely on the score of form resemblance ; it will not have escaped notice that
these are New Hebridean languages found much farther south than I have
been ready to admit within the sphere of Dyak raiders.

Now what is the animal to which the namu designation pertains?

The mosquito and none other, say all our authorities save the Maori and
the Marquesans, to whom it represents the sandfly and the gnat respec-
tively. But is the mosquito indigenous to this wide area of the Malay seas
and the Pacific?

This is a question for the biologist with his story of the migration of
species. In certain parts of Polynesia the mosquito was long unknown.
Read the historical record for Hawaii as set down by Prof. William DeWitt
Alexander in "The Brief History of the Hawaiian People" page 195 : "Dur-
ing this year (1826) mosquitos, hitherto unknown in the islands, were intro-
duced at Lahaina by the ship Wellington from San Bias, Mexico." Seventy
years later on the neighboring island of Hawaii I found them spread from
the port of Hilo no greater distance than to Olaa, less than a score of miles.
In Samoa we may not execrate the ship that brought the pest, but we have
equally valid record of the introduction of this and yet another insect. In
the manuscript of my "Samoa o le Vavau, " awaiting its due season, I have
preserved the tale of the king's daughter of Manu'a who stood on the shore
gazing out into the east and into the face of her hero coming over the
unknown seas to greet her. "I'll come back and marry you," he said,
"and meanwhile keep my treasure safe against my return, but on no account
open it." He gave her a tube of two joints of bamboo with their inter-
vening septum and with the open ends of each cavity plugged. Pandora
of the South Sea, from one end flew a swarm of flies, from the other a mist
of mosquitos, and her hero never did come back.

It may be objected that these migrations could never have preserved the
name when thev no longer had reason to remember it in new lands whither

388 THE POLYNESIAN WANDERINGS.

the mosquito had not penetrated. Then how shall the student of the Viti
account for the possession there of karavau as a name for cattle in a land
which knew no horned beasts until the missionaries came ? It seems to be
the carabao of the Philippines. Or how account for ngeli, the Viti name
for monkey, when the nearest quadrumane is no nearer than Indonesia?

329-

nate, nase, ate, atse, the banana, the plantain (plant and fruit).

Samoa, Futuna, Niue: futi, the banana. Tonga: fuji, the generic
term for all bananas. Fotuna: vuji, the banana. Moiki,
Nuguria, Nukuoro : huti, id. Maori: cf. hutiwai, a plant name
(Acoena sanguisorba) . Aniwa: hutshi, the banana. Fila:
butsh .
Viti: vundi, banana, plantain.

The following words all signify banana :

Fagani : fuki. Vaturanga, Nggela, Bugotu : vundi. Saa, Bululaha,
Ulawa: huti. Wango: hugi. Maewo: undi. Sesake:

audi. Baravon: vundu. Duke of York: wundu. Kalil:
'huddu. Pala, Namatanai: hudu. Lambell: 'hun. Laur:
hun. King : wun. Lamassa : wun. New Ireland (Carteret
Harbor) : un. Epi, Bierian : vihi. Paama : ahisi. Malekula :
na-vis. Malekula Uripiv: na-vits. Malekula Pangkumu:
ne-vij. Arag: ihi. Malekula: abus. Eromanga: no-bos.
Moanus: mbur. Ambrym, Aneityum: nohos. Dobu: udi.
Kiriwina: usi. Sariba: udu.

Ceram, Ahtiago: phitim. Matabello: phiidi. Ceram: judi. Gah:
fudia. Tobo: fud. Massaratty: fiiati. Mayapo: fuat.
Malagasy: hutsi, untsi. Timor: hudi. Caimarian: u'ki.
Batak: unchi. Macassar: unti. Sambawa: punti. Panga-
sinan: ponti. Sanguir: busa. Malay: pisang.

Arabic: muz', banana. Amharic: muz, id.

In Polynesia there are many names for banana and plantain and futi is
the least usual; but its existence is well established, and Tonga fuji marks
it as ancient, for generic terms are imperfectly developed in the intellectual
plane of our islanders. Incidentally to the mention of Tonga fuji I must
note that Codrington (Melanesian Languages, 55) has cited it incorrectly as
fugi and treats the g as a k derivative ; thereby he has unfortunately gone
astray in so much of his discussion as is based thereon, for t-j is the common
Tongan mutation before e and i (17 Journal of the Polynesian Society, 212).

Before taking up the Melanesian forms I cite, on the authority of Mr.
Christian, these Micronesian affinities : Ponape 'ut; Kusaie, Mortlocks, 'tis;
Uluthi, ut; Palawat, Uleai, Lamotrek, Satawal, uis.

Nowhere in Melanesia do we find exactly the primal form futi. The
nearest approximation is huti of Saa, Bululaha, and Ulawa. The mutation
t-g is found only in this one instance in Wango, but t-k is not only well
represented in our Melanesian material but is a presently active principle
in Hawaii and Samoa ; therefore we may associate huti, hugi, and huki as

DATA AND NOTES. 389

futi forms. The vnndi of Vaturanga, Nggela, and Bugotu, the same as Viti,
is also futi. We have, therefore, a perfect identification of futi in the
Solomon Islands, a series of crop colonies on the Samoa track. Baravon
vundu is easily associable with vundi and therefore with futi, and Kalil
'huddu represents the same secondary stem with the d not reinforced. I
have already commented on the borrowing at second hand of a reinforced
consonant by speakers to whom a double consonant was objectionable and
who, in their ignorance of the primal form, have ignorantly retained the
reinforcement and have dropped the radical. Thus from vundu of Baravon
and wundu of the Duke of York we pass to the 'hiin of Lambell, the hun
of Laur, the wan of Lamassa, and the un of Carteret Harbor in the Lamassa
region of New Ireland. These form a second, a more remote, stage on the
Samoa track ; they lie within the eastern gateway.

We are next to examine a group of forms having a for the earlier vowel,
a for u. These have all lost the initial consonant. Our studies in the east-
ern gateway have shown us the vanishing of v in progressive stages. Here
we have no record of progress in the dilapidation, yet we may accept the
result and view this group as representing an original fati variant of futi.
We find this in Efate ate, Sesake audi; and Maewo undi may be taken as
retaining somewhat of the transition phase which links fati and futi. Look-
ing back at Efate we find dtse. This is no unusual /-variant. In Polynesia
it is regular in Niue and Tongarewa; in this stem we find a blunter form
of it in Aniwa hutshi and Fila butsh. Then by reduction of the doubled
consonant and selection of the wrong member we come to ase.

We next find a group in which primal futi has become fiti. We do not
find fiti itself any more than we found fati. But its immediately secondary
phase is found in Malekula Pangkumu ne-vij and Uripiv na-vits. From
na-vits we pass as before to na-vis, and thence it is a normal step in one
direction to villi which we find in Bierian of Epi, in the other to hisi which
is preserved in Paama ahisi. Through vihi we are led simply to Arag ihi.

Paama ahisi has shown us a futi derivative assuming a prefixed a. We
have already become acquainted with the change of / to s. Therefore we
need have no hesitation in accepting Malekula abus as a futi derivative;
still less in joining therewith Eromanga no-bos, for its no- is readily seen
to be the Melanesian n-article chameleon-colored by the nearest radical
vowel. Twice already have we seen the primal / develop into h; therefore
the no-hos of Ambrym and Aneityum takes its place as a futi derivative.
Moanus mbur is the only form which for the present seems irreducible.

We find somewhat parallel variations in Indonesia. Primal futi is repre-
sented immediately by Ceram fudi, Matabello phiidi, Gah fudia, Timor hudi,
Malagasy hutsi, Tobo fud, Caimarian tiki, Malagasy untsi, Batak unchi,
Macassar unti, Samba wa punti, Pangasinan ponti. Sanguir busa introduces
the ^ which we have twice already developed. Massaratty fuati, Mayapo
fuat may represent the transition phase of futi to fati. The fiti form is
seen in Ceram and Ahtiago phitim, and with the ^--change may pass into
Malay pisang.

We have passed in review many variants of the initial consonant of futi,
v, p, b, w, h, and its complete vanishing. Not one case suggests the m
which this Semitic requires to bring it into kinship.

390 THE POLYNESIAN WANDERINGS.

330.
niu, the coconut palm.

Samoa, Fakaafo, Tonga, Futuna, Niue, Tahiti, Hawaii, Rarotonga,
Manahiki, Tongarewa, Paumotu, Rapanui, Fotuna, Nukuoro,
Nuguria, Moiki: niu, coconut. Mangareva: niu, the coconut
palm when young, ripening into nikau. Maori : niu, a means
of divination by sticks; nikau, a palm. Marquesas: niu, a
top, a species of coconut. Mangaia: nu, the coconut palm.

Viti: niu, the coconut palm. Rotuma: niu, id.

Sesake, Epi, Arag, Ulawa, Wango, Fagani, Saa, Bululaha, Vaturanga,
Nggela, Bugotu, Motu, Kabadi, Pokau, Doura, Sinaugoro, Hula,
Keapara, Galoma, Mugula, Suau, Sariba, Tubetube, Panaieti,
Nada, Dobu, Port Moresby, Moanus, Solomon Islands, Lifu:
niu, the coconut. Misima: nihu, id. Buka: neu, id. Alite:
liu, id. Mukawa: niura, id. Awalama, Taupota : neura, id.
Kwagila: diura, id. Kubiri, Kiviri: rura, id. Nengone,

Nifilole: nu, id. Kiriwina: nuia, luia, id. New Caledonia:
ni, id.

Malagasy: nihu, the coconut. Dyak: nju, id. Salibabo: nyu, id.
Salayer: nydrah, id. Malay: nior, id. Liang: nier, id. Gah:
niula, id. Bicol: niyog, id. Cajeli, Mayapo, Massaratty,
Amblaw: niivi, id. Morella: niwil, id. Lariko: nimil, id.
Batumerah, Caimarian : niwcli, id. Awaiya : liweli, id. Teluti :
nuelo, id. Wahai: luen, id. Java: nu, id. Sulu: nui, id.
Ahtiago: nuim, id. Tobo: niia, id. Malay: nw, id.

Arabic : nah'lu, nah'ilu, palms (general and collective respectively) .

In Polynesia the only form variant is Mangaia nu, an unusual loss and
to be met with again only in Nengone and Nifilole and in a part of Indonesia.
In the Marquesas niu is not the usual name of the coconut, but it does
occur and is recorded by Bishop Dordillon. The sense of top lies in the
fact that the bud end of a coconut shell is used for spinning, both in the
sport of children and as a means of applying to island life the practical side
of the doctrine of chances. Thus it may be that in New Zealand, in lati-
tudes higher than are grateful to the coconut, the divination sense has per-
sisted even to different implements whereby the arbitrament of fate may
be declared.

In Melanesia there is no deviation from sense and very few instances of
form variation. Buka neu is a very slight variant. In Alite the n-l muta-
tion is normal; thus liu is readily identified. The occurrence of nu has
already been mentioned. Such internal loss is not regular in these lan-
guages. Furthermore the ni of New Caledonia leads us to infer that niu
was anciently a composite in which ni carried at least some sort of generic
sense, it being understood that this refers to those characteristics which
might strike the islanders as indicating a genus. In composition with kau
(301) tree we should then see nikau, the m'-tree, serving in Mangareva for
the coconut palm, in New Zealand for the characteristic palm (Areca sapida)
of that land, in Tahiti as niau for coconut leaf, and as niau in Hawaii for
the leaf stalk of the coconut. The ?^'-form is found in Micronesia, and in
the Marshall Islands ni is the coconut.

DATA AND NOTES. 391

In Indonesia there is much form diversity, but we may distinguish two
classes, those in ni and those in nu.

ni-class. The closest approximation to the classic niu is Salibabo nyu
and Dyak nju, and Malagasy nihu comes next. From nihu we are led to
infer that in the East Indies this u was scarcely felt to be a pure vowel;
this naturally admits w as the semi-vowel form of u. Thereby we draw
into accord Cajeli and development forms of niwi. We find now a group
which picks up a final liquid, Malay nior, Liang nier, Morella niwil; and a
new final vowel, Gah niula, Salayer nydrah, Batumerah and Caimarian
niwcli. Lariko nimil is so strangely remote from niwil as to avoid explana-
tion. Awaiya liweli shows the n-l change already noted in Alite of Mela-
nesia. Bicol niyog is the only instance of the assumption of a final palatal.

nu-class. This appears baldly in Java. Sulu nui is at least in form
metathetic and if that be a permissible explanation nu becomes a secondary
abrasion form. We have, however, no transition forms in Polynesia and
Melanesia to point the way to the nu-iorms in those areas. Tobo una is a
nui variant. The assumption of the final liquid converts nu into Malay
nur, and nui into Teluti nuelo, Wahai being a metathetic nuel of the 4231
type. Ahtiago nuim is left without explanation.

The Semitic here proposed conies into closest resemblance with Teluti
nuelo. This we have seen to be a secondary development of the nu-c\ass,
and that class we have seen to be anomalously removed from niu. Even
niu itself we have been willing to suggest as an advanced phase of a primal
ni. Thus the chances of Semitic affinity become very small indeed.

ori, to rub, to grate, to make a creaking grating noise (as tree branches
rubbing) ; ari, to plane, to scrape off, to rub off.

Samoa: olo, to rub down, to smooth, to grate. Tonga: olo, to
rub, to brush, to scrub, to sharpen, to grind. Futuna: olo,
olol, to rub, to grate. Niue: olo, to rub, to plane; oloi, to
grind; holo, to rub. Uvea: holoholo, to rub. Hawaii: olo,
to rub, to grate. Maori: oro, to grind, to sharpen on a stone.
Tahiti: oro, to rub, to grate. Mangareva: oro, to rub, to
whet, to sharpen; ororo, to rub, to chafe. Fotuna: no-wurusi,
to plane, to shave ; no-soroa, to grate.

Viti: solota, to rub, to grind, to grate; thoronga, a stone or coral
on which cane is grated.

Motu: oria, to grate coconut; uro, a grindstone. Norbarbar:

heri-v, to rub. Re tan: hara-v, id. Mota: sara-v, id.

Malay: urut, to rub. Malagasy: utra, rubbed; urina, being rubbed.

Arabic : 'arata, 'arat'a, to rub. Syriac : gra', to scrape off, to shave.
Hebrew: gara', id.

It is impossible to consider this as wholly distinct from the stems gathered
in item 205.

In that we have Samoa solo. From this the Samoan derives a transitive
verb by the addition of the usual verb-formative. In abundance of cases
in this assemblage of data we have seen that this verbal i has sufficed to
protect and to preserve the final consonant of Proto-Samoan closed roots.

392 THE POLYNESIAN WANDERINGS.

The fact that here when added to solo it gives us soloi is sufficient evidence
that in solo we have to do with an open root. The open root is unmistak-
able in Motu hnria and Aneityum ruhoi. Now as to Mota. Codrington dis-
misses the suffixes to verbs as certain consonants which make them transi-
tive, but he offers no explanation of the principle which governs the choice
of one consonant over another. Consistently I have found reason in these
studies to regard them as the persistent final consonants of closed roots.
Mota, then, yields the roots sarag, sarav, not proper kin of the open Poly-
nesian solo. Viti is again a problem with solota. In general I have found
it possible to treat these verbal terminations as indicative of root closure ;
therefore we may infer a root solot.

Taking up now item 331, we find Samoan olo, which we may faintly
discern to be Proto-Samoan holo. That would give us solo and holo stems,
differing slightly as would be expected in my theory of consonantal modu-
lants. In Fotuna no-soroa would affiliate with solo. The other word,
no-wurusi, we may link with Samoa fiifalu (220). The Tonga fulufuluhaa
points to a fulus stem, and this Fotuna is in immediate accord therewith.
Yet in item 220 we developed a furun-stem. Now Viti thoronga argues a
Proto-Samoan stem of either solong or holong, and the latter is appreciably
proximate to furun. The Mota .yara^-stem already noted we here find con-
tinued to Norbarbar and Re tan.

We have assembled the following stems : solo, holo, fulus, fidun, sarag, sarav.
It is impossible as yet and in the scantiness of material to separate them.

The Malay and Malagasy might be adjusted if we could clear away the
difficulties in the Pacific area.

The triliteron of the Semitic is gr% and this is not identifiable with any
of the stems here isolated.

332-
oro, to grunt, to growl, to snarl; oro-maki, to bark at; bioro, to make a
confused murmuring noise, as a crowd of men all speaking at
once ; orooro, uru, uruuru, to growl, to grumble, to mutter, to
murmur,
(a) Samoa : ngongolo, a rushing sound, as of wind, waves or thunder.
Tonga: ngolo, to snuffle, to speak through the nose; taengolo,
to cough with a rattling hoarse sound; kokolo, a continuous
rumbling noise. Futuna, Niue: tungolo, to snore. Maori,
ngoro, to snore, to utter exclamations of surprise or admiration.
Mangareva: ngoro, to snore, to rattle in the throat; ngongo:
the noise of phlegm in the throat; ngoio, to snore, to breathe
through the nose ; tongoro, the noise of water shaken in a bottle.
Rapanui: ngorongoro, to grunt, to grumble, to snore, to sleep
soundly. Paumotu: ngooro, tagoro, to snore. Hawaii: nono,
to snore, to gurgle; nonoo, to snore; nonolo, to breathe hard, to
snore, the sound of singing birds ; hoononolo, to chirp, to coo, to
growl, to grunt, to snort. Tahiti: ooro, to snore.
Viti: cf. nggolou, nggonggolou, to shout.
Mota : ngora, to grunt, to snort, to snore.

Malay : ngorok, to snore. Malagasy : ngorodona, the scuffling sound
of many feet ; gogogogo, screaming with pain.

DATA AND NOTES. 393

(6) Samoa: tangulu, to snore, to emit a hollow sound. Tonga: ngulu,
to make a grumbling, grunting, muttering noise ; ngungidu, to
groan, to roar; fakangungidu, to breathe hard; fengului, to
mutter or murmur to one another; tangulu, to snore. Futuna :
ngungulu, to grunt, to emit a deep tone, a grunt, a growl, the
sound of rumbling thunder; songulu, to snore; iiidu, whistling
of the wind, any great noise, rumbling of the bowels. Niue:
ngungulu, to growl. Maori: nguru, to sigh, to grunt, to utter
a suppressed groan, to rumble, to hum. Rarotonga : ngurunguru,
to grunt, -to cry out, to wail. Mangareva: ngunguru, a far-
off noise, a heavy noise as of many voices; ngurunguru, to
murmur, to grunt, to stammer, to speak through the teeth.
Paumotu: ngurunguru, to gasp, to moan, to sigh, to breathe.
Hawaii : nunulu, to grunt, to growl, to sound as the singing of
birds, to chirp, to warble. Tahiti: unru, to groan, to grunt.
Fotuna: tangurunguru, thunder.
Viti: kuru, to thunder; ngguru, to scranch; langguru, to make a

scranching sound.
Nggela: ngurunguru, to roar. Motu: urn, deep groaning, sterto-
rous breathing. Duke of York: pakpakuru, thunder.
Malay: guruh,guroh, thunder; ngaluh, to sigh.; nguru, to sigh, to groan,
to rumble ; kurkur, to grunt. Ilocan : gurruud, a thunderclap,
(c) Samoa : ngu, to growl, to make a murmuring noise as distant voices;
ngungu, pa'angungu, to scranch; mangungu, to be scranched, to
make a grinding noise as when walking over gravel ; jenguingui,
to talk in a low tone. Tonga: ngu, to grunt; fakangungu, to
grumble, to mutter. Futuna : nguu, to grunt, to groan. Niue :
ngu, to moan, to grunt, to roar; ngungu, to scranch; fengui, to
murmur. Mangaia: mangungu, thunder. Hawaii: nu, to
groan, to roar as wind, to grunt; nunu, a moaning, grunting,
groaning, a dove. Fotuna: noh-ngu, to grunt, to groan.
Rapanui: henguhengu, to murmur; henguingui, to read.
(d) Maori: ngara, to snarl.

Nggela: ngangaralia, to scream.

Macassar: ngangara, to shriek; gagara, to speak roughly.
(e) Maori: ngerengerc, to growl. Mangareva: ngere, a loud confused
noise.
Viti : ngcngele, to sing.
(/) Maori: ngeringeri, to growl. Samoa: ngingili, to have a good voice
for singing or speaking.
Nggela: ngingili, shouting.

Arabic: nah'ara, to grunt; h'ara, to low; h'arh'ara, snort, snore;
harra, to growl, to snarl ; harharat, murmur or sound of copiously
flowing water.
We have assembled here a number of stems which have a common ele-
ment in form, ng-l, and in sense are united by the fact that they designate
things audible, almost wholly vocal noises, and uniformly they deal with
the inarticulate sounds. With the ng-l forms I have associated ngu, which
seems to be yet more elemental and to represent in itself a common factor.

394 THE POLYNESIAN WANDERINGS.

The ng-l forms vary in the vowel used, as follows : ngolo, ngulu, ngala,
ngele, ngili.

It is noted that the special form tangulu has already been examined in
item 248 and will be here omitted.

ngolo. It is only in Samoa ngongolo that we find this stem to deal with
sounds other than animal; in this language the common ngolo sense is
carried by 'o'olo of the stem kolo which is readily associable with this group.
Samoa 'o'olo, according to Pratt in one of his naif definitions, is "to have a
voice like a hen, to speak indistinctly." Uvea kokolo is to murmur. But
Tonga kokolo passes from the human sound to that in exterior nature which
the Samoan expresses by ngolo; it signifies a continuous rumbling noise.
In Futuna kokolo is the grumbling of the bowels, echo, dull noise ; sound
of water when bubbling or flowing or falling ; the sough of the wind in the
leaves of the ironwood. If Samoan ngongolo deviates from the general sense
of an animal noise it comes back to the norm in its derivative fa'angongolo
to curse, to utter a malediction, and this seems to err as far in the other
direction, for such curses as come to most of us depend for their accuracy
of impact upon their clear articulation. We find the same suggestion of
articulation in the second sense of Maori ngoro. In Tonga taengolo the
former element is toe, an alternative of talc to cough. In Mangareva and
Hawaii the forms lacking the liquid we postpone for consideration with the
ngu-stem. The Viti nggolou is set here for comparison ; ngg is the equivalent
of Proto-Samoan k, not of ng; the sense is not exactly in agreement with
ngolo or kolo. Mota ngora, despite vowel change in the unaccented final
syllable, is clearly of this kin. Here too we note for inclusion Efat6 forms of
oro; oromaki suggests a stem closure in m which is not elsewhere identified.

ngulu. As we have seen ngolo paralleled with kolo, so this stem has
its k-iovm kuht. That is found in Futuna iiiilu and probably in Samoa
fa'a'ulu'ulu (which Pratt records as fa'auluulu) to cry out, to shout either
from pain or fear; in Tonga uulu the sound of anything in motion, and in
Viti kuru. The basic signification of ngulu and kulu is the deep, confused
inarticulate sound, principally animal but capable of extension to outside
nature. In Hawaii, the northern limit of migration, ngolo and ngulu have
lost enough of the gruff sense to admit of the inclusion of the note of birds,
inarticulate but light and cheerful. With this belongs Efate uru.

ngala. This stem is of infrequent occurrence, but we have one excellent
identification in each of the three island areas, and the Macassar gagara
proves that for it also there exists a fc-form.

ngele. This must not be dissociated from ngili. This stem we identify
in Polynesia, in Viti, and in Melanesia.

ngu. This involves as well the stem ngo. The latter is found in Ma-
ngareva ngongo and ngoio, in Hawaii nono and nonoo; these are in the sense
of snoring or the rattling of phlegm, senses quite cognate with those we
find for ngu. This covers all the dull and deep inarticulate sounds from
the snore to the peal of thunder. I regard this stem as that which primarily
contains the idea of inarticulate sound. The compounds with a Zo-form
do not express specific differences. I therefore regard them as determinant
compounds, each member having inter alia one common signification, the
compound emphasizing that sense and restricting the meaning to it.

DATA AND NOTES. 395

Of the Semitic here offered the stem is hr. Of our Polynesian we have
seen that ng is a sufficient stem. The Semitic is not in accord therewith,
and if that difficulty could be obviated it looks toward the later develop-
ment and not toward the primordial ngo.

333-
ra, tra, ta, nra, blood; mita, to bleed.

Samoa, Fakaafo, Nukuoro, Tonga, Futuna, Niue, Uvea, Aniwa,
Fotuna, Nuguria, Sikayana, Vate, Maori, Tahiti, Marquesas,
Rapanui, Mangareva, Rarotonga, Manahiki: totb, blood. Pau-
motu: putotoi, bloody. Bukabuka: kura toto, dark red.
Hawaii: koko, blood.
Viti: ndra, blood. Rotuma: tot, id.

The following signify blood :

Sesake : nda. Makura : na-dah. Motlav, Volow, Norbarbar, Gog,
Merlav, Vuras: ndar. Arag: daga. Lo: ndara. Lakon:
data. Mosin: nar. Mota: nara. Pak, Sasar, Alo Teqel :
tar. Tanna: neta. Nengone: dra. Lifu: mondra. Pala,
Eromanga: de. Laur: da. New Caledonia : 'nte. Maewo:
ndai, ndei. Omba : ndai, ndrai. Malo : dai. Marina : tsae.
Tangoan Santo : rat. Motu : rara. Sinaugoro, Nala : lala.

Java, Malagasy: ra. Salayer: rara. Malay: darah. Matabello,
Teor: larah. Saparua, Awaiya: lalah. Cajeli, Liang, Morella,
Lariko, Caimarian: lala. Batumerah, Gah: lalai. Teluti:
lata. Tobo: lawa. Mayapo, Massaratty: raha. Baju:
lahah. Ahtiago: lahim. Wahai: lasin. Menado, Sanguir :
daha. Bouton: orah.

Arabic: dam', blood; damiya, to bleed. Hebrew: dam, blood.
Ethiopic: dam, id. Syriac: dem, id.

The Melanesian stem is evidently ta. I find it impossible to bring this
into accord with the Polynesian toto. There is at no point a vestige of a
transition form from a to o. On this point I cite Mr. Codrington's note
(Melanesian Languages, 59) :

The Polynesian word for blood is toto, which does not appear in the Malay Archipelago
at all, and in the Melanesian Vocabulary only as tot in Rotuma, where it is probably
an importation. But toto is used in Melanesia, and in such a way as to show that it is
at home there and can not possibly be an importation from Polynesia. In San Cristoval
toto is congealed blood; in Florida the disease haematuria is mimi-toto; and though
it is not unreasonable to maintain that toto may have been borrowed in those parts,
it would be very hard to conjecture how it had been done. But toto in the Banks Islands
is a poisoned arrow, and this can be shown to be the same word. The arrow is called
after the tree with the viscous sap of which it has been smeared, and the tree has its
name from the abundance of its sap, in Mota totoai, in a dialect of Fiji dotoa. The sap
of a tree is its blood, and it is very easy to conceive a word at one time more general in
its meaning being particularized to signify in one set of languages blood and in another
sap. It assists this view to observe that toto in San Cristoval is clotted thick blood,
like the thick sap totoa.

In Efate we have ta and ra {tra and nra being m-variants) .
Viti ndra argues a parent in uvular r, but this does not appear in Poly-
nesia at all. A mere glance at the chart will show that in our Melanesian

396 THE POLYNESIAN WANDERINGS.

identifications of the ra-stem we have Viti at the hither end and Motu at the
yonder end and that the intervening area lies altogether from the northern
New Hebrides southward. The only exception to that is Laur da on the
New Ireland coast. Omitting Laur, these three points mark out the south-
ern migration, that to which I have assigned the designation of the Viti
stream. The Melanesian occurrences of toto as presented in the foregoing
note almost wholly lie along the Samoa stream.

Similarly when we prick the Indonesian identifications upon the chart
we find them to lie nearest the southern gateway.

The Semitic proposed stems in dm. The latter is not found at all in the
Melanesian; and the former is not the radical r, not even a regular r-d
mutation, but the resultant of a second loan acting in a manner that has
already been sufficiently explained. There is, therefore, no reason to accept
this as an affinity.

334-

ran, water.

Samoa : lanu, liquor amnii, to wash off salt water (ja'alanu) . Tonga :

lanu, fakalanu, to wash off salt water with fresh. Futuna :

lanu, liquor amnii, to rinse off salt water with fresh {fakalanu) .

Niue : fakalanu, to wash in fresh, after bathing in salt, water.

Mangareva: rami, saliva, globules of foam.
Viti: ndranu, fresh (of water), to wash in fresh water after having

bathed in salt.
Motu: rami, water. Galoma: nalu, id. Sinaugoro, Hula,

Keapara, Rubi: nanu, id. Tanna: rani, fresh (water).
Malagasy : rano, water ; lano, swimming. Java : ranu, water. Kawi :

rami, danu, id. Ilocan: danum, id. Malay: danau, the

ocean.
Arabic: rahalu, water (of a kind).

Here we have one of the rare instances in which a Nuclear Polynesian
vocable reappears in Mangareva, although the sense is highly modified.
The word has been carried along the Viti stream, and in Torres Straits we
have several stations.

In Nuclear Polynesia the sense is specialized, the waters of birth being
recorded in Samoa and Futuna, and in every case the word refers to the
rinsing off of sea water. To wash away the saltwater is not only for appear-
ance sake in the tropical Pacific. The sea has a high degree of salinity and
the salt crystals are not only the clearly visible signs of an unkempt habit
when they glisten upon the brown skins, but they are uncomfortable as well.
This use stamps upon lanu a sense in which the fact that it is fresh water
is more prominent than the water sense.

While the Kawi has both rami and danu the extension of the latter in
Malay danau to the ocean signification estops us from considering ranu
and danu as variants ; they seem more likely to be differentiated in mean-
ing and form.

The Arabic rahalu, rhl, presents no close resemblance to an rn stem such
as we have in our island areas.

DATA AND NOTES. 397

335-
rau, leaves (for food to be cooked, and for putting food on, as on a plate,
when cooked).

Samoa : lau, a leaf ; lalau, to be in leaf ; laulau, a food tray plaited
from a coconut leaf, to set out food on such a tray or on a
table. Tonga: lau, lou, a leaf; laulau, a tray. Futuna,
Uvea, Nuguria: lau, a leaf. Niue: lau, a leaf; laulau, a table.
Hawaii : lau, a leaf ; laulau, the netting in which food is carried.
Maori, Tahiti, Rarotonga, Rapanui, Paumotu, Nukuoro, Fotuna:
rau, a leaf . Manga re va: rau, rou, id. Marquesas: au, ou, id.

Viti: ndrau, a leaf; rau, thatch, coconut leaves shaken in the water
to drive fish into the net. Rotuma: rau, a leaf.

Sesake, Sinaugoro, Hula, Keapara, Galoma : lau, a leaf. Makura :
lau, aplant. Moanus : laun- (always composite), a leaf. Marina,
Arag, Vaturanga, Nggela, Motu, Waima: rau, id. Natalava:
tharau, id. Aneityum: in-rau, a covering, a preparing of
food for a feast; in-raurau, coconut leaves for a net. Malo,
Roro, Kabadi: rauna, a leaf. Wango: raua, id. Mukawa:
raurau, id. Vaturanga: rarau, id. Malekula: raun, id.
Lakon: drawi, id. Mota: naui, id. Maewo: ndoui. Merlav,
Vuras : ndoi. Lo : hoi. Vuras, Gog : ndo. Duke of York :
dono. Mosin : no. Mota : nau, no, id. Mekeo : ngangaunga, id.
Motlav: ron. Volow: raren, ro. Nengone: ru. Deni:
leu. Guadalcanar : talu. Omba : raugi. Norbarbar : ndugi.
Pak, Sasar, Alo Teqel : togi.

Savu: rau, a leaf. Java: rou, id. Malagasy: ravina, id. Saparua:
laun, id. Baliyon, Baju: daun, id. Malay: daun, dawun, id.
Sulu: dahun, id.

Arabic: hadab', branches, leaves; hadiba, hadaba, to pluck, to have
long eyelashes, to have long branches.

In Efate rau- does not have the simple leaf sense, but the meaning of leaf
tray is found in Samoa, Tonga and Niue of Nuclear Polynesia, and in
Hawaii. Here it is accompanied with the meaning of leaf, and that is the
only signification in the Tonga fiti languages. That Hawaii laulau, if not
the food mat, is at least different from leaf and has to do with food is to
be attributed to that direct A^oyage from Nuclear Polynesia for which we
have gathered so many pieces of evidence. That Futuna and Uvea, though
geographically in Nuclear Polynesia, have not the food-tray signification is
a point to be considered with other evidence as tending to show that the
settlement of these two islands in such close proximity was principally
Tongafiti.

The Samoan lalau by its preduplication shows lau to be considered a com-
posite, la-u. The latter member we may not presume to identify, but the
former has a suggestive resemblance to la, lala, a branch, also a botanical
term. We note in Tonga, Mangareva, and the Marquesas alternative forms
with the vowel o. The fact of two forms in Viti of different sense points
to two distinct sources. The ndrau leaf is derived from a stem rau having
the r grasseye ; rau is derived from rau of the lighter r. This suggests that

398 THE POLYNESIAN WANDERINGS.

in Proto-Samoan the two forms were distinct, a distinction which in the
later weakening of the liquid it has been found impossible to preserve.
Thus may we account for the occurrence of the food-tray sense in Efate
and Aneityum without the leaf signification. It will be observed that
Aneityum has the fishing sense found in Viti, which further is found in
Samoa lauloa a surround of coconut leaves, and in Tongarewa raurau driv-
ing fish with coconut leaves.

In Melanesia rau a leaf is found somewhat widely: Sesake, Motu, Marina,
Arag, Vaturanga, Nggela ; Omba raugi is the same with a formative suffix.
Makura is only slightly removed in signification. Natalava tharau is a raw-
composite, as is Wango raua. Vaturanga rarau is in form a parallel of
Samoa lalau. Moanus and Malekula lann and Malo rauna introduce the
Indonesian final n; Moanus we might ascribe to that influence ; Malo and
Malekula lie beyond its reach. This n, therefore, may represent a persistent
radical w-closure.

We are next to examine a group of forms with i'-final, all New Hebridean.
This i is a suffix to noun stems when they are used absolutely. Of the
a-series we have Lakon drawi preserving the r grasseye by a preface as in
Viti, Mota naui nau. Of the o-series we have Maewo ndoui; Merlav, Vuras,
ndoi; Gog, Vuras, ndo; Mosin, Mota, no. These two series introduce that
r-nd-n-d mutation which we find it so hard to comprehend. Pak to-gi is a
direct derivative from Vuras ndo. Still on the o-series we have Duke of
York dono and Motlav ton, both with the n. Volow ro, and Nengone ru,
Guadalcanar ta-lu (suggesting Natalava tha-lau) and Norbarbar ndu-gi
belong in this series with still a further vowel change. Deni leu is probably
a form of lau, but unique. A still different mutation, r-h, is needed to
account for Lo ho-i, also unique.

In Indonesia we find rau, rou, laun and daun, all occurring in Melanesia;
and Malagasy ravina is very close to Lakon drawi.

The Semitic here proposed is represented without variation by the
triliteral hdb, which can have no relation with our island composite la-u
or la-un.

336.

rongo, rong i, tongi (dongi), nrong, tong i, to hear, to obey, to feel, to know ;
marongo, matrongo, manrong, to be idle, to amuse oneself at
the expense of someone.
Samoa : longo, to hear, to feel, to report, a sound ; longona, to hear ;
langona, to understand, to feel, to perceive by the senses;
fa'alongo, to hear, to obey. Fakaafo: jakalongo, to hear.
Futuna: longo, a report, tidings, to perceive, to feel. Niue:
longona, to hear, to perceive by the senses, to understand, to
smell; langona, to hear; ongo, a report. Uvea : longo, to hear,
to perceive, to feel, to comprehend, a message, a report ; jakalongo,
to obey. Nukuoro: longo, to hear; longo mai, to obey. Raro-
tonga: akalongo, to hear. Hawaii: lono, to hear, a report;
hoolono, to listen, to regard. Nuguria: ulono, to hear. Maori:
rongo, to hear, to feel, to smell, to taste, to obey, tidings, report,
fame, sound, noise. Mangaia: rongo, a report; akarongo, to
listen. Mangareva: rongo, to hear, to listen, to understand,

DATA AND NOTES. 399

to know, to apprehend. Aniwa : jakarongo, to hear. Fotuna:
no-rongona, to hear. Paumotu: rongo, to hear; fakarongo, to
cause to believe. Rapanui: rongo, to hear, to believe, to
comprehend, to understand; hakarongo, attention, to adhere,
to hear, to comprehend, to understand. Tonga: ongo, to
hear, to feel; ongoongo, to report; fakaongo, to hearken, to
await commands; longona, to inform, to report. Marquesas:
ono, oko, to listen, to understand; hakaono, to listen.
Viti: rongo, to hear, to be heard, to sound, a report, tidings; vaka-

rorongo, to listen, to obey.
New Britain : logor, to hear, to report. Lambell : longor, to hear.
King: longor 6, id. Duke of York: longor o, id. Baravon:
walongore, id. Raluana: ivalongore,7valawalongor, id. Bierian:
mlongo, id. Mekeo: longo, to hear, to know. Tagula: ilongwe,
to hear; kau-lungwe, to know. Nguna, Arag, Sesake: rongo,
to hear. Maewo: rongo, to believe. Omba: rongo, to feel
any sensation. Vaturanga : rongo, to feel a sensation. Malo :
rongo, to hear. Nggela: rongorongo, news; vaovarongo, the
hearers; rongovia, to hear. Vaturanga: rongovia, rongomi, to
hear, to listen to. Mukawa : nonora, to hear, to know. Tavara,
Wedau: nonori, id. Awalama, Taupota: nonori, to hear.
Oiun: nowara, id. Raqa: anoara, id. Kabadi: oonova, id.
Roro: ona, id. Uni: abai-oa, id. Nggao: rongi, to under-
stand. Volow: rong, to hear. Sesake: ndongo, to smell.
Omba : ronghogosi, to listen ; rorongtagi, to hear a sound. Arag :
rorongtai, to listen to. Marina, Mota: rogotag, to hear. Gog:
rongtag, id. Ambrym, Retan: rongta, to hear, to feel. Motlav,
Vuras: rongteg, to hear. Mosin: rongte, id. Norbarbar:

rongte, to feel. Lo: rongte, to hear, to feel. Santo: ronoa,
to hear. Maewo: roro, roroi, rore, a report. Mota: roro-i,
a report ; ronga, famous ; rongo, to apprehend by the senses, to
feel, hear, smell, taste. Baki: mjongi, to hear.
Java: rungu, to hear. Matu: langan, sound, noise. Malay:
dangar, to hear. Malagasy: rohona, a sound, as of thunder.
Arabic: 'adina, to hear, to know, to feel the smell of, to cause to
hear, to make known. Hebrew: 'azan, he'zin, to hear, to
listen, to obey.
In the Polynesian at least three senses are entangled in the stem longo,
to hear, a thing heard, and to preserve silence. The latter, fa'alolongo,
falls easily into place, for it is a composite with the fa'a of comparison, to be
as one listening, therefore to be silent, "be silent that ye may hear." We
find at least two Polynesian closed stems, longon of the sense perception
and longos of the production of a sound that may be heard. The latter
seems to appear in the Viti rongotha but with the longon sense to hear, the
former in Tonga longona but with the longos sense to report. Beyond these
two inverted instances the closed stems have left no definite record.

In the examination of this material we shall confine ourselves to so
much thereof as finds identification in our Melanesian records, namely, the
matter of sense perception.

400 THE POLYNESIAN WANDERINGS.

In item 324 we have discussed the stem mata the eye and the sense of
sight. This longo seems to serve for the other sense. This we shall
examine in general and in particular.

(1) Sense perception in general: Samoa, Futuna, Niue, Uvea, Maori,

Mangareva, Tonga, Marquesas ; Efate, Omba, Vaturanga, Nggao,
Ambrym, Retan, Norbarbar, Lo, Mota.

(2) To hear: Samoa, Fakaafo, Niue, Uvea, Nukuoro, Rarotonga,
Hawaii, Maori, Mangaia, Mangareva, Aniwa, Fotuna, Paumotu,
Tonga, Marquesas, Viti; Efate, New Britain, Lambell, King, Duke
of York, Baravon, Raluana, Bierian, Nguna, Arag, Sesake, Malo,
Nggela, Vaturanga, Volow, Omba, Marina, Mota, Gog, Ambrym,
Retan, Motlav, Vuras, Mosin, Lo, Santo, Baki ; Java, Malay.

(3) To smell: Niue, Maori; Sesake, Mota.

(4) To taste: Maori; Mota.

(5) To feel: Maori; Mota.

The last three deserve a little closer inspection.

(3) To smell. Niue: longona, to smell; hongi, to smell. The latter is
external ; it is the sniff with which the odor-emanations are welcomed to the
olfactory nerve-endings ; the apperception rests in longona. Maori: rongo,
to smell; hongi, to smell; whakamono, to smell. Whakamono is to sniff, to
sniff up, to smell, and probably in reference to a nauseous odor, for mono-
mono means unpleasant to the smell. It also is external ; the apperception
rests in rongo. Mota: rongo, to smell; punpun, to smell; soman, to smell.
Punpun in sense is the Mota equivalent of Polynesian songi as found in the
Niue and Maori preceding, this shown in the fact that it means to sniff
in general and in particular designates the salutation roughly called rubbing
noses. Soman means to put the nose to. These two, therefore, are
external; the apperception rests in rongo.

(4) To taste. Maori: rongo, to taste; wharakai, to taste. We have
only the definition of the latter and are without information as to the
extent and manner of its use. Mota : rongo, to taste ; nam, namis, to taste.
The latter is physical ; it means to touch with the tongue ; it is external ;
the apperception rests in rongo. If rongo in Maori and Mota had not such
a catholic applicability to all sense perception except sight we should
find an interesting text in the fact that these savage people had arrived
at a comprehension of the intimate association of taste and smell which
has not become at all commonly familiar to persons of far wider range of
information.

(5) To feel. In the extreme brevity of our vocabularies no attempt has
been made to note the distinction between feel in its tactile sense and the
broader signification of perception in general. We are therefore not justi-
fied in attempting a differentiation not warranted by our authorities.

The following general conclusions seem well supported by the facts of
this record.

At a period but narrowly anterior to the Tongafiti or later migration the
psychology of the Polynesian recognized two sense perceptions, the sense
of sight and the other sense. When this other sense began to differentiate
in the direction of the knowledge of the sense perceptions which our own

DATA AND NOTES. 401

race has so long possessed that only recently they were classed as instincts,
it was the sense of hearing which first came up for particular designation.
I am somewhat in doubt as to this statement ; it is hard to comprehend the
psychology of conditions so different from our own. The same material
might equally bear the interpretation that the other sense was most
commonly understood as that of hearing.

In a psychology of which the sense perception is at no higher speciali-
zation than sight and the other sense it will not surprise us to discover that
the sense of touch has not yet been isolated from the merely mechanical
pressure of the skin upon an exterior object. I do not say that tactile
sense is as yet unrecognized, but I will let the vocabulary tell its own story
in a few of the better-known Polynesian languages.

Samoa: tangotango, to feel; it is used in the sense to take hold of,
therefore it means no more than to have in the hands or to have
the hands upon.

Tonga : ala, alafi, to feel ; its particular use is to touch, to feel after
with the hand.

Viti : yamotha, to feel; to feel for with the hand, to run the hand over.

Maori: whawha, to feel; to handle, to grasp.

Hawaii : haha, to feel ; to move the hand over a thing.

These are representative of the central and of the distal languages of
Polynesia and they represent varying culture planes. It is clear that in
not one of them does the sense of feeling exist independently of the physical
impact which constitutes the touching.

In the matter of form the Polynesian variations of longo are strictly
according to rule, except that in Samoa and Niue a longona form is intro-
duced, not found elsewhere until we reach Indonesia.

In the eastern gateway we find an extended form longoro, which may be
found again in Malay dangar. In the New Hebrides the termination which
varies from tagi to te is verb-formative. The Omba ronghogosi may best
be understood as a composite. In Nggela and Vaturanga we have rongovia,
and in Vaturanga rongomia, consonant forms which agree neither with the
forms in the eastern gateway nor with those found in Nuclear Polynesia.

In Indonesia Java rungu is a satisfactory identification ; the Malay is
far less sure. Matu langan is in good form accord with Nuclear Polynesian
langona, but the sense is not so clearly of kin. The Malagasy rohona inter-
jects a syllable for which we have no means of accounting.

The skeleton of the Semitic is 'dn, 'zn, with which it is impossible to
establish our Polynesian longo in any sort of accord.

337-
sa, to be bad, evil.

Samoa: isa, an expletive of disapproval. Tonga: sa, an expression
of disgust and disapproval ; isa, to hiss, to disapprove. Fu-
tuna: saa, indecent, improper; isa, exclamation of indignation.
Fotuna: sa, bad.
Viti: tha, bad; isa, interjection of disapprobation. Rotuma:
raksa, bad.

402 THE POLYNESIAN WANDERINGS.

Sesake, Lakon: sa, bad. Pak, Leon, Sasar: se, id. Alo Teqel:
see, id. Malo, Eromanga: sat, id. Norbarbar: set, id. Motlav:
het, id. Volow: heat, id. Aneityum: has, id. Merlav:
sasat, id. Mota: tatas, id. Maewo: seseta, id. Wango,
Alite: faa, id. Tanna: ra, id. vSavo: isarongo, id.
Malay: /a/ia/, bad. Malagasy: ratsi, id.
Arabic: ^a', sawat, to be bad, evil.
Although all our identifications of sa in Nuclear Polynesia are inter-
jections they seem quite satisfactory.

In Viti we find the word in two forms. It would appear that isa is a
more modern appropriation of the word, for exclamations in any language
travel far and travel true. This will not be doubted by such as have had
occasion to welcome as an old friend in the most distant surroundings our
own truly Athanasian contribution to the vocabulary of the world's objur-
gation. The Rotuma raksa, like so much in that odd speech, remains for
the present inexplicable, but it is safe to identify the sa.

In Melanesia we find that the Proto-Samoan stem was sat. The abraded
form is found in Efate, Sesake, Leon, Lakon, Pak, Sasar, Alo Teqel. Malo,
Eromanga, and Norbarbar retain the stem form. Motlav het is an easy
variant of Norbarbar set. Volow heat ishat with the vowel interpolation cus-
tomary in that language. Aneityum is easily deducible as hat with the t-s
mutation. Merlav sasat and Maewo seseta are preduplications of sat and
set. Mota tatas is therefore metathetic. Savo isarongo is probably an isa-
composite. Wango and Alite taa is hard to adjust ; the fact that in each
language an inner t is frequently excised might warrant us in suggesting
lata, but even though that has some resemblance to Mota tatas it fails to be
convincing. On the possibility of s-r mutation in Tanna ra see note 239.
Malay jahat will pass muster as a /m/composite. Malagasy ratsi can not
be accepted without more evidence.
The Semitic is certainly a resemblance.

338.
saki, to ascend, to go up ; sakesake, to be up, to sit upon.

Samoa: a'e, upward, to go up; sa'e, to elevate one leg, as in the
act of falling in a club match ; 'a'e, to ascend, to rise. Tonga :
hake, upward, to ascend. Futuna: ake, up, to ascend; sake,
to raise the leg at one in derision or mockery; kake, to climb,
to ascend. Niue: hake, up, to go up. Uvea: ake, up; hake,
to go up. Maori: ake, upward; kake, to climb, to ascend.
Marquesas : ake, on high, upward ; kake, to ascend. Mangareva :
ake, upward. Bukabuka: ake, up. Tahiti: ae, up, to go
up, to ascend, to climb. Hawaii: ae, to raise, to lift up, to
mount. Fotuna: no-jikijiake, to lift up; no-tukake, to stand
upright. Nukuoro: kake, to go up. Nuguria: kake, up;
hanage, northwest. Rapanui: kake a, to go aboard.
Viti: thake, upward; thaketa, to dig or lift up.

Sesake, Tangoan Santo, Nguna: sake, up. Malo: sace, up. Vatu-
ranga: sake, id.; sahelia, to go up into it. Mota, Merlav:

sage, up. Omba: hage, id. Nggela: hage, to go up ; hagelia,
hagevia, to embark. Bugotu: hage, up; hagelia, to go up on

DATA AND NOTES. 403

a ship. Belaga: hage, to enter; hagelia, to embark. Gog:
sag, up. Norbarbar: hag, id. Eromanga: sah, id. Volow:
ha, id. Saa: ta'e, landwards. Motu: dae, up. Wedau:
g'ae, id. Sinaugoro: rage, id. Keakalo: agi, id.

Malay: daki, up. Malagasy: akatra, ascended, gone up.

Hebrew: nasak, up, to go up. Aramaic: nsak, sak, id.

By combining various elements in the material here assembled it is seen
that the Proto-Samoan has three stems, ake, sake, kake. Regarding ake
as the primitive stem the others devolve naturally therefrom in variant
significations conditioned by the prefixation of consonantal modulants.
The primal ake is evidenced by its existence in Futuna along with sake,
and in Uvea with hake. Sake is restricted to the Proto-Samoan migration ;
it is by no means clear whether it was originally hake or sake, for in that
early stage there seems to have been little distinction between aspiration
and sibilant; the employment of hake in Uvea (which has both h and s)
inclines for so much weight as it may possess toward hake over sake. From
the fact that kake is confined to Nuclear Polynesia and the Tongafiti area
it is evident that it was carried by the later migration ; it appears nowhere
in Melanesia. In the Solomon Islands hagelia (Nggela, Bugotu, Vaturanga,
Belaga), and hagevia (Nggela) suggest closed stems sakel, sakev, of which
no further trace exists.

Our Melanesian forms all derive from the sake stem and very few call for
comment. Volow ha is doubtful for the reason that we have no record
of the dropping of g in that language ; while the form seems akin to Nor-
barbar hag, there is not a single instance here of a terminal abrasion from
an open stem to other than a consonant closure. Saa ta'e is readily brought
into line, for k is commonly elided in that speech, and the s-t mutation is
seen again in the next item; the sense is quite comprehensible, for every
landward direction in Malanta is upward, and Mota sage makes this inland
sense plain as defined by Codrington "to go inland towards the inner upper
part of the country," but generally a'e means to windward.

The Indonesian identifications are too slight to have any particular value.

One of the Semitic forms bears a close resemblance to sake, but this
amounts to little if we may look back to a primal sakel.

339-

sela, road, path, landing place of a canoe.

Nukuoro: sailenga, ala, a road, a path. Tonga, Niue: hala, id.
Samoa, Futuna, Uvea, Hawaii: ala, id. Nuguria: hara, id.
Maori, Tahiti, Mangaia, Rapanui, Mangareva : ara, id. Pau-
motu: ear a, id. Marquesas: aa-nui, the highway.

Viti: sala, a path, road. Rotuma: sala, id.

Mota, Malo: sala, road, path. Deni: hala, id. Saa: tola, id.
Taupota : talaka, id. Awalama : talaha, id. Tavara : taeaha, id.
Roro: tai-ara, id. Motu: ariara, a road through a village;
dara, a road through forest. Uni, Pokau : dala, road. Wango :
tara, a road. Belaga: thalautu, id. Bugotu: hatautu, id.
Wedau, Kubiri, Kiviri, Oiun: eta, id. Raqa: eta-fu, id. Suau,
Dobu: eda, id. Galavi, Boniki, Mukawa: keta, id. Sariba,

404 THE POLYNESIAN WANDERINGS.

Murua, Kiriwina: keda, id. Nada: keza, id. Sinaugoro,
Rubi: da, id. Wedau: tete, id. Nggela: halautu, id. Male-
kula: ne-sar, id. Aneityum: ne-falaig, id. Mota, Mosin:
matesala, id. Pak: me'esal, id. Merlav: metsal, id. Volow:
metehal, id. Motlav: na-mtehal, id. Sesake : matakisala, id.
matawirsal, id. Vuras: matekpwersal, id. Lo: luwomejal, id.
Mekeo: ke-anga, inengea, id.

Malay: aZafe, course, direction; .raZefe, a road; jaZaw, id. Silong:
ja/on, id. Java, Ilocan: cZaZaw, id. Malagasy: aleha, id.

Arabic: s'ala', to proceed, to call.

The Proto-Samoan stem is hala.

The Nukuoro sailenga is incomprehensible.

In Melanesia we find two groups of composites. That which occurs in
the New Hebrides is in the simplest forms mata and sala, the evolutions
of the former being here negligible. Sala varies to sal, hal, and jal. The
last form is in Lo luwome-jal, of which the former member is obscure. The
Solomon Islands composite is kala-utu, of which the succeeding member
has so far escaped identification. In this group we have hala, thala, and
hata, and on the l-t mutation see note 312. Saa and Wango involve the
h-t mutation noted in 338, the same applying to Motu dala and Indonesian
dalan and jalan. If Aneityum na-falaig be admissible we have no direct
evidence in support of the h-f mutation, but h-v once occurs, sina'a (169)
Sesake vinaga.

The Indonesian forms are satisfactory except for Malagasy with the
anomaly of an interpolated syllable.

The Arabic has a form resemblance, but the sense is not at all satisfactory.

34o.
seme, sama, the outrigger, or, more accurately, the part of the outrigger,
shaped exactly like a canoe, which floats in the water.

Futuna : ama, the outrigger ; saynatasi, a small canoe. Tonga : hama,
the outrigger, the smaller part of a double canoe ; fakahama, to
put an outrigger on ; ama, the port side of a canoe (the outrigger
side). Samoa, Tahiti, Marquesas, Mangareva, Tongarewa,
Nuguria, Paumotu: ama, the outrigger. Maori: ama, the
outrigger, in a double canoe the space between the two hulls.
Hawaii : ama, the longitudinal stick of the outrigger. Fotuna :
ti-ama, id.

Viti: thama, the outrigger. Rotuma: sama, id.

Mota, Tubetube, Brierly Island : sama, outrigger. Malo: isama, id.
Bierian: ni-hama, id. Baki: ni-ame, id. Yaraikana : tama, id.
Koko-Yimidir : darman, id. Nada: Jam, id. Tangoan Santo :
tsama, id. Tanna: timen, id. Malekula: ni-jam, id. Anei-
tyum: jmaing, id. Kiviri, Oiun: rama, id. Raqa: aurama, id.
Kubiri: ramani, id. Wedau: g'ama, id. Galavi: gamanaki,id.
Mukawa: g' amani, id. Mugula, Kiwai: sarima, id. Sariba:
sarime, id. Mabuiag: saima, id. Motu: darima, id. Pokau:
dalima, id. Hula : ralima, id. Kiriwina: lamila, iamila, id.
Awalama, Taupota: haruma, id.

DATA AND NOTES. 405

Malay: sampan, a small boat. Malagasy: sambu, a ship.
Arabic: safinat, safin', a ship, a vessel. Hebrew, Syriac: sefina, id.

The Proto-Samoan stem is hama, which appears in Polynesia as ama,
sama, hama, thama.

In Melanesia we find sama, hama and ame. The h-ts mutation in Tangoan
Santo appears in Polynesia itself. The h-t mutation in timen is not else-
where recorded from Tanna, but a glance at the table will show it to be
not unusual in other parts of the Melanesian area. The same is to be said
of Malekula jam, and this leads readily to Aneityum jmaing. The Torres
Straits sarima and devolution forms lie outside the possibility of cordial
acceptance.

The Indonesian identifications which Dr. Macdonald presents depend for
their force upon the actual employment of an outrigger. This is not the
common naval architecture of the sampan ; it is quite out of the question
in the Malagasy.

The value of the Semitic lies in the establishment of so small a part for
so great a whole. Even were that acceptable the Semitic has a resemblance
in but a single consonant.

34i-

sili, to enter; sili-fi a, to enter into (as a spirit or [demon into a man);

sili-faki, sili-fiki, to make to enter into, to thrust or throw into.

Samoa : sulu, to thrust into, to plunge into (as a canoe into waves),

to take refuge in; sulufa'i, fesulua'i, fesuluna'i, to take refuge

with; suluma'i, to thrust or push through or into. Tonga:

hulu, to plunge, to push, to rush under water as a canoe in a

heavy sea; hulumaki, to push in or through. Futuna: sulu,

to put or place within, to thrust into. Niue: huhulu, to

thrust, to push. Nukuoro: sulu, suru, to dive. Fotuna:

sum, to dive, to swim under water.

vSamoa: sili, to lodge in, to stick in (as a pen behind the ear);

silinga, a penholder, the place to stick in the fish hook; sisili,

to shoot, to dart (as pain from one part of the body to another) .

Tonga : hili, hilifaki, to lay or put upon ; hilinga, a number of

spears tied together, a platform on which things are laid. Niue :

hilifaki, to stick in.

Viti : thuruma, to enter; thurumaka, to push a thing into or through.

thiliva, to cut or lance the body.
Mota: saro, to go in; saromag, sarovag, to sheathe.
Malay: julok, to thrust into. Malagasy: juluka, to enter.
Arabic: dahala, duh'uV, to enter (a house), to take refuge with, to
thrust in.
In form our Polynesian material preserves two distinct series, that in u
and that in i, but the consonant skeleton is the same and the significations
interlace. We are justified, therefore, in dealing with them, at least pro-
visionally, as homogenetic. Both series are confined to Nuclear Polynesia.
For the sulu series we find Proto-Samoan stems suluf, sulum, sulun.
For the sili series we find a Proto-Samoan stem silif.
Viti has both series as derivatives of sulum and silif.

406 THE POLYNESIAN WANDERINGS.

In Melanesia Efate" is of the silif-stem ; Mota, with vowel changes, derives
from the two stems suluf and sulum.

In Indonesia the two identifications are clearly of sulu with its proper
final consonants abraded and passed from memory and then treated with
a formative termination of Indonesian habit.

The Semitic must account for additional radical matter before we can
accept the partial resemblance.

Where we have sulu and sili in parallel series we find it difficult to make
a plain separation of the shades of signification. The utmost upon which
we may venture is to note that the sense varies generally in relation to the
size of the thing which is thrust, sili denoting the smaller or finer-pointed,
this distinction being best seen in Samoa and Viti.

342.

sina, sine, to shine, to be clear.

Sina, hina, ina, is used throughout Polynesia in the senses of
glistening, white, gray.

Samoa, Futuna, Fakaafo, Sesake : masina, the moon. Tonga, Niue,
Uvea, Hawaii, Maori, Tahiti, Rapanui, Marquesas: mahina, id.
Mangareva: maina, id. Nuguria: masina, mahina, id.

Viti: singa, the sun, day; thina, a torch, a lamp; singasingau, white.

Fagani: sina, the sun. Misima: sinasina, id. Nada: silasila, id.
Misima: hilahila, id. Dobu: sinara, id. Wango: sina, id.;
haasinaria, to dry in the sun. Nggela: hina, to shine (sun);
hinari, to burn. Belaga: thina, sunshine. Mota: singa, to
shine; singasingai, a shining; singesinge loa, sunshine (loa, the
sun). Lambell: singsinge, to dry (presumably by putting
out in the sunlight, as in Wango). King: sengesenga, id.
Lamassa: sdngsdng, id. Merlav: sing, to shine ; singsinge aloa,
sunshine. Gog: sing, to shine; singsinge walo, sunshine.
Aneityum: gesega, the sun. Doura, Motu: dina, daylight,
the sun. Uni: Una, the sun. Mekeo: hina, id. Uni: dia, id.

Malay: siyang, day, clear; sinar, a ray of light, luster; ber-sinar,
to shine; sinar-hulan, moonlight. Ilocan: sinamar, splen-
dor. Magindano : sinang, sun, midday. Java : rahina, day.
Sulu : jasina, the moon.

Arabic: saha', sun, day, to shine, to be clear. Ethiopic: sahawa, id.
Hebrew: sahah, id.

It is matter of no slight interest to find that a stem which in Polynesia
serves to designate the lesser luminary is used in Melanesia to denote the
sun. In this connection our linguistic material has left two records. One
that la, the general Polynesian word for the sun, was not carried in the
Proto-Samoan migration, for it has left no trace in the Melanesian halting-
places. The other that masina, the general Polynesian word for the moon,
was brought into Polynesian, in its present derivative form, by the Tonga-
fiti migration, for it is only in Sesake that we find masina as moon. Our
Polynesian records show us that sina was a sun name, i. e., the shiner. At
the same migration era the moon was also a shiner, see pupula 284. The
Tongafiti now comes to Nuclear Polynesia bringing la sun and masina moon,

DATA AND NOTES. 407

and the Proto-Samoan early colonists finding their earlier names not suffi-
ciently differentiated, for the root sense of shining is equally in use, adopt
the later forms. It is only in Niue that we find a persistent relic of the
older moon name, pupula the new moon. But Viti, untouched by the
Tongafiti swarm, preserves singa sun and vula moon.

In Melanesia sina designates the sun, but there is not lacking evidence
that its root signification of shining is close to the surface. Thus in Nggela,
Mota, Merlav, and Gog hina does not denote the sun as a name but is used
of its shining, and in three of these instances the Melanesian word for sun,
alo, walo, aloa is subjoined.

In Indonesia we find sina in the root sense of shining. Malay siyang is
not properly to be included, but sinar and its composites are unmistakable.
The Magindano sinang is the only instance outside of Melanesia in which
we encounter the sun sense. Java rahina seems to be a composite of ra
sun and sina to shine.

The Semitic is, as so often in Dr. Macdonald's material, a one-letter
resemblance and valueless.

343-
sinu, sislnu, to be hot, to burn (of the grass on the hills yearly) ; slnu, the
burning of the grass, a place on which the grass has been burned ;
tuni, to heat; tunu, to heat, to oppress or make to suffer;
bitunu, bitin, bitsin, to be hot, painful; tutun, to light up
torches or the evening cooking fires.
(a)Futuna: masinu, the smell of burning fat (feathers, hair).
Hawaii: ohinu, to roast meat; ohinuhinu, to be parched and
dried, as the outside of roast meat.
Viti : cf. sinunganga, a tree whose sap has a burning effect on the skin.
(b) Samoa: susunu, to burn up; sununga, a burning; sunusunu, the
burnt bush where a plantation is made; masunu, to singe (as
the hairs of a pig) . Futuna : sunusunu, to singe a bird ; susunu,
to burn, to broil, to put on the fire; masunu, burnt by the sun,
the odor of burning hair or feathers. Tonga: huhunu, to
singe, to sear; hunuhunu, to toast, to singe, to broil; mahunu,
blighted. Niue : huhunu, to burn ; mahunu, to blister with heat.
Uvea : mahunu, a burning. Maori : hunu, to char, to singe.
Viti : kuro-susunu, the name of a disease, an epidemic,
(c) Samoa: tunu, to roast, to toast, to broil, to fry, to boil. Tonga:
tunu, to toast, to broil. Futuna: tunu, to broil, to roast, to
toast, to boil. Niue: tunu, to broil, to boil, to roast. Maori:
tunu, to roast, to broil, to burn. Tahiti: tunu, to roast, to
boil. Rarotonga: tunu, to broil, to cook on embers. Mar-
quesas, Rapanui, Nuguria, Nukuoro: tunu, to cook. Hawaii:
kunu, to roast on embers. Uvea: tutu, to toast. Fotuna:
tuna, to roast, to cook.
(d) Samoa: tutu, tungia, to set fire to, to kindle, to light (as a
lamp); tunga'i, to light up a fire at night. Tonga: tutu, to
burn, to set on fire; tungaki, to light a beacon; tungia, to set
on fire. Futuna : tutu, to light, to burn, to set fire to ; tungaki,
to light a beacon; tungia, to kindle, to set fire to. Niue:

408 THE POLYNESIAN WANDERINGS.

tungi, to light a fire or a lamp, to burn. Maori, Rarotonga:
tungi, to kindle, to set on fire. Hawaii: kuni, id. Uvea:
tutu, to set on fire. Tahiti: tutui, id. Rapanui: tutu, to
kindle, to set on fire; tutunga, combustion.

Viti: tunu, warm; vakatunu, to warm up cold food; tutuvaka, to
light, to set on fire; tungiva, to kindle. Rotuma: sunu, hot;
fuf, to light, to kindle.

Buka: sinanga, warm. King: suniin, fire. Baur: songsong, hot.
King: misongot, id. Epi: pisusunu, id. Malo: tunu, to
roast, to heat. Motu: tunua, to bake pottery. Marina:
tutunu, hot. Gog, Lakon, Mota, Santo, Vuras, Mosin : tutun, id.
Belaga: tutun, to burn. Duke of York, New Britain: tun, id.
Pala: tun, to cook. Lo: tun, hot. Mota Maligo: tin, to
roast over embers ; Veverau : tun, id. Sasar, Alo Teqel : i'in, id.
Lambell, Lamassa : tuni, to cook. King: itutiin, id. Mota:
tutunsag, fever. Ulawa : tunga, fire. Bugotu : totha, to light
(a lamp).

Malay: tunu, to burn, to consume with fire. Macassar: tunu, to
bake, to roast. Savu : tunu, to roast. Bugi, Landa, Binue,
Baliyon: tunu, to burn. Sambawa: tunang, id. Sassac:
tulu, id. Lampong: tunkan, the hearth. Kandayan: tinu,
to burn. Malagasy: tono, to roast.

Arabic : sal/ana, suh'un', to be hot. Hebrew : s'ahan, id. Syriac :
s'hen, id.

Our Efate material groups several consimilars exhibiting both vowel and
consonant distinctions. These reduce to the four primitives sinu, sunu,
tunu, tung. It is not difficult to trace these to a common source particu-
larized by modulants. We shall examine these in detail.

sinu. This is the least frequent. It occurs in Efate, Viti, enters Nuclear
Polynesia by way of Futuna, and is found again in Hawaii, a landmark at
that end of the direct migration to that outlier which was reached only in
a roundabout circuit by its main, or Tongafiti, peopling. We lack Motu
data, but, so far as Efate-Viti-Futuna may determine the course, we are
disposed to charge this to the Viti stream. The Viti sinunganga is a com-
posite readily resolvable into this sinu hot and ngdngd which is variously
defined as poisonous, bitter, sour, salt, in other words highly objectionable.
In Futuna masinu is of the common composition with wa-conditional ; it is
a synonym of part of the meaning of masunu. Hawaii ohinu is influenced
by the local idiosyncrasy which prefixes to many verbs this distinctive o.
The only forms in i noted elsewhere in our material are the not wholly
satisfactory Buka sinanga and Kandayan tinu.

sunu. This also is a Nuclear Polynesian form, but as sinu pointed the
way along the direct northern migration this points out the southern migra-
tion direct from Nuclear Polynesia to New Zealand. We have seen reason
to credit sinu to the Viti stream; similarly we find reason to hold sunu to
have been a particular form of the Samoa stream. Sunu does not appear
in Efate or in Viti, but its presence in Rotuma is significant by reason of
the fact that this island lies northward from Viti. In Melanesia (except in

DATA AND NOTES. 409

the single instance of Epi pisusunu) it is found only in King and Laur in the
eastern gateway, and Buka sinanga may turn out to be a sunu form when
we have more abundant data from the northern Solomons. In Nuclear
Polynesia we have generally sunu and masunu with no great diversity; the
only wide division in apparent signification is Tonga mahunu, and as to
this the Baker vocabulary is too terse to enable us to draw any conclusion.
The Viti kuro-susunu (kuro being a pot) is of uncertain application.

tunu. This is the form common to the Proto-Samoan and Tongafiti migra-
tions ; it is found in Viti and Polynesia generally, largely in Melanesia, and
it has a good representation in Indonesia. Sinu and sunu have shown
senses which imply contact with the naked fire, and that idea is largely
predominant in tunu, as exemplified by the definitions of toasting, roasting,
broiling, cooking on embers. Disregarding the instances in which the word
is rendered by our general verb to cook we shall examine the exceptions
to this naked flame sense. It is used of boiling in Samoa, Futuna, Niue,
and Tahiti. It is significant that not one of these peoples had taken so
much as the first step in fictile art, and such heating of water as was needed
was performed by dropping hot stones into the water in a wooden bowl.
Equally significant it is that in Viti, where pottery had advanced to the
possession of luster and glaze, the word does not mean to boil (Viti : kere,
kerea, to boil; wevue, ndanda weruweru, to boil to pieces). I have no hesi-
tation, therefore, in ascribing this signification to the careless lack of pre-
cision of European influence. In Samoa alone is it used of frying and here
there can be no doubt whatever; the infamy of the frying pan is distinctly
European, the islanders had not that suicidal implement for the assassi-
nation of their digestion. In Melanesia the course of tunu may easily be
followed. The Mota Maligo tin suffices to account for an unrecorded pre-
duplication titin, and from this by excision of t, normal to those Vanua Lava
dialects, Sasar and Alo Teqel obtain i'in. The Indonesian forms are all
homogenetic with tunu.

tung. This also is common to the Proto-Samoan and Tongafiti migrations.
It also shares the sense of the open flame but with this difference : the other
forms have dealt with objects brought to the flame, this in its senses of
lighting and kindling deals with the flame brought to some object, exactly
paralleled by our English locution of setting fire to an object. The dis-
criminating final consonant is carried over into the languages of the Tonga-
fiti second swarming out of Samoa more distinctly than is commonly the
case. Rotuma fuf is characteristic ; Proto-Samoan t is commonly changed
to / in that speech; therefore fuf is tut, a simple abrasion of Samoa tutu.
Viti has the word in two forms, tungi and tutu, and has applied to each its
own formative terminations, these two forms representing the earlier epoch
when Proto-Samoan retained its closed stem and the later in which it
had undergone abrasion. This difference of development level also appears
in Ulawa and Bugotu of the Solomon Islands.

The Semitic has the shn skeleton, and that does not accord with any of
these island forms. Furthermore the meaning amounts to no more than
heat, whereas the stems here assembled are explicit in their insistence, not
upon warmth but upon the open flame.

410 THE POLYNESIAN WANDERINGS.

344-

susu, the breast, to suck the breast; bakasusu, to suckle.

Samoa: susu, the breast; fa'asusu, to suckle. Futuna: susu, to
suck; u, the breast. Tonga: huhu, the breast, to suck the
breast; fakahuhu, to suckle. Niue: huhu, the breast, nipple;
fakahuhu, to suckle. Uvea : huhu, the breast, the nipple, to give
milk, to suck. Nuguria: huhu, milk; u, the breast. Tahiti,
Rarotonga : u, the breast, milk. Maori, Hawaii, Marquesas,
Mangareva, Paumotu: u, the breast. Fotuna: ko-u, to suck ;
vai-u, milk; ta-u, nipple. Rapanui: u, the breast; vaihu, milk.

Viti : suthu, the breast, to suck the breast, to be born.

Mota, Lamassa, New Ireland (Port Praslin), Mugula, Dobu, Galavi,
Boniki, Mukawa, Kubiri, Kiviri: susu, the breast. Laur:
susu, id. Mabuiag: sus(u), id. Motu, Suau: susu, milk.
Aneityum : nasusu, an infant. New Ireland (Carteret Harbor) :
susun, the breast. Brumer Island : susuga, id. Buka : ssussuge,
the nipple. New Ireland (Duffield) : susung, kasiing, the breast.
Lambell, Pala, King: sus, id. Mota Veverau: sus, id.; Maligo:
sis, id. Nada: sese, id. Buka: mata na sus, the nipple.
Buka: suss, the breast. Waima: tutu, id. Kiriwina, Raqa :
nunu, id. Oiun: mini, id. Taupota: g'ug'u, id. Wedau:
gugu, id. New Britain: n, id. Malo: susu, the breast, to
suck. Baki: yu, milk, nipple. Malekula: mi sus, to suck;
na sus, milk; susin, the breast. Bierian: ohun, the breast.

Wango: haasusu, to lay (of a hen). Mota: mm,*-, to give birth.
Nggela: vahu, to be born. Norbarbar: visis, to give birth.
Retan: vusus, id. King: vusiis, id. Lambell: 'hiasus, id.
Lamassa: ia.yj<.y, id.

Malay: mra, the breast, milk. Java, Bugi, Pampangas, Macassar:
susu, the breast. Tagalog: suso, id. Matu: susan, id.
Visayas: dughan, id. Kayan: twofe, the breast; .yo, milk.

Waigiou : sus, the breast. Waigiou Alfuros : su, id. Malagasy :
nunu, id.

Hebrew: s'od, the breast. Arabic: t'idy', t'udiyy', id.

The data here collated will be found subject to a double complication.
In Viti we find suthu meaning the breast, to suck the breast, to be born.
If it were solitary we might strain analogies and strive to bring these two
into correlation, but we need go no further than Wango haasusu to find
reason to give over the effort, for it is only in botany that ornithogalum can
be conceived of as even a remote possibility. It will best be disposed of by
regarding susu of parturition as a distinctively Melanesian element which
chances to have the same form as the Proto-Samoan susu the breast. In
a separate paragraph I have set aside the Melanesian homogenetics of this
non- Polynesian susu. It will be seen that in general it appears in com-
position with the causative haa-va, and in Lambell and Lamassa it will be
observed that this has greatly disintegrated.

In the sense of the breast susu is distinctly a Proto-Samoan possession,
primarily the breast and secondarily to suck the breast; for in Efat6, Samoa,

DATA AND NOTES. 411

Tonga, and Niue we find a causative to convey the latter signification.
This form runs through Melanesia, and greatly to our surprise it prevails in
Indonesia, where in general the agreement is more close with the Tongafiti
stage of the Polynesian.

The Tongafiti word for breast is u. This we may not regard as a dilapi-
dated form of susu, su-hu-u, for in that case we should expect the aspiration
to be present in the languages in which it entails no difficulty. The only
way in which I can adjust it to the susu series is to regard u as primal and
su to have become a particularized form through the employment of a
consonantal coefficient. In Nuclear Polynesia this w-form appears only in
Futuna, a language in which Tongafiti influence is strongly marked — not
so strongly in this instance, however, as to obliterate the Proto-Samoan
susu to suck. At a distance, and in the western verge, u appears in Fotuna
in the sense of sucking, which the word nowhere else carries, for in exterior
Polynesia the words meaning to suck have no relation to the word for breast.
New Britain u is of the Tongafiti type, and in Epi the Bierian ohun sug-
gests a su form which undergoes yet further demolition in Baki yu.

The Indonesian identifications, as before remarked, are quite generally
of the susu or Proto-Samoan type. The Malagasy nunu is out of order,
for we have no data tending to establish an s-n mutation.

The Semitic proposed is sd, td, in structure a resemblance of but one
consonant and without attempting to account for the other, therefore not
to be accepted.

345-
ta i, tata, to chop, to cut, to speak or utter (as it were making a chopping
noise).

Samoa : to", to strike, to beat ; tata, to speak rapidly, to break fire-
wood (by hitting a tree or a stone with the billet to be broken) .
Tonga: to, to strike, to beat, to hew; tai, to strike, to beat.
Futuna : ta, to strike, to beat, to cut, to hew. Niue : ta, to
strike, to kill, to adze. Uvea: ta, to strike, to cut. Maori:
ta, to beat, to cleave, to split up. Tahiti: ta, to strike, to
repeat or tell a tale. Rarotonga, Marquesas: ta, to strike,
to beat, to kill. Mangareva: ta, to cut. Paumotu: ta, to
strike, to cut (in composition). Nuguria: taa, to hew. Sika-
yana, Moiki: ta, to strike. Nukuoro: ta, to cut. Hawaii:
ka, to strike, to beat, to cut or split wood.

Viti : ta, tata, to chop or cut with an axe.

New Britain : ta, to strike. Moanus : ta, to beat, to strike. Male-
kula : tai, to cut.

Malay: tatah, to chop. Macassar: tatta, to fell, to cut off. Mala-
gasy: ta, the sound of beating or knocking; tatana, to chop.

Arabic : hadda, to cut quickly, to utter speech quickly.

With this stem we have an irregular associate referring to speech. Efate-
ta to speak, to utter; Samoa tata to speak rapidly; Tahiti ta to tell a tale.
Dr. Macdonald's parenthetical suggestion "as it were making a chopping
noise" is far too childish for consideration. We must regard this as the
survival of a to-vocable elsewhere forgotten.

412 THE POLYNESIAN WANDERINGS.

The root ta I have already (27 American Journal of Philology, 383, 386)
treated by dissection to the seed and consonantal coefficient, and have
developed the suggestion that ' ' the root ta through its long series of known
combinations carries a strongly featured sense of action that is peripheral,
centrifugal, and there seems to be at least a suspicion of the further conno-
tation that the action is exerted downward." Thus in most of our Poly-
nesian occurrences of the simple root and in two out of the three Melanesian
identifications we find the primal sense to strike. The secondary sense of
cutting will easily be seen to be a striking with a specialized implement,
and we find this sense stated without recognition of the primal striking
only in Mangareva, Nukuoro, Viti, and Malekula.

In Indonesia this secondary sense is predominant, although Malagasy ta
may come somewhat close to the striking idea.

The Semitic is far too complex, hdd, to stand in kinship with the sim-
plicity of our ta.

346.

tangi, to wail, to cry, to ring, to sing, to clank, to hum ; tangi-si, to bewail.

Samoa: tangi, to chirp, to bellow, to roar, to wail, to lament, to
weep, to chant a poem, to complain ; tatangi, to tinkle, to ring.
Tonga: tangi, to cry, to weep, to call out; tatangi, to ring, to
tinkle; taangi, to chant a poem. Futuna: tangi, to weep,
to groan, the cry of any animal, the sound of any implement.
Niue : tangi, to weep, to cry. Uvea : tangi, to weep, to shed
tears, to howl, the cry of animals, to appeal to. Maori : tangi,
to cry, to sound, to coo, to roar, to weep, to wail; tatangi, to
rattle, to jingle; takitaki, to recite a song. Mangaia: tangi,
to sound, to cry, to sing, to wail, to weep, to lament. Manga-
reva : tangi, to mourn, to wail, to weep, to sing, to make a noise,
the sound of a bell or of a trumpet. Rapanui : tangi, to cry,
to howl, to groan, to weep, to pity, to covet. Paumotu:
tangi, tatangi, to weep, to bewail, to lament. Tongarewa,
Nukuoro, Manahiki, Fakaafo: tangi, to cry. Aniwa: noko-
tangi, id. Vate: tetangi, id. Nuguria: hakatani, to bewail.
Marquesas: tani, taki, to sound, to sing, to bark. Tahiti:
tai, to sound (as an instrument), to weep, to cry; tatai, to
rehearse or recapitulate the particulars of an agreement.
Hawaii: kani, to make a sound more or less musical, to hum
a tune, to sound (as a trumpet), to make a report (as a gun),
to crack (as a whip), to rumble (as thunder), to squeak, to
crow; kanikani, to tinkle.

Viti: tangi, to cry, to weep aloud, to lament, to crow, to sing.

Nguna: tangi, to weep. Omba: tangi, to cry; tangisi, to cry for
someone. Nggela: tangi, tangihia, to lament; tangitangi, to
sing (of birds) . Vaturanga : tangi, to weep ; tangisia, to weep
for. Bugotu : tangi, to weep ; tangihia, to weep for. Nggao :
tangi, to weep. Matupit: tangi, to cry. Baravon: tangi,
to weep. Belaga: tangi, sounding (of a bell). Lambell:
tangis, a tear. Maewo: tangtangisa, pitiful; tangtangisi, to

DATA AND NOTES. 413

feel pity for, to show mercy to. Mota : tangi, tangis, tangtang,
to cry, to weep. Vuras: tengteng, id. Motlav: teng, id.
Sinaugoro : tagi, to weep, to cry. Motu : tai, to cry, to howl,
to weep. Redscar Bay: tai, to cry. Aneityum: taing, to
cry, to weep ; ataingataing, to sing (as a kettle before boiling) .
Santo: toni, to cry. Malekula: teng, to cry; tengsi, to bewail.
Malo: tangtange, to cry; tangisia, to bewail. Baki: jengi, to
weep, to bewail. Tangoan Santo : tangi, to cry; tangsia, to
bewail. Pokau : kani, to weep, to cry. Uni : cani, id. Hula,
Keapara, Galoma : agi, id. Roro : hai, id. Waima : hat, to cry.

Malay: tangis, to weep. Siwa: tangi, to weep. Tagalog: taghoy,
to weep, to lament. Java : nangis, to weep. Kayan: nangi, id.
Ilocan: sangit, to cry, to weep floods of tears. Malagasy:
tany, a cry, a lamentation.

Arabic: tanna, to tinkle, to clank, to ring, to hum; tanien, din;
tantun, to tinkle, to jingle.

The Proto-Samoan stem is tangis.

The significations are polyphase; therefore the time will not be ill
employed in segregating them into such classes as may be found to exist.

A. Non-vocal. Samoa, Tonga, Futuna, Maori, Mangareva, Tahiti, Hawaii,

Efate, Belaga, Aneityum.

B. Vocal.

a. The animal cry: Samoa, Futuna, Uvea, Maori, Marquesas, Hawaii,

Viti, Nggela.

b. The human voice:

i. Sounds of pleasure: Samoa, Tonga, Maori, Mangaia, Mangareva,
Marquesas, Tahiti, Hawaii, Viti, Efate.

2. Sounds of pain: Samoa, Futuna, Niue, Uvea, Maori, Mangareva,
Mangaia, Paumotu, Tongarewa, Nukuoro, Manahiki, Fakaafo,
Aniwa, Tahiti, Viti, Efate, Omba, Motu, Redscar Bay, Santo,
Malekula, Nggela, Mota, Vuras, Motlav, Aneityum, Matupit,
Malo, Baki, Tangoan Santo, Tagalog, Ilocan, Malagasy.

2a. The howl and the tear: Samoa, Tonga, Futuna, Uvea, Maori,
Mangaia, Mangareva, Paumotu, Tahiti, Viti, Nguna, Vatu-
ranga, Bugotu, Nggao, Baravon, Mota, Aneityum, Baki, Malay,
Siwa, Tagalog, Java, Kayan, Ilocan.

ib. The wail of petition: Samoa, Uvea.

To these should be added Lambell tangis, a tear, and Maewo tangtangisi,
in which the tear drops in pity. Until we reach the ultimate items the
common factor of this wide range of significations is the inarticulate sound,
and even where it distinctly means to sing or chant the sense rests upon
the music and not the words.

The form calls for no extended comment beyond the general remark
upon the frequency with which the final .? is preserved. Aneityum taing
is metathetic. The a varies to e in several New Hebridean languages, and
in Santo to o. Java and Kayan show the t-n mutation which appears in
four New Ireland languages in 329. Ilocan sangit seems to be a most
unusual metathesis, type 52341.

414 THE POLYNESIAN WANDERINGS.

The Arabic approaches but a small sector of the sense of tangis, shows
no sign of the radical s, and is really in resemblance only so far as relates
to the initial consonant, no satisfactory showing of kinship.

347.
tai, excrement, filth.

Samoa: tae, excrement. Tonga: tae, excrement; tee, dung of

animals. Futuna: tae, excrement. Uvea: tae, id. Niue:
te, id. Maori: tae, gum exuding from plants; tahae, filth;
tutae, excrement. Tahiti: tutae, tiatae, id. Marquesas: kae,
slaver; tutae, excrement. Mangaia, Mangareva, Paumotu:
tutae, id. Rapanui: tutae, excrement, mildew. Moriori:
hokotae, disgusting, abominable. Hawaii: kukae, excrement,
dirt, filth; haukae, filthiness; kae, the exterior of the anus.
Fotuna: tah-tai, excrement.
Viti : nda, excrement of man ; ndc, of beast.

Motu: tage, excrement. Hula: kage, id. Matupit: taka, id.

Brierly Island : tai, dung. Malo : tai, excrement. Baki : ta, te, id.

Mota: ta, tae, id. Malekula: ne-ten, id. Aneityum: in-tin, id.

Malay, Macassar: tai, dung, ordure, excrement. Magindano: tae, id.

Malagasy: tay, id.
Hebrew: seah, excrement, filth, from yasa' to go out.
The simple stem tae is Proto-Samoan, the Tongafiti has tutae. This
discrimination gives us a time measure for the Moriori hokotae (Jioko is
the equivalent of }aka of resemblance). Once more we have a landmark
of the direct southern migration and we find it to be Proto-Samoan.

Tonga and Viti distinguish between man and beast, and by the same
means. A similar form distinction is found in Baki, but we lack informa-
tion of a corresponding source distinction. In Aneityum in-tin refers to
beasts and a word of another stem, nohok, refers to men.

That this should voluntarily be assumed as a personal name passes
comprehension. Yet in the ancient story of the swimming sisters (terato-
logically a Siamese twin monster) of Samoa the first object seen glistening
on a Tutuila beach on which they came to shore caused one of the sisters
to give herself the name Taenia (ma to shine). Exactly the same word
appears in Mota gak taenia as an endearing expression of admiration.

The Motu and Matupit forms with an inner palatal are listed for so much

resemblance as they possess ; I doubt their affinity. The Malekula ten and

Aneityum tin are of this stem with the local assumption of w-ephelkustic.

The Indonesian identifications are quite satisfactory. I note again the

accord with the Proto-Samoan rather than the Tongafiti and later form.

Except for the subject the Hebrew offering can have nothing whatever
to do with this stem.

348-
tale, a belt, rope, string.

Nukuoro : tali, a rope.

Viti: ndali, a rope or large cord.

Mota: tali, a cord. Wango: tali, a rope.

Malay: tali, a rope, string, bandage. Malagasy: tadi, a rope.

Arabic : dara, to go round, to whirl, to turn.

DATA AND NOTES. 415

Although established at few points this stem is of peculiar value. We
note its identification at two Indonesian points, at three in Melanesia, and
in Viti. In another subdivision of these data we have encountered several
instances of stems common to Indonesia, Melanesia, and Viti. This instance
very brilliantly illuminates those, for here we are prevented from assuming
a Polynesian migration which got no nearer its destination than Viti.
Tali is found nowhere in Nuclear Polynesia or exterior migration lands,
but it is found in Nukuoro. That distant island, almost a backward eddy,
we have already shown to have received its population from Samoa in a
secondary colony. The preservation of tali on Nukuoro shows it to have
existed in Samoa at the time of this northern voyage. Since the date of
that voyage tali has vanished from Samoa and in Nuclear Polynesia has
been replaced by maea (62).

The Semitic proposed has a consonantal resemblance, but there is abso-
lutely nothing in signification to support the resemblance.

349-
tale, the taro (so called because round; talo, taltal, round).

Samoa, Fakaafo, Tonga, Futuna, Niue: talo, the taro. Maori,
Tahiti, Mangareva, Bukabuka, Rarotonga, Fotuna, Aniwa, Vat6 :
taro, id. Nukuoro: taro, tao, id. Moiki: tango, id. Mar-
quesas: tao, id. Hawaii: kalo, id.

Viti : ndalo, the taro.

Motu: talo, the taro. Eromanga: tal, id. Aneityum: in-tal, id.
King, Rubi: kali, id. Sinaugoro: hare, id. Hula: hale, id.
Keapara: ale, id.

Malay: tolas, the taro.

Arabic: data, to go round, to whirl, to turn; da'ro, orb, round.

This staple food product is excellently identified in our three island areas.

The etymology in Efate "so called because round" is worse than childish;
it is a fabrication to establish an identity with the Arabic proposed. After
observing that Arabic data has just been employed (348) to account for
the cylindrical slimness of a cord and now for the roughly ovoid shape of
the taro underground stem, than which two objects could scarcely be more
dissimilar, one pauses in wonder at the mental process.

35o.
talfga na, linga na, nalfga na, the ear.

Samoa, Fakaafo, Futuna, Uvea : talinga, the ear. Nukuoro : talinga,
a horn; taninga, the ear. Maori, Rarotonga, Paumotu, Fotuna,
Manahiki, Aniwa, Fotuna, Rapanui : taringa, the ear. Nuguria :
tar ina, id. Sikayana: kautalina, id. Liueniua: karinga, id.
Tahiti: taria, id. Tonga, Niue: telinga, id. Mangareva,
Vate: teringa, id. Moriori: tiringa, id. Moiki: tanginga, id.
Marquesas: pu-aika, pu-aina, id.

Viti: ndalinga, the ear. Rotuma: faliang, id.

Pala, New Britain, Nguna, Duke of York : talinga, the ear. Laur :
talingd, id. King, Lambell: talngd, id. Buka: dalinga, talinga,
taliga, id. New Georgia : talenge, id. Sesake, Epi, Lo, Bougain-
ville, New Georgia : ndalinga, id. Wango, Fagani : karinga, id.

416 THE POLYNESIAN WANDERINGS.

Marina, Vella Lavella (Bilua) : salinga, id. Ulawa, Bululaha,
Alite: alinga, id. Saa: alinge, id. Kiriwina: taigila, id.
Tavara, Awalama, Taupota : taniga, id. Wedau: tanig'a, id.
Suau, Mukawa, Kwagila, Raqa: taina, id. Mugula, Kiviri:
teina, id. Oiun: tain, id. Galavi, Boniki: teini, id. Dobu:
iena, id. Murua: fe^a,id. Nada: fo'wa, id. Panaieti, Misima :
tanan, id. Motu: /<w'a, id. Redscar Bay: taf/j/a, id. Pokau,
Doura: &aza, id. Kabadi: kaina, id. Roro: haiana, id.
Mekeo: ama, id. New Ireland (Carteret Harbor): pralenhek,
palalignai, id. New Ireland (? Tregear) : plahingia, id.

Eromanga : teligo, id. Lakon : telngan, id. Lamassa : telngd, id.
Pak, Sasar, Alo Teqel: telnegi, id. Motlav: ndelngege, id.
Volow, Norbarbar: ndelngigi, id. Bierian: seligo, id. Moanus:
ndrilinga, id. Baki: tiline, id. Tanna: numateligen, id.
Aneityum: tiknga, id. Malekula Pangkumu: riringa, id.
Ambrym: ringi, id. Natalava: kulinda, id. Nggela: kuli, id.

Malay, Bouton, Dyak, Visayas: talinga, ear. Java, Basakrama:
talingan, id. Menankabau: talingo, id. Satawal: talinhe, id.
Chamorri: talanga (talanja), id. Ulea: talenga, id. Mala-
gasy: tadiny, id. Sulu, Tagalog: tainga, id. Silong: tengeh, id.
Teor: karin, id. Malay, Sula, Baju: telinga, id. Mayapo:
telingan, id. Morella: telina, id. Batumerah: telinawa, id.
Liang, Lariko: terina, id. Awaiya: terina-mo, id. Caimarian:
terinam, id. Saparua: terena, id. Cajeli: telila, id. Wahai:
teninare, id. North Borneo: linga, id. Matu: lingah, id.
Massaratty: linganani, id. Bugi: tjuling. Macassar: to/t, id.

Hebrew: 'ozew, 'azenaim, 'azene, the ear. Syriac: 'adna, id.
Chaldee: 'uden, 'udena, id. Arabic: 'udn', 'addn', 'udun', id.

In the study of this material we shall need to concern ourselves with form
variety. The signification is everywhere the same, the ear. This refers
to so much of the ear as is known to the rude anatomy of these islanders,
the outer ear. The externality is so distinct that in Nukuoro we find the
word serving for a horn. We shall examine the more regular of these
variants point by point, the vowel variety scarcely calling for comment.

T. This is found in all Polynesia but one, in Indonesia (except Teor,
North Borneo, Matu, Massaratty), and in Melanesia as follows:
Efate, New Britain, Duke of York, Nguna, Laur, King, Lambell,
Lamassa, Buka, New Georgia, New Guinea (Motu, Redscar Bay),
Eromanga, Lakon, Pak, Sasar, Alo Teqel, Baki, Tanna, Aneityum.

d, nd. Viti, Buka, Sesake, Epi, Lo, Bougainville, New Georgia, Motlav,
Volow, Norbarbar, Moanus.

k. Wango, Fagani, Natalava, Nggela, Teor.

s. Marina, Bierian.

f. Rotuma.

Vanished: Marquesas, Ulawa, Bululaha, Alite, Saa. To this are to
be added the instances of abrasion of the first syllable: Efat6,
Ambrym, Malekula Pangkumu (regarded as preduplicated n'wga),
North Borneo, Matu, Massaratty.

DATA AND NOTES. 417

L. As I or r it is present in all Polynesia but three, almost all Indonesia,
and a large part of Melanesia. The variants are here noted.

ng. Moiki. n. Nukuoro, Wahai. k. Aneityum. d. Malagasy.

Vanished: Marquesas, Motu, Redscar Bay, Sulu, Tagalog. In Silong
the whole syllable li has vanished. In the following languages
the / is preserved but the i has vanished : King, Lambell, Lamassa,
Lakon, Pak, Sasar, Alo Teqel, Motlav, Aneityum, Norbarbar,
Volow.
NG. This is most extensively preserved. We may not attempt to list
the <7-forms, for our vocabularies are uncertain in the employment
of g for g and for ng. The variants are as follows.

n. Pak, Sasar, Alo Teqel, Baki, Malagasy, Teor, Morella, Batumerah,
Liang, Lariko, Awaiya, Caimarian, Saparua, Wahai.

nd. Nggela.

1. Cajeli.

Vanished: Marquesas, Motu, Redscar Bay. The whole final syllable
nga has been lost through abrasion in Nggela and Macassar.

The following forms call for particular notice. Rotuma jaliang is taliang
and that is seen to be metathetic. Three New Ireland forms are anoma-
lous. One, plahingia, is cited by Tregear without memorandum of source.
The others are derived from Gaimard's Carteret Harbor in the "Voyage of
the Astrolabe," almost a century ago. The region is that covered more
closely in 1904 by Stephan, under the name of Lamassa. Allowing for
less skill in catching the sounds in Gaimard's brief sojourn, pralenhek and
telnga and palalignai are in fair accord as regards the linga element and
plahingia is not far removed. In the older forms we find initial t obscured,
but the obscurity takes the form of a reinforced liquid, pr, pi, pal. While
we may not hope at this lapse of time to reduce this difficulty we note
that in Moanus, in the same gateway area, the initial t is in the form of a
reinforced liquid.

The Semitic forms reduce to 'zn, *dn, neither of which may be at all
associated with ting, the universal stem of our island forms.

35i-
tanu e, tanua, tani, taniu, to spit ; tanua, taniu, spittle.

Samoa : anu, to spit ; anusanga, spittle ; anusalo, to hawk up phlegm.
Tonga: aanu, spittle, to spit; anuhia, to spit upon, to bedaub.
Futuna: aanu, adnusi, to spit. Uvea: aanu, id. Niue:
anu, id. Mangareva: anuanu, id. Marquesas: anuanu,
spittle. Rapanui: aanu, saliva, to spit. Maori: anuanu,
offensive, disgusting. Tahiti: manuanu, loathsome, surfeit-
ing, to be qualmish.

Viti: kanusi, to spit. Rotuma: onusi, id.

Nggela, Bugotu : angusu, to spit. Mota : anus, id. Lakon : anuh, id.
Lo: ngenguh, id. Wango, Fagani: ngusu, id. Saa, Ulawa:
ngisu, id. Bululaha: nisu, id. Nggao: misu, id. New
Georgia: kamisu, id. Boniki: kanu, to spit. Kubiri: kanu,
spittle. Mukawa: bi-kanu, to spit. Kubiri: kakanu, id.

418 THE POLYNESIAN WANDERINGS.

Kiviri: kakan, id.; kan, spittle. Volow: ngih, to spit. Motlav:
nih, id. New Ireland (Carteret Harbor): piniss, id. Alite:
ngilu, id. Motu: kanudi, id. Pokau: anodi, spittle, to spit.
Boniki: kanuna, spittle. Mukawa: kanuta, id. Sinaugoro:
kanunu, spittle, to spit. Rubi: kanuru, id. Hula: kaninu, id.
Keapara: aninu, id. Nada: kinura, id. Misima: kunruvi,
to spit. Wedau: anumai, spittle. Galoma: aniulu, to spit ;
atiulu, spittle, to spit. Kabadi : ainuku, to spit. Malekula :
rw£, id. Malo: lito, id.
Malay: ludah, to spit.
Arabic: rawwala, tarwilu, spittle.
Three stages of stem form are included in these data, kanus, anus, nus.
The longer form kanus is found most definitely in the Viti stream. Motu
gives us kanudi, the s-d mutation having already been noted in this lan-
guage in sina (342) to shine Motu dina and receiving support from similar
mutations to be found in the table. Efate with k-t mutation gives us
the abraded form tanu, and Viti comes back to normal in kannsi. New
Georgia, with an inner mutation later to be discussed, falls into this line,
and so, but less obviously, does Lo ngenguh. Gaimard's New Ireland form
maybe taken as an obscure record of a kanus form, but no particular reliance
is to be placed upon it.

The Proto-Samoan stem is anus, abraded to anu in modern Polynesian,
save where formative suffixes have availed to protect the terminal conso-
nant. This reaches into Rotuma onusi and with slight modification to
Nggela and Bugotu angusu. As a closed stem it is found in Mota anus and
Lakon anuh.

The nus stem is found in Wango and Fagani ngusu, Saa and Ulawa
ngisu, Bululaha nisu, Volow ngih, Motlav nih. Nggao misu involves an
n-m mutation for which we have no confirmatory data save such as may be
derived from the ng-m, m-n, m-ng listed in the table. It follows so
closely upon Bululaha nisu that it is probably safe to accept it. Alite
ngilu involves another mutation, s-I, for which we lack outside support
save as it may be found in s-r discussed in note 329.

Malekula rut and Malo lito fall into accord with Malay ludah, but it
would require a long line of well-established transition forms to bring that
into affinity with any part of the kanus stem.
And the Arabic is utterly out of place.

352.
tas, tasi, the sea, salt :

Tonga: tahi, the sea; fakatahitahi, to wet with salt water; taitai,
brackish. Niue: tahi, the sea, the seacoast, the tide; puke
tahi, vaitahi, salt water. Samoa, Maori, Rapanui, Nuguria :
tai, the sea, the tide. Futuna, Tahiti, Mangaia: tai, the sea,
the seacoast. Marquesas : tai, the sea, the shore ; taitai, slightly
salt. Mangareva : tai, the sea, the shore, salt water. Tonga -
rewa: tai, salt. Nukuoro: tai, the sea, salt water, the south.
Uvea, Rarotonga, Manahiki, Aniwa, Fotuna, Vate: tai, the sea.
Sikayana : waitai, salt water. Hawaii : kai, sea, salt water.
Viti: tathi, the sea. Rotuma: sasi, sosi, seas, id.

DATA AND NOTES. 419

(a) in the sense of sea :

Marina, Tanna, Sesake : tasi. Lambell, King : tasi. Buka: tassi,
dassi, tisi, itis. Santo: tost. Epi: tzi, tsi. Pala: tes.
Kalil: taas. Moanus: ndras. Laur: lontas. Belega: tathi.
Arag, Nggela : tahi. Makura : na-tah. Matupit : ta. Ambrym :
tei, tie. Epi (Southeast) : si. Wango : asi. Malekula :
nitis. Malo, Mota : tas. Baki : tei. Bierian : sahi. Ero-
manga Ura: de.

Kawi, Bugi, Java : tasik. Malay: tasek. Gah: tasok. Ceram,
Ahtiago : tasin. Tobo, Ahtiago, Ceram, Guaham : tasi. Togean
Islands: tasie. Chamorri: tahsi. Satawal: tali. Matabello:
tahi. Sunda: chai. Macassar: djai.

Arabic: ta's', the sea.

(b) in the sense of salt, salt water :

Sesake, Marina, Vaturanga: tasi. Arag, Omba, Nggela, Bugotu,
Nggao: tahi. Maewo: tas. Ulawa, Wango, Fagani, Saa,
Alite: asi. Bululaha: moi asi. Motu: tadi. Duke of
York: tai. New Britain: ta.
Amboyna: tasi. Awaiya, Caimarian: tasie. Buru: sasi. Tidore,
Gani, Galela: gasi. Sanguir: asing.

The Proto-Samoan stem is tahi. The form with aspirate is selected as
primal for the reason that a Proto-Samoan tasi would have remained tasi
in modern Samoan and would have become tahi in most of Polynesia.

It is only in Tonga tahi sea, taitai brackish, that we recognize any form
distinction of the two significations. While, therefore, they have been
segregated in the data collated we shall consider the form variants together.

In addition to the sense of sea and salt water, hardly to be dissociated,
we find in Polynesia the following added significations :

Tide: Samoa, Niue.

Shore: Niue, Futuna, Tahiti, Marquesas, Mangareva, Mangaia. This
is a development out of the system of direction nomenclature, for as ki iita
means landward, so ki tai, seaward, is limited by the shore as the end of
pedestrian locomotion.

South: Nukuoro. This has for us a peculiar interest. Nukuoro is but
a tiny speck of an island ; 360 degrees of its horizon show nothing but the
upward saucering of unbroken sea. It was over the southern sea of 3,000
miles that the people came voyaging from Samoa. It must have been a
bitter drift of castaways coming famished to land when generation after
generation has preserved the word for sea, the terrible sea over which they
struggled, in the sense of south. Other seas there about them, to north,
to east, to west ; but a race impressed by the agony of its castaway forebears
still recalls the one sea, the south.

The forms tasi, tahi, tathi are very clearly identifiable in Melanesia and
Indonesia. Epi tzi, tsi, are, of course, tasi with loss of the former vowel.
The tas by abrasion is found in Malo, Mota, and Maewo ; and Us in Buka
itis, Malekula ni-tis. Next we find tah in Makura na-tah, and a still further
progress in New Britain and Matupit ta, followed by Ambrym tei, tie, Baki
tei, Eromanga Ura de. Laur lontas, is a composite, Kalil taas is recogniz-

420 THE POLYNESIAN WANDERINGS.

able, and Moanus ndras presents the irregular reinforcement which has
before been noted in words from that speech. Bierian sahi corresponds
to Rotuma sasi and sosi, of which seas is a metathetic stage, and to Buru
sasi. Motu tadi finds a parallel in Satawal tati, and Duke of York tai
in the modern Polynesian form is not far remote from Sunda chai and
Macassar djai. Frontal abrasion yields asi in the Solomon Islands, Ulawa,
Wango, Fagani, Saa, Alite, and Bululaha, and asing in Sanguir. The Epi
si is another step in such abrasion readily reached through its tsi form.
The gasi of Tidore, Gani, and Galela smacks of the palatalization of t now
affecting Samoa.

The Arabic is a resemblance.

353-
tiko, a staff, a walking-staff, a pole by which a canoe is poled forward in
shallow water.

Nukuoro: tokotoko, a pole. Tonga: toko, a canoe-pole, to punt;
tokonaki, to rest the hand on anything. Futuna : toko, a canoe-
pole, to punt; tokotoko, a long staff, a cane; tokoi, to support.
Niue : toko, a prop, to support ; tokotoko, a staff. Uvea : tokotoko,
a staff; tokoni, to support, to prop. Maori: toko, a pole, to
punt ; tokotoko, to support oneself with a stick. Marquesas :
totoko, to prop up with a stick ; tokotoko, a stick, a cane. Manga-
reva : toko, the pole of a raft, to punt ; tokotoko, a stick. Rapanui,
Paumotu: tokotoko, a walking-stick. Samoa: to'o, a canoe-
pole, to punt a canoe; to'oto'o, a staff, a walking-stick; to'onai,
to lean upon a staff or anything for support. Tahiti: too, a
canoe-pole ; tootoo, a staff or walking-stick, to punt. Hawaii :
koo, to punt; kookoo, a staff.

Viti: ndoko, a punting-pole ; toko, a prop, a pole to boom out the
sail; tokona, to prop up. Rotuma: oi-tok, staff.

Mota : tigo, a pole ; tigonag, to pole a canoe ; tigotigo, to use a walking-
stick. Malo : tiko, a walking-stick.

Malay: takan, a staff. Kacassar: tokong, to push with a pole.
Malay: tongkat, a walking-stick. Tagalog: tungcod, id.

Pampangas: tucud, id. Malagasy: tehina, a staff; mitehina,
to walk with a staff, to walk leaning on a person.

Arabic: toka'at, a staff, a support, he who leans much on his side
and props himself up.

The Proto-Samoan stem is tokon.

The primal signification is the verbal one of supporting or propping.
This is made clear in Rotuma oi-tok, oi tree, tok to support, hence a staff.

In Melanesia the three identifications agree upon the vowel change to
tiko, and Mota preserves the radical n.

In Indonesia the Malay takan and Macassar tokong are satisfactorily
associated with tokon, and Malagasy tehina lags only a pace behind. But
Malay tongkat and the similar forms in Tagalog and Pampangas are not
readily reconcilable.

The Arabic shows no trace of the radical n, and in added material differs
from tokon to such an extent that we may not accept it as a resemblance.

DATA AND NOTES. 421

354-
tiri, riri, to fly; lea ki, to toss away, to throw down, to sweep, to drive
away (as the wind trees) ; leleaki, intensive.

Samoa : lele, to fly; jelelei, id. ; lelea, to be driven off or carried away
by the wind ; talele, to escape quickly ; jetulele, a shooting star ;
the root le seems to appear in leangiangi, an eminence exposed
to the breeze. Nuguria: ele, to fly soaring. Nukuoro: lele,
to leap, to jump; manulele, a bird. Tonga: le, to drive, to
affright fish into the net with a rod; lele, to run; lelea, to be
wind-driven; fakalele, to make a cast of birds, to fly a kite;
leleaki, leleekina, to run with; malele, to run. Futuna: lele,
to run swiftly, to fly; lelea, to be wind-driven. Niue: lele,
to fly, to flow (as water) . Uvea : lele, to fly, to run. Hawaii :
lele, to fly, to jump, to leap, to move swiftly, to move as a
meteor through the air; lelele, to leap, to jump, to fly quickly;
lelelele, to run off in haste. Maori: rere, to run (as water),
to fly, to flee, to sail, to leap, to go swiftly. Tahiti: rere, to
fly, to leap. Rapanui: rere, to fly; vae rere, to run; hakarere,
to abandon, to abjure. Mangareva : rere, to fly, to leap ; irere,
to run. Rarotonga: rere, to fly, to flee. Moriori: rere, to
run (as water), to fly. Paumotu: rere, to fly, to soar, to leap.
Fotuna : rere, lele, to fly. Marquesas : ee, to go, to go away ;
eepau, to leap; hee, to fly, to soar.

Viti: lele, to pass or go a short journey; lelele, a ferry canoe.
Rotuma: ferefere, to fly.

Motu: lele, to swim (as a fish). Mota: rere, set of sea, current.

Malay: leleh, to flow.

Arabic : tara, to ffy, to be swift, to move quickly.

In the present phase of Polynesian lele so much means to fly that the
plainest way of particularizing birds is to describe them as the flying animals,
manulele. But it is manifest that flight, an exercise or balancing of wings,
was by no means the primordial sense, for how could that give rise to a
description of water in the water-courses? It will be well to mass the
several significations which lele exhibits.

Flight of birds: Samoa, Nukuoro, Tonga, Futuna, Niue, Uvea, Hawaii,

Tahiti, Mangareva, Maori, Rarotonga, Moriori, Paumotu, Rapanui,

Fotuna, Efate\
Wind driven: Samoa, Tonga, Futuna, Efate.
Meteors: Samoa, Hawaii.
To leap: Nukuoro, Hawaii, Maori, Tahiti, Mangareva, Paumotu,

Marquesas.
To run: Samoa, Tonga, Futuna, Uvea, Hawaii, Maori, Mangareva,

Rarotonga, Rapanui.
Flow of water: Niue, Maori, Moriori, Mota, Malay.
To swim: Motu.
To sail: Maori, Viti.
These several activities are exercised in earth, air, and water. The
common factor is the swift motion. The means of motion cut no figure.

422 THE POLYNESIAN WANDERINGS.

It is an invisible means in the driving of the wind, the flash of the meteor
silent athwart the sky on its lethal errand, the slip and slide of the stream in
its deep course, the set of the sea, the gliding of the canoe upon its surface.
Samoa, Tonga, and Efate use the unduplicated stem and establish it
firmly. Therefore the Arabic tdra is not acceptable despite its resemblance
to the single Efate form tiri, a dialectic mutation of riri.

355-

titiro, to gaze into the sea looking for fish or molluscs, to look at oneself
in water or a mirror.

Samoa: tilotilo, to peep, to spy; tio, sharp looking (of the eyes).
Nukuoro: tiro, to see. Tonga: jio, to look, to stare; jio jio,
to gaze, to stare about; fakajio, to peep, to look, to stare, to
spy, to pry; kilo, to look askant; kikilo, to look from side to
side. Futuna: tilo, to gaze, to look with an optical instru-
ment; tiloika, to watch for fish; tiloata, to look at oneself in
any reflecting surface ; tio, to see, to look at, to observe, to spy.
Maori: tiro, to look, to look at; tirotiro, to gaze around, to look
about, to investigate. Tahiti: titiromatatia, to gaze, to look
steadfastly, to cast a lustful look. Hawaii: kilo, to look ear-
nestly at a thing ; kilokilo, to examine carefully ; hookilo, to spy.
Mangaia: tiro, to look. Rapanui: titiro, to admire. Atiu:
titiro, to look at. Marquesas: tiohi, id. Mangareva: tiho,
to examine, to regard attentively; matiro, to examine. Mana-
hiki: titi tiro, to see. Fotuna: jironga, a looking-glass.

Viti : tirova, to look at oneself in water, to peep at ; titiro, a viewing
indistinctly.

Buka : tara, tarra, to see. Santo : titiro, a looking-glass. Mota :
tiro, clear; tironin, to look into a pool or mirror.

Macassar: tiro, to spy. Malagasy : taratra, tarafina, to look, to gaze.

Arabic : nat'ara, to gaze, to look for, to consider, to spy.

Polynesian discloses three stems here, tio, tilo, and kilo. The kilo form
is found in Tonga only, for Hawaii kilo is the kappation of tilo and kilo
would have passed into the northern speech as ilo under its normal loss
of the true k. Tilo has a Proto-Samoan stem tilof, sufficient to preclude
the inclusion in this series of the open ilo, iloa, to see, to know. Tio has a
Proto-Samoan stem tiof, therefore it is well established as a derivative of
tilof, although the process of internal loss is quite rare in Polynesian except
when a letter is wholly discarded regardless of position. The Tongan
vocabulary shows no protected form of kilo whereby we might determine
its earlier stem. The sense of looking at oneself in water or in the lately
introduced mirror is found in Samoa, Futuna, Viti, Fotuna, Efate, Santo,
and Mota.

Though scanty, our Melanesian and Indonesian identifications are
satisfactory.

The Arabic stems in ntr, the Polynesian in trf, and these are by no means
in accord.

DATA AND NOTES. 423

356.
toa, t5wa, to', a domestic fowl, a bird.

Moiki: toa, a fowl. Samoa: toa, a cock. Rapanui: moa, toa, id.

Viti: toa, a fowl.

Sesake, Epi, Arag, Mota, Malo, Marina, Gog, Omba: toa, a fowl.
Ambrym, Malekula, Lakon, Norbarbar, Pak, Sasar, Vuras,
Mosin, Alo Teqel, Motlav, Lo : to, id. Ulawa, Fagani, Bululaha,
Alite, Wango, Saa: kua, id. Santo: toa, a bird. Baki: ju,
a fowl. Bierian: so, id.

Malay: ay am, a bird. Ceram: tofi, towim, id. Bum: tehui,
teput, id.

Hebrew: 'of, bird. Arabic: 'a'f, a fowl.

The solitary instance in Polynesian of toa a fowl in place of the common
moa is in Moiki, an island of the western verge. This is balanced by
Rotuma moa a cock. Of the Viti toa Hazlewood makes the note "a vosa
maivei beka na toa — where in the world does this word toa come from?"
As toa and to the word is abundantly established in the New Hebrides, a
sufficient answer to this question. Bierian so may be regarded as a to-
derivative, and Baki ju argues a tu variant of to. This suggest a tua, from
which we might derive kua in the Solomon Islands ; unfortunately the t-k
mutation, though common enough elsewhere, finds little confirmation in
these languages.

Not one of the Indonesian identifications seems to bear any relation to toa.

Dr. Macdonald would have found his task easier if he had essayed to
prove Dame Partlet of Rotuma a mother in Israel, for in that language the
hen is nof. Certainly his Semitic offerings suggest no kinship with toa.

357-
toko, to rest, to sit down, to dwell, to remain, to be.

Samoa : to' a, to subside, to settle down, to coagulate. Uvea : toka,
to fall; totoka, to coagulate, to curdle. Tonga: toka, to lie
down, to be, sediment. Futuna: toka, to fall, to lie down.
Niue: toka, to remain. Maori: totoka, to congeal.

Viti : tiko, to sit, to be (of persons) ; toka, to stand, to be placed
(chiefly of things inanimate) ; toka levu, to sit with arms and
legs abroad as one in pain.

Mota: toga, to abide. Volow: tonga, id. Arag: dogo, to sit.
Sesake: toko, ndoko, id. Nifilole: togoli, id. Mosin: toga,
to abide; tog, to stay with. Fagani: 'oga, to stay. Pak:
'o, 'oga, to abide. Wango: 'o'a, to stay. Merlav, Lo: tog,
to abide. Gog: tog, to live at. Vuras, Marina : tog, to stay.
Maewo: tog, to sit; togana, togaana, a sitting. Norbarbar, Ugi :
to, to stay. Nggela : to, to sit. Retan : ta, to stay. Sesake :
to, ndo, to be, to abide, to sit. Ambrym: dro, dru, to abide;
ro, to sit, to stay, to remain.

Malay, Java: dodok, duduk, to sit, to be. Malagasy: tuatra, id.

Hebrew: takah, to sit, to remain.

We find the word limited by Nuclear Polynesia and the direct migration
thence to New Zealand.

424 THE POLYNESIAN WANDERINGS.

Our Melanesian identifications are so abundant that it has been possible
to arrange them in a very satisfactory chain of form development, and the
several threads of meaning interlace and bind the whole.

The Malay and Java identifications are satisfactory; the Malagasy, in
the absence of transition phases, somewhat less so.

The Hebrew is a resemblance.

358.

tu, tsu, su, ru, to stand, to abide, to dwell, to be; tu-ri, to stand, to abide
with; tu-raki, to stand or abide for.

Samoa: tu, to stand, to arise; tula, a locality or habitat; tula, a
pigeon perch; tula'i, tutula'i, to rise up, to stand up. Nuguria:
tu, to stand. Tonga: tuu, to stand, to rise; tuula, a bird
roost. Rapanui: hakatuu, to erect, to build. Futuna: tuu,
to stand, to arise, to be situated or placed ; tuidanga, post, sta-
tion, place of residence; tuulaki, to arise, to stand up. Niue:
tu, tutu, to stand ; tuulupou, ground plates of a house. Uvea :
tutuu, tuulaki, to stand ; tuu, to arise. Nukuoro, Fotuna : tu,
to stand. Moiki : tu, to stand, to abide. Maori : tu, to stand ;
tunga, a standing place ; turanga, time, place or circumstance of
standing. Tahiti: tu, to stand erect, upright. Rarotonga:
tu, to stand, to arise. Marquesas: tu, to stand, to remain
fixed. Mangareva: tu, being, essential, to be the essence of,
to be upright, to stand up; turanga, a site, standing place.
Hawaii: ku, to arise, to stand erect; kulana, a place in a fort
where the men stand to throw their spears; kulanahale,avil\age.

Viti: tu, to arise, to stand, to be; tura, vakatura, viritu, to cause to
stand, to place; tutu, a stand, a place or thing to stand on.

Tangoan Santo, Malo, Duke of York, Raluana: turu, to stand.
Nggela: tuguru, id. Maewo: tura, id. Fagani, Wango:

'ura, id. Saa: 'ure, id. Marina: turi, id. Mota, Vuras,
Merlav, Gog, Pala, Baravon : tur, id. Lambell : itur, id. ; kamtur,
to stand up. Lamassa: itur, to stand; kaptiir, to stand up.
King: itur, to stand; ikatutur, to stand up. Matupit: tut, to
arise; wa-tut-pa, to make to rise. Vaturanga, Deni, Sesake,
Lo, Omba, Maewo : tu, to stand. Malekula : tu, tutu, id. Motu :
tutuka-tutuka, to stand firm, to be steady. Sesake: ndu, to
stand. Retan: tir, id. Tanna: (t)utul, id. Baki: jumolu, id.
Bierian: mtumau, id. Volow: tig, id. Norbarbar: ti, id.

Malay: diri, to stand; kukuh, stable, firm. Malagasy: juru, id.

Hebrew: nasa', so', s'et, to stand. Ethiopic: nasa' a, id. Arabic:
nas'a' , id.

Exactly at this point a note is proper as to the substantive verb, for one
of its most frequent substitutes is the sit of the preceding item, the other
is the stand of the present item. To the Polynesian and to the Melanesian
has come no concept of bare existence; he sees no need to say of himself
"I am," always "I am doing," "I am suffering." It is hard for the stranger
of alien culture to relinquish his nude idea of existence and to adopt the
island idea; it is far more difficult to acquire the feeling of the language

DATA AND NOTES. 425

and to accomplish elegance in the diction under these unfamiliar conditions.
Take for an illustrative example these two sentences from the Viti :

Sa tiko na tamata e kila: there are (sit) men who know.
Sa tu mai vale na yau: the goods are (stand) in the house.

The use of tu for tiko and of tiko for tu would not produce incomprehensi-
bility, but it would entail a loss of finish in diction, it would stamp the
speaker as vulgar, as a white man.

Again there is a use of these verbs as auxiliaries to some principal verb
in a fashion that is yet more difficult to grasp. Sa lako tu and sa lako tiko,
both "he is going." But sa lako tu suggests "he is going here in plain
sight" ; and sa lako tiko "he go-sits," to employ a barbarism in a speech of
culture representative of an elegance in a barbaric speech, produces the
sense "he is going steadily on and on."

Savage life is far too complex ; it is only in rich civilization that we can
rise to the simplicity of elemental concepts.

The Proto-Samoan stem is tul.

In Polynesia the only variant is Hawaii ku, the common kappation.

In Melanesia our identifications fall into groups as representing tul, tu,
tutu, and til. Turn, tura, 'ura, turi, 'uri, tur are the representatives of the tul
group, and probably to be associated with this is Tanna tutul. Tu, the
common Polynesian type, is found in Melanesia at scattered points, Efat£,
Vaturanga, Deni, Sesake, Lo, Omba, Maewo, and Malekula; and the two
Epi forms, Baki jumolu and Bierian mtumau, seem to be /^-composites.
Tutu, the normal duplication, is found only in Malekula; Motu has a
tutuka, a tutu enlarged and then duplicated in a manner which a true Poly-
nesian would never tolerate. An interesting form is Matupit tut, tu dupli-
cated and then abraded. The stem with vowel change, til, occurs but three
times and in a very narrowly restricted area of two small islands in the
north of the Banks Group. Re tan fir is clearly the simple stem and
Norbarbar ti stands to til as tu to tul. But Volow tig is anomalous; the
same anomaly is found in Nggela tuguru.

We have but two Indonesian identifications. Malagasy juru is of the
more usual type; Malay dirt reestablishes the til stem just under discus-
sion. It is uncertain whether the kappation is sufficiently well established
in Malay to cover kukuh as a tutu equivalent.

The Semitic offered here, ns' stem, can have no relationship with tul or til.

359-
turua, full of holes (as a rock through which water percolates) ; tuturu, to
drip (as eaves), to leak (of a roof); tuturu, a drop, dripping;
turu-ki, to drip or leak through; tirikit, to begin to drop or
spatter (of rain) .
Samoa: tutulu, to leak (of a roof), to weep (language of ceremony) ;
tulutulu, the eaves of a house; fa'atulutulu, to cause to drop;
tului, to drop a thing into; tuluvao, the drops from the trees
after rain. Tonga: tulutulu, to drop, to drip, to let in water,
the eaves; tutulu, to fall in drops, leaky; tuluta, a single drop,
a tear ; mokulu, to fall as tears ; tului, to drop into the eye or
into a wound. Futuna: tutulu, to fall drop by drop; tului,

426 THE POLYNESIAN WANDERINGS.

to drop into the eye or into a wound; tulu, tulutulu, to drip;

tulutulu, the eaves. Nukuoro : turuturu, to fall in drops : haka-

tulu, to pour out. Fotuna: turn, to leak, to drip. Maori:

tuturu, to leak, to drip; maturuturu, to fall in drops, to distil.

Rapanui: turu, to flow, to drip. Tahiti: tuturu, to drop, as

rain from a roof. Hawaii: kulu, a drop, to drop, to leak.
Viti : turu, the eaves ; turu, tuturu, tin, titiri, a drop, to drop ; turuva,

tiriva, to drop upon.
Motu: hetuturu, to drop (as water). Malekula: turitur, to leak,

to drip. Malo: tuture, id. Baki : mbari-drudruli, id. Mota:

tir, to drip, to drop.
Malay: jurus, to water, to irrigate.
Arabic : s'alla, to sew, to shed tears ; s'als'ala, to drip, to fall in drops ;

was'ala, to drip, to drop, to leak out.

The Proto-Samoan stem is tulu.

This is well established by the recurrence of tului in which the verb-
formative * is, as always, added to the stem and avails to preserve a final
radical consonant if the stem is a closed one. This leaves the v in Viti
turuva, tiriva, non-radical but to be explained in the constructive local
usage of that speech.

The vowel change u-i, found slightly in 358, here occurs in Viti, Efate,and
Mota. A similar change affecting the weak unaccented final vowel gives
us a turi stem in Malekula, ture in Malo, and the Baki composite is tuli
with a reinforcement of the initial mute. In Efate tirikit, tiri compounds
with kit (149) small.

The Malay jurus, accepting the common t-j mutation and assuming that
the final .y has the appropriate construction value, suggests kinship with
turu; but we can not definitely accept the identification on such slight data
as against the considerable variation in sense.

The Semitic and the island tongues have an / in common, far too little
on which to build an argument.

360.

ua, ueti, boua, fruit.

Samoa, Tonga, Futuna, Niue, Uvea, Fotuna, Moiki : fua, fruit. Sika-
yana: fua, an egg. Nuguria: hua, fruit, egg; te hua te rakau,
blossom. Maori, Hawaii, Nukuoro: hua, fruit. Aniwa:
no-hua, id. Rapanui: huaa, fruit, to bloom, to blossom.
Rarotonga: ua, id.

Viti: vua, fruit. Rotuma: hue, id.

Moanus: mbua, fruit. Eromanga: buwa, id. Uni, Pokau: vua, id.
Alite, Nggela : vuavua, id. Pokau: vuavua, flower. Kabadi:
ovaovana, id. Rubi: o-wuawua, id. Vaturanga: vuvua, fruit.
Baravon : vuei, to bear fruit. New Britain : vue, id. !'.Mekeo :
pua, fruit. Hula, Keapara, Galoma : hua, id. Roro: buana, id.
Rubi: guagua, id. Fagani: fua, id. Ulawa, Saa, Wango,
Doura, Bululaha: hua, id. Motu: huahua, id. Aneityum:
howa, id. Buka: hoa, id. Dobu, Taupota, Kubiri, Galavi:
ua, id. Suau: uaua, id. Mukawa: uwa, id. Taupota,

DATA AND NOTES. 427

Wedau, Awalama, Galavi : iuwa, id. Tubetube : uwe, id. Duke
of York: ua, to bear fruit. Malekula Pangkumu: mi uan,
fruit. Sesake: wa, id. Tanna: nowa, id. Nifilole: nua, id.
Deni, Marina: va, id. Norbarbar: wo, id. Mota: woai,rid.
Arag, Omba, Duke of York: wai, id. Merlav, Gog, Motlav:
we, id. Lakon: wehi, id. Mosin: weegi, id. Pak: wewegi, id.
Sasar, Alo Teqel: wowegi, id. Volow: wenen, id. Lo: w'», id.
Vuras: icrc'c, id. Ambrym: «*, id. Maewo: o^', id.

Salibabo : buwah, fruit. Bugi : buwa, id. Malay : bnah, id. Salayer,
Menado, Borneo, Liang, Baju : bna, id. Amblaw, Sanguir : buani,
id. Cajeli: buan,id. Ahtiago: uan, id. Mayapo, Massaratty:
fuan, id. Teor : fuin, id. Teluti, Wahai: /maw, id. Morella:
/iwa, id. Caimarian: huwai, id. Java: wo/f, id. Gah: •ze/oyajd.

Aramaic: /era, fira, to bear fruit. Hebrew: peri, para, id.

With this stem we should collate pua, of cognate meaning.

Maori: pua, a flower, a seed, to blossom. Tahiti: pua, to blossom.
Hawaii: pua, a flower, to blossom. Mangaia: pua, to blossom.
Mangareva: pua, a flower, to bloom. Marquesas: Ipua, a

flower, to blossom. Paumotu: pua, a flower.

We should also collate funga (246).

In Sesake we find the funga stem doing duty for blossom and for fruit.
In Indonesia we find funga in the fruit sense in Sulu, Tagalog, Ilocan,
Magindano, all Philippine tongues. This suffices to prove that funga and
fua were by some regarded as sufficiently akin to allow the sense of fua to
pass to funga.

So too of pua and fua. While the sense of pua is that of flowering, where-
ever it is found, the Maori is proof that it could have the fua fruit sense as
well. So far as we can now discover pua is an open stem, but its occurrence
is altogether in the Tongafiti area, a later stage of the language, so that this
is not conclusive.

And fua itself is susceptible of both senses. This comes to the surface
in Buka, where hoa is fruit and ua blossom. We find the confusion of mean-
ings also in Samoa, Maori. There is reason to infer that the Proto-Samoan
stem is fuat, but this does not necessarily militate against stem identity
with pua.

The data are so assembled in the foregoing collation as to indicate the
gradual deviation from the primal stem. The identity is found of con-
siderable extent in Indonesia.

The Semitic forms proposed are clearly uncoordinate.

361.

uango, uak, pig, swine ; bwakas, bwokas, a pig (not boar or sow) specially
reared and esteemed for its flesh.
Tonga : buaka, a pig. Futuna, Niue, Uvea, Nukuoro, Vate, Raro-
tonga, Mangareva: puaka, id. Rapanui: puaka, beast, animal,
but not pig. Samoa: pua1 a, a pig, swine, animals in general.
Maori: poaka, a pig. Tahiti, Marquesas, Hawaii: puaa, id.
Aniwa, Fotuna: pakasi, id.

428 THE POLYNESIAN WANDERINGS.

Viti: vuaka, vore, a pig. Rotuma: puaka, id.

New Georgia: boako, pig. Lifu, Nengone: puaka, id. Tanna:
puka', a pig, to grunt. Epi Bierian : bukahi, a pig. Sesake:
bokasi, flesh. Eromanga : mpokas, a pig. Aneityum : pigath, id.
Kabadi, Motu: boroma, id. Masig: boromo, id. Motu:

buruma, id. Kunini: 6/owe,id. Miriam: borom, id. Mabuiag:
burum, id. Pokau: boloma, id. Uni: buluma, id. Sariba,
Murua: buruka, id. Nada: buruku, id. Kiriwina: bunuka, id.
Lambell, Lamassa, Baravon: 6oroi, id. King, Tavara, Awa-
lama, Taupota, Wedau, Galavi, Boniki, Mukawa, Duke of York :
fcoro, id. Nggela: mbolo, animal, beast, pig. Savo: />o/o, pig.
Roro: aiporo, id. Kubiri, Raqa, Kiviri, Oiun: foro, id. Uni:
foloma, id. Bugotu: 60^0, id. Nggao: bosu, id. Pala:
ftdre, id. Laur: mbor, id. Solomon Islands: boa, id. Tube-
tube: £oa/o, id. Natalava: boalo, id. New Ireland (Carteret
Harbor): bouri, id. Moanus: pou, id. Fagani: 600, id. New
Ireland (Duffield) : bogh, id. Alite: bo, id. Brierly Island,
Tubetube, Tagula: bobo, id. Panaieti, Misima: bobu, id. Rua-
vatu: 6e, id. Ugi, Ulawa, Wango, Bululaha: po, id. Malo:
6oi, id. Marina, Nifilole: /50c, id. Ambrym, Epi Baki:
bue, id. Deni: 7«/pot, id. Omba, Maewo: kmbwoe, id. Merlav:
kmpwoe, id. Arag, Mota: kpwoe, id. Motlav, Gog, Norbarbar :
kmbwo, id. Volow: nggmbwo, id. AloTeqel: &m£wo,id. Pak,
Sasar: fe6wo,id. Lakon, Vuras, Mosin,Lo: £/>wo,id. Saa: />wo,
id. Buka: ^w, wgw, paum, apum, id. Sesake: wango, id.

Waigiou: 60, pig. Mysot: 6o/f, id. Buru: babue, id. Malay:
6a6i, id.

Arabic: fuka, fuak', fuwak', to emit hoarse guttural sounds; fakfaka,
to bark.

Both Pratt and Hazlewood, in the Samoan and Viti dictionaries respec-
tively, credit pua'a and vuaka with derivation from and corruption of
English pork. They little knew the extent of this ancient island word.
To show that I am not alone in my view I cite Mr. Tregear's interesting
note (Maori- Polynesian Comparative Dictionary, s. v. poaka) :

This word (generally supposed to be a corruption of the English word porker) is
genuine Polynesian. It was probably received by the Maoris from the Tahitian inter-
preter of Captain Cook, although the passage in Vol. ii, p. 372, of Cook's Voyages, ed.
1793, urges that the Maoris already knew the word. It is possible that the Maoris had
kept a traditional knowledge of the animal, just as in some smaller Polynesian islands
the natives called the dog kuri at sight, although the animal was not to be found amongst
them. The hogs were numerous in Tahiti, Hawaii, etc., before the arrival of the Euro-
peans, and the native hog appears to have been of a different species from the imported
breed. In the Hervey Group pigs were found on Atiu and Rarotonga, but not on the
neighboring islands of Mangaia and Aitutaki. However, at Mangaia, districts known
as pa-puaka (hog-pen) and puaka-ngunguru (grunting hog) are names in evidence of
the pig having once been known on the island.

In the Melanesian area we find at least three stems. Of one the full stem
is pakasi, as found at Aniwa and Fotuna. Sesake bokasi and Bierian
bukahi are but slight deviations therefrom. The first alteration is the

DATA AND NOTES. 429

abrasion of the final vowel, yielding Eromanga mpokas, Efate bwakas and
bwokas; and Aneityum pigath derives from this stage. The second altera-
tion is abrasion of the then final consonant yielding Tanna puka'. The
Efate bwakas affords the transition phase to the puaka type of Polynesia in
general, which is found in Lifu and Nengone (yet possibly a direct intro-
duction by Samoan missionaries) and in boako of New Georgia. Efate
uango, uak, Sesake wango, derive from puaka by frontal abrasion and k-ng
mutation.

The next stem is boro. In New Guinea we encounter boro with ra-accre-
tion, Kabadi boroma, Motu buruma, Miriam borom. As boro the stem is
found in the eastern gateway, New Britain, Duke of York, New Ireland,
in boroi, boro, mbolo, polo, boalo, bouri, botho, bosu, mbor.

The third stem is po. This appears in a wide range of forms. It is not
beyond probability that po may be an attrition form of boro. The Laur
mbor, in a district adjacent to boro forms, and the Solomon Island boa may
provide transition phases leading to po.

The forms found in Buka are not susceptible of derivation from any of
these three forms.

In Indonesia Waigiou bo and Mysot boh are satisfactorily identified with
the Melanesian po-stem. The Malay babi, Buru babue, offers a difficulty.
Codrington (Melanesian Languages, 86) offers Sanguir and Salibabo bawi
as a transition to boh and bo.

The Semitic, fk stem, can not be brought into relationship with our
primal pakasi.

362.
ula, a maggot.

Samoa, Futuna, Hawaii: ilo, a worm or maggot in putrid flesh or
ulcers. Maori, Tahiti, Mangareva: iro, id. Fotuna: eiro, id.
Viti: uloulo, a maggot.
Motu: uloulo, a maggot. Mota, Malo: ulo, id. Baki: ilo, id.

Malekula: na-uru, id.
Malay: ulat, a maggot. Malagasy: ulitra, id.
Ethiopic : 'es'e, a worm ; 'as'ya, to breed worms.

The identifications are quite satisfactory, the i-u vowel change being well
established. While the ilo forms are characteristically Polynesian and ulo
characteristically Melanesian, Baki ilo is a Melanesian occurrence of the
Polynesian form. The Efate ula has the same vowel scheme as Malay ulat.

The Semitic forms have no possible relation to the island forms.

363.
uose, uos, uohe, an oar, a paddle.

Tonga: fohe, a paddle; taufoe, a rope belonging to the steer oar.
Niue: fohe, a paddle. Samoa: foe, a paddle; foefoe, to paddle
briskly. Futuna: foe, an oar, a paddle. Fotuna: foi, id.
Maori, Tahiti, Marquesas, Nukuoro, Nuguria, Rapanui, Tonga-
rewa : hoe, a paddle, an oar, to row, to paddle. Mangareva :
hoe, id. ; ohe, an oar, a paddle. Hawaii : hoe, a paddle; hoehoe,
to row.
Viti : vothe, a paddle, to paddle.

430 THE POLYNESIAN WANDERINGS.

King: voso, a paddle. Epi: voho, id. Laur, Lamassa: vos, id.
Malekula: bos, id. Suau: vosc, id. Mugula: wosi, id. Sariba:
w<m, wo.?e, woase, id. Pokau : i^cZe, id. Roro : bote, id. Mekeo :
poke, id. Kabadi: ode, id. Solomon Islands : jose, id. New
Ireland (Duffield) : j^w, id. Buka: /zo^c, id. Motu: Ziorfe, id.
Pala: udsa, id. Mota: wore, id. Awalama, Taupota, Wedau,
Galavi, Boniki : voe, id. Mukawa,Kubiri,Oiun: 6oc,id. Kiviri:
boi,id. ~Raq&: bol'e, id. Tanna: wa,id. Aneityum : ne-hev, id .

Visayas: bugsai, paddle. Malay: dayung, id. Malagasy: /mu, id.

Arabic : mikda}' , migda}', mihdaf , mikdaf, an oar. Amharic : makzaj,
id. Arabic: "aduf, id.

The Proto-Samoan stem is fohe, as carried in Tonga and Niue, and oddly
seen in Mangareva ohe.

There is a certain agreement in these forms which compares very well
with the truly remarkable showing of the material offered in item 280.

In Melanesia we have most commonly the forms in s. The nearest to
the stem fohe is Solomon Islands fosc, Mota wose, Efate uose. A vowel
change in the unaccented final syllable is shown in King voso, Epi voho, at
widely separated points. The }-h mutation from jose yields Buka hosse,
and Motu hode shows still another change. Of the closed stem, an attri-
tion form, we have vos in Laur and Lamassa, and Duffield's unspecified
New Ireland vocabulary seems to draw from the same region for the form
fiss. Makekula yields us bos, and Efate uos. The Aneityum ne-hev is a
metathesis of some such attrition form, veh. This makes it within the
limit of possibility that Tanna vea may be included.

I can see no plan whereby Visayas bugsai can account for its inner palatal
and therefore become identifiable, and the other Indonesian forms are
hopelessly remote.

The Semitic offers no resemblance at all.

364-
ura, lobster, prawn.

Samoa: u la, lobster, prawn, shrimp, crayfish. Futuna: ida, lobster.
Rapanui, Nukuoro : ura, prawn, lobster. Mangareva : ura, cray-
fish. Maori: koura, id. Paumotu: cf. koranihi, prawn.
Marquesas: koua, crayfish. Tahiti: oura, prawn. Tonga:
uo, lobster. Niue: uo, crayfish.

Viti: ura, prawn.

Mota, Motu: ura, crayfish. Malekula: na-uru, id. Bierian:
ni-ua, id. Baki: Uo, id.

Java: urang, crayfish. Malagasy: urana, id. (uranurana, eating
greedily). Malay: udang, crayfish.

Hebrew: hawar, to be white, to become pale. Arabic: hara, to
be bleached, to eat greedily; hawar', haw, red leather.

Nuclear Polynesia has the simple stem ula; the languages of the Tonga-
fiti migration have a composite koula, except that Mangareva has the
Proto-Samoan form. Niue and Tonga have a secondary form uo, not to
be considered the ordinary elision of the liquid, for both these languages
make free use of /; this secondarv form is discoverable in Bierian.

DATA AND NOTES. 431

The Melanesian identifications are satisfactory, Baki ilo exhibiting the
same vowel change as in 362.

The three Indonesian identifications agree in the accretion of a nasal and
are satisfactory.

There is nothing in form and nothing in signification in this Semitic to
afford suspicion of relationship with ula.

365-
uti na, the penis.

Viti: uti, id.

New Guinea (Astrolabe Bay); uti, id.

Another root form is quite similar in the same sense and is adjoined for
comparison.

Samoa, Tonga, Niue, Nuguria, Hawaii: ule, the penis. Maori,

Tahiti, Mangareva, Paumotu: ure, id.
Motu: use, id. New Ireland (Port Praslin) : usik, id. New Ireland

(Carteret Harbor) : ausidi, id. Pala: «*, id. Lamotrek : ul, id .
Arabic: 'uss', id.

We have here forms in /, s, t. In the table of mutations we find very
scanty evidence of l-t mutation, but none at all of l-s. Starting from s we
find very scanty evidence of s-t mutation, none at all of s-l (s-r resting on
one unsatisfactory instance). Starting from t we find considerable proof
of t-s mutation, and, while /-/ does not appear, the similar t-r is found in a
few cases (see note 258).

It is, therefore, just possible if we regard uti as the parent form, to make
a plea for the derivation of use and ule therefrom. Apart from the fact
that our evidence in support is scanty, we find yet another difficulty. We
should be put in the position of finding in the Proto-Samoan a secondary
form, yet in all our material we have found that to be the primary form and
the Melanesian secondary or loan material. We are not warranted, there-
fore, in accepting this identification.

Not having anything in particular with which to identify it, the mere
presence of the Semitic here offered cuts no figure.

APPENDIX II.
THE SOUTHERN GATEWAY.

After the compilation of data and completion of the discussion thereof
in the foregoing work a few additional vocabularies became accessible.
The data collated therefrom have been incorporated in the proper places
in Appendix I, but it was found impracticable to make use thereof in the
several items of discussion. To remedy this omission the necessary check-
lists are here offered, together with such investigation of this added material
as seems advisable or likely to prove of interest.

Two of the items of additional material belong on the Polynesian side
of this inquiry, the vocabularies of Nuguria and Rapanui, one the most
distant of the islands of the Polynesian verge measured backward along
the migration track, the other the last outpost of Polynesian migration in
the Pacific. The following check-lists will afford access to this material.
The use of bold-face type serves to distinguish the vocables which are exclu-
sively Proto-Samoan, those noted in the common type being vocables
common to the two migration streams ; a single instance in the Rapanui
list is set off by italic numerals as exclusively Tongafiti.

Nuguria. 54 58 61 62 63 64 72 73 78 81 89 91 98 103

106 113 123 125 126 140 141 142 146 147 149 150 151 153

154 156 158 159 164 167 168 173 187 190 198 201 206 208

209 211 212 215 216 217 219 223 237 239 250 252 253 256
259 261 263 266 269 271 272 274 275 276 277 278 279 281
282 283 285 286 290 291 292 294 295 297 298 300 301 307
308 309 312 315 316 317 318 319 321 322 323 324 325 327
328 329 330 333 335 336 338 339 340 342 343 344 345 346
350 352 354 358 360 363 365

Rapanui. 24 54 58 61 64 73 75 76 80 81 94 103 106 107

113 118 119 120 122 125 126 128 132 133 135 137 141 142

143 144 146 148 149 151 153 154 158 159 160 164 166 167

168 171 183 184 187 196 197 198 201 205 206 207 208 209

210 211 212 213 214 215 216 218 222 228 229 230 233 234
238 239 240 241 253 256 258 259 261 263 264 266 267 269
270 271 272 273 274 275 276 278 281 282 284 285 287 288
289 290 291 292 293 295 296 297 298 300 301 302 304 305
306 307 308 312 313 316 317 318 319 320 321 322 323 324
325 326 327 330 332 333 335 336 338 339 342 343 344 346
347 35o 35i 352 353 354 355 356 358 359 360 361 363 364

We shall next examine each of these Polynesian languages for the dis-
covery of their mutations from the standard of the Proto-Samoan.

The vocabulary of Nuguria, collected by Dr. Thilenius, is probably not
standardized as would have been the case if we owed it to some resident
collector. Accordingly we may attach no particular importance to the
fact that / and r seem to be used with equal frequency, that i-f is almost
as frequent as f-h. In these lists it has been deemed unnecessary to indi-
cate the data upon which rest the more positively established mutations.

433

434 THE POLYNESIAN WANDERINGS.

a-o 61, 256

h— 206, 215, 278

e-i 62

o-a 201

u-i

s-s ng-n
s-h n-»g 259

Aspiration

t-h

f-v 91
assumed 159,

kg 113
k— 156

275-

PP
p-b

!85

The Rapanui speech diverges from Proto-Samoan in the loss of the true
aspirate, the mutation of the sibilant to an aspirate, and the mutation of
f to h . The irregular variants are as follows :

a-o 61, 233 i-u 321 o-a 137 u-u 125

lr I-/ 141 l-«. 154 1 — 103, 128 u-o 168

ng-»g ng-w 125 m-p 141

Aspiration assumed 158, 270.

Rapanui differs markedly in its speech from other languages of eastern
Polynesia. It carries a far larger element of distinctly Proto-Samoan mate-
rial; at the same time it has also a distinctly Tongafiti element, only one
item of the latter, however, being involved in the material here assembled.
The tradition is clear that Rapanui was peopled by a migrant race, said
to have come from a land in the east where periodical droughts killed the
vegetation, and since the time of the settler king there have been 56
successors. The eastern home is incomprehensible; nothing intervenes
between Rapanui and the coast of Chili and Peru in that direction, and
nothing in the ethnography of the races there found can be brought into
association with the Polynesians of this outlying colony of the South Sea.
Nor does this tradition fall into accord with the belief, supported by such
evidence as may exist in geographical names, that Rapanui, Great Rapa,
was settled from Rapaiti or Rapa the Less. The date of settlement by
reduction of the royal line counting four to the century is established at
500 a. d. This antedates by a considerable period the arrival of the Tonga-
fiti migration in the Pacific. The Tongafiti speech element is proof that
the Rapanui settlement must have followed the expulsion of the Tongafiti
from Nuclear Polynesia, an event which we have satisfactorily established
as synchronous with the Norman Conquest. The traditional date, there-
fore, is wholly untrustworthy. The presence of the remarkably large Proto-
Samoan content in the speech is clear evidence that in the place whence
the Rapanui colonists emigrated the two migration swarms were inter-
mingled in peaceful association. No island is now discoverable where the
language contains the like proportions of distinctively Proto-Samoan and
distinctively Tongafiti elements. The only conclusion permissible is that
Rapanui was peopled by a migration in which the two race elements were
evenly mixed, therefore a secondary migration from some point of contact
of the two migrations.

The remainder of this added material bears with considerable insistence
upon the station of Motu, which in the foregoing discussion has served as
the ultimate point in establishment of the Viti Stream of Proto-Samoan
migration. It is, accordingly, well worth the apparatus of check-lists and
phonetic tables here assembled, the material being associable with that
presented and discussed in Chapter VII.

THE SOUTHERN GATEWAY. 435

Awalama 46 74 190 199 217 239 247 254 258 259 265 272 276 290

297 298 300 306 317 321 324 328 330 336 339 340 350 300

361 363

Boniki . . 46 74 187 190 199 217 225 239 249 254 259 265 272 273

290 300 302 312 313 321 324 328 339 344 35o 35i 361 303

Dobu... 46 47 74 190 199 207 214 217 239 249 258 276 278 285

290 297 300 302 306 308 309 313 317 3i8 321 324 328 329

330 339 342 344 35o 360
Domara.312

Doura... 74 217 249 259 265 272 284 291 306 309 312 313 318 324

330 342 350 360

Duba 324

Galavi... 46 74 149 *99 217 225 239 249 254 259 265 272 273 282

290 297 312 313 321 324 328 339 340 344 35o 360 361 363

Galoma. . 46 47 74 109 122 149 187 190 206 217 247 249 254 265

276 284 285 290 291 292 294 298 306 308 312 3*3 3'7 321

324 328 330 334 335 346 351 360

Hula.... 46 47 74 109 122 149 187 190 214 217 247 249 265 276

290 291 292 294 298 306 308 312 313 317 321 324 328 330

334 335 34° 34^ 349 35 * 360

Kabadi.. 46 47 59 74 122 149 214 247 249 254 258 259 265 266

272 290 291 292 294 298 300 308 309 312 313 317 321 324

330 335 336 350 35i 36o 361 363

Keakalo.312 338

Keapara. 46 47 74 122 149 187 190 206 214 217 247 249 254 265

272 284 285 290 291 292 294 297 298 306 308 312 313 317

321 324 328 330 334 335 346 349 35* 360

Kiriwina 47 74 109 149 187 190 212 214 217 225 239 273 276 285

290 291 294 300 302 306 308 309 312 313 3^7 3i8 321 324

328 329 330 339 340 344 350 361

Kiviri... 59 74 190 199 217 225 249 278 282 285 297 300 306 312

313 321 324 328 330 339 340 344 35o 35i 361 363
Kiwai . . . 340
Koita. . .217
Koko-Yimidir 34°

Kubiri.. 74 217 225 249 278 282 300 302 306 312 3*3 3l6 3*7 321

324 328 330 339 340 344 35i 360 361 363
Kunini. .361

Kwagila.190 291 297 313 324 330 350

Mabuiag. 46 76 218 340 344 361
Mailu. . .312
Masig. . .361

Massim..26S 291 318

Mekeo... 46 74 109 190 206 207 214 217 247 254 259 272 276 290

292 294 306 309 312 313 318 321 324 328 335 336 339 342

350 360 363

Miriam.. 64 76 3^i (Murray Island, page 99)

Misima.. 47 109 190 199 217 247 259 291 297 308 312 321 324 328

330 342 350 35i 361

Mugula..i90 199 212 247 290 291 294 298 313 324 33° 34© 344 350

363

Mukawa. 46 74 19° *99 2o6 217 225 239 247 249 259 273 276 285

298 300 302 306 312 313 316 317 324 328 330 335 336 339

34° 344 350 35i 36o 361 363

Murua...i49 190 199 217 239 247 273 276 285 290 291 300 302 30b

308 309 312 317 3iS 321 324 328 339 350 361
Nada....io9 149 190 199 212 217 225 239 247 273 276 290 291 302

306 308 309 312 313 3i7 3i8 324 328 330 333 339 340 342

344 35° 35 l 3&l

Oiun 59 74 149 187 190 199 217 225 249 258 282 297 298 306

313 317 321 324 336 339 340 344 350 361

Panaieti. 47 74 109 190 214 217 225 239 247 249 259 291 292 308

309 312 313 32i 324 328 330 350 361

Pokau... 46 59 74 109 122 149 190 206 207 214 217 247 249 258

259 265 273 284 285 291 292 298 306 308 309 3X2 3i3 3i7

321 324 330 339 340 346 350 351 360 361 363

436 THE POLYNESIAN WANDERINGS.

Raqa 74 190 199 217 225 249 282 298 317 321 324 328 336 339

340 344 350 361 363

Roro. . . . 46 74 122 190 206 214 217 239 247 249 254 259 272 276

290 291 292 298 300 306 308 309 312 313 321 324 328 335

336 339 346 350 361 363

Rubi. . . . 46 109 187 190 199 217 247 249 265 276 278 284 285 290

292 298 306 312 313 317 321 324 328 334 339 349 35* 360

Sariba... 46 74 109 149 190 199 212 217 239 247 249 258 265 272

276 278 285 290 291 294 298 313 321 324 328 329 330 339

340 361 363

Sinaugoro 46 47 74 109 122 187 190 214 217 247 249 265 276 285

290 291 292 298 306 312 313 317 318 321 324 328 330 333

334 335 338 346 349 35*

Suau. ... 46 74 149 190 199 207 212 217 239 247 249 258 265 276

278 290 291 292 294 298 302 313 317 318 321 324 330 339

344 350 360 363

Tagula..i9o 212 217 247 258 259 294 302 312 313 324 336 361

Taupota. 46 74 190 199 217 225 239 254 259 265 272 282 290 291

297 298 300 302 306 312 313 317 321 324 328 330 336 339
340 344 350 360 361 363

Tavara. . 46 74 190 199 206 217 258 259 265 272 291 297 300 312

313 321 324 328 336 339 350 361

Tubetube 46 74 149 190 206 212 217 247 249 258 290 291 292 294

298 300 302 313 317 318 321 324 330 340 360 361

Uni 46 217 259 265 272 284 290 291 292 312 313 317 321 324

336 339 342 346 360 361

Wagawaga 74 217 290

Waima.. 47 308 312 321 335 344 346

Wedau.. 46 74 109 190 199 206 217 225 239 247 254 258 259 265

272 276 282 290 291 297 300 302 306 308 313 317 321 324

328 336 338 339 340 344 350 351 360 361 363

Yaraikana 340

THE SOUTHERN GATEWAY. 437

We shall next develop the phonetic variety of these languages in a
series of tables such as we have employed for other Melanesian speech
in Chapter VII.

AWALAMA.

a a, i, o
e e oo

i i, a, e, o u u, o

I I, r, n
ng n, g, m n n, g, h mm

h t, -
s h, —

v g
f v, w
k h, — t t, d, h, — p p, b

a-a 46, 74, 199, 217, 239, 258, k-h 297

265, 276, 290, 300, 317, 324, k— 46, 258, 306

339. 35°. 360 n-w 46, 74, 259, 290, 317, 330

a-i 328 n-g 254

a-o 199, 258, 291 n-^ 328

e-e 190, 276, 290, 363 h-t 339

\-i 46, 259, 290, 291, 298, 300, h— 363

350 s-h 239

i-a 254 s — 298

ie 254, 330 t-t 74, 199, 258, 306, 324, 350

\-o 254 t-d 276

0-0 247, 259, 336, 363 t-h 247

u-m 74. 239, 247, 258, 265, 298, t— 217

306, 317, 321, 330, 360 m-w 217, 258, 265, 317, 321, 324,

u-o 328 328

\-l 339 \-g 291

\-r 297 f-v 363

\-n 265, 336, 350 f-w 254, 259, 290, 360

ng-n 336 VP 190

ng-g 350 p-b 247

ng-w 199

Identical 3J7, 324

Vowel identity 46, 217, 265

Consonant mutation :

Vowel identity 239, 259, 276, 290, 306, 350, 363

Frontal abrasion 46, 306, 321

Frontal accretion 190, 239, 298, 360

Terminal accretion 190, 247, 254, 300, 321, 330, 336, 339

Metathesis 36°

438 THE POLYNESIAN WANDERINGS.

BONIKI.

a a, e, i, o

e e, a, u o o, e
i i, a, o U a, o

ng n, m n n, k, — m m

1 /, n, —
h

s s, —

V

f v, iv, p, k
k k t t, m p 6

a-a 46, 74, 199, 217, 239, 265, in 265

290, 300, 312, 324, 328, 351 I— 312, 350

a-e 350 ng-w 350

a-» 350 ng-m 199

a-o 199 k-^ 46, 225, 249, 351

e-e 190,290,363 n-n '46, 74,187,290,328,351

e-a 190, 249, 272 n-k 254

e-w 187 n— 259

i-i 46, 249, 273, 290, 300, 312, s-.y 344

350 s— 239

i-a 254 t-t 74, 199, 249, 302, 324, 350

\-o 259 t-w 217

0-0 187, 363 mm 217, 265, 312, 321, 324, 328

0-e 259 i-v 254, 273, 290,^363

u-w 74, 239, 265, 272, 273, 302, i-w 290

321, 344, 350 i-p 193

u-o 328 i-k 259

1-/ 225, 272 V-b 193

Identical 46, 74. 324, 344, 35 1

Consonant identity. . . .249, 328

Vowel identity 217, 239, 265, 273

Consonant mutation :

Vowel identity 290, 363

Frontal abrasion 259, 321

Frontal accretion 187, 239, 273, 302

Terminal accretion 217, 225, 249, 272, 300, 302, 321

THE SOUTHERN GATEWAY.

439

DOBU.

a a, e, o
e e, a o o, e

i i u u, a

1 /, r, n, —

ng n, g, m, — n n m m

h s

s s

V

f w, s, —
k g, s, — it, d, s, — p p, b

a-a 46, 74, 199, 207, 214, 239, ng~ 285

258, 276, 278, 290, 300, 308, k-<7 302

313, 3i7, 318, 324, 342, 360 k-s 297

&e 309, 328, 350 k— 46, 249, 258, 306

a-o 302 n-n 74, 290, 317, 318, 321, 324,

e-e 190, 276, 290 328

e-a 47, 249, 318 hs 47, 278

i-i 46, 47, 249, 285, 290, 300, s-j 239, 342, 344

302, 308, 313, 329, 330, 342 t-t 47, 74, 199, 207, 249, 258,

0-0 285 276, 306, 324, 350

O-e 309 t-d 329

u-u 74, 207, 239, 258, 278, 306, t-s 318

317, 321, 329, 330, 344, 360 t— 302

u-a 328 mm 258, 313, 317, 318, 321, 324,

1-/ 308 328

\-r 297 f-w 290

\-n 309, 313 f-s 290

1— 350 f— 214, 290, 329, 360

ng-n 309, 350 p-/> 190

ng-<7 3°8 P ^ 2°7» 285

ng-w 199

Identical 74, 190, 239, 276, 317, 324, 330, 342, 344

Consonant identity. . . .328

Vowel identity 46, 329, 360

Consonant mutation :

Vowel identity 199, 207, 278, 290, 306, 313

Frontal abrasion 46, 214, 249, 306

Frontal accession 46

Terminal accession 190, 278, 300, 317, 342

Metathesis 308

440 THE POLYNESIAN WANDERINGS.

DOMARA.

a-a 312 1 — 312

i-i 312 m-w 312

DOURA.

a a, e
e a o e

[ i XX u

1 I, r, i, —
ng — n n, — m m

h
s d

V V

f h, k
k— t k p h

a-a 74, 217, 265, 284, 309, 312, ng— 309, 350

3i3» 3i8, 324, 342, 350, 360 k— 249, 306

a-e 291 n-n 74. 33°. 342

e-a 249, 272, 318 n — 259

i-i 249, 259, 291, 312, 313, 330, s-d 342

342, 350 t-k 74, 217, 249, 306, 318, 324,

o-e 259, 309 35o

u-« 74,265,272,284,306,330, m-vi 217,265,312,313,318,324

360 v-v 291

1-/ 309 f-A 272, 360

\-r 265, 272 f-k 259

|-t 284, 309 V-h 284
1— 312, 313, 350

Identical 265, 330

Consonant mutation :

Vowel identity 74. 217, 3°6, 318, 324, 342, 360

Frontal abrasion 249, 259, 306

Terminal accretion 249

DUBA.

a-a 324 t-* 324

mm 324

THE SOUTHERN GATEWAY. 441

GAL AVI.

a a, e, i, o
e a, e o o, e

i i, a, o U u, o

1 /, r, n, —

ng n, m n n, k, — mm

h g

s s

V

f v, w, p, k,

k k, g t t, m p

a-a 46, 74, 199, 217, 239, 265, 1 — 313, 350

290, 312, 313, 324, 328, 340, ng-n 350

360, 363 ng-w 199

a-e 350 k-k 46, 149, 249

a-i 350 k-^ 225, 297

a-o 199 n-w 46, 74, 290, 328

e-e 290, 363 n-k 254

e-a 249, 272 n — 259

\-i 46, 149, 225, 249, 273, 290, h-g 340

312, 3i3> 350 S"J 239, 344

[-a 254 t-t 74, 199, 249, 324, 350

i-o 259 t-m 217

0-0 282 m-m 217, 265, 312, 313, 321, 324,

o-e 259 328, 34°

u-w 74, 239, 265, 272, 273, 282, f-v 254, 273, 282, 290

321, 344, 360 i-w 290, 360, 363

no 328 " f-p 272

1-/ 272 f-fe 259

1-r 225, 297 f— 360
\-n 265, 312, 313

Identical 46, 74. J49» 239. 324» 344

Consonant identity. . . .249, 328

Vowel identity 217, 265, 273, 360

Consonant mutation :

Vowel identity 225, 290, 312, 313, 363

Frontal abrasion 321

Frontal accretion 149, 239, 273, 360

Terminal accretion 149. 217, 225, 249, 254, 272, 321, 340

Metathesis 360

442

THE POLYNESIAN WANDERINGS.

GALOMA.

a a, e
e e, a o o

i i u u, i, o

1 /, r, g, n, —
ng g n n, g m m

h r
S r

\ b, w

f />, &
kg,— t /, — p £, 6

a-a 46, 74, 109, 217, 254, 265, ng-^ 285, 306, 346

290, 291, 292, 294, 308, 312, k-g 46, 149

313, 3i7, 324, 334, 335, 346, k— 249, 306

351, 360 n-n 46, 74, 187, 290, 292, 317,

a-e 284, 328 321, 328, 33°, 334, 35*

e-e 122, 187, 190, 276, 290 n-g 109, 254

e-a 47, 249 h-r 47, 206

\-i 46, 74, 149, 206, 249, 254, s-r 298

285, 290, 291 t-l 247

0-0 187, 206, 247, 285 t— 47, 74, 217, 249, 276, 294,

u-w 74, i°9, 247, 265, 284, 292, 306, 346

294, 298, 306, 317, 321, 330, mm 217, 254, 265, 312, 313, 317,

334, 335, 360 324, 328

u-i 351 v-6 109

u-o 328 v-w 291

M 308, 334, 335 f'b J22, 206, 29O, 292, 294, 36O,

\-r 149 f-p 254

\-g 308 V-P !9°, 285

\-n 265 p-b 247, 284, 285

1— 284, 312, 313

Identical 187, 190, 317, 321, 330, 334

Vowel identity 217, 265, 334

Consonant mutation :

Vowel identity 46, 74, 109, 122, 206, 254, 285, 290, 292, 308,

346, 360

Terminal abrasion 324

Frontal abrasion 47, 249, 306

Frontal accretion 276

Terminal accretion 321

Metathesis 149, 334

THE SOUTHERN GATEWAY.

443

HULA.

a a, e

e e, a o o

i i, e u u, i, o

1 /, r, g, n, —

ng g n n, k, — m m
h r
S r

V b, w

f b, v

k k, g, — t h, i, k, r, — p p, b

a-a 46, 74, 109, 214, 217, 265, k-k 149, 351

276, 290, 291, 292, 294, 308, kg 249, 306

312, 313, 317, 324, 334, 335, k— 46

346, 349, 351, 360 n-11 46, 74, 187, 290, 292, 317,

a-e 328 321, 328, 330, 334, 351

e-g 122, 187, 190, 276, 290, 349 n-fe 254

e-a 47, 249 n — 109

\-i 46, 47, 149, 249, 290, 291, h-r 47

298, 308, 312, 313, 321, 330, s-r 298

346 t-h 74

i-e 254 t-i 276

0-0 187, 247 t-k 349

U-tt 74, 109, 247, 265, 292, 294, t-r 347

298, 306, 317, 321, 330, 334, t— 47, 217, 249, 294, 346

335, 360 m-m 217, 265, 312, 313, 317, 324,

u-t 351 328

u-o 328 v-6 109

\l 335, 349 v"™ 291

\-r 149 i-b 360

\-g 313 i-v 122, 214, 254, 290, 292, 294

1-n 265, 334 V-P i9°

1— 308, 312 p-6 247
ng-9 308, 346

Identical 190, 330, 335

Consonant identity . . . .328

Vowel identity 46, 109, 187, 217, 265

Consonant mutation :

Vowel identity 74, 122, 214, 290, 291, 308, 313, 334, 346

Terminal abrasion 306, 324

Frontal abrasion 46, 47, 109, 308

Frontal accretion 187

Metathesis 149, 321

444 THE POLYNESIAN WANDERINGS.

KABADI.

a -a, e
e e, a o o, e

i i, e u u, o

1 r, —

ng «, — n n m w

h d

s ;

V V

f v, s, —
k — t k, s p p

a-a 46, 74, 214, 258, 265, 290, 1— 309, 312, 313, 336, 350
292, 294, 300, 308, 309, 312, ng-n 308, 336, 350

3i3, 3i7, 324. 335, 35o, 35i. nS— 309

360 k— 46, 249, 258

a-e 258, 291 n-n 46, 74, 259, 290, 292, 317,

e-e 122, 290, 363 321, 330, 351

e-a 47, 122, 249, 272 h-d 47, 363

i-i 46, 47, 122, 259, 290, 291, s-t 298

298, 308, 312, 313, 321, 330, t-k 47, 74, 247, 258, 294, 324,

350 35°

i-e 300 t-s 249

0-0 247, 309, 336, 363 mm 258, 265, 312, 313, 317, 324

o-e 259 v-v 122, 291

u-w 74, 247, 258, 265, 272, 292, f-v 122, 214, 290, 292, 294, 360

294, 298, 317, 321, 330, 335, f-s 259

35i f— 272, 363

VL-0 360 pp 247

\-r 265, 272, 308, 335

Identical 265, 317, 321, 330, 335

Consonant identity. . . .122

Vowel identity 46

Consonant mutation:

Vowel identity 74, 122, 214, 290, 292, 298, 308, 324

Frontal abrasion 46, 249, 336

Frontal accretion 300

Terminal accretion 249, 272, 294, 300, 321, 335, 336

Metathesis 360

KEAKALO.

a-a 312, 338 1— 312

e-i 338 k-a 338

i-i 312 m-m 312

THE SOUTHERN GATEWAY. 445

KEAPARA.

a a, e
e e, a o o
\ i u u, i, o

1 I, r, g, n, —
ng g n n, g mm

h r
s /

v w

f v, b, g
k k, g, h t I, — p p, b

a-a 46, 74, 214, 217, 254, 265, ng-# 285, 308, 346

290, 291, 292, 294, 308, 312, k-k 149

313, 317, 324, 334, 335, 346, kg 249, 297, 306

349, 351, 360 k-h 46, 214

a-e 284, 328 ^ n-w 46, 74, 187, 290, 292, 317,

e-e 122, 187, 190, 290, 349 321, 328, 330, 334

e-a 47, 249, 272 n-g 254

\-i 46, 47, 149, 206, 249, 254, h-r 47, 206

285, 290, 291, 298, 308, 312, s-l 298

313, 321, 330, 346 t-l 247

0-0 187, 206, 247, 285 t— 47, 74, 217, 249, 294, 346,

U-w 74, 247, 265, 272, 284, 292, 349

294, 298, 306, 317, 321, 330, mm 217, 254, 265, 312, 313, 317,

334. 335, 360 324, 328

n-i 351 v-w 291

u-o 328 f-v 122, 214, 254, 290, 292, 294

1-J 335, 349 f-& 272, 360

\-r 149 i-g 206

\-g 313 V-P 19°. 285

\-n 265, 334 p-b 247, 284
1 — 284, 297, 308, 312

Identical 187, 190, 317, 321, 330, 335

Consonant identity. . . .328

Vowel identity 217, 265

Consonant mutation:

Vowel identity 46, 74, 122, 214, 254, 290, 291, 298, 308, 313,

334, 346, 360

Terminal abrasion 306, 324

Frontal abrasion 47, 308

Terminal accretion 321

Metathesis 149

446 THE POLYNESIAN WANDERINGS.

KIRIWINA.

a a, e, i
e e, a, i o o, a, u
\ i u u

1 /, r, n, —
ng g n n, I m m

h s
s s, n

v w

f v, k, g, d, —
k k, kw, g t t, d, s, k p b

a-a 74, 214, 217, 276, 291, 294, k-k 149, 214, 225, 297, 306

302, 308, 309, 312, 313, 317, k-kw 225

318, 324 k-g 302

a-g 300 n-n 187, 317, 328, 330

a-i 290, 309, 324, 328, 350 n-/ 74

e-e 190, 297 h-s 47

e-a 47, 187, 190, 276, 290, 318 s-.? 239

e-z 318 s-n 344

i-i 47, 149, 273, 285, 290, 291, t-t 47, 74, 217, 276, 306, 318,

297, 300, 302, 308, 312, 313, 324, 350

329, 330, 350 t-d 302

00 285, 309 t-s 318, 329

o-a 187 t-k 294

o-u 309 m-m 217, 312, 313, 317, 318, 321,

u-w 47, 149, 273, 285, 290, 291, 324, 328

297, 300, 302, 308, 312, 313, \-w 291

329. 33°. 35° f-v 214, 273, 290

1-/ 212, 290, 308, 312, 313, 350 i-g 294

\-r 225, 297 f-& 212

I-w 309 i-d 294

1 — 291 f — 329

ng-£ 285, 308, 309, 350 p-6 190, 285

Identical 149, 217, 297, 306, 312, 317, 324

Consonant identity. . . .318, 328

Vowel identity 273, 329, 344

Consonant mutation :

Vowel identity 74, 212, 214, 285, 291, 302

Terminal abrasion 328

Frontal accretion 187, 273, 276

Terminal accretion 149, 300, 309, 317, 330

Metathesis 308, 313, 350

THE SOUTHERN GATEWAY.

447

KIVIRI.

a a, e, o, u

e e, a, too
\ i u u, a

1 r, n, —

ng n, m n n m m

h s, r

s .$•

V

f b

k k, g, s, — t t p p, f

a-a 74, 199, 217, 278, 313, 324, ng-w 199, 285

328, 340, 350 k-k 350

a-e 350 kg 225, 306

a-o 199 k-y 297

a-w 300 k — 249

e-e 190 n-n 74, 328, 350

e-a 249 h-^- 278

Q-i 363 h-r 340

\i 249, 300, 312, 313, 350 s-^ 344

0-0 285,363 t-t 74, 199, 2 1 7, 249, 306, 324,

u-w 74, 278, 306, 344 35°

u-a 306 mm 217, 312, 313, 324, 328, 340

\-r 225, 297 f-b 282, 363

1-n 312, 313 V-P 285

1— 350 Vf i9°

ng-n 350

Identical 74, 217, 324, 344

Terminal abrasion 350

Vowel identity 350

Consonant mutation :

Vowel identity 190, 278, 313, 340, 363

Terminal abrasion. . .285, 312, 328

Terminal abrasion 285, 312, 328, 350

Terminal accretion 190, 278

a-a 217

KOITA.
mm 217

t-m 2 1 7

448

THE POLYNESIAN WANDERINGS.

KUBIRI.

a-a

a-e

e-e

e-a

\-i

o-o

o-a

u-u

u-o
\-r

\-n

e e,

a a,

a

o,

u, o

ng

r, n

n, m
s, r

m m

b, —

k k, g, — t t

74, 217, 278, 3°°, 302, 312, k-k

3I3. 3!7, 324> 328, 340, 351, k-g

360 k —

316 n-n

363 n-m

249 h-s

249, 300, 302, 312, 313 h-r

363 S"J

282 U

74, 278, 282, 306, 316, 317, mm

344, 35i, 360

328 f-6

225, 316 f—

312, 313

Identical 74, 217, 324, 344, 351 Frontal abrasion 249,' 306

Consonant identity 316, 328 Terminal accretion . 249, 278, 300,

Consonant mutation : 302, 34°
Vowel identity.,278,306,312,313,363 Metathesis 302

35i

225, 302
249, 306

74, 328, 351
3J7
278

340

344

74, 217, 249, 302, 306, 324

217, 312, 313, 316, 317, 324,
328, 340
282, 363
360

KWAGILA.

a a, 0
e i o

\ i u u

1 r, -
ng 11 n d in m

h
s

v g
f
k k it p b

a-a 313, 324,

350

a-o 291

e-i 1 90

\-% 291, 313,

330, 350

u-u 330

\-r 297

1— 313, 350

Identical

-124.

Vowel identity .

350

ng-n

3.50

k-k

297

n-d

330

U

324,

350

mm

3i3,

324

\-g

291

V-b

190

Frontal accretion 190

Terminal accretion. . . . 190, 330

THE SOUTHERN GATEWAY. 449

MAILU.

a-a 312 1 — 312

\-i 312 m-m 312

MASSIM.

a-a 268, 291, 318 mm 318

e-i 318 t-t 268, 318

\-i 291 \-w 291

[r 268

MEKEO.

a a

e e, a o 0, e, u

\ i u u

1 *, -
ng ng, n n w, ng, — mm

h k

s k

v £

f /, A -
k— t k, — v 1

a-a 46, 74, 109, 207, 214, 217, k — 46

254, 276, 290, 292, 294, 309, n-n 259, 321, 342

312, 313, 318, 342, 350, 360 n-ng 46, 74, 254, 292
e-e 190, 276, 290, 318, 363 n — 109, 290

e-a 272 h-k 206, 363

\-i 46, 206, 254, 259, 290, 312, s-k 342

313, 321, 342, 350 t-k 247

0-0 246, 247, 336, 363 t— 74, 207, 217, 276, 294, 318,

o-e 259 35°

o-w 309 m-m 217, 254, 312, 313, 318

u-w 74, 109, 207, 247, 272, 292, \-p 109

294, 321, 360 f-/ 294

1-/ 336 f-p 206, 214, 254, 272, 290, 292,

1— 309. 312, 313. 350 360, 363

ng-na 309, 336 f— 259

ng-w 350 p-/ 190, 207, 247

Identical 321, 336

Vowel identity 46, 109, 217, 276, 318

Consonant mutation:

Vowel identity 190, 206, 207, 214, 254, 290, 342, 360

Frontal abrasion 46, 109

Frontal accretion 272

Terminal accretion 74, 254

450

THE POLYNESIAN WANDERINGS.

MABUIAG.
a a

1 i

ng

a-a 46, 218

i-i 46

U-w 344

u-a 76

u-a* 76

Identical 217, 344

Vowel identity 46

Frontal abrasion 46

1

n

h

s s

k—

mm
P-P

u z/, a, at
m m

v
f

P P

46

344

76

218

MIRIAM.
a

e i

"g

e-i

64

U-tt

64

u-a

76

u-e

76

Consonant mutation :

Vowel identity 64

1 /

n

h

s

M 64
mm 76

p-b 64

U w, a, e
m w

v
f

P b

THE SOUTHERN GATEWAY. 451

MISIMA.

a a, e, i, u
e e, a, too
\ i u it

1 I, n, —
ng n, m n n, I, k, — m m

h r
s s, k

v w, r
f
k k, g t /, n p p, b

a-a 109, 199, 217, 308, 312, 324, k-k 351

342, 350 kg 297

ZL-e 291 n-n 259, 330, 342, 351

at 328 n-Z 342

a-w 351 n-k 328

e-e 190 n — 109

e-a 47 h-r 47

e-i 190 s-j 342

i-t" 47, 259, 291, 312, 330, 342 s-h 342

0-0 247 t-t 47. i99» 217, 324. 35°

u-w 109, 247, 328, 330 t-n 217

U 308 mm 217, 312, 324, 328

\-n 312 v-w 291

I — 297, 350 v-r 109

ng-n 308, 350 p-/> 247

ng-m 199 p-6 190

Identical 217, 324, 342

Vowel identity 109, 330

Consonant mutation :

Vowel identity 190, 199, 308, 312

Terminal abrasion 259, 308

Frontal abrasion 109

Metathesis 247

452 THE POLYNESIAN WANDERINGS.

MUGULA.

a a, e, o
e e, a, i o o, i
\ i u u, oa

ng n, m n n

h s

1 I, n, ■

n n

h *

s .y, sh

m] m

v 9

f v,\w, s, k
t t, k, h p />, b

a-a 199, 3i3. 324 nn 29°' 330

a-e 294, 350 h-s 363

a-o 291 *-* 344

e-g 190 s-^/i 298

e-a 290 t-i 199. 324, 350

e-i 190, 363 i'k 294

i-i 290, 291, 298, 313, 330, 350 t-h 247

0-0 363 m-w 3i3, 324

o-i 247 v-^ 291

U-W 212, 294, 298, 33O, 344 f-V 294

u-oa 247 f-w 363

1-/ 212 f-tf 290

\-n 313 f-fe 2I2

1— 350 V-P 247

ng-w 350 P-b 19°
ng-ra 1 99

Identical 324. 33°. 344

Vowel identity 350

Consonant mutation :

Vowel identity 190, 199, 212, 298, 313

THE SOUTHERN GATEWAY. 453

MUKAWA.

a. a, e, o
e e, a o o, u

\ i u u, o

1 r, n, —
ng n, m n n, k mm

h s, g
s s, t, —

v

f p, b, w, s
k k t t, s p p, b

a-a 46, 74, 199, 217, 239, 276, k-k 46, 249, 306, 351

290, 300, 312, 313, 317, 324, n-n 46, 74, 259, 317, 328, 336,

328, 335. 340, 350» 35i. 360, 351

a-e 316 n-fe 290

a-o 199 h-j- 206

e-e 190, 276, 290, 363 h-^ 340

e-a 249 s-j- 239, 344

i-i 46, 206, 249, 259, 273, 290, s-t 351

298, 300, 312, 313, 330, 350 s— 298

0-0 259, 285, 336, 363 U 74, 199, 217, 249, 276, 302,

o-M 206, 247 3°6, 324. 35°

U-w 74. 239. 247, 273. 298. 3°2, t-j 247

306, 316, 317, 330, 335, 344. «nw 217, 312, 313, 316, 317, 324.

35i. 360 328, 340

U-o 328 f-& 259, 273, 363

\r 316, 335 f-P 2o6

1-n 312, 313, 336 f-w 290, 360

I— 350 f-J 290

ng-n 336, 350 V-P 247, 285

ng-m 199, 285 P-b 19°

Identical 46, 74, 217. 239, 276, 306, 317, 324, 330, 335,

344. 35i
Consonant identity. . . .249, 316, 328

Vowel identity 273, 350

Consonant mutation:

Vowel identity 190, 259, 312, 313

Frontal accretion 190, 239, 273, 276, 298, 302

Terminal accretion 190, 249, 285, 302, 330, 336, 340

Metathesis 360

454

THE POLYNESIAN WANDERINGS.

MURUA.

a a, e, t, u
e e, i, u o o, au
i i, u u u, i, o

1 11, y, —
ng g, m n n, s m m

h
s s

v r
f v
k k, g t t, d, s p p, b

a-a 199, 217, 276, 291, 308, 317, 1 — 35°

318, 324, 350 ng-g 285, 308, 309, 350

a-<? 199, 302, 350 ng-w 199

a-i 300, 309 k-k 149, 306

a-w 328 ' kg 302

e-c 190, 290 n-w 290, 317

e-i 190, 318 n-s 328

e-w 276 »-s 239

i-t 149, 273, 290, 291, 300, 302, t-t 199, 217, 276, 306, 318, 324,

308, 312 35°

i-w 285 t-d 302

0-0 247, 285 t-s 247

o-au 309 m-w 217, 312, 317, 318, 324, 328

u-m 273, 3°6 xr 29r

u-o 247, 328 f-v 273, 290

u-i 306 V-P 247

1-n 309, 312 V-b 190, 285

1-y 308

Identical 217, 324

Consonant identity. . . .306, 318

Vowel identity 273

Consonant mutation :

Vowel identity 190, 290, 317

Terminal abrasion. . .312

Terminal abrasion 312, 317

Frontal accretion 273, 276

Terminal accretion 149, 199, 285, 300

Metathesis 149

THE SOUTHERN GATEWAY. 455

NADA.

a a, e, i, o
e e, a, i o o, u
i i, a u h. e, o

1 /, r, n, —
ng g, ii, m, — n n, I, s, — m m

h d
s J

v pw, r
f v, k, —
k k, kw, g t t, J, s p p, b

a-a 109, 217, 276, 291, 302, 308, ng-vi 199

309, 312, 313, 317, 318, 324, ng— 308

340, 342, 350 k-k 149, 306, 351

ae 199 k-kw 225

a-t 290, 309. 328, 350, 351 k-# 302

a-o 199 n-n 290, 317, 330, 351

Q-e 190 n-/ 342

e-a 190, 276, 290 n-s 328

e-i 318 n — 109

\-i 149, 225, 290, 291, 308, 312, h-d 340

313, 330, 342 8-J 239, 342, 344

i-a 273 t-/ 199, 217, 276, 306, 318, 324,

0-0 247 35°

o-w 309 t-d 302

u-w 109, 212, 273, 306, 317, 328, t-s 247

33°, 35i mm 217, 312, 313, 317, 318, 324,

u-e 344 328, 340

u-o 247 v-/>w 109

1-/ 212, 308, 312 v-r 291

\-r 225 f-"y 273

I-m 309, 313 i-h 212

1— 350 f— 290

ng-^ 309 P-/> 247

ng-?z 350 p-6 190

Identical 149, 217, 306, 312, 317, 324, 330

Consonant identity. . . .318, 344
Consonant mutation:

Vowel identity 212, 225, 313, 342

Terminal abrasion ... 302, 340

Terminal abrasion 302, 340

Frontal accretion 273

Terminal accretion 149, 199

Metathesis 109

456

THE POLYNESIAN WANDERINGS.

01 UN.

a a, e, o
e e, a, i o a
i i, u u u, a, i

1 r, -
ng n, m n n, m m m

h r
s n, —

v

f b
k k, g, s, — t * p /

a-a 74. r99» 2I7. 258, 290, 313, ng-m 199

3i7, 324. 340, 35o kk J49. 225

a-e 217 kg 306

a-o 199 k-j 297

e-g 190, 290, 363 k — 249, 258

e-a 187 n-n 74, 187, 290

e-i 249 n-m 317

i-i 149, 249, 290, 298, 350 h-r 340

i-tt 313 s-n 344

o-a 187, 336, 363 s— 298

Vk-u 74. 258, 298, 306, 317, 344 t-t 74, 199, 217, 249, 258, 306,

u-a 306 324. 350

U-t 344 m-m 217, 258, 313, 317, 324, 340

\-r 225, 297 f-6 282, 290, 363

1— 3i3. 350 V-f J90

ng-n 336, 350

Identical 74, 149, 324

Consonant identity • • • 217

Vowel identity 317

Consonant mutation:

Vowel identity 190, 340, 363

Terminal abrasion . . .290

Terminal abrasion 290, 350

Frontal abrasion 249, 336

Frontal accretion 149, 187, 225

Terminal accretion 187, 190, 249, 336

THE SOUTHERN GATEWAY. 457

PANAIETI.

a a, e, u

e e, a, i o o, u
\ i u u

1 r, n, —

ng n, g n n, ng, g, — mm

h r

s s

v w, r

i P,-
k k t t, h, n p p, b

a-a 74, 109, 214, 217, 239, 292, k-k 225, 249

308, 309, 312, 313, 324, 350 n-n 74, 259, 330

a-e 291 n-ng 292

a-w 328 n-£ 328

e-e 190, 249 n — 109

e-a 47 h-r 47

e-i 190 s-j 239

\-i 47, 249, 259, 312, 313, 330 t-t 47, 74, 217, 249, 324, 350

0-0 247 t-h 247

o-M 309 t-n 217

u-t< 74, 109, 239, 247, 292, 328, mm 217, 312, 313, 324, 328

330 v-r 109

\-r 225, 308 v-w 291

1-w 309, 312, 313 *-P 2I4» 292

1— 350 f— 214

ng-n 308, 350 p-/» 247

ng-^ 309 P-b l9°

Identical 74. 217, 239, 249, 324, 330

Consonant mutation :

Vowel identity 190, 214, 292, 308, 312, 313

Terminal abrasion 259, 308

Frontal accretion 239

Terminal accretion 324

Metathesis 247

458 THE POLYNESIAN WANDERINGS.

POKAU.

a a, e
e e, a, i o o, e
i i, e u u, o

1 I, < —

ng n, I, — n n, — m m

h d

s d

V V

f v, b, k
k k, — t k, d p b, v

a a 46, 74, 109, 207, 214, 217, ng-u 285, 308, 346

258, 265, 284, 292, 308, 309, ng-/ 309

312, 313, 317, 324, 339, 346, ng— 350

35«» 35ij 360 k-k 149

a-e 291 k — 46, 249, 258, 306

e-e 190, 363 n-w 46, 74, 259, 292, 317, 321,

e-a 249 330, 351

e-* 122 n — 109

\-i 46, 149, 206, 249, 259, 285, h-d 206, 339, 363

291, 298, 3O8, 312, 3I3, 321, Srd 351

330, 346, 350 t-k 74, 207, 217, 249, 258, 306,

i-« 273 324, 346, 350

0-0 206, 285, 309, 363 t-d 298

o-e 259 m-w 217, 258, 265, 312, 313, 317,

\x u 74, 109, 207, 258, 265, 273, 324

284, 292, 298, 306, 317, 321, x-v 109, 291

330, 360 i-v 122, 206, 273, 292, 360, 363

u-o 351 f-6 214

1-/ 265, 284, 308, 309, 339 i-k 259

\-i 284 p-b 190, 207, 285

1— 312, 313, 350 pv 284

Identical. . 149, 265, 317, 321, 330

Vowel identity 46, 109

Consonant mutation :

Vowel identity 74, 190, 206, 207, 214, 217, 284, 285, 292, 298,

306, 308, 309, 324, 339, 360, 363

Frontal abrasion 46, 109, 306

Frontal accretion 1 90

Terminal accretion 190, 249, 321

Metathesis 273

THE SOUTHERN GATEWAY. 459

RAQA.

a a, o
e e, i o o, a

i * u u, o

1 r, -
ng n, m n n, m m m, d

h r, I
s 11, —

V

f b

k k, ka, — t t, d p 6

a-a 74. i99» 217, 317, 324, 328, k-ka 225

340, 350 k— 249

a-o 199 n-w 74, 328

e-e 190, 363 n-w 317

e-i 249 h-r 340

\-i 225, 249, 298, 350 h-l 363

0-0 336, 363 S-n 344

o-a 282, 336 s — 298

u-11 74, 282, 298, 306, 317, 344 t-t 74, 199, 217, 249, 306, 324,

u-o 328 350

\-r 225 t-d 217

1— 336, 350 m-ra 217, 317, 324. 328, 340

ng-w 336, 35° md 217

ng-m 199 f-6 282

k-k 306 p-6 190

Identical 74, 217, 306, 324

Consonant identity. . . .328

Vowel identity 225, 317, 344, 350, 363

Consonant mutation:

Vowel identity 190

Frontal abrasion 249, 336

Frontal accretion 190, 340

Terminal accretion 190, 336

460

THE POLYNESIAN WANDERINGS.

RORO.

a a, e
e e, a o o, a, e, u

\ i U u, i, o

1 I, r, -
ng n, — n n m m

h t, z, —
s k, z, t

v v, b
f v, b, h, —
k— t t, h p p, b

a-a 46, 74, 214, 217, 276, 290, ng— 309, 346, 350

292, 300, 308, 309, 312, 313, k— 46, 249, 306

324- 335. 339. 346, 35°. 360 n-w 46, 74, 254, 259, 290, 321

a-e 291 h-2 206, 363

e-e 190, 276, 290, 363 h-2 206

e-a 122, 249, 272 h — 339

\-i 46, 122, 206, 249, 254, 259, S-h 239

290, 291, 298, 300, 308, 312, s-z 298

313, 321, 346, 350 s-t 298

0-0 206, 247, 336, 363 t-t 247, 249

o-a 336 t-h 74, 217, 276, 306, 324, 346,

o-e 259 350

o-w 309 m-m 217, 312, 313, 324

u-w 74, 292, 292, 298, 306, 321, v-v 291

335> 360 v-6 122, 291

u-i 239 f-v 206, 254

u-0 247 f-6 214, 272, 290, 360, 363

I-/ 308 f-h 259

1-r 272, 308, 335, 339 f— 292

1— 3°9, 312, 313, 336, 350 p-/> 190

ng-w 308, 336 pb 247

Identical 190, 321, 335, 339

Vowel identity 46, 276

Consonant mutation :

Vowel identity 74, 206, 214, 217, 254, 290, 298, 308, 324, 360,

363

Frontal abrasion 46, 239, 249, 292, 306, 336

Frontal accretion 300, 339

Terminal accretion 190, 247, 249, 259, 335, 350, 360

THE SOUTHERN GATEWAY. 461

RUBI.

a a, e, i, o
e e, a, i oo
\ i u u, o

1 /, n, —
ng g, mo, — n n, g mm

h
s r

v —

i w, g, —
k k, g, — t t, k, r p p, b, w

a-o 46, 199. 217, 265, 276, 278, ng— 285

292, 312, 313, 317, 321, 324, kk 351

334. 349. 351, 360 kg 46

a-e 328 k — 249, 306

a-i 284, 290 n-w 46, 187, 290, 317, 321, 328,

a-o 109 334, 351

e-e 187, 190, 276, 290 n-<7 109

e-a 249 s-r 298

e-t 349 t-t 199, 217, 249, 276, 306, 324

\-i 46, 249, 285, 290, 298, 312, t-k 349

313, 321 t-r 278

0-0 187, 285 m-w 217, 265, 312, 313, 317, 324,

u-w 109, 265, 278, 284, 292, 298, 328

306, 317, 321, 334, 351, 360 v— 109

u-0 328 i-w 360

I-/ 349 t-g 360

I-n 334 f — 290, 292

1— 265, 284, 312, 313 p-p 285

ng-o 285 p-6 190, 285

ng-mo 199 p-w 284

Identical 217, 276, 317, 321, 324

Consonant identity. . . .328

Vowel identity 187, 265

Consonant mutation :

Vowel identity 46, 190, 199, 278, 285, 298, 306, 334

Terminal abrasion 292

Frontal abrasion 249, 265, 292, 306

Frontal accretion 187

Terminal accretion 46, 249, 321

Metathesis 360

462 THE POLYNESIAN WANDERINGS.

SARIBA.

a a, e, i, o

e c, a, i o o, i
i i, u u u, a, o

1 r, n

ng n, m n n, g m m

h s

s s

V IV, —
f V, u, w, s, k,—
k k, g, — t t, d, k, s p p, b

a-a 46, 74, 109, 199, 217, 239, k — 258

258, 265, 276, 291, 313, 324 n-w 74, 290, 321, 328, 330

a-e 278, 294 n-g 109

a-i 328 h-> 278, 363

a-o 258 s-,9 239, 298

e-e 190, 276, 290, 297, 363 t-t 74, 199, 217, 247, 249, 258,

e-a 249, 272 276, 324

e-i 363 t-(/ 329

\-i 46, 149, 249, 285, 290, 291. t-A' 294

297, 298, 313, 330 t-s 258

\ u 329 m-;» 217, 258, 265, 313, 321, 324,

0-0 285, 363 328

o-i 247 \-w 291

u-n 74, 109, 212, 239, 265, 278, v — 109

294, 298, 321, 329, 330 f-v 272

u-a 272, 328 i-w 294, 363

u-o 247 i-u 363

\-r 149, 212, 297 is 290

I-n 265, 313 f-& 212

ng-n 285 f — 329

ng-w 199 p-p 247

k-/v' 46, 249, 297 p-b 190, 285

k-a 149, 249, 272

Identical 46, 74, 217, 239, 276, 297, 298, 324, 330

Consonant identity. . . .249, 328

Vowel identity 265

Consonant mutation:

Vowel identity 190, 199, 212, 285, 291, 313, 363

Frontal abrasion 321

Frontal accretion 149, 239, 276, 278

Terminal accretion 109, 149, 272, 321, 328

Metathesis 149

THE SOUTHERN GATEWAY. 463

S1NAUGORO.

a a, e
e e, a o o

i i u u, o

1 /, r, n, g, —

ng g n n, g m m

h I

s /, r

v iv, —

f V

k k, g t t, k, s, I p b, —

a-a 46, 74, 109, 214, 217, 265, k-& 351

276, 290, 291, 292, 312, 313, kg 46, 214, 249, 306, 338

317, 318, 321, 324, 334, 335, n-n 46, 74, 187, 290, 292, 317,

338, 346, 349. 35i 321, 328

3L-e 328 n-g 109

e-^ 122, 187, 190, 276, 290, 318, h-l 47

338, 349 s-/ 298

e-a 47, 249 s-r 338

\-i 46, 47, 249, 285, 290, 291, t-t 47, 74, 217, 249, 306, 324,

298, 312, 313, 321, 330, 346 346

0-0 187, 247, 285 t-k 349

u-w 74, 109, 247, 265, 292, 298, t-l 247

306, 317, 330, 334, 335, 351 t-s 276, 318

u-o 328 mm 217, 265, 312, 313, 317, 318,

1-/ 335 324. 328

\-r 349 v-w 291

l-n 265, 334 v— 109

l-<7 313 f-v 122, 214, 290, 292

1— 312, 313 p-6 190, 285

ng-# 285, 346 p— 247

Identical 74, 217, 317, 321, 324, 330, 335

Consonant identity. . . .325

Vowel identity 109, 187, 265, 276, 318

Consonant mutation:

Vowel identity 46, 122, 190, 214, 285, 290, 291, 298, 306, 313,

334, 336

Frontal accretion 187, 190, 276

Terminal accretion 249, 321

464 THE POLYNESIAN WANDERINGS.

SUAU.

a a, e, o
e e, a o o

i i u u, o

1 r, n, —

ng n, m n n m m

h s

s j-

v g

f i>, w, ^, h, e, —
kg, — t t, s, — p b, —

a-a 46, 74, 199, 207, 217, 239, k— 46, 249, 258

258, 265, 276, 278, 290, 292, n-rc 74, 290, 292, 317, 321, 330

313, 317, 318, 324, 350, 360 h -s 278, 363

a-e 294 s-j 239, 298, 344

a-o 199, 291 t-t 74, 199, 207, 217, 249, 258,

e-e 190, 276, 290, 318, 363 276, 318, 324

e-a 249 t-s 247, 258

i-i 46, 149, 249, 290, 291, 298, t — 294

3i3» 33°, 35o m-ra 217, 258, 265, 313, 317, 318,

0-0 247, 363 321, 324

u-w 74, 207, 212, 239, 247, 258, \-g 291

265, 278, 292, 294, 298, 317, i-v 294, 363

330, 344, 360 i-w 290

U-o 321 i-h 290

\-r 149, 212 f-.? 290

1-w 265, 313 i-e 292

I— 350 f— 212, 360

ng-w 350 p-b 190, 207

ng-ra 199 p — 247
kg 149

Identical 74, 217, 239, 276, 298, 317, 318, 324, 330, 344

Vowel identity 46, 265

Consonant mutation:

Vowel identity 190, 207, 212, 278, 313, 350, 363

Terminal abrasion 290

Frontal abrasion 46, 247, 249, 321

Frontal accretion 149

Terminal accretion 207

Metathesis 149

THE SOUTHERN GATEWAY. 465

TAGULA.

a a, o
e e, i o o, u, we

i i, u u u, ia

1 /, n
ng ng n n, ng m m

h
s

V

f W, V

kg,— t d, r p b

a-a 217, 258, 294, 302, 312, 313, l-n 313

317, 324 ng-ng 336

a-o 258 k-<7 302

e-e 190 k — 258

e-i 190 n-w 259

i-z 247, 259, 312, 313 n-ng 259

«-« 259, 302 t-d 247, 258, 302

0-0 247, 336 t-r 217, 258, 294, 324

o-M 336 m-m 217, 258, 312, 313, 317, 324

o-we 336 i-v 294

U-W 212, 294 f-W 212

u-t'a 212 p-b 190, 247

I-/ 212, 312, 336

Identical 312

Vowel identity 217

Consonant mutation:

Vowel identity 190, 212, 247, 294, 313, 324

Terminal abrasion 259, 317

Frontal accretion 247, 336

466

THE POLYNESIAN WANDERINGS.

TAUPOTA.

a a, i, o
e e, a o o

i i, a, e u u, o

1 /, r, n, g
ng n, g, m n n, g, h, — m m

h t
s fe, g\ —

v w

f v, iv, b, —
k k, g, — t /, — p b

a-a 74, 199, 217, 239, 265, 290, ng-w 199

291, 300, 312, 313, 317, 324, k-k 225

339. 35°. 360 kg 297

a-* 328 k — 46, 306

a-o 199 n-x 46, 74, 290, 317, 330

e-e 190, 290 n-g 254

e-a 272 n-Ii 328

i-r 46, 259, 290, 291, 298, 300, n— 259

312, 313, 350 h-* 339

i-« 254, 330 $-h 239

i-a 254 s-^' 344

0-0 259, 282, 336 S— 298

u-u 74, 239, 265, 272, 282, 298, t-/ 74, 199, 302, 306, 324, 350

302, 306, 317, 321, 330, 344, t— 217

360 mm 217, 265, 312, 313, 317, 321,

u-o 328 324, 328

W 339. 35° V-w 291

\-r 225, 297 i-v 254, 259, 282, 290

1-w 265, 312, 313, 336 i-w 290, 360

l-<7 272 f-6 272

ng-w 336 f— 360

ng-9 350 p-b 190

Identical 74, 239, 317, 324

Vowel identity 46, 217, 259, 265, 344

Consonant mutation :

Vowel identity 190, 282, 290, 291, 306, 312, 313, 350

Frontal abrasion 46, 259, 306, 32

Frontal accretion 225, 239, 298

Terminal accretion 190, 225, 254, 272, 300, 302, 321, 330, 336, 339

Metathesis 360

THE SOUTHERN GATEWAY. 467

TAVARA.

a a, o
e e, a, u o o, u
i i, o u u, o

1 I, n
ng n, g, m n n m m

h //, /
s

v g

f />, w, 1 1
k gr, //, — t t, — p p

SL-a 46, 74,199,217,258,265, ng-w 199

290, 300, 313, 324, 339, 350 k-^ 272

a-o 199, 258, 291 k-h 297

e-e 290 k — 46, 258

e-a 272 n-n 46, 74, 259, 290

e-u 190 h-// 206

\-i 46, 259, 290, 291, 300, 312, ht 339

3!3» 35° t-' 74. 199- 258, 324, 350

i-o 206 t — 217

0-0 259, 336 mm 217, 258, 265, 312, 313, 321,

o-u 206 324, 328

u-M 74, 258, 265, 272, 321 x-g 291

11-0 328 f-/> 206

1-/ 297 i-w 259, 290

!-» 265, 312, 313, 336, 350 i-h 272

ng-w 336 p-/> 190

ng-9 350

Identical 74, 324

Vowel identity 46, 217, 265

Consonant mutation :

Vowel identity 259, 290, 313, 350

Terminal abrasion . . .312

Terminal abrasion 312

Frontal abrasion 46, 321

Frontal accretion 190

Terminal accretion 272, 300, 321, 336, 339

468 THE POLYNESIAN WANDERINGS.

TUBETUBE.

a a, e, i
e e, a, i o o, ua
i i u u, i

\ I, n, y
ng n n n m m

h s
s s

V w

f h, s, i, v, w, —
k k, g, — t t, s, k p p, b

a-a 46, 74, 217, 258, 265, 291, k— 258

292, 294, 298, 321, 330, 360 n-n 46, 74, 259, 290, 292, 317,

a-e 291, 294, 300, 360 330

a.-i 265 h-s 206, 340

e-g 190, 290, 318 s-s 298

e-o 249 t-t 74, 217, 249, 258, 318, 324

e-i 190 t-s 247, 258

\-i 46, 149, 206, 249, 259, 285, t-k 294

290, 291, 298, 300, 313, 330 mm 217, 258, 265, 313, 317, 318,

0-0 206, 285 321, 324, 340

o-ua 247 x-w 291

u-w 74, 212, 247, 258, 265, 292, i-h 212

294, 298, 321, 330, 360 f-s 290

u-i 317 i-i 292

\-l 212 f-i; 294

1-w 265, 313 f-w 360

\-y 265 f — 206

ng-w 285 p-p 247

k-k 46, 149 p-6 190, 285

k-g 249

Identical 46, 74, 149, 217, 298, 318, 324, 330

Consonant mutation :

Vowel identity 190, 206, 212, 285, 291, 313, 340

Terminal abrasion 259, 317

Frontal abrasion 321

Terminal accretion 149, 249, 321

Metathesis 360

THE SOUTHERN GATEWAY. 469

UNI.

a a, e, o
e e, a o o, eo, a

\ i u u

I I, r, i, -

ng w, — n w, — m w

h d

s /

V V, w
f v, b, g
k — t k, — p b

a-a 46, 217, 265, 284, 290, 292, 1— 312, 313, 336

312, 313, 317, 324, 339, 342, ng-w 346

346, 360 ng— 336

a-e 291 k — 46

a-o 284 n-w 46, 290, 292, 317, 321, 342

e-e 290 n— 259, 342

e-a 272 h-d 339

\-i 46, 259, 290, 291, 312, 313, s-t 342

321, 342, 346 t-k 324, 346

0-0 336 t — 217

o-eo 259 mm 217, 265, 312, 313, 317, 324

o-a 346 x-v 291

u-u 265, 272, 284, 292, 317, 321, x-w 291

360 f-v 290, 292, 360

l-l 265, 272, 284, 339 f-6 272, 290

l-r 265 i-g 259

I-t 284 p-6 284

Identical 265, 321

Vowel identity 46, 217, 339

Consonant mutation :

Vowel identity 290, 292, 324, 342, 360

Frontal abrasion 46, 259

Frontal accretion 321, 336

Terminal accretion 259, 321

470

THE POLYNESIAN WANDERINGS.

WAGAWAGA.

a-a

74, 217,

290

t-/

74

e-e

290

t—

217

it

290

m-m

217

\-u

74

f-w

290

n-n

74. 290

ng n, —

i'AIMA.

a a

a

0

1 /,

>', —

n n

h ft

s t

t t, h

u »

m m

a-a 308, 312, 335,

346

ng—

346

e-a 47

n-n

321

i-i 47, 308, 312,

321, 346

h-h

47

u-w 321, 335, 344

s-t

344

1-/ 308

tt

47

!->' 335

t-h

346

1— 312

mm

312

ng w 308

Identical

■ • 47. 321,

335

Consonant mutation

Vowel identity . .

• • • 308, 344

Metathesis

•■•47

THE SOUTHERN GATEWAY.

471

WEDAU.

a a, i, o
e e, a o o, u

i /'. a, c, o u u, o

1 /, r, n, g
ng n, q, m n n, g, — m m

h g, ~
s g, —

v 6, w
f w, />, v, b
k /,-, ^, - t t, — p 6

a-a 46, 74, 109, 199, 217, 239. ng-# 308, 350

258, 265, 276, 290, 291, 300, ng-;;/ 199

3°8, 313. 317, 324. 328, 340, k-k 225

350.35i.36o kg 297,338

a-i 328 k — 46, 258, 306, 338

a-o 199 n-;; 46, 74, 290, 317, 351

e-e 190, 276, 290, 338, 363 n-g 109

e-a 272 n — 254, 259, 328

\-i 46, 206, 259, 290, 291, 300, h-g 340

308, 313, 350 h — 206

i-a 254 s-g 344

i e 308 s— 239

\-o 206 t-t 74, 199, 258, 276, 302, 306,

0-0 247, 259, 282, 336, 363 324, 350

o-M 206 t— 217, 247

u-« 74. 109, 239, 247, 258, 265, m-;» 217, 258, 265, 313, 317, 321,

272, 282, 302, 306, 317, 321, 324, 328, 340

344, 351, 360 x-b 109

u-o 328 \-w 291

1-/ 272, 308 i-v 254, 259, 282, 290, 363

\-r 225, 297 i-b 272

\-n 265, 313, 336, 350 i-w 290, 360

\-g 272 i-p 206, 272

ng-n 308, 336 p-b 190, 247

Identical 74. 276, 317, 324

Vowel identity 46, 217, 239, 259, 265, 344

Consonant mutation:

Vowel identity 190, 282, 290, 291, 306, 308, 313, 340, 350, 350,

363

Frontal abrasion 46, 259, 306, 321

Frontal accretion 225, 239, 254, 360

Terminal accretion 109, 190, 225, 272, 300, 302, 308, 321, 336

Metathesis 360

472

THE POLYNESIAN WANDERINGS.

As was done in the preceding part of the work, we next group these
mutations under their several phonetic elements. Regarded locally the
following tables provide the means for an analytical study of the Poly-
nesian content of the languages which, on the northern shore of Torres
Straits, reach from the head of the Gulf of Papua to the Louisiade Archi-
pelago. Considered more generally they fall into place as supplementary
to the series of similar tables beginning on page 121.

Boniki

Dobu

Doura

Galavi

Galoma

Hula

Kabadi

Keapara

Kiriwina

Kiviri

Kubiri

Misima

Mugula

Mukawa

Murua

Nada

Oiun

Panaieti

Pokau

Roro

Rubi

Sariba

Sinaugoro

Suau

Tubetube

Uni

Awalama

Boniki

Galavi

Kiriwina

Misima

Murua

Nada
Rubi

Sariba
Taupota

Tubetube
Wedau

Awalama
Boniki
Dobu
Galavi

Kiviri

Mugula

Mukawa

Nada

Oiun
Raqa
Rubi
Sariba

Suau

Tagula

Taupota

Tavara

Uni

Wedau

Kiviri

Misima

Murua

Panaieti

i-c

Boniki

Dobu

Doura

Galavi

Galoma

Hula

Kabadi

Keapara

Kiriwina

Kiviri

Kubiri

Mekeo

Misima

Motu

Mugula

Mukawa

Nada

Oiun

Panaieti

Pokau

Roro

Rubi

Sariba

Sinaugoro

Suau

Taupota

Tavara

Tubetube

Uni

Waima

Wedau

Keakalo
Kiriwina
Kiviri
Kwagila

Miriam
Misima
Motu
Mugula

Murua
Nada
Oiun
Panaieti

Pokau

Raqa

Rubi

Sariba
Tagula
Tubetube

Motu

Boniki

Murua

Tavara

Awalama
Boniki

Galavi

Nada

Taupota

Wedau

Hula
Kabadi

Motu

Pokau

Taupota

Wedau

Awalama
Boniki

Galavi

Motu

Tavara

Wedau

Motu

Murua

Oiun

Sariba

Tagula

Kiriwina
Kubiri

Oiun

Raqa

Roro

Uni

Boniki
Dobu

Doura
Galavi

Kabadi
Mekeo

Pokau
Roro

Uni

THE SOUTHERN GATEWAY.

473

Motu

Mugula

Sariba

Kiriwina
Mekeo

Mukawa
Nada

Panaieti
Roro

Tagula
Tavara

Tubetube
Wedau

xx-a

Dobu
Kiviri

Mabuiag
Miriam

Motu

Oiun

Sariba

u-e

Miriam

Nada

u-i

Galoma
Hula

Keapara

Murua

Oiun

Tubetube

u-o

Awalama

Boniki

Galavi

Galoma

Hula

Kabadi

Keapara

Kubiri

Motu

Mugula

Mukawa
Murua
Nada
Pokau

Raqa
Rubi
Sariba
Sinaugoro

Suau
Taupota
Tavara
Wedau

H

Galoma
Hula

Keapara

Sinaugoro

Taupota

Wedau

It

Doura

Murua

Pokau

Tubetube

Uni

1 /

Awalama

Boniki

Dobu

Doura

Galavi

Hula

Keapara

Kiriwina

Mekeo

Miriam

Misima

Motu

Mugula

Nada

Pokau

Roro

Rubi

Sinaugoro

Tagula

Taupota

Tavara

Tubetube

Uni

Waima

Wedau

\-n

Awalama

Boniki

Dobu

Galavi

Galoma

Hula

Keapara

Kiriwina

Kiviri

Kubiri

Misima

Mugula

Mukawa

Murua

Nada

Panaieti

Rubi

Sariba

Sinaugoro

Suau

Tagula

Taupota

Tavara

Tubetube

Wedau

\-r

Awalama

Dobu

Doura

Galavi

Galoma

Hula

Kabadi

Keapara

Kiriwina

Kiviri

Kubiri

Kwagila

Massim

Motu

Mukawa

Nada

Oiun

Panaieti

Raqa

Roro

Sariba

Sinaugoro

Suau

Taupota

Uni

Waima

Wedau

1—

Boniki

Dobu

Domara

Doura

Galavi

Galoma

Hula

Kabadi

Keakalo

Keapara

Kiriwina

Kiviri

Kwagila

Mailu

Mekeo

Misima

Motu

Mugula

Mukawa

Murua

Nada

Oiun

Panaieti

Pokau

Raqa

Roro

Rubi

Sinaugoro

Suau

Uni

Waima

ng-g

Awalama

Dobu

Galoma

Hula

Keapara

Kiriwina

Murua

Nada

Panaieti

Rubi

Sinaugoro

Taupota

Tavara
Wedau

ng-l

Pokau

474

THE POLYNESIAN WANDERINGS.

ng-m

Awalama

Kiviri

Murua

Boniki

Misima

Nada

Dobu

Mugula

Oiun

Galavi

Mukawa

Raqa

ng-n

Awalama

Kwagila

Mukawa

Boniki

Mekeo

Nada

Dobu

Misima

Oiun

Galavi

Motu

Panaieti

Kabadi

Mugula

Pokau

Kiviri

ng-«g

Mekeo

Tagula

ng—

Dobu

Kabadi

Nada

Doura

Motu

Pokau

kg

Dobu

Keakalo

Motu

Galavi

Kiriwina

Murua

Galoma

Kiviri

Nada

Hula

Kubiri

Oiun

Keapara

Misima

Rubi

k-/j

Awalama

Keapara

Tavara

k-fe

Boniki

Kiviri

Mukawa

Galavi

Kubiri

Murua

Hula

Kwagila

Nada

Keapara

Misima

Oiun

Kiriwina

Motu

Panaieti

k-wt
k-s

Rubi

Sariba

Suau

Raqa

Roro

Sariba

Suau

Taupota

Roro
Rubi

Sariba
Sinaugoro
Suau
Tagula

Pokau
Raqa
Rubi
Sariba

Taupota

Tavara

Wedau

Tavara

Tubetube

Uni

Waima

Wedau

Uni
Waima

Taupota
Tavara
Tubetube
Wedau

Sinaugoro
Taupota
Tubetube
Wedau

Motu
Dobu

Kiviri

Oiun

n-d

n-g

n-k
n-k

n-l

Awalama

Dobu

Doura

Galoma

Hula

Kabadi

Kiviri

Kubiri

Mabuiag

Mekeo

Motu

Oiun

Pokau

Raqa

Roro

Rubi

Sariba

Suau

Tagula

Taupota

Tavara
Tubetube
Uni
Wedau

Kwagila

Awalama
Galoma

Keapara
Panaieti

Rubi
Sariba

Sinaugoro
Taupota

Wedau

Awalama

Taupota

Boniki

Galavi

Hula

Misima

Mukawa

Kiriwina

Misima

Nada

Kubiri

Oiun

Raqa

Awalama

Boniki

Dobu

Doura

Galavi

Galoma

Hula

Kabadi

Keapara

Kiriwina

Kiviri

Kubiri

Mekeo

Misima

Motu

Mugula

Mukawa

Murua

Nada

Oiun

Panaieti

Pokau

Raqa

Roro

Rubi

Sariba

Sinaugoro

Suau

Tagula

Taupota

Tavara

Tubetube

Uni

Wagawaga

Waima

Wedau

n-ng

Mekeo

Panaieti

Tagula

THE SOUTHERN GATEWAY.

475

Murua

Nada

h-d

h-2

h-h

h-k

h-l

h-r

hs

h-t

h-z

h—

S-d

s-g
s-h

Boniki
Doura
Galavi

Hula

Mekeo

Misima

Nada
Panaieti

Kabadi

Motu

Nada

Galavi

Mukawa

Wedau

Tavara

Waima

Mekeo

Raqa

Sinaugoro

Galoma
Hula

Keapara
Kiviri

Kubiri
Misima

Dobu
Kiriwina

Kiviri
Kubiri

Mugula
Mukawa

Awalama

Roro

Taupota

Roro

Awalama

Roro

Wedau

Doura

Motu

Pokau

Taupota

Wedau

Awalama

Misima

Motu

s-k

8-1

8-5

s-sh

s-t

Mekeo

t-d

t-h

Pokau
Taupota

Pokau

Oiun
Panaieti

Sariba
Suau

Tavara

Roro

Uni
Wedau

Uni

Raqa
Tubetube

Taupota

Keapara

Sinaugoro

Kiriwina

Oiun

Raqa

Galoma

Hula

Rubi

Sinaugoro

Boniki
Dobu
Galavi
Kiriwina

Kiviri
Kubiri
Mabuiag

Misima
Mugula
Mukawa

Murua

Nada

Panaieti

Sariba

Suau

Tubetube

Mugula

Kabadi

Mukawa

Roro

Uni

Waima

Roro

Awalama
Boniki

Mukawa
Oiun

Raqa

Taupota

Wedau

Awalama
Dobu

Kiriwina
Murua

Nada
Pokau

Raqa
Sariba

Tagula

Awalama
Hula

Mugula

Panaieti

Roro

Waima

476

THE POLYNESIAN WANDERINGS.

t-i

t-fe

Hula

Doura

Hula

Kabadi

Kiriwina

Mekeo

Mugula

Pokau
Rubi

Sariba
Sinaugoro

Tubetube
Uni

t-l

Galoma

Keapara

vSinaugoro

t-m

Boniki

Galavi

Koita

t-n

Misima

Panaieti

t-r

Hula

Rubi

Tagula

t-s

t-t

Dobu
Kabadi

Kiriwina
Mukawa

Murua
Nada

Sariba
Sinaugoro

Suau
Tubetube

Awalama

Boniki

Dobu

Duba

Galavi

Kiriwina

Kiviri

Kubiri

Kwagila

Massim

Misima

Motu

Mugula

Mukawa

Murua

Nada

Oiun

Panaieti

Raqa

Roro

Rubi

Sariba

Sinaugoro

Suau

Taupota

Tavara

Tubetube

Wagawaga

Waima

Wedau

t —

Awalama

Dobu

Galoma

Hula

Keapara

Mekeo

Suau
Taupota

Tavara
Uni

Wagawaga
Wedau

m-d

Raqa

\-b

v-p

f-b

t-d

Awalama

Boniki

Dobu

Domara

Doura

Duba

Galavi

Galoma

Hula

Kabadi

Keakalo

Keapara

Kiriwina

Kiviri

Koita

Kubiri

Kwagila

Mabuiag

Mailu

Massim

Mekeo

Miriam

Misima

Motu

Mugula

Mukawa

Murua

Nada

Oiun

Panaieti

Pokau

Raqa

Roro

Rubi

Sariba

Sinaugoro

Suau

Tagula

Taupota

Tavara

Tubetube

Uni

Wagawaga

Waima

Wedau

Galoma

Hula

Roro

Wedau

Awalama

Kwagila

Mugula

Suau

Tavara

Mekeo

Nada

Misima

Murua

Nada

Panaieti

Doura
Kabadi

Motu

Pokau

Roro

Uni

Galoma

Hula

Keapara

Kiriwina

Misima

Panaieti

Sariba
Sinaugoro

Taupota
Tubetube

Uni
Wedau

Rubi

Sariba

Sinaugoro

Galoma

Hula

Keapara

Kiviri
Kubiri
Mukawa

Oiun

Pokau

Raqa

Roro
Taupota

Uni
Wedau

Kiriwina

THE SOUTHERN GATEWAY.

477

f-e

f-g

t-h

ft

f-k

f-p

f-s

f-u

f-v

f-w

f—

p-6

Suau

Mekeo

Keapara

', Kiriwina

Rubi

Uni

Doura
Motu

Roro

Suau

Tavara

Tubetube

Tubetube

Boniki
Doura

Galavi
Kiriwina

Mugula
Nada

Pokau

Sariba

Boniki
Galavi

Galoma
Mekeo

Mukawa
Panaieti

Tavara

Wedau

Dobu
Kabadi

Mugula
Mukawa

Sariba

Suau

Tubetube

Sariba

Awalama

Boniki

Galavi

Hula

Kabadi

Keapara
Kiriwina
Motu
Mugula

Murua
Nada
Pokau
Roro

Sariba
Sinaugoro
Suau
Tagula

Taupota
Tubetube
Uni
Wedau

Awalama

Boniki

Dobu

Galavi

Mugula

Mukawa

Rubi

Sariba

Suau

Tagula

Taupota

Tavara

Tubetube

Wagawaga

Wedau

Dobu

Galavi

Kabadi

Kiriwina

Kubiri

Nada

Panaieti

Roro

Rubi

Sariba
Suau

Taupota
Tubetube

Awalama

Boniki

Dobu

Galoma

Hula

Keapara

Kiriwina

Kwagila

Miriam

Misima

Motu

Mugula

Mukawa

Murua

Nada

Panaieti

Pokau

Raqa

Roro

Rubi

Sariba

Sinaugoro

Suau

Tagula

Taupota

Tubetube

Uni

Wedau

p-/

p-h

PP

Kiviri

Doura

Mekeo

Oiun

Awalama

Kabadi

Misima

Murua

Rubi

Dobu

Keapara

Motu

Nada

Sariba

Galoma

Kiviri

Mugula

Panaieti

Tavara

Hula

Mabuiag

Mukawa

Roro

Tubetube

p-v

Pokau

p-w

Rubi

P—

Sinaugoro

Suau

478 THE POLYNESIAN WANDERINGS.

The elaboration of this New Guinea material, the more particularly in
consideration of its important bearing upon the Viti Stream in the southern
gateway out of Indonesia, has made it advisable to extend and, when there
is reason therefor, to amend the notes in several cases which have found
their place in Appendix I.

46.

Until this Torres Straits material became available the line of partition
between kai and kani was quite distinctly the division between Polynesia
and Melanesia, for kani found its utmost eastern extension in Viti. Lack-
ing support from areas of pure Polynesian, the Viti kani could be regarded
only in the light of a Melanesian component of that mixed speech. But
on the New Guinea coast, at the very portal of exit from the Malay Archi-
pelago, we find the two stems in use side by side. Sariba exhibits the
Polynesian kai unaltered ; Suau, Mabuiag, and Dobu show the loss of the
initial k, that loss being normal in those languages. The supposition that
a medial n has been lost, which would be necessary to the argument of the
devolution of kai from kani, has been opposed by the fact that in the four
languages not only is n never lost, but it retains its proper value far more
consistently than in most of the neighbor tongues. We may, then, safely
regard kani as an ancient Polynesian stem which maintained its existence
to a point just within Nuclear Polynesia and then was dropped from the
synonymy.

47-

With this considerable record of two forms appearing side by side in
each of the Pacific areas, we need have no hesitation in assuming tahi to
be an ancient variant of the tehi radical. This slight phonetic change is
satisfactorily explained on the theory of the neutral vowel. The series
which we find in New Guinea should offer a satisfactory identification of
the Indonesian material, which in the earlier note I denied. Waima halt
is the tahi radical under 3214 metathesis. For the mutation of the aspirate
to the sibilant in Dobu and Kiriwina tasi there is abundant confirmation.
The mutation hd in Motu and Kabadi is normal to those languages, and
the kappation in the latter has become very familar in these studies.

We now pass to a group of forms susceptible of less simple explanation,
that of which Misima tari is the type. Nowhere have we found evidence
in support of a mutation h-r, except in Sinaugoro with its similar s-/ muta-
tion. But we have some evidence in support of a t-r mutation, on the
nature of which refer to note 258. If, then, we are precluded from the
direct passage from tahi to tari we have no difficulty in finding a Motu
bridge which shall give us the series tahi-tadi-tari, each span of which may
be safely traversed. Thus we are carried over to the identification of the
tari of Misima and Panaieti, and to Sinaugoro tali. Thence to art is a safe
step, for t is normally dropped in Galoma, Hula, and Keapara. This shows
that the languages which use the forms tari-ari did not acquire the word
from the Polynesian wanderers through direct contact, but as a secondary
loan from the Motu-Kabadi folk. Thus having found the r in the h place
and the loss of stem t, and both within regions of Proto-Samoan influence,
we may safely accept the Indonesian identifications.

THE SOUTHERN GATEWAY. 479

74- •
It is now quite clear that our record shows the existence of a natu series
parallel with ati. Therefore the forms which exhibit an initial nasal with
the final i are to be regarded as natu attracted by ati. In this examination
of the New Guinea material it is safe to assume natu as a Polynesian stem.
Mekeo ngau-nga is readily establishable upon Galoma nan with a nasal
termination, and the latter may perhaps be seen in Malekula netin.

76.

Particular interest attaches to the discovery of the ainu type in Mabuiag
and Miriam, western and eastern islands of the straits and remote from
the New Guinea coast, for we shall make opportunity to discuss the paucity
of Polynesian material in these islands. The existence of amu in Fotuna
affords us reason to regard the type as ancient Proto-Samoan, and that
Mabuiag and Miriam received it directly and not on secondary loan from
Motu.

109.

We note here two series, yet reference to the several phonetic tables will
show that the essential mutations, while in general feasible, are in par-
ticular not well supported.

A. Retentive of the initial consonant: Galoma, Wedau; and with

obliteration of the inner labial, Sinaugoro, Sariba, Rubi.

B. Abrading the initial, but retentive of some representative of the

inner labial : Pokau, Mekeo, Hula, Motu, Panaieti, Misima.

Nada pwau may be regarded as apu under 213 metathesis, and the then
initial labial under such maltreatment as marks the Melanesian q, cf. 285.
But from pwau Kiriwina pwak conforms to no phonetic law as yet found
in this region.

122.

Kabadi vat and Roro bai may be taken as simple variants upon the
vei type which has effected a lodgment in Nuclear Polynesia in Viti and
Rotuma. Kabakada wara has already been under comment in the principal
note on this item. The other forms are clear variants upon the Proto-
Samoan fe.

149.

The Motu and Kabadi forms, referred to the simpler stem in Nifilole
laki, may be adjudged composites of a liki variant. The liki type is found
in New Guinea only in a metathetic variant : Keapara, Hula, Galoma, Suau,
and Sariba in series. The other forms are of the kiki type with various
foreign elements in composition.

190.

In the principal note on this item we have established an elemental pe;
in this district it is recoverable from composites in Awalama and Tavara,
in which the u suggests the intimately associable ula component elsewhere
widely extended. The duplicated stem pepe is found in Hula, Keapara,
Galoma, and in composition in Dobu. The ready mutant bebe is found
simply in Rubi, Suau, Sariba; Taupota and Wedau are consimilars of »-

480 THE POLYNESIAN WANDERINGS.

composites in Awalama and Tavara ; Motu and Sinaugoro of this stem are
initial composites in a class by themselves; Mukawa and Raqa are com-
posite initial and final, the latter not identifiable but probably not mata, the
former showing the ula component ; the doubly composite Pokau involves
the same liquid component, which has elsewhere been identified in each
position; and from this we pass easily to Roro. The slightly variant bebi
stem appears simply in Mugula, Tubetube, Tagula, Murua; the Panaieti
and Misima form is most readily explicable as preduplicated. The next
stem variant is beba, simply in Nada and Kiriwina, conduplicated in Boniki.
In Kwagila we release stem bibi with an initial kara suggestive of the
common liquid component. The stem feje appears simply in Mekeo, and
with a termination of unascertained value in Kiviri and Oiun. Thus is
established a very satisfactory series of evolution forms.

206.
The few New Guinea forms exhibit harsh but not wholly improbable
mutations of the consonants and, to a less degree, of the vowels of Proto-
Samoan hifo, the i being the most consistent element. The aspirate is
found only in Tavara ; it passes to the sibilant in Tubetube, Mukawa, and
Roro (zt) ; the mutation to d is abundantly established in Motu and Pokau,
passes easily to / in Roro, and thence is caught in the general kappation
movement to Mekeo kipo; these have all been downward mutations, the
change to r in Keapara and Galoma, upward in the lingual column, has
been discussed in the supplementary note 47 as diagnostic of secondary
borrowing; in Wedau the aspirate is extinct. The /-mutations to v, p, h, b,
are abundantly familiar ; the ^-mutant in Keapara looks like a secondary
borrowing from Galoma neighbors; in other items it appears in Kiriwina,
Rubi, and Uni; the peculiar siio of Tubetube is no doubt due to the fact
that this language has acquired neither v nor/. The vowel changes call for
no note except the occurrence of \-o in Tavara and Wedau; elsewhere in
these data this mutation is identified in Awalama, Boniki, Galavi, and Motu.

207.
These forms entail no difficulty until we reach ahu, which Motu uses
along with tapu. Inasmuch as the t is abundantly determined in this
language, ahu looks like a secondary borrowing from Mekeo, where t is
commonly elided.

212.

The New Guinea forms are all/w/w derivatives, except possibly Suau uru.
As that language has no/, this may derive from fulu; equally it may derive
from the ulu which we have well established as in Proto-Samoan posses-
sion during the Melanesian transit.

214.

As in the principal note on this item, so in the Torres Straits littoral we
have no difficulty in recognizing the two stems faka and fa.

217.
The presence of the form ama in Awalama, Taupota, Tavara and Wedau,
languages which possess the t, forms a connecting link for this state of the
stem between its occurrences in Melanesia and those in Indonesia.

THE SOUTHERN GATEWAY. 481

225.
In the Pacific we find as stems kill, kali, kalo, and kalu. These are found
in New Guinea as follows, but because of this consonantal fixity we shall
not list mutant vowels.

kili: Galavi, Nada, and perhaps Raqa.

kalo:Kubiri, Kiviri, Oiun, Motu, Taupota, Wedau and Boniki.

kali: Kiriwina.

To the above New Guinea contributes :
kuli: Kiriwina and Panaieti.

Wedau and Mukawa giai scarcely seems associable, the resemblance to
kili being only in the first syllable.

239-

Interesting results appear in the comparison of these New Guinea forms
with the variants already noted from the Proto-Samoan asu. Take first
the initial elements. The palatal is found in Moanus, the first landmark
of the Samoa Stream ; in Torres Straits, at the first landmark of the Viti
Stream, in Sariba, Panaieti, Mukawa and Awalama; the whole island bars
these two streams. The labial preface of Malekula Aulua is found in Galavi,
Taupota, Wedau, and Boniki. Ambrym walehi seems paralleled with Motu
kwalahu. The asu stem is identified in Suau and Dobu without alteration.
Roro hiavu is strikingly like Lo hiev, in which we distinguish hi as the asu
derivative and ev-av-avu as the a/i-fire derivative. Nada and Murua museu
(and the invert umseu in Kiriwina) seem like asu with the wz-preface not
elsewhere seen. In the phonetic tables regard has been had only of the
recoverable asu and its mutants.

247.

Tagula bibido is our sole New Guinea occurrence of the pito stem, all the
other examples in this area deriving from puto. These have been freely
dealt with by the New Guinea tribesmen, but close analysis will show
several series of variants, each consonant being strongly affected in the
mutation. For p we have p, b,f, m, and extinction; for t we have /, d, k, s,
h, r, and extinction. The ^-forms find distant support in Buka. The muta-
tion to / has a satisfactory extension in this area and affords valuable
support to the note on the subject under item 258.

249.
In the absence of t in the languages, Hula and Keapara gia and Galoma
ia derive from Sariba gita. As to Kabadi is' ana, Dr. Ray states distinctly
that s' is found only once in New Guinea and that once is not Kabadi;
however, it hangs quite properly upon ihana which he presents as a variant
of itana in Roro, and the latter connects through ita and gita with kita
and kite.

254-

For the most part these New Guinea identifications exhibit the wt/i-stem

before the acquisition of the conditional ma, which, however, appears in

Mekeo, Galoma, and Keapara. In Taupota and Awalama we have an

initial element which is probably formative, but we are without evidence

482 THE POLYNESIAN WANDERINGS.

of transition phases which may establish wo as a mutant of ma. In the
former consonant of the stem the mutation n-ng-g-k to extinction is steady
and well established ; so, too, the series of the latter consonant f-v-p-w.

258.

In the Proto-Samoan stem matakut we find four consonants to engage
our attention. The m, in fact the whole conditional preformative ma with
the exception of Kabadi me, exists unaltered through our New Guinea
identifications. The t in second place undergoes no greater change than
the kappa tion in Pokau and Kabadi and the change to r in Tagula, on
which see the principal note. In Tagula mar ode we shall arrive at the
most satisfactory conclusion by assuming that the stem syllable ku has
vanished, thus leaving d as representative of stem t. The stem k has van-
ished completely in all these New Guinea identifications. The final t is
preserved in Dobu, Oiun, Awalama ; becomes d in Tagula, s in Sariba ; has
vanished from Pokau and Kabadi.

259-
The development of these widely diverse forms is to be studied in the
mutations of the stem consonants. The former remains n in Mukawa,
Awalama, Ta vara, Roro, Kabadi, Mekeo, Pokau, Panaieti.Tubetube, Misima,
Tagula; it is no more than a breathing in Motu; is lost from Taupota,
Wedau, Doura, Uni, Galavi, Boniki. The / becomes v in Taupota and
Wedau, b in Mukawa, w in Awalama and Tavara, h in Roro, and vanishes
in Mekeo. From this stage we reach the extirpation of the second syllable
in Panaieti, Tubetube, and Misima; and Tagula I regard as a devolution
form of the latter. Returning to Roro we find the h passing to ^ in Motu
and Kabadi. Here comes a lacuna. If we imagine a passage to t, prac-
ticable but nowhere recorded, the not infrequent kappation would account
for Pokau, Doura, Galavi and Boniki. Thence to Uni is an easy passage.

265.

The forms in the Torres Straits area run in the series ruma-luma-numa-
numi-yuma-uma. With the exception of numi we find Melanesian instances
of each form.

272.

The probably more elemental stem e/uis traceable in Keapara and Mekeo.
The lefit stem appears in Motu, Doura, Kabadi, Roro, Uni, Wedau, Galavi,
Boniki. In the principal note I pronounced, on the material then acces-
sible, against kabu of the Duke of York. We now recognize a kefu stem,
either coordinate with lefu and nefu or possibly derivative from the latter.
This is found in Taupota, Wedau, Motu, Tavara, Awalama, and Sariba.

273-
The addition of this material removes the objection noted in the principal
note (b) against the form kuvi. The assumption of initial k is of wide
extent in New Guinea ; in fact but two identifications lack it, Motu uhe,
which is not the common yam name, and the variant in Pokau, which is
metathetic of the 231 type.

THE SOUTHERN GATEWAY. 483

276.

This series of identifications runs very smoothly and calls for little com-
ment. In a group of the languages we encounter the preface k which has
been noted in 273. So far as comparison may be made this appears in the
same languages in the two items.

278.

The identifications are very simple and faithful to the uha stem. Sariba
exhibits the preface k as in 276. The Rubi form with the liquid is com-
parable with matching forms in the other gateway.

282.

These forms are simple until we reach Kiviri-Oiun bobu. This seems to
be a duplication of fou degraded either to fo or to/w, the former being intrin-
sically the more likely form, but we lack the data to make a determination.

284.

The course of the initial consonant in this area is to b, to v, to h, to w.
The liquid passes through i to extinction. This passage from consonant
to vowel has not been segregated in Melanesian phonology, but in New
Guinea it is found in five languages and in several items.

285.

The absence of complication in these identifications in the southern gate-
way affords me the opportunity to note here, as satisfactorily as anywhere,
my surprise that in this New Guinea material we find so marked an absence
of the final consonants of the closed stems. In the hypothesis we are to
regard these landings on the shore of Torres Straits as among the most
ancient way-ports of the migration of the Proto-Samoan swarm, yet in this
particular we find a predominance of the junior forms which in Polynesia
we are to regard as developed in general since the Melanesian transit. It
may well be that the New Guinea languages shared the tendency of objec-
tion to closed stems and had not yet advanced to the employment of
formative suffixes, which has been the chief agent in preserving once final
consonants in Nuclear Polynesian. In this case, having no means of pro-
tection for closing consonants, the process of final abrasion proceeded with
scant interruption to remove them.

290.

Here, as in remoter Polynesian migration, we find the trace (Murua vine)
of the earlier stem -fine. Another simple stem calls for comment, sina in
Mugula, sine in Suau, Sariba, Tubetube, Dobu. I incline to regard this as
fine under a first mutation (hine) to the aspiration of its own series ; then,
by the floating of h, swung into secondary mutation normal to the lingual
series. Thus we may establish the transition phase by which Sina has
become a woman's name in Polynesia and is thus removed from possibility
of association with the shining sina (342).

In the general fafine stem this material gives us reason to suspect that
the former and the latter/ are not of the same potency or quality; for
that reason I note separately their mutation series for comparison. For

484 THE POLYNESIAN WANDERINGS.

fa we find fai in Suau, confirming its occurrence at three Solomon Island
stations :

Former/: to h, to w, to p, to b, to v, to extinction.
Latter/: to h, to s, to w, to p, to 6, to -y, to extinction.
Diversely mutant: w-v, Taupota, Wedau, Galavi, Boniki; w-s, Mu-
kawa ; w-h, Suau ; w — , Dobu ; — h, Redscar Bay.
The only changes of n are to k in Mukawa, to / in Kiriwina, to extinction
in Mekeo. In Rubi gaiine I regard ga as a foreign composition member
and iine as the remnant of fifine after extirpation of/, as also in Nada.

291.

The final liquid hitherto recognized in Indonesia is of wide extent in
New Guinea. It occurs in these languages: Tubetube, Massim, Taupota,
Wedau, Misima, Panaieti, Mugula, Suau, Awalama, Kwagila, Tavara. It
will tend toward simplicity to include therewith Kiriwina waia and Kabadi
veina. The p-b forms discovered in Melanesia account for Roro bei. The
k forms, sparsely found in Melanesia, are here represented by goila in five
languages. The vowel change from vai to vei has already been discovered
and accepted. In Nada and Murua rai offers a form without parallel,
unless we are prepared to establish the \-r mutation upon this instance
and a single form in Misima and Panaieti in item 109.

292.

These identifications are very satisfactory. We note two which are
unusual. In Tubetube and Suau the initial labial is replaced by a light
vowel sound, evidently transitional to its entire absence in Rubi and
Roro. In Sinaugoro, Hula, and Keapara we lack explanation of the inter-
jected palatal. The loss of the final vowel in Rubi is rare, but not without
parallel in Melanesia.

297.

The Proto-Samoan stem nowhere exactly appears in any of these New
Guinea languages save Sariba and Kiriwina. The closest approximation
is stem gila recoverable from Keapara and Misima. Next we find a gala
stem with its range of consonant mutation. At the last we find a karo
stem in Kwagila. I have collated only for the consonantal elements and
with no great insistence on the validity of these groups.

298.

A very good series is here exposed. The only unusual element is the
preface in Awalama, Taupota, and Mukawa; and this recurs in Nggao.
In the principal note I expressed doubt as to the Malay and Javanese ; we
come nearer to them with this New Guinea material, where idu and ilu
are regular in their mutation.

300.

The prevalence of n in these New Guinea identifications inclines me now
to the establishment of the Proto-Samoan stem ikan, the more particularly
as this region of the great island seems to have been little visited by Indo-
nesian rovers. The central k is altogether absent, except perhaps that the

THE SOUTHERN GATEWAY. 485

extra * in Tubetube He is some such recognition of recent loss as is carried
by the ' in Samoan. Inexplicable and unrelated prefixes are found in Roro
and Kabadi.

302.

The kali stem is easily recognizable in Kiriwina and Nada, and the change
of final vowel does not interrupt the identification in Tagula. The Dobu
form seems associable, yet we have in that speech no confirming instance
of a t elided under full protection. The former element of Kubiri gitaboni
I regard as gati under 1432 metathesis, and thus admissible. Murua gedi
is a simple variant of gadi. The Tubetube and Suau forms are not to be
explained.

In Boniki, Mukawa, Taupota, and Wedau I recognize a different root,
the Proto-Samoan ut of the same meaning. In the two former it is modu-
lated by the preface k which has already been seen in this area. In Nuclear
Polynesia we find this root as follows :

Samoa : u,feu, utia, to bite. Futuna : u, uti,id. Uvea : uu, uusi,id.
Tonga: uu, uji, feuuji, feuutaki, id.

The Taupota utai and forms associable therewith derive quite satisfac-
torily from this Proto-Samoan ut, except that Melanesia, so far, has exhib-
ited no identifications which might bridge the great gap between Torres
Straits and Nuclear Polynesia.

306.

The New Guinea identifications are truer to type than many which are
found in Melanesia. The series is kutu-gutu-utu-uku-uhu-uu-u, and from
gutu along an offshoot to gu.

308.

In this Torres Straits material we find no occurrence of the sky sense,
which is the only meaning which the stem carries in Indonesia and in
Polynesia. In several cases the stem designates day and its light, more
commonly it is the wind, and these two meanings are found in Melanesia.
Yet on the eastern coast of New Guinea, at Gorendu in Astrolabe Bay,
Miklucho Maclay has recorded lang in the sense of sky.

3°9-

Lacking in the simplicity which marks many of the New Guinea identi-
fications, we shall best account for the variety in this item by assuming
that the loan has been made in several instances at second or third hand.
Thus, while Dobu nene is remote from lango, the latter element in the
composite nene-wara points to a borrowing of the nigau-wari of Murua ;
this is easily derivative from Panaieti nagunagu, which may readily come
from the same parent as Mekeo angu, and that parent is an easy variant
of lango. In one devolution series we find lalo-lao-ao and angu-au, in
another nagu-nigu-nigo-nene. The acquisition of these forms clears up
some of the difficulties noted in the principal note: Guadalcanar ango is
abundantly established ; and the Solomon Islands lau-au no longer requires
assignment to the Post- Polynesian raiders.

486

THE POLYNESIAN WANDERINGS.

312.

Not one of these forms fails of recurrence in Polynesia itself except nim,
and the Polynesian intolerance of closed syllables precludes that. In this
added material we note the following instances where the five word and
the hand word diverge :

Five.

Hand.

Language.

nim

lima
ima

lima

nima
nima
gima
ia-mila

Tavara, Kiviri.
Tagula, Nada.
Keapara, Hula, Sinaugoro.
Kiriwina.

1

313-

In Kiriwina ia-mila the metathetic form, of the 3214 type, has the same
prefix ia as in ia-gila in item 308 where also it is associated with a meta-
thetic form.

316.

Mukawa and Kubiri exhibit the stem as melu, which reappears in Santo,
Lo, and Volow. They differ slightly in their treatment of it, Mukawa con-
duplicating, Kubiri preduplicating the stem. The syllable na may have
locally a formative function, or it may represent the final m of the stem.

317-
The word remains almost without change in this area until we reach the
tnamu form in three languages neighboring in a single bay. The n-ra muta-
tion has been observed but once, in the New Georgia and Nggao forms in
item 351. Tagula ma has lost distinctive form, yet it is reasonable to
associate it with manu as a product of excessive degradation.

321.
There are two distinct loan series in this area. The one borrows the
form inu and subjects it to no change greater than the suffixation of local
formative elements. The metathesis in Hula is patent, 2 1 3 type. The Uni
bibinu is clearly a preduplication of binu, but we have no other example of a
prefix to this stem. It would undoubtedly have served Judge Fornander
as a link between Polynesian and Greek. It should, however, be noted
that similar preface consonants have elsewhere been noted in New Guinea
despite their anomaly in Polynesian. The second loan series is based upon
the stem inum with a consistent sacrifice of the former vowel. This series
runs clear from numa to nom; then with a further frontal abrasion we have
the series from uma to im, and probably Kiriwina mum adjusts itself there-
with. Then with a preface / it is possible that we can trace yet another
series in tanuma, toman by metathesis, toma and torn. Murua amomu may
be related in some obscure fashion to mum.

324-
Excellent identifications are found in the series mata-maka-mati-matan-
mara-maha-ma. We encounter a group based on a mani stem which we
can not reconcile with mata; they are included in the list for so much value
as may lie in the resemblance to ma.

THE SOUTHERN GATEWAY. 487

328.
We should remove the Mekeo and Roro words from this item; they are
variants of lango (309) and have already been discussed in their proper
place. So far as we have words with the n-m skeleton we may feel secure
in the identification. At the same time it should be noted that the vowel
treatment is the less usual one in each of these New Guinea languages, but
each finds confirmation in Melanesia, and the namo form is found in the
Polynesian of Fotuna. The more complicated forms call for examination in
detail. In the composites two alien elements are involved. The iodi of
Awalama is so clearly the eoti of Tavara that we are assured that the mo of
the latter is a wasted form of Awalama himo. As between the succeeding
members of the other group, kini of Taupota and Wedau, and the nika and
niku of Nada and Murua, there is an evident metathesis, but it is imprac-
ticable to determine from our material the direction in which it has acted.
We then encounter namu in the series giwiu-kimu-sitmo-simu-kimo-imo-mo.
The gumu form illuminates Tangoan Santo moke, and himo the Malo mohe,

both by metathesis.

329-

It is interesting to note in this area the close proximity of udi and udu,

the latter of which is so strongly prevalent in the eastern gateway.

330.
As soon as we pass from the identical niu in New Guinea we encounter
some very interesting series of Indonesian resemblances. Misima niku,
with the inner assumption of the aspirate, is identical with the Malagasy.
The forms with final liquid in Mukawa, Awalama, Taupota, and Kwagila
are readily comparable with forms in Indonesia subject to the same modifi-
cation. The riira of Kubiri and Kiviri is a doubtful form, but the presence of
the final liquid syllable may serve as a link to the foregoing. The Kiriwina
forms seem associable with the Sulu and Ahtiago, metathesis of the 132 type
affecting the two vowels.

The Motu identification is so distinct (and Galoma is clearly metathetic,
of the 3214 type) that it carries the nanu forms. Furthermore the muta-
tion \-n is well established, not only in New Guinea and in Melanesia but
in Polynesia as well.

All these identifications are satisfactory until we encounter Mekeo nga-
ngau-nga. I have admitted it for inspection because it has the lau stem
vowels properly placed, but the consonant mutation (which, however, is
normal in Moiki and has a certain standing in Melanesia) and the dupli-
cation lack confirmation in the speech and in the life-history of the word.

336.
After the excellent identifications in Mekeo and Tagula we shall have to
grope our way through local modifiers and degradation forms. Excising
the formative elements, we pick out nono in Mukawa, Tavara, Wedau,
Awalama, and Taupota ; and this is cut down to no in Oiun. The loss of
the initial liquid (a well-supported movement in the four languages) gives
ona in Roro, ano (321 metathesis) in Raqa, oono in Kabadi, which is yet
further reduced to oa in Uni.

488 THE POLYNESIAN WANDERINGS.

338.

In this collection Motu dae would scarcely find a place if it were not for
the Saa tae to account for the initial t, and for the establishment in other
items of the extinction of k. Wedau also has established the k loss, and
the initial g may be regarded as a modified kappation of the Motu /. Both
these forms may proceed from the stem sake. So, too, does Sinaugoro
rage, for the s-l mutation found in 298 confirms the r in this instance. The
Keakalo agi rests upon the simpler ake stem.

339-
The forms which exhibit the mutation of the initial aspirate to /, d, and
extinction are satisfactory. Thus we have no difficulty in accepting Tau-
pota, Awalama, Tavara, Roro, Motu, Uni, and Pokau. In Tavara the e
in the / place may be an error in transcription and the word be identical
with the Awalama form, or it may be comprehended as a phase of the
mutation from the liquid to the vowel noted in item 284. The remaining
forms are too remote from type to establish themselves, except that we
note the presence of a series keta-eta-ta.

340.

A considerable group of these languages has a vocable of sarima type,
but we have no explanation of the interjection of a syllable within any
Polynesian word ; we note the curious resemblance and constant difference,
but may venture on no identification. Divested of extraneous matter,
the other New Guinea forms rest upon the mutation series for the aspirate
of s-t-d-r. It is interesting to note the persistency of an n either in the
stem itself or characteristically associated therewith. This is found in
Kubiri, Koko-Yimidir, Galavi, and Mukawa, and in Astrolabe Bay on the
east coast in Gorendu saman-mole. It comes again to light in Tanna and
Aneityum.

342.

To these satisfactory identifications we add from Astrolabe Bay the
equally satisfactory Gorendu sing, the sun, a form by no means uncommon
in Melanesia.

344-

The mutation proceeds regularly from s to /, to n, to g. Vowel variety
appears but twice, in Nada and Oiun, and is associable with the only
Melanesian instance of vowel change, Norbarbar visis.

349-

The fact that Motu talo is of the present Polynesian type and that the
other New Guinea identifications are of the Efate type is proof that during
the traverse of Torres Straits both forms were in the possession of the
migration.

350-

There is little difficulty in these identifications. Kiriwina taigila is so
clearly metathetic that the intrusive i need give us no concern. We have
next a series in which the \-n mutation dominates. Next comes the loss
of I in taina, tain, teina, tcini, kaina. Then follows the loss of the vowel

THE SOUTHERN GATEWAY. 489

in the second syllable. The extirpation of an inner syllable in these lan-
guages is rare, but in this case it is incontestable ; this gives us the tertiary
series tena, tega, Una, tenan. Simultaneous loss of / and ng gives a sec-
ondary series of taia, taiya, kaia, haia-na. Mekeo aina shows loss of t and /,
but retains a trace of ng.

Forms associable with the stem kanus hitherto postulated are kanu, kan,

kanudi, kanuna, kanuta, kanunu, kaninu, kanuru, kinura, kunruvi. The

series of the anus type runs aninu, anumai, anodi, aniulu, atiulu, ainuku.

It may be that the final s of the two stems appears in kanudi, kanuta,

anodi; but it is unsafe to attempt to trace it beyond these well-established

mutations. „*-„

360.

A critical point is the position of the stem /. From it we obtain two
series : the more direct is vua, pua, bua, gua, hua; the other, affected by
metathesis quasi ufa, is uwa, uwe, ova, owu, awu; this accounts for the
Kabadi and Rubi forms. It is a matter of no great moment, save in the
establishment of radiant foci for this loan material, that while ua might
devolve from the other series it follows more normally in the languages
where it is found as a reduction form of uwa.

361.

Of the three stems conjoined in these data, New Guinea lacks the Poly-
nesian puaka. Therefore I have omitted the phonetic collation of the
material. The data are so ordered, however, that it will be easy to follow
the ramifications of the stems poro and po in this region. Nowhere in this
mass of material does any middle term appear by which we may certify
ourselves that poro and po are homogenetic.

What, now, is the position of these three dozen languages of New Guinea
from which we have extracted material, some more and some less satis-
factory? In 1892, and again in 1894, Dr. Sidney Herbert Ray declared
them Melanesian. In 1907, in the linguistic volume of the Cambridge
reports, he continues this designation with renewed argument in its behalf.
In 1 9 10 Dr. Seligmann, working upon this material and a valuable supple-
ment of his later collection, follows this leadership and designates the people
of the eastern Gulf of Papua and the continuing coast and islands well along
into the Louisiade Archipelago as Melanesian. At the point of demarca-
tion in the Gulf of Papua, approximately at Cape Possession, set down by
these students, there is a difference in the people ; more than the language
shows it. They have decided that the difference is one between Papuans
and Melanesians.

In this inquiry we shall devote no attention to the differences of physical
appearance, to the diversity of custom of life; we are limited to the evidence
which may be found to lie in language.

What, then, is a Melanesian? In Dr. Ray's studies the men eastward of
Cape Possession who differ from the men living westward of that boundary
point are the Melanesians of New Guinea. We are then to discover whence
he derives his knowledge of the Melanesians. It is from Dr. Codrington
that he has drawn — from the same work which we have been so glad to
use in this work, albeit to a far different conclusion.

490 THE POLYNESIAN WANDERINGS.

Assuming for the present, and only for the present, a standard Mela-
nesian, we must consider how that element could be communicated to the
New Guinea coast. Either the inhabitants of the Gulf of Papua voyaged
to some place of Melanesian culture, remained sufficiently long to acquire
the Melanesian element they are considered now to possess, and voyaged
home again; or else the Melanesians put to sea from their own lands,
colonized the New Guinea coast, and survive in these settlements discrete
from the Papuan autochthons. Either movement predicates navigation.
The Melanesians are notoriously not a seagoing race ; with very few excep-
tions their art of canoe building has advanced no greater distance than is
required for the construction of light dugouts in which the single paddler
scarcely dares follow the fish outside the still lagoon water within his reef.

The seamanship of the coastal peoples of New Guinea is of no better
order ; when the most adventurous of all this race sets forth upon its annual
sago voyage, the putting to sea remains so unusual to their custom on the
water that it is surrounded with all the formality of religious rites. It is
not like the free adventure of the hardy seaman who hoists sail on the
canoe in which he has confidence, trims the sheet to force the wind to work
for him, and gaily sails to a distant port, all in the day's work. The voyage
of the New Guinea traders of pots for sago is no more than across the head
of the gulf, from Motu to Elema, yet it can be accomplished only with the
favor of fair winds; the return voyage can not be made until the wind
changes in the seasonal break of the monsoon. Even the name of the vessel
which has specially to be compacted for this voyage shows that its navi-
gation is a foreign art, for lakatoi is a survival from something borrowed ;
it is a mutant of vaka-tolu, which we have no difficulty in interpreting as
"three boats," and that is true to the naval architecture, for the Motu man
fears the sea so cordially that he will not allow even his zeal for trading to
trust himself to the waves in anything less stable than three boats lashed
together. Whether the fleets of Motu went to northern Melanesia or Mela-
nesia came to the Gulf of Papua, we are confronted by the need of canoe-
craft and seamanship which has utterly vanished. Neither voyage could
be repeated in these days ; the people at each end have not the boats nor
the ability to sail them ; they fear the sea.

On either side these folk who fear the sea we have the clear record of a
race who made the sea their own, even to its empty limit, who have adven-
tured such voyages to distant lands and a safe return as proved beyond
the power of our own race until four centuries ago.

Dr. Ray rests his Melanesian identification upon the Melanesian of Dr.
Codrington's study. In this volume we have worked intimately over that
Melanesian possession. We have found no language which we might estab-
lish as the standard Melanesian, no Ursprache from which the jangling
multiplicity of languages might be shown to derive. There is none such ;
there is no standard Melanesian. In these data we have examined the
element common to the languages from Moanus to the southern tip of New
Caledonia. We have found this element everywhere reducible to its par-
entage ; we have found it Polynesian material borrowed by ruder people.

The same identification holds when we examine the data collected in
Torres Straits. It is Polynesian material. It would have been possible

THE SOUTHERN GATEWAY. 491

for voyagers from the Gulf of Papua to some Melanesian island, for colonists
coming out of Melanesia, to bring this Polynesian content of the Melanesian
speech. But the language record militates against this. As between dif-
ferent languages on the New Guinea coast we have found differences which
indicate secondary borrowing, but these readily reduce to well-marked foci
on the same coast; as to the material segregated at these foci, their char-
acter is clear: they are primary loans of the Polynesian.

On the strength of such material as was then available from the Motu,
I had no hesitation in adopting that language as establishing a proof point
of the swarm of Polynesian migration through Torres Straits, the first point
fixing the course to which, from its ultimate destination, I have assigned
the designation of the Viti Stream. This newly accessible material but
confirms the former conclusion; instead of a single early station on that
course we now have a number and all confirmatory.

The Torres Straits stations, with two exceptions, are all along the coast.
The two exceptions are Mabuiag and Miriam, which Dr. Ray assigns respec-
tively to the provinces of Australian and Papuan speech. Although the
vocabularies are the largest in our possession from the region, the material
which they have contributed to these studies is the least ; six vocables from
Mabuiag and three from Miriam are all that may be associated with the
Proto-Samoan. On the map these islands seem to lie in the fairway for
any voyage through the straits; it would seem that there the Polynesian
influence should be at its maximum. But in the conditions of Proto-
Samoan sailing these islands were remote. The Polynesians were sailors,
but they were not bigoted in their navigation. Hardy to risk the unknown
expanses of open sea when the sky lay empty before them, they preferred
the greater certainty of a coast to follow; any shore, even when it lay alee,
served to deflect their course. This is a part of their seamanship which we
have already had to consider when examining the courses through the
Solomons and the New Hebrides. It is this facility of coastwise voyaging
which has established for us so many points of Polynesian influence along
the New Guinea shore from the great gulf to the remote eastern islands.

In conclusion we judge that these languages of New Guinea that have
been called Melanesian are susceptible of correlation with languages situ-
ated geographically in Melanesia ; that such correlation rests almost wholly
on the vocables here examined ; that this element common to Torres Straits
and Melanesia is common to Melanesia and to Polynesia. Therefore the
result of the investigation is not the establishment of a distinctively Mela-
nesian content in the languages limited by Motu and Nada, but the asso-
ciation of that content through the Melanesian with the earliest type of
the Polynesian. As the result of this added study we now find the southern
gateway out of Indonesia most satisfactorily established.

APPENDIX III

BIBLIOGRAPHY.

I have been permitted to publish in the Bulletin of the American Geo-
graphical Society (vol. xli, pages 305-343) a far more extended list of the
works which have contributed to the foregoing studies. Since that publi-
cation is readily accessible I have restricted this bibliography to the purely
linguistic works upon which I have drawn for this material. A few of
general applicability are grouped as General, the remainder will be found
under the name of the language upon which they bear, with such repetitions
of authorities cited as will be found to facilitate reference. — W. C.

General

Brandstetter, Ren ward. Eine Prodromus zu einem vergleichenden Worterbuch der

malaio-polynesischen Sprachen fur Sprachforscher und Ethnographen.

Luzem: E. Haag, 1906, pp. 74.
Churchill, William. Duplication by Dissimilation. American Journal of Philology,

xxx, 171.
Duplication Mechanics in Samoan and their Functional Values. American

Journal of Philology, xxix, 33.
Principles of Samoan Word Composition. Journal 0} the Polynesian Society,

xiv, 24.
Root Reducibility in Polynesian. American Journal of Philology, xxvii,

369-
Samoan Phonetics in the Broader Relation. Journal of the Polynesian

Society, xvii, 79, etc.
Weather Words of Polynesia. Memoirs of the American Anthropological

Association, ii, r, 1907.
Codrington, R. H. The Melanesian Languages. Oxford: Clarendon Press, 1885, pp.

-> via, 572.

Crawfurd, John. On the Malayan and Polynesian Languages and Races. Journal
of the Ethnological Society, London, i, 330.

Finck, Franz Nikolaus. Die Wanderungen der Polynesier nach dem Zeugnis ihrer
Sprachen. Berlin: Nachrichien von der Koniglichen Gesellschaft der Wis-
senschaften zu Gottingen, philologisch-historische Klasse, 1909, Heft 3, 308.

Qabelentz, Hans Conon von der. Die melanesischen Sprachen nach ihrem gram-
matischen Bau und ihrer Verwandtschaft unter sich und mit den malaiisch-
polynesischen Sprachen. Aus dem VIII Bande der Abhandlungen der
Kbniglich. Sachsischen Gesellschaft der Wissenschaften. Leipzig: S.Hirzel,
i860, pp. 266.

Id., zweite Abhandlung. Des VII Bandes der Abhandlungen der philologisch-

historischen Classe der Kb'niglich. Sachsischen Gesellschaft der Wissen-
schaften No. I. Leipzig: S. Hirzel, 1873, pp. 186.

Qabelentz, Georg von der, und Adolf Bernard Meyer. Beitrage zur Kenntniss der
melanesischen, mikronesischen und papuanischen Sprachen, ein erster
Nachtrag zu Hans Conon von der Gabelentz Werke "Die melanesischen
Sprachen." Des VIII Bandes der A bhandlungen der philologisch-historischen
Classe der Kb'nigl. Sachsischen Gesellschaft der Wissenschaften No. IV.
Leipzig: S. Hirzel, 1882, pp. 170.

493

494 THE POLYNESIAN WANDERINGS.

General — continued

Macdonald, Daniel. New Hebrides Linguistics, Introductory, Three New Hebrides

Languages: Efatese, Eromangan, Santo. Melbourne, 1889: Printed at

the Expense of the Trustees 0} the Melbourne Public Library, pp. 134.
Oceania, Linguistic and Anthropological. Melbourne: M. L. Hutchinson,

1889, pp. xii, 218, plates 7.
The Oceanic Languages, their Grammatical Structure, Vocabulary and

Origin. London: Henry Frowde, 1907, pp. xv., 3,52, 2 maps.
South Sea Languages, a Series of Studies on the Language of the New

Hebrides and Other South Sea Islands. Vol. II: Tangoan-Santo, Malo,

Malekula, Epi (Baki and Bierian), Tanna and Futuna. Melbourne, 1891 :

Printed at the Expense 0} the Trustees of the Public Library, Museum and

National Gallery of Victoria, pp. xxvi, 281.
MUller, Friedrich. Reise der osterreichischen Fregatte Novara um die Erde in

den Jahre 1857, 1858, 1859 under den Befehlen des Commodore B. von

Wiillerstorf-Urbair: LinguistischerTheil. Wien: Kaiserliche Akademie der

Wissenschaften, 1867. (IV Abtheilung: Malay o-polynesische Sprachen,

pp. 267-357.)
Quatrefages, A. de. Les Polynesiens et Leurs Migrations. Paris: Arthus Bertrand,

n. d., pp. 200, 4 charts.
Ray, Sidney H. The Common Origin of the Oceanic Languages. Extract Journal

of the Polynesian Society.
Schmidt, P. W. liber das Verhaltniss des melanesischen Sprachen zu den poly-

nesischen und untereinander. Wien, 1899: Sitzungsberichte der Kais.

Akademie der Wissenschaften in Wien, philosophisch-historische Classe,

Band CXLI, 6, pp. 93.
Smith, S. Percy. Hawaiki: The Whence of the Maori: Being an Introduction

to the Native History of Rarotonga. Wellington: Whitcombe and Tombs,

1898, pp. 128.
Hawaiki: The Original Home of the Maori, with a Sketch of Polynesian

History. Second edition, enlarged and mostly rewritten. Wellington:

Whitcombe and Tombs, 1904, pp. 223.
The same, third edition. Wellington: Whitcombe and Tombs Limited, 1910,

pp. 301.
Tregear, Edward. The Maori-Polynesian Comparative Dictionary. Wellington,

N. Z.: Whitcombe and Tombs, n. d., pp. xxiv, 676.
Turner, George. Samoa a hundred years ago and long before, together with notes

on the cults and customs of 23 other islands in the Pacific. With a preface

by E. B. Tylor. London: Macmillan and Co., 1884, pp. xvi, 395.

Ahtiago

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

Alite

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,
P- 39-

Alo Teqel

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-
Grammar. Melanesian Languages, p. 355.

Amblaw

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.
Amboina

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

BIBLIOGRAPHY. 495

Ambrym

Codrington, R. H. Comparative vocabulary of 38 words. Melanesian Languages,

P- 39-

Grammar. Melanesian Languages, p. 449.

Gabelentz, H. C. von der. Die melanesischen Sprachen.

Aneityum

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-

Grammar. Melanesian Languages, p. 477.

Gabelentz, H. C. von der. Die melanesischen Sprachen.

Inglis, John. A dictionary of the Aneityumese Language; in two parts: I. Anei-
tyumese and English; II. English and Aneityumese; also, Outlines of
Aneityumese Grammar; and an Introduction Containing Notices of
the Missions to the Native Races and Illustrations of the Principles and
Peculiarities of the Aneityumese Language. London: Williams and Nor gate,
1882, pp. 200.

Aniwa

Turner, George. Comparative vocabulary of 123 words. Samoa, p. 354.
Comparative vocabulary of 15 words, source unknown.

Arag

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-

Grammar. Melanesian Languages, p. 431.

Gabelentz, H. C. von der. Die melanesischen Sprachen (Vun Marama).

AWAIYA

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

AWALAMA

Ray, Sidney H. Expedition to Torres Straits III, 482.

Baki

Fraser, R. M. A Grammar of the Baki Language of Epi, New Hebrides. (With
vocabulary.) (In Macdonald's South Sea Languages II, 73.)

Balade

Gabelentz, H. C. von der. Die melanesischen Sprachen.
Comparative vocabulary of 15 words, source unknown.

BATUMERAH

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

BlERIAN

Fraser, R. M. Grammar of the Bierian Language of the Island of Epi, New Hebrides.
(With vocabulary.) (In Macdonald's South Sea Languages II, 98.)

BOLANGHITAM

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

BONGU

Hanke, A. Grammatik und Vokabularium der Bongu-Sprache (Astrolabe-bai,
Kaiser-Wilhelmsland) ; mit einer Karte, einer wortvergleichenden Tabelle
von neun Orten des Astrolabegebietes und einem Vokabularium der
Sungumana-Sprache. Berlin: Archiv fur das Studium deutscher Kolonial-
sprachen, Band viii, Commissionsverlag von Georg Reimer, 1909, pp. xii,
252, 1 map.

BONIKI

Ray, Sidney H. Expedition to Torres Straits III, 482.

BOUTON

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

496 THE POLYNESIAN WANDERINGS.

BUGOTU

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-

Grammar. Melanesian Languages, p. 546.

BUKA

Ray, Sidney H. Mittheilungen iiber drei Dialekte der Salomon-Inseln. Zeitschrift
fur ajrikanischc und oceanische Sprachen ii, 54.
BUKABUKA

Hutchin, J. J. K. Traditions and some wordsof the language of Danger or Pukapuka
Island. Journal 0} the Polynesian Society, xiii, 173.
BlJLULAHA

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,
P- 39-

Buru Islands

Turner, George. Comparative vocabulary of 59 words in the dialects of Cajeli,
Massaratty and Amblaw. Samoa, p. 354.
Caimarian

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

Cajeli

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

Celebes

Turner, George. Comparative vocabulary of 59 words in the dialects of Bolang-
hitam, Bouton, Menado, Salayer. Samoa, p. 354.

Ceram

Turner, George. Comparative vocabulary of 59 words in the dialects of Ahtiago,
Awaiya, Caimarian, Gah, Teluti, Tobo, Wahai. Samoa, p. 354.

Chamorro

Fritz, Georg. Chamorro- Worterbuch: in zwei Theilen, Deutsch-Chamorro und
Chamorro-Deutsch. Auf der Insel Saipan, Marianen, gesammelt. Berlin:
Archiv fur das Studium deutscher Kolonial sprachen, Band ii, Commissions-
verlag von Georg Reimer, 1904, pp. vi, 124.
Deni

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-
Grammar. Melanesian Languages, p. 486.

Dobu

Ray, Sidney H. Expedition to Torres Straits III, 482.
DOURA

Ray, Sidney H. Expedition to Torres Straits III, 482.
DUAURU

Gabelentz, H. C. von der. Die melanesischen Sprachen.

Comparative vocabulary of 15 words, source unknown.

Duke of York

Brown, G. Notes on the Duke of York Group, New Britain and New Ireland
(contains a vocabulary of 58 words.) Journal 0} the Royal Geographical
Society, 1877, *37-

Codrington, R. H. Comparative vocabulary of 69 words. Melanesian Languages,

P- 39-
Ray, Sidney H. Texts in the languages of the Bismarck Archipelago (Duke of
York texts with bibliography.) Zeitschrift fur afrikanische und oceanische
Sprachen, i, 334.

Ebon

Turner, George. Comparative vocabulary of 123 words. Samoa, p. 354.

bibliography. 497

Efate

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-

Grammar. Melanesian Languages, p. 471.

Qabelentz, H. C. von der. Die melanesischen Sprachen.

Macdonald, Daniel. The Oceanic Languages, their Grammatical Structure, Vocab-
ulary and Origin (Efate" dictionary). London: Henry Frowde, 1907, pp.
xv> 352> 2 maps.

Turner, George. Comparative vocabularies of 119 and 121 words, dissimilar, one
showing a strong Polynesian content. Samoa, p. 354.
Epi (see Baki, Bierian.)

Codrington, R. H. Comparative vocabulary of 51 words. Melanesian Languages,

P- 39-
Grammar. Melanesian Languages, p. 469.

Eromanga

Codrington, R. H. Comparative vocabulary of 43 words. Melanesian Languages,

P- 39-
Gabelentz, H. C. von der. Die melanesischen Sprachen.
Gordon, J. D. Sketch of the Eromangan grammar. (In Macdonald's Three New

Hebrides Languages.)
Turner, George. Comparative vocabulary of 128 words. Samoa, p. 354.
Comparative vocabulary of 15 words, source unknown.

Espiritu Santo

Codrington, R. H. Comparative vocabulary of 69 words. Melanesian Languages,

P- 39-

Grammar. Melanesian Languages, p. 441.

Gordon, J. D. Sketch of the Santo grammar and vocabulary. (In Macdonald's

Three New Hebrides Languages.)

Fagani

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-
Grammar. Melanesian Languages, p. 499.

Fakaafo

Turner, George. Comparative vocabulary of 1 13 words. Samoa, p. 354.
FOTUNA

Gunn, Dr. William. Grammar of the language of Futuna. (With vocabulary.)
(In Macdonald's South Sea Languages II, 163.)
FUTUNA

Gre"zel, Pere. Dictionnaire Futunien-Francais avec notes grammaticales. Paris:
Maisonneuve et Cie., 1878, pp. 304.

Gah

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

Galavi

Ray, Sidney H. Expedition to Torres Straits III, 482.

Galela

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

Galoma

Ray, Sidney H. Expedition to Torres Straits III, 482.
Gani

Turner, George. Comparative vocabulary of ,59 words. Samoa, p. 354.

Gilbert Islands

Bingham, Hiram. A Gilbertese-English Dictionary. Boston: American Board 0}
Commissioners for Foreign Missions, 1908, pp. viii, 179.

498 THE POLYNESIAN WANDERINGS.

Gilolo

Turner, George. Comparative vocabulary of 59 words in the Dialects of Galela
and Gani. Samoa, p. 354.

Gog

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-
Grammar. Melanesian Laguages, p. 367.

GUADALCANAR

Gabelentz, H. C. von der. Die melanesischen Sprachen.
Guam

Safford, William Edwin. The Chamorro language of Guam. American Anthro-
pologist, vols. 5-7.
Hawaii

Andrews, Lorrin. A dictionary of the Hawaiian language, to which is appended
an English-Hawaiian vocabulary and a chronological table of remarkable
events. Honolulu, H. I.: printed by Henry M. Whitney, 1865, pp. xvi,

559-
Emerson, J. S. He hoakakaolelo no na huaoleloBeritania i mea kokua i na kanaka
Hawaii e ao ana ia olelo. Lahainaluna: Kulanui, 1845, pp. x, 184.

Hula

Ray, Sidney H. Expedition to Torres Straits III, 482.

Java

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.
Kabadi

Ray, Sidney H. Expedition to Torres Straits III, 482.

Kalil

See Stephan und Grabner: New Ireland.
Keapara

Ray, Sidney H. Expedition to Torres Straits III, 482.
King

See Stephan und Grabner: New Ireland.
KlRIWINA

Ray, Sidney H. Expedition to Torres Straits III, 482.
KlVIRI

Ray, Sidney H. Expedition to Torres Straits III, 482.
KUBIRI

Ray, Sidney H. Expedition to Torres Straits III, 482.
KWAGILA

Ray, Sidney H. Expedition to Torres Straits III, 482.

Lakon

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-
Grammar. Melanesian Languages, p. 377.

Lamassa

See Stephan und Grabner: New Ireland.

Lambell

See Stephan und Grabner: New Ireland.

Lariko

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

Laur

See Stephan und Grabner: New Ireland.

Liang

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

BIBLIOGRAPHY. 499

LlFU

A. C. Notes grammaticales sur la langue de Lifu d'apres les manuscrits du P. F. P.

Parts: Maisonneuve et Cie., 1882, pp. 72.
Gabelentz, H. C. von der. Die melanesischen Sprachen.
Turner, George. Comparative vocabulary of 122 words. Samoa, p. 354.
Comparative vocabulary of 15 words, source unknown.

LlUENIUA

Parkinson, R. Die ostlichen Inseln. (Dreissig Jahre in der Sudsee, pp. 513-564.)
Thilenius, Dr. G. Ethnographische Ergebnisse aus Melanesien: erster Theil,

Reisebericht; die polynesischen Inseln an der Ostgrenze Melanesiens.

Halle: Nova Acta, Abhandlungen der Kaiserlichen Leopold-Carolinischen

deutschen Akademie der Naturjorscher, Band Ixxx, Nr. 1, 1902, pp. 102,

4 plates, 1 map, 9 text illustrations.

ho

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-
Grammar. Melanesian Languages, p. 391.

Mabuiag

Ray, Sidney H. Expedition to Torres Straits III, 88.

Madagascar

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-
Marre\ Vocabulaire syst&natique, comparatif, des principales racines des langues

Malgache et Malayo-Polynesiennes, n. p., 1885, pp. 83-214.
Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.
Comparative vocabulary of 15 words, source unknown.
Maewo

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-

Grammar. Melanesian Languages, p. 408.

Makura

Tregear, Edward. _ The Language of Makura. Journal 0} the Polynesian Society,

v, 187.

Maeekula

Gabelentz, H. C. von der. Die melanesischen Sprachen.

Morton, Alex. Grammar of the Language Spoken at Pangkumu, Malekula, Noticing
Occasionally a Dialect Spoken North of Pangkumu and Beginning at a
Village Called Rukumbu. (With vocabulary.) (In Macdonald's South
Sea Languages, II, 34.)
Malo

Landels, J. D. Outline Grammar of Maloese as Spoken on the West Side of Malo,
New Hebrides. (With vocabulary.) (In Macdonald's South Sea Lan-
guages, II, 15.)
Manahiki

Turner, George. Comparative vocabulary of 126 words. Samoa, p. 354.
Mangareva

Tregear, Edward. A Dictionary of the Mangareva (or Gambier) Islands. Published
under the Authority of the Board of Directors of the New Zealand Institute.
Wellington: John Mackay, Government Printing Office, 1899, pp. 121.

Marquesas

Dordillon, I. R. Grammaire et dictionnaire de la langue des lies Marquises. Paris:
Imprimerie Belin Freres, 1904, pp. 294, 205.

500 THE POLYNESIAN WANDERINGS.

Marshall Islands

Erdland, August. Worterbuch und Grammatik der Marshall-Sprache nebst ethno-
graphischen Erlauterungen und kurzen Sprachiibungen. Berlin: Archiv
fur das Studium deutschen Kolonialsprachen. Band iv, Druck und Kommis-
sionsverlag von Georg Reimer, 1906, pp. xii, 248.

Qabelentz, G. von der, und A. Meyer. Beitrage zur Kenntniss der melanesischen,
mikronesischen und papuanischen Sprachen, 1882.

Gabelentz, H. C. von der. Die melanesischen Sprachen.

Hernsheim, Franz. Beitrag zur Sprache der Marshall-Inseln. Leipzig: 1880,
pp. 28.

Senfft, A. Worterverzeichniss der Sprache der Marshall-Insulaner. Zeitschrift
fur afrikanische und oceanische Sprachen, v, 79.

Steinbach-Grbsser. Worterbuch der Marshall-Sprache nach hinterlassenen Papieren
des verstorbenen Stabsarztes Dr. Edwin Steinbach (vom 1891 bis 1894,
Regierungsarzt in Jaluit, Marshall-Inseln) umgearbeitet und herausge-
geben von Herman Grosser, Direktor der Jaluit-Gesellschaft zu Hamburg.
Hamburg: L. Friederichsen und Co., 1902, pp. x, 126.

Massaratty

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

Matabello

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.
Mayapo

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

Mekeo

Ray, Sidney H. Expedition to Torres Straits III, 482.
Menado

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

Merlav

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,
P- 39-

Grammar. Melanesian Languages, p. 357.

MlLLE

Comparative vocabulary of 15 words, source unknown.
Miriam

Ray, Sidney H. Expedition to Torres Straits III, 131.
MlSIMA

Ray, Sidney H. Expedition to Torres Straits III, 482.
MOANUS

Meier, Jos. Berichtigungen zu Dr. Schnee's Mitteilungen liber die Sprache der
Moanus. Anthropos, i, 210, 472.
Moiki

Ray, Sidney H. Mittheilungen iiber drei Dialekte der Salomon-Inseln. Zeitschrift
fur afrikanische und oceanische Sprachen, ii, 54.
MORELLA

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

Mosin

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-

Grammar. Melanesian Languages, p. 350.

MOTA

Codrington, R. H., and J. Palmer. A Dictionary of the Language of Mota,

Sugarloaf Island, Banks Islands, with a Short Grammar and Index.

London: Society for Promoting Christian Knowledge, 1896, pp, xxiii, 312.

BIBLIOGRAPHY.

501

MOTLAV

Codrington, R. H

P- 39-
Grammar.

Comparative vocabulary of 70 words.
Melanesian Languages, p. 310.

Comparative vocabulary of 55 words.
Expedition to Torres Straits III, 482.

Motu

Codrington, R. H.

P- 39-
Ray, Sidney H.

MlJGULA

Ray, Sidney H. Expedition to Torres Straits III, 482.

MUKAWA

Ray, Sidney H. Expedition to Torres Straits III, 482.

MURARE

Comparative vocabulary of 15 words, source unknown.

Murray Island (see Miriam.)

Codrington, R. H. Comparative vocabulary of 52 words.
P- 39-

MURUA

Ray, Sidney H.

Mysol

Turner, George.

Nada

Ray, Sidney H.

Melanesian Languages,

Melanesian Languages,

Melanesian Languages,

Expedition to Torres Straits III, 482.

Comparative vocabulary of 59 words. Samoa, p. 354.

Expedition to Torres Straits III, 482.

Nengone

Codrington, R. H. Comparative vocabulary of 69 words. Melanesian Languages,

P- 39-

Grammar. Melanesian Languages, p. 478.

Gabelentz, H. C. von der. Die melanesischen Sprachen.

Turner, George. Comparative vocabulary of 125 words. Samoa, p. 354.

Comparative vocabulary of 15 words, source unknown.

New Britain

Bley, B. Grundziige der Grammatik der Neu-Pommerschen Sprache an der
Nordkiiste der Gazellen-Halbinsel. Zeitschrijt jiir ajrikanische und ocean-
ische Sprachen, Hi, 85.

Parkinson, R. Dressig Jahre in der Sudsee: Land und Leute, Sitten und Gebraiiche
im Bismarckarchipel und auf den deutschen Salamoinseln. Herausge-
geben von Dr. B. Ankermann, Direktorial-Assistent am Koniglichen
Museum fur Volkerkunde zu Berlin. Stuttgart: Strecker und Schroder,
1907, pp. xxii, 876, 4 maps, 56 plates, 141 text illustrations.

Ray, Sidney H. Texts in the Languages of the Bismarck Archipelago. Zeitschrijt
fur afrikanische und oceanische Sprachen, i, 334.

New Caledonia

Gabelentz, H. C. von der. Die melanesischen Sprachen.
Turner, George. Comparative vocabulary of 117 words. Samoa, p. 354.
Comparative vocabulary, source unknown, of 15 words in the languages of Balade,
Duauru, Murare, Nikete and Yengin.

New Georgia

Codrington, R. H. Comparative vocabulary of 61 words. Melaresian Languages,
P- 39-

502 THE POLYNESIAN WANDERINGS.

New Guinea

Qabelentz und Meyer. Die Sprache von Errub und Maer.

Die Sprache in cler Bai von Segaar.

Miklucho-Maclay, N. Papua-Dialekte der Maclay-Kiiste in New Guinea. (In

Gabelentz und Meyer.)
Timoteo. Notes on the Kabadi dialect of New Guinea. Journal of the Polynesian

Society. (?)

New Ireland

Brown, Q. Notes on the Duke of York Group, New Britain and New Ireland
(contains a vocabulary of 19 words). Journal of the Royal Geographical
Society, 1877, 137.

Duffield, A. J. Notes on the inhabitants of New Ireland and its archipelago, their
fine and industrial arts, customs and language (contains a vocabulary of
350 words). Proceedings of the Royal Society of Queensland, i, 115.

D'Urville, Dumont. Vocabulary of Port Praslin. Voyage de I' Astrolabe, vol. ii,
Philologie, p. 143.

Gaimard. Vocabulary of Carteret Harbor. Voyage de I' Astrolabe, ii, 143.

Stephan, Emil, und Fritz Graebner. Neu-Mecklenburg: die Kiiste von Urauddu
bis Kap St. Georg. Forschungsergebnisse bei den Vermessungsfahrten
von S. M. S. Mowe im Jahre 1904. Berlin: Dietrich Reimer, 1907, pp.
xii, 242.

Nggao

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-

Grammar. Melanesian Languages, p. 555.

Gabelentz, H. C. von der. Die melanesischen Sprachen (Mahaga) .

Nggela

Ray, Sidney H. Texts in the languages of the Solomon Islands. Zeitschrift fur

afrikanische und oceanische Sprachen, Hi, 103.
Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-

Grammar. Melanesian Languages, p. 522.

Gabelentz, H. C. von der. Die melanesischen Sprachen (Anudha).

NlFILOLE

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-
Grammar. Melanesian Languages, p. 493.

NlKETE

Comparative vocabulary of 15 words, source unknown.

NiuE

Tregear, Edward, and S. Percy Smith. A vocabulary and grammar of the Niue
dialect of the Polynesian language. Wellington: By authority, John Mackay,
Government Printer, 1907, pp. 179.

NORBARBAR

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-

Grammar. Melanesian Languages, p. 384.

NUGURIA

See Liueniua.

NUKUMANU

See Liueniua.

BIBLIOGRAPHY. 503

NUKUORO

Christian, F. W. Nukuoro vocabulary. Journal of the Polynesian Society (?).

OlUN

Ray, Sidney H. Expedition to Torres Straits III, 482.

Omba

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-
Grammar. Melanesian Languages, p. 420.

Paama

Gabelentz, H. C. von der. Die melanesischen Sprachen.

Pak

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-
Grammar. Melanesian Languages, p. 332.

Pala

Peekel, P. Gerhard. Grammatik der Neu-Mecklenburgischen Sprache speziell der
Pala-Sprache. Berlin: Archiv fur das Studium deutscher Kolonialsprachen,
Band ix, Commissionsverlag von Georg Reimer, 1909, pp. xiv, 2i6,si /nap.

Panaieti

Ray, Sidney H. Expedition to Torres Straits III, 482.

POKAU

Ray, Sidney H. Expedition to Torres Straits III, 482.

Port Moresby

Turner, George. Comparative vocabulary of 35 words. Samoa, p. 354.

Rapanui

Roussel, Pere Hippolyte. Yocabulaire de la langue de l'lle-de-Paques ou Rapanui.
Louvain: he Musion, nouvellc serie vol. ix, 159.

Raoa

Ray, Sidney H. Expedition to Torres Straits III, 482.

Rarotonga

Turner, George. Comparative vocabulary of 122 words. Samoa, p. 534.

Roro

Ray, Sidney H. Expedition to Torres Straits III, 482.

ROTUMA

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-

Grammar. Melanesian Languages, p. 401.

Turner, George. Comparative vocabulary of 116 words. Samoa, p. 3,54.
Comparative vocabulary of 15 words, source unknown.

Rubi

Ray, Sidney H. Expedition to Torres Straits III, 482.

Saa

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-
Grammar. Melanesian Languages, p. 516.

Salayer

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

Salibabo

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

504

THE POLYNESIAN WANDERINGS.

Samoa

Churchill, William. A dictionary of Satnoan speech. (MSS.)

Samoa o le Vavau: Samoan contributions to Polynesian history. (MSS.)

Samoan grammar: an evolutionary essay toward the comprehension of

the speech method and linguistic mechanism of the Polynesian, a primordial
language of the isolating type. (MSS.)

Funk, Bernhard. Kurze Anleitung zum Verstandniss der samoanischen Sprache;
Grammatik und Vokabularium ; nebst einem Anhange, meteorologische
Notizen; mit einen Plan von Apia. Berlin: Ernst Siegfried Mittler
undSohn, 1893, pp. vi, 82.

Neffgen, H. Deutsch-Samoanisches Konversationsbuch. 'O le tusi-fetalia'i fa'asia-
mani ma fa'asamoa. Leipzig: Otto Ficker, 1904, pp. Hi, 64.

Grammatik der samoanischen Sprache nebst Lesestiicken und Worterbuch.

Wien und Leipzig: A. Hartleben, n. d., pp. viii, 168.

Newell, J. E. English and Samoan vocabulary ('0 le fa'asologa'upu Peritania
'ua fa'asamoaina) . Being Part Hi of the Grammar and Dictionary of
the Samoan Language, fourth edition, enlarged and revised. Malua,
Samoa: London Missionary Society, 1905, pp. 158.

Pratt, George. A grammar and dictionary of the Samoan language, with English
and Samoan vocabulary. Third and revised edition. (N. P.) : Printed
by the Religious Tract Society for the London Missionary Society, 1893,
pp. viii, 416.

Violette, L. Dictionnaire Samoa-Francais-Anglais et Francais-Samoa- Anglais pre-
cede d'une grammaire de la langue Samoa. Paris: Maisonneuve et Cie.,
1879, pp. xcii, 468.

Sanguir

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

Saparua

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

Sariba

Ray, Sidney H. Expedition to Torres Straits III, 482.

Sasar

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-
Grammar. Melanesian Languages, p. 337.

Savo

Codrington, R. H.

P- 39-
Gram mar .

SESAKE

Codrington, R. H.

P- 39-

Grammar .

Gabelentz, H. C. von der Die melanesischen Sprachen.

SlNAUGORO

Ray, Sidney H. Expedition to Torres Straits III, 482.

SUAU

Ray, Sidney H. Expedition to Torres Straits Til, 482.

Sulu Islands

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

Tagula

Ray, Sidney H.

Comparative vocabulary' of 70 words.
Melanesian Languages, p. 559.

Melanesian Languages,

Comparative vocabulary of 65 words. Melanesian Languages,
Melanesian Languages, p. 459.

Expedition to Torres Straits III, 482.

BIBLIOGRAPHY. 505

Tahiti

Jaussen, Tepano. Grammaire et dictionnaire de la langue Maorie, dialecte Tahitien.
Paris: Neia i te Nenei Raa no Belin, 1898, pp. 388.

Tami

Bamler, Missionar. Bemerkungen zur Gramtnatik der Tamisprache : Vokabular
der Tamisprache. Zeitschrijt fur afrikanische und oceanische Sprachen,
v, 19S.

Tangoan Santo

Annand, J. A grammar of the Tangoan Santo language. (In Macdonald's South
Sea Languages, II, 1.)

Tanna

Gray, W. Grammar of the Weasisi-Tanna language, with notices of other Tanna
dialects. (With vocabulary.) (In Macdonald's South Sea Languages,
II, 108.)

Gabelentz, H. C. von der. Die melanesischen Sprachen.

Turner, George. Comparative vocabulary of 121 words. Samoa, p. 354.

Comparative vocabulary of 15 words, source unknown.

Taupota

Ray, Sidney H. Expedition to Torres Straits III, 482.

Tauu

See Liueniua.

Tavara

Ray, Sidney H. Expedition to Torres Straits III, 482.

Teluti

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

Teor

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

TlDORE

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

Tobo

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354

Togean Islands

Rosenberg, C. B. H. von. De Togean-Eilanden (contains a vocabulary of 88
words). Amsterdam: Werken van het koninklijk Institut voor Taal-
Land- en Volkerkunde van Ncderlandsch-Indie; tweede Ajdeeling; ajzonder-
lijke Werken, 1865.

Tonga

Baker, Shirley Waldemar. An English and Tongan vocabulary, also a Tongan and
English vocabulary, with a list of idiomatic phrases; and Tongan grammar.
Aucklatid, N. Z.: pp. vi, 212, 44.
TONGAREWA

Smith, S. Percy. Tongarewa, or Penrhyn Island, and its people. Transactions
0} the New Zealand Institute, xxii, 85.
TUBETUBE

Ray, Sidney H. Expedition to Torres Straits III, 482.

UGI

Ray, Sidney H. Mittheilungen iiber drei Dialekte der Salomon-Iuseln. Zeitschrijt
fur afrikanische und oceanische Sprachen, ii, 54.

Ulawa

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-
Grammar. Melanesian Languages, p. 512.

506 THE POLYNESIAN WANDERINGS.

Uni

Ray, Sidney H. Expedition to Torres Straits III, 482.

Uvea

A. C. Dictionnaire Latin-Uvea a l'usage des eleves du college de Lano par les
missionaires maristes revu par le P. A. C, pretre mariste. Paris: Librairie
Poussielgue Frhes, 1886, pp. iv, 186.

Vaturanga

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-
Grammar. Melanesian Languages, p. 539.

VlTI

A. C. Essai de grammaire de la langue de Viti d'apres les manuscnts des mission-
aires maristes coordines. Paris: Poussielgue Freres, 1884.

Hazlewood, David, and James Calvert. A Fijian-English dictionary and a grammar
of the language. By the late David Hazlewood, Wesleyan missionary.
Second edition, with map, edited by James Calvert. London: Sampson
Low, Marston and Co. (1872), pp. 282, 64.

Moore, William. Handbook of the Fijian Language. Levuka: G. L. Griffiths,
1881, pp. 40.

VoivOW

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-
Grammar. Melanesian Languages, p. 322.

VURAS

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-
Grammar. Melanesian Languages, p. 345.

Wahai

Turner, George. Comparative vocabulary of 59 words. Samoa, p. 354.

Wango

Codrington, R. H. Comparative vocabulary of 70 words. Melanesian Languages,

P- 39-
Grammar. Melanesian Languages, p. 505.

Wedatj

Ray, Sidney H. Expedition to Torres Straits III, 482.

YENGIN

Comparative vocabulary of 15 words, source unknown.

INDEX.

PAGE.

o-mutations 50, 121, 154, 472

a 208

abrasion, final 200, 203,

205,228, 252,377

abrasion, frontal 192, 198, 206, 208,

213,225,235,254,273

abrasion, medial 191, 212, 213, 225,

237.259.302,385,422

a}apnc 337

a/t- 277

afisi 243

age of Melanesian-Polynesian . . 147

Ahtiago, check-list 152

ai 208

a'a 318

a'au 197

a'e 216, 402, 488

aki, verb-formative 291

ala 403, 488

alei 206

Alexander, William DeWitt 387

Alite, check-list 55

phonetic table 60

Allor, check-list 152

alo 193

alo 326

Alo Teqel, check-list 55

phonetic table 61

alunga 241

ama j.04 , 488

Amblaw, check-list 152

Amboyna, check-list 152

Ambrym, check-list 55

phonetic table 62

Amharic, check-list 175

atno 201

Andrews, Lorrin 278, 287

Aneityum, check-list 55

phonetic table 63

animals, introduced 16

Aniwa 3, 142

anu 417, 489

angi 317

apula 259

apungaleveleve 361

Arabia, Oceanic motherland ... 27

Arabic, check-list 175

Arag, check-list 55

phonetic table 64

Aramaic, check-list 175

Arishtan Shar 169

Aru, check-list 152

Aryan origin 25, 176, 178

aspiration 318, 346

Melanesian mutations 135

triply placed 133, 303

asu 286,481

ata 197

atali'i 194

PAGE.

ate 320, 483

ati 198, 479

aue 289

Avaiki-te-Varinga 21

Awaiya, check-list 152

Awalama, check-list 435

phonetic table 437

Baju, check-list 152

Baker, Shirley Waldemar 8, 225,409

Baki, check-list 55

phonetic table 65

Baliyon, check-list 152

Baravon, check-list 55

phonetic table 66

Basakrama, check-list 152

Batak, check-list 152

Batavia, check-list 152

Batumerah, check-list 152

Bauro, check-list 55

phonetic table 66

be 424

Belaga, check-list 56

phonetic table 67

Bellona Island, see Moiki.

Beu, check-list 152

Bicol, check-list 152

Bierian, check-list 56

phonetic table 68

Binue, check-list 152

bird of paradise 22

Bismarck Archipelago, change

of names 2, 14

black 331

blind 331

boiling 248, 409

Bolanghitam, check-list 152

Bongu 182

Boniki, check-list 435

phonetic table 438

bonito, courtesy language 352

Bopp, Franz; Polynesian theory 19, 25,

27,32, 183

Borneo, check-list 152, 172

borrowing, second-hand 199, 264,

323,324,389

Bougainville Id., check-list 56

phonetic table. 69

Bouton, check-list 152, 172

brain 224

bridegoom 6

Brierly Island, check-list 56

phonetic table. . 69

Brissi West, check-list 152, 172

brown color 330

Brumer Island, check-list 56

phonetic table. . 69

Bual, check-list 152

Buelow, W. von 4

507

508

INDEX.

PAGE.

Bugi, check-list *52

Bugotu, check-list 56

phonetic table 69

Buka, check-list 5°

phonetic table 70

traces of Polynesians ... 15

Bululaha, check-list 5°

phonetic table 71

Bum, check-list 152, 172

Caimarian, check-list 152

Cajeli, check-list 152

canoe voyaging conditions. .139. 49°. 49 *

cardinal points 2 J6

Caroline Islands, Polynesian mi-
gration 14, 16, 17

castaway drift 14, 45. I38> *44. x48

Celebes 21, *72

Ceram, check-list 152

mentioned 21, 172, 373

cerumen 284

Chaldee, check-list 175

Chamorri, check-list 152

Champa, check-list 152

cicatricial ornament 22

circumcision 265

closed roots, a Proto-Samoan

distinction 48, 148

coconut derivatives 224

Codrington, R. H 45, 268, 296, 297,

303, 307, 332, 357, 365, 366,
371,375.388, 392. 395, 403

concrete expressions 266

consonantal modulants 201, 219, 268,

298, 314, 320, 349, 35o, 403, 4"

consonants, easy to acquire. ... 36

mutation direction. 133, 177

cookery 24S, 409

count by fingers 3°5

course, determination of. 140, 141, 146, 216

courtesy speech 234, 236, 352, 374

crop colonies 139

currents, influence of 16

cutting 261

Dana, James Dwight iii

darkness 331

day 332

deafness 284

death 29i,374

Deeken, Richard, "Manuia

Samoa" 18

definitions, clumsy 8

squinting 8

Deni, check-list 56

phonetic table 71

determinant composition 255,

268,341,394

diphthongs 339

direction constants 216

Dobu, check-list 435

phonetic table 439

Domara 440

Dorey, check-list 152

Doura, check-list 435

phonetic table 440

PAGE.

drift theory of migration 14, 45,

138, 144, 148

Duba 440

Dufaure Island, check-list 56

phonetic table . 72

Duke of York, check-list 56

phonetic table . . 72

duplication 2 10

Dyak, check-list 152

e-mutations 51, 12

echo words

Efate, Turner's two word-lists. .

under Proto-Samoan in-
fluence

Polynesian phonetics. . . .

Viti phonetics

Viti-Samoan phonetics . .

Macdonald's dictionary. .

e}u

Ellice Group, migration

eneene

English borrowings from Poly-
nesian

Enua-kura

Ende, check-list

Epi, check-list

phonetic table

equatorial currents

Erakor vocabulary

Eromanga, check-list

phonetic table

Ethiopic, check-list

ethnic swarm

exclamation

/-mutations 54, I29> x33>

160, 168, 178,276

fafanga

fafangu

fafano

jafau

fafine

Fagani, check-list

phonetic table

fat

faititili

fa'a

fa'afisi

fa'afou

fa'alanu

fa'alangilangi

fa'alelefu

fa'alolo

fa'alolongo

fa'alongo

fa'apoa

Ja'asau

ja'asusu

fa'atuai

fa'atulutulu

Fanning Island, migration

fanofano

fano

fangufangu

famia

fasi

, 154,472

302

10

45,46,47

42

35,37,42

38,41

6

313,482

17, 180

194

264
22

152
56
73
16
10
56
74

175

170
289,402

, 136, 137,

303,476
269
217
194
328

337,483

56

75

218

240

269,480
277

327,483
396
197

313
224

399
398
276
282
410
3"
425
i/
194
226
201

341,484
203

INDEX.

509

PAGE.

fata. 343

jataninga 194

fati 219

fatu 228

fatu 344

fatu'uhi 228

fatulau 229

fatumanava 344

fatunga 229

fan 328

h 211,479

feel 400

fe'au 218

felelei 42 1

fenguingui 393

fesulua'i 405

fesuliina'i 405

fetu 199

fetulele 42 1

fia 271

fifi 277

Fiji, see Viti.

Fileni 4

fili 178, 290

fingers in counting 365

fisi 277

fiso 196

five, hand 364

flight 295

flora, provinces 17

foanga 204

foe 429

folau 334

fonoh 333

foreign novelty 234, 259

Fornander, Aryan theory 178, 184

fotu 245

Fotuna, differenced in spelling. 3

settlement 142, 285

fou 327.483

four, the perfect number 19

frying ' 409

fua 297, 426, 489

fuata 321,427

fufula 202

Jufulu 275

fufuti 276

fui 292

fula 202

fuli 335

fulu 267, 480

fulua 275

funga 292, 427

fusi 292

futi 276

futi 388,487

Futuna, Tongafiti settlement. . . 287

397,408,411

Gah, check-list 152

Galavi, check-list 435

phonetic table 441

Galela, check-list 152

Galoma, check-list 435

phonetic table 442

Gani, check-list 152

general migration theory 19

PAGE.

Gilbert Archipelago 17, 180, 274

Gilolo, check-list 152

mentioned 21, 172

Gog, check-list 56

Gog, phonetic table 76

Goram, check-list 152

Guadalcanar, check-list 56

phonetic table ... 77

Guaham, check-list 152

A-mutations 53, 126, 132, 135, 157,

166,227,287,323,324,475

hack 9, 205

Halmahera 17

hand activity 350

hand, as five 364

Havannah Harbor mission sta-
tion 10, 11

Hawaii, direct migration 44, 49, 262,

2 79.337,346,408
Sumatra voyage to. . . . 19

Hawaiki 13, 17, 49, 184

"Hawaiki" 20,26, 179, 184

hearing 401

Hebrew, check-list 175

bog 427

Hula, check-list 435

phonetic table 443

Humboldt, Wilhelm von 19, 183

i-mutations 51, 122, 155, 472

Iai, check-list 56

phonetic table 77

ifo 262, 480

i'a 350,484

i'amanu 372

ilo 429

Ilocan, check-list 152

Indonesia, common to both mi-
gration theories . . 19
Malayan settlement . 169
Indonesian, community of speech

with Polynesian. 151, 170
contact with Poly-
nesian 169

not a linguistic link. 20
Polynesian exit

(Thilenius) 16

infixation 212

inu 376, 486

isa 401

isu 348, 484

iti 230

Java 21, 169, 184

check-list 152

Jobi, check-list 152

^-mutations 52, 124, 131, 135,

136,156, 164,255,474

'in Tongan 225

Kabadi, check-list 56, 435

phonetic table 78, 444

Kabakada, check-list 56

phonetic table 78

'ai 191, 478

Kaili, check-list 152

510

INDEX.

PAGE.

Kaioa Island, check-list 152

'a'a 318

'a'asa 251

'a'e 402, 488

Kalil, check-list 56

phonetic table 78

'ama'ama 260

'amu 277

Kandayan, check-list 152

kani 191, 478

'apa'au 295

'apatd 295

Kapingamarangi 15, 17

'apu 246

karavau 388

'aii 355, 485

'ato 255

'au 218

'an 278

'au 295

'auali'i 353

'aualuma 295

'au'auli 353

Kawi, check-list 152

Kayan, check-list 152

Keakalo 444

Keapara, check-list 435

phonetic table 445

'eli 347,484

'ete 256

'* 219

'iato 230

'ie 196

'i'ila 246

'i'ite 294, 481

'Hi 278,481

'ili'ili 355

Kilinailau, Polynesian settle-
ment 15

'iliola 356

'imoa 296

King, census 366

check-list 56

phonetic table 79

'ini 279

'iole 296

Kiriwina, check-list 435

phonetic table 446

Kisa, check-list 152

'isumu 148, 296

Kiviri, check-list 435

phonetic table 447

Kiwa, sea of 13, 184

'oju 220

Koita 447

'o'ai 204

'olo 220

'ona 205

'oti 256

'oto 256

Kramer, Dr. Augustin 265

Kubiri, check-list 435

phonetic table 448

'«'« 349

'uli 255

'upenga 231

'Upolu 23

PAGE.

'utu 357- 485

Kwagila, check-list 435

phonetic table 448

/-mutations 52, 123, 131, 135,

136, 157. 165, 219, 223, 234, 268,
300, 304, 305, 334. 361, 385. 473, 487

la 304

labials, last acquired in speech.. 332

lingual mutation 136

palatal mutation 136

Melanesian difficulty. . . 42, 134,
212, 263,332

lac 304, 369

La Farge, John 330

Lakatoi language 186, 490

la'au 353

la'ei 209

Lakon, check-list 56

phonetic table 80

la'u 209

laid 304

lalau 397

lalo 213,216

Lamassa, census 366

check-list 56

phonetic table 81

Lambell, check-list 56

phonetic table 82

Lampong, check-list 152

Landa, check-list 152

lanu 396, 487

langa 197

langi 318,324,359,485

lango 360, 485, 487

lango 257

langofuju 384

langomumu 384

langona 398

Lariko, check-list 153

Laur, check-list 56

phonetic table 83

lase 209

last 246

laso 297

latele 238

lau, sea 215

lau 397,487

lau' a' a 3*8

laulau 397

Lauta iv

lauulu 268

lava 358

lavalava 28

lavasi 28

lave 28

lavea 231

leangiangi 421

lefu 3T3,482

lele 42 x

lelefu 313

Lemaroro, check-list 56

phonetic table 84

leo 232

leo 363

Leon, check-list 56

phonetic table 84

INDEX.

511

229,

PACE.

153

217

56

84

176

479
234
486

233
135
136

234
142

56

85
153
197
368

26

Liang, check-list

lielievale

Lifu, check-list

phonetic table

like-/^c

li'i

liliu

lima 363.367.

lingi

Unguals, labial mutation

palatal mutation

liu

Liueniua 3, 15, 17,22,45,46,

Lo, check-list

phonetic table

Lobo, check-list

lofa 28,

lofia

Logan, J. R., Polynesian theory.

lolo 223

lolo 368

lolo'u 221

longo 398, 487

longona 398

lua'i 279

lue 235

Lukunor, Polynesian settlement 22

lulu 235

lulua 279

w-mutations 54, 128, 132, 136,

137. 159. 167,253,378,476

ma 298

Mabuiag, check-list 435

phonetic table 450

Macassar, check-list 153

Macdonald, Rev. Daniel. Efate

dictionary 1,5

Madura, check-list 153

maea 195

Maewo, check-list 56

phonetic table 86

Mai 4, 45, 46, 47, 142

check-list 56

phonetic table 86

Polynesian settlement 46

maja 194

mafine 337

mafuli 335

Magindano, check-list 153

Mahri, check-list 175

Mailu 449

ma'i 379

ma'ini 194

Makura, check-list 57

phonetic table 87

malae 305, 369

Malagasy, check-list 152

malama 378

malamalama 378

Malanta, check-list 57

phonetic table 87

Malay, check-list 151

race, early distory 169

Malayan influence Post-Poly-
nesian 322,324,335,351,357,

360, 361, 373, 375, 380, 387, 398

PAGE.

Malayo-Polynesian family 20, 25, 26,

170, 171, 173, 183

Malekula, check -list 57

phonetic table 87

Malietoa, Savea the first 45

Laupepa iv

malingi 233

Malo, check-list 57

phonetic table 89

main 375

malil 370, 486

malu 369

mama 280

mamac 194

Mame, check-list 153

Manahune traditions 22

Manatolo, check-list 153

manifinifi 29S, 481

mano 195

matin 218, 372, 486

manu'a 234

Manu'a 23

Sumatra voyage to 19

manga 280

mangungu 393

mao 203

mao 206

mapelu 217

Marina, check-list 57

phonetic table 90

Marshall Islands 14

masa 257

masi 281

masina 406

Massaratty, check-list 153

Massim 449

masunu 407

masiisii 207

maia 299

tnata 380, 486

Matabello, check-list 153

matai 195

mataisau 195

mata'u 301 , 482

matala 223

matali'i 195

matangi 317

Matamatame 45, 180, 379

mate 373

Matema 142

mati 281

matolutolu 199, 238

Matu, check-list 153

matua 271,311

Matupit, check-list 57

phonetic table 91

man 207

maua 207

mdni 258

maunga 195

Mayapo, check-list 153

Mekeo, check-list 435

phonetic table 449

Melanesia, a geographical unit . . 2
dialects abundant. . . 10
early history un-
known 13

512

INDEX.

PAGE.

Melanesia, little attractive to

voyagers 3

mutation tables. ... 59

problems of i

Melanesian contribution to Poly-
nesia 148

q 332

races 13. lSl

Mele, check -list 57

phonetic table 91

vocabulary 10

mele'i 253

melomelo 3°°

Menado, check-list 153

Menankabau, check-list 153

Merlav, check-list 57

phonetic table 92

metathesis 218, 223, 227, 230, 239, 241,

262, 269, 275, 292, 295, 314, 317, 330,
332, 346, 361, 371, 376, 377, 39i. 4i3. 417

mianga 375

mimi 375

Miriam, check-list 435

phonetic table 450

Misima, check-list 435

phonetic table 45 *

Moanus 147. 360

check-list 57

phonetic table 93

Moiki 4, 45, 46, 142, 146

molemole 244

Molucca i5i 17

check-list 153

momona 281

monkey 388

Mon-Khmer 25

mono i97i 333

moon 33°

Morella, check-list 153

Moriori migration 271, 335, 414

Mosin, check-list 57

phonetic table 94

Moso 299

mosquito 386

Mota, check -list 57

phonetic table 96

transitive verb suffix. . . . 305, 392

Motlav, check-list 57

phonetic table 95

motu 383

Motu, check-list 57

phonetic table 98

Polynesian settlement. . . 21, 147
eastern Polynesian asso-
ciations 349,361, 362,385

Mugula, check-list 435

phonetic table 452

mui 383

Mukawa, check-list 435

phonetic table 453

muli 384

Miiller, Friedrich 19, 26, 183

Miiller, Max, Polynesian theory . 25, 27,

32,34.183

mumu 383

mumu 383

PAGE.

muna 383

Murray Island, check-list 57

phonetic table. . 99

Murua, check-list 435

phonetic table 454

musumusu 383

mutation, basic principles 177

Indonesia-Melanesia-
Polynesia 163

by series 133

Mysot, check-list 153

myth fabrication 19

ra-mutations 53, 125, 132.. 135,

137,157, 165,346,386,474

Nada, check-list 435

phonetic table 455

nafa 234

namu 386, 487

namu 221

Natalava, check-list 57

phonetic table 99

national life unknown 170

natu 198, 479

navigation, Polynesian 21

Navigators' Islands 139

navu 20S, 479

neju 313

net 200

nenefu 313,482

Nengone, check-list 57

phonetic table 99

neutral vowel 35, 36

New Britain, check-list 57

New Caledonia, check-list 57

New Georgia, check-list 57

phonetic table. . . 100

New Guinea 14, 21, 22, 26, 141,

142, 172, 173, 180, 182

check-list 57

north coast voyages 2 1
obstacle to Malayan

settlement 141, 173

New Hanover 17

New Hebrides 137, 140, 141, 145, 180

New Ireland, check-list 57

New Zealand, direct migration.. 44, 49,
262, 279, 369, 370, 379, 408, 423

Nicobar, check-list 153

Nifilole, check-list 58

phonetic table 104

nijo 3°2, 482

Ninigo 15, l7

Nisan, Polynesian settlement . . 15

niu 390,487

Niu-Sisila 19

nofo 259

nonu 254

nonufi'afi'a 254

Norbarbar, check-list 58

phonetic table 105

nuanga 209

Nuclear Polynesia. ... 2, 19, 25, 42, 43, 179

Ef ate identifications . . 43

Nuguria 4, 15, 16, 17, 142

check list 433

INDEX.

513

PAGE.

Nukumanu 4

Nukuoro, Samoan settlement .22, 42, 346,

360,385,415,419

Sumatra voyage to . . 19

nunu 209

Nggao, check-list 57

phonetic table 101

Nggela, check-list 57

phonetic table 102

ng-mutations 52, 124, 131, 136, 137,

156,163,233,349,473

ngafa 204

nga'e 204

ngalolo. 224

ngalu 245

ngasu 221

ngata 258

ngeli 388

ngingili 393

ngongo 196

ngongolo 392

ngu 393

Nguna, check-list 58

phonetic table 103

ngutu 349

o-mutations 51, 122, 155, 472

Oceanic language, Macdonald's

statement 26

phonology 27

race 26,31,32

Oiun, check-list 435

phonetic table 456

oh 39i

Omba, check-list 58

phonetic table 106

one 250

Ongtong Java, see Liueniua.

opeope 249

otaola 207

oti 274, 373

/^-mutations 54, 130, 133, 135,

136, 161, 168,264,477

pa 274

Paama, check-list ... 58

phonetic table . 107

Pak, check-list 58

phonetic table 107

pa'angungu. ._ 393

Pala, phonetic table 108

palatals, labial mutation 136

lingual mutation 136

Palau Islands 16, 17

palolo 8, 224

Pampangas, check-list 153

Panaieti, check-list 435

phonetic table 457

Pangasinan, check-list 153

Pani, check-list 153

papa 325

particles 29

Paumotu, a primitive type .... 311

peau 244

pepe 251,479

pelu 217

pelu'i 217

PAGE.

Philippine Islands 16, 172, 173

Phcenix Group 180

Pig--.- 427

pi'opi'o 243

pili 289

Pilot chart, U. S N. Hydro-
graphic Office 18

pito 293, 481

plants, introduced 16

Pleiades 195

P° 330,483

poa 276

poapoa 221, 276

Pokau, check-list 435

phonetic table 458

po'u 194

pola 253

polata 253

poloa'i 291

Polynesian expulsion from In-
donesia 170

inclusions 4, 42, 45, 141

migration 13, 25

mutation tables ... 50

origin 179,183

verge. . . .4, 15, 42, 45, 142, 181

pongipongi 330

pongisa 330

pork 428

pottery 409

pronouns 29

prosit. . 259

Proto-Samoan migration . . 38, 43, 45, 48,

49, 50, 138, 179, 184, 246, 259,

262, 285, 303, 312, 362, 379, 431

pua'a 427

pula 329,483

pule 196

Pulotu 13

pulou 206

pulu 336

pulupulu 336

puni 333

pupil 202

pupula 329

pupulu 336

pupuni 333

puna 244

pute 293, 481

Qakea, Polynesian settlement. . 46

quality of Polynesian content. . 143

r-grasseye" 38, 121,315,327

rain, sensitiveness to 16

Raluana, check-list 58

phonetic table 109

Rapanui, check-list 433

Raqa, check-list 436

phonetic table 459

Rason, Capt. Ernest, R. N. . . . 11

Ray, Sidney H., Melanesian

studies 4, 182, 252, 489

reinforced consonants 38, 199,

223,264,324

Rennel Island 4,45,46, 142, 146

Retan, check-list 58

514

INDEX.

PAGE.
109
436
460
153
58
109
436
461
uma 307 1 482

Retan, phonetic table . .
Roro, check-list

phonetic table . . . .

Rotti check-list

Ruavatu, check-list

phonetic table
Rubi, check-list

phonetic table.

.^-mutations 53. I2°\

158, 166, 203, 247

Saa, check-list

Saa, phonetic table

Saba myth

sailing quality

sola

Salayer, check-list

motu o Salaia

Salibabo, check-list

Salu, check-list

Sambawa, check-list

Samoa 14, 17, 19, 23. 49

Samoa stream 147

samusamu

Sandol, check-list

Sanguir, check-list

Santa Cruz Group

Santo, check-list

phonetic table

Sapartia, check-list . .

Sariba, check-list

phonetic table
Saru, check-list

Sasar, check-list

phonetic table
sasau

sasau

Sassac, check-list . .
Satawal, check-list

saupapa

sailpapa

Savai'i

savali

Savea, the first Malietoa

Savo, check-list

Savo, check-list

phonetic table

Savu, check-list

Sawaiori, proposed by Whitmee
seamanship, Polynesian ... 18, 21

seed of speech

sele

selu

Semitic theory discussed

sense perception

Sesake, check-list

phonetic table

mentioned

seven not mystic

Sheppard Group

sieve theory 14

132, 137,
,287,475

58

1 10

13
216

402,488

247
260

153
241

153

153

153

, 138, 180

, 173,408

203

153

153

140, 180

58

1 11
247
153
436
462

153
247
58
112
282
283
153
153
282

195, 283

283

17.23

210

45

153

58

112

153

35, 169

, 138, 140

350

260

218

26, 176

399

58

113

47

19

45

, 138, 148

PAGE.

Sikayana 3, 15, 22, 142, 180

si'i 261

sili 405

stlinga 405

Silong, check-list 153

sina 406, 488

Sina 483

sina'aiunga 242

Sinaugoro, check-list 436

phonetic table 463

Sirang, check-list 153

sisili 405

Siwa, check-list 153

skull 224

sky 359

smell 400

Smith, S. Percy, general migra-
tion theory 20, 21, 26,

180,184,383

"Hawaiki" 20, 26, 179, 184

Tongafiti migration 180

smoke 286

soa 306

Sokotra, check-list 175

<;5loi 262

Solomon Islands, check-list 58

mentioned. ... 18, 21, 137, 140, 180

Solor, check-list 153

songi 306

spirant mutation 340

strike 266

sit 221

sua 221

Suau, check-list 436

phonetic table 464

substantive verb 260, 424

sn'i 225

Sula, check-list 153

suit 211

Sulu, check-list 153

sulu 247

sulu 405

sulufa' i 405

sulunia'i 405

Sumatra 18, 19, 25, 169, 172

Sunda, check-list 153

sununga 407

susu 410,488

susulu 247

susunu 407

Swallow Group 45, 46

sword 217

Syriac, check-list 175

/-mutations 53, 127, 132, 135, 137,

158, 167, 177, 221, 238, 256,
264, 265, 302, 316, 321, 345, 374, 475

ta 411

tabu in speech 4S, 299

tae 414

Taema 138,414

tafa 211

taje 264

tafuli 335

Tagalog, check-list 153

Tagula, check-list 436

phonetic table 465

INDEX.

515

PAGE.

tai 216, 418

taimasa 257

ta'alaelae 304

ta'ili 355

ta'u 225

tola 223

tala 238

talafa'aoli 274

talai 310

talele 42 1

tali 200

talinga 4i5> 488

talingatuli 284

talo 415.488

talosanga 236

tama 272, 480

tamafafine 337

Tanna, check-list 58

phonetic table 1 14

tanu 308

ianga 265

tangata 243

tangi 412

tango 401

Tangoan Santo, check-list 58

phonetic table. 115

tangulu 393

tangulu 293

tao 248

tapa 248

tapu 263,480

taste 400

tata 411

tatala 223

tatalo 236

tatangi 412

tau 237

tail 248

tan 283

tau 309

tau 283

Taui 15

taula 237

tatitnafa 236

Taupota, check-list 436

phonetic table 466

tausanga 309

Tauu 4. x42

tauvale 237

Tavara, check-list 436

phonetic table 467

teje 265

tei 193.478

telaicla 238

tele 238

telea'i 211

Teluti, check-list 153

Teor, check-list 153

tepa 201

Ternati, check-list 153

"thee and thou" 302

Thai, as a Polynesian source. ... 25

Thilcnius, Dr. G., sieve theory . . 14, 18, 19,

21,23, 138, 140, 141

Ticopia 3, 16, 45, 46, 142

Tidore, check-list 153

Tigre, check-list 175

tila 239

PAGE.

Tilafainga 138

tilotilo 422

Timor, check-list 153

tio 422

toa 423

Tobo, check-list 153

Togean Islands, check-list 153

to' a 423

to'elau 215

Tokelau Islands 14

to'i 310

to'o 420

Tongariti migration 43, 45, 48, 49, 138, 179,

180, 245, 259, 262, 285, 2S6,

295,297,303,312,362,379

Tongariti, traces in Melanesia . . 138, 275

toto 395

touch 401

Treasury Island, check-list 58

Tregear, Edward. ... 20, 21, 151, 219, 231,
236,237,241,245,268,279,

280, 282, 297, 307, 312, 362, 428

triliteralism 28, 176

Tringanu, check-list 153

tu 424

tualafine 337

tuai 311

Tuatau, driftwood idol 144

Tubetube, check-list 436

phonetic table 468

tufa 1S8, 367

tui 225

tu'i 266

tu'u 226

tula 424

tula j.24

tuli 284

tuliii 425

tulutulu 425

tuluvao 425

tuma 200

tunu 407

tungia 407

tupunga 214

Turanian origin 25

tutu 407

Tutuila 23

tutula'i 424

tutulu 425

w-mutations 51, 122, 155, 305, 473

u 200

« 485

ua 322

ua 322,483

Uea 45, 145

check-list 58

phonetic table 115

uii 316, 482

ufi 315

Ugi, check-list 58

phonetic table 115

uila 345

ula 430

Ulawa, check-list 58

phonetic table 116

ule 431

Ulea, check-list 153

516

INDEX.

PAGE.

uluulu 267, 480

uma 235

umelolo 224

umu 199, 479

una 312

anafi 312

Uni, check-list 436

phonetic table 469

unusi 312

iita 216

uta 285

Utanata, check-list 153

utu 242

Uvea 397

•v-mutations 54, 128, 133,

135, 136, 160, 167,476

vaevae 201

vai 339, 484

Vaiqueno East, check-list 153

vaito'elau 216

vale 251

valu 326

vdlu 326

valusanga 326

Vanikoro, check-list 58

phonetic table 1 16

Vanua Lava, check-list 58

vatele 238

Vaturanga, check-list 58

phonetic table. ... 117

telo 202

Verge, Polynesian; defined .... 4

Thilenius theory of set-
tlement 14

windward position 142

veve 194

Vila, seat of administration .... 10

PAGE.

vili 220

Visayas, check-list 153

Viti 20, 23, 25, 138, 180

race mixture 2

Efat6 concords 42

speech mixture 35 > 47

stream 147, 173,211,290,

328,350,396,408,491

Viti, Samoan phonetics 39

vivini 290

Volow, check-list 58

phonetic table 118

volu 346

vowels, mutation frequency ... 36

difficult to acquire 36

Vuras, check-list 58

phonetic table 119

Wagawaga 470

Wahai, check-list 153

Waigiou, check-list 153

Waima, check list 436

phonetic table 470

Wango. check-list 58

phonetic table 120

Wawn, Capt. William T 4

weather gage 140

Wedau, check-list 436

phonetic table 471

white 329

Whitney, William Dwight 183

winds and migration 16, 216

obstacles to migration . . 21

set the course 141

windward islands, significant set-
tlement 142

^-initial, Viti 319.321

"".si

A I S T II \ I I A

MIGRATION TRACKS l!li:oi(,ll MELANESIA

■

The Polynesian Wanderings

Tracks of the Migration Deduced from an Examination

of the Proto-Samoan Content of Efate and

other Languages of Melanesia

BY

WILLIAM CHURCHILL

Sometime Consul-General of the United Stales in Samoa and Tonga,

M ember of the Polynesian Society, the Hawaiian Historical

Society, the American Philological Association

PUBLISHED BY

THE CARNEGIE INSTITUTION OF WASHINGTON
1911

m^m mt ' ■

MBL/WHOI LIBRARY

UH 1SJE 5